CN115014821A - Method and device for detecting abnormality of working machine, and working machine - Google Patents
Method and device for detecting abnormality of working machine, and working machine Download PDFInfo
- Publication number
- CN115014821A CN115014821A CN202210616307.9A CN202210616307A CN115014821A CN 115014821 A CN115014821 A CN 115014821A CN 202210616307 A CN202210616307 A CN 202210616307A CN 115014821 A CN115014821 A CN 115014821A
- Authority
- CN
- China
- Prior art keywords
- model
- operation data
- abnormality detection
- data
- candidate
- 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.)
- Pending
Links
- 230000005856 abnormality Effects 0.000 title claims abstract description 98
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000001514 detection method Methods 0.000 claims abstract description 129
- 230000002159 abnormal effect Effects 0.000 claims abstract description 24
- 238000012549 training Methods 0.000 claims abstract description 19
- 238000012360 testing method Methods 0.000 claims description 49
- 238000004590 computer program Methods 0.000 claims description 12
- 238000004140 cleaning Methods 0.000 claims description 7
- 230000007613 environmental effect Effects 0.000 claims description 5
- 238000004891 communication Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000002547 anomalous effect Effects 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
- G01M99/005—Testing of complete machines, e.g. washing-machines or mobile phones
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to the technical field of machine detection, and provides a method and a device for detecting the abnormality of a working machine and the working machine, wherein the method comprises the following steps: determining the operation data of the operation machine to be detected; inputting the operation data into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model; the anomaly detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different. According to the invention, whether the operation machine to be detected is abnormal is determined by representing the operation data of whether the operation machine operates normally, so that the state of the operation machine to be detected can be analyzed more comprehensively and accurately, and the abnormality detection model can learn from a large amount of sample operation data, so that the abnormality detection result can be obtained at high precision.
Description
Technical Field
The invention relates to the technical field of machine detection, in particular to a method and a device for detecting an abnormality of a working machine and the working machine.
Background
In the working process of the working machine, abnormal working conditions such as pressure abnormality, flow abnormality and the like may occur, and if the working machine works under the abnormal working conditions for a long time, the normal operation of the working machine may be affected.
At present, operation data related to the abnormality of the working machine are obtained, a corresponding threshold value is set, and if the operation data is larger than the threshold value, the abnormality of the working machine is judged.
Disclosure of Invention
The invention provides a method and a device for detecting the abnormity of a working machine and the working machine, which are used for solving the defect of low abnormity detection precision of the working machine in the prior art.
The invention provides a method for detecting an abnormality of a working machine, which includes:
determining operation data of the operation machine to be detected;
inputting the operation data into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model;
the abnormal detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different; the score of each candidate model is used to characterize the performance of each candidate model.
According to the method for detecting the abnormality of the working machine, the abnormality detection model is obtained by training based on the following steps:
respectively inputting the sample operation data into each initial model to obtain a sample prediction result output by each initial model;
performing parameter iteration on each initial model based on the difference between the sample prediction result and the sample label to obtain a candidate model corresponding to each initial model;
and determining the score of each candidate model based on the test operation data and the test label, and taking the candidate model with the highest score as the abnormality detection model.
According to the work machine abnormality detection method provided by the present invention, determining the score of each candidate model based on the test operation data and the test label includes:
inputting the test operation data into each candidate model respectively to obtain a test prediction result output by each candidate model;
determining a score for each candidate model based on the test prediction and the test label.
According to the work machine abnormality detection method of the present invention, the candidate model having the highest score is used as the abnormality detection model, and the method further includes:
determining the influence degree of each type of data in the sample operation data based on the abnormal detection model;
and taking the corresponding type data with the influence degree larger than a threshold value as optimized sample operation data, and updating the abnormality detection model based on the optimized sample operation data and the sample label corresponding to the optimized sample operation data.
According to the method for detecting an abnormality of a working machine provided by the present invention, the step of inputting the sample operation data to each of the initial models further includes: and performing data cleaning on the sample operation data.
According to the work machine abnormality detection method provided by the present invention, the sample operation data includes at least one of power data, hydraulic data, electrical data, environmental data, and position data.
The present invention also provides a work machine abnormality detection device including:
the determining unit is used for determining the operation data of the to-be-detected working machine;
the detection unit is used for inputting the operation data into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model;
wherein the anomaly detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different; the score of each candidate model is used to characterize the performance of each candidate model.
The present invention also provides a work machine comprising: the work machine abnormality detection device described above.
The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the above-mentioned work machine abnormality detection methods when executing the computer program.
The present disclosure also provides a non-transitory computer-readable storage medium having stored thereon a computer program that, when executed by a processor, implements a work machine abnormality detection method as in any one of the above.
The present invention also provides a computer program product comprising a computer program which, when executed by a processor, implements a method of detecting an abnormality in a work machine as defined in any one of the above.
According to the method and the device for detecting the abnormity of the working machine and the working machine, whether the working machine to be detected is abnormal or not is determined by representing the operation data of whether the working machine operates normally or not, so that the state of the working machine to be detected can be analyzed more comprehensively and accurately. In addition, the anomaly detection model can learn from a large amount of sample operation data, so that an anomaly detection result can be obtained at high precision, and the problem of low detection precision caused by judgment based on a threshold value in the traditional method is solved.
Drawings
In order to more clearly illustrate the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art 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 can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a schematic flow diagram of a work machine anomaly detection method provided by the present disclosure;
FIG. 2 is a schematic flow chart of a method for training an anomaly detection model according to the present invention;
fig. 3 is a schematic structural view of a work machine abnormality detection device according to the present invention;
fig. 4 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
At present, when the abnormality of the working machine is detected, operation data related to the abnormality of the working machine is obtained and a corresponding threshold value is set, and if the operation data is larger than the threshold value, the abnormality of the working machine is judged, but the method is easy to cause misjudgment, and further the detection precision is low.
For example, when detecting a pressure abnormality of a working machine, it is generally considered that the working temperature of the working machine is related to the pressure abnormality, and therefore, in the prior art, by acquiring the current working temperature, it is determined that the working machine is in a pressure abnormality when the current working temperature is greater than a temperature threshold value. However, in actual operation, there are many factors that affect the pressure of the working machine, not only the working temperature, but also other factors that may exist, and sometimes the pressure of the working machine does not become abnormal when the current working temperature is greater than the temperature threshold, which may cause erroneous determination, affect the detection accuracy, and in the working process of the working machine, the abnormality of the working machine may not be accurately determined, which not only affects the working performance of the equipment, but also increases the maintenance cost of the equipment.
In view of this, the present invention provides a method for detecting an abnormality in a work machine. Fig. 1 is a schematic flow chart of a work machine abnormality detection method according to the present invention, and as shown in fig. 1, the method includes the steps of:
and step 110, determining the operation data of the to-be-detected working machine.
Here, the work machine to be detected is a work machine that needs abnormality detection. The operation data refers to data that can represent whether the operation of the work machine to be detected is normal, for example, the operation data may include oil temperature, water temperature, voltage, pressure, and the like, and when the operation data is abnormal (e.g., exceeds a preset range, fluctuates greatly, and the like), it indicates that the work machine may be abnormal. The operation data may be obtained by arranging a corresponding sensor on the to-be-detected operation machine, or may be acquired from a vehicle-mounted controller on the to-be-detected operation machine, which is not specifically limited in this embodiment of the present invention.
the anomaly detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different; the score of each candidate model is used to characterize the performance of each candidate model.
Specifically, the score of each candidate model is used for representing the performance of each candidate model, and the higher the score is, the better the performance of the corresponding candidate model is, and the higher the accuracy is when abnormality detection is performed; conversely, the lower the score is, the worse the performance of the corresponding candidate model is, and the lower the accuracy is when anomaly detection is performed. The score of each candidate model can be represented by the accuracy of each candidate model, that is, the higher the score of each candidate model is, the higher the accuracy of each candidate model is in anomaly detection. In addition, the score of each candidate model may also be represented by other evaluation indicators (such as a recall rate), which is not specifically limited in the embodiment of the present invention.
And training a plurality of initial models with different network structures respectively on the basis of the sample operation data and the sample labels corresponding to the sample operation data to obtain each candidate model. Because the network structures of the initial models are different, the obtained network structures of the candidate models are also different, and the performances of the candidate models are also different. The structure of each initial model may be a tree model structure, a Support Vector Machine (SVM) structure, a logistic regression model (LR) structure, or the like.
After the abnormity detection model is determined, the operation data is input into the abnormity detection model, the abnormity detection model carries out prediction, and an abnormity detection result output by the abnormity detection model is determined. Compared with the conventional method for determining whether the operation machine is abnormal through single data, the operation machine abnormality detection method provided by the embodiment of the invention determines whether the operation machine to be detected is abnormal through the operation data representing whether the operation of the operation machine is normal, so that the state of the operation machine to be detected can be analyzed more comprehensively and accurately. In addition, the anomaly detection model in the embodiment of the invention learns from a large amount of sample operation data, so that the anomaly detection result can be obtained at high precision, and the problem of low detection precision caused by judgment based on a single threshold in the traditional method is solved.
Based on the above embodiment, the anomaly detection model is obtained by training based on the following steps:
respectively inputting the sample operation data into each initial model to obtain a sample prediction result output by each initial model;
performing parameter iteration on each initial model based on the difference between the sample prediction result and the sample label to obtain a candidate model corresponding to each initial model;
and determining the score of each candidate model based on the test operation data and the test label, and taking the candidate model with the highest score as the abnormal detection model.
Specifically, each initial model is obtained based on sample running data and sample label training, and the specific process is as follows: and respectively inputting the sample operation data into each initial model, and predicting by each initial model to obtain a sample prediction result output by each initial model. The sample prediction result is a result obtained by prediction, that is, whether the operation machine is abnormal or not is predicted but not a real result, and the sample label is used for representing whether the operation machine corresponding to the sample operation data is abnormal or not, so that based on a difference between the sample prediction result output by each initial model and the sample label, a loss value of each initial model can be determined, and parameter iterative optimization is performed on each initial model based on the loss value until a convergence condition is reached (for example, the iteration number reaches a threshold value, the loss value is smaller than a preset loss value, and the like), and a candidate model corresponding to each initial model is obtained.
After obtaining each candidate model, the performance of each candidate model needs to be evaluated, and the candidate model with the best performance is selected as the abnormality detection model, that is, the score of each candidate model needs to be determined at this time. In the embodiment of the invention, the performance of each candidate model is verified based on the test operation data and the test label, the score of each candidate model is determined, and the candidate model with the highest score is used as the abnormality detection model. The test label is used for representing whether the operation machine corresponding to the test operation data is really abnormal or not, the test operation data and the sample operation data can be obtained from the originally acquired operation data, and for example, the originally acquired operation data are divided according to a ratio of 3:7 to respectively obtain the test operation data and the sample operation data.
Based on any of the above embodiments, determining a score for each candidate model based on the test run data and the test label includes:
respectively inputting the test operation data into each candidate model to obtain a test prediction result output by each candidate model;
and determining the score of each candidate model based on the test prediction result and the test label.
Specifically, test operation data is respectively input into each candidate model, and prediction is performed by each candidate model to obtain a test prediction result output by each candidate model. The test prediction result is a result obtained through prediction, that is, whether the working machine is abnormal or not is predicted but not a real result, and the test tag is used for representing whether the working machine corresponding to the test operation data is abnormal or not, so that the score of each candidate model can be determined based on the difference between the test prediction result output by each candidate model and the test tag, for example, the accuracy of each candidate model can be determined based on the difference between the test prediction result and the test tag, and the accuracy is used as the score of each candidate model, so that a model with the best performance can be selected from each candidate model as an abnormality detection model based on the score.
Based on any of the above embodiments, the candidate model with the highest score is taken as the abnormality detection model, and then the method further includes:
determining the influence degree of each type of data in the sample operation data based on the abnormal detection model;
and taking the corresponding type data with the influence degree larger than the threshold value as optimized sample operation data, and updating the abnormal detection model based on the optimized sample operation data and the sample label corresponding to the optimized sample operation data.
Specifically, the sample operation data includes various different types of data, such as pressure data, temperature data, and rotation speed data, and the different types of data have different degrees of influence on the operation of the work machine, that is, some types of data have a large influence on the work machine, some types of data have a small influence on the work machine, and the influence on the operation process of the work machine due to the types of data having the small influence can be ignored.
If training is performed based on all types of data, the obtained anomaly detection model has a large calculation amount during prediction, the calculation complexity is higher, and the cost of hardware needing to be deployed is higher. In contrast, the embodiment of the invention can determine the influence degree of each type of data in the sample operation data based on the anomaly detection model, screen out the type data with the influence degree larger than the threshold value as the optimized sample operation data, and update the anomaly detection model based on the optimized sample operation data and the sample label corresponding to the optimized sample operation data, thereby not only reducing the calculation amount and complexity of the anomaly detection model, but also ensuring the detection accuracy of the anomaly detection model.
Based on any of the above embodiments, the method further includes the step of inputting the sample operation data to each initial model, respectively: and performing data cleaning on the sample operation data.
Specifically, some invalid data and missing data may exist in the sample operation data, and if training is performed based on these data, interference may be caused, thereby affecting the accuracy of each candidate model.
In contrast, in the embodiment of the present invention, before the sample operation data is input to each initial model, the sample operation data is subjected to data cleaning to verify the sample operation data and filter out invalid data, actual data, and the like, so that the accuracy of each candidate model is ensured.
In any of the above embodiments, the sample operational data includes at least one of power data, hydraulic data, electrical data, environmental data, and position data.
Specifically, the abnormality detection model may be used to perform abnormality detection on the work machine, such as may detect whether the pressure of the work machine is abnormal. If the pressure anomaly detection is performed on the working machine, the sample operation data to be acquired is data affecting the pressure, such as at least one of power data (rotating speed, torque and the like), hydraulic data (pilot pressure, cavity pressure and the like), electrical data (voltage, current and the like), environmental data (oil temperature, water temperature and the like) and position data (geographical position, altitude and the like), and each initial model is trained based on the data, so that the finally obtained anomaly detection model can accurately determine whether the working machine is anomalous.
Based on any of the above embodiments, the present invention further provides an anomaly detection model training method, as shown in fig. 2, the method including:
a large amount of original operation data are collected, and after the original operation data are subjected to data cleaning, the original operation data are divided into sample operation data and test operation data. And then, training each initial model by adopting the sample operation data and the corresponding sample label to obtain each candidate model. And then, verifying each candidate model by adopting the test operation data and the corresponding test label thereof, determining the score of each candidate model to evaluate the performance of each candidate model, and taking the candidate model with the highest score as an abnormality detection model for performing abnormality detection on the working machine.
In the following description of the work machine abnormality detection apparatus according to the present invention, the work machine abnormality detection apparatus described below and the work machine abnormality detection method described above may be referred to in correspondence with each other.
Based on any of the embodiments described above, the present invention also provides a work machine abnormality detection apparatus, as shown in fig. 3, the apparatus including:
a determination unit 310 for determining operational data of the work machine to be inspected;
the detection unit 320 is configured to input the operation data to an anomaly detection model, so as to obtain an anomaly detection result output by the anomaly detection model;
wherein the anomaly detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different; the score of each candidate model is used to characterize the performance of each candidate model.
Based on any embodiment above, the apparatus further comprises:
the sample prediction unit is used for respectively inputting the sample operation data into each initial model to obtain a sample prediction result output by each initial model;
the parameter iteration unit is used for carrying out parameter iteration on each initial model based on the difference between the sample prediction result and the sample label to obtain a candidate model corresponding to each initial model;
and the score determining unit is used for determining the score of each candidate model based on the test operation data and the test label, and taking the candidate model with the highest score as the abnormality detection model.
Based on any embodiment above, the score determining unit includes:
the test prediction unit is used for respectively inputting the test operation data into each candidate model to obtain a test prediction result output by each candidate model;
and the calculating unit is used for determining the score of each candidate model based on the test prediction result and the test label.
Based on any embodiment above, the apparatus further comprises:
the influence degree determining unit is used for determining the influence degree of each type of data in the sample running data based on the abnormal detection model after the candidate model with the highest score is taken as the abnormal detection model;
and the updating unit is used for taking the corresponding type data with the influence degree larger than the threshold value as optimized sample operation data and updating the abnormal detection model based on the optimized sample operation data and the sample label corresponding to the optimized sample operation data.
Based on any embodiment above, the apparatus further comprises:
and the data cleaning unit is used for respectively inputting the sample operation data into each initial model and cleaning the sample operation data before.
In any of the above embodiments, the sample operational data includes at least one of power data, hydraulic data, electrical data, environmental data, and position data.
Based on any of the above embodiments, the present invention also provides a working machine, including: the work machine abnormality detection device according to any one of the above embodiments.
Here, the working machine may be a construction machine such as a crane, an excavator, a pile machine, or the like, or a construction vehicle such as a climbing vehicle, a fire truck, a mixer truck, or the like. The work machine provided with the work machine abnormality detection device according to any of the embodiments described above can accurately detect an abnormality in the work machine and obtain an abnormality detection result.
Fig. 4 is a schematic structural diagram of an electronic device provided in the present invention, and as shown in fig. 4, the electronic device may include: a processor (processor)410, a communication Interface 420, a memory (memory)430 and a communication bus 440, wherein the processor 410, the communication Interface 420 and the memory 430 are communicated with each other via the communication bus 440. Processor 410 may invoke logic instructions in memory 430 to perform a work machine anomaly detection method that includes: determining operation data of the operation machine to be detected; inputting the operation data into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model; wherein the anomaly detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different; the score of each candidate model is used to characterize the performance of each candidate model.
In addition, the logic instructions in the memory 430 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.
In another aspect, the present disclosure also provides a computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions that, when executed by a computer, enable the computer to perform the method for detecting abnormality of a work machine provided by the above methods, the method comprising: determining operation data of the operation machine to be detected; inputting the operation data into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model; wherein the anomaly detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different; the score of each candidate model is used to characterize the performance of each candidate model.
In yet another aspect, the present disclosure also provides a non-transitory computer-readable storage medium having stored thereon a computer program that, when executed by a processor, is implemented to perform the method for detecting an abnormality of a work machine provided in each of the above aspects, the method including: determining the operation data of the operation machine to be detected; inputting the operation data into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model; wherein the anomaly detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different; the score of each candidate model is used to characterize the performance of each candidate model.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention 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 solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method of detecting an abnormality in a work machine, comprising:
determining operation data of the operation machine to be detected;
inputting the operation data into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model;
wherein the anomaly detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different; the score of each candidate model is used to characterize the performance of each candidate model.
2. The work machine abnormality detection method according to claim 1, characterized in that the abnormality detection model is trained based on the steps of:
respectively inputting the sample operation data into each initial model to obtain a sample prediction result output by each initial model;
performing parameter iteration on each initial model based on the difference between the sample prediction result and the sample label to obtain a candidate model corresponding to each initial model;
and determining the score of each candidate model based on the test operation data and the test label, and taking the candidate model with the highest score as the abnormality detection model.
3. The work machine abnormality detection method according to claim 2, wherein determining the score of each candidate model based on the test run data and the test label includes:
inputting the test operation data into each candidate model respectively to obtain a test prediction result output by each candidate model;
determining a score for each candidate model based on the test prediction and the test label.
4. The work machine abnormality detection method according to claim 2, wherein the candidate model with the highest score is used as the abnormality detection model, and thereafter further comprises:
determining the influence degree of each type of data in the sample operation data based on the abnormal detection model;
and taking the corresponding type data with the influence degree larger than a threshold value as optimized sample operation data, and updating the abnormality detection model based on the optimized sample operation data and the sample label corresponding to the optimized sample operation data.
5. The work machine abnormality detection method according to claim 2, wherein the step of inputting the sample operation data to each of the initial models further comprises: and performing data cleaning on the sample operation data.
6. The work machine abnormality detection method according to any one of claims 1 to 5, characterized in that the sample operation data includes at least one of power data, hydraulic data, electrical data, environmental data, and position data.
7. A work machine abnormality detection device characterized by comprising:
the determining unit is used for determining the operation data of the to-be-detected working machine;
the detection unit is used for inputting the operation data into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model;
wherein the anomaly detection model is a model with the highest score selected from a plurality of candidate models; the candidate models are obtained by respectively training a plurality of initial models based on sample operation data and corresponding sample labels, and the network structures of the initial models are different; the score of each candidate model is used to characterize the performance of each candidate model.
8. A work machine, comprising: the work machine abnormality detection device according to claim 7.
9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the work machine abnormality detection method according to any one of claims 1 to 6 when executing the program.
10. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the work machine abnormality detection method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210616307.9A CN115014821A (en) | 2022-05-31 | 2022-05-31 | Method and device for detecting abnormality of working machine, and working machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210616307.9A CN115014821A (en) | 2022-05-31 | 2022-05-31 | Method and device for detecting abnormality of working machine, and working machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115014821A true CN115014821A (en) | 2022-09-06 |
Family
ID=83070673
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210616307.9A Pending CN115014821A (en) | 2022-05-31 | 2022-05-31 | Method and device for detecting abnormality of working machine, and working machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115014821A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110059775A (en) * | 2019-05-22 | 2019-07-26 | 湃方科技(北京)有限责任公司 | Rotary-type mechanical equipment method for detecting abnormality and device |
| CN110119787A (en) * | 2019-05-23 | 2019-08-13 | 湃方科技(北京)有限责任公司 | A kind of rotary-type operating condition of mechanical equipment detection method and equipment |
| CN110751227A (en) * | 2019-10-28 | 2020-02-04 | 中国建设银行股份有限公司 | Data processing method, device, equipment and storage medium |
| CN111311338A (en) * | 2020-03-30 | 2020-06-19 | 网易（杭州）网络有限公司 | User value prediction method and user value prediction model training method |
| CN113218537A (en) * | 2021-05-25 | 2021-08-06 | 中国南方电网有限责任公司超高压输电公司广州局 | Training method, device, equipment and storage medium of temperature anomaly detection model |
| CN113239970A (en) * | 2021-04-16 | 2021-08-10 | 首钢集团有限公司 | Model training method, equipment vibration abnormity detection method and device |
| CN113778766A (en) * | 2021-08-17 | 2021-12-10 | 华中科技大学 | Hard disk failure prediction model establishing method based on multi-dimensional characteristics and application thereof |
| CN113989367A (en) * | 2021-10-12 | 2022-01-28 | 三一重机有限公司 | Method and device for estimating attitude of working machine, and working machine |
| CN114090601A (en) * | 2021-11-23 | 2022-02-25 | 北京百度网讯科技有限公司 | Data screening method, device, equipment and storage medium |
| CN114330135A (en) * | 2021-12-30 | 2022-04-12 | 国网浙江省电力有限公司信息通信分公司 | Classification model construction method and device, storage medium and electronic equipment |
| CN114461499A (en) * | 2022-02-11 | 2022-05-10 | 中国工商银行股份有限公司 | Abnormal information detection model construction method and gray scale environment abnormal detection method |
-
2022
- 2022-05-31 CN CN202210616307.9A patent/CN115014821A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110059775A (en) * | 2019-05-22 | 2019-07-26 | 湃方科技(北京)有限责任公司 | Rotary-type mechanical equipment method for detecting abnormality and device |
| CN110119787A (en) * | 2019-05-23 | 2019-08-13 | 湃方科技(北京)有限责任公司 | A kind of rotary-type operating condition of mechanical equipment detection method and equipment |
| CN110751227A (en) * | 2019-10-28 | 2020-02-04 | 中国建设银行股份有限公司 | Data processing method, device, equipment and storage medium |
| CN111311338A (en) * | 2020-03-30 | 2020-06-19 | 网易（杭州）网络有限公司 | User value prediction method and user value prediction model training method |
| CN113239970A (en) * | 2021-04-16 | 2021-08-10 | 首钢集团有限公司 | Model training method, equipment vibration abnormity detection method and device |
| CN113218537A (en) * | 2021-05-25 | 2021-08-06 | 中国南方电网有限责任公司超高压输电公司广州局 | Training method, device, equipment and storage medium of temperature anomaly detection model |
| CN113778766A (en) * | 2021-08-17 | 2021-12-10 | 华中科技大学 | Hard disk failure prediction model establishing method based on multi-dimensional characteristics and application thereof |
| CN113989367A (en) * | 2021-10-12 | 2022-01-28 | 三一重机有限公司 | Method and device for estimating attitude of working machine, and working machine |
| CN114090601A (en) * | 2021-11-23 | 2022-02-25 | 北京百度网讯科技有限公司 | Data screening method, device, equipment and storage medium |
| CN114330135A (en) * | 2021-12-30 | 2022-04-12 | 国网浙江省电力有限公司信息通信分公司 | Classification model construction method and device, storage medium and electronic equipment |
| CN114461499A (en) * | 2022-02-11 | 2022-05-10 | 中国工商银行股份有限公司 | Abnormal information detection model construction method and gray scale environment abnormal detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10760995B2 (en) | Failure detection device for detecting an issue with a part of a machine | |
| CN108027611B (en) | Decision assistance system and method for machine maintenance using expert opinion supervised decision mode learning | |
| CN113467420B (en) | Method and device for detecting zone controller fault | |
| KR20190025474A (en) | Apparatus and Method for Predicting Plant Data | |
| CN112799898B (en) | Interconnection system fault node positioning method and system based on distributed fault detection | |
| CN109470946B (en) | Power generation equipment fault detection method and system | |
| CN114738205B (en) | Method, device, equipment and medium for monitoring state of floating fan foundation | |
| CN112565187B (en) | Power grid attack detection method, system, equipment and medium based on logistic regression | |
| CN112712305B (en) | Aircraft system and health assessment method and device thereof | |
| CN117786578A (en) | Method, device, equipment and storage medium for detecting electric power running state | |
| KR102110319B1 (en) | System for generating learning data | |
| CN114548280A (en) | Fault diagnosis model training method, fault diagnosis method and electronic equipment | |
| CN115952081A (en) | Software testing method, device, storage medium and equipment | |
| CN115014821A (en) | Method and device for detecting abnormality of working machine, and working machine | |
| CN113485898A (en) | Vibration measuring point display method, device, equipment and storage medium | |
| CN114322446B (en) | Cooling system fault early warning method and device, cooling system and working machine | |
| CN114341814A (en) | Efficient fault analysis by simulating faults in digital twinning | |
| CN108965053B (en) | Method, device and system for testing software pressure of upper computer | |
| CN112162528B (en) | Fault diagnosis method, device, equipment and storage medium of numerical control machine tool | |
| CN113094249A (en) | Node abnormity detection method, device and medium | |
| CN114528662A (en) | Method and device for predicting residual life of structural part and operating machine | |
| CN112907114A (en) | Oil leakage fault detection method and device, electronic equipment and storage medium | |
| WO2023044907A1 (en) | Method and device for monitoring equipment health and computer readable storage medium | |
| CN111965609A (en) | Radar reliability evaluation method and device, electronic equipment and readable storage medium | |
| CN117387925B (en) | Damper performance continuous test 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 |