CN118134210A - Carbon footprint management method and system for steel production - Google Patents
Carbon footprint management method and system for steel production Download PDFInfo
- Publication number
- CN118134210A CN118134210A CN202410550914.9A CN202410550914A CN118134210A CN 118134210 A CN118134210 A CN 118134210A CN 202410550914 A CN202410550914 A CN 202410550914A CN 118134210 A CN118134210 A CN 118134210A
- Authority
- CN
- China
- Prior art keywords
- carbon footprint
- fluctuation
- emission reduction
- data
- reduction potential
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 263
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 263
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 192
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 86
- 239000010959 steel Substances 0.000 title claims abstract description 86
- 238000007726 management method Methods 0.000 title claims abstract description 71
- 230000009467 reduction Effects 0.000 claims abstract description 151
- 238000000034 method Methods 0.000 claims abstract description 59
- 230000002068 genetic effect Effects 0.000 claims abstract description 45
- 238000005265 energy consumption Methods 0.000 claims abstract description 22
- 239000005431 greenhouse gas Substances 0.000 claims abstract description 18
- 238000004458 analytical method Methods 0.000 claims description 39
- 238000004140 cleaning Methods 0.000 claims description 24
- 238000013507 mapping Methods 0.000 claims description 24
- 238000011156 evaluation Methods 0.000 claims description 22
- 238000009826 distribution Methods 0.000 claims description 19
- 238000005457 optimization Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 18
- 230000015654 memory Effects 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 15
- 238000012300 Sequence Analysis Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 8
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 5
- 230000008859 change Effects 0.000 description 19
- 230000008569 process Effects 0.000 description 19
- 238000012544 monitoring process Methods 0.000 description 10
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 8
- 230000035772 mutation Effects 0.000 description 8
- 230000000737 periodic effect Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 5
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 4
- 229910052742 iron Inorganic materials 0.000 description 4
- 230000007423 decrease Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 229910002092 carbon dioxide Inorganic materials 0.000 description 2
- 239000001569 carbon dioxide Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 230000001932 seasonal effect Effects 0.000 description 2
- 238000012731 temporal analysis Methods 0.000 description 2
- 238000000700 time series analysis Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000002759 chromosomal effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000002737 fuel gas Substances 0.000 description 1
- 230000008303 genetic mechanism Effects 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000003334 potential effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- Biophysics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Genetics & Genomics (AREA)
- Physiology (AREA)
- Educational Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of carbon emission, and discloses a carbon footprint management method and system for steel production. The method comprises the following steps: acquiring energy consumption data, greenhouse gas emission data and production line state data in the steel production environment in real time to obtain an original production data set; extracting key features from the original production data set to obtain a key feature set; inputting the key feature set into a preset genetic programming algorithm to construct a carbon footprint prediction model, so as to obtain a target carbon footprint prediction model; inputting the original production data set into a target carbon footprint prediction model to predict the carbon footprint quantity, so as to obtain a predicted carbon footprint quantity; the predicted carbon footprint is input into a preset self-adaptive resonance theoretical model to identify emission reduction potential areas, a plurality of emission reduction potential areas are obtained, carbon footprint management is carried out on the steel production environment based on the emission reduction potential areas, and the accuracy and the efficiency of the carbon footprint management of steel production are improved.
Description
Technical Field
The application relates to the field of carbon emission, in particular to a carbon footprint management method and system for steel production.
Background
With the increasing emphasis on environmental protection and climate change worldwide, the iron and steel industry is one of the largest carbon dioxide emissions sources worldwide, and the carbon footprint management thereof is becoming the focus of industry attention. The prior art evaluates and manages carbon emissions by monitoring and recording information such as energy consumption data, greenhouse gas emissions, and line status during steel production. These methods often rely on traditional statistical analysis techniques to calculate the overall carbon emissions by post-processing the collected data, and empirical emission reduction measures to attempt to reduce the carbon footprint.
However, the prior art suffers from several significant drawbacks. Due to the lack of highly refined carbon emission monitoring and real-time data analysis capabilities, existing methods have difficulty in accurately identifying high carbon emission areas and times during production, thereby limiting the pertinence and effectiveness of emission abatement measures. Secondly, when complex production data are processed by the traditional method, hidden carbon emission reduction potential in the data cannot be fully mined, and deep understanding of a carbon emission fluctuation mode in the production process is lacking, so that a carbon footprint management strategy is not intelligent and adaptive enough. The existing carbon footprint management strategy lacks systematic optimization and real-time feedback adjustment mechanisms, and is difficult to adapt to dynamic changes of the production process and fluctuation of environmental conditions.
Disclosure of Invention
The application provides a carbon footprint management method and a system for steel production, which are used for the accuracy and the efficiency of carbon footprint management of steel production.
In a first aspect, the present application provides a carbon footprint management method for steel production, the carbon footprint management method for steel production comprising: acquiring energy consumption data, greenhouse gas emission data and production line state data in the steel production environment in real time to obtain an original production data set;
Extracting key features from the original production data set to obtain a key feature set;
Inputting the key feature set into a preset genetic programming algorithm to construct a carbon footprint prediction model, so as to obtain a target carbon footprint prediction model;
inputting the original production data set into the target carbon footprint prediction model to predict the carbon footprint quantity, so as to obtain a predicted carbon footprint quantity;
inputting the predicted carbon footprint quantity into a preset self-adaptive resonance theoretical model to identify emission reduction potential areas, obtaining a plurality of emission reduction potential areas, and performing carbon footprint management on the steel production environment based on the emission reduction potential areas.
With reference to the first aspect, in a first implementation manner of the first aspect of the present application, the extracting key features from the raw production dataset to obtain a key feature set includes:
performing data cleaning on the original production data set to obtain a cleaning data set;
performing time sequence analysis on the cleaning data set to obtain a time sequence analysis result;
Based on the time sequence analysis result, extracting frequency domain features of the cleaning data set to obtain a frequency domain feature set;
Inputting the frequency domain feature set and the cleaning data set into a preset long-short-time memory network to extract short-term fluctuation features, so as to obtain an initial feature set;
And performing dimension reduction processing on the initial feature set through a t-distribution random neighborhood embedding algorithm to obtain the key feature set.
With reference to the first aspect, in a second implementation manner of the first aspect of the present application, inputting the key feature set into a preset initial genetic programming algorithm to perform carbon footprint prediction model construction, to obtain a target carbon footprint prediction model, includes:
and initializing parameters of the genetic programming algorithm, wherein the initialization parameters comprise: algorithm crossover probability, variation probability and stopping standard value;
performing coding processing on the key feature set to obtain a coding feature set corresponding to the key feature set;
inputting the coding feature set into an initialized initial genetic programming algorithm to carry out data evolution to obtain a target program solution;
Performing performance evaluation on the target program solution through a preset fitness function to obtain a performance evaluation value of the target program solution;
and when the performance evaluation value meets the stopping standard value, constructing a carbon footprint prediction model of the initial genetic programming algorithm according to the target program solution to obtain a target carbon footprint prediction model.
With reference to the first aspect, in a third implementation manner of the first aspect of the present application, when the performance evaluation value meets the stopping standard value, constructing a carbon footprint prediction model of the initial genetic programming algorithm according to the target program solution, to obtain a target carbon footprint prediction model, including:
When the performance evaluation value meets the stop standard value, extracting relevant features of the target program solution to obtain a relevant feature set of the target program solution;
generating an expression structure of the target program solution, parameter information of the target program solution and operation logic of the target program solution according to the related feature set;
Constructing a carbon footprint mapping relation according to the expression structure of the target program solution, the parameter information of the target program solution and the operation logic of the target program solution to obtain a target mapping relation;
carrying out algorithm optimization parameter analysis on the initial genetic programming algorithm according to the target mapping relation to obtain an algorithm optimization parameter set;
and carrying out carbon footprint prediction model construction on the initial genetic programming algorithm based on the algorithm optimization parameter set to obtain the target carbon footprint prediction model.
With reference to the first aspect, in a fourth implementation manner of the first aspect of the present application, the inputting the raw production dataset into the target carbon footprint prediction model to perform carbon footprint prediction, to obtain a predicted carbon footprint includes:
inputting the original production data set into the target carbon footprint prediction model for L1 regularization treatment to obtain regularized data;
Performing nonlinear feature mapping on the regularized data to obtain nonlinear features;
according to the nonlinear characteristics, carrying out evolution trend analysis on the original production data set to obtain evolution trend data;
Carrying out data enhancement processing on the original production data set according to the evolution trend data to obtain an enhanced data set;
Inputting the enhanced data set into the target carbon footprint prediction model for stacking parameter analysis to obtain a plurality of stacking parameter sets;
Respectively carrying out stacking parameter feature recognition on each stacking parameter set to obtain stacking feature sets;
Performing carbon footprint distribution analysis according to the stacking feature set to obtain carbon footprint distribution data;
and predicting the carbon footprint according to the carbon footprint distribution data to obtain the predicted carbon footprint.
With reference to the first aspect, in a fifth implementation manner of the first aspect of the present application, the inputting the predicted carbon footprint into a preset adaptive resonance theoretical model to identify an emission reduction potential area, to obtain a plurality of emission reduction potential areas, and performing carbon footprint management on the steel production environment based on the plurality of emission reduction potential areas, includes:
inputting the predicted carbon footprint into the self-adaptive resonance theoretical model to identify footprint fluctuation points, so as to obtain a plurality of footprint fluctuation points;
respectively carrying out footprint fluctuation amplitude identification on each footprint fluctuation point to obtain fluctuation amplitude data of each footprint fluctuation point;
based on the fluctuation amplitude data of each footprint fluctuation point, respectively carrying out fluctuation category identification on each footprint fluctuation point to obtain the fluctuation category of each footprint fluctuation point;
Based on the fluctuation category of each footprint fluctuation point, carrying out fluctuation area segmentation on the predicted carbon footprint to obtain a plurality of fluctuation areas;
And carrying out emission reduction potential area identification on a plurality of fluctuation areas based on a preset emission reduction fluctuation standard type set to obtain a plurality of emission reduction potential areas, and carrying out carbon footprint management on the steel production environment based on the emission reduction potential areas.
With reference to the first aspect, in a sixth implementation manner of the first aspect of the present application, the performing emission reduction potential area identification on the plurality of fluctuation areas based on the preset emission reduction fluctuation standard type set to obtain a plurality of emission reduction potential areas, and performing carbon footprint management on the steel production environment based on the plurality of emission reduction potential areas includes:
Carrying out fluctuation mode identification on the emission reduction fluctuation standard type set to obtain a fluctuation mode set;
Carrying out fluctuation pattern matching on each fluctuation area based on the fluctuation pattern set to obtain a target fluctuation pattern of each fluctuation area;
Carrying out emission reduction potential area identification on a plurality of fluctuation areas through a target fluctuation mode of each fluctuation area to obtain a plurality of initial emission reduction potential areas;
Respectively carrying out emission reduction analysis on each initial emission reduction potential area to obtain emission reduction of each initial emission reduction potential area;
And carrying out area screening on the initial emission reduction potential areas through the emission reduction amount of each initial emission reduction potential area to obtain a plurality of emission reduction potential areas, and carrying out carbon footprint management on the steel production environment based on the emission reduction potential areas.
In a second aspect, the present application provides a carbon footprint management system for steel production, the carbon footprint management system for steel production comprising:
The acquisition module is used for acquiring energy consumption data, greenhouse gas emission data and production line state data in the steel production environment in real time to obtain an original production data set;
the extraction module is used for extracting key features of the original production data set to obtain a key feature set;
The construction module is used for inputting the key feature set into a preset genetic programming algorithm to construct a carbon footprint prediction model, so as to obtain a target carbon footprint prediction model;
The prediction module is used for inputting the original production data set into the target carbon footprint prediction model to predict the carbon footprint quantity so as to obtain a predicted carbon footprint quantity;
the identification module is used for inputting the predicted carbon footprint quantity into a preset self-adaptive resonance theoretical model to identify the emission reduction potential areas, so that a plurality of emission reduction potential areas are obtained, and carbon footprint management is performed on the steel production environment based on the emission reduction potential areas.
A third aspect of the present application provides a carbon footprint management apparatus for steel production, comprising: a memory and at least one processor, the memory having instructions stored therein; the at least one processor invokes the instructions in the memory to cause the steel production carbon footprint management device to perform the steel production carbon footprint management method described above.
A fourth aspect of the present application provides a computer-readable storage medium having instructions stored therein, which when run on a computer, cause the computer to perform the above-described carbon footprint management method of steel production.
According to the technical scheme provided by the application, through collecting the energy consumption data, the greenhouse gas emission data and the production line data in the steel production environment in real time and combining a genetic programming algorithm and a self-adaptive resonance theoretical model, the method can realize accurate monitoring and real-time prediction of the carbon footprint in the steel production process, and the accuracy and timeliness of the carbon footprint management are greatly improved. Secondly, by carrying out deep analysis and key feature extraction on the original production data set, the method not only can identify key influencing factors of carbon emission, but also can intelligently identify areas with obvious emission reduction potential, and provides scientific basis for making targeted emission reduction measures for iron and steel enterprises. In addition, based on a preset emission reduction fluctuation standard type set, the method can effectively manage carbon emission fluctuation in the production process, and the strategy of carbon footprint management is further refined through precise identification and classification of fluctuation areas, so that emission reduction measures are more vector. The method not only can effectively reduce the carbon emission in the steel production process, but also can optimize the production efficiency and the energy use structure, and promote the green and sustainable development of the production process. According to the method, a dynamic model adjustment and feedback optimization mechanism is introduced, so that a carbon footprint management strategy can be continuously adjusted and optimized according to real-time monitoring and prediction results, and the implementation effect of carbon emission reduction measures and the dynamic change of the production process are kept synchronous.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained based on these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of one embodiment of a method for carbon footprint management for steel production in accordance with an embodiment of the present application;
FIG. 2 is a schematic diagram of one embodiment of a carbon footprint management system for steel production in an embodiment of the present application.
Detailed Description
The embodiment of the application provides a carbon footprint management method and system for steel production. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, the following describes a specific flow of an embodiment of the present application, referring to fig. 1, and one embodiment of a carbon footprint management method for steel production in the embodiment of the present application includes:
Step S101, acquiring energy consumption data, greenhouse gas emission data and production line state data in a steel production environment in real time to obtain an original production data set;
It is to be understood that the execution subject of the present application may be a carbon footprint management system for steel production, and may also be a terminal or a server, which is not limited herein. The embodiment of the application is described by taking a server as an execution main body as an example.
Specifically, a plurality of sensors and data acquisition equipment are deployed at key positions of a production line, energy consumption, greenhouse gas emission and the running state of the production line are monitored and recorded, wherein the energy consumption monitoring equipment can be used for recording the consumption of electric power, fuel gas and water, the greenhouse gas emission sensors are specially used for measuring the emission levels of key greenhouse gases such as carbon dioxide, methane and the like, and the production line state monitoring equipment can track the running efficiency of the production line, the running state of the equipment and abnormal conditions in the production process in real time. By establishing an integrated data acquisition system, data acquired by the sensors and monitoring devices are collected and transmitted in real time to a central data processing center. At the data processing center, the collected raw data is initially processed and stored to form a raw production dataset.
Step S102, extracting key features of an original production data set to obtain a key feature set;
Specifically, the original data is subjected to data cleaning, noise and irrelevant information in the data, such as erroneous data input, repeated records, missing values and the like, are removed, and the accuracy and the reliability of subsequent analysis are ensured. After the data cleaning is completed, a cleaning data set is obtained. And carrying out time sequence analysis on the cleaning data set, and helping to understand the dynamic change characteristics of energy consumption and emission in the production process by analyzing the rule and trend of the data change along with time. The results of the time sequence analysis can reveal the relevant characteristics such as periodicity, trend and the like of the data, and provide important time dimension information for subsequent characteristic extraction. And carrying out frequency domain feature extraction on the cleaning data set based on the result of the time sequence analysis. By converting the data to the frequency domain, analyzing the data's performance at different frequencies, the main frequency components and frequency domain features in the data are identified. The frequency domain features reflect the periodicity and volatility in the data set, which helps to understand the fluctuation pattern of energy consumption and emissions during production. The frequency domain feature set and the cleaning data set are input to a pre-set long and short term memory network (LSTM) adapted to handle and predict long term dependencies and short term fluctuations in the time series data. Short-term fluctuation features in the data are extracted through LSTM network processing, and an initial feature set is obtained. And adopting a t-distribution random neighborhood embedding algorithm (t-SNE) to perform dimension reduction treatment on the initial feature set. the t-distributed random neighborhood embedding algorithm is an effective dimension reduction technique that maps high-dimensional data into low-dimensional space by maintaining the relative distance between the original data points, while maintaining as much as possible the local and global structures between the data points. And finally obtaining the key feature set through dimension reduction processing of the t-SNE algorithm.
Step S103, inputting the key feature set into a preset genetic programming algorithm to construct a carbon footprint prediction model, so as to obtain a target carbon footprint prediction model;
specifically, the genetic programming algorithm is subjected to parameter initialization, and the initialized parameters comprise algorithm crossover probability, variation probability and stop standard value. Crossover probability and mutation probability are probabilities for controlling gene crossover and mutation in the algorithm evolution process, and have direct influence on the searching efficiency and the diversity of the algorithm. The stop criteria are then used to determine when the algorithm is to terminate operation, typically based on a performance rating of the solution or the number of iterations. And carrying out coding processing on the key feature set, and converting the actual feature set into a coding feature set which can be processed by an algorithm. And inputting the coded feature set into an initialized genetic programming algorithm to carry out data evolution. Algorithms continually perform crossover and mutation operations on the encoded feature sets by modeling natural selection and genetic mechanisms, generating new program solutions. The evolution process is dynamic, and the quality of the solution is gradually optimized through continuous iteration, so that the optimal or near-optimal program solution is found. And evaluating the performance of each generated target program solution through a preset fitness function. The fitness function is a standard for measuring the quality of a program solution, and gives a performance evaluation value according to the accuracy and efficiency of the program solution on the carbon footprint prediction. Judging whether the algorithm finds a high-quality solution, wherein the higher the performance evaluation value is, the better the quality of the program solution is, and the better the performance of the prediction model is. And when the performance evaluation value of the found target program solution meets a preset stop standard value, stopping the operation of the algorithm, and constructing a final carbon footprint prediction model according to the target program solution. The model integrates the data processing and analysis capability obtained through genetic programming algorithm evolution, and can accurately predict the carbon trace amount in the steel production process.
And when the performance evaluation value meets the stop standard value, extracting relevant features of the target program solution to obtain a relevant feature set of the target program solution. And analyzing variables and functions with the largest contribution to the carbon footprint prediction in the program solution, and extracting the key elements as related feature sets. And generating an expression structure, parameter information and operation logic of the target program solution according to the related feature set. The expression structure reveals how the target program solution predicts the carbon footprint by combining different operations and functions, the parameter information describes the specific values of these operations, and the arithmetic logic describes the logical relationship and execution order among the operations. And constructing a carbon footprint mapping relation according to the expression structure of the target program solution, the parameter information of the target program solution and the operation logic of the target program solution. The mapping relation can predict the corresponding carbon trace amount according to the input production data. By comprehensively considering the expression structure, the parameter information and the operation logic, the model is ensured to not only accurately predict the carbon footprint, but also have certain generalization capability, and can cope with the change and uncertainty in the production process. And carrying out algorithm optimization parameter analysis on the initial genetic programming algorithm according to the target mapping relation, and finding out the optimal algorithm parameter configuration by analyzing the influence of different parameter settings on the performance of the prediction model. The method comprises the steps of adjusting the crossover probability, the mutation probability and the like, and improving the operation efficiency and the stability of the model while ensuring the accuracy of the model. And (3) adjusting and optimizing an initial genetic programming algorithm based on the algorithm optimization parameter set, and finally constructing a target carbon footprint prediction model.
Step S104, inputting an original production data set into a target carbon footprint prediction model to predict the carbon footprint quantity, so as to obtain a predicted carbon footprint quantity;
Specifically, the original production data set is input into the target carbon footprint prediction model to carry out L1 regularization treatment, so that the complexity of the data is reduced, overfitting is avoided, and the regularization treatment is carried out on the data, so that the generalization capability of the model is effectively improved, and regularized data is obtained. And carrying out nonlinear feature mapping on the regularized data. By introducing nonlinear transformation, mapping the data into a higher dimensional space, complex features and patterns hidden in the original data are obtained. And carrying out evolution trend analysis on the original production data set according to the nonlinear characteristics, and helping to identify the trend and mode of the data change along with time. The evolution trend data reflects the dynamic change of the carbon footprint in the production process and provides an important basis for predicting the future carbon footprint. And carrying out data enhancement processing on the original production data set based on the evolution trend data. Data enhancement increases the diversity and richness of data sets by introducing additional information or modifying existing data. And inputting the enhanced data set into a target carbon footprint prediction model for stack parameter analysis to obtain a plurality of stack parameter sets. Each stack parameter set represents one possible model configuration, and the optimal model parameter combination is found through comprehensive analysis of the configuration, so that the prediction accuracy is improved. And carrying out feature recognition on each stacking parameter set to generate a stacking feature set. Carbon footprint distribution analysis is performed based on the stacked feature set, and the carbon footprint is predicted by analyzing the distribution characteristics of the data by utilizing the previously identified key features. And predicting based on the data characteristics, and considering the potential distribution rule of the carbon trace amount to finally obtain the predicted carbon trace amount.
And step 105, inputting the predicted carbon footprint quantity into a preset self-adaptive resonance theoretical model to identify the emission reduction potential areas, obtaining a plurality of emission reduction potential areas, and managing the carbon footprint of the steel production environment based on the emission reduction potential areas.
Specifically, the predicted carbon footprint is input into an adaptive resonance theoretical model to identify the footprint fluctuation point. The carbon footprint time series data is analyzed to find those points in time that exhibit significant changes, which are identified as carbon footprint fluctuation points. And respectively carrying out footprint fluctuation amplitude identification on each footprint fluctuation point. The fluctuation amplitude data is calculated by comparing the change of the trace amount of the carbon before and after the fluctuation point, and the data reflects the change degree of each fluctuation point. Then, the fluctuation category identification is performed for the fluctuation point based on the fluctuation amplitude data of each fluctuation point. The fluctuation points are classified according to their fluctuation characteristics, such as periodic fluctuation, sudden increase or sudden decrease, etc. And the model carries out fluctuation region segmentation on the predicted carbon footprint according to the identified fluctuation category. The overall carbon footprint time series is divided into a plurality of regions, and the carbon footprint variation characteristics in each region are similar, which helps to clearly identify key emission reduction potential regions in the production process. And the model carries out emission reduction potential area identification on the fluctuation area based on a preset emission reduction fluctuation standard type set. Each fluctuation area is matched with a plurality of emission reduction fluctuation standards, and which areas have high emission reduction potential are determined. Once multiple emission reduction potential areas are identified, targeted carbon footprint management of the steel production environment is performed based on these areas, such as by adjusting production flows, optimizing energy usage, or implementing new emission reduction techniques.
Further, the emission reduction fluctuation standard type set is subjected to fluctuation mode identification. Different types of fluctuation modes, such as periodic fluctuation, sudden increase and decrease and the like, are analyzed and understood to obtain a fluctuation mode set. Each fluctuation zone is subjected to fluctuation pattern matching based on the fluctuation pattern set. By comparing the carbon footprint fluctuation characteristics within the fluctuation zones to a predefined set of fluctuation patterns, a target fluctuation pattern is identified that best matches each zone. And carrying out emission reduction potential area identification according to the target fluctuation mode of each fluctuation area, wherein all areas identified to have a specific fluctuation mode are regarded as initial emission reduction potential areas. These initial emission reduction potential areas represent key links in the steel production process that may be optimized or improved to achieve carbon emission reduction. And carrying out emission reduction analysis on each initial emission reduction potential area, and estimating the emission reduction possible in each area by adopting proper emission reduction measures. Analysis takes into account a number of factors such as existing production technology, energy efficiency, and viable improvements, etc., quantifying the emission reduction potential of each zone. Region screening is performed based on the emission reduction amount of each initial emission reduction potential region, and those regions with the greatest emission reduction potential are determined. By comparing the emission reduction of different areas, several areas with the largest emission reduction potential are selected, and the areas become finally determined emission reduction potential areas. Based on these areas, targeted carbon footprint management measures are formulated, such as adjusting production flows, adopting cleaner energy sources or introducing efficient emission reduction technologies, etc., so that effective carbon footprint management is realized in the steel production environment.
According to the embodiment of the application, through collecting the energy consumption data, the greenhouse gas emission data and the production line data in the steel production environment in real time and combining a genetic programming algorithm and the self-adaptive resonance theoretical model, the method can realize accurate monitoring and real-time prediction of the carbon footprint in the steel production process, and the accuracy and timeliness of the carbon footprint management are greatly improved. Secondly, by carrying out deep analysis and key feature extraction on the original production data set, the method not only can identify key influencing factors of carbon emission, but also can intelligently identify areas with obvious emission reduction potential, and provides scientific basis for making targeted emission reduction measures for iron and steel enterprises. In addition, based on a preset emission reduction fluctuation standard type set, the method can effectively manage carbon emission fluctuation in the production process, and the strategy of carbon footprint management is further refined through precise identification and classification of fluctuation areas, so that emission reduction measures are more vector. The method not only can effectively reduce the carbon emission in the steel production process, but also can optimize the production efficiency and the energy use structure, and promote the green and sustainable development of the production process. According to the method, a dynamic model adjustment and feedback optimization mechanism is introduced, so that a carbon footprint management strategy can be continuously adjusted and optimized according to real-time monitoring and prediction results, and the implementation effect of carbon emission reduction measures and the dynamic change of the production process are kept synchronous.
In a specific embodiment, the process of executing step S102 may specifically include the following steps:
(1) Performing data cleaning on the original production data set to obtain a cleaning data set;
(2) Performing time sequence analysis on the cleaning data set to obtain a time sequence analysis result;
(3) Based on a time sequence analysis result, carrying out frequency domain feature extraction on the cleaning data set to obtain a frequency domain feature set;
(4) Inputting the frequency domain feature set and the cleaning data set into a preset long-short-time memory network to extract short-term fluctuation features, so as to obtain an initial feature set;
(5) And performing dimension reduction processing on the initial feature set through a t-distribution random neighborhood embedding algorithm to obtain a key feature set.
Specifically, the data cleaning is performed on the original production data set, including removing invalid, erroneous or incomplete data records, such as energy consumption data, greenhouse gas emission data and production line data recorded in the steel production process, and abnormal values may exist due to equipment failure or human error. And (3) carrying out time sequence analysis on the cleaning data set, and identifying the rule and trend of the data change along with time. Time series analysis can reveal periodic changes in energy consumption and emissions during production, seasonal effects, and long-term trends. For example, by time series analysis of daily greenhouse gas emissions over the past year, a significant increase in emissions during a particular season or particular production phase may be found. And carrying out frequency domain feature extraction on the cleaning data set based on the time sequence analysis result. The frequency domain analysis converts the perspective of the time series data, revealing the periodicity and ripple characteristics of the data from a frequency perspective. The time series data is converted into a frequency domain representation by applying fourier transform or other frequency domain analysis methods, and the main frequency components affecting the trace amount of change of carbon are identified. For example, if the amplitude of a certain frequency component is found to be particularly large, this may indicate that a periodically varying factor corresponding to that frequency exists during production. And inputting the extracted frequency domain feature set and the cleaned data set into a preset long-short-time memory network to extract short-term fluctuation features. The long-time and short-time memory network is a deep learning model suitable for processing time series data, and can capture long-term dependency and short-term fluctuation in the data. And (3) identifying key factors and modes influencing short-term changes of the carbon footprint by training a long-and-short-term memory network model, and obtaining an initial feature set. And performing dimension reduction processing on the initial feature set by using a t-distribution random neighborhood embedding algorithm. the t-distribution random neighborhood embedding algorithm is an efficient machine learning algorithm and is specially used for visualization and dimension reduction of high-dimensional data, mapping is carried out on the low-dimensional space by maintaining the relative distance between original high-dimensional data points, and the internal structure and mode of the data are effectively revealed. After applying the t-SNE to the initial feature set, a more refined and representative key feature set is extracted from the high-dimensional complex data.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
(1) Parameter initialization is carried out on the genetic programming algorithm, wherein the initialization parameters comprise: algorithm crossover probability, variation probability and stopping standard value;
(2) Performing coding processing on the key feature set to obtain a coding feature set corresponding to the key feature set;
(3) Inputting the coding feature set into an initialized initial genetic programming algorithm to carry out data evolution to obtain a target program solution;
(4) Performing performance evaluation on the target program solution through a preset fitness function to obtain a performance evaluation value of the target program solution;
(5) And when the performance evaluation value meets the stop standard value, constructing a carbon footprint prediction model of the initial genetic programming algorithm according to the target program solution to obtain a target carbon footprint prediction model.
Specifically, the genetic programming algorithm is subjected to parameter initialization, and the initialization parameters mainly comprise the crossover probability, the variation probability and the stop standard value of the algorithm. The crossover probability determines the frequency with which the algorithm selects two programs (called parents) in each generation to crossover (i.e., recombine their codes), a process that mimics chromosomal crossover in biological genetics to produce new offspring. The probability of variation then determines the frequency of a certain part (e.g., operator or value) of the randomly altered program in each generation, mimicking mutations in biological inheritance. The stopping criterion is typically a predefined performance criterion, such as reaching a certain fitness score or stopping the algorithm after a certain number of iterations. And (3) carrying out coding treatment on the key feature set collected in the steel production process to obtain a coding feature set suitable for genetic programming algorithm treatment. Each feature is converted into a form that the algorithm can understand and operate, for example, converting greenhouse gas emissions and energy consumption data into a series of symbols or numerical codes. Inputting the coded feature set into an initialized genetic programming algorithm, and carrying out a data evolution process. The algorithm simulates the process of natural selection, gradually improving the program solution by repeated crossover, mutation and selection operations to better adapt it to the environment (i.e., more accurately predict the carbon footprint). For example, an algorithm may initially generate a series of random program solutions, each attempting to predict a carbon footprint using a feature set in a different manner. By crossover and mutation, these program solutions are continually improved and optimized. To evaluate the performance of these program solutions, they are evaluated by a preset fitness function. The fitness function calculates a performance evaluation value based on the difference between the predicted result of the program solution and the actual carbon footprint data. For example, a program solution may have a higher fitness score if it can accurately predict changes in the carbon footprint caused by a particular production activity. And stopping iteration of the algorithm when the performance evaluation value reaches or exceeds a preset stopping standard value, wherein the obtained optimal program solution is used for constructing a final carbon footprint prediction model. The model is based on a program solution optimized by a natural selection mechanism through a genetic programming algorithm, and key data characteristics in the steel production process can be effectively utilized to predict the carbon footprint. For example, if during steel production it is found that a particular production parameter variation can lead to significant fluctuations in the carbon footprint, as an increase in production speed can lead to an increase in energy consumption and greenhouse gas emissions, this relationship can be captured and simulated by a procedural solution in a genetic programming algorithm. By optimizing the program solutions, the finally constructed carbon footprint prediction model can accurately predict the carbon footprint quantity under different production parameter settings.
In a specific embodiment, the process of performing the carbon footprint prediction model building step for the initial genetic programming algorithm according to the target program solution may specifically include the following steps:
(1) When the performance evaluation value meets a stop standard value, extracting relevant features of the target program solution to obtain a relevant feature set of the target program solution;
(2) Generating an expression structure of the target program solution, parameter information of the target program solution and operation logic of the target program solution according to the related feature set;
(3) Constructing a carbon footprint mapping relation according to the expression structure of the target program solution, the parameter information of the target program solution and the operation logic of the target program solution to obtain a target mapping relation;
(4) Carrying out algorithm optimization parameter analysis on the initial genetic programming algorithm according to the target mapping relation to obtain an algorithm optimization parameter set;
(5) And (3) constructing a carbon footprint prediction model for the initial genetic programming algorithm based on the algorithm optimization parameter set to obtain a target carbon footprint prediction model.
Specifically, when the performance evaluation value satisfies the stop standard value, the relevant feature extraction is performed on the target program solution. In genetic programming, the target program solution is typically composed of a series of operations and functions that directly act on the input data features to predict the carbon footprint. By analyzing the constitution of these operations and functions, the data features that have the greatest influence on the prediction result, i.e., the relevant feature set of the target program solution, are identified. Based on the extracted relevant feature set, an expression structure, parameter information and operation logic of the target program solution are generated. The expression structure describes how the different features and functions are organized and nested, the parameter information describes the specific parameter settings of each function, and the operational logic describes how these operations and functions work in concert to achieve carbon footprint predictions. And constructing a carbon footprint mapping relation according to the expression structure, the parameter information and the operation logic of the target program solution. The mapping relationship can predict the carbon trace amount based on key data characteristics in the production process. And carrying out algorithm optimization parameter analysis on the initial genetic programming algorithm based on the constructed target mapping relation. The efficiency and accuracy of the algorithm in finding the optimal carbon footprint prediction model are improved by adjusting parameter settings (such as crossover probability, mutation probability and the like) of the genetic programming algorithm. By testing different parameter configurations, the algorithm parameter set most favorable for generating the high-performance carbon footprint prediction model is identified. Based on the algorithm optimization parameter set, the initial genetic programming algorithm is adjusted, and the construction process of the carbon footprint prediction model is conducted again. The optimized genetic programming algorithm searches and optimizes the program solution more efficiently, and finally a target carbon footprint prediction model with higher accuracy and stronger generalization capability is obtained.
In a specific embodiment, the process of executing step S104 may specifically include the following steps:
(1) Inputting the original production data set into a target carbon footprint prediction model for L1 regularization treatment to obtain regularized data;
(2) Carrying out nonlinear feature mapping on the regularized data to obtain nonlinear features;
(3) According to the nonlinear characteristics, carrying out evolution trend analysis on the original production data set to obtain evolution trend data;
(4) Carrying out data enhancement processing on the original production data set according to the evolution trend data to obtain an enhanced data set;
(5) Inputting the enhanced data set into a target carbon footprint prediction model for stacking parameter analysis to obtain a plurality of stacking parameter sets;
(6) Respectively carrying out stacking parameter feature recognition on each stacking parameter set to obtain stacking feature sets;
(7) Performing carbon footprint distribution analysis according to the stacking feature set to obtain carbon footprint distribution data;
(8) And predicting the carbon footprint according to the carbon footprint distribution data to obtain predicted carbon footprint.
Specifically, the original production dataset is input into a target carbon footprint prediction model for L1 regularization treatment. L1 regularization is a technique commonly used for optimization and data preprocessing that reduces the complexity of the data, prevents model overfitting, and at the same time encourages the model to focus on the most important features. By applying L1 regularization to each feature, those features that contribute less to the predicted target are effectively removed, resulting in a more compact and focused regularized dataset. And carrying out nonlinear feature mapping on the regularized data. The nonlinear feature mapping is a process of converting original features into complex structures capable of better reflecting data, and by introducing nonlinear transformation, a model can capture complex modes and relations which cannot be identified by a linear method. The nonlinear mapping is implemented using an activation function such as in kernel methods or deep learning to obtain a nonlinear feature set that reveals the deep structure of the original data. And carrying out evolution trend analysis on the original production data set according to the nonlinear characteristics, and identifying the trend of the data change along with time. By analyzing the pattern of the non-linear characteristic over time, data describing the evolution trend of the data is obtained, for example, by trend analysis, a trend may be found in which the amount of carbon emissions caused by the production activity in a particular season increases significantly. And based on evolution trend data, carrying out data enhancement processing on the original production data set, and improving generalization capability and prediction accuracy of the model. Data enhancement can be achieved by introducing noise, interpolating the data, or generating new data points using knowledge based on trend analysis. The enhanced data set contains all the information of the original data, and additional knowledge obtained through trend analysis is added, so that the model can learn more information about the data evolution in the training process. And inputting the enhanced data set into a target carbon footprint prediction model for stack parameter analysis, optimizing model parameters, and improving the prediction accuracy. By analyzing the behavior of the model under different parameter configurations, a plurality of stacked parameter sets are obtained, each set representing a set of potential model configurations. For example, different stacking levels or different regularization strengths may have an impact on the predictive performance of the model. And carrying out feature recognition on each stacking parameter set, and finding out which parameter configurations can most effectively capture key modes in the data to form a stacking feature set. By analyzing the behavior of the different parameter sets when processing the enhancement data set, it is identified which parameter settings are most significant to the model performance improvement. And carrying out carbon footprint distribution analysis based on the stacking feature set, and refining the prediction capability of the model.
In a specific embodiment, the process of executing step S105 may specifically include the following steps:
(1) Inputting the predicted carbon footprint into a self-adaptive resonance theoretical model to identify footprint fluctuation points, so as to obtain a plurality of footprint fluctuation points;
(2) Respectively carrying out footprint fluctuation amplitude identification on each footprint fluctuation point to obtain fluctuation amplitude data of each footprint fluctuation point;
(3) Based on the fluctuation amplitude data of each footprint fluctuation point, respectively carrying out fluctuation category identification on each footprint fluctuation point to obtain the fluctuation category of each footprint fluctuation point;
(4) Based on the fluctuation category of each footprint fluctuation point, carrying out fluctuation area segmentation on the predicted carbon footprint to obtain a plurality of fluctuation areas;
(5) And carrying out emission reduction potential area identification on the plurality of fluctuation areas based on a preset emission reduction fluctuation standard type set to obtain a plurality of emission reduction potential areas, and carrying out carbon footprint management on the steel production environment based on the plurality of emission reduction potential areas.
Specifically, the predicted carbon footprint is input into an adaptive resonance theoretical model to identify the footprint fluctuation point. By analyzing the change of the carbon footprint over time, the model can accurately identify the specific location where the fluctuation occurs, e.g., a sudden increase in energy consumption during production results in a sharp rise in the carbon footprint. And carrying out fluctuation amplitude identification on each identified footprint fluctuation point, and quantifying the fluctuation intensity, namely the degree of the change of the carbon footprint before and after the fluctuation point. For example, if the identified carbon footprint fluctuation amplitude is particularly large at a particular production stage, this may indicate that there is significant space for optimization of the production efficiency at that stage. And carrying out category identification on the fluctuation points based on the fluctuation amplitude data of each fluctuation point. By analyzing the pattern and characteristics of the fluctuations, the fluctuation points are classified into different fluctuation categories, such as periodic fluctuations, sporadic fluctuations, and the like. For example, periodic fluctuations may be related to the production cycle, while sporadic fluctuations may be caused by specific abnormal events. And according to the fluctuation category of each fluctuation point, carrying out fluctuation area segmentation on the predicted carbon footprint. The carbon footprint time series is divided into several regions, with the fluctuations within each region having the same or similar fluctuation categories. The division enables the management of the fluctuation of the carbon footprint to be more targeted, and corresponding management measures can be formulated according to different fluctuation categories. For example, for periodic fluctuation zones, energy consumption and carbon emissions can be smoothed by adjusting the production schedule. And carrying out emission reduction potential area identification on each fluctuation area based on a preset emission reduction fluctuation standard type set. The characteristics of the fluctuation zones are matched to emission abatement fluctuation criteria to determine which zones have the greatest emission abatement potential. By identifying the potential area for emission reduction, emission reduction measures such as optimizing production process, improving energy utilization efficiency or adopting clean energy are targeted, so that the carbon footprint is reduced to the greatest extent while the production efficiency is ensured.
In a specific embodiment, the process of performing the emission reduction potential region identification step on the plurality of fluctuation regions may specifically include the following steps:
(1) Carrying out fluctuation mode identification on the emission reduction fluctuation standard type set to obtain a fluctuation mode set;
(2) Carrying out fluctuation pattern matching on each fluctuation area based on the fluctuation pattern set to obtain a target fluctuation pattern of each fluctuation area;
(3) Carrying out emission reduction potential area identification on a plurality of fluctuation areas through a target fluctuation mode of each fluctuation area to obtain a plurality of initial emission reduction potential areas;
(4) Respectively carrying out emission reduction analysis on each initial emission reduction potential area to obtain emission reduction of each initial emission reduction potential area;
(5) And carrying out area screening on the plurality of initial emission reduction potential areas through emission reduction of each initial emission reduction potential area to obtain a plurality of emission reduction potential areas, and carrying out carbon footprint management on the steel production environment based on the plurality of emission reduction potential areas.
Specifically, the emission reduction fluctuation standard type set is subjected to fluctuation mode identification, a fluctuation mode set is obtained, the set comprises all modes which can influence the change of the carbon emission, the modes can comprise fluctuation caused from periodical change to emergency, and each mode corresponds to a specific carbon emission change rule. For example, the periodic pattern may correspond to natural fluctuations in production activity, such as increases or decreases in energy consumption due to seasonal production demand changes, while the incident pattern may reflect abnormal emissions due to equipment failure or production accidents. And carrying out fluctuation pattern matching on each identified fluctuation area based on the fluctuation pattern set, and determining main reasons and characteristics of carbon emission fluctuation in the area. By comparing the carbon emission fluctuation data in the region with the patterns in the fluctuation pattern set, it is determined which pattern is the most capable of interpreting the fluctuation situation in the region. And carrying out emission reduction potential area identification on a plurality of fluctuation areas through the target fluctuation mode of each fluctuation area. The fluctuation zones that have been matched with the fluctuation pattern are evaluated to determine which zones exhibit a higher emission reduction potential due to having a particular fluctuation pattern. The initial emission abatement potential zone represents a zone where significant carbon emission abatement is possible by taking specific abatement measures. And carrying out emission reduction analysis on the initial emission reduction potential areas, and quantifying the reduction amount of carbon emission which is possible to be achieved after each area is subjected to emission reduction measures. Analysis considers the potential effects of various emission abatement strategies, such as employing more efficient production techniques, improving energy management, or introducing clean energy. By emission reduction analysis, the emission reduction target and potential of each region are defined, and a quantitative basis is provided for subsequent emission reduction measures. Comprehensive evaluation and region screening are performed based on the emission reduction of each region, and those regions with the greatest emission reduction potential are determined and become the focus of implementing carbon footprint management. In this embodiment, the method includes evaluating emission reduction potential in different production lines or processes, and considering implementation cost and feasibility of different emission reduction measures.
The method for managing carbon footprint of steel production in the embodiment of the present application is described above, and the following describes a carbon footprint management system for steel production in the embodiment of the present application, referring to fig. 2, one embodiment of the carbon footprint management system for steel production in the embodiment of the present application includes:
The acquisition module 201 is used for acquiring energy consumption data, greenhouse gas emission data and production line state data in the steel production environment in real time to obtain an original production data set;
an extraction module 202, configured to extract key features from the original production dataset to obtain a key feature set;
the construction module 203 is configured to input the key feature set into a preset genetic programming algorithm to perform carbon footprint prediction model construction, so as to obtain a target carbon footprint prediction model;
the prediction module 204 is configured to input the raw production dataset into the target carbon footprint prediction model to perform carbon footprint prediction, so as to obtain a predicted carbon footprint;
The identification module 205 is configured to input the predicted carbon footprint amount into a preset adaptive resonance theoretical model to identify an emission reduction potential area, obtain a plurality of emission reduction potential areas, and manage carbon footprint of the steel production environment based on the plurality of emission reduction potential areas.
Through the cooperation of the components, the method can realize accurate monitoring and real-time prediction of the carbon footprint in the steel production process by collecting the energy consumption data, the greenhouse gas emission data and the production line data in the steel production environment in real time and combining a genetic programming algorithm and a self-adaptive resonance theoretical model, and the accuracy and timeliness of the carbon footprint management are greatly improved. Secondly, by carrying out deep analysis and key feature extraction on the original production data set, the method not only can identify key influencing factors of carbon emission, but also can intelligently identify areas with obvious emission reduction potential, and provides scientific basis for making targeted emission reduction measures for iron and steel enterprises. In addition, based on a preset emission reduction fluctuation standard type set, the method can effectively manage carbon emission fluctuation in the production process, and the strategy of carbon footprint management is further refined through precise identification and classification of fluctuation areas, so that emission reduction measures are more vector. The method not only can effectively reduce the carbon emission in the steel production process, but also can optimize the production efficiency and the energy use structure, and promote the green and sustainable development of the production process. According to the method, a dynamic model adjustment and feedback optimization mechanism is introduced, so that a carbon footprint management strategy can be continuously adjusted and optimized according to real-time monitoring and prediction results, and the implementation effect of carbon emission reduction measures and the dynamic change of the production process are kept synchronous.
The application also provides a carbon footprint management device for steel production, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, cause the processor to execute the steps of the carbon footprint management method for steel production in the above embodiments.
The present application also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and may also be a volatile computer readable storage medium, in which instructions are stored which, when executed on a computer, cause the computer to perform the steps of the carbon footprint management method of steel production.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, systems and units may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random acceS memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (10)
1. The carbon footprint management method for steel production is characterized by comprising the following steps of:
acquiring energy consumption data, greenhouse gas emission data and production line state data in the steel production environment in real time to obtain an original production data set;
Extracting key features from the original production data set to obtain a key feature set;
Inputting the key feature set into a preset genetic programming algorithm to construct a carbon footprint prediction model, so as to obtain a target carbon footprint prediction model;
inputting the original production data set into the target carbon footprint prediction model to predict the carbon footprint quantity, so as to obtain a predicted carbon footprint quantity;
inputting the predicted carbon footprint quantity into a preset self-adaptive resonance theoretical model to identify emission reduction potential areas, obtaining a plurality of emission reduction potential areas, and performing carbon footprint management on the steel production environment based on the emission reduction potential areas.
2. The method for carbon footprint management for steel production according to claim 1, wherein the performing key feature extraction on the raw production dataset to obtain a key feature set comprises:
performing data cleaning on the original production data set to obtain a cleaning data set;
performing time sequence analysis on the cleaning data set to obtain a time sequence analysis result;
Based on the time sequence analysis result, extracting frequency domain features of the cleaning data set to obtain a frequency domain feature set;
Inputting the frequency domain feature set and the cleaning data set into a preset long-short-time memory network to extract short-term fluctuation features, so as to obtain an initial feature set;
And performing dimension reduction processing on the initial feature set through a t-distribution random neighborhood embedding algorithm to obtain the key feature set.
3. The method for carbon footprint management in steel production according to claim 1, wherein inputting the key feature set into a preset initial genetic programming algorithm for carbon footprint prediction model construction to obtain a target carbon footprint prediction model comprises:
and initializing parameters of the genetic programming algorithm, wherein the initialization parameters comprise: algorithm crossover probability, variation probability and stopping standard value;
performing coding processing on the key feature set to obtain a coding feature set corresponding to the key feature set;
inputting the coding feature set into an initialized initial genetic programming algorithm to carry out data evolution to obtain a target program solution;
Performing performance evaluation on the target program solution through a preset fitness function to obtain a performance evaluation value of the target program solution;
and when the performance evaluation value meets the stopping standard value, constructing a carbon footprint prediction model of the initial genetic programming algorithm according to the target program solution to obtain a target carbon footprint prediction model.
4. The method for carbon footprint management in steel production according to claim 3, wherein when the performance evaluation value meets the stop standard value, constructing a carbon footprint prediction model of the initial genetic programming algorithm according to the target program solution to obtain a target carbon footprint prediction model, comprising:
When the performance evaluation value meets the stop standard value, extracting relevant features of the target program solution to obtain a relevant feature set of the target program solution;
generating an expression structure of the target program solution, parameter information of the target program solution and operation logic of the target program solution according to the related feature set;
Constructing a carbon footprint mapping relation according to the expression structure of the target program solution, the parameter information of the target program solution and the operation logic of the target program solution to obtain a target mapping relation;
carrying out algorithm optimization parameter analysis on the initial genetic programming algorithm according to the target mapping relation to obtain an algorithm optimization parameter set;
and carrying out carbon footprint prediction model construction on the initial genetic programming algorithm based on the algorithm optimization parameter set to obtain the target carbon footprint prediction model.
5. The method of carbon footprint management for steel production of claim 1, wherein said inputting said raw production dataset into said target carbon footprint prediction model for carbon footprint prediction, resulting in a predicted carbon footprint, comprises:
inputting the original production data set into the target carbon footprint prediction model for L1 regularization treatment to obtain regularized data;
Performing nonlinear feature mapping on the regularized data to obtain nonlinear features;
according to the nonlinear characteristics, carrying out evolution trend analysis on the original production data set to obtain evolution trend data;
Carrying out data enhancement processing on the original production data set according to the evolution trend data to obtain an enhanced data set;
Inputting the enhanced data set into the target carbon footprint prediction model for stacking parameter analysis to obtain a plurality of stacking parameter sets;
Respectively carrying out stacking parameter feature recognition on each stacking parameter set to obtain stacking feature sets;
Performing carbon footprint distribution analysis according to the stacking feature set to obtain carbon footprint distribution data;
and predicting the carbon footprint according to the carbon footprint distribution data to obtain the predicted carbon footprint.
6. The method for carbon footprint management for steel production according to claim 1, wherein the inputting the predicted carbon footprint into a preset adaptive resonance theoretical model for emission reduction potential region identification to obtain a plurality of emission reduction potential regions, and performing carbon footprint management on the steel production environment based on the plurality of emission reduction potential regions comprises:
inputting the predicted carbon footprint into the self-adaptive resonance theoretical model to identify footprint fluctuation points, so as to obtain a plurality of footprint fluctuation points;
respectively carrying out footprint fluctuation amplitude identification on each footprint fluctuation point to obtain fluctuation amplitude data of each footprint fluctuation point;
based on the fluctuation amplitude data of each footprint fluctuation point, respectively carrying out fluctuation category identification on each footprint fluctuation point to obtain the fluctuation category of each footprint fluctuation point;
Based on the fluctuation category of each footprint fluctuation point, carrying out fluctuation area segmentation on the predicted carbon footprint to obtain a plurality of fluctuation areas;
And carrying out emission reduction potential area identification on a plurality of fluctuation areas based on a preset emission reduction fluctuation standard type set to obtain a plurality of emission reduction potential areas, and carrying out carbon footprint management on the steel production environment based on the emission reduction potential areas.
7. The method for carbon footprint management for steel production according to claim 6, wherein the performing emission reduction potential area identification on the plurality of fluctuation areas based on the preset emission reduction fluctuation standard type set to obtain a plurality of emission reduction potential areas, performing carbon footprint management on the steel production environment based on the plurality of emission reduction potential areas, comprises:
Carrying out fluctuation mode identification on the emission reduction fluctuation standard type set to obtain a fluctuation mode set;
Carrying out fluctuation pattern matching on each fluctuation area based on the fluctuation pattern set to obtain a target fluctuation pattern of each fluctuation area;
Carrying out emission reduction potential area identification on a plurality of fluctuation areas through a target fluctuation mode of each fluctuation area to obtain a plurality of initial emission reduction potential areas;
Respectively carrying out emission reduction analysis on each initial emission reduction potential area to obtain emission reduction of each initial emission reduction potential area;
And carrying out area screening on the initial emission reduction potential areas through the emission reduction amount of each initial emission reduction potential area to obtain a plurality of emission reduction potential areas, and carrying out carbon footprint management on the steel production environment based on the emission reduction potential areas.
8. A carbon footprint management system for steel production, the carbon footprint management system for steel production comprising:
The acquisition module is used for acquiring energy consumption data, greenhouse gas emission data and production line state data in the steel production environment in real time to obtain an original production data set;
the extraction module is used for extracting key features of the original production data set to obtain a key feature set;
The construction module is used for inputting the key feature set into a preset genetic programming algorithm to construct a carbon footprint prediction model, so as to obtain a target carbon footprint prediction model;
The prediction module is used for inputting the original production data set into the target carbon footprint prediction model to predict the carbon footprint quantity so as to obtain a predicted carbon footprint quantity;
the identification module is used for inputting the predicted carbon footprint quantity into a preset self-adaptive resonance theoretical model to identify the emission reduction potential areas, so that a plurality of emission reduction potential areas are obtained, and carbon footprint management is performed on the steel production environment based on the emission reduction potential areas.
9. A carbon footprint management apparatus for steel production, characterized in that the carbon footprint management apparatus for steel production comprises: a memory and at least one processor, the memory having instructions stored therein;
The at least one processor invoking the instructions in the memory to cause the steel production carbon footprint management device to perform the steel production carbon footprint management method of any one of claims 1-7.
10. A computer readable storage medium having instructions stored thereon, which when executed by a processor, implement the carbon footprint management method of steel production of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410550914.9A CN118134210B (en) | 2024-05-07 | 2024-05-07 | Carbon footprint management method and system for steel production |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410550914.9A CN118134210B (en) | 2024-05-07 | 2024-05-07 | Carbon footprint management method and system for steel production |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118134210A true CN118134210A (en) | 2024-06-04 |
| CN118134210B CN118134210B (en) | 2024-07-05 |
Family
ID=91238039
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410550914.9A Active CN118134210B (en) | 2024-05-07 | 2024-05-07 | Carbon footprint management method and system for steel production |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118134210B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118644049A (en) * | 2024-08-13 | 2024-09-13 | 山东浪潮智慧能源科技有限公司 | Production line management and control system and method, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230015077A1 (en) * | 2021-07-19 | 2023-01-19 | Ford Global Technologies, Llc | Real-time carbon footprint estimation |
| CN116763299A (en) * | 2023-06-16 | 2023-09-19 | 西北工业大学 | Fall detection system and method based on wearable sensor |
| CN117522654A (en) * | 2024-01-08 | 2024-02-06 | 珠江水利委员会珠江水利科学研究院 | Pollution and carbon reduction synergy analysis method based on gray water footprint |
| CN117541272A (en) * | 2024-01-09 | 2024-02-09 | 小象飞羊(北京)科技有限公司 | Method, system, equipment and storage medium for determining digital engineering carbon emission data |
-
2024
- 2024-05-07 CN CN202410550914.9A patent/CN118134210B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230015077A1 (en) * | 2021-07-19 | 2023-01-19 | Ford Global Technologies, Llc | Real-time carbon footprint estimation |
| CN116763299A (en) * | 2023-06-16 | 2023-09-19 | 西北工业大学 | Fall detection system and method based on wearable sensor |
| CN117522654A (en) * | 2024-01-08 | 2024-02-06 | 珠江水利委员会珠江水利科学研究院 | Pollution and carbon reduction synergy analysis method based on gray water footprint |
| CN117541272A (en) * | 2024-01-09 | 2024-02-09 | 小象飞羊(北京)科技有限公司 | Method, system, equipment and storage medium for determining digital engineering carbon emission data |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118644049A (en) * | 2024-08-13 | 2024-09-13 | 山东浪潮智慧能源科技有限公司 | Production line management and control system and method, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118134210B (en) | 2024-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tan et al. | Ultra-short-term industrial power demand forecasting using LSTM based hybrid ensemble learning | |
| Singh et al. | An efficient time series forecasting model based on fuzzy time series | |
| CN117670066B (en) | Questor management method, system, equipment and storage medium based on intelligent decision | |
| CN118134210B (en) | Carbon footprint management method and system for steel production | |
| CN111553543A (en) | Power load prediction method based on TPA-Seq2Seq and related assembly | |
| CN117725537A (en) | Real-time metering data processing platform | |
| Ceylan et al. | Estimation of healthcare expenditure per capita of Turkey using artificial intelligence techniques with genetic algorithm‐based feature selection | |
| Lin et al. | Revenue prediction for integrated renewable energy and energy storage system using machine learning techniques | |
| Kaboli et al. | An expression-driven approach for long-term electric power consumption forecasting | |
| CN105844353A (en) | Aquatic product price prediction method and device | |
| Wang et al. | Causal carbon price interval prediction using lower upper bound estimation combined with asymmetric multi-objective evolutionary algorithm and long short-term memory | |
| CN118245733B (en) | Data preprocessing method based on operation error monitoring model and intelligent ammeter | |
| CN117977028B (en) | Intelligent charging method and system based on vehicle battery monitoring | |
| CN117422167A (en) | Electric power carbon emission predictive analysis method based on tree model | |
| CN116108963A (en) | Electric power carbon emission prediction method and equipment based on integrated learning module | |
| CN116561569A (en) | Industrial power load identification method based on EO feature selection and AdaBoost algorithm | |
| CN116701846A (en) | Hydropower station dispatching operation data cleaning method based on unsupervised learning | |
| CN116780530A (en) | Large-scale new energy access power grid weak link identification prediction method and system | |
| CN114330485A (en) | Power grid investment capacity prediction method based on PLS-SVM-GA algorithm | |
| CN117494071B (en) | Life prediction method based on motor rotation speed monitoring and related device | |
| CN117439146B (en) | Data analysis control method and system for charging pile | |
| CN117575258B (en) | Coal-fired power plant deep peak shaving method and device considering wastewater treatment | |
| CN117807114B (en) | Logistics information intelligent retrieval method, system, equipment and storage medium | |
| Kourentzes et al. | Advances in forecasting with artificial neural networks | |
| CN118332765A (en) | Multi-time scale electric power transaction market simulation method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |