Electric vehicle parking lot fire prevention and control system based on multi-source information fusion

Abstract

The invention discloses an electric vehicle parking lot fire prevention and control system based on multi-source information fusion. A sensing layer captures the environment state of an electric vehicle parking lot and the operation parameters of a vehicle battery in real time through a multi-mode sensor network; the network layer adopts an intelligent fusion network architecture to realize communication between the sensing layer and the data processing layer; the data processing layer is used for performing fusion analysis on the environment state and the operation parameters, judging a fire risk level and generating a dynamic control instruction; the data processing layer is composed of a multi-dimensional analysis engine, and a hierarchical progressive intelligent analysis architecture is adopted to realize multi-level research and judgment of fire risks; and the application layer receives the dynamic control instruction generated by the data processing layer, and realizes full-process closed-loop management of fire prevention and control through a multi-mode terminal device. The method not only improves the fire prevention and control precision of the electric vehicle parking lot, but also has good interpretability, and can be directly applied to a fire prevention and control system of the electric vehicle parking lot.

Description

Technical Field

[0001] The present invention relates to the field of fire prevention and control, and in particular to a fire prevention and control system for electric vehicle parking lots based on multi-source information fusion. Background Art

[0002] With the vigorous development of the global electric vehicle industry, the number of electric vehicles has increased dramatically, and the fire safety problem in their parking lots has become increasingly serious. As the core energy storage component, electric vehicle batteries have the characteristics of highly concentrated energy and fragile thermal stability. Once thermal runaway occurs, the fire will spread rapidly in a short period of time, releasing a large amount of toxic and harmful gases. Traditional firefighting methods are difficult to effectively contain the fire.

[0003] At present, conventional fire alarm systems have exposed many shortcomings when dealing with fires in electric vehicle parking lots. Its monitoring method based on a single or a small number of parameters cannot fully capture the complex early characteristics of electric vehicle fires, resulting in frequent false alarms and missed alarms. At the same time, the system is slow to respond, and there is often a long delay from the occurrence of a fire to the triggering of an alarm, delaying the best time to extinguish the fire. The fire extinguishing system lacks pertinence in the selection of extinguishing agents and injection control, making it difficult to accurately match the unique needs of electric vehicle fires and unable to extinguish fires efficiently. These problems seriously threaten the safety of life and property, and restrict the sustainable development of the electric vehicle industry. Therefore, it is urgent to develop an accurate, efficient, and intelligent fire prevention and control system for electric vehicle parking lots, which is of great practical significance for ensuring social public safety and promoting the steady development of the electric vehicle industry. Summary of the invention

[0004] The purpose of the present invention is to provide a fire prevention and control method for electric vehicle parking lots based on multi-source information fusion.

[0005] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0006] The present invention comprises the following steps:

[0007] The perception layer captures the environmental status of electric vehicle parking spaces and the operating parameters of vehicle batteries in real time through a multimodal sensor network;

[0008] The network layer adopts an intelligent fusion network architecture to achieve communication between the perception layer and the data processing layer;

[0009] The data processing layer is used to perform a fusion analysis of the environmental status and the operating parameters, determine the fire risk level, and generate dynamic control instructions; the data processing layer is composed of a multi-dimensional analysis engine, and adopts a hierarchical and progressive intelligent analysis architecture to achieve multi-level analysis and judgment of fire risks; the data processing layer includes a heterogeneous data preprocessing module, a multi-dimensional feature extraction module, a fire risk assessment module, and a dynamic control strategy generation module; including:

[0010] Based on the feature information extracted by the multidimensional feature extraction module, a fire risk assessment model is constructed using a three-level progressive analysis framework to assess the fire risk level of electric vehicle parking lots; the fire risk level includes primary assessment, secondary assessment and tertiary assessment;

[0011] The application layer receives the dynamic control instructions generated by the data processing layer, and realizes the full-process closed-loop management of fire prevention and control through multimodal terminal equipment; the application layer includes an audio and visual alarm module, a smoke exhaust system control module, a fire extinguishing system control module, a remote monitoring platform, and an emergency evacuation and fire extinguishing and rescue module.

[0012] Furthermore, the first-level assessment includes:

[0013] Use the deep neural network pre-trained on historical environment states and operating parameters to load its weight parameters;

[0014] Collect local data of the target parking lot and expand the sample size through data enhancement. Use transfer learning technology to freeze the underlying feature extraction layer of the model and only adjust the parameters of the top fully connected layer. The loss function is designed to be a combination of cross entropy loss and KL divergence.

[0015] The real-time sensor data is input into the fine-tuned model to output a fire risk probability value in the range of 0-1; a threshold is set, marked as a first-level high risk, triggering a second-level assessment.

[0016] Furthermore, the secondary assessment includes:

[0017] The Kalman filter is used to fuse the data of the same type of sensors to eliminate the isolated point error; the time series alignment algorithm is used to synchronize the BMS data and environmental data to the millisecond level;

[0018] Construct a Bayesian network topology, where nodes include sensor data, equipment health, and fire status; calculate the posterior probability based on the prior probability and conditional probability table;

[0019] The improved DS evidence theory is used to perform uncertainty reasoning on multi-source sensor data. Combined with the adaptive weighting algorithm, the trust weights of different sensors are dynamically allocated. The expression is:

[0020]

[0021] The health of the i-th device is The environmental adaptation coefficient is β, the dynamic correlation adjustment coefficient is μ, the global data set is U, and the i-th sensor data is D i , sensor data D i The historical correlation with the global dataset U is C orr , the correlation change rate is

[0022] Set a threshold, mark it as "secondary confirmation risk", and trigger the third-level assessment; use a sliding window mechanism to count the distribution of sensor data, analyze the timeliness of data through information entropy, and update the Bayesian network conditional probability table in real time.

[0023] Furthermore, the three-level evaluation method further includes:

[0024] Define entities, including battery types, fire causes, environmental conditions, and extinguishing agent characteristics, and establish causal chains based on historical cases and domain knowledge;

[0025] Input the current risk characteristics, traverse the knowledge graph through the graph neural network, match similar subgraphs; calculate the subgraph confidence and identify potential risk paths;

[0026] Combining the first-level probability output and the second-level confidence, the cause of the fire is deduced through the logic rule engine; if the knowledge graph reasoning result is consistent with the first-level and second-level assessments, a high risk level is determined; if there is a contradiction, manual review is triggered.

[0027] Furthermore, the method for screening the multidimensional features further includes:

[0028] Calculate the correlation between the multidimensional features and the fire, sort the multidimensional features in descending order according to the correlation, and take the multidimensional features with a correlation greater than 0.469 as the candidate feature set;

[0029] Combine candidate features according to the univariate perturbation principle according to their dimensions to obtain a multidimensional feature set, introduce a particle swarm, and use the multidimensional feature set as particles;

[0030] The prediction error of different multidimensional feature sets on the training set is calculated using the base model, and the fitness value of the particle is calculated based on the prediction error:

[0031]

[0032] The fitness function of the cth particle is The number of dimensions is N1, the multidimensional feature set is Q, and the minimum prediction error is y min , the prediction error of the cth multidimensional feature set is y c , the a-th multidimensional feature of the s-th dimension is u a (s), the z-th multidimensional feature of the k-th dimension is u z (k), multi-dimensional feature u a (s) and multidimensional features u z (k) Joint probability distribution, multi-dimensional feature u a The marginal probability distribution of (s) is p(u a (s)), multi-dimensional feature u zThe marginal probability distribution of (k) is p(u z (k)), the number of multidimensional features in the multidimensional feature set is N1;

[0033] The particle with the largest fitness value is taken as the attacker, and the average Euclidean distance between the particle and the attacker is calculated;

[0034] Select particles whose Euclidean distance to the attacker is greater than the average Euclidean distance, and update the particle position. The expression is:

[0035]

[0036] The position of the cth particle in the t+1th iteration is The upper bound of the search space is K max , the lower bound of the search space is K min , the position of the cth particle in the tth iteration is The current iteration number is t, the random number from 0 to 1 is τ1, and the position of the attacker is

[0037] After updating the particle, calculate the fitness value, divide the attackers twice according to the fitness value, and update the particle position according to the search step of the particle to obtain the mutation position. The expression is:

[0038]

[0039] The random numbers from 0 to 1 are τ4 and τ5, and the search step length is The mutation position of the cth particle in the t+1th iteration is The maximum number of iterations is t max ;

[0040] Until the fitness reaches the maximum, the corresponding multi-dimensional feature set is used as the output result.

[0041] Furthermore, the heterogeneous data preprocessing module and the multidimensional feature extraction module include:

[0042] The heterogeneous data preprocessing module adopts the configuration spectrum-spatial domain joint noise reduction algorithm, establishes a noise model based on sensor characteristics, and uses a sliding window mechanism to implement adaptive wavelet denoising on time series data such as temperature / smoke to achieve dynamic threshold noise reduction; Kalman filtering is used to fuse sensor data of the same type to eliminate isolated point errors for multi-source calibration; convolutional neural networks are used to identify interference sources such as device shadows and water vapor in video streams, and false trigger signals are corrected for pattern recognition correction;

[0043] The multidimensional feature extraction module is used to extract fire-related feature information from the preprocessed data and build a three-dimensional spatiotemporal feature extraction system, which includes time domain features, frequency domain features, spatial correlation features, and cross-modal features;

[0044] The time domain characteristics calculate the dynamic indicators of battery temperature gradient and second-order derivative of VOCs concentration; the pressure sensor spectrum energy distribution is analyzed by FFT, and the characteristic frequency of micro-detonation inside the battery is detected as a spatial correlation feature; a sensor topology model is constructed to analyze the spatial propagation path of abnormal signals and determine the spatial correlation characteristics; a twin network is used to align the feature space of thermal imaging and voltage fluctuation data as a cross-modal feature.

[0045] Furthermore, the dynamic control strategy generation module includes:

[0046] Based on the risk level and feature analysis results output by the fire risk assessment module, dynamic control instructions are generated through the multimodal decision engine; the dynamic control strategy generation module includes hierarchical response strategy and instruction generation logic and dynamic strategy optimization mechanism;

[0047] The hierarchical response strategy and instruction generation logic generates differentiated control instructions based on the fire risk level output by the three-level assessment and the causal reasoning results of the knowledge graph: the risk levels include low risk, medium risk and high risk;

[0048] When the risk is low, start the ventilation system in the target area to reduce the concentration of combustible gas; send a battery health check instruction to the BMS to force the charging current to be limited to a safe threshold; push yellow warning information to management personnel through the remote monitoring platform to prompt manual review; when the risk is medium, start the global sound and light alarm, and link the charging pile power off protection; activate the smoke exhaust system preparation mode, calculate the smoke diffusion path in real time and pre-adjust the direction of the smoke exhaust valve; when the risk is high, locate the fire source according to the thermal imaging data, and release fine water mist to suppress the initial fire; if the temperature gradient continues to rise, switch to full coverage fire extinguishing with perfluorohexanone gas; simultaneously cut off the main power supply of the parking lot, and enable the emergency lighting and evacuation guidance system; send three-dimensional thermal maps, vehicle location and battery type data to the city fire command center through the blockchain encrypted channel;

[0049] A dynamic strategy optimization mechanism is used to establish a control instruction execution effect evaluation model, and the fire extinguishing efficiency is calculated through real-time data from the fire extinguishing system pressure sensor and camera flame recognition results. If the efficiency is lower than the preset threshold, the strategy adjustment is automatically triggered: the historical case library in the knowledge graph is called to match the optimal strategy for similar scenarios; otherwise, the Q-Learning reinforcement learning algorithm is used to dynamically update the control strategy weights to achieve adaptive optimization.

[0050] In the second aspect, a fire prevention and control system for electric vehicle parking lots based on multi-source information fusion includes:

[0051] Data acquisition module: used for the perception layer to capture the environmental status of electric vehicle parking spaces and the operating parameters of vehicle batteries in real time through a multimodal sensor network;

[0052] Communication module: used for the network layer to adopt intelligent converged network architecture to achieve communication between the perception layer and the data processing layer;

[0053] Evaluation strategy generation module: used in the data processing layer to perform fusion analysis on the environmental status and the operating parameters, determine the fire risk level, and generate dynamic control instructions; the data processing layer is composed of a multi-dimensional analysis engine, and adopts a hierarchical and progressive intelligent analysis architecture to achieve multi-level analysis and judgment of fire risks; the data processing layer includes a heterogeneous data preprocessing module, a multi-dimensional feature extraction module, a fire risk assessment module, and a dynamic control strategy generation module; including:

[0054] Based on the feature information extracted by the multidimensional feature extraction module, a fire risk assessment model is constructed using a three-level progressive analysis framework to assess the fire risk level of electric vehicle parking lots; the fire risk level includes primary assessment, secondary assessment and tertiary assessment;

[0055] Positioning and rescue module: used for the application layer to receive the dynamic control instructions generated by the data processing layer, and to achieve full-process closed-loop management of fire prevention and control through multimodal terminal equipment; the application layer includes sound and light alarm module, smoke exhaust system control module, fire extinguishing system control module, remote monitoring platform, emergency evacuation and fire extinguishing rescue module

[0056] The beneficial effects of the present invention are:

[0057] The present invention is a fire prevention and control system for electric vehicle parking lots based on multi-source information fusion. Compared with the prior art, the present invention has the following technical effects:

[0058] The present invention can improve the accuracy of fire prevention and control in electric vehicle parking lots through preprocessing, fusion analysis, generation of dynamic control instructions, multi-level analysis of fire risks, assessment of fire risk levels and full-process closed-loop management steps, thereby improving the accuracy of fire prevention and control in electric vehicle parking lots, and optimizing fire prevention and control in electric vehicle parking lots, which can greatly save resources and improve work efficiency, and can realize intelligent prevention and control of fires in electric vehicle parking lots, and can perform fusion analysis and assessment of fire risk levels of electric vehicle parking lots in real time, which is of great significance to fire prevention and control in electric vehicle parking lots, and can adapt to fire prevention and control in electric vehicle parking lots of different standards and different fire prevention and control needs of electric vehicle parking lots, and has a certain universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 The present invention is a flowchart of the steps of a fire prevention and control method for electric vehicle parking lots based on multi-source information fusion. DETAILED DESCRIPTION

[0060] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0061] The present invention provides a fire prevention and control system for electric vehicle parking lots based on multi-source information fusion, comprising the following steps:

[0062] like Figure 1 As shown, in this embodiment, the following steps are included:

[0063] The perception layer captures the environmental status of electric vehicle parking spaces and the operating parameters of vehicle batteries in real time through a multimodal sensor network;

[0064] The network layer adopts an intelligent fusion network architecture to achieve communication between the perception layer and the data processing layer;

[0065] The data processing layer is used to perform a fusion analysis of the environmental status and the operating parameters, determine the fire risk level, and generate dynamic control instructions; the data processing layer is composed of a multi-dimensional analysis engine, and adopts a hierarchical and progressive intelligent analysis architecture to achieve multi-level analysis and judgment of fire risks; the data processing layer includes a heterogeneous data preprocessing module, a multi-dimensional feature extraction module, a fire risk assessment module, and a dynamic control strategy generation module; including:

[0066] Based on the feature information extracted by the multidimensional feature extraction module, a fire risk assessment model is constructed using a three-level progressive analysis framework to assess the fire risk level of electric vehicle parking lots; the fire risk level includes primary assessment, secondary assessment and tertiary assessment;

[0067] The application layer receives the dynamic control instructions generated by the data processing layer, and implements the full-process closed-loop management of fire prevention and control through multimodal terminal equipment; the application layer includes an audio-visual alarm module, a smoke exhaust system control module, a fire extinguishing system control module, a remote monitoring platform, and an emergency evacuation and fire extinguishing rescue module;

[0068] In the actual evaluation, β is the environmental adaptation coefficient, which represents the adjustment factor of the equipment health to the credibility weight. Its value is dynamically adjusted by the environmental conditions and can be dynamically fitted through the regression model of the historical failure rate of environmental parameters;

[0069] μ is the dynamic correlation adjustment coefficient, which represents the contribution ratio of the control history correlation change rate to the credibility, reflects the sensitivity of the system to the timeliness of data, and is based on the variance analysis of the correlation change within the sliding window and is optimized in combination with the information entropy theory;

[0070] is the health of the i-th device, which quantifies the operating status of the sensor itself, including calibration accuracy, fault history, signal stability, etc., and is normalized to [0,1], where 1 indicates complete health and 0 indicates failure;

[0071] C orr (D i ,U) is the historical correlation, representing the i-th sensor data D i The statistical correlation with the global data set U reflects the consistency of the sensor data with the overall state of the system;

[0072] is the correlation change rate, the first-order derivative of the historical correlation over time, which characterizes the dynamic change trend of the consistency between sensor data and the global state;

[0073] Temperature and humidity monitoring submodule: ground temperature 25°C, air temperature 26°C, ambient humidity 48% RH; gas monitoring submodule: CO concentration 25ppm, H2 concentration 53ppm, VOCs concentration 20ppm, HF concentration not detected;

[0074] Flame and smoke detection submodule: no flame spectrum characteristics, smoke concentration 6.73mg / m 3 ;

[0075] Visual perception submodule: battery pack surface temperature 30°C, no overheating of charging plug and battery bulging; battery body monitoring submodule: single cell temperature 32°C, battery deformation pressure 50kPa, battery internal resistance 0.01Ω;

[0076] Electrical parameter monitoring submodule: battery charge and discharge current 7.2A, battery pack total voltage 400V, single cell voltage 3.3V;

[0077] BMS data fusion interface: SOC estimation value 58%, no fault code, thermal management system operating normally;

[0078] Hybrid network architecture module: Assume that the 5G network signal strength is -80dBm, the Wi-Fi 6E network rate is 800Mbps, the ZigBee network node transmission success rate is 98%, the LoRa network node transmission success rate is 95%, and there is no packet loss in the optical fiber ring network;

[0079] Adaptive network dynamic adjustment module: Set the initial bandwidth allocation to 800Mbps for video stream, 100Mbps for battery status data, and 50Mbps for environmental monitoring data; the channel quality indicators are 0.01% bit error rate for video stream, 0.001% bit error rate for battery status data, and 0.05% bit error rate for environmental monitoring data;

[0080] Heterogeneous data preprocessing module: The sensor noise level is set to ±0.3°C for temperature sensor noise, ±2ppm for gas sensor noise (taking CO as an example), and ±1mg / m for smoke sensor noise 3 ;

[0081] Multidimensional feature extraction module: Calculate the battery temperature gradient of 0.1℃ / min and the second-order derivative of VOCs concentration of 0.01ppm / min 2 ;

[0082] Fire risk assessment module: In the first-level assessment, the accuracy of the model before fine-tuning based on transfer learning on local data is 70%; in the second-level assessment, the initial sensor credibility weights are: temperature sensor 0.3, gas sensor 0.3, smoke sensor 0.2, and visual sensor 0.2;

[0083] Sound and light alarm module: alarm volume 80dB(A), alarm light brightness 500cd / m 2 ;

[0084] Smoke exhaust system control module: the initial speed of the smoke exhaust fan is 800r / min, and the initial opening of the smoke exhaust valve is 30%;

[0085] Fire extinguishing system control module: water mist nozzle initial pressure 2MPa, perfluorohexanone gas storage pressure 4MPa.

[0086] In this embodiment, the perception layer includes:

[0087] The perception layer includes an environmental information collection module and an electric vehicle status information collection module;

[0088] The environmental information acquisition module includes a temperature and humidity monitoring submodule and a gas monitoring submodule;

[0089] The temperature and humidity monitoring submodule uses distributed optical fiber temperature sensors, which are laid along the parking space layout to monitor the ground and air temperature in real time with an accuracy of ±0.5°C, and eliminate electromagnetic interference through wavelength demodulation technology; an integrated capacitive humidity sensor detects ambient humidity with a range of 0-100% RH and an accuracy of ±2%;

[0090] The gas monitoring submodule is equipped with a multi-spectral gas sensor array to simultaneously detect CO, H2 (range 0-5000ppm), VOCs and HF with a resolution of ppb level;

[0091] The flame and smoke detection submodule uses a dual-band infrared flame detector, 3-5μm & 8-14μm, to achieve multiple verification of flame spectrum characteristics; deploys photoelectric smoke fire detectors to analyze particle size distribution through the scattering principle and distinguish fire smoke from dust / water mist interference;

[0092] The visual perception submodule is equipped with an infrared thermal imaging camera with a resolution of 640×480 and a temperature measurement range of -20℃ to 1200℃, which can capture abnormal temperature fields on the surface of the battery pack in real time; it also integrates an AI visual analysis unit to identify overheating of the charging plug and battery bulging;

[0093] The electric vehicle status information acquisition module includes a battery body monitoring submodule, an electrical parameter monitoring submodule and a BMS data fusion interface;

[0094] The battery body monitoring submodule uses a multi-channel thermocouple array embedded in the battery module to monitor the temperature of the single battery, with a sampling rate of 10Hz and an accuracy of ±0.2℃, and builds a three-dimensional temperature field model; the fiber grating strain sensor is attached to the battery shell to measure the deformation pressure, with a range of 0-500kPa and a resolution of 0.1% FS, to warn of the risk of battery expansion; the AC impedance spectrum analyzer uses a high-frequency excitation signal, 1kHz-10MHz, to detect the change in the internal resistance of the battery and quantify the degree of aging, with an SOH error of less than 2%;

[0095] The electrical parameter monitoring submodule uses a wide-band Hall sensor to collect the battery charge and discharge current in real time, with a range of ±500A and a bandwidth of DC-1MHz, to capture micro-short circuit transient characteristics; a high-precision voltage acquisition unit is used to synchronously monitor the total voltage of the battery pack and the single cell voltage, with an accuracy of ±0.05%, to identify the trend of voltage consistency degradation;

[0096] The BMS data fusion interface uses the CAN bus protocol to obtain SOC estimation values, fault codes and thermal management status in real time; it uses a time series alignment algorithm to synchronize BMS data with environmental sensor data to the millisecond level;

[0097] In the actual assessment, CO, range 0-1000ppm; H2, range 0-5000ppm; VOCs, range 0-1000ppm; HF, range 0-50ppm.

[0098] In this embodiment, the network layer includes:

[0099] The network layer includes a hybrid network architecture module, an adaptive network dynamic adjustment module, an edge-fog-cloud collaborative computing module, and blockchain-enabled network security;

[0100] The hybrid network architecture module includes multimodal communication protocol integration and network slicing and QoS classification;

[0101] The multimodal communication protocol integration includes high-speed data transmission channels, low-power wide-area coverage, and wired redundant backup. Through the high-speed data transmission channel, 5G and Wi-Fi 6E technologies are used to provide >10Gbps transmission rate for large-bandwidth data such as video streaming and three-dimensional heat maps, and the end-to-end delay is less than 5ms; low-power wide-area coverage deploys ZigBee and LoRa dual-mode base stations, supports wide-area coverage of low-frequency sensor data such as smoke concentration and temperature, and reduces power consumption to μW level; wired redundant backup lays optical fiber ring network based on GPON standard, providing 100% packet loss-free transmission guarantee in areas with severe electromagnetic interference;

[0102] Network slicing and QoS classification are based on SDN technology, which divides the physical network into three logical slices: emergency command slice, real-time monitoring slice and routine maintenance slice. The emergency command slice exclusively uses bandwidth resources, giving priority to the transmission of key data such as fire extinguishing instructions and evacuation alarms, ensuring latency <10ms. The real-time monitoring slice allocates dynamic bandwidth for video streams and battery status data, and supports QoS adaptive adjustment. The routine maintenance slice is used for non-real-time services such as device status heartbeat packets and log uploads.

[0103] The adaptive network dynamic adjustment module includes reinforcement learning-driven bandwidth allocation and multi-path fault-tolerant transmission;

[0104] Reinforcement learning-driven bandwidth allocation includes defining the state space, action space, and reward function; using a deep deterministic policy gradient algorithm to optimize network configuration in real time;

[0105] Multi-path fault-tolerant transmission adopts the multi-armed bandit model to dynamically select the optimal transmission path and seamlessly switch when the channel is congested or the signal is attenuated, with a packet loss rate of less than 0.01%; integrated forward error correction and ARQ retransmission protocol to ensure 100% accessibility of key instructions;

[0106] The edge-fog-cloud collaborative computing module includes edge computing nodes, fog computing clusters, and cloud collaborative mechanisms. The edge computing nodes deploy AI acceleration chips in the charging piles to perform sensor data preprocessing and feature extraction. The features include temperature gradient and gas concentration change rate. A lightweight LSTM model is used to achieve local fire risk preliminary judgment with an accuracy rate greater than 85%. Only high-risk data is uploaded to the fog computing layer.

[0107] The fog computing cluster is deployed on the parking lot area server to aggregate data from multiple edge nodes and eliminate spatiotemporal deviations through multi-source data fusion based on Kalman filtering. It runs a simplified fire risk assessment model to generate a pre-control strategy. It ensures that the command issuance delay is less than 50ms through the time-sensitive network protocol.

[0108] The cloud collaboration mechanism only calls cloud resources when uploading complex analysis tasks or when cross-regional collaboration is required; a federated learning framework is used to aggregate data from multiple parking lots to update the global model while protecting local data privacy;

[0109] Blockchain empowers network security, including data integrity protection and identity authentication and access control; data integrity protection generates a Merkle tree hash value for each data packet, verifies the hash chain consistency through smart contracts, and the tampering detection rate is greater than 99.99%; the national secret SM4 algorithm is used to encrypt sensitive data, and the key is dynamically updated through the quantum key distribution protocol; identity authentication and access control are based on the issuance of device digital certificates on the blockchain to achieve two-way authentication; an attribute-based encryption model is constructed to dynamically authorize data access rights by role;

[0110] In the actual evaluation, ZigBee, 2.4GHz; LoRa, Sub-1GHz; the wide area coverage radius is greater than 1km; the priority of QoS adaptive adjustment is 1080P video>battery temperature>ambient humidity.

[0111] In this embodiment, the first-level assessment includes:

[0112] Use the deep neural network pre-trained on historical environment states and operating parameters to load its weight parameters;

[0113] Collect local data of the target parking lot and expand the sample size through data enhancement. Use transfer learning technology to freeze the underlying feature extraction layer of the model and only adjust the parameters of the top fully connected layer. The loss function is designed to be a combination of cross entropy loss and KL divergence.

[0114] The real-time sensor data is input into the fine-tuned model to output a fire risk probability value in the range of 0-1; the threshold is set: probability>0.7, marked as level 1 high risk, triggering a level 2 assessment.

[0115] In this embodiment, the secondary evaluation includes:

[0116] The Kalman filter is used to fuse the data of the same type of sensors to eliminate the isolated point error; the time series alignment algorithm is used to synchronize the BMS data and environmental data to the millisecond level;

[0117] Construct a Bayesian network topology, where nodes include sensor data, equipment health, and fire status; calculate the posterior probability based on the prior probability and conditional probability table;

[0118] The improved DS evidence theory is used to perform uncertainty reasoning on multi-source sensor data. Combined with the adaptive weighting algorithm, the trust weights of different sensors are dynamically allocated. The expression is:

[0119]

[0120] The health of the i-th device is The environmental adaptation coefficient is β, the dynamic correlation adjustment coefficient is μ, the global data set is U, and the i-th sensor data is D i , sensor data D i The historical correlation with the global dataset U is C orr , the correlation change rate is

[0121] Set the threshold: if the evidence support is greater than 0.7, it will be marked as "secondary confirmation risk" and trigger the third-level assessment; use the sliding window mechanism to count the distribution of sensor data, analyze the timeliness of the data through information entropy, and update the Bayesian network conditional probability table in real time.

[0122] In this embodiment, the three-level evaluation method further includes:

[0123] Define entities, including battery types, fire causes, environmental conditions, and extinguishing agent characteristics, and establish causal chains based on historical cases and domain knowledge;

[0124] Input the current risk characteristics, traverse the knowledge graph through the graph neural network, match similar subgraphs; calculate the subgraph confidence and identify potential risk paths;

[0125] Combining the first-level probability output and the second-level confidence, the cause of the fire is deduced through the logic rule engine; if the knowledge graph reasoning result is consistent with the first-level and second-level assessments, a high risk level is determined; if there is a contradiction, manual review is triggered.

[0126] In this embodiment, the method for screening the multi-dimensional features further includes:

[0127] Calculate the correlation between the multidimensional features and the fire, sort the multidimensional features in descending order according to the correlation, and take the multidimensional features with a correlation greater than 0.469 as the candidate feature set;

[0128] Combine candidate features according to the univariate perturbation principle according to their dimensions to obtain a multidimensional feature set, introduce a particle swarm, and use the multidimensional feature set as particles;

[0129] The prediction error of different multidimensional feature sets on the training set is calculated using the base model, and the fitness value of the particle is calculated based on the prediction error:

[0130]

[0131] The fitness function of the cth particle is The number of dimensions is N1, the multidimensional feature set is Q, and the minimum prediction error is y min , the prediction error of the cth multidimensional feature set is y c, the a-th multidimensional feature of the s-th dimension is u a (s), the z-th multidimensional feature of the k-th dimension is u z (k), multi-dimensional feature u a (s) and multidimensional features u z (k) Joint probability distribution, multi-dimensional feature u a The marginal probability distribution of (s) is p(u a (s)), multi-dimensional feature u z The marginal probability distribution of (k) is p(u z (k)), the number of multidimensional features in the multidimensional feature set is N1;

[0132] The particle with the largest fitness value is taken as the attacker, and the average Euclidean distance between the particle and the attacker is calculated;

[0133] Select particles whose Euclidean distance to the attacker is greater than the average Euclidean distance, and update the particle position. The expression is:

[0134]

[0135] The position of the cth particle in the t+1th iteration is The upper bound of the search space is K max , the lower bound of the search space is K min , the position of the cth particle in the tth iteration is The current iteration number is t, the random number from 0 to 1 is τ1, and the position of the attacker is

[0136] After updating the particle, calculate the fitness value, divide the attackers twice according to the fitness value, and update the particle position according to the search step of the particle to obtain the mutation position. The expression is:

[0137]

[0138] The random numbers from 0 to 1 are τ4 and τ5, and the search step length is The mutation position of the cth particle in the t+1th iteration is The maximum number of iterations is t max ;

[0139] Until the fitness reaches the maximum, the corresponding multi-dimensional feature set is used as the output result.

[0140] In this embodiment, the heterogeneous data preprocessing module and the multidimensional feature extraction module include:

[0141] The heterogeneous data preprocessing module adopts the configuration spectrum-spatial domain joint noise reduction algorithm, establishes a noise model based on sensor characteristics, and uses a sliding window mechanism to implement adaptive wavelet denoising on time series data such as temperature / smoke to achieve dynamic threshold noise reduction; Kalman filtering is used to fuse sensor data of the same type to eliminate isolated point errors for multi-source calibration; convolutional neural networks are used to identify interference sources such as device shadows and water vapor in video streams, and false trigger signals are corrected for pattern recognition correction;

[0142] The multidimensional feature extraction module is used to extract fire-related feature information from the preprocessed data and build a three-dimensional spatiotemporal feature extraction system, which includes time domain features, frequency domain features, spatial correlation features, and cross-modal features;

[0143] The time domain features calculate the dynamic indicators of battery temperature gradient and second-order derivative of VOCs concentration; analyze the spectral energy distribution of the pressure sensor through FFT, and detect the characteristic frequency of micro-detonation inside the battery as a spatial correlation feature; build a sensor topology model, analyze the spatial propagation path of abnormal signals, and determine the spatial correlation features; use the twin network to align the feature space of thermal imaging and voltage fluctuation data as a cross-modal feature;

[0144] In actual evaluation, characteristic information includes temperature change rate, smoke concentration change rate, combustible gas concentration change rate, flame characteristics, abnormal battery temperature, abnormal battery voltage, abnormal battery current, abnormal battery deformation pressure, and abnormal battery internal resistance.

[0145] In this embodiment, the dynamic control strategy generation module includes:

[0146] Based on the risk level and feature analysis results output by the fire risk assessment module, dynamic control instructions are generated through the multimodal decision engine; the dynamic control strategy generation module includes hierarchical response strategy and instruction generation logic and dynamic strategy optimization mechanism;

[0147] The hierarchical response strategy and instruction generation logic generates differentiated control instructions based on the fire risk level output by the three-level assessment and the causal reasoning results of the knowledge graph: the risk levels include low risk, medium risk and high risk;

[0148] When the risk is low, start the ventilation system in the target area to reduce the concentration of combustible gas; send a battery health check instruction to the BMS to force the charging current to be limited to a safe threshold; push yellow warning information to management personnel through the remote monitoring platform to prompt manual review; when the risk is medium, start the global sound and light alarm, and link the charging pile power-off protection; activate the smoke exhaust system preparation mode, calculate the smoke diffusion path in real time and pre-adjust the direction of the smoke exhaust valve; when the risk is high, locate the fire source according to the thermal imaging data, and release fine water mist to suppress the initial fire; if the temperature gradient continues to rise and the temperature gradient is less than 5℃ / s, switch to full coverage fire extinguishing with perfluorohexanone gas; simultaneously cut off the main power supply of the parking lot, and enable the emergency lighting and evacuation guidance system; send three-dimensional thermal maps, vehicle location and battery type data to the city fire command center through the blockchain encrypted channel;

[0149] A dynamic strategy optimization mechanism is used to establish a control instruction execution effect evaluation model, and the fire extinguishing efficiency is calculated through real-time data from the fire extinguishing system pressure sensor and camera flame recognition results. If the efficiency is lower than the preset threshold, the strategy adjustment is automatically triggered: the historical case library in the knowledge graph is called to match the optimal strategy for similar scenarios; otherwise, the Q-Learning reinforcement learning algorithm is used to dynamically update the control strategy weights to achieve adaptive optimization.

[0150] In the second aspect, a fire prevention and control system for electric vehicle parking lots based on multi-source information fusion includes:

[0151] Data acquisition module: used for the perception layer to capture the environmental status of electric vehicle parking spaces and the operating parameters of vehicle batteries in real time through a multimodal sensor network;

[0152] Communication module: used for the network layer to adopt intelligent converged network architecture to achieve communication between the perception layer and the data processing layer;

[0153] Evaluation strategy generation module: used in the data processing layer to perform fusion analysis on the environmental status and the operating parameters, determine the fire risk level, and generate dynamic control instructions; the data processing layer is composed of a multi-dimensional analysis engine, and adopts a hierarchical and progressive intelligent analysis architecture to achieve multi-level analysis and judgment of fire risks; the data processing layer includes a heterogeneous data preprocessing module, a multi-dimensional feature extraction module, a fire risk assessment module, and a dynamic control strategy generation module; including:

[0154] Based on the feature information extracted by the multidimensional feature extraction module, a fire risk assessment model is constructed using a three-level progressive analysis framework to assess the fire risk level of electric vehicle parking lots; the fire risk level includes primary assessment, secondary assessment and tertiary assessment;

[0155] Positioning and rescue module: used for the application layer to receive the dynamic control instructions generated by the data processing layer, and realize the full-process closed-loop management of fire prevention and control through multimodal terminal equipment; the application layer includes sound and light alarm module, smoke exhaust system control module, fire extinguishing system control module, remote monitoring platform, emergency evacuation and fire extinguishing and rescue module.

[0156] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A fire prevention and control method for electric vehicle parking lots based on multi-source information fusion, characterized in that: The following steps are involved: The perception layer captures the environmental status of electric vehicle parking spaces and the operating parameters of vehicle batteries in real time through a multimodal sensor network; The network layer adopts an intelligent fusion network architecture to achieve communication between the perception layer and the data processing layer; The data processing layer is used to perform a fusion analysis on the environmental status and the operating parameters, determine the fire risk level, and generate dynamic control instructions; The data processing layer is composed of a multi-dimensional analysis engine, and adopts a hierarchical and progressive intelligent analysis architecture to achieve multi-level analysis and judgment of fire risks; the data processing layer includes a heterogeneous data preprocessing module, a multi-dimensional feature extraction module, a fire risk assessment module and a dynamic control strategy generation module; including: Based on the feature information extracted by the multidimensional feature extraction module, a fire risk assessment model is constructed using a three-level progressive analysis framework to assess the fire risk level of electric vehicle parking lots; the fire risk level includes primary assessment, secondary assessment and tertiary assessment; The application layer receives the dynamic control instructions generated by the data processing layer, and realizes the full-process closed-loop management of fire prevention and control through multimodal terminal equipment; the application layer includes an audio and visual alarm module, a smoke exhaust system control module, a fire extinguishing system control module, a remote monitoring platform, and an emergency evacuation and fire extinguishing and rescue module.

2. According to claim 1, a method for fire prevention and control in electric vehicle parking lots based on multi-source information fusion, characterized in that: The first-level assessment includes: Use the deep neural network pre-trained on historical environment states and operating parameters to load its weight parameters; Collect local data of the target parking lot and expand the sample size through data enhancement. Use transfer learning technology to freeze the underlying feature extraction layer of the model and only adjust the parameters of the top fully connected layer. The loss function is designed to be a combination of cross entropy loss and KL divergence. The real-time sensor data is input into the fine-tuned model to output a fire risk probability value in the range of 0-1; a threshold is set, marked as a first-level high risk, triggering a second-level assessment.

3. According to claim 1, a method for fire prevention and control in electric vehicle parking lots based on multi-source information fusion, characterized in that: The secondary assessment includes: The Kalman filter is used to fuse the data of the same type of sensors to eliminate the isolated point error; the time series alignment algorithm is used to synchronize the BMS data and environmental data to the millisecond level; Construct a Bayesian network topology, where nodes include sensor data, equipment health, and fire status; calculate the posterior probability based on the prior probability and conditional probability table; The improved DS evidence theory is used to perform uncertainty reasoning on multi-source sensor data. Combined with the adaptive weighting algorithm, the trust weights of different sensors are dynamically allocated. The expression is: The health of the i-th device is The environmental adaptation coefficient is β, the dynamic correlation adjustment coefficient is μ, the global data set is U, and the i-th sensor data is D i , sensor data D i The historical correlation with the global dataset U is C orr , the correlation change rate is Set a threshold, mark it as "secondary confirmation risk", and trigger the third-level assessment; use the sliding window mechanism to count the distribution of sensor data, analyze the timeliness of data through information entropy, and update the Bayesian network conditional probability table in real time.

4. According to claim 1, a method for fire prevention and control in electric vehicle parking lots based on multi-source information fusion, characterized in that: The three-level evaluation method further includes: Define entities, including battery types, fire causes, environmental conditions, and extinguishing agent characteristics, and establish causal chains based on historical cases and domain knowledge; Input the current risk characteristics, traverse the knowledge graph through the graph neural network, match similar subgraphs; calculate the subgraph confidence and identify potential risk paths; Combining the first-level probability output and the second-level confidence, the cause of the fire is deduced through the logic rule engine; if the knowledge graph reasoning result is consistent with the first-level and second-level assessments, a high risk level is determined; if there is a contradiction, manual review is triggered.

5. According to claim 1, a method for fire prevention and control in electric vehicle parking lots based on multi-source information fusion, characterized in that: The method for screening the multidimensional features further includes: Calculate the correlation between multidimensional features and fire, sort the multidimensional features in descending order according to the correlation, and take the multidimensional features with a correlation greater than 0.469 as the candidate feature set; Combine candidate features according to the univariate perturbation principle according to their dimensions to obtain a multidimensional feature set, introduce a particle swarm, and use the multidimensional feature set as particles; The prediction error of different multidimensional feature sets on the training set is calculated using the base model, and the fitness value of the particle is calculated based on the prediction error: The fitness function of the cth particle is The number of dimensions is N1, the multidimensional feature set is Q, and the minimum prediction error is y min , the prediction error of the cth multidimensional feature set is y c , the a-th multidimensional feature of the s-th dimension is u a (s), the z-th multidimensional feature of the k-th dimension is u z (k), multi-dimensional feature u a (s) and multidimensional features u z (k) Joint probability distribution, multi-dimensional feature u a The marginal probability distribution of (s) is p(u a (s)), multi-dimensional feature u z The marginal probability distribution of (k) is p(u z (k)), the number of multidimensional features in the multidimensional feature set is N1; The particle with the largest fitness value is taken as the attacker, and the average Euclidean distance between the particle and the attacker is calculated; Select particles whose Euclidean distance to the attacker is greater than the average Euclidean distance, and update the particle position. The expression is: The position of the cth particle in the t+1th iteration is The upper bound of the search space is K max , the lower bound of the search space is K min , the position of the cth particle in the tth iteration is The current iteration number is t, the random number from 0 to 1 is τ1, and the position of the attacker is After updating the particle, calculate the fitness value, divide the attackers twice according to the fitness value, and update the particle position according to the search step of the particle to obtain the mutation position. The expression is: The random numbers from 0 to 1 are τ4 and τ5, and the search step length is The mutation position of the cth particle in the t+1th iteration is The maximum number of iterations is t max ; Until the fitness reaches the maximum, the corresponding multi-dimensional feature set is used as the output result.

6. According to claim 1, a method for fire prevention and control in electric vehicle parking lots based on multi-source information fusion, characterized in that: The heterogeneous data preprocessing module and the multidimensional feature extraction module include: The heterogeneous data preprocessing module adopts the configuration spectrum-spatial domain joint noise reduction algorithm, establishes a noise model based on sensor characteristics, and uses a sliding window mechanism to implement adaptive wavelet denoising on time series data such as temperature / smoke to achieve dynamic threshold noise reduction; Kalman filtering is used to fuse sensor data of the same type to eliminate isolated point errors for multi-source calibration; convolutional neural networks are used to identify interference sources such as device shadows and water vapor in video streams, and false trigger signals are corrected for pattern recognition correction; The multidimensional feature extraction module is used to extract fire-related feature information from the preprocessed data and build a three-dimensional spatiotemporal feature extraction system, which includes time domain features, frequency domain features, spatial correlation features, and cross-modal features; The time domain characteristics calculate the dynamic indicators of battery temperature gradient and second-order derivative of VOCs concentration; the pressure sensor spectrum energy distribution is analyzed by FFT, and the characteristic frequency of micro-detonation inside the battery is detected as a spatial correlation feature; a sensor topology model is constructed to analyze the spatial propagation path of abnormal signals and determine the spatial correlation characteristics; a twin network is used to align the feature space of thermal imaging and voltage fluctuation data as a cross-modal feature.

7. According to claim 1, a method for fire prevention and control in electric vehicle parking lots based on multi-source information fusion, characterized in that: The dynamic control strategy generation module includes: Based on the risk level and feature analysis results output by the fire risk assessment module, dynamic control instructions are generated through the multimodal decision engine; the dynamic control strategy generation module includes hierarchical response strategy and instruction generation logic and dynamic strategy optimization mechanism; The hierarchical response strategy and instruction generation logic generates differentiated control instructions based on the fire risk level output by the three-level assessment and the causal reasoning results of the knowledge graph: the risk levels include low risk, medium risk and high risk; When the risk is low, start the ventilation system in the target area to reduce the concentration of combustible gas; send a battery health check instruction to the BMS to force the charging current to be limited to a safe threshold; push yellow warning information to management personnel through the remote monitoring platform to prompt manual review; when the risk is medium, start the global sound and light alarm, and link the charging pile power off protection; activate the smoke exhaust system preparation mode, calculate the smoke diffusion path in real time and pre-adjust the direction of the smoke exhaust valve; when the risk is high, locate the fire source according to the thermal imaging data, and release fine water mist to suppress the initial fire; if the temperature gradient continues to rise, switch to full coverage fire extinguishing with perfluorohexanone gas; simultaneously cut off the main power supply of the parking lot, and enable the emergency lighting and evacuation guidance system; send three-dimensional thermal maps, vehicle location and battery type data to the city fire command center through the blockchain encrypted channel; A dynamic strategy optimization mechanism is used to establish a control instruction execution effect evaluation model, and the fire extinguishing efficiency is calculated through real-time data from the fire extinguishing system pressure sensor and camera flame recognition results. If the efficiency is lower than the preset threshold, the strategy adjustment is automatically triggered: the historical case library in the knowledge graph is called to match the optimal strategy for similar scenarios; otherwise, the Q-Learning reinforcement learning algorithm is used to dynamically update the control strategy weights to achieve adaptive optimization.

8. A fire prevention and control system for electric vehicle parking lots based on multi-source information fusion, used to execute the method described in any one of claims 1 to 7, characterized in that: include: Data acquisition module: used for the perception layer to capture the environmental status of electric vehicle parking spaces and the operating parameters of vehicle batteries in real time through a multimodal sensor network; Communication module: used for the network layer to adopt intelligent converged network architecture to achieve communication between the perception layer and the data processing layer; Evaluation strategy generation module: used in the data processing layer to perform fusion analysis on the environmental status and the operating parameters, determine the fire risk level, and generate dynamic control instructions; The data processing layer is composed of a multi-dimensional analysis engine, and adopts a hierarchical and progressive intelligent analysis architecture to achieve multi-level analysis and judgment of fire risks; the data processing layer includes a heterogeneous data preprocessing module, a multi-dimensional feature extraction module, a fire risk assessment module and a dynamic control strategy generation module; including: Based on the feature information extracted by the multidimensional feature extraction module, a fire risk assessment model is constructed using a three-level progressive analysis framework to assess the fire risk level of electric vehicle parking lots; the fire risk level includes primary assessment, secondary assessment and tertiary assessment; Positioning and rescue module: used for the application layer to receive the dynamic control instructions generated by the data processing layer, and realize the full-process closed-loop management of fire prevention and control through multimodal terminal equipment; the application layer includes sound and light alarm module, smoke exhaust system control module, fire extinguishing system control module, remote monitoring platform, emergency evacuation and fire extinguishing and rescue module.

Images