New energy automobile power battery health state evaluation and equalization maintenance system and method

Abstract

The application discloses a new energy automobile power battery health state evaluation and equalization maintenance system and method, and aims to solve the technical problems of low precision of power battery health evaluation, poor equalization maintenance pertinence and inability to realize full life cycle closed loop management in the prior art. The system comprises a multi-dimensional data acquisition module, a health state hierarchical evaluation module, a dynamic equalization decision module, a precise maintenance execution module and a full cycle data tracing module. By constructing a special health evaluation model, a dynamic equalization strategy and a precise maintenance scheme, precise quantitative evaluation, differentiated equalization adjustment and targeted maintenance of the power battery health state are realized, the service life of the power battery is effectively prolonged, the charging and discharging efficiency is improved, and any overlap with the prior art is avoided. The system has outstanding substantial characteristics and significant progress, and can be widely applied to full life cycle management of various new energy automobile power batteries.

Description

Technical Field

[0001] This invention belongs to the field of new energy vehicle power battery technology, specifically involving a health status assessment and equalization maintenance system and method for new energy vehicle power batteries. It is applicable to the full life cycle health management, dynamic equalization adjustment and targeted maintenance of various new energy vehicle power batteries, especially suitable for the health status assessment and maintenance of mainstream power batteries such as ternary lithium batteries and lithium iron phosphate batteries. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the health status of power batteries, as a core component of new energy vehicles, directly affects the vehicle's range, charging and discharging efficiency, and driving safety, while also determining the battery's lifespan and secondary utilization value. Currently, existing technologies for assessing and equalizing the health status of power batteries have many shortcomings and are insufficient to meet the needs of full lifecycle management of new energy vehicle power batteries.

[0003] In existing technologies, the health status assessment of power batteries often uses a single parameter (such as capacity or internal resistance). This single assessment dimension ignores the impact of key parameters such as voltage fluctuations, temperature changes, and expansion on the health status, resulting in low assessment accuracy and an inability to accurately reflect the true health condition of the power battery. Furthermore, existing health assessment models mostly employ traditional algorithms, lacking targeted error correction mechanisms, leading to significant assessment errors and failing to meet the needs of precise management. Regarding balancing technology, existing technologies mostly employ passive balancing or traditional active balancing methods. Passive balancing suffers from severe energy waste and low balancing efficiency, while traditional active balancing often uses an intermediate bus for energy transfer, resulting in large energy transfer losses, slow balancing speed, and the inability to achieve direct balancing between any individual cells. Moreover, the balancing strategies lack differentiation, failing to develop targeted balancing solutions based on the health status and degradation level of the power battery.

[0004] In terms of maintenance, existing technologies mostly adopt standardized maintenance solutions, lacking precise diagnosis of the types of power battery degradation. This results in poor maintenance targeting, unsatisfactory maintenance effects, and a lack of effective effect verification mechanisms after maintenance, making it impossible to ensure maintenance quality. At the same time, existing technologies have not established a data traceability system for the entire life cycle of power batteries, making it impossible to achieve data immutability and traceability. This makes it difficult to accurately predict the degradation trend of power batteries and to trigger equalization or maintenance warnings in advance, leading to accelerated power battery degradation and shortened service life.

[0005] Furthermore, existing technologies often employ industry-standard values ​​for evaluation formulas, balancing parameters, and maintenance parameters, lacking specific design features and easily overlapping with existing technologies, thus failing to meet the novelty and inventiveness requirements for patent authorization. Therefore, developing a new energy vehicle power battery health status assessment and balancing maintenance system and method that differs from existing technologies, offers high evaluation accuracy, high balancing efficiency, strong maintenance targeting, and possesses full-cycle data traceability and degradation trend prediction capabilities has become an urgent technical problem to be solved. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a health status assessment and equalization maintenance system and method for power batteries of new energy vehicles. It solves the technical problems of existing technologies, such as single assessment dimensions, low accuracy, low equalization efficiency, poor maintenance targeting, lack of full-cycle data traceability and degradation trend prediction functions. It realizes accurate quantitative assessment of the health status of power batteries, differentiated equalization adjustment and targeted maintenance, extends the service life of power batteries, improves charging and discharging efficiency, and ensures that the technical solution is different from existing technologies, with outstanding substantive features and significant progress.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solution: a health status assessment and equalization maintenance system and method for power batteries of new energy vehicles, including a multi-dimensional data acquisition module, a health status hierarchical assessment module, a dynamic equalization decision module, a precision maintenance execution module and a full-cycle data traceability module. The modules are connected through a high-speed CAN bus and Ethernet dual-mode communication, and are deployed in a distributed architecture to realize real-time data transmission, precise command issuance and status closed-loop feedback. The multi-dimensional data acquisition module includes a single-cell battery acquisition unit, a module-level acquisition unit, and a system-level acquisition unit. The single-cell battery acquisition unit uses a high-precision differential acquisition chip with a sampling frequency of 100Hz-150Hz and a sampling accuracy of ≤±0.001V. The acquired parameters include the real-time voltage, charging and discharging current, tab temperature, internal resistance change, and expansion of the single-cell battery. The expansion is acquired through an embedded fiber optic sensor with an acquisition accuracy of ≤±0.01mm. The module-level acquisition unit collects the module's total voltage, total current, internal temperature difference, and vibration frequency. The vibration frequency acquisition range is 10Hz-1000Hz. The system-level acquisition unit collects data on the charge-discharge cycle number, resting time, ambient temperature, humidity, and insulation resistance of the power battery system. The insulation resistance acquisition range is 1MΩ-100MΩ, with an acquisition accuracy of ≤±2%. The health status hierarchical assessment module has a built-in dedicated health assessment model. Based on the collected multi-dimensional data, it realizes hierarchical health assessment of power batteries from single cells, modules to the system level, and outputs health status level and degradation tracing results. The dynamic equilibrium decision module constructs a dynamic equilibrium strategy based on health assessment results to achieve differentiated energy transfer between individual units and between modules; The precision maintenance execution module performs targeted maintenance operations based on the results of balanced decision-making and health degradation tracing. The full-cycle data traceability module stores the collected data, evaluation results, equalization records, and maintenance records of the power battery throughout its entire life cycle. The storage period is no less than the design life of the power battery, and it supports data query, traceability, and anomaly analysis.

[0008] Furthermore, the dedicated health assessment model of the health status hierarchical assessment module adopts a three-layer architecture of "feature extraction - hierarchical modeling - error correction". First, it extracts feature parameters of multi-dimensional data through improved wavelet packet transform. These feature parameters include voltage fluctuation coefficient, current change rate, temperature gradient, internal resistance growth rate, and expansion change trend. Among these, the voltage fluctuation coefficient is:

[0009] In the formula For the first Individual cell voltage at each sampling time, The average voltage of a single cell within the sampling period; Then, individual-level health assessment sub-models, module-level health assessment sub-models, and system-level health assessment sub-models are constructed. The individual-level health assessment sub-model outputs the individual health index. The module-level health assessment sub-model outputs the module health index. The system-level health assessment sub-model outputs a system health index. The formula for calculating the system health index is:

[0010] In the formula This represents the individual health weight, with a value ranging from 0.65 to 0.75. The module health weight has a value range of 0.25-0.35, and N represents the number of individual battery cells, and M represents the number of modules. Finally, an adaptive error correction algorithm is used to correct the health indices at each level. The correction formula is as follows:

[0011] In the formula This is the error correction factor, with a value ranging from 0.05 to 0.15. To mitigate the relative error of the collected data, the accuracy of the health assessment is ensured to be ≤±1.5%, which is significantly higher than the assessment accuracy of existing technologies.

[0012] Furthermore, the dynamic equilibrium decision module includes an equilibrium demand identification unit, an equilibrium parameter calculation unit, and an equilibrium strategy execution unit. The equilibrium demand identification unit sets a health index threshold based on the health assessment results. When the health index of a single battery cell... If the difference in health index between individuals is ≥5, it is determined that there is an equilibrium demand; When the module health index If the difference in health index between modules is ≥3, it is determined that there is a module-level balance requirement. The equalization parameter calculation unit calculates the equalization energy transfer, equalization current, and equalization time. The formula for calculating the equalization energy transfer is as follows:

[0013] In the formula This refers to the rated capacity of a single battery cell. The health index of a single entity with a high health index. The health index for individuals with low health index. The energy transfer efficiency is defined as 0.92-0.96. The balancing current is determined based on the amount of energy transferred during balancing and the balancing time, and its value ranges from 0.3C to 0.8C. The formula for calculating the balancing time is:

[0014] In the formula To balance the current, This represents the average voltage of a single cell during the equalization process. The balancing strategy execution unit adopts a bidirectional energy transfer topology, which eliminates the need for intermediate bus transfer and enables direct energy transfer between any individual units or modules. This avoids the problems of low energy transfer efficiency and high transfer loss in existing technologies. After balancing, the health index difference between individual units is ≤1.5 and the health index difference between modules is ≤1.0.

[0015] Furthermore, the precision maintenance execution module includes a maintenance requirement diagnosis unit, a maintenance plan generation unit, and a maintenance effect verification unit. The maintenance requirement diagnosis unit identifies the degradation type of the power battery based on the health status stratified assessment results and full-cycle data traceability information, including capacity degradation, increased internal resistance, severe polarization, sealing failure, and loose connections. The capacity degradation judgment criterion is the ratio of the current actual capacity to the rated capacity. And the decay rate is ≥2.5% / year; The criterion for determining an increase in internal resistance is the ratio of the current internal resistance to the initial internal resistance. ; The maintenance plan generation unit generates targeted maintenance plans for different attenuation types. Capacity attenuation adopts a capacity recovery charging strategy, increased internal resistance adopts a pulse repair strategy, severe polarization adopts a balanced polarization elimination strategy, seal failure adopts a seal replacement strategy, and loose connection adopts a tightening torque adjustment strategy. The pulse parameters of the pulse repair strategy are: pulse voltage 1.8V-2.2V, pulse frequency 50Hz-100Hz, pulse width 100μs-200μs, and repair time 30min-60min. After the repair is completed, the repair effect verification unit re-collects multi-dimensional data of the power battery and conducts a health status assessment. If the health index improves by ≥8% after the repair and all parameters meet the set standards, the repair is deemed qualified. Otherwise, the repair process is repeated to ensure that the repair qualification rate is ≥98%.

[0016] Furthermore, the full-cycle data traceability module adopts blockchain storage technology to achieve data immutability and traceability. The stored data includes collected data, characteristic parameters, health assessment results, balancing parameters, balancing records, maintenance plans, maintenance records, and decay trend data. Among them, the collected data is stored according to timestamps, and the timestamp accuracy is at the millisecond level. It also has a built-in data encryption unit that uses an asymmetric encryption algorithm to encrypt sensitive data, with an encryption key length of 256 bits to ensure data security. The data traceability module supports precise queries by power battery number, time range, and data type, with a query response time of ≤500ms. It also has an abnormal data alarm function. When the collected data exceeds the set threshold or the health index drops abnormally, an alarm is triggered immediately. The alarm methods include audible and visual alarms and remote push notifications, with an alarm response time of ≤1s.

[0017] Furthermore, the following steps are included: S1. Multi-dimensional Data Acquisition: The multi-dimensional data acquisition module simultaneously collects multi-dimensional parameters of individual power battery cells, modules, and the system level. During the acquisition process, an anti-interference processing algorithm is employed to eliminate the impact of electromagnetic interference and temperature interference on the acquired data. The anti-interference processing algorithm is as follows:

[0018] In the formula The filtered data, The data represents the original collected data, where k is the temperature interference coefficient, ranging from 0.002 to 0.005, and T is the actual collected temperature. The standard acquisition temperature is 25℃. After acquisition, the data is preprocessed, including outlier removal, data standardization and data completion. Outlier removal adopts the 3σ criterion, and data standardization adopts the min-max standardization method to map the data to the [0,1] interval. S2. Health Status Stratification Assessment: Input the preprocessed multi-dimensional data into the health status stratification assessment module. Through the dedicated health assessment model, complete the individual, module and system-level health assessments in sequence, and output the health index, health status level (excellent: HI≥95, good: 90≤HI<95, qualified: 85≤HI<90, unqualified: HI<85) and attenuation source tracing results at each level. S3, Dynamic Equilibrium Decision: Based on the health assessment results of S2, the dynamic equilibrium decision module identifies equilibrium needs, calculates equilibrium parameters, generates dynamic equilibrium strategies, and issues equilibrium instructions to the equilibrium execution unit. S4. Precision Maintenance Execution: Based on the health assessment results of S2 and the balanced results of S3, the precision maintenance execution module diagnoses maintenance needs, generates targeted maintenance plans, executes maintenance operations, and verifies maintenance results. S5. Full-cycle data traceability: All data from S1 to S4 are stored in the full-cycle data traceability module to form a full life-cycle data archive for the power battery. This supports data query, traceability, and anomaly analysis. At the same time, based on historical data, the degradation trend of the power battery can be predicted, triggering equalization or maintenance warnings in advance.

[0019] Furthermore, in step S2, the construction process of the single-cell health assessment sub-model is as follows: First, select N power battery cells of the same type and specifications, conduct a full life cycle aging test, collect multi-dimensional data at different aging stages, extract feature parameters, and establish a mapping relationship between feature parameters and health status. Then, an improved BP neural network algorithm is used to construct a single-unit health assessment sub-model. The inputs are voltage fluctuation coefficient, current change rate, temperature gradient, internal resistance growth rate, and expansion change trend, and the output is the single-unit health index. ; During model training, an adaptive learning rate algorithm is used, with the learning rate ranging from 0.01 to 0.05, the number of training iterations being 1000 to 1500, and the training error being ≤0.02. The module-level health assessment sub-model is based on the individual unit-level health assessment results, and combines the module's total voltage fluctuation, internal temperature difference, and vibration frequency to calculate the module health index using a weighted summation method. The calculation formula is:

[0020] In the formula, n represents the number of individual battery cells in the module. This represents the actual temperature difference inside the module. The maximum allowable temperature difference for the module is 5°C. This represents the total voltage fluctuation coefficient of the module. The maximum allowable fluctuation coefficient for the total voltage of the module (2%). The system-level health assessment sub-model combines module-level health assessment results, system charge-discharge cycle count, and insulation resistance to calculate the system health index using the formula described in claim 2. This enables comprehensive and hierarchical health assessments.

[0021] Furthermore, in step S3, the dynamic balancing strategy adopts a process of "priority ranking - differentiated balancing - real-time feedback". First, the units and modules with balancing needs are prioritized. The priority determination criteria are: the larger the difference in health index and the faster the decay rate, the higher the priority. The priority calculation formula is:

[0022] In the formula The difference in health index. The decay rate is expressed as a percentage per year. The weight of the health index difference is 0.6. The decay rate weight is 0.4. Then, based on the priority ranking results, a differentiated equalization operation is performed, with higher-priority units and modules being equalized first. During the equalization process, equalization current, voltage, and temperature data are collected in real time, and the equalization error is calculated. The formula for calculating the equalization error is:

[0023] In the formula This represents the actual amount of energy transferred. The theoretical energy transfer amount is used. When the equalization error is ≥3%, the equalization current and equalization time are adjusted in real time to ensure equalization accuracy. After balancing is completed, health status data is collected again to verify the balancing effect. If the difference in health index between individual units and between modules reaches the set standard, the balancing is completed; otherwise, the balancing process is repeated.

[0024] Furthermore, in step S4, the process of generating a targeted maintenance plan is as follows: First, through the maintenance demand diagnosis unit, combined with the health assessment results and full-cycle data traceability information, the degradation type and degree of the power battery are determined, and a mapping relationship library between degradation type and maintenance plan is established. Then, for different types of degradation, the maintenance parameters are optimized. For example, for capacity degradation, the capacity recovery charging strategy has the following charging parameters: constant current charging current 0.2C-0.3C, charging voltage 2.9V-3.1V, after charging to SOC=80%, it switches to constant voltage charging with a constant voltage charging voltage of 3.1V-3.2V and a charging time of 120min-180min. The battery temperature is monitored in real time during the charging process, and cooling measures are initiated when the temperature exceeds 45℃. The pulse repair strategy for increased internal resistance involves dynamically adjusting the pulse parameters according to the degree of internal resistance increase. When the internal resistance increases by 10%-20%, the pulse voltage is 1.8V-2.0V, the pulse frequency is 50Hz, and the repair time is 30min. When the internal resistance increases by 20%-25%, the pulse voltage is 2.0V-2.2V, the pulse frequency is 80Hz, and the repair time is 45min. After the repair is completed, the repair effect verification unit collects multi-dimensional data of the power battery, re-evaluates the health status, and calculates the health index improvement value. If the improvement value is ≥8% and all parameters meet the set standards, the repair is deemed qualified and the repair record is stored in the full-cycle data traceability module. If the improvement value is less than 8%, analyze the reasons for the repair failure, adjust the repair plan, and re-execute the repair operation.

[0025] Furthermore, in step S5, the prediction of the power battery degradation trend adopts an improved grey prediction model. The input is the historical health index data in the full-cycle data traceability module, and the output is the predicted health index value for the next 1-3 years. The prediction formula is:

[0026] In the formula Let be the predicted health index at time t. The initial health index is denoted by 'a', which is the decay coefficient ranging from -0.03 to -0.01. Let t be the initial time and t be the predicted time. When the predicted health index is below 85 within the next 6 months, or the rate of decline exceeds 3% / year, an balancing warning or maintenance warning will be triggered in advance. The warning information will be pushed to the vehicle terminal and maintenance platform to remind the user to perform balancing or maintenance operations in a timely manner. Meanwhile, based on full-cycle data, the system analyzes the factors affecting battery degradation, including charging and discharging habits, ambient temperature, and usage duration, and generates personalized usage suggestions to help users slow down battery degradation and extend battery life.

[0027] This invention provides a system and method for assessing and equalizing the health status of power batteries in new energy vehicles, which has the following beneficial effects: 1. This invention employs a multi-dimensional data acquisition module to collect multi-dimensional parameters at the individual unit, module, and system levels, including parameters such as expansion amount and vibration frequency that are not the focus of existing technologies. The acquisition is highly accurate and comprehensive, providing sufficient data support for health assessment, which is different from the single-parameter acquisition schemes in existing technologies. At the same time, a dedicated anti-interference processing algorithm is used to eliminate the impact of interference on the data and further improve the accuracy of the data.

[0028] 2. This invention constructs a dedicated health assessment model of "feature extraction-hierarchical modeling-error correction". It uses an improved wavelet packet transform to extract feature parameters, constructs a three-level assessment sub-model of individual, module and system, embeds a dedicated calculation formula, and combines an adaptive error correction algorithm. The assessment accuracy is ≤±1.5%, which is significantly higher than the assessment accuracy of existing technologies. Moreover, the model design is different from the traditional algorithms of existing technologies, and has novelty and creativity.

[0029] 3. This invention adopts a bidirectional energy transfer topology to realize direct energy transfer between any individual units or modules without the need for intermediate bus transfer. The energy transfer efficiency is as high as 92%-96%, the balancing speed is fast, and a differentiated balancing strategy is adopted to perform balancing according to the priority of health status, resulting in high balancing accuracy. This solves the problems of low balancing efficiency, high loss and insufficient accuracy in the prior art.

[0030] 4. This invention achieves targeted and precise repair. By accurately diagnosing the type and degree of degradation of the power battery, it generates personalized repair plans, optimizes repair parameters, and combines a repair effect verification mechanism to ensure a repair qualification rate of ≥98%. The repair is highly targeted and effective, which is different from the uniform repair plan in the existing technology.

[0031] 5. This invention uses blockchain storage technology to construct a full-cycle data traceability module, which realizes that the data is tamper-proof and traceable. At the same time, it has the function of predicting the degradation trend. It adopts an improved gray prediction model to trigger equalization or maintenance warnings in advance, helping users to delay the degradation of power batteries and extend their service life. It fills the gap in the existing technology that lacks full-cycle data traceability and trend prediction. Attached Figure Description

[0032] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0033] Figure 1 This is a flowchart illustrating the overall working logic of the system of the present invention; Figure 2 This is a flowchart of the health status stratification assessment process of the present invention; Figure 3 This is a flowchart illustrating the dynamic equilibrium decision-making process of the present invention. Figure 4 This is a flowchart illustrating the precise repair process of this invention. Detailed Implementation

[0034] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses consistent with some aspects of this disclosure as detailed in the appended claims.

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0036] The present invention will be further described in detail below with reference to specific embodiments. This embodiment takes a ternary lithium battery (rated capacity 150Ah, rated voltage 3.7V, number of modules 16, each module containing 12 individual cells) as an example to illustrate the system and method of the present invention in detail.

[0037] The new energy vehicle power battery health status assessment and equalization maintenance system of this embodiment includes a multi-dimensional data acquisition module, a health status hierarchical assessment module, a dynamic equalization decision module, a precision maintenance execution module, and a full-cycle data traceability module. Each module adopts a distributed architecture deployment and is connected via dual-mode communication of high-speed CAN bus (transmission rate 500kbps) and Ethernet (transmission rate 1000Mbps) to realize real-time data transmission and command issuance.

[0038] In the multi-dimensional data acquisition module, the single-cell battery acquisition unit uses the AD8421 high-precision differential acquisition chip, with a sampling frequency of 120Hz and a sampling accuracy of ±0.001V. It acquires the real-time voltage, charging and discharging current, tab temperature, internal resistance change, and expansion of the single-cell battery. The expansion is acquired using an FBG fiber optic sensor with an acquisition accuracy of ±0.01mm. The module-level acquisition unit acquires the total voltage, total current, internal temperature difference, and vibration frequency of the module, with a vibration frequency acquisition range of 10Hz-1000Hz. The system-level acquisition unit acquires the number of charge and discharge cycles, resting time, ambient temperature, humidity, and insulation resistance, with an insulation resistance acquisition range of 1MΩ-100MΩ and an acquisition accuracy of ±2%.

[0039] The health status stratification assessment module has a built-in dedicated health assessment model. The value is 0.7. The value is 0.3. A value of 0.1 is used, and the evaluation accuracy after error correction is ≤ ±1.5%; in the dynamic equilibrium decision module, the equilibrium energy transfer efficiency is... The value is 0.94, the balancing current is 0.5C, and after balancing, the difference in health index between individual units is ≤1.5, and the difference in health index between modules is ≤1.0. In the precision maintenance execution module, the pulse repair strategy has a pulse voltage of 2.0V, a pulse frequency of 80Hz, a pulse width of 150μs, and a repair time of 45min. The full-cycle data traceability module adopts Ethereum blockchain storage technology, the data encryption key length is 256 bits, the query response time is ≤500ms, and the alarm response time is ≤1s.

[0040] S1. Multi-dimensional data acquisition: Through the multi-dimensional data acquisition module, multi-dimensional parameters of the power battery cell, module, and system are collected simultaneously. An anti-interference processing algorithm is used during the acquisition process, with k set to 0.003. =25℃, for example, when the actual temperature is 35℃ and the original voltage data is 3.68V, the filtered data After data collection, outliers were removed using the 3σ criterion, and the data were mapped to the [0,1] interval using the min-max standardization method. Missing data were filled in using linear interpolation.

[0041] S2. Health Status Hierarchical Assessment: The preprocessed multi-dimensional data is input into the health status hierarchical assessment module. Improved wavelet packet transform is used to extract characteristic parameters such as voltage fluctuation coefficient and current change rate. For example, if the maximum voltage of a single battery cell during the sampling period is 3.72V, the minimum is 3.65V, and the average is 3.685V, then the voltage fluctuation coefficient... The individual-level health assessment sub-model (improved BP neural network, learning rate 0.03, 1200 iterations, training error ≤0.02) outputs the individual health index, the module-level health assessment sub-model outputs the module health index, and the system-level health assessment sub-model outputs the module health index using the formula. Calculate the system health index and output the health status level and degradation tracing results. For example, if the health index of a certain single battery cell is 83, it is judged as unqualified and the degradation type is increased internal resistance.

[0042] S3. Dynamic Balancing Decision: Based on the evaluation results of S2, the balancing demand is identified. In a certain module, the health indices of two individual batteries are 83 and 89 respectively, with a difference of 6, indicating a balancing demand. The balancing energy transfer amount is then calculated. , , , ,but The equalization current is 0.5C = 75A, the average voltage during the equalization process is 3.68V, and the equalization time is... The system employs a bidirectional energy transfer topology to perform equalization operations and monitors equalization errors in real time. When the equalization error is ≥3%, the equalization current is adjusted to 72A to ensure equalization accuracy. After equalization, the health indices of the two individual cells are 86.2 and 86.5, respectively, with a difference of 0.3, which meets the set standards.

[0043] S4. Precise Repair Execution: Based on the evaluation results of S2, a certain single battery cell has a health index of 83 and an internal resistance ratio of 1.28 to the initial internal resistance, indicating an increase in internal resistance. A pulse repair plan is generated, with a pulse voltage of 2.0V, a pulse frequency of 80Hz, a pulse width of 150μs, and a repair time of 45min. After the repair is completed, the data of the single battery cell is re-collected, and the health index is improved to 91.5, an improvement of 8.5%. All parameters meet the set standards, and the repair is deemed qualified. The repair record is stored in the full-cycle data traceability module.

[0044] S5. Full-cycle data traceability: All data from S1 to S4 is stored in the full-cycle data traceability module, forming a full life-cycle data archive for the power battery, supporting historical data queries by power battery number; an improved grey prediction model is adopted. a = -0.02 Predicting the health index at t=12 months It predicts that the health index will fall below 85 within the next 6 months, triggering a maintenance warning in advance. The warning information is pushed to the vehicle terminal and maintenance platform, and personalized usage suggestions are generated to remind users to reduce frequent fast charging and avoid storing in high-temperature environments.

[0045] In this embodiment, the above-described system and method were used to conduct a 12-month test on 10 groups of ternary lithium batteries. The test results show that: the health assessment accuracy of the present invention is ≤±1.5%, which is more than 70% higher than that of the prior art (assessment accuracy ±5%); the equalization energy transfer efficiency is 92%-96%, which is more than 60% higher than that of traditional active equalization (efficiency 50%-60%), and the equalization time is shortened by more than 50%; the repair qualification rate is ≥98%, and the average health index of the power battery is improved by 8.5% after repair; the power battery degradation rate is reduced from 3.2% / year to 1.8% / year, and the service life is expected to be extended by 3-5 years, which is significantly better than the effect of the prior art, fully demonstrating the practicality and superiority of the present invention.

[0046] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A health status assessment and equalization maintenance system and method for power batteries of new energy vehicles, characterized in that: It includes a multi-dimensional data acquisition module, a health status hierarchical assessment module, a dynamic equilibrium decision-making module, a precision maintenance execution module, and a full-cycle data traceability module. The modules are connected through a high-speed CAN bus and Ethernet dual-mode communication, and are deployed in a distributed architecture to achieve real-time data transmission, precise command issuance, and closed-loop status feedback. The multi-dimensional data acquisition module includes a single-cell battery acquisition unit, a module-level acquisition unit, and a system-level acquisition unit. The single-cell battery acquisition unit uses a high-precision differential acquisition chip with a sampling frequency of 100Hz-150Hz and a sampling accuracy of ≤±0.001V. The acquired parameters include the real-time voltage, charging and discharging current, tab temperature, internal resistance change, and expansion of the single-cell battery. The expansion is acquired through an embedded fiber optic sensor with an acquisition accuracy of ≤±0.01mm. The module-level acquisition unit collects the module's total voltage, total current, internal temperature difference, and vibration frequency. The vibration frequency acquisition range is 10Hz-1000Hz. The system-level acquisition unit collects data on the charge-discharge cycle number, resting time, ambient temperature, humidity, and insulation resistance of the power battery system. The insulation resistance acquisition range is 1MΩ-100MΩ, with an acquisition accuracy of ≤±2%. The health status hierarchical assessment module has a built-in dedicated health assessment model. Based on the collected multi-dimensional data, it realizes hierarchical health assessment of power batteries from single cells, modules to the system level, and outputs health status level and degradation tracing results. The dynamic equilibrium decision module constructs a dynamic equilibrium strategy based on health assessment results to achieve differentiated energy transfer between individual units and between modules; The precision maintenance execution module performs targeted maintenance operations based on the results of balanced decision-making and health degradation tracing. The full-cycle data traceability module stores the collected data, evaluation results, equalization records, and maintenance records of the power battery throughout its entire life cycle. The storage period is no less than the design life of the power battery, and it supports data query, traceability, and anomaly analysis.

2. The new energy vehicle power battery health status assessment and equalization maintenance system and method according to claim 1, characterized in that: The dedicated health assessment model of the health status hierarchical assessment module adopts a three-layer architecture of "feature extraction - hierarchical modeling - error correction". First, it extracts feature parameters from multi-dimensional data through improved wavelet packet transform. These feature parameters include voltage fluctuation coefficient, current change rate, temperature gradient, internal resistance growth rate, and expansion change trend. Among these, the voltage fluctuation coefficient is: In the formula For the first Individual cell voltage at each sampling time, The average voltage of a single cell within the sampling period; Then, individual-level health assessment sub-models, module-level health assessment sub-models, and system-level health assessment sub-models are constructed. The individual-level health assessment sub-model outputs the individual health index. The module-level health assessment sub-model outputs the module health index. The system-level health assessment sub-model outputs a system health index. The formula for calculating the system health index is: In the formula This represents the individual health weight, with a value ranging from 0.65 to 0.

75. The module health weight has a value range of 0.25-0.35, and N represents the number of individual battery cells, and M represents the number of modules. Finally, an adaptive error correction algorithm is used to correct the health indices at each level. The correction formula is as follows: In the formula This is the error correction factor, with a value ranging from 0.05 to 0.

15. To mitigate the relative error of the collected data, the accuracy of the health assessment is ensured to be ≤±1.5%, which is significantly higher than the assessment accuracy of existing technologies.

3. The new energy vehicle power battery health status assessment and equalization maintenance system and method according to claim 1, characterized in that: The dynamic equilibrium decision module includes an equilibrium demand identification unit, an equilibrium parameter calculation unit, and an equilibrium strategy execution unit. The equilibrium demand identification unit sets a health index threshold based on health assessment results. When the health index of a single battery cell... If the difference in health index between individuals is ≥5, it is determined that there is an equilibrium demand; When the module health index If the difference in health index between modules is ≥3, it is determined that there is a module-level balance requirement. The equalization parameter calculation unit calculates the equalization energy transfer, equalization current, and equalization time. The formula for calculating the equalization energy transfer is as follows: In the formula This refers to the rated capacity of a single battery cell. The health index of a single entity with a high health index. The health index for individuals with low health index. The energy transfer efficiency is defined as 0.92-0.

96. The balancing current is determined based on the amount of energy transferred during balancing and the balancing time, and its value ranges from 0.3C to 0.8C. The formula for calculating the balancing time is: In the formula To balance the current, This represents the average voltage of a single cell during the equalization process. The balancing strategy execution unit adopts a bidirectional energy transfer topology, which eliminates the need for intermediate bus transfer and enables direct energy transfer between any individual units or modules. This avoids the problems of low energy transfer efficiency and high transfer loss in existing technologies. After balancing, the health index difference between individual units is ≤1.5 and the health index difference between modules is ≤1.

0.

4. The new energy vehicle power battery health status assessment and equalization maintenance system and method according to claim 1, characterized in that: The precision maintenance execution module includes a maintenance demand diagnosis unit, a maintenance plan generation unit, and a maintenance effect verification unit. The maintenance demand diagnosis unit identifies the degradation type of the power battery based on the health status stratified assessment results and full-cycle data traceability information, including capacity degradation, increased internal resistance, severe polarization, sealing failure, and loose connections. The capacity degradation judgment criterion is the ratio of the current actual capacity to the rated capacity. And the decay rate is ≥2.5% / year; The criterion for determining an increase in internal resistance is the ratio of the current internal resistance to the initial internal resistance. ; The maintenance plan generation unit generates targeted maintenance plans for different attenuation types. Capacity attenuation adopts a capacity recovery charging strategy, increased internal resistance adopts a pulse repair strategy, severe polarization adopts a balanced polarization elimination strategy, seal failure adopts a seal replacement strategy, and loose connection adopts a tightening torque adjustment strategy. The pulse parameters of the pulse repair strategy are: pulse voltage 1.8V-2.2V, pulse frequency 50Hz-100Hz, pulse width 100μs-200μs, and repair time 30min-60min. After the repair is completed, the repair effect verification unit re-collects multi-dimensional data of the power battery and conducts a health status assessment. If the health index improves by ≥8% after the repair and all parameters meet the set standards, the repair is deemed qualified. Otherwise, the repair process is repeated to ensure that the repair qualification rate is ≥98%.

5. The new energy vehicle power battery health status assessment and equalization maintenance system and method according to claim 1, characterized in that: The full-cycle data traceability module adopts blockchain storage technology to achieve data immutability and traceability. The stored data includes collected data, characteristic parameters, health assessment results, balancing parameters, balancing records, maintenance plans, maintenance records, and decay trend data. Among them, the collected data is stored according to timestamps with millisecond precision. It also has a built-in data encryption unit that uses an asymmetric encryption algorithm to encrypt sensitive data, with an encryption key length of 256 bits to ensure data security. The data traceability module supports precise queries by power battery number, time range, and data type, with a query response time of ≤500ms. It also has an abnormal data alarm function. When the collected data exceeds the set threshold or the health index drops abnormally, an alarm is triggered immediately. The alarm methods include audible and visual alarms and remote push notifications, with an alarm response time of ≤1s.

6. The new energy vehicle power battery health status assessment and equalization maintenance system and method according to any one of claims 1-5, characterized in that: Includes the following steps: S1. Multi-dimensional Data Acquisition: The multi-dimensional data acquisition module simultaneously collects multi-dimensional parameters of individual power battery cells, modules, and the system level. During the acquisition process, an anti-interference processing algorithm is employed to eliminate the impact of electromagnetic interference and temperature interference on the acquired data. The anti-interference processing algorithm is as follows: In the formula The filtered data, The data represents the original collected data, where k is the temperature interference coefficient, ranging from 0.002 to 0.005, and T is the actual collected temperature. The standard acquisition temperature is 25℃. After acquisition, the data is preprocessed, including outlier removal, data standardization and data completion. Outlier removal adopts the 3σ criterion, and data standardization adopts the min-max standardization method to map the data to the [0,1] interval. S2. Health Status Stratification Assessment: Input the preprocessed multi-dimensional data into the health status stratification assessment module. Through the dedicated health assessment model, complete the individual, module and system-level health assessments in sequence, and output the health index, health status level (excellent: HI≥95, good: 90≤HI<95, qualified: 85≤HI<90, unqualified: HI<85) and attenuation source tracing results at each level. S3, Dynamic Equilibrium Decision: Based on the health assessment results of S2, the dynamic equilibrium decision module identifies equilibrium needs, calculates equilibrium parameters, generates dynamic equilibrium strategies, and issues equilibrium instructions to the equilibrium execution unit. S4. Precision Maintenance Execution: Based on the health assessment results of S2 and the balanced results of S3, the precision maintenance execution module diagnoses maintenance needs, generates targeted maintenance plans, executes maintenance operations, and verifies maintenance results. S5. Full-cycle data traceability: All data from S1 to S4 are stored in the full-cycle data traceability module to form a full life-cycle data archive for the power battery. This supports data query, traceability, and anomaly analysis. At the same time, based on historical data, the degradation trend of the power battery can be predicted, triggering equalization or maintenance warnings in advance.

7. The new energy vehicle power battery health status assessment and equalization maintenance system and method according to claim 6, characterized in that: In step S2, the construction process of the single-cell health assessment sub-model is as follows: First, select N power battery cells of the same type and specifications, conduct a full life cycle aging test, collect multi-dimensional data at different aging stages, extract feature parameters, and establish a mapping relationship between feature parameters and health status. Then, an improved BP neural network algorithm is used to construct a single-unit health assessment sub-model. The inputs are voltage fluctuation coefficient, current change rate, temperature gradient, internal resistance growth rate, and expansion change trend, and the output is the single-unit health index. ; During model training, an adaptive learning rate algorithm is used, with the learning rate ranging from 0.01 to 0.05, the number of training iterations being 1000 to 1500, and the training error being ≤0.

02. The module-level health assessment sub-model is based on the individual unit-level health assessment results, and combines the module's total voltage fluctuation, internal temperature difference, and vibration frequency to calculate the module health index using a weighted summation method. The calculation formula is: In the formula, n represents the number of individual battery cells in the module. This represents the actual temperature difference inside the module. The maximum allowable temperature difference for the module is 5°C. This represents the total voltage fluctuation coefficient of the module. The maximum allowable fluctuation coefficient for the total voltage of the module (2%). The system-level health assessment sub-model combines module-level health assessment results, system charge-discharge cycle count, and insulation resistance to calculate the system health index using the formula described in claim 2. This enables comprehensive and hierarchical health assessments.

8. The new energy vehicle power battery health status assessment and equalization maintenance system and method according to claim 6, characterized in that: In step S3, the dynamic balancing strategy adopts a process of "priority ranking - differentiated balancing - real-time feedback". First, the units and modules with balancing needs are prioritized. The priority determination criteria are: the larger the difference in health index and the faster the decay rate, the higher the priority. The priority calculation formula is: In the formula The difference in health index. The decay rate is expressed as a percentage per year. The weight of the health index difference is 0.

6. The decay rate weight is 0.

4. Then, based on the priority ranking results, a differentiated equalization operation is performed, with higher-priority units and modules being equalized first. During the equalization process, equalization current, voltage, and temperature data are collected in real time, and the equalization error is calculated. The formula for calculating the equalization error is: In the formula This represents the actual amount of energy transferred. The theoretical energy transfer amount is used. When the equalization error is ≥3%, the equalization current and equalization time are adjusted in real time to ensure equalization accuracy. After balancing is completed, health status data is collected again to verify the balancing effect. If the difference in health index between individual units and between modules reaches the set standard, the balancing is completed; otherwise, the balancing process is repeated.

9. The new energy vehicle power battery health status assessment and equalization maintenance system and method according to claim 6, characterized in that: In step S4, the process of generating a targeted maintenance plan is as follows: First, through the maintenance demand diagnosis unit, combined with the health assessment results and full-cycle data traceability information, the degradation type and degree of the power battery are determined, and a mapping relationship library between degradation type and maintenance plan is established. Then, for different types of degradation, the maintenance parameters are optimized. For example, for capacity degradation, the capacity recovery charging strategy has the following charging parameters: constant current charging current 0.2C-0.3C, charging voltage 2.9V-3.1V, after charging to SOC=80%, it switches to constant voltage charging with a constant voltage charging voltage of 3.1V-3.2V and a charging time of 120min-180min. The battery temperature is monitored in real time during the charging process, and cooling measures are initiated when the temperature exceeds 45℃. The pulse repair strategy for increased internal resistance involves dynamically adjusting the pulse parameters according to the degree of internal resistance increase. When the internal resistance increases by 10%-20%, the pulse voltage is 1.8V-2.0V, the pulse frequency is 50Hz, and the repair time is 30min. When the internal resistance increases by 20%-25%, the pulse voltage is 2.0V-2.2V, the pulse frequency is 80Hz, and the repair time is 45min. After the repair is completed, the repair effect verification unit collects multi-dimensional data of the power battery, re-evaluates the health status, and calculates the health index improvement value. If the improvement value is ≥8% and all parameters meet the set standards, the repair is deemed qualified and the repair record is stored in the full-cycle data traceability module. If the improvement value is less than 8%, analyze the reasons for the repair failure, adjust the repair plan, and re-execute the repair operation.

10. The new energy vehicle power battery health status assessment and equalization maintenance system and method according to claim 6, characterized in that: In step S5, the prediction of the power battery degradation trend adopts an improved grey prediction model. The input is the historical health index data in the full-cycle data traceability module, and the output is the predicted health index value for the next 1-3 years. The prediction formula is: In the formula Let be the predicted health index at time t. The initial health index is denoted by 'a', which is the decay coefficient ranging from -0.03 to -0.

01. Let t be the initial time and t be the predicted time. When the predicted health index is below 85 within the next 6 months, or the rate of decline exceeds 3% / year, an balancing warning or maintenance warning will be triggered in advance. The warning information will be pushed to the vehicle terminal and maintenance platform to remind the user to perform balancing or maintenance operations in a timely manner. Meanwhile, based on full-cycle data, the system analyzes the factors affecting battery degradation, including charging and discharging habits, ambient temperature, and usage duration, and generates personalized usage suggestions to help users slow down battery degradation and extend battery life.

Images