CN111127811A - Fire early detection alarm system and method - Google Patents
Fire early detection alarm system and method Download PDFInfo
- Publication number
- CN111127811A CN111127811A CN201911357941.XA CN201911357941A CN111127811A CN 111127811 A CN111127811 A CN 111127811A CN 201911357941 A CN201911357941 A CN 201911357941A CN 111127811 A CN111127811 A CN 111127811A
- Authority
- CN
- China
- Prior art keywords
- alarm
- value
- fire
- temperature
- neural network
- 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
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/06—Electric actuation of the alarm, e.g. using a thermally-operated switch
-
- 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
- G06N3/043—Architecture, e.g. interconnection topology based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS]
-
- 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
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/10—Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/10—Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means
- G08B17/117—Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means by using a detection device for specific gases, e.g. combustion products, produced by the fire
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Emergency Management (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Automation & Control Theory (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Fuzzy Systems (AREA)
- Computational Mathematics (AREA)
- Fire Alarms (AREA)
Abstract
The invention provides a fire early detection alarm system and a fire early detection alarm method, wherein a plurality of active air suction type smoke detectors are adopted for sampling air to obtain the smoke concentration of a protection part, a plurality of temperature detectors are adopted for detecting the temperature to obtain the temperature value of the protection part, and a plurality of CO detectors are adopted for detecting CO to obtain the CO value of the protection part; and receiving the smoke concentration by adopting an alarm expansion card, receiving the temperature value and the CO value by adopting a loop card, sending the smoke concentration, the temperature value and the CO value to a main CPU, judging the fire alarm grade by the main CPU and displaying the judgment result. The response time of the electric smoldering fire which is difficult to find in the early stage in the closed cabin of the ship is prolonged, the characteristic quantity CO of the combustion products of the smoldering electric fire is fused, the parameters of various detectors are fused and judged through a fuzzy neural network and a genetic algorithm, the early warning function of the fire is realized, and meanwhile, the false alarm resistance is improved.
Description
Technical Field
The invention relates to a fire automatic alarm technology of a marine fire-fighting system, in particular to a fire early detection alarm system and a fire early detection alarm method.
Background
In the marine fire-fighting system in China, the front-end detection of fire detection alarm mainly adopts a point-type smoke-temperature composite detector and a difference constant-temperature detector, and core board cards such as an alarm controller mainboard, a loop card, a display card and the like mainly adopt imported marine products. Because most cabins in the ship are closed, electromechanical equipment is dense, personnel are numerous, an internal passage is narrow, the ship needs to sail on water for a long time, rescue is difficult after a fire disaster occurs, the problems of long detection alarm response time and weak false alarm resistance can be caused by only adopting a point-type passive smoke-sensing temperature-sensing detection means, and hidden danger exists for ensuring the fire safety of the ship. Meanwhile, the adoption of imported products is not beneficial to mastering the underlying communication protocol and the core technology, and once production stoppage or trade dispute occurs, great hidden danger also exists for supply guarantee of subsequent products.
The prior art related to the application is patent document CN 103839367a, and discloses a black box fire monitoring alarm system for a ship, which comprises a plurality of fire monitoring alarm devices arranged on the ship, and a black box and a fire monitoring computer which are arranged in a steering room, wherein each fire monitoring alarm device comprises a fire monitoring controller module, a first CAN bus communication circuit module, a Flash memory module, a smoke detector, a temperature-sensitive detector and a manual alarm button; the black box is internally integrated with a main control panel, a power supply, a data storage unit and a communication unit, the main control panel is provided with a central processing unit, a chip set, a CAN bus interface circuit module, an RJ45 interface circuit module and a memory, and the communication unit comprises a second CAN bus communication circuit module, a short message gateway and a network interface card.
Disclosure of Invention
In view of the shortcomings in the prior art, it is an object of the present invention to provide an early fire detection alarm system and method.
According to the invention, the fire early detection alarm system comprises:
a fire detection module: sampling air by adopting a plurality of active air suction type smoke detectors to obtain smoke concentration of a protection part, detecting temperature by adopting a plurality of temperature detectors to obtain a temperature value of the protection part, and detecting CO by adopting a plurality of CO detectors to obtain a CO value of the protection part;
fire alarm module: and receiving the smoke concentration by adopting an alarm expansion card, receiving the temperature value and the CO value by adopting a loop card, sending the smoke concentration, the temperature value and the CO value to a main CPU, judging the fire alarm grade by the main CPU and displaying the judgment result.
Preferably, the fire detection module includes:
a smoke sampling module: the smoke detectors are connected in series, communication networking is carried out through a 485 bus, the smoke concentration is sent to the alarm expansion card in real time through the PC-LINK, and the PC-LINK and the alarm expansion card are communicated through an RS-232 serial port;
a temperature sampling module: the temperature detector and the CO detector are communicated through a power signal multiplexing two bus, the temperature value and the CO value are sent to the loop card, and the CO detector and the loop card are communicated in a loop two bus mode.
Preferably, the fire alarm module includes:
an input module: the alarm expansion card sends the received smoke concentration to the main CPU through the CAN bus, the loop card sends the received temperature value and the received CO value to the main CPU through the CAN bus, and normalization processing is respectively carried out on the smoke concentration, the temperature value and the CO value;
a fuzzy module: fuzzification processing is carried out on the obtained three input parameters through a set membership function, three fuzzy sets are established, and input fuzzification is realized;
an inference module: multiplying elements in the three fuzzy sets respectively to obtain new parameter values, and realizing fuzzy reasoning;
a judgment module: and solving the fuzzy by using a weighted average mode, establishing a fuzzy neural network containing parameters, and substituting the three input parameters into the fuzzy neural network to obtain the fire alarm grade.
Preferably, the fuzzy neural network constructs a random number matrix N, assigns the uncertain membership function variable in the fuzzy neural network and the weight in the weighted average calculation to the corresponding elements in the matrix N, substituting the temperature, smoke and CO concentration variables in the training sample into a fuzzy neural network for reasoning calculation, comparing the obtained value with the actual fire alarm grade, retaining the parameters in a matrix N with good fitness by utilizing the selection, exchange and variation operations of a genetic algorithm, and optimizing the parameters with poor fitness, substituting corresponding elements in the generated new matrix N into the fuzzy neural network, performing reasoned calculation to solve an output result, performing iterative calculation on membership function variables until the output of the fuzzy neural network and the actual alarm level in the training sample meet the error requirement, solving to obtain a weight value, and establishing the fuzzy neural network.
The invention provides a fire early detection alarm method, which comprises the following steps:
and a fire detection step: sampling air by adopting a plurality of active air suction type smoke detectors to obtain smoke concentration of a protection part, detecting temperature by adopting a plurality of temperature detectors to obtain a temperature value of the protection part, and detecting CO by adopting a plurality of CO detectors to obtain a CO value of the protection part;
and fire alarm step: and receiving the smoke concentration by adopting an alarm expansion card, receiving the temperature value and the CO value by adopting a loop card, sending the smoke concentration, the temperature value and the CO value to a main CPU, judging the fire alarm grade by the main CPU and displaying the judgment result.
Preferably, the fire detecting step includes:
smoke sampling: the smoke detectors are connected in series, communication networking is carried out through a 485 bus, the smoke concentration is sent to the alarm expansion card in real time through the PC-LINK, and the PC-LINK and the alarm expansion card are communicated through an RS-232 serial port;
temperature sampling: the temperature detector and the CO detector are communicated through a power signal multiplexing two bus, the temperature value and the CO value are sent to the loop card, and the CO detector and the loop card are communicated in a loop two bus mode.
Preferably, the fire alarming step includes:
an input step: the alarm expansion card sends the received smoke concentration to the main CPU through the CAN bus, the loop card sends the received temperature value and the received CO value to the main CPU through the CAN bus, and normalization processing is respectively carried out on the smoke concentration, the temperature value and the CO value;
a fuzzy step: fuzzification processing is carried out on the obtained three input parameters through a set membership function, three fuzzy sets are established, and input fuzzification is realized;
the inference step comprises: multiplying elements in the three fuzzy sets respectively to obtain new parameter values, and realizing fuzzy reasoning;
and a judgment step: and solving the fuzzy by using a weighted average mode, establishing a fuzzy neural network containing parameters, and substituting the three input parameters into the fuzzy neural network to obtain the fire alarm grade.
Compared with the prior art, the invention has the following beneficial effects:
1. aiming at the electric smoldering fire which is difficult to find in the early stage in the closed cabin of the ship, the problem that the response time is slow due to the fact that the alarm time of the point-type passive smoke detector is influenced by factors such as a smoke propagation path, a fire source distance and air flow can be solved to the greatest extent;
2. the invention fuses the characteristic quantity CO of the combustion products of smoldering electrical fires, fuses and judges the parameters of various detectors through a fuzzy neural network and a genetic algorithm, realizes the early warning function of the fire and simultaneously improves the false warning resistance.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a hardware composition diagram of the early fire detection alarm method of the present invention.
FIG. 2 is a schematic diagram of the early fire detection and alarm method of the present invention.
FIG. 3 is a fuzzy neural network implementation method for early fire detection and alarm.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
In order to improve the viability of a ship and solve the problems of weak early detection capability of a fire and supply guarantee, the early detection and alarm method for the fire for the ship is provided. The hardware mainly comprises an air sampling smoke detector, a CO detector, a temperature detector and a fire alarm controller. The method mainly aims at electrical smoldering fire which is difficult to find in an early stage in a closed cabin of a ship, the problem that the alarm time of a point-type passive smoke detector in the conventional system is influenced by factors such as a smoke propagation path, a fire source distance and air flow, so that the response time is slow is solved to the greatest extent by adopting the active air-breathing type air sampling smoke detector, and meanwhile, the method integrates the characteristic quantity CO of combustion products of smoldering electrical fire (including cables, circuit boards and power distribution cabinets), performs fusion judgment on various detector parameters through a fuzzy neural network and a genetic algorithm, realizes the early fire alarm function, and improves the false alarm resistance.
The invention discloses an active air suction type early detection and alarm method for a marine fire based on fuzzy neural network multi-data fusion. By adopting active detection and multi-parameter fusion fuzzy neural network and genetic algorithm, the fire condition can be timely and accurately detected in the early stage of the fire, the survival capability of the ship is improved, and the problems of weak detection capability and supply guarantee in the early stage of the fire are solved. The active air-breathing smoke-sensing detector is mainly used for solving the problem that the alarm time of a point-type passive smoke-sensing detector in the conventional system is influenced by factors such as smoke propagation path, fire source distance, air flow and the like to the greatest extent and the response time is slow, and meanwhile, the invention integrates the characteristic quantity CO of combustion products of smoldering electrical fires (including cables, circuit boards and power distribution cabinets), fuses and judges various detector parameters through a fuzzy neural network and a genetic algorithm, realizes the early fire alarm function and improves the false alarm prevention capability.
As shown in fig. 1, the hardware is mainly composed of a fire detector at the front end and a fire alarm controller at the rear end. The fire detector mainly comprises an air sampling smoke detector and related accessories, a temperature sensing detector, a CO detector and a PC-LINK. The number and type of detectors can be selected according to the analysis and discrimination of the fire hazard source at the protected location. The air sampling smoke detector is networked through a 485 bus, and the temperature detector and the CO detector are communicated through two buses for multiplexing power signals. The fire alarm controller mainly comprises a main CPU, a display unit, an alarm expansion card and a loop card. The fire detector and the fire alarm controller are communicated in a RS-232 serial port and loop two-bus mode. And the board cards in the fire alarm controller are communicated through a CAN bus.
The air sampling smoke detector is an active air suction type smoke detector, smoke is sucked into a detection cavity through a sampling pipe and a capillary pipe which are arranged on a protection part, so that a smoke concentration value of the protection part is obtained and is used as one of characteristic quantities for fuzzy reasoning during early detection and alarm of fire. The PC-LINK is an interface card of the air sampling smoke-sensitive detector and other equipment. The interface card can realize the communication networking of a plurality of air sampling smoke detectors through a 485 bus, and can send parameters such as smoke concentration to other related equipment through an RS-232 interface. The protection parts can select different types of air sampling smoke detectors according to the size of the protection space, the detectors are connected through a 485 bus, and meanwhile, the PC-LINK is also in networking communication with the detectors through the 485 bus. The temperature-sensing detector is mainly used for monitoring temperature change in the environment, the environment temperature can not obviously rise for smoldering electrical fire, but the temperature-sensing detector is added into an early detection alarm system as an input parameter as a typical characteristic quantity of fire, so that once open fire, flash fire and other phenomena occur at a protected part of the systemAnd the fire type can be more accurately distinguished and early fire alarm can be realized. The CO detector is mainly used to monitor the concentration change of CO in the environment. In normal environment, the content of CO is less than 8.7mg/m3The CO content is less than 17.5mg/m in the kitchen and other places3The CO content in the smoke is 43.6mg/m3The above. Experiments show that smoldering fires can produce more CO than open fires and that CO is produced earlier and diffuses faster than smoke particles, so that they are well suited for fire detection, especially for early detection of smoldering fires which are not easily detected early. Meanwhile, the density of the carbon monoxide is slightly lighter than that of air, and the carbon monoxide is more diffusive than smoke, so that the carbon monoxide is beneficial to detection at the top of the ship. And the steam, dust, haze and the like can not generate CO, and the CO can be used as an important parameter for improving the false alarm resistance of the system and distinguishing real fire from false fire. The temperature-sensing detector and the CO detector are communicated with the loop card through a loop two-bus. The loop two bus adopts a bus communication form of power supply and signal multiplexing.
The main CPU is mainly responsible for completing fuzzy reasoning and deblurring by taking smoke concentration parameters sent by the alarm expansion card and temperature and CO concentration parameters sent by the loop card as input signals of the established fuzzy neural network, and outputting a corresponding fire alarm grade. And meanwhile, the main CPU is responsible for sending the related fire alarm information to the display unit for information display. The display unit mainly completes the functions of displaying and inquiring fire alarm information. The alarm expansion card is mainly responsible for sending relevant information such as smoke concentration of the air sampling smoke detector to the main CPU through the PC-LINK. The loop card is mainly responsible for communication with the temperature detector and the CO detector, and simultaneously sends the obtained temperature and CO concentration parameters to the main CPU through the CAN bus.
As shown in fig. 2, the smoke concentration data is sent to the alarm expansion card by the air sampling smoke-sensing detection port communication mode, and the temperature detector and the CO detector send the temperature data and the CO concentration data to the loop card by the loop two-bus mode. The alarm expansion card and the loop card in the fire alarm controller send the obtained smoke, temperature and CO concentration values to the main CPU, the main CPU is mainly responsible for fuzzifying smoke concentration parameters sent by the alarm expansion card and temperature and CO concentration parameters sent by the loop card, fuzzifying reasoning and defuzzifying are carried out on the smoke concentration parameters and the temperature and CO concentration parameters, and a genetic algorithm is used for training a neural network to determine relevant parameters of a fuzzy reasoning system, so that the grade judgment of whether a fire disaster occurs is made. And finally, displaying the judged result through a display unit.
As shown in fig. 3, the main CPU of the fire alarm controller is responsible for establishing a fuzzy neural network and collecting data in real time to determine the fire alarm level. The training samples determine the accuracy of early detection and alarm of fire, and the standard fire and the actual fire training samples of specific fire scenes need to be obtained according to the analysis of the fire hazard sources of the protected parts. The training sample needs to be trained through a fuzzy neural network, the invention designs a four-layer fuzzy neural network implementation method, the first layer is an input layer, and normalization processing is carried out on input temperature, smoke concentration and CO concentration parameters. The second layer is a fuzzy layer, and selects proper membership functions for fuzzification processing on the three input parameters, and establishes three fuzzy sets to realize fuzzification of input. The third layer is a fuzzy inference layer, and elements in the three fuzzy sets are multiplied respectively to obtain new parameter values. The fourth layer is a fuzzy decision layer, and the fuzzy decision layer is used for solving the fuzzy by using a weighted average method and initially establishing a fuzzy neural network containing parameters. And constructing a random number matrix N, assigning uncertain membership function variables in the fuzzy neural network and weights in weighted average calculation by corresponding elements in the N, substituting temperature, smoke and CO concentration variables in training samples into the fuzzy neural network to carry out inference calculation, comparing the solved values with the actual fire alarm level, reserving N parameters with good fitness by utilizing selection and exchange of a genetic algorithm and mutation operation, and optimizing parameters with poor fitness. And substituting corresponding elements in the generated new matrix N into the fuzzy neural network, and carrying out reasoned calculation to solve an output result. And (4) iteratively calculating the membership function variable until the output of the fuzzy neural network and the actual alarm level in the training sample meet the error requirement, and solving to obtain a weight value, thereby establishing the complete fuzzy neural network. After the establishment is completed, the main CPU substitutes the temperature, smoke concentration and CO concentration parameters acquired by the sensor in real time into the fuzzy neural network to calculate and obtain the corresponding fire alarm grade.
Those skilled in the art will appreciate that, in addition to implementing the systems, apparatus, and various modules thereof provided by the present invention in purely computer readable program code, the same procedures can be implemented entirely by logically programming method steps such that the systems, apparatus, and various modules thereof are provided in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system, the device and the modules thereof provided by the present invention can be considered as a hardware component, and the modules included in the system, the device and the modules thereof for implementing various programs can also be considered as structures in the hardware component; modules for performing various functions may also be considered to be both software programs for performing the methods and structures within hardware components.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (8)
1. An early fire detection alarm system, comprising:
a fire detection module: sampling air by adopting a plurality of active air suction type smoke detectors to obtain smoke concentration of a protection part, detecting temperature by adopting a plurality of temperature detectors to obtain a temperature value of the protection part, and detecting CO by adopting a plurality of CO detectors to obtain a CO value of the protection part;
fire alarm module: and receiving the smoke concentration by adopting an alarm expansion card, receiving the temperature value and the CO value by adopting a loop card, sending the smoke concentration, the temperature value and the CO value to a main CPU, judging the fire alarm grade by the main CPU and displaying the judgment result.
2. The early fire detection alarm system of claim 1, wherein the fire detection module comprises:
a smoke sampling module: the smoke detectors are connected in series, communication networking is carried out through a 485 bus, the smoke concentration is sent to the alarm expansion card in real time through the PC-LINK, and the PC-LINK and the alarm expansion card are communicated through an RS-232 serial port;
a temperature sampling module: the temperature detector and the CO detector are communicated through a power signal multiplexing two bus, the temperature value and the CO value are sent to the loop card, and the CO detector and the loop card are communicated in a loop two bus mode.
3. The early fire detection alarm system of claim 1, wherein the fire alarm module comprises:
an input module: the alarm expansion card sends the received smoke concentration to the main CPU through the CAN bus, the loop card sends the received temperature value and the received CO value to the main CPU through the CAN bus, and normalization processing is respectively carried out on the smoke concentration, the temperature value and the CO value;
a fuzzy module: fuzzification processing is carried out on the obtained three input parameters through a set membership function, three fuzzy sets are established, and input fuzzification is realized;
an inference module: multiplying elements in the three fuzzy sets respectively to obtain new parameter values, and realizing fuzzy reasoning;
a judgment module: and solving the fuzzy by using a weighted average mode, establishing a fuzzy neural network containing parameters, and substituting the three input parameters into the fuzzy neural network to obtain the fire alarm grade.
4. The early fire detection and alarm system of claim 3, wherein the fuzzy neural network is constructed by constructing a random number matrix N, assigning the corresponding elements in the matrix N to uncertain membership function variables in the fuzzy neural network and weights in the weighted average calculation, substituting the temperature, smoke and CO concentration variables in the training samples into the fuzzy neural network for inference calculation, comparing the solved values with the actual fire alarm level, retaining the parameters in the matrix N with good fitness by using the selection, exchange and mutation operations of the genetic algorithm, optimizing the parameters with poor fitness, substituting the corresponding elements in the new matrix N into the fuzzy neural network, performing inference calculation again to solve the output result, and iteratively calculating the membership function variables until the output of the fuzzy neural network and the actual alarm level in the training samples meet the error requirement, solving the weight value and establishing a fuzzy neural network.
5. An early fire detection alarm method, comprising:
and a fire detection step: sampling air by adopting a plurality of active air suction type smoke detectors to obtain smoke concentration of a protection part, detecting temperature by adopting a plurality of temperature detectors to obtain a temperature value of the protection part, and detecting CO by adopting a plurality of CO detectors to obtain a CO value of the protection part;
and fire alarm step: and receiving the smoke concentration by adopting an alarm expansion card, receiving the temperature value and the CO value by adopting a loop card, sending the smoke concentration, the temperature value and the CO value to a main CPU, judging the fire alarm grade by the main CPU and displaying the judgment result.
6. The early fire detection alarm method according to claim 5, wherein the fire detection step comprises:
smoke sampling: the smoke detectors are connected in series, communication networking is carried out through a 485 bus, the smoke concentration is sent to the alarm expansion card in real time through the PC-LINK, and the PC-LINK and the alarm expansion card are communicated through an RS-232 serial port;
temperature sampling: the temperature detector and the CO detector are communicated through a power signal multiplexing two bus, the temperature value and the CO value are sent to the loop card, and the CO detector and the loop card are communicated in a loop two bus mode.
7. The early fire detection alarm method according to claim 5, wherein the fire alarm step comprises:
an input step: the alarm expansion card sends the received smoke concentration to the main CPU through the CAN bus, the loop card sends the received temperature value and the received CO value to the main CPU through the CAN bus, and normalization processing is respectively carried out on the smoke concentration, the temperature value and the CO value;
a fuzzy step: fuzzification processing is carried out on the obtained three input parameters through a set membership function, three fuzzy sets are established, and input fuzzification is realized;
the inference step comprises: multiplying elements in the three fuzzy sets respectively to obtain new parameter values, and realizing fuzzy reasoning;
and a judgment step: and solving the fuzzy by using a weighted average mode, establishing a fuzzy neural network containing parameters, and substituting the three input parameters into the fuzzy neural network to obtain the fire alarm grade.
8. The method of claim 7, wherein the fuzzy neural network is constructed by constructing a random number matrix N, assigning the uncertain membership function variable in the fuzzy neural network and the weight in the weighted average calculation to corresponding elements in the matrix N, substituting the temperature, smoke and CO concentration variables in the training sample into the fuzzy neural network for inference calculation, comparing the solved value with the actual fire alarm level, retaining the parameters in the matrix N with good fitness by using the selection, exchange and mutation operations of the genetic algorithm, optimizing the parameters with poor fitness, substituting the corresponding elements in the new matrix N into the fuzzy neural network, performing inference calculation again to solve the output result, and performing iterative calculation on the membership function variable until the output of the fuzzy neural network and the actual alarm level in the training sample satisfy the error requirement, solving the weight value and establishing a fuzzy neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911357941.XA CN111127811B (en) | 2019-12-25 | 2019-12-25 | Fire early detection alarm system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911357941.XA CN111127811B (en) | 2019-12-25 | 2019-12-25 | Fire early detection alarm system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111127811A true CN111127811A (en) | 2020-05-08 |
| CN111127811B CN111127811B (en) | 2022-02-08 |
Family
ID=70502223
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911357941.XA Active CN111127811B (en) | 2019-12-25 | 2019-12-25 | Fire early detection alarm system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111127811B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111783940A (en) * | 2020-06-05 | 2020-10-16 | 广西民族大学 | Method for reducing fire false alarm rate based on GA-BP neural network algorithm |
| CN111915833A (en) * | 2020-07-15 | 2020-11-10 | 山东省科学院自动化研究所 | Fire multi-element detection device and detection method in limited space |
| CN113516824A (en) * | 2021-04-14 | 2021-10-19 | 汉威科技集团股份有限公司 | Composite fire detector and detection method thereof |
| CN113554843A (en) * | 2021-07-29 | 2021-10-26 | 无锡圣敏传感科技股份有限公司 | Pyrolytic particle fire detection method and detector |
| CN113944512A (en) * | 2021-10-18 | 2022-01-18 | 中煤科工集团重庆研究院有限公司 | Gas disaster accurate prediction method based on drilling multivariate information |
| CN114399881A (en) * | 2021-10-21 | 2022-04-26 | 国网山东省电力公司电力科学研究院 | Early fire recognition method and system |
| CN114419831A (en) * | 2022-01-06 | 2022-04-29 | 博迈科海洋工程股份有限公司 | Rectangular offshore inter-electric early-stage smoke generation source positioning method |
| CN114895211A (en) * | 2022-04-25 | 2022-08-12 | 中铁建设集团机电安装有限公司 | Theme park electromechanical engineering debugging system |
| CN115569338A (en) * | 2022-11-09 | 2023-01-06 | 国网安徽省电力有限公司蚌埠供电公司 | Multi-class fire early warning data distributed training and local fire extinguishing method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3032402B2 (en) * | 1993-05-18 | 2000-04-17 | ホーチキ株式会社 | Fire judging device and neural network learning method |
| CN103092074A (en) * | 2012-12-30 | 2013-05-08 | 重庆邮电大学 | Parameter optimization control method of semiconductor advance process control |
| CN103093574A (en) * | 2013-01-08 | 2013-05-08 | 中国科学技术大学 | Active inspiration type point type photoelectric smoke detector |
| CN104766433A (en) * | 2015-04-23 | 2015-07-08 | 河南理工大学 | Electrical fire warning system based on data fusion |
| CN207970693U (en) * | 2017-12-06 | 2018-10-16 | 西安科技大学 | A kind of building fire smoke absorption device |
| CN109243130A (en) * | 2018-09-07 | 2019-01-18 | 公安部四川消防研究所 | More methods physics fusion fire monitoring system and its predict fire probability |
| EP3489886A1 (en) * | 2017-11-22 | 2019-05-29 | Accenture Global Solutions Limited | Utilizing a machine learning model to identify activities and deviations from the activities by an individual |
-
2019
- 2019-12-25 CN CN201911357941.XA patent/CN111127811B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3032402B2 (en) * | 1993-05-18 | 2000-04-17 | ホーチキ株式会社 | Fire judging device and neural network learning method |
| CN103092074A (en) * | 2012-12-30 | 2013-05-08 | 重庆邮电大学 | Parameter optimization control method of semiconductor advance process control |
| CN103093574A (en) * | 2013-01-08 | 2013-05-08 | 中国科学技术大学 | Active inspiration type point type photoelectric smoke detector |
| CN104766433A (en) * | 2015-04-23 | 2015-07-08 | 河南理工大学 | Electrical fire warning system based on data fusion |
| EP3489886A1 (en) * | 2017-11-22 | 2019-05-29 | Accenture Global Solutions Limited | Utilizing a machine learning model to identify activities and deviations from the activities by an individual |
| CN207970693U (en) * | 2017-12-06 | 2018-10-16 | 西安科技大学 | A kind of building fire smoke absorption device |
| CN109243130A (en) * | 2018-09-07 | 2019-01-18 | 公安部四川消防研究所 | More methods physics fusion fire monitoring system and its predict fire probability |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111783940A (en) * | 2020-06-05 | 2020-10-16 | 广西民族大学 | Method for reducing fire false alarm rate based on GA-BP neural network algorithm |
| CN111915833A (en) * | 2020-07-15 | 2020-11-10 | 山东省科学院自动化研究所 | Fire multi-element detection device and detection method in limited space |
| CN113516824A (en) * | 2021-04-14 | 2021-10-19 | 汉威科技集团股份有限公司 | Composite fire detector and detection method thereof |
| CN113554843A (en) * | 2021-07-29 | 2021-10-26 | 无锡圣敏传感科技股份有限公司 | Pyrolytic particle fire detection method and detector |
| CN113944512A (en) * | 2021-10-18 | 2022-01-18 | 中煤科工集团重庆研究院有限公司 | Gas disaster accurate prediction method based on drilling multivariate information |
| CN113944512B (en) * | 2021-10-18 | 2024-01-26 | 中煤科工集团重庆研究院有限公司 | Accurate gas disaster prediction method based on drilling multi-element information |
| CN114399881B (en) * | 2021-10-21 | 2023-08-22 | 国网山东省电力公司电力科学研究院 | Early fire disaster identification method and system |
| CN114399881A (en) * | 2021-10-21 | 2022-04-26 | 国网山东省电力公司电力科学研究院 | Early fire recognition method and system |
| CN114419831A (en) * | 2022-01-06 | 2022-04-29 | 博迈科海洋工程股份有限公司 | Rectangular offshore inter-electric early-stage smoke generation source positioning method |
| CN114419831B (en) * | 2022-01-06 | 2023-12-19 | 博迈科海洋工程股份有限公司 | Rectangular marine electrical room early smoke generation source positioning method |
| CN114895211B (en) * | 2022-04-25 | 2022-10-25 | 中铁建设集团机电安装有限公司 | Theme park electromechanical engineering debugging system |
| CN114895211A (en) * | 2022-04-25 | 2022-08-12 | 中铁建设集团机电安装有限公司 | Theme park electromechanical engineering debugging system |
| CN115569338A (en) * | 2022-11-09 | 2023-01-06 | 国网安徽省电力有限公司蚌埠供电公司 | Multi-class fire early warning data distributed training and local fire extinguishing method and system |
| CN115569338B (en) * | 2022-11-09 | 2023-10-13 | 国网安徽省电力有限公司蚌埠供电公司 | Multi-category fire early warning data distributed training and local extinguishing method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111127811B (en) | 2022-02-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111127811B (en) | Fire early detection alarm system and method | |
| CN107564231B (en) | Building fire early warning and fire situation assessment system and method based on Internet of things | |
| JP6942029B2 (en) | Fire monitoring system | |
| US5483222A (en) | Multiple sensor apparatus and method | |
| CN110648490B (en) | Multi-factor flame identification method suitable for embedded platform | |
| JP6857007B2 (en) | Fire monitoring system | |
| CN112002095A (en) | Fire early warning method in mine tunnel | |
| CN113223264B (en) | Fire disaster intelligent early warning system and method based on QPSO-BP neural network | |
| CN110147762A (en) | A kind of embedded type fire control wrong report elimination system | |
| CN109948937B (en) | Method, system, equipment and medium for detecting potential safety hazard of artificial intelligence self-learning | |
| CN113192283B (en) | Wireless fire early warning system with multi-sensor information fusion | |
| CN112750272B (en) | Intelligent control method and system of fire alarm | |
| CN109920193A (en) | Intelligent detection method, system, equipment and medium for electrical fire hazard | |
| CN114580829A (en) | Power utilization safety sensing method, equipment and medium based on random forest algorithm | |
| CN117854013A (en) | Fire monitoring system and method for electrical equipment | |
| Muller et al. | A robust fire detection algorithm for temperature and optical smoke density using fuzzy logic | |
| Wang et al. | Forest fire detection system based on Fuzzy Kalman filter | |
| CN111508182A (en) | Novel monitoring alarm system for community fire fighting through Internet of things | |
| CN117010187A (en) | Analysis system for traditional rural fire prevention and control facility | |
| CN104240432B (en) | Mn-rich slag production safety based on information fusion monitoring method | |
| CN216249484U (en) | Airplane cargo compartment fire smoke detection system based on distributed multi-sensors | |
| Zhang et al. | Forecasting backdraft with multimodal method: Fusion of fire image and sensor data | |
| CN112229435A (en) | Ship main engine accidental fault detection method and system based on standard variable analysis | |
| JP2003036289A (en) | House evacuation safety evaluation method and house evacuation safety evaluation system | |
| US20240230136A9 (en) | System for diagnosis and management of indoor air quality using machine learning |
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 |