New energy automobile battery pack intelligent equalization control method and system

Abstract

The invention discloses an intelligent equalization control method and system for a new energy automobile battery pack, and the method comprises the steps: obtaining the charge state data of each cell from a battery management system, and calculating the charge state difference value between cells through a preset feature extraction algorithm; the charge state difference value between the battery cells is compared with a first preset threshold value, and if the charge state difference value between the battery cells is larger than the first preset threshold value, a target battery cell combination is determined; according to the target battery cell combination, predicting a dynamic energy transfer path between the battery cells by adopting a decision tree model, and generating an energy transfer scheme; acquiring real-time monitoring data through an energy transfer scheme, and calculating energy transfer efficiency; and comparing the energy transfer efficiency with a second preset threshold value, and if the energy transfer efficiency reaches the second preset threshold value, triggering the active equalization circuit to perform energy distribution. Equalization efficiency and control precision are improved, energy distribution is optimized, energy consumption is reduced, and the service life of the battery pack is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and particularly discloses an intelligent equalization control method and system for a new energy vehicle battery pack. Background Art

[0002] The battery management system is crucial in electric vehicles and energy storage systems, directly affecting the performance, safety, and service life of the battery pack.

[0003] With the rapid development of the new energy industry, the equalization control of the battery pack has become a core technology to ensure battery consistency and extend the cycle life. Effective equalization control can not only improve energy utilization efficiency but also reduce potential safety hazards, providing guarantee for large-scale commercial applications. However, many current equalization methods have obvious deficiencies in practical applications. Traditional passive equalization dissipates energy through resistors, with low efficiency and generating a large amount of heat; active equalization can transfer energy, but the control is complex and the cost is high. These limitations make it difficult for the battery pack to achieve efficient and accurate state-of-charge balance under complex working conditions, restricting the overall performance of the system.

[0004] In the field of battery equalization, the core challenges focus on how to efficiently achieve energy distribution among battery cells while taking into account the computational complexity of the system and the hardware cost.

[0005] The difference in state of charge among battery cells is the primary problem, which directly leads to overcharging or over-discharging of some battery cells, accelerating aging.

[0006] Due to the need for real-time monitoring and dynamic adjustment, existing methods often face the problem of a sharp increase in computational load when dealing with large-scale battery cell groups, increasing system latency and energy consumption.

[0007] Furthermore, the increase in computational complexity requires more powerful hardware support, which raises the design and maintenance costs of the system. Especially under dynamic working conditions, it is difficult to quickly respond to changes in the state of battery cells.

[0008] Therefore, how to reduce the computational complexity and hardware cost while ensuring efficient energy transfer has become a key issue in battery equalization control research. Summary of the Invention

[0009] The present invention provides an intelligent equalization control method and system for a new energy vehicle battery pack, aiming to solve at least one defect existing in the above-mentioned prior art.

[0010] One aspect of the present invention relates to an intelligent equalization control method for a new energy vehicle battery pack, including the following steps: Obtain the state-of-charge data of each battery cell from the battery management system, and calculate the state-of-charge difference value among battery cells by using a preset feature extraction algorithm; Compare the state-of-charge difference value between cells with a first preset threshold. If the state-of-charge difference value between cells is greater than the first preset threshold, determine the target cell combination; According to the target cell combination, use a decision tree model to predict the dynamic energy transfer path between cells and generate an energy transfer scheme; Obtain real-time monitoring data through the energy transfer scheme and calculate the energy transfer efficiency; Compare the energy transfer efficiency with a second preset threshold. If the energy transfer efficiency reaches the second preset threshold, trigger the active balancing circuit to perform energy distribution.

[0011] Further, the steps of obtaining the state-of-charge data of each cell from the battery management system and calculating the state-of-charge difference value between cells using a preset feature extraction algorithm include: Obtain the state-of-charge data of each cell from the battery management system, classify and store the data according to the cell numbers to obtain a state data set; Use a preset feature extraction algorithm to extract the state-of-charge feature values of each cell from the state data set to obtain a feature value set; Through the first difference calculation formula D(i,j)=|F(i)-F(j)|, calculate the state-of-charge difference value between cells for the state-of-charge feature values of each cell in the feature value set, where D(i,j) represents the difference value between cell i and cell j, F(i) represents the feature value of cell i, and F(j) represents the feature value of cell j.

[0012] Further, the steps of comparing the state-of-charge difference value between cells with a first preset threshold and determining the target cell combination if the state-of-charge difference value between cells is greater than the first preset threshold include: Obtain the voltage data of each cell from the battery management system, classify and store the voltage data according to the cell numbers to obtain a voltage data set; Use a preset feature extraction algorithm to extract the voltage feature values of each cell from the voltage data set to obtain a voltage feature value set; Through the second difference calculation formula V(i,j)=|V(i)-V(j)|, calculate the voltage difference value between cells for the voltage feature values of each cell in the voltage feature value set to obtain a voltage difference value set, where V(i,j) represents the voltage difference value between cell i and cell j, V(i) represents the voltage feature value of cell i, and V(j) represents the voltage feature value of cell j; If any voltage difference value in the voltage difference value set is greater than the second preset threshold, determine the corresponding cells i and j as the target cell combination to obtain a target cell combination set.

[0013] Further, according to the target battery cell combination, the steps of predicting the dynamic energy transfer path between battery cells using a decision tree model and generating an energy transfer scheme include: Obtain the operating data of each battery cell from the battery management system, store it classified according to the battery cell number, and extract the dynamic voltage characteristic values of the operating data using a preset feature extraction algorithm to obtain a set of dynamic voltage characteristic values; If at least one characteristic value in the set of dynamic voltage characteristic values exceeds a preset dynamic threshold, classify and predict the target battery cell combination through the decision tree model to obtain a preliminary set of energy transfer paths; According to the preliminary set of energy transfer paths, use a path optimization algorithm to sort the paths, calculate the transfer time and energy loss values for the paths, and obtain an optimized set of energy transfer paths; Generate a corresponding energy transfer scheme through the optimized set of energy transfer paths, determine the transfer priority and execution order of the target battery cell combination, and obtain the final energy transfer scheme.

[0014] Further, the steps of obtaining real-time monitoring data through the energy transfer scheme and calculating the energy transfer efficiency include: Obtain the real-time monitoring data of the energy transfer scheme, calculate the energy transfer efficiency of each path using a preset efficiency calculation formula, and obtain a set of energy transfer efficiencies; Extract the paths with efficiencies lower than a preset threshold from the set of energy transfer efficiencies. If the efficiency of at least one path is lower than the threshold, use a path optimization algorithm to re-sort the paths to obtain an optimized set of paths; For the optimized set of paths, calculate the transfer time and energy loss values of each path. If the transfer time exceeds a preset time threshold, eliminate the corresponding path to obtain a filtered set of paths; According to the filtered set of paths, use a preset priority allocation algorithm to determine the transfer priority of each path and obtain a priority sorting result; Based on the priority sorting result and combined with the dynamic voltage characteristics in the battery operating data, judge the execution order of each path. If the dynamic voltage characteristic value exceeds a preset range, adjust the execution order to obtain a final set of execution orders; Extract the preferred path from the final set of execution orders, obtain its real-time monitoring data, and recalculate the energy transfer efficiency using the efficiency calculation formula.

[0015] Further, the steps of comparing the energy transfer efficiency with a second preset threshold and triggering the active balancing circuit for energy distribution if the energy transfer efficiency reaches the second preset threshold include: Obtain the energy transfer efficiency of each path from the real-time monitoring data, and use a preset efficiency calculation formula to calculate the energy transfer efficiency set of each path; the efficiency calculation formula is E = P_out / P_in, where E represents the energy transfer efficiency, P_out represents the output power, and P_in represents the input power; If the energy transfer efficiency of at least one path reaches the second preset threshold, a trigger signal is generated through a preset trigger signal generation algorithm to determine the startup state of the active equalization circuit; According to the trigger signal, using the control logic of the active equalization circuit, extract the dynamic voltage characteristics from the battery operation data, judge the energy distribution ratio of each path, and obtain the distribution ratio set; Through the distribution ratio set, combined with a preset path selection algorithm, adjust the energy transmission amount of each path to determine the final energy distribution execution order set.

[0016] Another aspect of the present invention relates to an intelligent equalization control system for a new energy vehicle battery pack, which is used to implement the above-mentioned intelligent equalization control method for a new energy vehicle battery pack. The intelligent equalization control system for a new energy vehicle battery pack includes: A first calculation module, which is used to obtain the state of charge data of each battery cell from the battery management system and calculate the state of charge difference value between battery cells by using a preset feature extraction algorithm; A determination module, which is used to compare the state of charge difference value between battery cells with a first preset threshold. If the state of charge difference value between battery cells is greater than the first preset threshold, determine the target battery cell combination; A generation module, which is used to predict the dynamic energy transfer path between battery cells according to the target battery cell combination by using a decision tree model and generate an energy transfer scheme; A second calculation module, which is used to obtain real-time monitoring data through the energy transfer scheme and calculate the energy transfer efficiency; A trigger module, which is used to compare the energy transfer efficiency with a second preset threshold. If the energy transfer efficiency reaches the second preset threshold, trigger the active equalization circuit to perform energy distribution.

[0017] Furthermore, the first calculation module includes: A first acquisition unit, which is used to obtain the state of charge data of each battery cell from the battery management system, classify and store the data according to the battery cell number, and obtain a state data set; A second acquisition unit, which is used to extract the state of charge characteristic values of each battery cell from the state data set by using a preset feature extraction algorithm and obtain a characteristic value set; A first calculation unit, configured to calculate the state-of-charge difference value between cells for the state-of-charge characteristic values of each cell in the set of characteristic values through a first difference calculation formula D(i,j)=|F(i)-F(j)|, where D(i,j) represents the difference value between cell i and cell j, F(i) represents the characteristic value of cell i, and F(j) represents the characteristic value of cell j.

[0018] Further, the determination module includes: A third acquisition unit, configured to acquire the voltage data of each cell from the battery management system, classify and store the voltage data according to the cell numbers, and obtain a voltage data set; A fourth acquisition unit, configured to extract the voltage characteristic values of each cell from the voltage data set by using a preset feature extraction algorithm, and obtain a set of voltage characteristic values; A second calculation unit, configured to calculate the voltage difference value between cells for the voltage characteristic values of each cell in the set of voltage characteristic values through a second difference calculation formula V(i,j)=|V(i)-V(j)|, and obtain a set of voltage difference values, where V(i,j) represents the voltage difference value between cell i and cell j, V(i) represents the voltage characteristic value of cell i, and V(j) represents the voltage characteristic value of cell j; A fifth acquisition unit, configured to determine the corresponding cells i and j as a target cell combination if any voltage difference value in the set of voltage difference values is greater than a second preset threshold, and obtain a set of target cell combinations.

[0019] Further, the generation module includes: A sixth acquisition unit, configured to acquire the operation data of each cell from the battery management system, classify and store the operation data according to the cell numbers, and extract the dynamic voltage characteristic values of the operation data by using a preset feature extraction algorithm, and obtain a set of dynamic voltage characteristic values; A seventh acquisition unit, configured to perform classification prediction on the target cell combination through a decision tree model if at least one characteristic value in the set of dynamic voltage characteristic values exceeds a preset dynamic threshold, and obtain a preliminary energy transfer path set; An eighth acquisition unit, configured to sort the paths according to a path optimization algorithm according to the preliminary energy transfer path set, calculate the transfer time and energy loss value for the paths, and obtain an optimized energy transfer path set; A ninth acquisition unit, configured to generate a corresponding energy transfer scheme through the optimized energy transfer path set, determine the transfer priority and execution order of the target cell combination, and obtain a final energy transfer scheme.

[0020] The beneficial effects achieved by the present invention are: The present invention provides an intelligent equalization control method and system for a new energy vehicle battery pack, which obtains the state of charge data of each battery cell from the battery management system and calculates the state of charge difference value between battery cells by using a preset feature extraction algorithm; compares the state of charge difference value between battery cells with a first preset threshold, and if the state of charge difference value between battery cells is greater than the first preset threshold, determines the target battery cell combination; according to the target battery cell combination, uses a decision tree model to predict the dynamic energy transfer path between battery cells and generates an energy transfer scheme; obtains real-time monitoring data through the energy transfer scheme and calculates the energy transfer efficiency; compares the energy transfer efficiency with a second preset threshold, and if the energy transfer efficiency reaches the second preset threshold, triggers the active equalization circuit to perform energy distribution. The intelligent equalization control method and system for the new energy vehicle battery pack provided by the present invention have the following specific beneficial effects: I. Improve equalization efficiency and control accuracy By calculating the state of charge (SOC) difference value between battery cells in real time through the feature extraction algorithm and combining the decision tree model to predict the dynamic energy transfer path, high-difference battery cell combinations can be accurately identified. Compared with the traditional passive equalization that relies on a fixed threshold to trigger, the grading threshold judgment (the first and second preset thresholds) realizes the quantitative evaluation of the difference degree, thereby dynamically adjusting the equalization strategy.

[0021] II. Optimize energy distribution and reduce energy consumption By adopting the active equalization circuit combined with the dynamic energy transfer path planning, the energy of the high-SOC battery cells can be directly transferred to the low-SOC battery cells instead of being dissipated through resistance, reducing energy loss. The real-time monitoring and feedback mechanism of the energy transfer efficiency further ensures the minimization of energy consumption during the equalization process.

[0022] III. Prolong the service life of the battery pack By continuously monitoring the SOC difference and quickly triggering equalization, the "barrel effect" caused by inconsistent capacity attenuation between battery cells is effectively alleviated, delaying the overall aging of the battery pack. Combining the analysis ability of the decision tree model for historical data, the degradation trend of battery cell performance can also be predicted to achieve preventive equalization. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic flowchart of an embodiment of an intelligent equalization control method for a new energy vehicle battery pack according to the present invention. DETAILED DESCRIPTION

[0024] In order to better understand the above technical solutions, the above technical solutions will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0025] As Figure 1 shown, a first embodiment of the present invention proposes an intelligent equalization control method for a new energy vehicle battery pack, including the following steps: Step S100: Obtain the state of charge (SOC) data of each battery cell from the battery management system, and calculate the SOC difference value between battery cells using a preset feature extraction algorithm.

[0026] The battery management system (BMS) is the core control system used to monitor, manage, and protect battery packs (such as lithium-ion batteries, lead-acid batteries, etc.), and is widely used in new energy vehicles, energy storage systems, consumer electronics and other fields. Its core goal is to ensure the safe, efficient, and reliable operation of the battery pack and extend the battery life.

[0027] The state of charge (SOC) data is the core parameter that describes the ratio of the remaining battery charge to the full charge capacity.

[0028] The feature extraction algorithm is a mathematical method for identifying and extracting key indicators or patterns from raw data, used to simplify the data dimension, highlight the core features to support subsequent analysis or decision-making.

[0029] The SOC difference value between battery cells is the core indicator to measure the dispersion degree of the remaining charge of each single battery cell in the same battery pack, defined as the difference (range) between the maximum and minimum SOC values of the battery cells in the group or quantization parameters such as the statistical distribution standard deviation.

[0030] Step S200: Compare the SOC difference value between battery cells with the first preset threshold. If the SOC difference value between battery cells is greater than the first preset threshold, determine the target battery cell combination.

[0031] The target battery cell combination is to integrate single battery cells into a battery module that meets the consistency requirements through a specific series-parallel connection method, aiming to optimize the overall performance and adapt to the application scenario requirements.

[0032] Step S300: According to the target battery cell combination, use a decision tree model to predict the dynamic energy transfer path between battery cells and generate an energy transfer scheme.

[0033] The decision tree model is a supervised learning algorithm based on a tree structure, which realizes classification or regression tasks by recursively dividing the data feature space. Its core logic is to simulate the human decision-making process and construct an interpretable prediction path with an "if-then" rule chain. In the field of battery management, this model is often used for battery cell state evaluation, fault diagnosis, and balance strategy optimization.

[0034] The dynamic energy transfer path refers to an intelligent management architecture that dynamically adjusts the energy flow direction based on the real-time system state (such as battery SOC, load demand, input power fluctuation, etc.). Its core goal is to balance system efficiency and safety by optimizing the energy distribution strategy.

[0035] An energy transfer scheme is a collection of strategies and technologies designed to achieve efficient energy flow and balanced distribution in a battery system or energy network. Its core lies in eliminating the energy differences between battery cells / components through a controllable energy transfer mechanism to ensure the overall performance and safety of the system.

[0036] Step S400: Obtain real-time monitoring data through the energy transfer scheme and calculate the energy transfer efficiency.

[0037] Real-time monitoring data refers to the operating parameters of the target system (such as energy network, industrial equipment) continuously collected by hardware devices such as sensors and intelligent terminals with a time resolution from milliseconds to seconds, and the data is instantaneously transmitted and analyzed through wired / wireless communication technologies to form a dynamic visual feedback and decision support for the system state.

[0038] The energy transfer efficiency refers to the ratio of the available energy effectively output to the total input energy during the energy transmission or conversion process, which is used to quantify the effectiveness of the energy transfer process.

[0039] Step S500: Compare the energy transfer efficiency with the second preset threshold. If the energy transfer efficiency reaches the second preset threshold, trigger the active equalization circuit to perform energy distribution.

[0040] The active equalization circuit is an electronic system that eliminates the energy differences between individual cells in a battery pack through controllable energy transfer technology. Its core lies in using active components (such as capacitors, inductors or DC-DC converters) to achieve lossless or low-loss energy redistribution, rather than simply dissipating excess energy.

[0041] Energy distribution refers to the dynamic adjustment and optimal configuration of energy between different units (such as battery cells, energy storage modules or load devices) in the system through active or passive control means to maintain the overall performance of the system, extend its life and improve energy utilization efficiency.

[0042] Furthermore, for the intelligent equalization control method of the energy vehicle battery pack provided in this embodiment, step S100 includes: Step S110: Obtain the state-of-charge data of each battery cell from the battery management system, classify and store the data according to the battery cell number to obtain a state data set.

[0043] Obtaining the state-of-charge data of each battery cell from the battery management system is one of the core functions of the battery management system. The state of charge reflects the percentage of the remaining capacity of the battery cell in the total capacity, which directly affects the performance and safety of the battery pack.

[0044] Exemplarily, an electric vehicle battery pack includes 100 battery cells. The battery management system collects data of each battery cell in real time through voltage, current, and temperature sensors, and generates state of charge values. For example, the state of charge of cell 1 is 80% and that of cell 2 is 78%. These data are stored in a database with timestamps and cell numbers as identifiers.

[0045] In a possible implementation, the state of charge data is classified and stored according to the cell number to generate a state data set.

[0046] It should be noted that classified storage needs to ensure data traceability. For example, the database uses the cell number as the primary key to create a table to store the state of charge data of each battery cell. The data of cell 1 is stored in the record with number 1, and the data of cell 2 is stored in the record with number 2. This method facilitates subsequent data retrieval and analysis, and can quickly locate abnormal cells, improving management efficiency.

[0047] Step S120: Use a preset feature extraction algorithm to extract the state of charge feature values of each battery cell from the state data set to obtain a feature value set.

[0048] Specifically, a preset feature extraction algorithm is used to extract the state of charge feature values from the state data set. The feature extraction algorithm can be a statistical analysis method, such as calculating the mean or change rate of the state of charge over a period of time.

[0049] In one embodiment, for cell 1, the state of charge data for the past 1 hour is selected, and the calculated mean is 79.5% and the standard deviation is 0.5%, which are used as feature values. Such feature values reflect the state of charge stability of the battery cell and help to identify battery cells with abnormal fluctuations.

[0050] Preferably, feature extraction can also combine the influence of temperature to generate comprehensive feature values, improving the analysis accuracy.

[0051] Step S130: Through the first difference calculation formula D(i,j)=|F(i)-F(j)|, for the state of charge feature values of each battery cell in the feature value set, calculate the state of charge difference value between battery cells, where D(i,j) represents the difference value between battery cell i and battery cell j, F(i) represents the feature value of battery cell i, and F(j) represents the feature value of battery cell j.

[0052] For example, calculate the state of charge difference value between battery cells through the difference calculation formula. The difference value reflects the balance of the state of charge between battery cells and is crucial for the balanced management of the battery pack.

[0053] In one embodiment, the characteristic value of cell 1 is 79.5%, and that of cell 2 is 78.5%. Then the difference value is |79.5 - 78.5| = 1%. If the difference value exceeds the threshold, such as 2%, the equalization circuit is triggered to discharge the cell with a high state of charge. The difference value analysis can effectively prevent the cells from overcharging or over-discharging and extend the battery life.

[0054] It can be understood that the above process needs to ensure the real-time and accuracy of data acquisition. For example, the sensor sampling frequency is set to 1 time per second, and the state of charge data is filtered by a filtering algorithm to remove noise, ensuring the reliability of feature extraction.

[0055] In addition, the automated processing of classified storage and feature extraction can significantly reduce the cost of manual intervention and improve the system response speed.

[0056] In a possible implementation, the difference value result can be used to dynamically adjust the battery management strategy. For example, when it is detected that the difference value between cell 3 and cell 4 is 3%, the system automatically allocates more charging current to cell 4 with a lower state of charge.

[0057] This adaptive management can optimize the overall performance of the battery pack and reduce energy waste.

[0058] Preferably, the implementation of the above technical solution can significantly improve the safety and service life of the battery pack. For example, through real-time monitoring and difference analysis, cells with abnormal states of charge can be detected in advance to avoid the risk of thermal runaway.

[0059] At the same time, the equalization management is accurately regulated based on the difference value, reducing the performance deviation between cells and ensuring the stability of the battery pack under high-load scenarios. These technical effects jointly support the reliability and efficiency of the battery management system.

[0060] Furthermore, the intelligent equalization control method for the energy vehicle battery pack provided in this embodiment, step S200 includes: Step S210, obtain the voltage data of each cell from the battery management system, and classify and store the voltage data according to the cell number to obtain a voltage data set.

[0061] In a possible implementation, classify and store the voltage data according to the cell number to generate a voltage data set.

[0062] It should be noted that the classified storage needs to ensure the traceability and efficiency of the data. Exemplarily, the database uses the cell number as the primary key to create a table to store the voltage data. The voltage data of cell 1 is stored in the record numbered 1, and that of cell 2 is stored in the record numbered 2. This method is convenient for quickly retrieving and analyzing abnormal cells.

[0063] Obtaining the voltage data of each battery cell from the battery management system is one of the core functions of the battery management system. The voltage data reflects the working state of the battery cell and directly affects the performance of the battery pack. For example, a battery pack of an electric vehicle contains 50 battery cells. The battery management system collects the voltage data of each battery cell in real time through high-precision voltage sensors. For example, the voltage of battery cell 1 is 3.65V and that of battery cell 2 is 3.62V. These data are stored in the database with timestamps and battery cell numbers as identifiers.

[0064] Step S220: Use a preset feature extraction algorithm to extract the voltage feature values of each battery cell from the voltage dataset to obtain a voltage feature value set.

[0065] Specifically, use a preset feature extraction algorithm to extract the voltage feature values of each battery cell from the voltage dataset. The feature extraction algorithm can be a statistical analysis method, such as calculating the mean or fluctuation range of the voltage over a period of time.

[0066] In one embodiment, for battery cell 1, select the voltage data of the past 30 minutes, calculate the mean value as 3.64V, the maximum value as 3.66V, and the minimum value as 3.62V, and use the mean value as the voltage feature value. This feature value reflects the stability of the battery cell voltage and helps to identify abnormal battery cells.

[0067] Preferably, feature extraction can be combined with current data to generate a comprehensive feature value to improve the analysis accuracy.

[0068] Step S230: Through the second difference calculation formula V(i,j)=|V(i)-V(j)|, for the voltage feature values of each battery cell in the voltage feature value set, calculate the voltage difference value between the battery cells to obtain a voltage difference value set, where V(i,j) represents the voltage difference value between battery cell i and battery cell j, V(i) represents the voltage feature value of battery cell i, and V(j) represents the voltage feature value of battery cell j.

[0069] Calculate the voltage difference value between the battery cells through the second difference calculation formula to obtain a voltage difference value set. For example, if the voltage feature value of battery cell 1 is 3.64V and that of battery cell 2 is 3.61V, then the difference value is |3.64 - 3.61| = 0.03V. If the second preset threshold is 0.05V, then this difference value does not exceed the threshold.

[0070] In one embodiment, if the feature value of battery cell 3 is 3.68V and that of battery cell 4 is 3.60V, and the difference value is 0.08V, exceeding the threshold, then battery cell 3 and battery cell 4 are determined as the target battery cell combination.

[0071] Step S240: If any voltage difference value in the voltage difference value set is greater than the second preset threshold, then determine the corresponding battery cells i and j as the target battery cell combination to obtain a target battery cell combination set.

[0072] It is understandable that the target battery cell combination set is used to identify battery cell pairs with voltage imbalance and trigger subsequent management strategies.

[0073] In a possible implementation, the analysis of the voltage difference value set can dynamically adjust the battery management strategy. For example, when it is detected that the difference value between battery cell 5 and battery cell 6 is 0.07V, exceeding the threshold, the system automatically allocates more charging current to the battery cell 6 with a lower voltage. This adaptive management can optimize the performance of the battery pack.

[0074] Preferably, the automated processing of classification storage and feature extraction can reduce the cost of manual intervention and improve the system response speed. For example, the sensor sampling frequency is set to 2 times per second, and the voltage data is processed by a filtering algorithm to remove noise, ensuring the reliability of feature extraction.

[0075] This method can quickly locate the battery cells with abnormal voltages and improve the management efficiency.

[0076] Furthermore, for the intelligent equalization control method of the energy vehicle battery pack provided in this embodiment, step S300 includes: Step S310: Obtain the operating data of each battery cell from the battery management system, classify and store it according to the battery cell number, and extract the dynamic voltage feature values of the operating data using a preset feature extraction algorithm to obtain a set of dynamic voltage feature values.

[0077] Exemplarily, obtaining the operating data of each battery cell from the battery management system is a key link to achieve battery performance optimization.

[0078] The operating data usually includes information such as voltage, current, and temperature, and these data are collected in real time through high-precision sensors. For example, a battery pack of an electric vehicle contains 100 battery cells, and the battery management system collects data once per second, recording the voltage of battery cell 1 as 3.63V, the current as 2.5A, and the temperature as 25°C. These data are identified by the battery cell number and timestamp to ensure traceability.

[0079] In a possible implementation, classifying and storing the operating data according to the battery cell number facilitates subsequent analysis. The database uses the battery cell number as the primary key to create a table to store the data. The operating data of battery cell 1 is stored in the record numbered 1, and the operating data of battery cell 2 is stored in the record numbered 2.

[0080] It should be noted that the classification storage needs to ensure efficient retrieval, and an indexing mechanism is used to optimize the query speed. For example, querying the voltage data of battery cell 50 in the past hour can be completed in only a few milliseconds.

[0081] Specifically, a preset feature extraction algorithm is used to extract the dynamic voltage feature values to generate a set of dynamic voltage feature values. The feature extraction algorithm can be based on time series analysis to extract the fluctuation frequency or change rate of the voltage.

[0082] In one embodiment, for the battery cell 1, the voltage data in the past 10 minutes is analyzed, and the voltage change rate is calculated to be 0.01 V / min as the dynamic voltage characteristic value. This characteristic value reflects the dynamic performance of the battery cell.

[0083] If the characteristic value of a certain battery cell is 0.03 V / min, exceeding the preset dynamic threshold of 0.02 V / min, it is marked as abnormal.

[0084] Step S320: If at least one characteristic value in the dynamic voltage characteristic value set exceeds the preset dynamic threshold, the target battery cell combination is classified and predicted through a decision tree model to obtain a preliminary energy transfer path set.

[0085] Preferably, the target battery cell combination is classified and predicted through a decision tree model to generate a preliminary energy transfer path set. For example, the dynamic characteristic values of battery cell 3 and battery cell 4 are 0.025 V / min and 0.015 V / min respectively, and the decision tree model predicts that energy transfer is required between the two, generating the path "battery cell 3 → battery cell 4".

[0086] It can be understood that the decision tree quickly identifies the battery cell pairs that need to be transferred through eigenvalue comparison and historical data training.

[0087] Step S330: According to the preliminary energy transfer path set, a path optimization algorithm is used to sort the paths, and the transfer time and energy loss value are calculated for the paths to obtain an optimized energy transfer path set.

[0088] In one embodiment, a path optimization algorithm is used to sort the preliminary energy transfer paths, and the transfer time and energy loss value are calculated. For example, the expected transfer time of the path "battery cell 3 → battery cell 4" is 5 minutes, and the energy loss is 0.02 Wh; the transfer time of another path "battery cell 5 → battery cell 6" is 6 minutes, and the loss is 0.03 Wh.

[0089] The optimization algorithm preferentially selects the path with lower loss to generate an optimized energy transfer path set.

[0090] Step S340: Through the optimized energy transfer path set, a corresponding energy transfer scheme is generated, the transfer priority and execution order of the target battery cell combination are determined, and the final energy transfer scheme is obtained.

[0091] For example, a final energy transfer scheme is generated through the optimized energy transfer path set to determine the transfer priority and execution order. The path from battery cell 3 to battery cell 4 is preferentially executed due to low loss, followed by the path from battery cell 5 to battery cell 6.

[0092] It should be noted that the solution can also adjust the priority in combination with the cell temperature data. For example, the priority of the cells with higher temperature is reduced. This adaptive solution can effectively balance the cell state and improve the overall performance of the battery pack.

[0093] Furthermore, for the intelligent equalization control method of the energy vehicle battery pack provided in this embodiment, step S400 includes: Step S410: Obtain the real-time monitoring data of the energy transfer solution, and calculate the energy transfer efficiency of each path using a preset efficiency calculation formula to obtain an energy transfer efficiency set.

[0094] Exemplarily, obtaining the real-time monitoring data of the energy transfer solution is the basis for ensuring the efficient operation of the battery pack. The real-time monitoring data includes the voltage, current, temperature of each cell, and the real-time state of the energy transfer path.

[0095] In a possible implementation, the battery management system collects data every second through high-precision sensors. For example, the voltage of cell 7 in a certain battery pack is 3.65V, the current is 2.3A, and the temperature is 26°C. These data are stored with timestamps and path numbers for subsequent efficiency calculation. It should be noted that real-time monitoring needs to ensure data consistency, and data loss or errors are avoided through a verification mechanism.

[0096] Specifically, calculate the energy transfer efficiency of each path using a preset efficiency calculation formula.

[0097] Efficiency calculation is usually based on the ratio of input energy to output energy. For example, for the path "cell 7 → cell 8", the input energy is 10Wh and the output energy is 9.5Wh, and the calculated efficiency is 95%. The efficiencies of all paths form an energy transfer efficiency set.

[0098] Step S420: Extract the paths with efficiencies lower than the preset threshold from the energy transfer efficiency set. If the efficiencies of at least one path are lower than the threshold, use a path optimization algorithm to reorder the paths to obtain an optimized path set.

[0099] Preferably, extract the paths with efficiencies lower than the preset threshold (such as 90%) from the efficiency set. For example, the efficiency of the path "cell 9 → cell 10" is 88%, which is lower than the threshold and needs further optimization.

[0100] In one embodiment, reorder the paths with efficiencies lower than the threshold using a path optimization algorithm. The path optimization algorithm can be based on factors such as energy loss and transmission distance. For example, for the path "cell 9 → cell 10", by adjusting the current distribution, the efficiency is increased to 92% after optimization. The optimized paths form a new set.

[0101] It should be noted that the optimization algorithm needs to balance efficiency and calculation speed to ensure real-time performance.

[0102] Step S430: For the optimized path set, calculate the transfer time and energy loss value of each path. If the transfer time exceeds the preset time threshold, eliminate the corresponding path to obtain the filtered path set.

[0103] It can be understood that for the optimized path set, calculate the transfer time and energy loss value of each path. For example, for the path "Cell 7 → Cell 8", the transfer time is 4 minutes and the loss is 0.015 Wh; for the path "Cell 9 → Cell 10", the transfer time is 7 minutes and the loss is 0.025 Wh.

[0104] If the transfer time exceeds the preset threshold (such as 5 minutes), eliminate the corresponding path to obtain the filtered path set. The path "Cell 9 → Cell 10" is eliminated due to exceeding the time limit.

[0105] Step S440: According to the filtered path set, use the preset priority assignment algorithm to determine the transfer priority of each path to obtain the priority sorting result.

[0106] For example, based on the filtered path set, use the priority assignment algorithm to determine the transfer priority.

[0107] The priority algorithm can sort according to the state of charge and efficiency of the battery cells. For example, the state of charge of Cell 7 is relatively high, and the path "Cell 7 → Cell 8" is given a high priority. The priority sorting result guides the path execution order.

[0108] Step S450: Through the priority sorting result, combined with the dynamic voltage characteristics in the battery operation data, judge the execution order of each path. If the dynamic voltage characteristic value exceeds the preset range, adjust the execution order to obtain the final execution order set.

[0109] In a possible implementation, adjust the execution order in combination with the dynamic voltage characteristic value. For example, the dynamic voltage characteristic value of Cell 8 is 0.02 V / min, which exceeds the preset range of 0.015 V / min, and its execution priority needs to be reduced. In the final execution order set, the path "Cell 7 → Cell 8" is ranked first.

[0110] Step S460: Extract the preferred path from the final execution order set, obtain its real-time monitoring data, and recalculate the energy transfer efficiency using the efficiency calculation formula.

[0111] Preferably, extract the preferred path from the final execution order set, re-obtain its real-time monitoring data and calculate the efficiency.

[0112] For example, the latest data of the path "Cell 7 → Cell 8" shows that the input energy is 10.2 Wh, the output energy is 9.7 Wh, and the efficiency is 95.1%. This kind of real-time verification ensures the accuracy of the scheme.

[0113] It should be noted that dynamically adjusting the execution order can effectively adapt to the changes in the cell state and improve the robustness of the solution.

[0114] Furthermore, for the intelligent equalization control method of the energy vehicle battery pack provided in this embodiment, step S500 includes: Step S510: Obtain the energy transfer efficiency of each path from the real-time monitoring data, and use a preset efficiency calculation formula to calculate the energy transfer efficiency set of each path; the efficiency calculation formula is E = P_out / P_in, where E represents the energy transfer efficiency, P_out represents the output power, and P_in represents the input power.

[0115] Exemplarily, obtaining the energy transfer efficiency of each path from the real-time monitoring data is one of the core functions of the battery management system. The energy transfer efficiency reflects the effectiveness of energy transmission among the cells in the battery pack. The real-time monitoring data usually includes parameters such as voltage, current, and power. For example, for the path "Cell 1 → Cell 2" in a certain battery pack, the input power is 100W and the output power is 95W. Through the preset formula E = P_out / P_in, the calculated efficiency is 95%. Similarly, for the path "Cell 3 → Cell 4", the input power is 120W and the output power is 108W, and the efficiency is 90%. These efficiency values form an efficiency set, providing the basis for subsequent optimization.

[0116] Step S520: If the energy transfer efficiency of at least one path reaches the second preset threshold, then generate a trigger signal through a preset trigger signal generation algorithm to determine the startup state of the active equalization circuit.

[0117] In a possible implementation manner, if the energy transfer efficiency of at least one path reaches the second preset threshold, such as 95%, then the active equalization circuit is started through the trigger signal generation algorithm.

[0118] The generation of the trigger signal is based on a comprehensive judgment of the efficiency data and the cell state. For example, when the efficiency of the path "Cell 1 → Cell 2" reaches 95%, the trigger algorithm analyzes that the state of charge of Cell 1 is 80% and that of Cell 2 is 60%, and generates a high-priority trigger signal to indicate the startup of the active equalization circuit to balance the energy distribution among the cells. This mechanism ensures that the circuit operates preferentially on the high-efficiency path.

[0119] Step S530: According to the trigger signal, adopt the control logic of the active equalization circuit to extract the dynamic voltage characteristics from the battery operation data, judge the energy distribution ratio of each path, and obtain the distribution ratio set.

[0120] Specifically, the control logic of the active balancing circuit extracts dynamic voltage characteristics from the battery operation data for judging the energy distribution ratio. The dynamic voltage characteristics reflect the voltage change trend of the battery cells during operation. For example, the dynamic voltage characteristic value of cell 1 is 0.01V / min, indicating that its state is stable and it is suitable as the energy output terminal; The characteristic value of cell 2 is 0.03V / min, indicating that its voltage changes relatively fast and it needs to receive energy preferentially.

[0121] Based on this, the control logic determines that the energy distribution ratio of the path "cell 1 → cell 2" is 70%, and the path "cell 3 → cell 4" is 30%, forming a set of distribution ratios. This distribution method effectively balances the energy demands among the battery cells.

[0122] Step S540: Through the set of distribution ratios, combined with a preset path selection algorithm, adjust the energy transmission amounts of each path to determine the final set of energy distribution execution sequences.

[0123] Preferably, the energy transmission amounts of each path are adjusted by combining the set of distribution ratios with the path selection algorithm.

[0124] The path selection algorithm comprehensively considers efficiency, distance, and the state of the battery cells. For example, for the path "cell 1 → cell 2", the algorithm preferentially allocates a higher transmission amount, such as 70W; the path "cell 3 → cell 4" is allocated 30W.

[0125] After adjustment, determine the final set of energy distribution execution sequences. For example, the path "cell 1 → cell 2" is ranked first due to high efficiency and high priority, and the path "cell 3 → cell 4" follows. This sequence ensures that the priority of energy transmission is consistent with the system requirements.

[0126] It can be understood that through multi-faceted analysis, the above embodiments form a rigorous logical chain from efficiency calculation to trigger signal generation, then to dynamic voltage characteristic extraction and path selection.

[0127] Each link supports each other, ensuring the accuracy and efficiency of the energy distribution of the battery pack. For example, the generation of the trigger signal depends on the accuracy of the efficiency data, and the determination of the distribution ratio is based on the real-time analysis of the dynamic voltage characteristics. This multi-level collaborative work improves the overall performance of the system and provides a reliable guarantee for the efficient operation of the battery pack.

[0128] Another aspect of the present invention relates to an intelligent equalization control system for a new energy vehicle battery pack, which is used to implement the above-mentioned intelligent equalization control method for a new energy vehicle battery pack. The intelligent equalization control system for a new energy vehicle battery pack includes a first calculation module, a determination module, a generation module, a second calculation module, and a trigger module. Among them, the first calculation module is used to obtain the state of charge data of each battery cell from the battery management system and calculate the state of charge difference value between battery cells by using a preset feature extraction algorithm; the determination module is used to compare the state of charge difference value between battery cells with a first preset threshold. If the state of charge difference value between battery cells is greater than the first preset threshold, the target battery cell combination is determined; the generation module is used to predict the dynamic energy transfer path between battery cells according to the target battery cell combination by using a decision tree model and generate an energy transfer scheme; the second calculation module is used to obtain real-time monitoring data through the energy transfer scheme and calculate the energy transfer efficiency; the trigger module is used to compare the energy transfer efficiency with a second preset threshold. If the energy transfer efficiency reaches the second preset threshold, the active equalization circuit is triggered to perform energy distribution.

[0129] Further, for the intelligent equalization control system for a new energy vehicle battery pack provided in this embodiment, the first calculation module includes a first acquisition unit, a second acquisition unit, and a first calculation unit. Among them, the first acquisition unit is used to obtain the state of charge data of each battery cell from the battery management system, classify and store the data according to the battery cell number, and obtain a state data set; the second acquisition unit is used to extract the state of charge characteristic values of each battery cell from the state data set by using a preset feature extraction algorithm to obtain a characteristic value set; the first calculation unit is used to calculate the state of charge difference value between battery cells for the state of charge characteristic values of each battery cell in the characteristic value set through the first difference calculation formula D(i,j)=|F(i)-F(j)|, where D(i,j) represents the difference value between battery cell i and battery cell j, F(i) represents the characteristic value of battery cell i, and F(j) represents the characteristic value of battery cell j.

[0130] Preferably, for the intelligent equalization control system of the new energy vehicle battery pack provided in this embodiment, the determination module includes a third acquisition unit, a fourth acquisition unit, a second calculation unit, and a fifth acquisition unit. Among them, the third acquisition unit is configured to acquire the voltage data of each battery cell from the battery management system, classify and store the voltage data according to the battery cell numbers, and obtain a voltage data set; the fourth acquisition unit is configured to extract the voltage characteristic values of each battery cell from the voltage data set by using a preset feature extraction algorithm, and obtain a voltage characteristic value set; the second calculation unit is configured to calculate the voltage difference value between battery cells for the voltage characteristic values of each battery cell in the voltage characteristic value set through the second difference calculation formula V(i,j)=|V(i)-V(j)|, and obtain a voltage difference value set, where V(i,j) represents the voltage difference value between battery cell i and battery cell j, V(i) represents the voltage characteristic value of battery cell i, and V(j) represents the voltage characteristic value of battery cell j; the fifth acquisition unit is configured to, if any voltage difference value in the voltage difference value set is greater than the second preset threshold, determine the corresponding battery cells i and j as the target battery cell combination, and obtain a target battery cell combination set.

[0131] Further, for the intelligent equalization control system of the new energy vehicle battery pack provided in this embodiment, the generation module includes a sixth acquisition unit, a seventh acquisition unit, an eighth acquisition unit, and a ninth acquisition unit. Among them, the sixth acquisition unit is configured to acquire the operation data of each battery cell from the battery management system, classify and store it according to the battery cell numbers, and extract the dynamic voltage characteristic values of the operation data by using a preset feature extraction algorithm, and obtain a dynamic voltage characteristic value set; the seventh acquisition unit is configured to, if at least one characteristic value in the dynamic voltage characteristic value set exceeds the preset dynamic threshold, classify and predict the target battery cell combination through a decision tree model, and obtain a preliminary energy transfer path set; the eighth acquisition unit is configured to sort the paths according to the preliminary energy transfer path set by using a path optimization algorithm, calculate the transfer time and energy loss value for the paths, and obtain an optimized energy transfer path set; the ninth acquisition unit is configured to generate a corresponding energy transfer scheme through the optimized energy transfer path set, determine the transfer priority and execution order of the target battery cell combination, and obtain a final energy transfer scheme.

[0132] The intelligent equalization control method and system for a new energy vehicle battery pack provided in this embodiment, compared with the prior art, obtain the state of charge data of each battery cell from the battery management system, and calculate the state of charge difference value between battery cells by using a preset feature extraction algorithm; compare the state of charge difference value between battery cells with a first preset threshold, and if the state of charge difference value between battery cells is greater than the first preset threshold, determine the target battery cell combination; according to the target battery cell combination, use a decision tree model to predict the dynamic energy transfer path between battery cells and generate an energy transfer scheme; obtain real-time monitoring data through the energy transfer scheme and calculate the energy transfer efficiency; compare the energy transfer efficiency with a second preset threshold, and if the energy transfer efficiency reaches the second preset threshold, trigger the active equalization circuit to perform energy distribution. The intelligent equalization control method and system for a new energy vehicle battery pack provided in this embodiment have the following specific beneficial effects: I. Improve equalization efficiency and control accuracy By calculating the state of charge (SOC) difference value between battery cells in real time through a feature extraction algorithm and combining a decision tree model to predict the dynamic energy transfer path, high-difference battery cell combinations can be accurately identified. Compared with traditional passive equalization that relies on a fixed threshold trigger, the quantitative evaluation of the difference degree is realized through hierarchical threshold judgment (the first and second preset thresholds), so as to dynamically adjust the equalization strategy.

[0133] II. Optimize energy distribution and reduce energy consumption By adopting an active equalization circuit combined with dynamic energy transfer path planning, the energy of high-SOC battery cells can be directly transferred to low-SOC battery cells instead of being dissipated through resistors, reducing energy loss. The real-time monitoring and feedback mechanism of energy transfer efficiency further ensures the minimization of energy consumption during the equalization process.

[0134] III. Prolong the service life of the battery pack By continuously monitoring the SOC difference and quickly triggering equalization, the "barrel effect" caused by inconsistent capacity attenuation between battery cells is effectively alleviated, delaying the overall aging of the battery pack. Combining the analysis ability of the decision tree model for historical data, the degradation trend of battery cell performance can also be predicted to achieve preventive equalization.

[0135] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention. Obviously, those skilled in the art can make various changes and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A new energy vehicle battery pack intelligent balancing control method, characterized in that: The following steps are involved: Obtain the state of charge data of each battery cell from the battery management system, and use a preset feature extraction algorithm to calculate the difference in state of charge between the batteries; Comparing the state of charge difference value between the battery cells with a first preset threshold value, and determining a target battery cell combination if the state of charge difference value between the battery cells is greater than the first preset threshold value; According to the target battery cell combination, a decision tree model is used to predict the dynamic energy transfer path between the battery cells and generate an energy transfer plan; Acquire real-time monitoring data through the energy transfer scheme and calculate energy transfer efficiency; The energy transfer efficiency is compared with a second preset threshold value, and if the energy transfer efficiency reaches the second preset threshold value, the active balancing circuit is triggered to perform energy distribution.

2. The intelligent balancing control method for a new energy vehicle battery pack according to claim 1, characterized in that: The step of acquiring the state of charge data of each battery cell from the battery management system and calculating the state of charge difference value between the battery cells using a preset feature extraction algorithm comprises: Acquire the state of charge data of each battery cell from the battery management system, classify and store the data according to the battery cell number, and obtain a state data set; Using a preset feature extraction algorithm, extracting the state of charge characteristic value of each battery cell from the state data set to obtain a characteristic value set; The first difference calculation formula D(i,j)=|F(i)-F(j)| is used to calculate the state of charge difference between the battery cells for the state of charge characteristic value of each battery cell in the characteristic value set, where D(i,j) represents the difference between battery cell i and battery cell j, F(i) represents the characteristic value of battery cell i, and F(j) represents the characteristic value of battery cell j.

3. The intelligent balancing control method for a new energy vehicle battery pack according to claim 1, characterized in that: Comparing the state of charge difference value between the battery cells with a first preset threshold value, if the state of charge difference value between the battery cells is greater than the first preset threshold value, the step of determining the target battery cell combination includes: Obtaining voltage data of each battery cell from a battery management system, and classifying and storing the voltage data according to the battery cell number to obtain a voltage data set; Using a preset feature extraction algorithm, extracting a voltage feature value of each battery cell from the voltage data set to obtain a voltage feature value set; The voltage difference value between the battery cells is calculated for the voltage characteristic value of each battery cell in the voltage characteristic value set by using the second difference calculation formula V(i,j)=|V(i)-V(j)|, to obtain a voltage difference value set, wherein V(i,j) represents the voltage difference value between battery cell i and battery cell j, V(i) represents the voltage characteristic value of battery cell i, and V(j) represents the voltage characteristic value of battery cell j; If any voltage difference value in the voltage difference value set is greater than a second preset threshold, the corresponding battery cell i and battery cell j are determined to be a target battery cell combination, and a target battery cell combination set is obtained.

4. The intelligent balancing control method for a new energy vehicle battery pack according to claim 1, characterized in that: According to the target battery cell combination, a decision tree model is used to predict the dynamic energy transfer path between the battery cells, and the steps of generating an energy transfer scheme include: Acquire the operating data of each battery cell from the battery management system, classify and store the data according to the battery cell number, and use a preset feature extraction algorithm to extract the dynamic voltage feature value of the operating data to obtain a dynamic voltage feature value set; If at least one characteristic value in the dynamic voltage characteristic value set exceeds a preset dynamic threshold, the target battery cell combination is classified and predicted by a decision tree model to obtain a preliminary energy transfer path set; According to the preliminary energy transfer path set, a path optimization algorithm is used to sort the paths, and the transfer time and energy loss value are calculated for the paths to obtain an optimized energy transfer path set; By optimizing the energy transfer path set, a corresponding energy transfer plan is generated, the transfer priority and execution order of the target battery cell combination are determined, and the final energy transfer plan is obtained.

5. The intelligent balancing control method for a new energy vehicle battery pack according to claim 1, characterized in that: The steps of obtaining real-time monitoring data through the energy transfer scheme and calculating the energy transfer efficiency include: Acquire real-time monitoring data of the energy transfer scheme, calculate the energy transfer efficiency of each path using a preset efficiency calculation formula, and obtain an energy transfer efficiency set; Extracting paths with a lower than a preset threshold value from the energy transfer efficiency set, and if the efficiency of at least one path is lower than the threshold value, reordering the paths using a path optimization algorithm to obtain an optimized path set; For the optimized path set, the transfer time and energy loss value of each path are calculated. If the transfer time exceeds the preset time threshold, the corresponding path is eliminated to obtain the filtered path set; According to the filtered path set, a preset priority allocation algorithm is used to determine the transfer priority of each path and obtain a priority sorting result; The execution order of each path is determined by combining the priority sorting result with the dynamic voltage characteristics in the battery operation data. If the dynamic voltage characteristic value exceeds a preset range, the execution order is adjusted to obtain a final execution order set. A preferred path is extracted from the final execution order set, its real-time monitoring data is obtained, and the energy transfer efficiency is recalculated using an efficiency calculation formula.

6. The intelligent balancing control method for a new energy vehicle battery pack according to claim 1, characterized in that: The step of comparing the energy transfer efficiency with a second preset threshold, and if the energy transfer efficiency reaches the second preset threshold, triggering the active balancing circuit to perform energy distribution includes: The energy transfer efficiency of each path is obtained from the real-time monitoring data, and a preset efficiency calculation formula is used to calculate the energy transfer efficiency set of each path; the efficiency calculation formula is E=P_out / P_in, where E represents the energy transfer efficiency, P_out represents the output power, and P_in represents the input power, and the energy transfer efficiency set of each path is calculated; If the energy transfer efficiency of at least one path reaches a second preset threshold, a trigger signal is generated through a preset trigger signal generation algorithm to determine the start-up state of the active balancing circuit; According to the trigger signal, the control logic of the active balancing circuit is used to extract the dynamic voltage characteristics from the battery operation data, determine the energy distribution ratio of each path, and obtain a distribution ratio set; The energy transmission amount of each path is adjusted through the allocation ratio set in combination with a preset path selection algorithm to determine the final energy allocation execution order set.

7. An intelligent balancing control system for a new energy vehicle battery pack, used to implement the intelligent balancing control method for a new energy vehicle battery pack as claimed in any one of claims 1 to 6, characterized in that: The new energy vehicle battery pack intelligent balancing control system includes: The first calculation module is used to obtain the state of charge data of each battery cell from the battery management system and calculate the difference value of the state of charge between the battery cells using a preset feature extraction algorithm; A determination module, configured to compare the state of charge difference value between the battery cells with a first preset threshold value, and determine a target battery cell combination if the state of charge difference value between the battery cells is greater than the first preset threshold value; A generation module, used to predict the dynamic energy transfer path between the cells using a decision tree model according to the target cell combination, and generate an energy transfer plan; A second calculation module, used to obtain real-time monitoring data through the energy transfer scheme and calculate the energy transfer efficiency; The trigger module is used to compare the energy transfer efficiency with a second preset threshold value, and if the energy transfer efficiency reaches the second preset threshold value, trigger the active balancing circuit to perform energy distribution.

8. The intelligent balancing control system for new energy vehicle battery packs as claimed in claim 7, characterized in that: The first calculation module includes: A first acquisition unit is used to acquire the state of charge data of each battery cell from the battery management system, classify and store the data according to the battery cell number, and obtain a state data set; A second acquisition unit is used to extract the charge state characteristic value of each battery cell from the state data set by using a preset feature extraction algorithm to obtain a characteristic value set; The first calculation unit is used to calculate the state of charge difference value between the battery cells according to the first difference calculation formula D(i,j)=|F(i)-F(j)|, for the state of charge characteristic value of each battery cell in the characteristic value set, wherein D(i,j) represents the difference value between battery cell i and battery cell j, F(i) represents the characteristic value of battery cell i, and F(j) represents the characteristic value of battery cell j.

9. The intelligent balancing control system for new energy vehicle battery packs as claimed in claim 7, characterized in that: The determination module comprises: A third acquisition unit is used to acquire voltage data of each battery cell from the battery management system, and classify and store the voltage data according to the battery cell number to obtain a voltage data set; A fourth acquisition unit, configured to extract a voltage characteristic value of each battery cell from the voltage data set by using a preset characteristic extraction algorithm to obtain a voltage characteristic value set; A second calculation unit is used to calculate the voltage difference value between the battery cells according to the voltage characteristic value of each battery cell in the voltage characteristic value set by a second difference calculation formula V(i,j)=|V(i)-V(j)|, so as to obtain a voltage difference value set, wherein V(i,j) represents the voltage difference value between battery cell i and battery cell j, V(i) represents the voltage characteristic value of battery cell i, and V(j) represents the voltage characteristic value of battery cell j; The fifth acquisition unit is configured to determine that the corresponding battery cell i and battery cell j are the target battery cell combination if any voltage difference value in the voltage difference value set is greater than a second preset threshold value, so as to obtain the target battery cell combination set.

10. The intelligent balancing control system for new energy vehicle battery packs as claimed in claim 7, characterized in that: The generation module comprises: a sixth acquisition unit, configured to acquire operation data of each battery cell from the battery management system, classify and store the data according to the battery cell number, and extract dynamic voltage characteristic values ​​of the operation data using a preset characteristic extraction algorithm to obtain a dynamic voltage characteristic value set; a seventh acquisition unit, configured to classify and predict the target battery cell combination through a decision tree model to obtain a preliminary energy transfer path set if at least one characteristic value in the dynamic voltage characteristic value set exceeds a preset dynamic threshold; An eighth acquisition unit is used to sort the paths according to the preliminary energy transfer path set by using a path optimization algorithm, calculate the transfer time and energy loss value for the paths, and obtain an optimized energy transfer path set; The ninth acquisition unit is used to generate a corresponding energy transfer plan through an optimized energy transfer path set, determine the transfer priority and execution order of the target battery cell combination, and obtain a final energy transfer plan.

Images