Power battery state real-time monitoring system

Abstract

The invention discloses a power battery state real-time monitoring system, and relates to the technical field of new energy automobiles, and the system comprises the following components: sensor network construction: installing a voltage sensor, a temperature sensor, a current sensor and an internal resistance monitoring sensor on each single battery of a power battery pack to form an omnibearing and high-density monitoring network; according to the integrated fault diagnosis technology, the comprehensive and high-density sensor network is constructed, key parameters of voltage, temperature, current and internal resistance of each single battery in the battery pack are collected in real time, comprehensiveness and accuracy of data are ensured, multiple advanced fault diagnosis algorithms are combined, and the fault diagnosis accuracy of the battery pack is improved. According to the method, characteristics can be rapidly extracted from mass data according to a battery comprehensive fault identification index formula, a battery physical model prediction index and an abnormal value detection index, and pattern identification is carried out, so that accurate judgment of a fault type of a single battery and accurate positioning of a fault position are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and in particular to a real-time monitoring system for power battery status. Background Art

[0002] With the rapid development of the new energy vehicle industry, power batteries, as core components of electric vehicles, have a direct impact on the vehicle's endurance and driving safety through their performance and safety. Therefore, real-time monitoring of the power battery status and timely detection and handling of potential faults have become the key to ensuring the safe operation of electric vehicles. Traditional battery monitoring technology often relies on the monitoring of a few parameters, which makes it difficult to fully reflect the health status of the battery pack. In addition, the efficiency and accuracy of fault diagnosis and isolation are limited, and cannot meet the high safety and high reliability requirements of modern electric vehicles.

[0003] In traditional technology, fault diagnosis and isolation of power batteries often rely on manual inspections and simple sensor monitoring. This method is not only inefficient, but also prone to missing potential faults. At the same time, due to the lack of integrated fault diagnosis and isolation technology, it is difficult for the system to achieve comprehensive analysis of multiple sensor data, and it is impossible to accurately and quickly identify the fault type and location of the single cell. In addition, when faced with complex and changeable working conditions, traditional systems are often unable to take timely and effective isolation measures to prevent the spread of faults, thereby increasing the risk of vehicle safety accidents.

[0004] Therefore, in response to the above problems, a real-time monitoring system for power battery status is developed. Integrated fault diagnosis and isolation technology not only improves the accuracy and efficiency of fault diagnosis, but also enhances the safety and reliability of electric vehicles, providing strong support for the sustainable development of the new energy vehicle industry. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and provide a real-time monitoring system for the status of a power battery. It can realize real-time monitoring of the status of each single cell in the power battery pack by building a comprehensive sensor network and combining advanced algorithms and data processing technologies. The system can quickly and accurately identify various fault types of single cells and their specific locations, and automatically trigger an isolation mechanism to effectively prevent the spread of faults and ensure the safe operation of the vehicle. In addition, the system also supports remote monitoring and maintenance functions, which is convenient for manufacturers and service providers to perform fault warnings and preventive maintenance, thereby improving the reliability and service life of the power battery.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a real-time monitoring system for power battery status, the system comprising the following components:

[0007] Sensor network construction: Install voltage sensors, temperature sensors, current sensors and internal resistance monitoring sensors on each single cell of the power battery pack to form an all-round, high-density monitoring network;

[0008] Data acquisition and processing module: The sensor data is aggregated to the central processing unit through wireless transmission, and the collected data is pre-processed and feature extracted using algorithm formulas;

[0009] Integrated fault diagnosis module: Integrates multiple fault diagnosis algorithms, comprehensively analyzes sensor data, and quickly and accurately identifies various fault types and specific locations of single cells;

[0010] Fault isolation and emergency response module: Once a fault is diagnosed, the system automatically triggers the isolation mechanism, quickly disconnecting the faulty single cell from the rest of the battery pack through intelligent relays and electronic switches to prevent the fault from spreading. At the same time, an alarm signal is sent to the vehicle control system to initiate an emergency plan.

[0011] Remote monitoring and maintenance: The system supports remote data transmission function, which can upload monitoring data and fault diagnosis reports to the cloud server in real time, making it convenient for manufacturers and service providers to conduct remote monitoring, fault warning and preventive maintenance.

[0012] Furthermore, the integrated fault diagnosis module extracts key feature vectors that can reflect the battery status from the preprocessed data, integrates multiple fault diagnosis algorithm formulas, uses the battery fault identification index to identify the fault type and location of the single cell, establishes a battery physical model according to the electrochemical, thermal and mechanical characteristics of the battery, and predicts the behavior of the battery under different working conditions through simulation, compares it with real-time data, predicts potential fault types, detects outliers in the feature vectors through outlier detection indicators, identifies abnormal conditions that deviate from the normal range, and uses them as the basis for fault diagnosis. Combined with the output results of multiple algorithms, the fault type and specific location of the single cell are determined through formulas, and the diagnosis results are organized into a diagnosis report.

[0013] Furthermore, the integrated fault diagnosis module uses the battery fault identification index to identify the fault type and location of the single cell. Assuming that there are n single cells in the battery pack, for the i-th single cell, the voltage characteristic value V i Indicates the real-time voltage and current characteristic value I of the single battery i Indicates the real-time current flowing through the single cell, the temperature characteristic value T i Indicates the real-time temperature and internal resistance characteristic value of the single battery Indicates the internal resistance of the single battery, and the battery fault identification index is F i , the calculation formula is: Among them, are the average values of the voltages, currents, temperatures, and internal resistances of all single cells respectively, and σ V、 σ I、 σ T、 σ R are the corresponding standard deviations respectively, and α, β, γ, δ are weight coefficients. For the judgment of fault types, different threshold intervals are set. F i When < t1, it is judged as the normal state, t1 ≤ F i When < t2, it is judged as the minor fault type 1, t2 ≤ F i When < t3, it is judged as the moderate fault type 2, F i When ≥ t3, it is judged as the severe fault type 3. According to the serial number of the single cell with the largest F i value, the fault location is determined, and the fault location is the j-th single cell.

[0014] Furthermore, the specific steps for the integrated fault diagnosis module to predict the battery performance and identify potential faults by constructing a battery physical model combined with real-time data are as follows:

[0015] (1) Construct electrochemical, thermal, and mechanical models to describe the internal electrochemical reactions, thermal behaviors, and mechanical stress deformations of the battery respectively, covering factors such as lithium-ion migration, heat conduction, and volume change;

[0016] (2) Collect performance data under different working conditions through experiments, and use the performance data to calibrate the constructed physical model;

[0017] (3) Design a data interface to access real-time data and preprocess it, and input the preprocessed real-time data into the calibrated physical model to predict the performance of the battery in the future for a period of time;

[0018] (4) Set a fault threshold, compare the model prediction result with the preset normal range, and identify potential fault modes;

[0019] (5) Organize the identified potential fault types, locations, severities, and predicted development trends into a report.

[0020] Furthermore, the integrated fault diagnosis module detects the outliers in the feature vector through outlier detection indicators. Suppose there are n single cells in the battery pack, and the feature vector of each single cell is where m is the number of features, and calculate the average value of the feature vectors of all single cells as the central feature vector Among them, a dispersion matrix D is constructed. Among them, for the feature vector of the i-th single cell Define the outlier detection indicator as A i , and the calculation formula is: Set the abnormal threshold T, when A i >T, the ith single cell is considered to be in an abnormal state, and the abnormal position is the position of the ith single cell in the battery pack.

[0021] Furthermore, the integrated fault diagnosis module determines the fault type of the single battery through a fault judgment formula. Assuming that the output based on the battery fault identification index is A, the output of the fault prediction algorithm based on the battery physical model is B, and the output based on the abnormal value detection index is C, the fault judgment formula is defined as F, and the calculation formula is: Among them, W A ,W B ,W C are the weight vectors output by the corresponding algorithms respectively, α, β, and γ are the weight coefficients of the three algorithms, satisfying α+β+γ=1, ∥·∥ represents the modulus of the vector, and the fault type of the single cell is judged to be the fault type represented by the corresponding subscript with the largest value in F.

[0022] Furthermore, the integrated fault diagnosis module determines the specific location of the single battery fault through the location judgment formula, and the probability distribution of each location fault given by the battery fault identification index is assumed to be a vector M = (m1, m2, ..., m k ), where k is the total number of possible positions, m i represents the probability of a fault occurring at the i-th position. The fault index value of each position given by the fault prediction algorithm based on the battery physical model is the vector P = (p1, p2, ..., p k ), which can reflect the fault possibility of each location. The abnormality degree value of each location given by the abnormal value detection index is a vector S = (s1, s2, ..., s k ), define the position judgment formula as G, and the calculation formula is: Among them, λ M , P , S are weight adjustment factors of the output results of the three algorithms respectively. The specific fault location of the single cell is determined as the position subscript that makes the G value maximum. When i=j, G obtains the maximum value, and the specific fault location of the single cell is determined to be the jth position.

[0023] Furthermore, after receiving the fault report from the fault diagnosis module, the fault isolation and emergency response module automatically triggers the isolation mechanism, and quickly and accurately cuts off the electrical connection between the faulty single cell and other parts of the battery pack by controlling the intelligent relay and the electronic switch to prevent the fault from spreading. At the same time, when the fault isolation operation is performed, an alarm signal containing information on the fault type, location and severity is sent to the vehicle control system. After receiving the alarm signal, an appropriate emergency plan is selected through an evaluation formula, and the corresponding emergency measures are initiated after confirmation by the driver. The fault data is recorded and a fault notification is sent to the owner and the service center. After the fault isolation and emergency response operations are completed, the battery pack status is re-tested to verify whether the fault is effectively isolated and whether the emergency response measures are appropriate, and feedback information is collected for system optimization and performance improvement.

[0024] Furthermore, the fault isolation and emergency response module selects a suitable emergency plan through an evaluation formula and defines evaluation parameters, S i It represents the fault isolation speed, which is the time from detecting the fault to completely isolating the faulty single cell. i Indicates the degree of risk reduction, P i Indicates the degree of protection of vehicle performance, C i Represents the cost factor and constructs the evaluation formula: E i =αS i +βR i +γP i -δC i , where α, β, γ, and δ are weight coefficients, satisfying α+β+γ+δ=1, and comparing the E of different emergency response strategies i Value, E i The strategy with the largest value is the optimal emergency response strategy.

[0025] Furthermore, the specific steps of optimizing the fault isolation and emergency response of the fault isolation and emergency response module are as follows:

[0026] (1) After fault isolation and emergency response, immediately start the overall status re-detection of the battery pack and use high-precision sensors to collect key parameters and operating status information;

[0027] (2) Verify the fault isolation effect by comparing data changes and check the boundaries of the isolation area;

[0028] (3) Evaluate the effectiveness and appropriateness of emergency response measures, analyze their timeliness of initiation, execution and impact on vehicle safety, and automatically record relevant data, logs and feedback information;

[0029] (4) Generate a verification and feedback report based on the collected information, listing the isolation effects, evaluation results, improvement issues and optimization suggestions;

[0030] (5) Optimize and improve the system based on the report, covering algorithms, control strategies, emergency plan processes, and sensitivity response speed;

[0031] (6) Update the lessons learned information to the knowledge base as a reference for handling similar situations in the future.

[0032] Compared with the existing technology, the real-time monitoring system for power battery status has the following beneficial effects:

[0033] 1. The present invention uses integrated fault diagnosis technology to build an all-round, high-density sensor network to collect key parameters such as voltage, temperature, current and internal resistance of each single cell in the battery pack in real time, ensuring the comprehensiveness and accuracy of the data. Combined with a variety of advanced fault diagnosis algorithms, such as the battery comprehensive fault identification index formula, battery physical model prediction and abnormal value detection indicators, it can quickly extract features from massive data and perform pattern recognition, thereby achieving accurate judgment of the single cell fault type and precise positioning of the fault location. This integrated diagnosis method not only improves the accuracy of fault diagnosis, but also greatly shortens the diagnosis time, providing strong support for timely response measures.

[0034] 2. The present invention uses intelligent relays and electronic switches to accurately cut off the electrical connection between the faulty single cell and other parts of the battery pack, effectively preventing the fault from spreading to the entire battery pack and avoiding the chain reaction caused by the single cell failure. At the same time, the system can also send an alarm signal to the vehicle control system in time to activate the emergency plan to ensure the safe and stable operation of the vehicle. In addition, the fault isolation and emergency response module also has self-optimization capabilities. By continuously collecting feedback information, it continuously optimizes the algorithm, control strategy and plan process, further improving the reliability and safety of the system. This integrated fault isolation and emergency response mechanism provides a solid guarantee for the safety of the power battery pack and the entire vehicle.

[0035] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1This is the process operation diagram of the real-time monitoring system for the power battery status;

[0038] Figure 2 It is the process framework diagram of the integrated fault diagnosis module;

[0039] Figure 3 This is the process framework diagram of the fault isolation and emergency response module. DETAILED DESCRIPTION

[0040] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Embodiment 1:

[0042] This embodiment provides a real-time monitoring system for the status of a power battery. The system realizes real-time monitoring of the voltage, temperature, current and internal resistance of a single cell by constructing a sensor network. The data acquisition and processing module is used to wirelessly aggregate and pre-process data, and perform feature extraction and pattern recognition. The integrated fault diagnosis module integrates multiple algorithms to quickly and accurately identify the type and location of battery faults, and constructs a battery physical model to predict potential faults. The fault isolation and emergency response module quickly cuts off the connection of the faulty battery to prevent the fault from spreading, and initiates the emergency plan. The system supports remote monitoring, which is convenient for manufacturers and service providers to carry out fault warning and maintenance. The overall design improves the safety and reliability of the electric vehicle battery system.

[0043] In the specific implementation, the sensor network is constructed according to the overall layout and design requirements of the power battery pack, and the specific position and spacing of each single battery are planned to ensure the feasibility of sensor installation and the comprehensiveness of the monitoring network. High-precision and high-sensitivity voltage sensors, temperature sensors, current sensors and internal resistance monitoring sensors are selected. For each single battery, the optimal installation position of the voltage sensor, temperature sensor, current sensor and internal resistance monitoring sensor is determined. According to the planned sensor position, the sensor network is wired, and high-temperature resistant, corrosion-resistant, and anti-electromagnetic interference cables are used to ensure the stability and security of data transmission. After all sensors are installed, network debugging and testing are carried out, and the data collected by each sensor is integrated and processed to form a unified data format and protocol.

[0044] The data acquisition and processing module sends the monitored data to the central processing unit in real time through wireless communication technology. After receiving the raw data from each sensor, the central processing unit first cleans the data to remove noise, outliers and incomplete data to ensure the data quality for subsequent analysis. Based on the pre-processed data, the feature extraction algorithm is used to extract key information useful for fault diagnosis from a large amount of raw data, such as voltage fluctuations, temperature change rates, and current mutation points. The extracted data is compressed and stored in the cloud database.

[0045] The integrated fault diagnosis module extracts key feature vectors that can reflect the battery status from the preprocessed data, and uses the battery comprehensive fault identification index formula to combine the characteristic values ​​of voltage, current, temperature and internal resistance to calculate the comprehensive fault identification index of each single cell. By comparing the index with the preset threshold range, the health status of the single cell is judged, including normal, minor fault, moderate fault and severe fault and location. Based on the electrochemical, thermal and mechanical characteristics of the battery, a battery physical model is constructed. Performance data under different working conditions are collected through experiments and the model is calibrated. Subsequently, real-time data is input into the calibrated model to predict the performance of the battery in the future and compare it with the preset normal range to identify potential fault modes and locations. The outliers in the feature vector are detected by the outlier detection index, the average value and discrete degree matrix of all the feature vectors of the single cells are calculated, the anomaly detection index is defined, and the anomaly threshold is set. When the feature vector of a single cell exceeds the threshold, it is considered to be in an abnormal state and its location is determined. Combined with the battery comprehensive fault identification index, the prediction results of the battery physical model and the anomaly detection results, the fault type and specific location of each single cell are calculated through a comprehensive judgment formula, and a detailed diagnosis report is generated.

[0046] After receiving the fault report from the fault diagnosis module, the fault isolation and emergency response module immediately controls the intelligent relay and electronic switch to quickly and accurately cut off the electrical connection between the faulty single cell and the rest of the battery pack. While the fault isolation operation is being performed, the system sends an alarm signal containing the fault type, location and severity information to the vehicle control system. This signal not only reminds the driver to pay attention, but also provides a basis for the selection of subsequent emergency response measures. After receiving the alarm signal, the appropriate emergency plan is selected through the selection formula. These plans are customized according to the fault type, severity and current status of the vehicle. They can be started after confirmation by the driver to start the corresponding emergency measures to ensure vehicle safety and reduce the impact of the fault on vehicle performance. At the same time, the system records the fault data and sends a fault notification to the owner and the service center for timely maintenance and follow-up. After the fault isolation and emergency response operations are completed, the battery pack status is re-tested to verify whether the fault is effectively isolated and whether the emergency response measures are appropriate. The fault isolation effect is verified by comparing data changes, the isolation area boundaries are checked, and the effectiveness and appropriateness of the emergency response measures are evaluated.

[0047] Remote monitoring and maintenance: The system encrypts the power battery status data and fault diagnosis report and uploads it to the cloud server through wireless communication. The cloud performs data processing and storage. Manufacturers and service providers can remotely monitor through the monitoring platform and mobile applications. The system has an intelligent early warning mechanism that can send early warning notifications in time and perform preventive maintenance based on historical data. The fault diagnosis report records information in detail and supports online feedback.

[0048] Embodiment 2:

[0049] The real-time monitoring system for the power battery status in this embodiment places special emphasis on real-time monitoring of the status of each single battery in the battery pack through advanced algorithms and sensor networks, quickly and accurately identifying the type and location of the fault, and automatically taking isolation measures to prevent the fault from spreading.

[0050] Voltage sensors, temperature sensors, current sensors and internal resistance monitoring sensors are installed on each single cell of the power battery pack to form an all-round, high-density monitoring network. These sensors collect the voltage (V), temperature (T), current (I) and internal resistance (R) data of the single cell in real time.

[0051] The sensor data is wirelessly aggregated to the central processing unit, and the collected data is pre-processed, including data cleaning and denoising, and then feature extraction is performed to provide basic data for subsequent fault diagnosis.

[0052] The battery fault identification index is used to identify the fault type and location of the single cell. Assume that there are n single cells in the battery pack. For the i-th single cell, the voltage characteristic value Vi represents the real-time voltage of the single cell, and the current eigenvalue I i represents the real-time current flowing through the single cell, and the temperature eigenvalue T i represents the real-time temperature of the single cell, and the internal resistance eigenvalue represents the internal resistance of the single cell, and the comprehensive fault identification index of the single cell is F i , and the calculation formula is: Set different threshold intervals for F i When t1, it is judged to be in a normal state, t1 ≤ F i When t2, it is judged to be a minor fault type 1, t2 ≤ F i When t3, it is judged to be a moderate fault type 2, F i When ≥ t3, it is judged to be a severe fault type 3. Determine the fault location according to the number of the single cell with the largest F value, and the fault location is the j-th single cell. i The fault location is the j-th single cell according to the number of the single cell with the largest F value.

[0053] Construct a mathematical model to describe the internal electrochemical reaction process of the battery, including but not limited to the migration of lithium ions between the positive and negative electrodes, charge transfer reactions, the formation and evolution of the solid electrolyte interface (SEI), establish a mathematical model of the battery's thermal behavior, considering factors such as heat generated inside the battery, heat conduction, heat convection, and heat radiation, to predict the battery temperature distribution and changes, construct a model to describe the internal mechanical stress and deformation of the battery, considering the stress distribution and possible structural failures caused by volume changes and thermal expansion during the charging and discharging process of the battery, obtain the electrochemical performance, thermal performance, and possible mechanical performance data of the battery under different working conditions through experimental tests, calibrate the constructed physical model using the experimental data, design a data interface to receive real-time sensor data of voltage, current, and temperature from the battery management system in real time, preprocess the real-time data, input the preprocessed real-time data into the calibrated physical model, calculate the performance of the battery in the next period of time based on the current data and prediction algorithms, including voltage changes, temperature distribution, and mechanical stress, set fault thresholds, compare the model prediction results with the preset normal range and historical data, identify potential fault modes, such as rapid capacity decay, abnormal temperature rise, and mechanical stress concentration, and organize the identified potential fault types, locations, severities, and predicted development trends into a report.

[0054] Detect outliers in the eigenvector through outlier detection metrics. Suppose there are n single cells in the battery pack, and the eigenvector of each single cell is where m is the number of features, and calculate the average value of the eigenvectors of all single cells as the central eigenvector Among them, construct a dispersion matrix D. Among them, for the eigenvector of the i-th single cell Define the anomaly detection indicator as A i , the calculation formula is: Set the abnormal threshold T, when A i >T, the ith single cell is considered to be in an abnormal state, and the abnormal position is the position of the ith single cell in the battery pack.

[0055] The fault type of the single battery is determined by the fault judgment formula. Assume that the output based on the battery fault identification index is A, the output of the fault prediction algorithm based on the battery physical model is B, and the output based on the abnormal value detection index is C. The fault judgment formula is defined as F, and the calculation formula is: Among them, W A ,W B ,W C are the weight vectors output by the corresponding algorithms respectively, α, β, and γ are the weight coefficients of the three algorithms, satisfying α+β+γ=1, ∥·∥ represents the modulus of the vector, and the fault type of the single cell is judged to be the fault type represented by the corresponding subscript with the largest value in F.

[0056] The specific location of the single battery fault is determined by the location judgment formula. The probability distribution of each location fault given by the battery fault identification index is assumed to be a vector M = (m1, m2, ..., m k ), where k is the total number of possible positions, m i represents the probability of a fault occurring at the i-th position. The fault index value of each position given by the fault prediction algorithm based on the battery physical model is the vector P = (p1, p2, ..., p k ), which can reflect the fault possibility of each location. The abnormality degree value of each location given by the abnormal value detection index is a vector S = (s1, s2, ..., s k ), define the position judgment formula as G, and the calculation formula is: Among them, λ M , P , S are weight adjustment factors of the output results of the three algorithms respectively. The specific fault location of the single cell is determined as the position subscript that makes the G value maximum. When i=j, G obtains the maximum value, and the specific fault location of the single cell is determined to be the jth position.

[0057] The diagnostic results of the fault type and its specific location are organized into a diagnostic report.

[0058] After receiving the fault report from the fault diagnosis module, the isolation mechanism is automatically triggered to quickly and accurately cut off the electrical connection between the faulty single cell and the rest of the battery pack by controlling the intelligent relay and electronic switch to prevent the fault from spreading. At the same time, when the fault isolation operation is performed, an alarm signal containing information on the fault type, location and severity is sent to the vehicle control system.

[0059] After receiving the alarm signal, the appropriate emergency plan is selected through the evaluation formula, and the evaluation parameters are defined. i It indicates the fault isolation speed, which is the time from detecting the fault to completely isolating the faulty single cell. The specific value is obtained through experiments. The faster the isolation speed, the larger this value should be. i It indicates the degree of risk reduction, which can be quantified according to the degree of reduction in the possibility of fault propagation. This value is determined by comparing the change in the probability of fault propagation before and after the emergency strategy is adopted. i Indicates the degree of protection of vehicle performance, which is determined by comparing the changes in the performance parameters of the vehicle's power output and driving speed before and after the emergency strategy is adopted. The higher the degree of protection, the larger this value is. i It represents the cost factor, including the hardware cost and time cost required to implement the emergency strategy. The lower the cost, the greater the value. The evaluation formula is constructed: E i =αS i +βR i +γP i -δC i , where α, β, γ, and δ are weight coefficients, satisfying α+β+γ+δ=1. These weight coefficients are based on the actual application of different factors to compare the E of different emergency response strategies. i Value, E i The strategy with the largest value is the optimal emergency response strategy. After the driver confirms, the corresponding emergency measures are initiated, the fault data is recorded, and a fault notification is sent to the owner and the service center.

[0060] After the fault isolation and emergency response operations are completed, the system will immediately start re-detection of the overall status of the battery pack, using high-precision sensors and advanced monitoring technology to collect key parameters of single cells and overall operating status information of the battery pack. Then, by comparing and analyzing data changes, the fault isolation effect is verified, and the boundaries of the isolation area are checked to ensure that the fault has not spread. At the same time, the effectiveness and appropriateness of the emergency response measures are evaluated, including the analysis of its timeliness of initiation, execution, and impact on vehicle and passenger safety. The system will also automatically collect and record relevant data, logs and feedback information, and generate a detailed verification and feedback report based on this information, listing the fault isolation effect, emergency response measures evaluation results, and problems that need to be improved and future optimization suggestions. Then, based on the suggestions and solutions in the report, the system is continuously optimized and improved, including optimizing the fault diagnosis algorithm, adjusting the isolation mechanism control strategy, improving the emergency plan process, and improving the system's sensitivity and response speed to abnormal situations. Finally, the lessons learned from each fault isolation and emergency response, successful cases and failed cases are updated to the knowledge base as a reference for similar situations in the future, helping the system to make decisions and respond faster and more accurately.

[0061] The system also supports remote data transmission, uploading monitoring data and fault diagnosis reports to the cloud server in real time, making it easier for manufacturers and service providers to conduct remote monitoring, fault warning and preventive maintenance.

[0062] It will be apparent to those skilled in the art that the 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 the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. The real-time monitoring system for power battery status is characterized by: The system consists of the following components: Sensor network construction: Install voltage sensors, temperature sensors, current sensors and internal resistance monitoring sensors on each single cell of the power battery pack to form an all-round, high-density monitoring network; Data acquisition and processing module: The sensor data is aggregated to the central processing unit through wireless transmission, and the collected data is pre-processed and feature extracted using algorithm formulas; Integrated fault diagnosis module: Integrates multiple fault diagnosis algorithms, comprehensively analyzes sensor data, and quickly and accurately identifies various fault types and specific locations of single cells; Fault isolation and emergency response module: Once a fault is diagnosed, the system automatically triggers the isolation mechanism, quickly disconnecting the faulty single cell from the rest of the battery pack through intelligent relays and electronic switches to prevent the fault from spreading. At the same time, an alarm signal is sent to the vehicle control system to initiate an emergency plan. Remote monitoring and maintenance: The system supports remote data transmission function, which can upload monitoring data and fault diagnosis reports to the cloud server in real time, making it convenient for manufacturers and service providers to conduct remote monitoring, fault warning and preventive maintenance.

2. The power battery status real-time monitoring system according to claim 1, characterized in that: The integrated fault diagnosis module extracts key feature vectors that can reflect the battery status from the preprocessed data, integrates multiple fault diagnosis algorithm formulas, uses the battery fault identification index to identify the fault type and location of the single cell, establishes a battery physical model according to the electrochemical, thermal and mechanical characteristics of the battery, and predicts the behavior of the battery under different working conditions through simulation, compares it with real-time data, predicts potential fault types, detects abnormal values ​​in the feature vector through anomaly detection indicators, identifies abnormal states that deviate from the normal range, and uses them as the basis for fault diagnosis. In combination with the output results of multiple algorithms, the fault type and specific location of the single cell are determined through formulas, and the diagnosis results are compiled into a diagnosis report.

3. The power battery status real-time monitoring system according to claim 2, characterized in that: The integrated fault diagnosis module uses the battery fault identification index to identify the fault type and location of each single battery. Suppose there are n single batteries in the battery pack. For the i-th single battery, the voltage characteristic value V i represents the real-time voltage of this single battery, and the current characteristic value I i represents the real-time current flowing through this single battery, and the temperature characteristic value T i represents the real-time temperature of this single battery, and the internal resistance characteristic value represents the internal resistance of this single battery. The battery fault identification index is F i , and the calculation formula is: where, are the average values of the voltages, currents, temperatures, and internal resistances of all single batteries respectively, and σ V、 σ I、 σ T、 σ R are the corresponding standard deviations respectively, and α, β, γ, δ are weight coefficients. For the judgment of the fault type, different threshold intervals are set. When F i < t1, it is judged as the normal state. When t1 ≤ F i < t2, it is judged as the minor fault type 1. When t2 ≤ F i < t3, it is judged as the moderate fault type 2. When F i ≥ t3, it is judged as the severe fault type 3. According to the serial number of the single battery with the largest F i value, the fault location is determined, and the fault location is the j-th single battery.

4. The power battery status real-time monitoring system according to claim 2, characterized in that: The integrated fault diagnosis module predicts battery performance and identifies potential faults by building a battery physical model combined with real-time data: (1) Construct electrochemical, thermal and mechanical models to describe the electrochemical reaction, thermal behavior and mechanical stress deformation inside the battery, covering the factors of lithium ion migration, heat conduction and volume change; (2) Collect performance data under different working conditions through experiments and use the performance data to calibrate the constructed physical model; (3) Design a data interface to access real-time data and preprocess it, input the preprocessed real-time data into the calibrated physical model, and predict the performance of the battery in the future; (4) Setting the fault threshold, comparing the model prediction results with the preset normal range, and identifying potential failure modes; (5) Organize the identified potential fault types, locations, severity and predicted development trends into a report.

5. The power battery status real-time monitoring system according to claim 2, characterized in that: The integrated fault diagnosis module detects abnormal values ​​in the feature vector through an abnormal value detection index. Assuming that there are n single cells in the battery pack, the feature vector of each single cell is Where m is the number of features, and the average value of all single battery feature vectors is calculated as the central feature vector Among them, the discrete degree matrix D is constructed, where the characteristic vector of the i-th single cell is Define the anomaly detection indicator as A i , the calculation formula is: Set the abnormal threshold T, when A i >T, the ith single cell is considered to be in an abnormal state, and the abnormal position is the position of the ith single cell in the battery pack.

6. The power battery status real-time monitoring system according to claim 2, characterized in that: The integrated fault diagnosis module determines the fault type of the single battery through the fault judgment formula. Assume that the output based on the battery fault identification index is A, the output of the fault prediction algorithm based on the battery physical model is B, and the output based on the abnormal value detection index is C. The fault judgment formula is defined as F, and the calculation formula is: Among them, W A ,W B ,W C are the weight vectors output by the corresponding algorithms respectively, α, β, and γ are the weight coefficients of the three algorithms, satisfying α+β+γ=1, ∥·∥ represents the modulus of the vector, and the fault type of the single cell is judged to be the fault type represented by the corresponding subscript with the largest value in F.

7. The power battery status real-time monitoring system according to claim 2, characterized in that: The integrated fault diagnosis module determines the specific location of the single battery fault through the location judgment formula. The probability distribution of each location fault given by the battery fault identification index is assumed to be a vector M = (m1, m2, ..., m k ), where k is the total number of possible positions, m i represents the probability of a fault occurring at the i-th position. The fault index value of each position given by the fault prediction algorithm based on the battery physical model is the vector P = (p1, p2, ..., p k ), which can reflect the fault possibility of each location. The abnormality degree value of each location given by the abnormal value detection index is the vector S = (s1, s2, ..., s k ), define the position judgment formula as G, and the calculation formula is: G = Among them, λ M , P , S are weight adjustment factors of the output results of the three algorithms respectively. The specific fault location of the single cell is determined as the position subscript that makes the G value maximum. When i=j, G obtains the maximum value, and the specific fault location of the single cell is determined to be the jth position.

8. The power battery status real-time monitoring system according to claim 1, characterized in that: After receiving the fault report from the fault diagnosis module, the fault isolation and emergency response module automatically triggers the isolation mechanism, and quickly and accurately cuts off the electrical connection between the faulty single cell and other parts of the battery pack by controlling the intelligent relay and the electronic switch to prevent the fault from spreading. At the same time, when the fault isolation operation is performed, an alarm signal containing information on the fault type, location and severity is sent to the vehicle control system. After receiving the alarm signal, an appropriate emergency plan is selected through an evaluation formula, and the corresponding emergency measures are initiated after confirmation by the driver. The fault data is recorded and a fault notification is sent to the owner and the service center. After the fault isolation and emergency response operations are completed, the battery pack status is re-tested to verify whether the fault is effectively isolated and whether the emergency response measures are appropriate, and feedback information is collected for system optimization and performance improvement.

9. The power battery status real-time monitoring system according to claim 8, characterized in that: The fault isolation and emergency response module selects a suitable emergency plan through an evaluation formula and defines evaluation parameters, S i It represents the fault isolation speed, which is the time from detecting the fault to completely isolating the faulty single cell. i Indicates the degree of risk reduction, P i Indicates the degree of protection of vehicle performance, C i Represents the cost factor and constructs the evaluation formula: E i =αS i +βR i +γP i -δC i , where α, β, γ, and δ are weight coefficients, satisfying α+β+γ+δ=1, and comparing the E of different emergency response strategies i Value, E i The strategy with the largest value is the optimal emergency response strategy.

10. The power battery status real-time monitoring system according to claim 8, characterized in that: The specific steps of the fault isolation and emergency response module for optimizing fault isolation and emergency response are as follows: (1) After fault isolation and emergency response, immediately start the overall status re-detection of the battery pack and use high-precision sensors to collect key parameters and operating status information; (2) Verify the fault isolation effect by comparing data changes and check the boundaries of the isolation area; (3) Evaluate the effectiveness and appropriateness of emergency response measures, analyze their timeliness of initiation, execution and impact on vehicle safety, and automatically record relevant data, logs and feedback information; (4) Generate a verification and feedback report based on the collected information, listing the isolation effects, evaluation results, improvement issues and optimization suggestions; (5) Optimize and improve the system based on the report, covering algorithms, control strategies, emergency plan processes, and sensitivity response speed; (6) Update the lessons learned information to the knowledge base as a reference for handling similar situations in the future.

Images