A BMS battery safety management control system

Abstract

The present invention relates to the technical field of battery management, and discloses a BMS battery safety management control system, comprising: a data acquisition module, used for acquiring voltage, current and temperature data of a battery; a data processing module, used for optimizing the data acquired by the data acquisition module, reducing the influence of noise on difference data, and reducing electromagnetic interference; the specific steps of optimizing the acquired data by the data processing module are: a state evaluation module, used for estimating the charge state, power state and health state of the battery; an energy management module, used for controlling the charging and discharging of the battery, and preventing overcharging and overdischarging; a safety protection module, used for monitoring whether the voltage, current and temperature of the battery exceed a set threshold; the present invention constructs a new data processing method to process the acquired data, and reduces the influence of electromagnetic interference on battery data through abnormal data detection and processing, thereby improving the accuracy and reliability of the data.

Description

Technical Field

[0001] The present invention relates to the technical field of battery management, and in particular to a BMS battery safety management control system. Background Art

[0002] The battery management system (BMS) is an indispensable component of electric vehicles. It is the hub for managing and monitoring power batteries. It manages, maintains, and monitors each battery module. It is responsible for preventing battery overcharge and over-discharge, extending battery life, and helping batteries operate normally.

[0003] The battery management system (BMS) is an important link between the on-board battery and the electric vehicle. Its main functions include: real-time monitoring of battery physical parameters, battery status estimation, online diagnosis and early warning, charge and discharge and pre-charge control balance management, thermal management, etc. If any of the above functions is not implemented well, it will cause fatal harm to the battery.

[0004] However, the electronic modules and high-voltage systems inside electric vehicles generate electromagnetic radiation when working. These radiations propagate through space or wires and couple with the circuits in the battery management system (BMS), causing interference and affecting the accuracy and real-time performance of data transmission. This may cause the battery management system (BMS) to be unable to receive instructions and information from the vehicle control system or other related systems in a timely manner, thereby affecting the battery management effect. Summary of the invention

[0005] The object of the present invention is to provide a BMS battery safety management and control system to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a BMS battery safety management and control system, comprising:

[0007] Data acquisition module, used to collect battery voltage, current and temperature data;

[0008] A data processing module, used to optimize the data collected by the data collection module, reduce the impact of noise on the difference data, and reduce electromagnetic interference;

[0009] The specific steps of optimizing the collected data by the data processing module are as follows:

[0010] S1. Data preprocessing: The collected raw data is preprocessed by the sliding window averaging method. The formula of the sliding window averaging method is:

[0011]

[0012] In the formula, p t represents the value after preprocessing at time t, xt represents the value collected at time t, x t-1 represents the value collected at time t-1, x t+1 represents the value collected at time t+1, and n represents the sliding window radius;

[0013] S2, feature extraction, extract the mean (P), variance (P) and standard deviation std (P) after data preprocessing;

[0014] S3, anomaly detection, identifies outliers in the preprocessed data through statistical methods and removes them;

[0015] A state assessment module, which estimates the battery's state of charge, state of power, and state of health;

[0016] Energy management module, used to control the charging and discharging of the battery to prevent overcharging and over-discharging;

[0017] Safety protection module, used to monitor whether the voltage, current and temperature of the battery exceed the set threshold;

[0018] Data communication module, used for data interaction with the background system;

[0019] The MCU control module is electrically connected to the data acquisition module, the data processing module, the state evaluation module, the energy management module, the safety protection module and the data communication module respectively.

[0020] Wherein, the calculation formula of the mean mean(P) is:

[0021]

[0022] Where P t Represents the value after preprocessing at time t.

[0023] Wherein, the calculation formula of the variance (P) is:

[0024]

[0025] Wherein, the calculation formula of the standard deviation std(P) is:

[0026]

[0027] The formula of the statistical method is:

[0028]

[0029] In the formula, mean(P) represents the data mean, std(P) represents the data standard deviation, when |Z|>Zstd(P) / 2 , the value P t is an outlier.

[0030] The data acquisition module includes a voltage sensor, a current sensor and a temperature sensor. The voltage sensor is used to detect the voltage parameter of the battery pack, the current sensor is used to detect the current parameter of the battery pack, and the temperature sensor is used to detect the temperature parameter of the battery pack.

[0031] The method for estimating the state of charge is as follows:

[0032] S1. Model the battery system as a fractional-order kinetic model and fit the model parameters by optimizing the processed data;

[0033] S2. Based on the UKFMIUKF algorithm, the multi-innovation theory is introduced to form the UKFMIUKF-S0C estimation algorithm;

[0034] S3, use TensorFlow machine learning library to train UKFMIUKF-S0C estimation algorithm;

[0035] S4. Estimate the battery state of charge in real time through the trained UKFMIUKF-S0C estimation algorithm.

[0036] The power state estimation method is as follows:

[0037] S1, extracting the charging capacity, discharging capacity and internal resistance change of the battery as characteristic values;

[0038] S2, selecting an SVR mathematical model, and then using the characteristic value as input and the actually measured power state value as output to train the SVR mathematical model;

[0039] S3. Apply the trained SVR mathematical model to real-time data collection and estimate the power state in real time.

[0040] The health status estimation method is as follows:

[0041] S1, extracting the charging capacity, discharging capacity and internal resistance change of the battery as characteristic values;

[0042] S2, select the SVR mathematical model, and then use the characteristic value as input and the actual measured health status value as output to train the SVR mathematical model;

[0043] S3. Apply the trained SVR mathematical model to real-time data collection and estimate the health status in real time.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention first collects the voltage, current and temperature data of the battery through a data acquisition module; then pre-processes the collected raw data through a sliding window averaging method, and then extracts the mean, variance and standard deviation of the pre-processed data; preferably, outliers of the pre-processed data are identified through a statistical method, and the outliers are removed, and a new data processing method is constructed to process the collected data. Through abnormal data detection and processing, the influence of electromagnetic interference on battery data is reduced, the accuracy and reliability of the data are improved, thereby improving the overall performance and reliability of the battery management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a block diagram of the hardware control system of the present invention;

[0047] Figure 2 This is a schematic diagram of the process flow for optimizing the data of the present invention;

[0048] Figure 3 It is a schematic flow chart of the method for estimating the state of charge of the present invention;

[0049] Figure 4 It is a schematic flow chart of a method for estimating power state of the present invention;

[0050] Figure 5 Schematic diagram of the flow of the method for estimating health status of the present invention. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0052] See also Figure 1-5 The present invention provides a technical solution: a BMS battery safety management and control system, comprising:

[0053] The data acquisition module is used to collect the voltage, current and temperature data of the battery. The data acquisition module includes a voltage sensor, a current sensor and a temperature sensor. The voltage sensor is used to detect the voltage parameters of the battery pack, the current sensor is used to detect the current parameters of the battery pack, and the temperature sensor is used to detect the temperature parameters of the battery pack.

[0054] The data processing module is used to optimize the data collected by the data acquisition module, reduce the influence of noise on the difference data, and reduce electromagnetic interference. The voltage data, current data, and temperature data are independent databases. When processing the data, the voltage data, current data, and temperature data are processed separately.

[0055] like Figure 2As shown, the specific steps of optimizing the collected data by the data processing module are as follows:

[0056] S1. Data preprocessing: The collected raw data is preprocessed by the sliding window averaging method to reduce the impact of random fluctuations in the data. The formula of the sliding window averaging method is:

[0057]

[0058] In the formula, p t represents the value after preprocessing at time t, x t represents the value collected at time t, x t-1 represents the value collected at time t-1, x t+1 represents the value collected at time t+1, and n represents the sliding window radius;

[0059] S2, feature extraction, extract the mean (P), variance (P) and standard deviation std (P) after data preprocessing;

[0060] Wherein, the calculation formula of the mean mean(P) is:

[0061]

[0062] Where P t Represents the value after preprocessing at time t.

[0063] Wherein, the calculation formula of the variance (P) is:

[0064]

[0065] Wherein, the calculation formula of the standard deviation std(P) is:

[0066]

[0067] The formula of the statistical method is:

[0068]

[0069] In the formula, mean(P) represents the data mean, std(P) represents the data standard deviation, when |Z|>Z std(P) / 2 , the value P t is an outlier;

[0070] S3, anomaly detection, identifies the outliers of the preprocessed data through statistical methods and removes them.

[0071] When the voltage database needs to be processed, first, data preprocessing is performed by using the sliding window averaging method to preprocess the collected voltage raw data to reduce the impact of random data fluctuations. Next, feature extraction is performed to extract the mean (P), variance (P) and standard deviation (P) of the preprocessed voltage data. Then, anomaly detection is performed to identify abnormal values ​​of the preprocessed voltage data through statistical methods and remove abnormal voltage values.

[0072] When the current database needs to be processed, first, data preprocessing is performed by using the sliding window averaging method to preprocess the collected raw current data to reduce the impact of random fluctuations in the data. Next, feature extraction is performed to extract the mean (P), variance (P) and standard deviation (P) of the preprocessed current data. Then, anomaly detection is performed to identify abnormal values ​​of the preprocessed current data through statistical methods, and to remove abnormal current values.

[0073] Among them, when the temperature database needs to be processed, first of all, data preprocessing is performed, and the collected raw temperature data is preprocessed by the sliding window averaging method to reduce the impact of random data fluctuations: then, feature extraction is performed, and the mean mean(P), variance variance(P) and standard deviation std(P) of the preprocessed temperature data are extracted; then, anomaly detection is performed, and the abnormal values ​​of the preprocessed temperature data are identified by statistical methods, and the abnormal temperature values ​​are removed.

[0074] A state assessment module, which estimates the battery's state of charge, state of power, and state of health;

[0075] The energy management module is used to control the charging and discharging of the battery to prevent overcharging and over-discharging. Specifically:

[0076] 1. Estimate the battery's state of charge (SOC), state of health (SOH), and state of function (SOF) to provide information such as the remaining battery power and remaining usage time;

[0077] 2. When the battery has abnormal conditions, such as overcharge, over discharge, over temperature, etc., the system will automatically cut off the charge and discharge to avoid safety accidents;

[0078] 3. In the case of multiple series or parallel connections, due to the differences between each single battery, some single cells are prone to overcharge or over discharge. The energy management module is used to balance and make the status of each single cell as consistent as possible;

[0079] 4. Determine the intensity of active heating or heat dissipation based on the temperature distribution information in the battery pack and the charging and discharging requirements, so that the battery can work at the most suitable temperature as much as possible;

[0080] 5. Control the battery charging and discharging process according to external load requirements and real-time battery status.

[0081] A safety protection module is used to monitor whether the voltage, current and temperature of the battery exceed the set thresholds; the voltage thresholds include a maximum voltage threshold and a minimum voltage threshold, the current thresholds include a maximum current threshold and a minimum current threshold, and the temperature thresholds include a maximum temperature threshold and a minimum temperature threshold;

[0082] When the processed voltage exceeds the set maximum voltage threshold, charging is stopped and an alarm signal is issued; when the processed voltage is lower than the set minimum voltage threshold, discharging is stopped and an alarm signal is issued;

[0083] When the processed current exceeds the set maximum current threshold, charging or discharging is stopped and an alarm signal is issued; when the processed current is lower than the set minimum current threshold, the circuit may be short-circuited, the power supply is automatically disconnected and an alarm signal is issued;

[0084] When the treated temperature exceeds the set maximum temperature threshold or is lower than the set minimum temperature threshold, charging or discharging is stopped and an alarm signal is issued.

[0085] Data communication module, used for data interaction with the background system;

[0086] Among them, the data communication module adopts CAN bus technology and wireless communication technology. CAN bus technology is a real-time, safe and reliable communication technology. Through a distributed network structure, it can realize data transmission and communication between multiple battery modules to ensure the data interconnection of the vehicle itself; wireless communication technology can realize the interconnection of the system with external background data.

[0087] MCU control module, the MCU control module is electrically connected to the data acquisition module, data processing module, state evaluation module, energy management module, safety protection module and data communication module respectively, and the data acquisition module, data processing module, state evaluation module, energy management module, safety protection module and data communication module are controlled by the MCU control module.

[0088] Among them, Figure 3 As shown, the state of charge is estimated as follows:

[0089] S1. Model the battery system as a fractional-order kinetic model and fit the model parameters by optimizing the processed data;

[0090] Among them, the fractional-order kinetic model is constructed based on the physical and chemical mechanisms inside the battery, such as the second-order RC equivalent circuit model, which contains fractional-order capacitance elements;

[0091] The fractional order dynamics model can be expressed as x(t) = A(t) + Bu(t), where x(t) is the state vector, which contains all the necessary information to describe the current state of the system, u(t) is the control input vector of the system at time t, and the control input vector represents the external control applied to the system. A is the state matrix, which describes the internal dynamic characteristics of the system. Matrix A determines how the system state changes over time when there is no control input, and B is the input matrix, which describes how the control input affects the system state. Matrix B maps the control input to the state space.

[0092] Fractional-order kinetic models require experimental data or system identification methods to determine the unknown parameters in the model. Common methods include the least squares method, which is an optimization algorithm that solves the parameters by minimizing the error between the predicted value and the actual measured value. Its objective function is usually the sum of the squares of the error between the predicted voltage and the actual voltage;

[0093] After the parameter identification is completed, the model needs to be verified to ensure that it can accurately describe the dynamic characteristics of the battery. The verified fractional-order kinetic model can be applied to the battery management system for state of charge (SOC) estimation.

[0094] S2. Based on the UKFMIUKF algorithm, the multi-innovation theory is introduced to form the UKFMIUKF-S0C estimation algorithm;

[0095] UKFMIUKF algorithm is a Bayesian filter for nonlinear systems. Its basic steps include initialization, Sigma point calculation, prediction and update.

[0096] Multi-innovation theory: Improve the accuracy of state of charge (SOC) estimation by introducing information from multiple time points. The specific implementation steps include multi-time observation fusion and coefficient design

[0097] S3. Use TensorFlow machine learning library to train UKFMIUKF-S0C estimation algorithm. During the training process, a large amount of historical data is needed as training samples, and the model parameters are optimized through cross-validation method.

[0098] S4, real-time estimation of the battery state of charge through the trained UKFMIUKF-S0C estimation algorithm;

[0099] According to the estimation results, corresponding protection measures are triggered, such as disconnecting the power supply and alarming. At the same time, abnormal events are recorded for subsequent analysis and algorithm improvement. The self-learning mechanism enables the model to be updated in real time by regularly retraining the model and using online learning algorithms, improving the adaptability and accuracy of the algorithm.

[0100] Among them, Figure 4As shown, the estimation of battery power state (SOP) is crucial for applications such as electric vehicles, because it can predict the available power of the battery in real time, thereby optimizing the battery's efficiency and life. The power state (SOP) estimation method is as follows:

[0101] S1. Extracting the charging capacity, discharging capacity and internal resistance change of the battery as feature values, specifically extracting key features related to the state of power (SOP) from the processed voltage database, current database and temperature database, such as charging capacity, discharging capacity and internal resistance change;

[0102] S2, selecting an SVR mathematical model, and then using the characteristic value as input and the actually measured power state value as output to train the SVR mathematical model;

[0103] The training data set of the SVR mathematical model is Where X i ∈R is the feature vector, Y i ∈R is the corresponding state of power (SOP) value.

[0104] S3. Apply the trained SVR mathematical model to real-time data collection, estimate the power status in real time, and take corresponding maintenance measures based on the estimation results. For example, when the power exceeds the set threshold, limit the charging power; when the power is lower than the set threshold, adjust the charging power to ensure the safe operation of the battery and extend the battery life.

[0105] Among them, Figure 5 As shown in the figure, the estimation of battery health state (SOH) is an important part of the battery management system. It reflects the remaining capacity and aging degree of the battery. The estimation method of health state is as follows:

[0106] S1. Extracting the charging capacity, discharging capacity and internal resistance change of the battery as feature values, specifically extracting key features related to the state of health (SOH) from the processed voltage database, current database and temperature database, such as charging capacity, discharging capacity and internal resistance change;

[0107] S2, selecting an SVR mathematical model, and then using the characteristic value as input and the actually measured power state value as output to train the SVR mathematical model;

[0108] The training data set of the SVR mathematical model is Where X i ∈R is the feature vector, Y i ∈R is the corresponding state of power (SOP) value.

[0109] S3. Apply the trained SVR mathematical model to real-time data collection, estimate the state of health (SOH) in real time, and take corresponding maintenance measures based on the estimation results. For example, when the state of health exceeds the set threshold, limit the charging power; when the state of health is lower than the set threshold, adjust the charging power to ensure the safe operation of the battery and extend the battery life.

[0110] To sum up: the present invention first collects the voltage, current and temperature data of the battery through the data acquisition module; then preprocesses the collected raw data through the sliding window averaging method, and then extracts the mean, variance and standard deviation of the preprocessed data; preferably, the outliers of the preprocessed data are identified through statistical methods, and the outliers are removed, and a new data processing method is constructed to process the collected data. Through abnormal data detection and processing, the impact of electromagnetic interference on battery data is reduced, the accuracy and reliability of the data are improved, thereby improving the overall performance and reliability of the battery management system.

[0111] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0112] Finally, it should be noted that: Obviously, the above embodiments are only examples for clearly explaining the present invention, and are not intended to limit the implementation methods. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from this are still within the scope of protection of the present invention.

Claims

1. A BMS battery safety management and control system, characterized in that: include: Data acquisition module, used to collect battery voltage, current and temperature data; A data processing module, used to optimize the data collected by the data collection module, reduce the impact of noise on the difference data, and reduce electromagnetic interference; The specific steps of optimizing the collected data by the data processing module are as follows: S1. Data preprocessing: The collected raw data is preprocessed by the sliding window averaging method. The formula of the sliding window averaging method is: In the formula, p t represents the value after preprocessing at time t, x t represents the value collected at time t, x t-1 represents the value collected at time t-1, x t+1 represents the value collected at time t+1, and n represents the sliding window radius; S2, feature extraction, extract the mean (P), variance (P) and standard deviation std (P) after data preprocessing; S3, anomaly detection, identifies outliers in the preprocessed data through statistical methods and removes them; A state assessment module, which estimates the battery's state of charge, state of power, and state of health; Energy management module, used to control the charging and discharging of the battery to prevent overcharging and over-discharging; Safety protection module, used to monitor whether the voltage, current and temperature of the battery exceed the set threshold; Data communication module, used for data interaction with the background system; The MCU control module is electrically connected to the data acquisition module, the data processing module, the state evaluation module, the energy management module, the safety protection module and the data communication module respectively.

2. A BMS battery safety management and control system according to claim 1, characterized in that: The calculation formula of the mean mean(P) is: Where P t Represents the value after preprocessing at time t.

3. A BMS battery safety management and control system according to claim 2, characterized in that: The calculation formula of the variance (P) is:

4. A BMS battery safety management and control system according to claim 3, characterized in that: The calculation formula of the standard deviation std(P) is:

5. A BMS battery safety management and control system according to claim 4, characterized in that: The formula of the statistical method is: In the formula, mean(P) represents the data mean, std(P) represents the data standard deviation, when |Z|>Z std(P) / 2 , the value P t is an outlier.

6. A BMS battery safety management and control system according to claim 1, characterized in that: The data acquisition module includes a voltage sensor, a current sensor and a temperature sensor. The voltage sensor is used to detect the voltage parameter of the battery pack, the current sensor is used to detect the current parameter of the battery pack, and the temperature sensor is used to detect the temperature parameter of the battery pack.

7. A BMS battery safety management and control system according to claim 1, characterized in that: The state of charge estimation method is as follows: S1. Model the battery system as a fractional-order kinetic model and fit the model parameters by optimizing the processed data; S2. Based on the UKFMIUKF algorithm, the multi-innovation theory is introduced to form the UKFMIUKF-S0C estimation algorithm; S3, use TensorFlow machine learning library to train UKFMIUKF-S0C estimation algorithm; S4. Estimate the battery state of charge in real time through the trained UKFMIUKF-S0C estimation algorithm.

8. A BMS battery safety management and control system according to claim 1, characterized in that: The power state is estimated as follows: S1, extracting the charging capacity, discharging capacity and internal resistance change of the battery as characteristic values; S2, selecting an SVR mathematical model, and then using the characteristic value as input and the actually measured power state value as output to train the SVR mathematical model; S3. Apply the trained SVR mathematical model to real-time data collection and estimate the power state in real time.

9. A BMS battery safety management and control system according to claim 1, characterized in that: The health status is estimated as follows: S1, extracting the charging capacity, discharging capacity and internal resistance change of the battery as characteristic values; S2, select the SVR mathematical model, and then use the characteristic value as input and the actual measured health status value as output to train the SVR mathematical model; S3. Apply the trained SVR mathematical model to real-time data collection and estimate the health status in real time.

Images