A method and device for automatically generating a universal BMS control model

By automatically generating BMS control models using deep learning algorithms, the shortcomings of traditional BMS systems in terms of battery array adaptability and efficiency are solved, realizing efficient, flexible and reliable charge and discharge control of the battery management system, which is suitable for the production and application of new energy battery packs.

CN117767462BActive Publication Date: 2024-11-01YANTAI HEYING CULTURAL IND DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311642448.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-01
Publication Date
2024-11-01
Estimated Expiration
2043-12-01

AI Technical Summary

Technical Problem

Existing BMS control systems struggle to achieve optimal charge and discharge control when faced with different battery models and varying battery array sizes. Furthermore, traditional manual modeling methods are time-consuming and costly, and cannot adapt to the complex nonlinear characteristics and variability of new energy batteries.

Method used

A deep learning algorithm is used to establish a control neural network model for an intelligent battery management system, which automatically generates a BMS control model adapted to the battery array. Through real-time data analysis and model adjustment, the optimal charging and discharging strategy is achieved.

Benefits of technology

It improves battery charging efficiency, extends battery life, reduces human error, adapts to different battery types and sizes, provides powerful battery management capabilities, and is suitable for industrial, transportation, and energy storage applications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117767462B_ABST
    Figure CN117767462B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of new energy battery management system (BMS), and in particular to a method and device for automatically generating a general BMS control model. Obtain battery data from several battery array units, attach battery number information, and upload to an information processing center; in the information processing center, store, sort, count, and analyze the obtained battery data in real time through a battery learning mode; then, based on the analysis of the battery data, use deep learning to establish an intelligent battery management system control neural network model to obtain an initial BMS control model; use the obtained initial BMS control model to generate a core working program and a calling interface; complete data verification during charging and discharging, compare with theoretical simulation results, and record the deviation value of parameter indicators; make judgments and adjust the parameters of the BMS control model; solidify the adjusted BMS control model parameters to generate a general BMS control model. More accurate, customized, and efficient battery management and control are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of new energy battery management systems (BMS), and in particular to a method and device for automatically generating a universal BMS control model. Background Art

[0002] With the development and application of new energy technologies, batteries, as important energy storage and supply devices, are widely used in electric vehicles, renewable energy systems and other fields. In order to ensure the safety, performance and life of new energy batteries, BMS, or battery management system, plays a key role. BMS monitors, controls and protects battery packs to achieve battery charge and discharge management, state estimation and fault diagnosis.

[0003] The development of existing BMS control systems has the following limitations: Different models of batteries, or even the same model of batteries, have different electrical characteristics when combined into battery arrays of different capacities, forming different charge and discharge curves; and battery packs produced by different manufacturers may have their own BMS systems, or be configured by specialized third-party technical entities. For battery arrays of different sizes, traditional BMS control model design usually uses manual modeling and manual programming parameter adjustment methods, which requires a lot of time and professional knowledge accumulation, and is difficult to adapt to the above-mentioned different types and array sizes of battery systems. New energy batteries have complex nonlinear characteristics, capacity attenuation, temperature changes and other factors. The design and debugging methods of traditional methods have a long time cycle. The finalization of the BMS system of a set of new energy vehicle battery packs takes at least several months and at most more than a year, and the labor cost remains high.

[0004] At the same time, the traditional BMS system also has the following problems: it is unable to provide the optimal charge and discharge control solution, i.e., the micro-control solution, at the battery cell level; when a debugged battery array BMS is applied to battery packs with the same model of batteries but different quantities and sizes, it is unable to ensure the optimization of battery performance and output power.

[0005] Therefore, there is an urgent need to design and implement a general BMS control model automatic generation method and equipment. Summary of the invention

[0006] The present invention provides a method and device for automatically generating a universal BMS control model, with the aim of automatically generating a BMS control model adapted to the current battery array according to the characteristics and monitoring data of the battery during the charging and discharging process, thereby realizing the optimal charging and discharging control strategy for the battery array, improving the charging efficiency as much as possible, protecting the battery health, maximizing the utilization rate of the energy of the overall battery array, and extending the battery life; at the same time, the present invention can be widely used in the production of new energy battery packs.

[0007] The technical solution of the present invention is specifically as follows:

[0008] A general BMS control model automatic generation method comprises the following steps:

[0009] S1. Obtain battery data of an uncertain number of battery array units, attach battery number information, and upload to the information processing center module in real time;

[0010] S2. In the information processing center module, the battery data obtained is stored, sorted, counted and analyzed in real time through the learning mode;

[0011] S3. Based on the analysis of battery data, an intelligent battery management system control neural network model is established based on deep learning to obtain an initial BMS control model;

[0012] S4. Generate the core working program and calling interface using the initial BMS control model obtained; complete the data verification during charging and discharging, compare it with the theoretical simulation effect, and record the deviation value of the parameter index;

[0013] S5. Determine the deviation value and adjust the parameters of the BMS control model;

[0014] S6. Solidify the adjusted BMS control model parameters to generate an overall model that adapts to the usage of different batteries, that is, a universal BMS control model.

[0015] Further, step S1 specifically includes:

[0016] The information processing center module will send a battery operating mode instruction, which includes: discharging, charging, and decoding. The core control module will transmit the battery operating mode instruction to the array unit.

[0017] Further, step S2 specifically includes:

[0018] The battery data information is stored in a database in real time; the stored battery data is sorted and organized in chronological order; the stored battery data is used for statistics and analysis to calculate the overall charging time, discharging time and battery capacity indicators; and the overall charging time, discharging time and battery capacity of the battery array are obtained.

[0019] Further, step S3 specifically includes:

[0020] The battery data is processed through a deep learning algorithm to build an intelligent battery management system control neural network model to perform model parameter operation analysis, provide multiple sets of parameter recommended configurations, and obtain the initial BMS control model.

[0021] Furthermore, the intelligent battery management system control neural network model includes an input layer, a hidden layer, and an output layer.

[0022] Further, step S4 specifically includes:

[0023] According to the trained intelligent battery management system control neural network model, the core control module controls the array unit to complete the charge and discharge data verification during multiple rounds of battery operation, compare it with the theoretical simulation data of the initial BMS control model, and record the deviation value of the parameter index.

[0024] Furthermore, when the deviation value exceeds the threshold range, the model needs to be retrained until the deviation value meets the threshold range requirements; when the deviation value meets the threshold range, the BMS control model parameters are fine-tuned.

[0025] A general BMS control model automatic generation system includes the following contents:

[0026] Transmission bus, array group module, core control module, information processing center module;

[0027] The transmission bus is used for data transmission and command transmission between the core control module of the battery pack and each battery array. It can select a two-way data communication link that supports multiple communication methods, including IO connection, MODBUS / 485, and LAN;

[0028] The array group module consists of multiple array units and is responsible for uploading the collected array unit summary data and issuing work instructions to the array units;

[0029] The array unit includes a microcontroller system, an expandable IO board, a wiring port, a battery electrode group, and a communication port; the array unit can be connected to electrodes and can be connected to a test battery object; the universal BMS control model automatic generation system can transmit the battery data of the real-time remaining power, working current, voltage, and temperature of the battery to the core control unit at an adjustable time interval, providing real-time monitoring and feedback of the battery status; at the same time, signal transmission is completed through a transmission bus;

[0030] The core control module monitors the overall status of the battery pack in real time and takes corresponding control measures according to the predefined control strategy; the core control module includes a bus control unit, a battery data buffer unit, a current / voltage control unit, an array unit control unit, a communication unit, and a power supply management unit;

[0031] The information processing center module is mainly responsible for data processing, analysis and modeling, and establishes a database of collected battery-related data; analyzes current, voltage, remaining power, real-time power consumption, and temperature battery data through the learning unit; and generates a battery management system control model in the model generation unit to help predict battery performance, status, and change trends; the information processing center module includes a real-time battery database, a learning unit, a model generation unit, and a parameter adjustment unit.

[0032] Furthermore, in the array unit, the microcontroller system, which is embedded in the array unit, is responsible for executing various operation instructions for battery management. The microcontroller system is the brain of the array unit and is used to control and collect the working status of the batteries connected to the array;

[0033] Expandable IO board, which provides expandable input and output interfaces, allowing the system to be connected to other devices or sensors;

[0034] Wiring ports, used to connect cables and lines of array units, and connect array units to battery electrode groups;

[0035] The battery electrode group is the core component of the battery, including the positive and negative electrodes of the battery; the array unit manages and controls the battery by connecting to these electrode groups;

[0036] The communication port provides a variety of communication interface or communication protocol options, including 485, MODBUS, and LAN.

[0037] Furthermore, in the core control module, the bus control unit is responsible for managing the communication between the internal modules and realizing the transmission of data and the sending of control instructions;

[0038] A battery data buffer unit, which is used to store and manage data from each single cell of the battery pack, including voltage, current, and temperature information;

[0039] Current / voltage control unit, which monitors and controls the current and voltage of the battery pack to ensure operation within a safe range and controls the charge and discharge of the battery pack as needed;

[0040] An array unit control unit is used to manage multiple battery packs or battery chips in the battery pack and coordinate the working status between them;

[0041] The communication unit is responsible for communicating with the external system to monitor and control the battery pack;

[0042] The power management unit manages the power supply of the core control module to ensure normal operation and provide backup power or fault protection when needed.

[0043] Beneficial effects:

[0044] 1. The intelligent battery management system of the present invention controls the neural network model to learn the patterns and rules of battery data through a deep learning algorithm, thereby achieving adaptive optimization; the model can adjust the operating parameters of the battery according to real-time battery data input to maximize the performance and life of the battery, improve the efficiency and reliability of the battery, and optimize energy management and use; the operating parameters are automatically adjusted according to battery data and environmental conditions without the need for human intervention, thereby reducing human errors and workload and improving the stability and efficiency of the system; at the same time, the model can select the best working strategy according to different scenarios and needs to achieve the optimal operation of the battery system; it can also promptly detect abnormal conditions in battery operation, such as overheating, overvoltage, over-discharge, etc., and adopt corresponding control strategies, and real-time monitoring and control help protect the safety and stability of the battery.

[0045] 2. The present invention sends the battery working mode instruction, and the information processing center can optimize the battery working mode according to real-time needs to achieve efficient energy management; transmit the battery working mode instruction in real time to achieve rapid response and real-time adjustment; upload the battery data to the information processing center, and the management personnel can remotely monitor the battery status and performance at any time. By comparing the actual charging and discharging data with the simulation data of the model, the accuracy and reliability of the intelligent battery management system control model can be verified. By comparing the difference between the actual data and the simulation data, the performance of the system can be evaluated to ensure the reliability and stability of the system in actual work.

[0046] 3. The general BMS control model automatic generation device of the present invention does not depend on the type of battery, but is only related to the number of battery arrays, the series connection method, and the working power support mode. With wide voltage and wide current adaptability, the number of battery arrays supported can be expanded to more than 4096 pcs. Therefore, the present invention can provide powerful battery management and control capabilities, adapt to various battery pack configurations and application scenarios, whether in the fields of industry, transportation, energy storage, etc., a battery management system with wide voltage, wide current adaptability and expandable battery array support will play an important role and improve the reliability, efficiency and performance of the system.

[0047] 4. In the present invention, after the universal BMS control model is generated, it is connected and operated by simulating load / real load to evaluate the battery working efficiency and determine the optimal state of BMS. If necessary, the parameters can be adjusted, and the closed-loop test is performed to lock the parameters after determining the optimal state. This means that the most suitable control parameter setting for the battery system is found, so that it can provide the best working efficiency and life in practical applications. Based on the universal BMS control model, the control strategy of the battery management system can be optimized, and the battery charging and discharging process can be more effectively managed by accurately predicting the behavior of the battery, thereby improving the efficiency and performance of the system.

[0048] 5. Traditionally, the development of BMS requires a lot of time and resources, including modeling of battery characteristics, design and verification of algorithms, etc. The general model automatic generation method of the present invention can automate these processes, reduce development time and workload, and has better scalability, which can quickly adapt to the needs of new battery technologies and different models. It has great potential in the field of new energy vehicles, can improve the performance of BMS, reduce costs, and support intelligent and optimized control, thereby promoting the development and application of new energy vehicle technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flowchart of a method for automatically generating a general BMS control model according to the present invention;

[0050] Figure 2 A structural diagram of a general BMS control model automatic generation system of the present invention;

[0051] Figure 3 It is an error curve diagram of the intelligent battery management system of the present invention controlling the neural network training process;

[0052] Figure 4 This is a diagram of the deviation value determination process of the present invention;

[0053] Figure 5 It is a schematic diagram of the composition of the array unit of the present invention. DETAILED DESCRIPTION

[0054] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. It should also be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0055] Refer to the attached Figure 1 , this embodiment provides a method for automatically generating a general BMS control model, comprising the following steps:

[0056] S1. Obtain battery data from several battery array units, attach battery number information, and upload to the information processing center in real time.

[0057] The battery-related data is obtained through sensors and other means, and the charge and discharge control of the battery pack is implemented. The battery data is uploaded in real time, and the time interval will be reduced accordingly. The battery data includes current, voltage, remaining power, battery cell temperature, average current, remaining power, implemented power consumption and change trends of related quantities; the change trends of related quantities include but are not limited to predictions, etc. The average current includes but is not limited to 1-second average current, 6-second average current, 12-second average current, etc.

[0058] At the same time, the information processing center module will send battery working mode instructions, which include: discharging, charging, decoding, etc. The core control module will transmit the battery working mode instructions to the array unit; the information transmission method can be wireless, wired, standard bus or custom bus.

[0059] By sending the battery working mode instruction, the information processing center can optimize the battery working mode according to the real-time demand to achieve efficient energy management; the battery working mode instruction is transmitted in real time to achieve rapid response and real-time adjustment; according to the actual situation, the wireless, wired, standard bus or custom bus information transmission mode is selected to ensure the reliability of data transmission; by uploading the battery data to the information processing center, the management personnel can remotely monitor the battery status and performance at any time.

[0060] S2. In the information processing center, the obtained battery data is pre-processed in real time by storage, sorting, statistics and analysis through the battery learning mode, and the battery charging and discharging process is repeatedly implemented.

[0061] The battery data information is stored in the database in real time; the stored battery data is sorted and organized in chronological order for subsequent analysis and statistics; the stored battery data is used for statistics and analysis to calculate indicators such as the overall charging time, discharging time and battery capacity; and finally the overall charging time, discharging time and battery capacity of the battery array are obtained.

[0062] Among them, the overall charging time is the total charging time of the battery array calculated according to the charging timestamp in the battery data, the overall discharging time is the total discharging time of the battery array calculated according to the discharging timestamp in the battery data, and the battery capacity is the capacity of the battery estimated by analyzing the battery discharge curve and current data, which can be calculated based on the battery discharge rate and the total discharge time of the battery pack.

[0063] The calculated overall charging time, discharging time, and battery capacity information are displayed and applied to the battery management system, which can help monitor the performance of the battery array, optimize the charging and discharging strategy, and predict the battery life and health status.

[0064] S3. According to the analysis of battery data, an intelligent battery management system control neural network model is established based on deep learning. The established intelligent battery management system control neural network model is used to complete the training process and obtain the initial BMS control model.

[0065] Through deep learning algorithms, the battery data is analyzed by building an intelligent battery management system control neural network model to perform model parameter operation analysis, and multiple sets of recommended optimal parameter configurations are given, including the optimal operating parameters in high power output, general power output and power-saving power output modes; among which, the optimal operating parameters include current, voltage, output time segment length, etc.

[0066] First, normalize each input data of the training sample:

[0067]

[0068] Among them, x0 is the original data, and x is the normalized data.

[0069] In this embodiment, the intelligent battery management system control neural network model includes an input layer, a hidden layer, and an output layer. The dimension of the input vector is 6, and the dimension of the output vector is 1; the number of neurons q in the hidden layer neural network and the number of neurons M in the input layer generally have the following approximate relationship: q = 2M + 1, so the number of neurons in the hidden layer is set to 13.

[0070] The transfer function of the neurons in the hidden layer uses the S-type tangent function tansig, and the transfer function of the neurons in the output layer uses the S-type logarithmic function logsig. The training function tranlm uses the Levenberg-Marquardt algorithm to train the network. Finally, the output of the output layer is the battery cell SOC. SOC (State of Charge) is an indicator used to represent the charging state of the battery, usually expressed as a percentage of the ratio between the current capacity of the battery and its rated capacity. Set the maximum number of iterations to 1000, the minimum expected error to 0.0001, and the learning efficiency to 0.1.

[0071] In this embodiment, the intelligent battery management system control neural network model is trained, wherein the input of the input layer is the battery information stored in the database, including the battery cell temperature, voltage, real-time current, 1-second average current, 6-second average current, and 12-second average current (including but not limited to this time interval); the battery cell temperature is temperature_val, the voltage is voltage_val, the real-time current is current_val, the 1-second average current is current_1s_val, the 6-second average current is current_6s_val, and the 12-second average current is current_12s_val.

[0072] Data synchronization is performed based on the time axis of the temperature signal, and the temperature time Temp_time and the current time current_time are further obtained, which represent the temperature voltage time voltage_time;

[0073] First, the outer loop for i=2:(size(temperature_val,1)-1) traverses the temperature data from the second time point to the second to last time point. The inner loop for j=1:size(voltage_time,1) is used to find the voltage time point synchronized with the current temperature time point (Temp_time(i,1)); by comparing the values ​​of voltage_time(j,1) and Temp_time(i,1), the position in the voltage time array is determined; once the matching position is found, a ratio can be calculated for linear interpolation at that time point to obtain the corresponding voltage value (voltage_temp) and SOC value (soc_temp).

[0074] Similarly, the inner loop for j=1:size(current_time,1) is used to find the current time point that is synchronized with the current temperature time point (Temp_time(i,1)). By comparing the values ​​of current_time(j,1) and Temp_time(i,1), the position in the current time array is determined. Once the matching position is found, a ratio can be calculated to perform linear interpolation at that time point to obtain the corresponding current value (current_temp) and other current values, such as current_1s_temp, current_6s_temp, and current12s_temp.

[0075] The present invention synchronously obtains the voltage, SOC and current values ​​at the corresponding time points according to the temperature time points, and uses the interpolation method to perform linear interpolation according to time to obtain the corresponding voltage and current values ​​at the current temperature, which can fill the gaps between the data to obtain a more continuous and accurate data sequence. Temperature is one of the important factors affecting battery performance. By synchronously obtaining the values ​​of other parameters according to the temperature time points, a complete data set can be obtained; accurately obtaining the voltage and current values ​​at the current temperature can provide more refined data support for battery management.

[0076] Finally, the denormalization is performed, and the SOC estimation result obtained by the neural network test is Test_INCA_simulation. After denormalization, Test_INCA_simulation can be compared with the test data Test_INCA_y, which is the real-time estimation of the cell SOC by the BMS controller based on the extended Kalman filter algorithm.

[0077] The anti-normalization formula is:

[0078]

[0079] The Kalman filter formula is used in the team leader prediction process: after 1000 iterations, the error is reduced to 0.0031253. Figure 3 ,The SOC estimation result obtained by the neural network test is compared with the SOC value obtained by the extended Kalman filter algorithm, and it is concluded that the estimation accuracy of SOC in steady-state conditions can be maintained within 5%.

[0080] In this embodiment, when the current changes dynamically, the estimated value of SOC fluctuates over a large range. Thus, by repeatedly implementing the above process several times (at least a hundred times) through the present invention, an array of fluctuation curves can be obtained to provide more comprehensive information, and the evaluation of battery charging and discharging behavior can be more robust and accurate. At the same time, in the time axis direction, the variance value is taken as an indicator reflecting the current volatility. Through the variance change on the time axis, the optimal working state of battery discharge can be identified.

[0081] After modeling and training in this step, the initial BMS control model is obtained, making the BMS control model more flexible and scalable, able to adapt to batteries of different types, specifications and usage conditions, and achieve more accurate and effective control.

[0082] The intelligent battery management system of the present invention controls the neural network model to learn the patterns and rules of battery data through a deep learning algorithm, thereby achieving adaptive optimization; the model can adjust the operating parameters of the battery according to real-time battery data input to maximize the performance and life of the battery, improve the efficiency and reliability of the battery, and optimize energy management and use; the operating parameters are automatically adjusted according to battery data and environmental conditions without the need for human intervention, thereby reducing human errors and workload and improving the stability and efficiency of the system; at the same time, the model can select the best working strategy according to different scenarios and needs to achieve the optimal operation of the battery system; it can also promptly detect abnormal conditions in battery operation, such as overheating, overvoltage, over-discharge, etc., and adopt corresponding control strategies, and real-time monitoring and control help protect the safety and stability of the battery.

[0083] S4. Generate the core working program and calling interface using the obtained initial BMS control model; and complete the battery data verification during charging and discharging through the obtained working parameters of the battery cell array, compare with the theoretical simulation effect, and record the deviation value of the parameter index.

[0084] Based on the obtained initial BMS control model, the core working program and calling interface are generated. The core control module controls the array unit to complete the charging and discharging data sampling and verification during multiple rounds of battery operation, and compares them with the theoretical simulation data of the model, while recording the parameter indicators and their deviation values.

[0085] Furthermore, a plan for multiple rounds of charging and discharging experiments is developed, including charging and discharging tests under different current, voltage and temperature conditions; experiments are conducted and actual charging and discharging data are recorded; and then charging and discharging under the same conditions are simulated and simulation data is generated.

[0086] Compare and analyze the simulated data and actual data, compare their consistency and accuracy during the charge and discharge process, and evaluate the differences between the simulated data and the actual data. Adjust and optimize the parameters of the model based on the comparison results between the simulated data and the actual data. Repeat multiple rounds of charge and discharge data verification and parameter adjustment until the model can meet the expected performance requirements in actual applications.

[0087] The present invention verifies the accuracy and reliability of the intelligent battery management system control model by comparing the actual charge and discharge data with the simulation data of the model. By comparing the differences between the actual data and the simulation data, the performance of the system can be evaluated, the prediction ability of the model and the effectiveness of the control strategy can be checked, and the reliability and stability of the system in actual work can be ensured. The charge and discharge rate, voltage range, charging cut-off condition and other parameters can also be adjusted according to the characteristics of the actual data and the analysis results of the simulation data to optimize the performance and life of the battery.

[0088] S5. Determine the deviation value and adjust the parameters of the BMS control model.

[0089] The parameters that can be adjusted here include: the number of hidden layers; the number of neurons in each layer; setting the training target error; the number of time intervals for displaying training results; the number of training times; the allowed training time; the learning rate; the power factor; the minimum performance gradient, etc.

[0090] According to the results of multiple rounds of charge and discharge data verification, the prediction deviation of the model is analyzed and determined; then according to the determination results, the model parameters or algorithms are adjusted to reduce the prediction deviation. Figure 4 Specifically, a threshold range is given in advance based on a large amount of experimental data and expert experience. When the deviation value is too large or too small and exceeds the range, the model needs to be retrained to improve the prediction accuracy of the model. The training set and the validation set are used for training, and the performance of the model on the validation set is observed until the accuracy and performance goals required by the model are met. When the deviation value meets the threshold range, it means that the accuracy required by the model is met, and there is no need for iterative training. The BMS control model parameters can be fine-tuned by introducing regularization or improving the learning rate, so as to further improve the performance and generalization ability of the model.

[0091] The present invention analyzes and determines the deviation of the model, adjusts the model parameters or algorithm according to the determination results, and performs parameter adjustment and verification through iteration, thereby improving prediction accuracy, optimizing control strategies, improving system efficiency and performance, and achieving sustainable optimization; it can enhance the reliability, efficiency and performance of the intelligent battery management system and provide more reliable battery management and control functions.

[0092] S6. Solidify the adjusted BMS control model parameters to generate an overall model that adapts to the usage of different batteries, that is, a universal BMS control model.

[0093] Furthermore, according to the adjusted BMS control model, its parameters are further corrected and solidified; and the use of the battery may be affected by many factors, including charging method, discharge load, ambient temperature, etc. By generating a general BMS control model, these factors can be comprehensively considered to more comprehensively analyze the working characteristics and performance of the battery; provide important reference for battery operation and management, and help make decisions, such as battery replacement time, charging strategy adjustment, etc., to maximize the battery life and performance.

[0094] Based on the general BMS control model, the control strategy of the battery management system can be optimized. By accurately predicting the behavior of the battery, the battery charging and discharging process can be more effectively managed to improve the efficiency and performance of the system.

[0095] In summary, in this embodiment, after the general BMS control model is generated, it is connected and operated by simulating load / real load, and runs in the downward deviation curve / matching curve / upward deviation curve respectively; the battery system is operated in the downward deviation curve, which means that during the charging and discharging process, the behavior of the battery system will be slightly lower than the expected curve; the battery system is operated in the completely matching curve, which means that the behavior of the battery system is completely consistent with the expected curve; the battery system is operated in the upward deviation curve, which means that the behavior of the battery system is slightly higher than the expected curve. Evaluate the battery working efficiency, determine the optimal state of BMS, adjust the parameters if necessary, conduct closed-loop testing, and lock the parameters after determining the optimal; this means that the most suitable control parameter setting for the battery system is found, so that it can provide the best working efficiency and life in actual applications.

[0096] See attached Figure 2 , this embodiment provides a general BMS control model automatic generation system, including the following contents:

[0097] Transmission bus, array group module, core control module, information processing center module;

[0098] The transmission bus is used to transmit data and instructions between various modules of the BMS. It supports two-way data communication links with multiple communication methods, including instructions, MODBUS / 485, LAN, etc.

[0099] The array group module is composed of array units and is responsible for uploading the collected array unit summary data and issuing work instructions to the array units.

[0100] Refer to the attached Figure 5 , the array unit includes a microcontroller system, an expandable IO board, a wiring port, a battery electrode group, and a communication port. All array units are first grouped in multiples of 8 (8, 16, 24, etc.), and the groups that are multiples of 2 (such as 4, 6, 8, 10, etc.) are linked to the core control module through a transmission bus. An array unit generally refers to a small package in which multiple battery columns are encapsulated together.

[0101] The array unit can be connected to electrodes and can be connected to test battery objects (since the number of test battery objects is variable, the system is scalable and can therefore adapt to battery arrays of different sizes); it also includes a microcontroller system, i.e. a programmable microcontroller, which is responsible for executing charging, discharging and other working instructions conveyed by the bus, so that the system can operate and control the battery as needed; and the system can transmit the battery's real-time remaining power, operating current, voltage, temperature and other battery data to the core control module at an adjustable time interval, providing real-time monitoring and feedback of the battery status; at the same time, signal transmission is completed through the transmission bus, which helps to ensure timely data transmission and monitoring.

[0102] The microcontroller system is embedded in the array unit and is responsible for executing various operation instructions for battery management, such as charging and discharging. The microcontroller system is the brain of the array unit and is used to control and collect the working status of the batteries connected to the array.

[0103] Expandable IO board, which provides expandable input and output interfaces, allowing the system to be connected to other devices or sensors, increasing the flexibility of the array unit to adapt to different external devices and input sources;

[0104] Wiring ports are used to connect cables and lines of array units, connect array units with other systems, battery electrode groups, etc., and ensure the physical connection of the system;

[0105] The battery electrode group is the core component of the battery, including the positive and negative electrodes of the battery; the array unit manages and controls the battery by connecting to these electrode groups;

[0106] The communication port provides a variety of communication interface or communication protocol options, including 485, MODBUS, LAN, etc.; it enables the array unit to communicate with other devices, systems or networks to achieve information exchange and sharing.

[0107] The core control module monitors the overall status of the battery pack in real time and takes corresponding control measures according to the predefined control strategy. It is the main control center of the BMS. The core control module includes the bus control unit, battery data buffer unit, current / voltage control unit, array unit control unit, communication unit, and power supply management unit.

[0108] The bus control unit is responsible for managing the communication between various internal modules and realizing the transmission of data and the sending of control instructions.

[0109] The battery data buffer unit is used to store and manage data from each single cell of the battery pack, including voltage, current, temperature and other information.

[0110] The current / voltage control unit monitors and controls the current and voltage of the battery pack to ensure operation within a safe range and controls the charge and discharge of the battery pack as needed.

[0111] The array unit control unit is used to manage multiple battery packs or battery chips in the battery pack and coordinate their working status to ensure the balance and stability of the overall battery pack.

[0112] The communication unit is responsible for communicating with external systems, such as data exchange between the on-board battery management system and the vehicle control system, to achieve monitoring and control of the battery pack.

[0113] The power management unit manages the power supply of the core control module to ensure its normal operation and provide backup power or fault protection when necessary.

[0114] The information processing center module is mainly responsible for data processing, analysis and modeling, and establishing a database of collected battery-related data. It is generally a computing server with GPU computing power; the learning unit analyzes battery data such as current, voltage, remaining power, real-time power consumption, temperature, etc.; and generates a battery management system control model in the model generation unit to help predict the performance, status and change trend of the battery; among which, the accuracy requirements are ma (milliampere), mv (millivolt), 1 / 10 degree, mah (milliampere hour), ms (milliseconds). The information processing center module includes a real-time battery database, a learning unit, a model generation unit, and a parameter adjustment unit. Highly accurate processing and analysis of battery data in order to achieve accurate monitoring and control of battery status, the integration of the learning unit and the model generation unit helps to achieve intelligent battery management and improve system performance and efficiency.

[0115] The automatic generation device implements unspecified multiple rounds of charge and discharge process detection on different battery array units, and establishes the BMS charge and discharge management control algorithm model of the battery array of this specification through the detection data, including the following components:

[0116] Battery array unit, core control center, information processing center, transmission bus;

[0117] The battery array unit is a collection of batteries. Each battery unit has a specific capacity and characteristics. It is a specific object for implementing charge and discharge control and obtaining charge and discharge data of each battery unit.

[0118] The core control center is responsible for monitoring and controlling the charging and discharging process of the battery array, and has the ability to connect and coordinate multiple battery array control units, as well as data reception and temporary storage functions;

[0119] The information processing center is responsible for processing the acquired battery data and executing algorithms to analyze and optimize charge and discharge management;

[0120] Transmission bus is a communication channel used to transmit data and instructions between various components.

[0121] The general BMS control model automatic generation device of the present invention does not depend on the type of battery, but is only related to the number of battery arrays, the series connection method, and the working power support mode. With wide voltage and wide current adaptability, the number of battery arrays supported can be expanded to more than 4096 pcs. Therefore, the present invention can provide powerful battery management and control capabilities, adapt to various battery pack configurations and application scenarios, whether in the fields of industry, transportation, energy storage, etc., a battery management system with wide voltage, wide current adaptability and expandable number of battery arrays supported will play an important role and improve the reliability, efficiency and performance of the system.

[0122] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0123] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.

[0124] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0125] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0126] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A general BMS control model automatic generation method, characterized in that: The following steps are involved: S1. Obtain battery data from multiple battery array units, attach battery number information, and upload to the information processing center module in real time; S2. In the information processing center module, the battery data obtained is stored, sorted, counted and analyzed in real time through the learning mode; Store battery data information in real time in the database; Sort the stored battery data and organize the data in chronological order; use the stored battery data to perform statistics and analysis, calculate the overall charging time, discharging time and battery capacity indicators; obtain the overall charging time, discharging time and battery capacity of the battery array; S3. Based on the analysis of battery data, an intelligent battery management system control neural network model is established based on deep learning to obtain an initial BMS control model; The intelligent battery management system control neural network model includes an input layer, a hidden layer, and an output layer. The battery data is subjected to a deep learning algorithm to construct an intelligent battery management system control neural network model for model parameter operation analysis, and multiple sets of parameter recommended configurations are given to obtain the initial BMS control model. S4. Generate the core working program and calling interface using the initial BMS control model obtained; complete the data verification during charging and discharging, compare it with the theoretical simulation effect, and record the deviation value of the parameter index; S5. Determine the deviation value and adjust the parameters of the BMS control model; S6. Solidify the adjusted BMS control model parameters to generate an overall model that adapts to the usage of different batteries, that is, a universal BMS control model.

2. A general BMS control model automatic generation method according to claim 1, characterized in that: The step S1 specifically includes: The information processing center module will send a battery operating mode instruction, which includes: discharging, charging, and decoding. The core control module will transmit the battery operating mode instruction to the array unit.

3. The method for automatically generating a general BMS control model according to claim 1, characterized in that: The step S4 specifically includes: According to the trained intelligent battery management system control neural network model, the core control module controls the array unit to complete the charge and discharge data verification during multiple rounds of battery operation, compare it with the theoretical simulation data of the initial BMS control model, and record the deviation value of the parameter index.

4. A method for automatically generating a general BMS control model according to claim 3, characterized in that: When the deviation value exceeds the threshold range, the model needs to be retrained until the deviation value meets the threshold range requirements; when the deviation value meets the threshold range, the BMS control model parameters are fine-tuned.

5. A general BMS control model automatic generation system, applied to the general BMS control model automatic generation method according to claim 1, characterized in that: Includes the following: Transmission bus, array group module, core control module, information processing center module; The transmission bus is used for data transmission and command communication between the core control module of the battery pack and each battery array, and can support a two-way data communication link with multiple communication modes, including IO connection, MODBUS / 485, and LAN; The array group module is composed of multiple array units and is responsible for uploading the collected array unit summary data and issuing work instructions for the array units; The array unit includes a microcontroller system, an expandable IO board, a wiring port, a battery electrode group, and a communication port; the array unit can be connected to electrodes and can be connected to a test battery object; the universal BMS control model automatic generation system can transmit the battery data of real-time remaining power, working current, voltage, and temperature of the battery to the core control unit at an adjustable time interval, providing real-time monitoring and feedback of the battery status; at the same time, signal transmission is completed through a transmission bus; The core control module monitors the overall status of the battery pack in real time and takes corresponding control measures according to the predefined control strategy; the core control module includes a bus control unit, a battery data buffer unit, a current / voltage control unit, an array unit control unit, a communication unit, and a power supply management unit; The information processing center module is mainly responsible for data processing, analysis and modeling, and establishing a database of collected battery-related data; analyzing current, voltage, remaining power, real-time power consumption, and temperature battery data through a learning unit; and generating a battery management system control model in a model generation unit to help predict battery performance, status, and change trends; the information processing center module includes a real-time battery database, a learning unit, a model generation unit, and a parameter adjustment unit.

6. A general BMS control model automatic generation system according to claim 5, characterized in that: In the array unit, the microcontrol system is embedded in the array unit and is responsible for executing various operation instructions for battery management. The microcontrol system is the brain of the array unit and is used to control and collect the working status of the battery connected to the array; The expandable IO board provides an expandable input and output interface, so that the system can be connected to other devices or sensors; The wiring port is used to connect the cables and lines of the array unit, and connect the array unit with the battery electrode group; The battery electrode group is the core component of the battery, including the positive electrode and the negative electrode of the battery; the array unit manages and controls the battery by connecting to these electrode groups; The communication port provides a variety of communication interface or communication protocol options, including 485, MODBUS, and LAN.

7. A general BMS control model automatic generation system according to claim 5, characterized in that: In the core control module, the bus control unit is responsible for managing the communication between the internal modules and realizing the transmission of data and the sending of control instructions; The battery data buffer unit is used to store and manage data from each single cell of the battery pack, including voltage, current, and temperature information; The current / voltage control unit monitors and controls the current and voltage of the battery pack to ensure operation within a safe range, and controls the charge and discharge of the battery pack as required; The array unit control unit is used to manage multiple battery packs or battery chips in the battery pack and coordinate the working states between them; The communication unit is responsible for communicating with the external system to monitor and control the battery pack; The power supply management unit manages the power supply of the core control module to ensure normal operation and provide backup power or fault protection when necessary.