Method and device for managing fire composite detector

Abstract

The invention provides a fire composite detector management method and device, and the method comprises the steps: obtaining the real-time data of a plurality of fire composite detectors for detecting at least one detection object in a specific simulation scene after a target combustible is combusted at a preset fire position of a bullet train equipment cabin for a preset time; and based on the real-time data of at least one detection object detected by the plurality of fire composite detectors, the coverage range of each fire composite detector and the arrangement position of each fire composite detector, determining target arrangement measurement points for arranging the fire composite detectors in the bullet train equipment compartment in a layered manner. According to the application, the arrangement of the fire composite detector in the bullet train equipment compartment can be realized, early warning and position sensing of an early fire can be carried out by using the arranged fire composite detector in a fusion mode, and the response speed and accuracy of early warning are effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of design of fire detection devices, and in particular to a management method and device of a fire composite detector. Background Art

[0002] At present, with the continuous development of science and technology, EMUs have gradually become the focus of my country's high-speed railway development with their comfort and economy. Their service quality has been continuously improved, making them the first choice for passengers. However, once a fire occurs, due to the characteristics of EMUs such as high speed, untimely rescue, slow evacuation, fast burning speed, and limited rescue space, it will bring great safety risks to the lives and property of passengers.

[0003] Existing fire detection systems are mostly concentrated in the EMU carriages, toilets, and battery boxes, and are also equipped with traditional fire detection systems that mainly use single smoke or temperature sensing cables. The existing pyrotechnic systems of rail vehicles are not suitable for early detection of fires in complex environmental conditions (high ventilation volume, greatly affected by the external environment) of transformers in EMU equipment compartments. Summary of the invention

[0004] In view of this, the embodiment of the present application at least provides a method and device for managing a fire composite detector. Through the present application, not only can the deployment of fire composite detectors in the equipment compartment of a motor vehicle be realized, but also the deployed fire composite detectors can be used in a fusion manner to perform early warning and location perception of fire, effectively improving the response speed and accuracy of the warning. .

[0005] In a first aspect, an exemplary embodiment of the present application provides a method for deploying a fire composite detector, comprising: in a specific simulation scenario, after the target combustible material has burned for a predetermined time at a predetermined fire ignition position in a motor vehicle equipment compartment, obtaining real-time data of multiple fire composite detectors detecting at least one detection object; the fire composite detectors are dispersedly deployed at predetermined positions in the motor vehicle equipment compartment; based on the real-time data of multiple fire composite detectors detecting at least one detection object, the coverage range of each fire composite detector and the deployment position of each fire composite detector, determining the target deployment measurement points for layered deployment of the fire composite detectors in the motor vehicle equipment compartment; the target deployment measurement points are the optimal positions for deploying the fire composite detectors in the motor vehicle equipment compartment.

[0006] In the second aspect, an embodiment of the present application also provides a fire composite detector deployment device, comprising: a sensing unit, which, under a specific simulation scenario, obtains real-time data of multiple fire composite detectors detecting at least one detection object after the target combustible material has burned for a predetermined time at a predetermined fire ignition position in the EMU equipment compartment; the fire composite detectors are dispersedly deployed at predetermined positions in the EMU equipment compartment; a measuring point determination unit, which determines the target deployment measuring points for layered deployment of the fire composite detectors in the EMU equipment compartment based on the real-time data of multiple fire composite detectors detecting at least one detection object, the coverage range of each fire composite detector and the deployment position of each fire composite detector; the target deployment measuring points are the optimal positions for deploying the fire composite detectors in the EMU equipment compartment.

[0007] In a third aspect, an embodiment of the present application also provides a fire prediction system, which includes: an acquisition module, which obtains real-time data of the corresponding detection object in the current EMU equipment compartment collected by one or more target sensors in the fire composite detector arranged in the EMU equipment compartment; an AI decision module, which inputs the real-time data of the fire composite detector into the AI ​​decision engine for each fire composite detector to obtain the AI ​​predicted alarm level of the fire composite detector; a rule decision module, which matches the real-time data of the fire composite detector with the alarm level combination condition for each fire composite detector to obtain the rule predicted alarm level of the fire composite detector; a joint decision module, which performs joint decision processing on the AI ​​predicted alarm level and the rule predicted alarm level of the fire composite detector to obtain the alarm level of the fire composite detector.

[0008] In a fourth aspect, an embodiment of the present application also provides a fire prediction method including: obtaining real-time data of corresponding detection objects in the current EMU equipment compartment collected by one or more target sensors in a fire composite detector arranged in the EMU equipment compartment; for each fire composite detector, inputting the real-time data of the fire composite detector into an AI decision engine to obtain the AI ​​predicted alarm level of the fire composite detector; for each fire composite detector, matching the real-time data of the fire composite detector with the alarm level combination condition to obtain the rule predicted alarm level of the fire composite detector; performing joint decision processing on the AI ​​predicted alarm level and the rule predicted alarm level of the fire composite detector to obtain the alarm level of the fire composite detector.

[0009] In the fifth aspect, an embodiment of the present application also provides an electronic device, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of the above method.

[0010] In a sixth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are executed.

[0011] The management method and device of the fire composite detector provided in the embodiment of the present application can not only realize the deployment of the fire composite detector in the equipment compartment of the motor vehicle, but also use the deployed fire composite detector in a fusion manner to perform early warning and location perception of fire, effectively improving the response speed and accuracy of the warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0013] Figure 1 A flow chart showing a method for deploying a composite fire detector provided by an exemplary embodiment of the present application;

[0014] Figure 2 A schematic diagram showing that a combustible material provided by an exemplary embodiment of the present application is arranged at a predetermined ignition position;

[0015] Figure 3 One of the schematic diagrams showing the arrangement of the layers provided by the exemplary embodiment of the present application at predetermined positions;

[0016] Figure 4 A second schematic diagram showing the arrangement of the layers provided by the exemplary embodiment of the present application at predetermined positions;

[0017] Figure 5 A schematic diagram showing that a composite fire detector provided by an exemplary embodiment of the present application is arranged at a predetermined position;

[0018] Figure 6 A block diagram showing a device for deploying a composite fire detector provided by an exemplary embodiment of the present application;

[0019] Figure 7 A block diagram of a fire prediction system provided by an exemplary embodiment of the present application is shown;

[0020] Figure 8 A flow chart showing a fire prediction method provided by an exemplary embodiment of the present application;

[0021] Fig. 9 A schematic diagram of the structure of an electronic device provided by an exemplary embodiment of the present application is shown. DETAILED DESCRIPTION

[0022] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art, under the guidance of the content of the present application, can add one or more other operations to the flowchart, or remove one or more operations from the flowchart.

[0023] The terms "a", "an", "the" and "said" are used in this specification to indicate the presence of one or more elements / components / etc.; the terms "including" and "having" are used to express an open-ended inclusion and mean that additional elements / components / etc. may exist in addition to the listed elements / components / etc.; the terms "first" and "second" etc. are used only as labels and are not intended to limit the quantity of their objects.

[0024] It should be understood that in the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" is merely a way to describe the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the previous and subsequent associated objects are in an "or" relationship. "Including A, B and / or C" means including any one, any two, or any three of A, B, and C.

[0025] It should be understood that in the embodiments of the present application, "B corresponding to A", "B corresponding to A", "A corresponds to B", or "B corresponds to A" means that B is associated with A, and B can be determined according to A. Determining B according to A does not mean determining B only according to A, and B can also be determined according to A and / or other information.

[0026] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0027] At present, with the continuous development of science and technology, EMUs have gradually become the focus of my country's high-speed railway development with their comfort and economy. Their service quality has been continuously improved, making them the first choice for passengers. However, once a fire occurs, due to the characteristics of EMUs such as high speed, untimely rescue, slow evacuation, fast burning speed, and limited rescue space, it will bring great safety risks to the lives and property of passengers.

[0028] Existing fire detection systems are mostly concentrated in the EMU carriages, toilets, and battery boxes, and are also equipped with traditional fire detection systems that mainly use single smoke or temperature sensing cables. The existing pyrotechnic systems of rail vehicles are not suitable for early detection of fires in complex environmental conditions (high ventilation volume, greatly affected by the external environment) of transformers in EMU equipment compartments.

[0029] In response to at least one of the above problems, the present application proposes a design method for designing the layout position of fire composite detectors using the idea of ​​"uniform layout + layered layout", and also provides a solution for graded alarm using the laid out fire composite detectors.

[0030] To facilitate understanding of the present application, the management method and device of the fire composite detector provided in the embodiments of the present application are described in detail below.

[0031] See also Figure 1 , which is a flow chart of a method for deploying a composite fire detector provided in an exemplary embodiment of the present application.

[0032] like Figure 1 As shown, the method for deploying the fire composite detector of the exemplary embodiment of the present application specifically includes:

[0033] Step S101: In a specific simulation scenario, after the target combustible material burns at a predetermined fire location in the EMU equipment compartment for a predetermined time, real-time data of at least one detection object detected by multiple fire composite detectors is obtained; the fire composite detectors are dispersedly arranged at predetermined locations in the EMU equipment compartment.

[0034] Here, the specific simulation scenario may include: a scenario of simulating the operation of the EMU under the influence of the wind speed of the internal flow field during the operation of N meters per second. Here, N may be a speed value designed according to actual working conditions. For example, N may be 350, and the specific simulation scenario may be a scenario of simulating the operation of the EMU under the influence of the wind speed of the internal flow field during the operation of 350 meters per second.

[0035] In the embodiments of the present application, the target combustible material may be one or more of grease, PA6 nylon wire clamp seal, elbow plug nylon material, traction transformer cable, fan high-speed power cable, and transformer high-voltage insulation terminal. In addition, the target combustible material may also be other combustible materials set according to actual conditions, which will not be described in detail in the present application.

[0036] In the embodiment of the present application, through the analysis of the transformer composition materials, 7 non-metallic substances in the train can be determined, which are distributed in the six predetermined fire locations L1-L6. Figure 2 The table below shows the locations of the above six types of combustibles in the EMU.

[0037]

[0038] Multiple fire composite detectors are dispersedly arranged at predetermined positions in the equipment compartment of the motor vehicle. Here, the predetermined positions can be positions arranged based on experience or positions arranged in advance based on coverage. For example, the positions that can be arranged are as follows: Figure 2 shown.

[0039] In addition, the present application also uniquely proposes a layered layout method, which is based on the internal flow field simulation analysis: the dominant factor controlling the flow field characteristics in the EMU equipment compartment is the air flow caused by the cabin fan, which is limitedly affected by the train running speed and the position of the carriage. Combined with experimental verification, the non-metallic combustible materials near the transformer in the EMU equipment compartment all show flame retardant properties, and have a large density and a slow rising speed after combustion. At the same time, affected by the wind speed of the cooling unit in the EMU equipment compartment, once burning, the smoke moves horizontally at the burning position toward the air inlet of the cooling unit (i.e., cooling equipment). Therefore, two layered fire composite detection devices (such as P9 and P10) are set near the transformer and the cooling unit (where the five-pointed star is) Figure 3 shown).

[0040] In addition, as an example, the predetermined position may also include positions arranged in layers at predetermined layer heights at the same position of the motor vehicle equipment compartment. Figure 4As shown, a fire composite detector can be arranged upward at a predetermined interval at the same position. For example, since the internal height of the EMU equipment compartment is 775cm, the lowest fire composite detector in the layered arrangement is 30cm away from the bottom of the equipment compartment. Therefore, a fire composite detector can be arranged upward at an interval of 10cm.

[0041] Furthermore, as an example, the detection object may include at least one of the following items: CO concentration, smoke concentration, temperature, and VOC concentration.

[0042] Assuming that the target combustible is grease, the predetermined fire location is L2, and the predetermined time is 60 seconds, the objects that can be detected are CO concentration, smoke concentration, and temperature, etc., after the grease burns at L2 in the EMU equipment compartment for 60 seconds. Figure 5 The real-time data detected by the fire composite detectors deployed at 17 predetermined locations in the detection results are shown in the following table:

[0043]

[0044]

[0045] Step S102: Based on the real-time data of at least one detection object detected by multiple fire composite detectors, the coverage of each fire composite detector and the layout position of each fire composite detector, determine the target layout measurement points for layered layout of fire composite detectors in the EMU equipment compartment; the target layout measurement points are the optimal positions for laying out fire composite detectors in the EMU equipment compartment.

[0046] Regarding step S102, in specific implementation, the target measurement points can be determined by the following steps:

[0047] Step S1021, based on the real-time data of multiple fire composite detectors detecting at least one detection object, determining the layout position corresponding to the fire composite detector whose detection effect meets the first requirement.

[0048] Specifically, first, for each detection object, the fire composite detectors whose real-time data for detecting the detection object is greater than the detection threshold corresponding to the detection object can be screened out from all fire composite detectors. Here, the detection threshold corresponding to the detection object can be a threshold set based on experience. Then, the screened fire composite detectors are intersected to obtain fire composite detectors that meet the first requirement, that is, fire detectors that meet the first requirement are all fire composite detectors whose detection effect meets the requirement. Finally, the layout positions corresponding to the fire composite detectors that meet the first requirement are obtained.

[0049] Through the above method, the location for deploying the fire composite detector with effective detection effect can be screened out from multiple predetermined fire locations.

[0050] Step S1022, determining the layout position corresponding to the composite fire detector whose coverage range meets the second requirement.

[0051] Specifically, first, a fire composite detector that is less than the first distance from the target device is determined as a fire composite detector that meets the second requirement. Here, as an example, the target device may include at least one of the following items: a fan high-speed cable, a transformer cooling fan inlet. For example, a fire composite detector that is less than the first distance from the fan high-speed cable or the transformer cooling fan inlet is determined as a fire composite detector that meets the second requirement, and then the layout position corresponding to the fire composite detector that meets the second requirement is obtained.

[0052] Step S1023, determining that the layout position of the fire composite detector corresponds to the layout position of the fire composite detector that meets the third requirement.

[0053] Specifically, first, the fire composite detector whose distance to the cooling device is less than the second distance is determined as the fire composite detector that meets the third requirement; then, the layout position corresponding to the fire composite detector that meets the third requirement is obtained.

[0054] Step S1024, determining the layout positions corresponding to the fire composite detectors that meet the first requirement, the layout positions corresponding to the fire composite detectors whose coverage meets the second requirement, and the layout positions corresponding to the fire composite detectors whose layout positions meet the third requirement as target layout positions.

[0055] Through the above method, the target layout position of the fire composite detector can be determined. Through the above layout method, the optimal way to lay out the fire composite detector in the equipment compartment of the electric vehicle can be obtained, which effectively saves the layout cost. However, those skilled in the art should understand that, when necessary, the fire composite detector can also be laid out at the layoutable position to ensure the coverage of the detector.

[0056] Based on the same application concept, the embodiments of the present application also provide a fire composite detector deployment device corresponding to the method provided in the above embodiments. Since the principle of solving the problem by the device in the embodiments of the present application is similar to the fire composite detector deployment method in the above embodiments of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0057] Figure 4 The schematic diagram of the structure of the deployment device of the fire composite detector provided by the exemplary embodiment of the present application. Figure 4 As shown, the deployment device 200 of the fire composite detector includes:

[0058] The sensing unit 210, in a specific simulation scenario, obtains real-time data of at least one detection object detected by multiple fire composite detectors after the target combustible material burns at a predetermined fire location of the motor vehicle equipment compartment for a predetermined time; the fire composite detectors are dispersedly arranged at predetermined locations of the motor vehicle equipment compartment;

[0059] The measuring point determination unit 220 determines the target measuring points for arranging the fire composite detectors in layers in the EMU equipment compartment based on the real-time data of at least one detection object detected by multiple fire composite detectors, the coverage of each fire composite detector and the arrangement position of each fire composite detector; the target measuring points are the optimal positions for arranging the fire composite detectors in the EMU equipment compartment.

[0060] See also Figure 5 , Figure 5 A block diagram of a fire prediction system provided for an exemplary embodiment of the present application.

[0061] like Figure 5 As shown, the fire prediction system 800 described in the present application includes: a collection module 801, an AI decision module 802, a rule decision module 803 and a joint decision module 804.

[0062] The acquisition module 801 acquires the real-time data of the corresponding detection object in the current motor vehicle equipment compartment collected by one or more target sensors in the fire composite detector arranged in the motor vehicle equipment compartment.

[0063] here, Figure 6 The structural block diagram of a composite fire detector provided by an exemplary embodiment of the present application is shown, wherein the composite fire detector described in the present application may include one or more of a CO gas sensor, a VOC gas sensor, a smoke sensor, and a digital temperature sensor, a sampling circuit, and a main control chip. As an example, in the present application, the composite fire detector may include a CO gas sensor, a VOC gas sensor, a smoke sensor, and a digital temperature sensor.

[0064] The acquisition module 801 can obtain real-time data of the corresponding detection object in the current motor vehicle equipment compartment collected by one or more target sensors in the fire composite detector in the motor vehicle equipment compartment. Here, as an example, the target sensor may include at least one of the following items: a CO gas sensor, a VOC gas sensor, a photoelectric smoke sensor, and a digital temperature sensor.

[0065] Here, the CO gas sensor can collect the CO concentration in the EMU equipment compartment, the VOC gas sensor 302 can collect the VOC concentration in the EMU equipment compartment, the photoelectric smoke sensor 303 can collect the smoke in the EMU equipment compartment, and the digital temperature sensor 304 can collect the temperature of the EMU equipment compartment.

[0066] The acquisition module 801 can obtain the CO gas concentration, VOC concentration, smoke concentration and temperature collected in real time in the current EMU equipment compartment from the CO gas sensor 301, the VOC gas sensor 302, the photoelectric smoke sensor 303 and the digital temperature sensor 304, and send the real-time data of the CO gas concentration, VOC concentration, smoke concentration and temperature collected above to the AI ​​decision module 802 and the rule decision module 803.

[0067] For each composite fire detector, the AI ​​decision module 802 inputs the real-time data of the composite fire detector into the AI ​​decision engine to obtain the AI ​​predicted alarm level of the composite fire detector.

[0068] Specifically, as an example, the AI ​​decision module 802 may include: an XGBoost module, a LightGBM module, a CatBoost module, and a Stcacking module.

[0069] The real-time data of the fire composite detector can be input into the pre-trained XGBoost module, LightGBM module, CatBoost module and Stcacking module respectively, and each model performs fire graded warning processing to obtain the fire detection level of the fire composite detector. Here, the XGBoost module, LightGBM module, CatBoost module and Stcacking module are pre-trained warning models. After training, they can be saved through the dump function of the joblib library. When the warning model is running, the trained warning model can be loaded and called through the load function of the joblib library.

[0070] Specifically, the fire classification prediction model can be obtained in the following way, that is, the AI ​​decision-making module.

[0071] Step 1: Since there is still an uneven distribution of categories in the classified samples in the data set of the fire composite detector, the classified samples in the data set of the fire composite detector can be processed for data balancing to obtain feature data.

[0072] The second step is to normalize the feature data. For example, Python's MinMaxScaler is used to normalize the feature data. This step aims to improve the effect and accuracy of model training by scaling all feature values ​​to the same scale. For the CNN-LSTM deep learning model based on the attention mechanism, StandardScaler is used to standardize the features, and the to_categorical method is used to convert the label data into a classification format suitable for neural network processing.

[0073] The third step is to segment the data. In the gradient boosting algorithm and deep learning model, the train_test_split method is used to divide the data set into training set and test set, and K-fold iteration is performed. This method effectively ensures that the model has effective generalization ability on unseen data, thereby achieving better performance in practical applications.

[0074] The preprocessed data can be used for model training, but in order to improve the efficiency of model training and effectively reduce the model training time, it is also necessary to perform tentative optimization of hyperparameters. Hyperparameters determine the structure, complexity and performance of the model. Adjusting hyperparameters can optimize the model. Common hyperparameter optimization algorithms include grid search, random search, Bayesian optimization and other methods. This system uses random search to tune hyperparameters, which is more efficient than the traditional grid search method. James Bergstra and Yoshua Bengio compared grid search and random search in their published papers, and empirical evidence shows that random search is often a better method than grid search. After the system determines the optimal parameters of different algorithms, it further uses Bagging integration technology to fuse the models on different random seeds to become the deterministic model of XGBoost Bagging, LightGBM Bagging and CatBoost Bagging branch algorithms, and then uses Stacking fusion technology to further fuse the model. After completing the training of the fire classification prediction model, the model is saved through the dump function of the joblib library in Python. When the model is warned, the trained model is directly called through the load function of the joblib library to carry out the fire classification warning process.

[0075] In the specific implementation, the real-time data of CO gas concentration, VOC concentration, smoke concentration and temperature collected by the sensor in the fire composite detector A at the current moment can be used as the feature data of the XGBoost model 821. The XGBoost algorithm is used to perform parameter tuning and model training on the test data set to obtain the first branch fire prediction level of the fire composite detector A at the current moment.

[0076] At the same time, the real-time data of CO gas concentration, VOC concentration, smoke concentration and temperature collected by the sensor in the fire composite detector A at the current moment is used as the feature data of the LightGBM model 822. The LightGBM algorithm is used to perform parameter tuning and model training on the test data set to obtain the second branch fire prediction level of the fire composite detector A at the current moment.

[0077] At the same time, the real-time data of CO gas concentration, VOC concentration, smoke concentration and temperature collected by the sensor in the fire composite detector A at the current moment is used as the feature data of the CatBoost module 823. The CatBoost algorithm is used to perform parameter tuning and model training in the test data set to obtain the third branch fire prediction level of the fire composite detector A at the current moment.

[0078] The first branch fire prediction level, the second branch fire prediction level and the third branch fire prediction level obtained are input into the Stcacking model for logistic regression fusion judgment to obtain the AI ​​prediction alarm level of the fire composite detector A.

[0079] For example, the fire prediction level of the first branch is level one alarm, the fire prediction level of the second branch is level two alarm, and the fire prediction level of the third branch is level three alarm. After the above results are input into the Stcacking model for logistic regression fusion judgment, the AI ​​prediction alarm level of the fire composite detector A is level one alarm.

[0080] Since real fire test data cannot cover all possible fire scenarios, in order to ensure that no false alarms or missed alarms occur, this application also uses a rule engine design to expand the warning judgment conditions to achieve mandatory alarm conditions and directly output the warning level.

[0081] Therefore, in the process of determining the AI ​​predicted alarm level, at the same time, the present application also determines the alarm level through the preset alarm rules. Here, the rule decision module 803 matches the real-time data of each fire composite detector with the alarm level combination condition to obtain the rule predicted alarm level of the fire composite detector.

[0082] Specifically, the real-time data of CO gas concentration, VOC concentration, smoke concentration and temperature collected by the sensor in the fire composite detector A at the current moment are matched with the alarm level combination conditions to determine the rule-predicted alarm level of the fire composite detector A, wherein the matching relationship between the alarm level, the alarm level combination conditions and the threshold judgment rule can be shown in the following table:

[0083]

[0084]

[0085] For example, when only the VOC concentration reaches the alarm threshold, or only the smoke concentration reaches the alarm threshold, a first-level alarm (early warning) is issued; when one or more of the following conditions are met among the temperature reaching the primary threshold, the VOC concentration reaching the alarm threshold and the smoke reaching the alarm threshold, the CO concentration reaching the alarm threshold and the smoke reaching the alarm threshold, the VOC concentration reaching the alarm threshold and the CO reaching the alarm threshold, smoldering is detected and the smoke reaches the alarm threshold, a second-level alarm (fire alarm) is issued; when the temperature reaches the intermediate threshold; the smoke concentration, VOC concentration, and CO concentration reach the alarm threshold at the same time; the VOC concentration reaches the alarm threshold and the CO concentration reaches the alarm threshold and smoldering is detected; the smoke concentration reaches the alarm threshold and the CO concentration reaches the alarm threshold and smoldering is detected and the smoke concentration reaches the alarm threshold VOC When one or more of the conditions in which the concentration reaches the alarm threshold and smoldering is detected is met, a third-level alarm (fire alarm) is issued; when the temperature reaches the advanced threshold, the VOC concentration reaches the alarm threshold and the CO concentration reaches the alarm threshold and a rapid temperature rise is detected, a rapid temperature rise is detected and the VOC concentration reaches the alarm threshold and the smoke concentration reaches the alarm threshold, a rapid temperature rise is detected and the CO concentration reaches the alarm threshold and the smoke concentration reaches the alarm threshold, the CO concentration reaches the alarm threshold and the VOC concentration reaches the alarm threshold and the temperature reaches the intermediate threshold, the VOC concentration reaches the alarm threshold and the smoke concentration reaches the alarm threshold and the temperature reaches the intermediate threshold, and the CO concentration reaches the alarm threshold and the smoke reaches the alarm threshold and the temperature reaches the intermediate threshold, a fourth-level alarm (fire alarm) is issued. Here, the actual forms of the first-level alarm, the second-level alarm, the third-level alarm and the fourth-level alarm can be preset according to actual conditions or experience, and this application does not make any limitations here.

[0086] In this way, the rule-predicted alarm level of the fire composite detector can be obtained, and the interference of external combustion products can also be avoided. Taking straw burning as an example, the combustion products generated by external combustion may cause false alarms in the fire detector after entering the EMU equipment compartment. In order to reduce this influence, a straw burning interference experiment was carried out to simulate the external straw burning environment. The experiment showed that there were obvious differences in the characteristics of the smoke generated by external straw burning and the fire conditions in the EMU equipment compartment, which can be eliminated by the rule engine. The specific difference characteristics of the smoke generated by external straw burning flowing into the EMU equipment compartment are as follows: (1) No obvious temperature rise was found in the composite detectors in the area; (2) There was no obvious temperature rise in the temperature sensing cable; (3) The smoke concentration and VOC concentration of a large number (for example, 4 or more) of composite detectors increased at the same time.

[0087] In order to improve the response time and accuracy of the early warning, this application innovatively proposes to determine the alarm level of any fire composite detector through the joint decision-making of AI algorithm and rule engine.

[0088] Specifically, after obtaining the AI ​​predicted alarm level and the rule predicted alarm level of the fire composite detector, the joint decision module 804 performs joint decision processing on the AI ​​predicted alarm level and the rule predicted alarm level of the fire composite detector to obtain the alarm level of the fire composite detector.

[0089] As an example, the joint decision module 804 may match the AI ​​predicted alarm level and the rule predicted alarm level of the fire composite detector with the alarm level of the fire composite detector to obtain the alarm level of the fire composite detector.

[0090] Specifically, the obtained AI predicted alarm level and rule predicted alarm level can be matched according to the matching relationship shown in the following table to obtain the alarm level of the fire composite detector.

[0091]

[0092] Here, "0" represents consideration of fire, indicating that the alarm level has not been reached; "1" represents a level one alarm (early warning); "2" represents a level two alarm (fire alarm); "3" represents a level three alarm (fire alarm); and "4" represents a level four alarm (fire alarm).

[0093] For example, for fire composite detector A, when AI predicts the alarm level to be level 2 alarm and the rule predicts the alarm level to be level 1 alarm, by matching according to the contents in the above table, it can be seen that the alarm level of fire composite detector A is level 2 alarm.

[0094] The introduction of the rule decision module 803 can not only make up for the omissions or false alarms caused by insufficient original data learning and training of the AI ​​decision module 802, but also the joint decision-making of the AI ​​decision module 802 and the rule decision module 803 can effectively improve the response time and accuracy of the warning and solve the problem of fire warning in the confined space of the EMU.

[0095] In addition, since multiple fire composite detectors are installed in the EMU equipment compartment, and the wind speed will also affect the fire composite detectors at certain positions among the fire composite detectors installed in the EMU equipment compartment, it is necessary to make a final warning level for the alarm status of different fire composite detectors under the same working conditions.

[0096] Therefore, the fire prediction system of the present application further includes: a coordination unit ( Figure 5 (not shown) is used to determine the highest alarm level among the alarm levels of all the fire composite detectors as the current alarm level of the motor vehicle equipment compartment after obtaining the alarm levels of all the fire composite detectors.

[0097] According to the fire prediction system provided in the embodiment of the present application, a fusion method can be adopted to utilize the deployed composite fire detectors for early warning and location perception of fires, thereby effectively improving the response speed and accuracy of the warning.

[0098] Based on the same application concept, the embodiments of the present application also provide a fire prediction method provided in the above embodiments. Since the principle of solving the problem by the fire prediction method in the embodiments of the present application is similar to the fire prediction system in the above embodiments of the present application, the implementation of the fire prediction method can refer to the implementation of the fire prediction system, and the repeated parts will not be repeated.

[0099] The fire prediction method provided in the embodiment of the present application is as follows: Figure 8 As shown, including:

[0100] Step 301: acquiring real-time data of corresponding detection objects in the current motor vehicle equipment compartment collected by one or more target sensors in the fire composite detector arranged in the motor vehicle equipment compartment;

[0101] Step 302: for each composite fire detector, input the real-time data of the composite fire detector into the AI ​​decision engine to obtain the AI ​​predicted alarm level of the composite fire detector;

[0102] Step 303: for each composite fire detector, matching the real-time data of the composite fire detector with the alarm level combination condition to obtain the rule-predicted alarm level of the composite fire detector;

[0103] Step 304: Perform joint decision processing on the AI ​​predicted alarm level and the rule predicted alarm level of the fire composite detector to obtain the alarm level of the fire composite detector.

[0104] See also Fig. 9 , Fig. 9 The schematic diagram of the structure of the electronic device provided by the exemplary embodiment of the present application. Fig. 9 As shown, the electronic device 300 includes a processor 310 , a memory 320 and a bus 330 .

[0105] The memory 320 stores machine-readable instructions executable by the processor 310. When the electronic device 300 is running, the processor 310 communicates with the memory 320 via the bus 330. When the machine-readable instructions are executed by the processor 310, the steps of the method for deploying a composite fire detector in any of the above embodiments can be executed.

[0106] According to the fire prediction method provided in the embodiment of the present application, a fusion approach can be adopted to utilize the deployed composite fire detectors for early warning and location perception of fires, thereby effectively improving the response speed and accuracy of the warning.

[0107] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method in any of the above embodiments can be executed.

[0108] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0109] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0110] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0111] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0112] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for deploying a composite fire detector, characterized in that: include: In a specific simulation scenario, after the target combustible burns for a predetermined time at a predetermined fire location in the motor vehicle equipment compartment, real-time data of at least one detection object detected by multiple fire composite detectors is obtained; the fire composite detectors are dispersedly arranged at predetermined locations in the motor vehicle equipment compartment; Based on the real-time data of at least one detection object detected by multiple fire composite detectors, the coverage of each fire composite detector and the layout position of each fire composite detector, the target layout measurement points for layered layout of the fire composite detectors in the EMU equipment compartment are determined; the target layout measurement points are the optimal positions for laying the fire composite detectors in the EMU equipment compartment.

2. The deployment method according to claim 1, characterized in that: The predetermined positions include positions arranged in layers at predetermined heights at the same position of the motor vehicle equipment cabin.

3. The deployment method according to claim 1, characterized in that: The steps of determining target deployment measurement points for layered deployment of fire composite detectors in a motor vehicle equipment compartment based on real-time data of at least one detection object detected by multiple fire composite detectors, the coverage of each fire composite detector, and the deployment position of each fire composite detector include: Determine, based on real-time data of multiple fire composite detectors detecting at least one detection object, a layout position corresponding to the fire composite detector whose detection effect meets the first requirement; Determine the layout position of the fire composite detector corresponding to the fire composite detector whose coverage meets the second requirement; Determine that the layout position of the fire composite detector meets the layout position of the fire composite detector corresponding to the third requirement; The layout positions corresponding to the fire composite detectors that meet the first requirement, the layout positions corresponding to the fire composite detectors whose coverage meets the second requirement, and the layout positions corresponding to the fire composite detectors whose layout positions meet the third requirement are determined as target layout positions.

4. The arrangement position according to claim 3, characterized in that: The step of determining the layout position corresponding to the fire composite detector whose detection effect meets the first requirement based on the real-time data of at least one detection object detected by multiple fire composite detectors comprises: For each detection object, screen out the fire composite detectors whose real-time data for detecting the detection object is greater than the detection threshold corresponding to the detection object from all fire composite detectors; Taking the intersection of all the screened fire composite detectors to obtain the fire composite detectors that meet the first requirement; Obtain the layout position corresponding to the fire composite detector that meets the first requirement.

5. The management method according to claim 5, characterized in that: The step of determining the layout position of the composite fire detector corresponding to the composite fire detector whose coverage range meets the second requirement includes: Determine the fire composite detector that is located at a distance less than the first distance from the target device as the fire composite detector that meets the second requirement; Obtain the layout position corresponding to the fire composite detector that meets the second requirement.

6. The management method according to claim 5, characterized in that: The target device includes at least one of the following items: a fan high-speed cable, a transformer cooling fan inlet.

7. The management method according to claim 3, characterized in that: The step of determining the layout position of the fire composite detector corresponding to the layout position of the fire composite detector that meets the third requirement includes: Determine the fire composite detector that is located at a distance less than the second distance from the cooling device as the fire composite detector that meets the third requirement; Obtain the layout position of the fire composite detector corresponding to the third requirement.

8. The deployment method according to claim 1, characterized in that: The specific simulation scenario includes: simulating a scenario in which a train is running under the influence of the wind speed of the internal flow field during the running process of N meters per second.

9. The deployment method according to claim 1, characterized in that: The detection object includes at least one of the following items: CO concentration, flue gas concentration, temperature and VOC concentration; The target combustibles include at least one of the following: grease, nylon wire clamps, seals, elbow plug nylon materials, traction transformer cables, fan high-speed power cables and transformer high-voltage insulation terminals.

10. A deployment device for a composite fire detector, characterized in that: include: The sensing unit, in a specific simulation scenario, obtains real-time data of at least one detection object detected by multiple fire composite detectors after the target combustible material burns at a predetermined fire location in the motor vehicle equipment compartment for a predetermined time; the fire composite detectors are dispersedly arranged at predetermined locations in the motor vehicle equipment compartment; The measuring point determination unit determines the target measuring points for arranging the fire composite detectors in layers in the EMU equipment compartment based on the real-time data of at least one detection object detected by multiple fire composite detectors, the coverage of each fire composite detector and the arrangement position of each fire composite detector; the target measuring points are the optimal positions for arranging the fire composite detectors in the EMU equipment compartment.

11. A fire prediction system, characterized in that: The fire prediction system comprises: The acquisition module acquires the real-time data of the corresponding detection object in the current motor vehicle equipment compartment collected by one or more target sensors of the fire composite detector arranged in the motor vehicle equipment compartment; The AI ​​decision module inputs the real-time data of each fire composite detector into the AI ​​decision engine to obtain the AI ​​predicted alarm level of the fire composite detector; The rule decision module matches the real-time data of each fire composite detector with the alarm level combination condition to obtain the rule-predicted alarm level of the fire composite detector; The joint decision-making module performs joint decision-making processing on the AI ​​predicted alarm level and the rule predicted alarm level of the fire composite detector to obtain the alarm level of the fire composite detector.

12. The fire prediction system according to claim 11, characterized in that: The AI ​​decision-making module includes: The XGBoost module inputs the real-time data of the fire composite detector into the XGBoost model to obtain the first branch fire prediction level of the fire composite detector at the current moment; The LightGBM module inputs the real-time data of the fire composite detector into the LightGBM model to obtain the second branch fire prediction level of the fire composite detector at the current moment; CatBoost module, inputs the real-time data of the fire composite detector into the CatBoost model to obtain the third branch fire prediction level of the fire composite detector at the current moment; The Stcacking module inputs the first branch fire prediction level, the second branch fire prediction level and the third branch fire prediction level into the Stcacking model for logistic regression fusion judgment to obtain the AI ​​prediction alarm level of the fire composite detector.

13. The fire prediction system according to claim 11, characterized in that: The fire prediction system also includes: The coordination unit, after obtaining the alarm levels of all current fire composite detectors, determines the highest alarm level among the alarm levels of all fire composite detectors as the current alarm level of the EMU equipment compartment.

14. The fire prediction system according to claim 11, characterized in that: The target sensor includes at least one of the following items: a CO gas sensor, a VOC gas sensor, a photoelectric smoke sensor, and a digital temperature sensor.

15. A fire prediction method, characterized in that: The fire prediction method comprises: Acquire real-time data of corresponding detection objects in the current motor vehicle equipment compartment collected by one or more target sensors of the fire composite detector arranged in the motor vehicle equipment compartment; For each composite fire detector, the real-time data of the composite fire detector is input into the AI ​​decision engine to obtain the AI ​​predicted alarm level of the composite fire detector; For each fire composite detector, the real-time data of the fire composite detector is matched with the alarm level combination condition to obtain the rule-predicted alarm level of the fire composite detector; The AI ​​predicted alarm level and the rule predicted alarm level of the fire composite detector are jointly decided and processed to obtain the alarm level of the fire composite detector.

16. An electronic device, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform the steps of any method as claimed in claim 1 to 9 or 15.

17. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 or 15 are executed.

Images