A power battery pack safety monitoring system and a monitoring method

Abstract

The application discloses a kind of power battery pack safety monitoring system and monitoring method, belong to battery safety monitoring technical field.The system includes multidimensional sensing acquisition module, data fusion processing module, safety state analysis module, risk prediction and early warning module and communication and display module, by multidimensional parameter such as power battery pack voltage, current, temperature, gas concentration, internal pressure and vibration, real-time acquisition and fusion processing are carried out, construct battery safety state analysis model, realize the comprehensive evaluation and risk prediction early warning of power battery pack operating state.The monitoring method is through multi-parameter acquisition, data fusion processing, safety feature extraction, safety state evaluation and risk trend prediction etc., to the potential safety hazard of power battery pack is identified in advance and is linked control.The application effectively improves the accuracy and timeliness of power battery pack safety monitoring, can reduce the probability of thermal runaway and sudden safety accident, with good application prospect.

Description

Technical Field

[0001] This invention relates to the field of battery safety technology, specifically to a power battery pack safety monitoring system and monitoring method. Background Technology

[0002] With the rapid development of new energy vehicles, the safety of power batteries, as the core energy unit of vehicles, is directly related to the reliability of vehicle operation and the safety of passengers. At present, most mainstream power batteries adopt the form of lithium-ion battery packs for integrated packaging, which has the characteristics of high energy density and strong electrochemical activity, but also has safety hazards such as thermal runaway, internal short circuit, abnormal voltage, and structural damage.

[0003] Current power battery pack safety monitoring technologies mainly rely on battery management systems (BMS), which perform simple threshold judgments by collecting voltage, current, and temperature parameters of individual cells. However, this type of monitoring method has the following shortcomings: The monitoring parameters are limited: it only focuses on basic electrical parameters and lacks awareness of hidden risks such as gas leaks, internal pressure changes, vibration and shock. Strong response lag: Most of them adopt post-event alarm mechanisms, making it difficult to make effective predictions in the early stages of thermal runaway; Insufficient data fusion capabilities: Failure to perform correlation analysis on multi-source sensor data results in low accuracy of risk assessment; Lack of intelligent early warning mechanism: unable to dynamically predict battery aging trends and abnormal evolution.

[0004] Therefore, there is an urgent need for a power battery pack safety monitoring system and method with multi-dimensional perception, intelligent analysis and real-time early warning capabilities, so as to achieve comprehensive monitoring and risk prevention of the power battery pack's operating status. Summary of the Invention

[0005] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a power battery pack safety monitoring system and method, solving the following problems: Limited monitoring parameters: focusing only on basic electrical parameters and lacking awareness of hidden risks such as gas leaks, internal pressure changes, vibration, and shock; Strong response lag: most systems rely on post-event alarm mechanisms, making it difficult to effectively predict thermal runaway in its early stages; Insufficient data fusion capabilities: failing to perform correlation analysis on multi-source sensor data, resulting in low accuracy in risk assessment; Lack of intelligent early warning mechanisms: unable to dynamically predict battery aging trends and abnormal evolution.

[0006] (II) Technical Solution The purpose of this invention is to provide a power battery pack safety monitoring system and method. Through multi-sensor collaborative data acquisition, intelligent battery status analysis, and dynamic risk level assessment, it enables early warning of potential safety hazards in the power battery pack, effectively reducing the probability of thermal runaway and sudden safety accidents. This is achieved through the following technical solution: A power battery pack safety monitoring system, characterized in that it includes: A multi-dimensional sensing and acquisition module is used to acquire the operating status parameters of the power battery pack in real time. The operating status parameters include at least voltage parameters, current parameters, temperature parameters, gas concentration parameters, internal pressure parameters, and vibration parameters. The data fusion processing module is connected to the multi-dimensional sensing acquisition module and is used to perform synchronous processing, filtering processing and feature fusion on the acquired multi-source operating status data to form a unified status dataset. The safety status analysis module is connected to the data fusion and processing module and is used to build a battery safety status analysis model based on the status dataset to evaluate the safety level of the power battery pack's operating status. The risk prediction and early warning module is connected to the safety status analysis module. It is used to predict the development trend of safety risks of the power battery pack based on historical operating data and current safety status assessment results, and to trigger an early warning signal when the prediction result reaches a preset threshold. The communication and display module is connected to the risk prediction and early warning module and is used to send the monitoring results to the vehicle control system or remote monitoring platform and display them visually.

[0007] As a further preferred embodiment of the present invention, the multidimensional sensing acquisition module includes: Voltage acquisition unit, current acquisition unit, multiple temperature acquisition units, gas detection unit, pressure detection unit and vibration detection unit.

[0008] As a further preferred embodiment of the present invention, the data fusion processing module uses a time synchronization mechanism to align data collected by different sensors and uses an adaptive filtering algorithm to remove noise data.

[0009] As a further preferred embodiment of the present invention, the safety status analysis module includes a feature extraction unit and a status discrimination unit. The feature extraction unit is used to extract temperature change rate, voltage deviation rate, gas generation rate and pressure change rate as safety feature parameters.

[0010] As a further preferred embodiment of the present invention, the safety status analysis module performs safety status assessment based on a hybrid analysis method combining rule models and machine learning models, and evaluates the safety level of the power battery pack's operating status. The specific implementation steps are as follows: First, the collected parameters of temperature change rate, voltage deviation rate, gas generation rate, pressure change rate, and vibration intensity are standardized to convert parameters of different dimensions to a unified numerical range. The standardization process uses the following formula: X′= (X - Xmin) / (Xmax - Xmin) Where X is the original value of the current safety feature parameter, Xmin and Xmax are the minimum and maximum values ​​of the parameter under historical safe operating conditions, respectively, and X′ is the standardized feature parameter value; Secondly, the corresponding weighting coefficients are determined based on the degree of influence of each safety characteristic parameter on the safety risk of the power battery pack. These weighting coefficients are obtained through statistical analysis of historical operating data or the analytic hierarchy process (AHP), and include weights for temperature changes, voltage deviations, gas changes, pressure changes, and vibration effects. The sum of these weights satisfies the following: w1 + w2 + w3 + w4 + w5 = 1 Then, a comprehensive safety risk assessment model for the power battery pack is constructed. The weighted summation of each safety characteristic parameter yields the comprehensive safety risk score under the current operating state. The calculation formula is as follows: R = w1·T +w2·V +w3·G +w4·P + w5·S in: T is the standardized characteristic value of the rate of temperature change; V is the standardized voltage deviation rate characteristic value; G is the standardized characteristic value of the gas generation rate; P is the standardized characteristic value of the rate of change of pressure; S is the standardized characteristic value of vibration intensity; R represents the current comprehensive safety risk score of the power battery pack. Subsequently, based on the comprehensive security risk score R, multiple security level determination ranges were set: When R is less than the first safety threshold, the power battery pack is determined to be in a normal safety state. When R is between the first safety threshold and the second safety threshold, the power battery pack is determined to be in a warning state. When R is greater than or equal to the second safety threshold, the power battery pack is determined to be in a dangerous state.

[0011] As a further preferred embodiment of the present invention, the risk prediction and early warning module uses a time series prediction algorithm to predict and analyze the changing trends of safety characteristics and the development trend of safety risks in the power battery pack. The specific implementation steps are as follows: First, the comprehensive safety risk scores output by the safety status analysis module during continuous operation cycles are used to construct a time-series dataset in chronological order: R(1), R(2), ..., R(n) Where R(t) represents the comprehensive safety risk score of the power battery pack at time t; Secondly, stationarity analysis and trend feature extraction are performed on the risk time series, and continuous historical risk data are selected as prediction input samples by using a sliding time window method. A time series prediction model is used to model and analyze the changing trend of safety risks. The prediction model includes an autoregressive moving average model, a grey prediction model, or a neural network prediction model. Preferably, a long short-term memory neural network model is used to learn the nonlinear characteristics of risk changes. During the forecasting process, continuous historical risk scores are input into the forecasting model to obtain the risk forecast values ​​for the future forecast period: R(t + k) Where k is the prediction time step. Subsequently, the predicted future risk values ​​are compared and analyzed with preset risk thresholds. When the following conditions are met, the risk development trend is determined to enter an early warning state: When R(t + k) is greater than the first warning threshold, it is determined to be an upward risk trend and a level one warning is triggered; When R(t + k) is greater than or equal to the second danger threshold, a high-risk trend is identified and a level-two warning is triggered. The first warning threshold and the second danger threshold are determined based on statistical analysis of historical operational data. To improve the stability and accuracy of trend prediction, the risk prediction and early warning module simultaneously smooths the risk time series, eliminating short-term fluctuations through a moving average algorithm. The calculation formula is as follows: R(t) = [R(t) + R(t - 1) + …… + R(t - m + 1)] / m Where R(t) is the smoothed risk score, and m is the smoothing window length.

[0012] As a further preferred embodiment of the present invention, the risk prediction and early warning module sends a safety control command to the vehicle control system at the same time as triggering the early warning signal.

[0013] As a further preferred embodiment of the present invention, the safety control command includes at least one of the following: power-limiting operation command, battery power-off protection command, and auxiliary heat dissipation start command.

[0014] As a further preferred embodiment of the present invention, the communication and display module supports vehicle terminal display and cloud-based remote monitoring, and the temperature acquisition unit of the multi-dimensional sensing acquisition module is located between the individual battery cells and in the middle of the battery pack casing.

[0015] As a further preferred embodiment of the present invention, the method for monitoring the safety of a power battery pack in a power battery pack safety monitoring system is characterized by comprising the following steps: S1. Real-time acquisition of voltage, current, temperature, gas concentration, internal pressure and vibration data of the power battery pack; S2. Preprocess and fuse the collected data to form a unified state dataset; S3. Extract safety characteristic parameters and conduct a safety status assessment; S4. Predict and analyze the development trend of security risks; S5. When the prediction result reaches the warning threshold, an early warning is triggered and safety linkage control is executed.

[0016] (III) Beneficial Effects This invention provides a power battery pack safety monitoring system and method, which has the following beneficial effects: By employing multiple sensing units working collaboratively to detect voltage, current, temperature, gas concentration, internal pressure, and vibration, this system overcomes the limitations of traditional methods that rely solely on electrical parameter monitoring. It achieves comprehensive perception of the physical and chemical states of the power battery pack, thus expanding the coverage of safety monitoring. Furthermore, the data fusion processing module synchronously processes and fuses features from multiple data sources, effectively eliminating misjudgments caused by fluctuations in data from a single sensor, making the safety status assessment results more reliable.

[0017] The risk prediction and early warning module analyzes the changing trends of safety characteristics based on historical operating data, and can issue early warnings before thermal runaway or serious failures occur, realizing a shift from post-event alarm to pre-event prevention. When a high-risk state is detected, it automatically sends power limiting, power-off, or auxiliary cooling commands to the vehicle control system to achieve proactive safety control and prevent further escalation of the accident. The system supports remote monitoring via onboard terminals and the cloud, and is applicable to various application scenarios such as new energy vehicles, energy storage equipment, and battery swapping facilities, possessing excellent scalability. Attached Figure Description

[0018] Figure 1 : A schematic diagram of the system principle framework of the present invention.

[0019] Figure 2 : Schematic diagram of the monitoring method of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1-2 This invention provides a technical solution: a power battery pack safety monitoring system, comprising: a multi-dimensional sensing acquisition module for real-time acquisition of power battery pack operating status parameters, including: The voltage acquisition unit is used to obtain the voltage of each individual battery cell and the total voltage of the battery pack; through the combined deployment of multiple types of sensors, a comprehensive perception of the multi-dimensional safety status of the power battery pack can be achieved.

[0022] The current acquisition unit is used to acquire charging and discharging current data; Temperature acquisition unit is used to monitor the temperature distribution on the surface and in the middle area inside the battery; Gas detection unit is used to detect the concentration of combustible gases inside the battery pack; The pressure detection unit is used to monitor changes in the internal expansion pressure of the battery pack; Vibration detection unit is used to sense external impacts and abnormal structural vibrations.

[0023] Data fusion processing module It connects to a multi-dimensional sensing acquisition module to synchronously process, filter, and extract features from the acquired multi-source data, forming a unified state dataset.

[0024] Security Status Analysis Module A battery safety status model is established based on fused data. The temperature rise rate, voltage deviation rate, gas concentration change rate and pressure anomaly are comprehensively analyzed to determine the battery pack's operational safety level.

[0025] Specific implementation method of safety status analysis module for power battery pack safety level assessment The safety status analysis module assesses the safety level of the power battery pack's operating status based on the status dataset output by the data fusion processing module. The specific implementation steps are as follows: First, the collected parameters of temperature change rate, voltage deviation rate, gas generation rate, pressure change rate, and vibration intensity are standardized to convert parameters with different dimensions to a unified numerical range. The standardization process uses the following formula: X′ = (X - Xmin) / (Xmax - Xmin) Where X is the original value of the current safety feature parameter, Xmin and Xmax are the minimum and maximum values ​​of the parameter under historical safe operating conditions, respectively, and X′ is the standardized feature parameter value.

[0026] Secondly, the corresponding weighting coefficients are determined based on the degree of influence of each safety characteristic parameter on the safety risk of the power battery pack. These weighting coefficients are obtained through statistical analysis of historical operating data or the analytic hierarchy process (AHP), and include weights for temperature changes, voltage deviations, gas changes, pressure changes, and vibration effects. The sum of these weights satisfies the following: w1 + w2 + w3 + w4 + w5 = 1 Then, a comprehensive safety risk assessment model for the power battery pack is constructed. The weighted summation of each safety characteristic parameter yields the comprehensive safety risk score under the current operating state. The calculation formula is as follows: R = w1·T +w2·V +w3·G +w4·P + w5·S in: T is the standardized characteristic value of the rate of temperature change; V is the standardized voltage deviation rate characteristic value; G is the standardized characteristic value of the gas generation rate; P is the standardized characteristic value of the rate of change of pressure; S is the standardized characteristic value of vibration intensity; R represents the current comprehensive safety risk score of the power battery pack.

[0027] Subsequently, based on the comprehensive security risk score R, multiple security level determination ranges were set: When R is less than the first safety threshold, the power battery pack is determined to be in a normal safety state. When R is between the first safety threshold and the second safety threshold, the power battery pack is determined to be in a warning state. When R is greater than or equal to the second safety threshold, the power battery pack is determined to be in a dangerous state.

[0028] The first and second safety thresholds are determined through statistical analysis of historical operational data and accident samples. To improve the accuracy and robustness of the safety level assessment, the safety status analysis module also incorporates a machine learning classification model to assist in correcting the evaluation results of the aforementioned rule-based model. The machine learning model takes each safety feature parameter as input and the safety level as output, and is trained using a random forest model, support vector machine model, or neural network model. When there is a discrepancy between the output results of the rule-based model and the machine learning model, a weighted fusion method is used to obtain the final safety level assessment result.

[0029] Risk prediction and early warning module Intelligent algorithms trained on historical data predict battery operating trends, and trigger warning signals when the prediction results exceed preset safety thresholds.

[0030] The risk prediction and early warning module, based on the comprehensive safety risk score and various safety characteristic parameters output by the safety status analysis module, predicts and analyzes the development trend of safety risks in the power battery pack. Its specific implementation steps are as follows: First, the comprehensive safety risk scores output by the safety status analysis module during continuous operation cycles are used to construct a time-series dataset in chronological order: R(1), R(2), ..., R(n) Where R(t) represents the comprehensive safety risk score of the power battery pack at time t.

[0031] Secondly, stationarity analysis and trend feature extraction are performed on the risk time series, and continuous historical risk data are selected as prediction input samples by using a sliding time window method.

[0032] A time series prediction model is used to model and analyze the changing trends of safety risks. The prediction model includes an autoregressive moving average model, a grey prediction model, or a neural network prediction model. Preferably, a long short-term memory neural network model is used to learn the nonlinear characteristics of risk changes.

[0033] During the forecasting process, continuous historical risk scores are input into the forecasting model to obtain the risk forecast values ​​for the future forecast period: R(t + k) Where k is the prediction time step.

[0034] Subsequently, the predicted future risk values ​​are compared and analyzed with preset risk thresholds. When the following conditions are met, the risk development trend is determined to enter an early warning state: When R(t + k) is greater than the first warning threshold, it is determined to be an upward risk trend and a level one warning is triggered; When R(t + k) is greater than or equal to the second danger threshold, it is judged as a high-risk trend and a level 2 warning is triggered.

[0035] The first warning threshold and the second danger threshold are determined based on statistical analysis of historical operating data.

[0036] To improve the stability and accuracy of trend prediction, the risk prediction and early warning module simultaneously smooths the risk time series, eliminating short-term fluctuations through a moving average algorithm. The calculation formula is as follows: R(t) = [R(t) + R(t-1) + …… + R(tm + 1)] / m Where R(t) is the smoothed risk score, and m is the smoothing window length.

[0037] Finally, the risk prediction and early warning module combines the smoothing results with the prediction model output to determine the future risk development trend of the power battery pack. When the predicted risk reaches the corresponding threshold, it triggers an audible and visual alarm signal and sends a safety linkage control command to the vehicle control system.

[0038] Communication and display module It is used to upload monitoring results to the vehicle control system or cloud platform in real time, and to display them visually through the vehicle terminal or mobile device.

[0039] The multi-dimensional sensing acquisition module is used to collect intermediate safety parameters in real time during the operation of the power battery pack. Among them, the voltage parameter reflects the electrical energy status of individual cells and battery packs, the current parameter reflects the changes in charging and discharging conditions, the temperature parameter is used to monitor the thermal status of the battery, the gas concentration parameter is used to sense the combustible gas generated by the decomposition or leakage of electrolyte inside the battery, the internal pressure parameter is used to monitor the battery expansion, and the vibration parameter is used to sense external impact or structural abnormalities.

[0040] The data fusion processing module aligns various types of sensor data through a time synchronization mechanism and eliminates noise interference through a filtering algorithm, so that data from different sources form a unified state dataset, providing a reliable foundation for subsequent security analysis.

[0041] The safety status analysis module constructs a battery safety status analysis model based on the status dataset. By comprehensively evaluating multi-dimensional safety characteristic parameters, it divides the operating status of the power battery pack into normal status, warning status, and dangerous status, thereby achieving refined safety management.

[0042] The risk prediction and early warning module predicts and analyzes potential risks based on historical operational data and current safety status trends, and triggers early warning signals before the risks reach preset thresholds to achieve early prevention and control.

[0043] The communication and display module is used to transmit monitoring results to the vehicle control system or cloud platform and display them graphically, so that drivers or managers can keep track of the battery safety status in real time.

[0044] A method for safety monitoring of a power battery pack includes the following steps: S1: Real-time acquisition of multiple parameters The multi-dimensional sensing acquisition module acquires real-time data on battery voltage, current, temperature, gas concentration, internal pressure, and vibration.

[0045] S2: Data Preprocessing and Fusion The collected data is denoised, synchronized with time, and normalized, and a unified state data matrix is ​​constructed.

[0046] S3: Security Feature Extraction Intermediate safety feature parameters extracted from the fused data include temperature gradient, voltage deviation, gas generation rate, and pressure change rate.

[0047] S4: Safety Status Assessment The extracted features are comprehensively evaluated using a safety status analysis model, and the corresponding safety level is output, such as normal status, warning status, and dangerous status.

[0048] S5: Risk Trend Forecasting By combining historical operational data, we can predict and analyze the current trends in safety characteristics to determine whether there is a risk of thermal runaway.

[0049] S6: Early Warning and Linkage Control When the predicted results meet the warning conditions, an audible and visual alarm is triggered, and a control command is sent to the vehicle control system to achieve power-limited operation, power-off protection, or emergency cooling. The monitoring method constructs a complete safety monitoring closed loop through steps such as real-time acquisition of multiple parameters, data fusion processing, safety feature extraction, safety status assessment, and risk trend prediction, enabling continuous monitoring and early prevention of safety risks to the power battery pack.

[0050] Example 1: Real-time Safety Monitoring of Multi-dimensional Parameters of Vehicle Power Battery Pack This embodiment uses a new energy vehicle power battery pack as the application object, and arranges a multi-dimensional sensing and acquisition module in the middle of the battery pack's interior and exterior. Voltage acquisition units are set on both sides of each battery cell, connected to a data acquisition interface via wires, to acquire the working voltage of each individual battery cell in real time. A current acquisition unit is set in the main circuit of the battery pack to monitor the overall charging and discharging current changes. Multiple temperature acquisition units are set in the gaps between battery cells and on the inner wall of the battery pack shell, forming a multi-point temperature monitoring network to achieve real-time perception of the internal temperature field distribution of the battery pack. A gas detection unit is set in the ventilation channel at the top inside the battery pack to detect changes in the concentration of combustible gases generated by electrolyte decomposition; a pressure detection unit is set in the reserved space inside the battery pack to sense internal pressure changes caused by thermal expansion or gas accumulation; a vibration detection unit is installed at the middle support position outside the battery pack shell to monitor vibrations and collision impacts generated during vehicle operation. All sensing units are connected to a data fusion processing module via a bus. The data fusion processing module uses a time synchronization mechanism to uniformly sample various types of data and uses an adaptive filtering algorithm to eliminate noise interference. The safety status analysis module processes the fused data, extracting temperature change rate, voltage deviation rate, gas generation rate, and pressure change rate as safety characteristic parameters. Based on a preset safety model, it determines whether the battery pack is currently in a normal or warning state. This embodiment achieves multi-dimensional real-time perception of the power battery pack's operating status, providing fundamental data support for subsequent risk prediction and coordinated control.

[0051] Example 2: Security Protection Example Based on Risk Prediction and Linkage Control Building upon Example 1, this example further introduces a risk prediction and early warning module and a safety linkage control mechanism. During system operation, the risk prediction and early warning module performs time-series modeling on the safety characteristic parameters output by the safety status analysis module, forming a battery safety risk trend prediction model. When the system detects a continuous rise in temperature in a certain battery area and a gas concentration change rate significantly higher than historical normal values, the risk prediction and early warning module predicts that this trend will develop into a dangerous state within a preset time window. When the prediction result reaches the early warning threshold, the system automatically triggers an audible and visual alarm device and simultaneously sends a safety control command to the vehicle control system. Specifically, when the predicted risk level is medium, the vehicle control system executes a power-limiting operation command, reducing charging and discharging power to reduce heat generation; when the predicted risk level reaches high, a battery power-off protection command is executed and the auxiliary cooling system is activated to forcibly cool the battery pack. Through this example, a closed-loop safety protection mechanism from risk prediction to active intervention is realized, effectively preventing the occurrence of thermal runaway accidents.

[0052] Example 3: Cloud-based Remote Monitoring and Big Data Security Analysis Example This embodiment, building upon the previous embodiments, connects the communication and display module to a cloud monitoring platform to achieve remote centralized management of the power battery pack's safety status. During vehicle operation, the communication and display module uploads data collected by the multi-dimensional sensing acquisition module and safety status analysis results to the cloud server in real time. The cloud platform centrally stores and analyzes the operating data of power battery packs from multiple new energy vehicles, establishing a battery safety behavior model through big data analysis algorithms. When the cloud platform detects high-frequency abnormal characteristics in a certain type of battery pack under specific operating conditions, it issues optimized operating strategies or safety warning commands to the corresponding vehicle. Simultaneously, the cloud platform uses long-term operating data to optimize the parameters of the risk prediction model, enabling the onboard risk prediction and warning module to achieve higher prediction accuracy. Through this embodiment, the safety monitoring of the power battery pack is upgraded from independent operation of a single vehicle to vehicle-cloud collaborative intelligent management, improving the overall safety protection level.

[0053] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or basic characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the same elements of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0054] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A power battery pack safety monitoring system, characterized in that: include: A multi-dimensional sensing and acquisition module is used to acquire the operating status parameters of the power battery pack in real time. The operating status parameters include at least voltage parameters, current parameters, temperature parameters, gas concentration parameters, internal pressure parameters, and vibration parameters. The data fusion processing module is connected to the multi-dimensional sensing acquisition module and is used to perform synchronous processing, filtering processing and feature fusion on the acquired multi-source operating status data to form a unified status dataset. The safety status analysis module is connected to the data fusion and processing module and is used to build a battery safety status analysis model based on the status dataset to evaluate the safety level of the power battery pack's operating status. The risk prediction and early warning module is connected to the safety status analysis module. It is used to predict the development trend of safety risks of the power battery pack based on historical operating data and current safety status assessment results, and to trigger an early warning signal when the prediction result reaches a preset threshold. The communication and display module is connected to the risk prediction and early warning module and is used to send the monitoring results to the vehicle control system or remote monitoring platform and display them visually.

2. The power battery pack safety monitoring system according to claim 1, characterized in that: The multi-dimensional sensing acquisition module includes: Voltage acquisition unit, current acquisition unit, multiple temperature acquisition units, gas detection unit, pressure detection unit and vibration detection unit.

3. The power battery pack safety monitoring system according to claim 1, characterized in that: The data fusion processing module uses a time synchronization mechanism to align data collected by different sensors and an adaptive filtering algorithm to remove noise data.

4. The power battery pack safety monitoring system according to claim 1, characterized in that: The safety status analysis module includes a feature extraction unit and a status discrimination unit. The feature extraction unit is used to extract temperature change rate, voltage deviation rate, gas generation rate and pressure change rate as safety feature parameters.

5. The power battery pack safety monitoring system according to claim 1, characterized in that: The safety status analysis module performs safety status assessment based on a hybrid analysis method combining rule-based models and machine learning models, and evaluates the safety level of the power battery pack's operating status. The specific implementation steps are as follows: First, the collected parameters of temperature change rate, voltage deviation rate, gas generation rate, pressure change rate, and vibration intensity are standardized to convert parameters of different dimensions to a unified numerical range. The standardization process uses the following formula: X′ = (X - Xmin) / (Xmax - Xmin) Where X is the original value of the current safety feature parameter, Xmin and Xmax are the minimum and maximum values ​​of the parameter under historical safe operating conditions, respectively, and X′ is the standardized feature parameter value; Secondly, the corresponding weighting coefficients are determined based on the degree of influence of each safety characteristic parameter on the safety risk of the power battery pack. These weighting coefficients are obtained through statistical analysis of historical operating data or the analytic hierarchy process (AHP), and include weights for temperature changes, voltage deviations, gas changes, pressure changes, and vibration effects. The sum of these weights satisfies the following: w1 + w2 + w3 + w4 + w5 = 1 Then, a comprehensive safety risk assessment model for the power battery pack is constructed. The weighted summation of each safety characteristic parameter yields the comprehensive safety risk score under the current operating state. The calculation formula is as follows: R = w1·T +w2·V +w3·G +w4·P + w5·S in: T is the standardized characteristic value of the rate of temperature change; V is the standardized voltage deviation rate characteristic value; G is the standardized characteristic value of the gas generation rate; P is the standardized characteristic value of the rate of change of pressure; S is the standardized characteristic value of vibration intensity; R represents the current comprehensive safety risk score of the power battery pack. Subsequently, based on the comprehensive security risk score R, multiple security level determination ranges were set: When R is less than the first safety threshold, the power battery pack is determined to be in a normal safety state. When R is between the first safety threshold and the second safety threshold, the power battery pack is determined to be in a warning state. When R is greater than or equal to the second safety threshold, the power battery pack is determined to be in a dangerous state.

6. The power battery pack safety monitoring system according to claim 1, characterized in that: The risk prediction and early warning module uses a time series prediction algorithm to predict and analyze the changing trends of safety characteristics and the development trend of safety risks in the power battery pack. Its specific implementation steps are as follows: First, the comprehensive safety risk scores output by the safety status analysis module during continuous operation cycles are used to construct a time-series dataset in chronological order: R(1), R(2), ..., R(n) Where R(t) represents the comprehensive safety risk score of the power battery pack at time t; Secondly, stationarity analysis and trend feature extraction are performed on the risk time series, and continuous historical risk data are selected as prediction input samples by using a sliding time window method. A time series prediction model is used to model and analyze the changing trend of safety risks. The prediction model includes an autoregressive moving average model, a grey prediction model, or a neural network prediction model. Preferably, a long short-term memory neural network model is used to learn the nonlinear characteristics of risk changes. During the forecasting process, continuous historical risk scores are input into the forecasting model to obtain the risk forecast values ​​for the future forecast period: R(t + k) Where k is the prediction time step. Subsequently, the predicted future risk values ​​are compared and analyzed with preset risk thresholds. When the following conditions are met, the risk development trend is determined to enter an early warning state: When R(t + k) is greater than the first warning threshold, it is determined to be an upward risk trend and a level one warning is triggered; When R(t + k) is greater than or equal to the second danger threshold, a high-risk trend is identified and a level-two warning is triggered. The first warning threshold and the second danger threshold are determined based on statistical analysis of historical operational data. To improve the stability and accuracy of trend prediction, the risk prediction and early warning module simultaneously smooths the risk time series, eliminating short-term fluctuations through a moving average algorithm. The calculation formula is as follows: R(t) = [R(t) + R(t-1) + …… + R(tm + 1)] / m Where R(t) is the smoothed risk score, and m is the smoothing window length.

7. The power battery pack safety monitoring system according to claim 1, characterized in that: The risk prediction and early warning module sends a safety control command to the vehicle control system at the same time as triggering the early warning signal.

8. The power battery pack safety monitoring system according to claim 1, characterized in that: The safety control commands include at least one of the following: power-limiting operation command, battery power-off protection command, and auxiliary heat dissipation start command.

9. The power battery pack safety monitoring system according to claim 1, characterized in that: The communication and display module supports in-vehicle terminal display and cloud-based remote monitoring. The temperature acquisition unit of the multi-dimensional sensing acquisition module is located between the individual battery cells and in the middle of the battery pack casing.

10. A method for monitoring the safety of a power battery pack based on the power battery pack safety monitoring system according to any one of claims 1 to 10, characterized in that, Includes the following steps: S1. Real-time acquisition of voltage, current, temperature, gas concentration, internal pressure and vibration data of the power battery pack; S2. Preprocess and fuse the collected data to form a unified state dataset; S3. Extract safety characteristic parameters and conduct a safety status assessment; S4. Predict and analyze the development trend of security risks; S5. When the prediction result reaches the warning threshold, an early warning is triggered and safety linkage control is executed.

Images