Power assembly health management and control system of new energy automobile

Abstract

The invention discloses a new energy automobile power assembly health management and control system, and relates to the technical field of battery health, and the system comprises an initial state obtaining module which obtains the initial state parameter of a battery pack and determines the initial health state; the health state prediction module is used for collecting charging and discharging data for health assessment and generating a reference health state; the equalization simulation module is used for establishing a simulation model according to the equalization strategy and generating equalization data; the health state correction module is used for evaluating and correcting the reference health state by using the balanced data and acquiring a correction state updated in real time; and the health management and control module performs management and control based on the corrected health state and generates a health management and control strategy. Therefore, the technical effects of improving the monitoring reliability and reducing the monitoring difficulty and the management and control cost are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery health, and particularly to a power-train health management and control system for new energy vehicles. Background Art

[0002] The power train of a new energy vehicle is the core part of the vehicle, and its performance and health status are crucial to the performance and operation safety of the vehicle. The current power-train health management system usually relies on regular maintenance and the status monitoring of each single battery cell in the battery pack. The cost of sensor layout is high, the amount of generated data is large, and data processing is difficult, which not only increases the maintenance cost but also affects the energy density of the power train. There are technical problems such as great monitoring difficulty, low reliability, and high control cost. Summary of the Invention

[0003] The present invention provides a power-train health management and control system for new energy vehicles to solve the technical problems of great monitoring difficulty, low reliability, and high control cost in the prior art, and achieve the technical effects of improving monitoring reliability, reducing monitoring difficulty, and control cost.

[0004] The power-train health management and control system for new energy vehicles provided by the present invention includes: An initial state acquisition module, which is used to acquire the initial state parameters of a target battery pack and determine the initial health status of the battery pack, where the target battery pack includes multiple battery cells.

[0005] A health status prediction module, which is used to collect the charge and discharge data of the battery pack, evaluate the health status in combination with the initial health status, and generate a reference health status.

[0006] An equalization simulation module, which is used to establish an equalization simulation model according to the charge and discharge equalization strategy of the target battery pack in combination with the initial health status, and perform equalization simulation based on the equalization simulation model to generate simulation equalization data.

[0007] A health status correction module, which is used to evaluate and correct the reference health status by using the simulation equalization data to obtain a corrected health status, where the corrected health status is updated in real time.

[0008] A health management and control module, which is used to perform health management and control of the target battery pack based on the corrected health status and generate corresponding health management and control strategies.

[0009] In a feasible implementation manner, the steps of acquiring the initial state parameters of the target battery pack and determining the initial health status of the battery pack include: An interactive battery management component extracts the initial state parameters of multiple battery cells in a target battery pack, where the initial state parameters at least include capacity and internal resistance.

[0010] According to the initial state parameters, the initial cell health states of multiple battery cells are calculated iteratively, where the initial cell health states include absolute cell states and relative cell states.

[0011] Output the multiple initial cell health states as the initial health state.

[0012] In a feasible implementation, before collecting the charge and discharge data of the battery pack, combining the initial health state to perform a health state assessment, and generating a reference health state, the steps to be executed include: Obtain the sample aging data of the battery cells in the target battery pack.

[0013] Analyze the sample aging data, and extract a sample temperature sequence, a sample energy flow sequence, and a sample health state sequence, where the sequence interval of the sample health state sequence is greater than the sequence intervals of the extracted sample temperature sequence and the sample energy flow sequence.

[0014] Taking the sample health state sequence as the partitioning target, synchronously partition the sample temperature sequence and the sample energy flow sequence, and associate the sample health state sequence with the synchronous partitioning result to obtain multiple standard sample data groups.

[0015] Using the multiple standard sample data groups as training data, construct and train a health state prediction model.

[0016] In a feasible implementation, when collecting the charge and discharge data of the battery pack, combining the initial health state to perform a health state assessment, and generating a reference health state, the steps to be executed include: Interactively obtain the charge and discharge data, where the charge and discharge data includes a historical temperature sequence and a historical energy flow sequence.

[0017] Based on the elbow method, analyze the dispersion degree of the temperature data in the historical temperature sequence, determine the number of clustering clusters, and perform clustering partitioning on the charge and discharge data according to the number of clustering clusters to obtain multiple environmental charge and discharge data groups.

[0018] Input the multiple environmental charge and discharge data groups into the health state prediction model respectively, obtain multiple discrete prediction results, and fit the multiple discrete prediction results to the initial health state to obtain the reference health state.

[0019] In a feasible implementation manner, according to the charge and discharge equalization strategy of the target battery pack, an equalization simulation model is established in combination with the initial health state, and equalization simulation is performed based on the equalization simulation model to generate simulation equalization data. The execution steps include: Combining the initial health state and the reference health state, an equivalent health state is defined, where the equivalent health state is between the initial health state and the reference health state.

[0020] Construct a digital model of the target battery pack, and initialize the digital model of the target battery pack according to the charge and discharge equalization strategy to obtain the equalization simulation model.

[0021] Input the charge and discharge data into the equalization simulation model for equalization simulation of the target battery pack, and synchronously record the equalization data, and the output is the simulation equalization data, where the simulation equalization data includes an equalization temperature sequence and an equalization energy flow sequence.

[0022] In a feasible implementation manner, the reference health state is evaluated and corrected by using the simulation equalization data to obtain a corrected health state. The execution steps include: Input the simulation equalization data into the health state prediction model to obtain an equalization prediction result.

[0023] Fit the equalization prediction result to the reference health state to obtain the corrected health state, where the corrected health state includes the health states of multiple battery cells in the target battery pack.

[0024] In a feasible implementation manner, health control of the target battery pack is performed based on the corrected health state, and a corresponding health control strategy is generated, including: Calculate the remaining service life, where the remaining service life is the difference between the preset life control limit and the corrected health state.

[0025] Perform health control discrimination according to the remaining service life, and call the health control solution library to generate the health control strategy, where the health control discrimination includes consistency evaluation and control direction discrimination.

[0026] In a feasible implementation manner, the sample aging data includes intrinsic aging data and homologous aging data, where the homologous aging data is the aging data of the same type of battery cells as the battery cells of the target battery pack.

[0027] The present invention discloses a power train health control system for new energy vehicles, which solves the technical problems of large monitoring difficulty, low reliability, and high control cost, and realizes the technical effects of improving monitoring reliability, reducing monitoring difficulty and control cost. Description of the Drawings

[0028] Figure 1 It is a schematic structural diagram of the power-train health management and control system for the new energy vehicle of the present invention; Figure 2 It is a schematic process diagram for health status assessment in the power-train health management and control system of the new energy vehicle of the present invention.

[0029] Explanation of reference numerals: Initial state acquisition module 11, health status prediction module 12, balancing simulation module 13, health status correction module 14, health management and control module 15. Specific implementation manners

[0030] In the embodiments provided by the present invention, for the technical solution to solve the technical problems of difficult monitoring and low reliability, and high management and control cost existing in the prior art, the overall idea adopted is as follows: First, the initial state acquisition module acquires the initial state parameters of the target battery pack and determines the initial health status of the battery pack, where the target battery pack includes multiple battery cells. Then, the health status prediction module collects the charge and discharge data of the battery pack, combines the initial health status to conduct health status assessment, and generates a reference health status; then, the balancing simulation module establishes a balancing simulation model according to the charge and discharge balancing strategy of the target battery pack, combines the initial health status, and conducts balancing simulation based on this balancing simulation model to generate simulation balancing data; next, the health status correction module uses the simulation balancing data to evaluate and correct the reference health status to obtain the corrected health status, and this corrected health status is updated in real time; furthermore, the health management and control module conducts health management and control on the target battery pack based on the corrected health status and generates corresponding health management and control strategies; finally, through this health management and control strategy, the whole-life cycle health management of the battery pack is realized.

[0031] The above technical solution will be described in detail below in combination with the specification drawings and specific implementation manners to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments only used to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention. In addition, it should be noted that for the sake of description, only the parts related to the present invention are shown in the drawings rather than all of them.

[0032] Embodiment 1 Figure 1 It is a schematic process diagram of the power-train health management and control system for the new energy vehicle of the present invention, where the system includes: Initial state acquisition module 11, which is used to acquire the initial state parameters of the target battery pack and determine the initial health state of the battery pack. The target battery pack includes multiple battery cells.

[0033] Specifically, the target battery pack is the monitoring object of the system in the powertrain of a new energy vehicle. It consists of multiple individual battery cells, usually connected in series or parallel, and the multiple battery cells jointly provide the required power output. Therefore, the performance and health status of a single battery cell will directly affect the overall performance of the battery pack.

[0034] Specifically, the initial state parameters refer to the cell state parameters detected and recorded at the initial stage of battery pack assembly or charge and discharge. By obtaining these parameters, the system can evaluate the operating condition of the battery pack to obtain the health level of the battery cells in the initial state, which is the basic data of the battery cell performance.

[0035] In some embodiments, to acquire the initial state parameters of the target battery pack and determine the initial health state of the battery pack, the execution steps of the initial state acquisition module 11 include: Interact with the battery management component to extract the initial state parameters of multiple battery cells in the target battery pack. The initial state parameters at least include capacity and internal resistance.

[0036] According to the initial state parameters, traverse and calculate the initial cell health states of multiple battery cells. The initial cell health states include absolute cell state and relative cell state.

[0037] Output the multiple initial cell health states as the initial health state.

[0038] Specifically, the system acquires the initial state parameters of each battery cell in the target battery pack by interacting with the battery management component (BMS). The initial state parameters are the state parameters of the battery cells when the battery pack is encapsulated or manufactured, which reflect the basic performance indicators of each battery cell in the battery pack. In other words, the initial state parameters are the physical parameters of multiple battery cells in the target battery pack.

[0039] Specifically, for the extracted initial state parameters, calculate the health states of multiple battery cells in the target battery pack one by one. The health state evaluation of each battery cell is divided into absolute cell state and relative cell state, which respectively evaluate the initial state of the battery cell and its performance relative to the standard. Among them, the absolute cell state refers to the actual performance data of the battery cell. For example, the capacity of the battery cell (such as the actual capacity is 10 Ah) is its absolute cell state, which reflects the direct performance of the battery cell. Preferably, the cell capacity or cell internal resistance is used as the absolute cell state to further determine the absolute health state of the battery cell.

[0040] Specifically, the relative cell state refers to the performance of the absolute cell state of the cell relative to the rated state. For example, the achievement rate of the cell capacity relative to its rated capacity is the relative health state of the cell. If the actual capacity of the cell is 10 Ah (absolute cell state) and the rated capacity corresponding to the cell model is 9 Ah, then the relative cell state of this cell is 111%.

[0041] Exemplarily, in addition to the capacity, the relative change in internal resistance is also an important indicator for evaluating the health of the cell. Therefore, the initial internal resistance of the cell can be compared with the standard value to determine the relative health state of the cell.

[0042] Specifically, each cell is traversed and calculated to determine its absolute health state and relative health state respectively. After completing the traversal calculation of multiple cells, the system aggregates the health state results of each cell to generate the initial health state of the target battery pack. This initial health state reflects the health condition and performance level when the battery pack is manufactured, that is, the health state of the battery pack before use, which is a key reference for subsequent use.

[0043] The health state prediction module 12, which is used to collect the charge and discharge data of the battery pack, combine with the initial health state to conduct a health state assessment, and generate a benchmark health state.

[0044] Specifically, the charge and discharge data of the battery pack is obtained. This charge and discharge data contains all the charge and discharge parameters of the target battery pack during use, such as current, voltage, temperature, and the corresponding timestamps, which reflects the energy flow and load conditions during the use of the target battery pack.

[0045] Specifically, by combining the initial health state with the charge and discharge data, according to the energy flow, load level, and environmental conditions, the health state of the target battery pack after experiencing the charge and discharge data is predicted as the benchmark health state. The above steps, compared with collecting the state parameters of each cell, omit the process of collecting data from a large number of cell monomers and analyzing and processing the obtained large amount of data, which helps to reduce the computational cost and sensing cost required to obtain the state of the target battery pack.

[0046] In some embodiments, before collecting the charge and discharge data of the battery pack, combining with the initial health state to conduct a health state assessment, and generating a benchmark health state, the execution steps of the system include: Obtain the sample aging data of the cells in the target battery pack.

[0047] Parse the sample aging data, and extract the sample temperature sequence, sample energy flow sequence, and sample health state sequence. Among them, the sequence interval of the sample health state sequence is greater than the sequence intervals of the extracted sample temperature sequence and the sample energy flow sequence.

[0048] Using the sample health status sequence as the partitioning target, synchronously partition the sample temperature sequence and the sample energy flow sequence, and associate the sample health status sequence with the synchronous partitioning result to obtain multiple standard sample data sets.

[0049] Using multiple said standard sample data sets as training data, construct and train a health status prediction model.

[0050] Specifically, the sample aging data is historical data containing the characteristics of cell degradation, including cell records from experimental tests or long-term operation. By analyzing and learning the sample aging data of the cells in the target battery pack, the mapping relationship between the working state of the battery and the aging process of the cells can be understood.

[0051] Specifically, the sample aging data includes the following core parameters: Temperature, which is used to reflect the temperature change of the cell under different operating environments. Temperature has a significant impact on cell aging, and high temperature conditions will accelerate aging. Energy flow, the energy flow of the cell refers to the change of electrical energy during the charge and discharge process, including charging energy and discharging energy. These data can reflect the load level of the cell during cyclic use. Health status, which is used to reflect the aging degree of the cell with cycles, such as cell capacity, cell internal resistance, etc.

[0052] Specifically, after obtaining the sample aging data, parse and extract the time series of each key index, including temperature, energy flow, and health status, for subsequent model training. Among them, the temperature sequence is the temperature change data of the cell at different time periods, formed by arranging the temperature values in the aging data in chronological order. The temperature sequence is used to introduce the contribution of the temperature environment to the performance degradation of the cell. The energy flow sequence is the energy change during the charge and discharge process of the cell, reflecting the charge and discharge rate, voltage, and current conditions of the cell in each cycle, representing the load level of the cell during the charge and discharge process (such as charge and discharge rate). Charge and discharge under high load will accelerate the decline of the cell health. The sample health status sequence is the change of the health status of the cell during aging. Exemplarily, it is represented by the ratio of the actual capacity of the cell to the rated capacity.

[0053] Optionally, the data interval of the sample health status sequence is larger than that of the temperature sequence and the energy flow sequence because the health status of the cell usually changes over a longer time period, while the temperature and energy flow data change more frequently. At the same time, the measurement of the actual capacity of the cell is relatively complex and not suitable for a high data acquisition frequency.

[0054] Specifically, taking the time points of each data in the sample health status sequence as landmark points, the sample temperature sequence and the sample energy flow sequence are synchronously divided within the corresponding time periods, and the synchronous division results are aligned with the sample health status sequence to form a standardized data set. The standardized data set includes multiple data groups, and each data group corresponds to a measurable and perceivable change in the cell health status. In other words, each standard sample data group contains the temperature, energy flow characteristics, and health status of the cell within a specific time period.

[0055] Through the above division process, it helps to split complex and continuous sample data into multiple relatively simple and small-volume data groupings, which is convenient for subsequent training.

[0056] Furthermore, a health status prediction model is constructed and trained based on the standard sample data groups. This model is used to predict the future health status of the battery pack, that is, based on data such as temperature and energy flow during the use of the battery pack, predict the health status of the battery pack at the time point after use. Among them, the input features of the model include the sample temperature sequence and the sample energy flow sequence, and the sample health status sequence is used as the target output. By learning the implicit mapping relationship in the standard sample data groups, the aging pattern of the cell is gradually mastered.

[0057] Exemplarily, the health status prediction model can adopt a variety of machine learning or deep learning algorithms, such as random forest, support vector machine (SVM), neural network (such as LSTM or RNN), etc. The specific choice depends on the scale and complexity of the data set.

[0058] In some implementation manners, the sample aging data includes intrinsic aging data and homologous aging data, where the homologous aging data is the aging data of the same type of cells as the cells of the target battery pack.

[0059] Specifically, the intrinsic aging data is the aging data accumulated by the internal cells of the target battery pack itself during long-term charge and discharge cycles. During use, each cell will gradually age with the increase of use time. The intrinsic aging data reflects the process of capacity attenuation and internal resistance increase of multiple cells belonging to different battery pack individuals in the same target battery pack model.

[0060] Specifically, the homologous aging data is the aging data of other battery cells of the same model or series as the cells in the target battery pack. Through homologous data, the data volume and data richness of the obtained sample aging data can be expanded, which helps to infer the aging pattern of the target cell under similar conditions. Especially when some battery packs have not gone through a complete aging cycle, homologous data can be used to fill in the gaps to ensure the coverage integrity and prediction accuracy of the generated health status prediction model for the cell working conditions.

[0061] In some embodiments, such asFigure 2 As shown, collect the charge and discharge data of the battery pack, combine with the initial health state for health state assessment, and generate a reference health state. The execution steps include: Interactively obtain the charge and discharge data, where the charge and discharge data includes a historical temperature sequence and a historical energy flow sequence.

[0062] Based on the elbow method, analyze the dispersion degree of the temperature data in the historical temperature sequence, determine the number of clustering clusters, and perform clustering division on the charge and discharge data according to the number of clustering clusters to obtain multiple environmental charge and discharge data groups.

[0063] Respectively input multiple environmental charge and discharge data groups into the health state prediction model, obtain multiple discrete prediction results, and fit multiple discrete prediction results to the initial health state to obtain the reference health state.

[0064] Specifically, first interact with the battery management system (BMS) or other data collection devices to obtain the historical charge and discharge data of the target battery pack. These data include key parameters recorded during multiple charge and discharge processes since the battery pack was put into use, such as temperature and energy flow data.

[0065] Specifically, use the elbow method to analyze the dispersion degree of the temperature data in the historical temperature sequence, determine the appropriate number of clustering clusters, and perform clustering division on the charge and discharge data, so as to structurally obtain the historical charge and discharge data and reduce the data complexity. Exemplarily, calculate the sum of squared errors (SSE) of clustering under different numbers of clusters and draw a curve graph. As the number of clusters increases, the sum of squared errors (SSE) of clustering will gradually decrease. When the number of clusters increases to a certain critical point, the decline rate of SSE significantly slows down. Reflected in the curve, the turning point where the curve gradually flattens is the elbow, corresponding to the optimal number of clustering clusters.

[0066] Furthermore, according to the determined number of clustering clusters, perform clustering division on the charge and discharge data. By grouping the charge and discharge data according to different working conditions, multiple environmental charge and discharge data groups with similar environmental conditions can be obtained, which is convenient for improving the continuity of mapping relationship calls in subsequent predictions.

[0067] Specifically, after obtaining multiple environmental charge and discharge data groups, input these data groups into the health state prediction model respectively, predict the influence degree of each environmental charge and discharge data group on the cell health state respectively, and the output is a discrete prediction result. In other words, the discrete prediction result can be understood as the influence amount of multiple charge and discharge working conditions on the cell health state (such as a 1% decrease in SOH). Optionally, each set of environmental charge-discharge data groups corresponds to calling different prediction layers or prediction channels in the state of health prediction model. The different prediction layers or prediction channels are obtained after specialized enhanced training for different environments and working conditions, which helps to improve the prediction accuracy under different environments and working conditions.

[0068] Furthermore, after obtaining multiple discrete prediction results, these results are fitted with the initial state of health of the battery pack. For example, the vector sum of the initial state of health and the multiple discrete prediction results is calculated to generate the reference state of health of the battery pack. The reference state of health is based on the comprehensive prediction results under multiple different environments and load conditions, reflecting the health level of the battery pack in the current usage cycle.

[0069] An equalization simulation module 13, which is used to establish an equalization simulation model according to the charge-discharge equalization strategy of the target battery pack in combination with the initial state of health, and perform equalization simulation based on the equalization simulation model to generate simulation equalization data.

[0070] Specifically, the equalization strategy is used to ensure that each battery cell maintains consistent voltage and capacity during charge and discharge, avoiding the situation of too fast deterioration or inconsistent capacity of the battery cells. The equalization strategy can be active equalization or passive equalization, both of which involve the charge and discharge of the battery cells in the battery pack, thus also affecting the state of health of the battery cells.

[0071] In some embodiments, according to the charge-discharge equalization strategy of the target battery pack, an equalization simulation model is established in combination with the initial state of health, and equalization simulation is performed based on the equalization simulation model to generate simulation equalization data. The execution steps include: Define an equivalent state of health by combining the initial state of health and the reference state of health, where the equivalent state of health is located between the initial state of health and the reference state of health.

[0072] Construct a digital model of the target battery pack, and initialize the digital model of the target battery pack according to the charge-discharge equalization strategy to obtain the equalization simulation model.

[0073] Input the charge-discharge data into the equalization simulation model for equalization simulation of the target battery pack, and synchronously record the equalization data, and the output is the simulation equalization data, where the simulation equalization data includes an equalization temperature sequence and an equalization energy flow sequence.

[0074] Specifically, in actual use, the health state of the battery pack changes dynamically. Therefore, by combining the initial health state and the reference health state, an equivalent health state (such as the weighted average result of the initial health state and the reference health state) is defined between the two. The equivalent health state reflects the dynamic balance health condition of the battery pack during the simulation process, which helps to improve the accuracy of the balancing simulation. Optionally, the equivalent health state is defined based on the degree of dispersion of historical charge and discharge data (such as the entropy, variance, etc. of the data).

[0075] Specifically, according to the physical characteristics of the battery pack and the charge and discharge balancing strategy, a digital model of the target battery pack is constructed. This model can accurately simulate the behavior of the battery pack under different charge and discharge conditions and serve as the basis for the balancing simulation model. Exemplarily, first, physical characteristic modeling is performed to reflect the actual physical structure of the battery pack, including the arrangement of multiple battery cells, the capacity of each battery cell, the internal resistance, and other key parameters. Then, according to the predetermined balancing strategy, the balancing working mode of the battery pack is set, such as the balancing rate, the balancing current, etc., to achieve initialization. The generated simulation model reflecting the overall balancing state of the battery pack is used to simulate the balancing process under different charge and discharge conditions, so as to determine the battery cell cycling situation caused by balancing.

[0076] Furthermore, the charge and discharge data is input into the balancing simulation model to perform the balancing simulation of the target battery pack. During the simulation process, the simulation balancing data is synchronously recorded and output, including the balancing temperature sequence and the balancing energy flow sequence. In other words, the simulation balancing data reflects the battery cell cycling situation caused by balancing, such as the actual charge and discharge times of each battery cell in the battery pack during the balancing process. This simulation balancing data is used to analyze the cycling situations of different battery cells and evaluate the impact of the balancing strategy on the battery cell life.

[0077] The health state correction module 14 is used to evaluate and correct the reference health state by using the simulation balancing data to obtain a corrected health state, where the corrected health state is updated in real time.

[0078] Specifically, since some battery cells may experience faster capacity decay or internal resistance increase due to frequent participation in the balancing operation during the balancing process, it is necessary to correct the reference health state based on the balancing prediction result to obtain a more realistic health state of the battery pack. The corrected health state takes into account the losses caused by the balancing operation between battery cells and obtains a more accurate health state of the battery pack.

[0079] In some embodiments, the steps of evaluating and correcting the reference health state by using the simulation balancing data to obtain a corrected health state include: Input the simulation balancing data into the health state prediction model to obtain a balancing prediction result.

[0080] Fit the balanced prediction result to the reference health state to obtain the corrected health state, where the corrected health state includes the health states of multiple battery cells in the target battery pack.

[0081] Specifically, first, input the simulation balance data into the trained health state prediction model to obtain the change amount of the battery pack health state caused by factors such as energy transfer and temperature change during balancing, and the output is the balanced prediction result. Then, based on the same principle steps as above, fit the balanced prediction result to the previous reference health state, and correct the overall health state of the battery pack by considering the losses brought by the balancing operation to generate the corrected health state. This corrected health state can more accurately reflect the current usage of the battery pack, especially after multiple balancing operations.

[0082] Health management and control module 15, which is used to perform health management and control of the target battery pack based on the corrected health state and generate corresponding health management and control strategies.

[0083] In some embodiments, performing health management and control of the target battery pack based on the corrected health state and generating corresponding health management and control strategies includes: Calculate the remaining service life, where the remaining service life is the difference between the preset life control limit and the corrected health state.

[0084] Perform health management and control discrimination according to the remaining service life, and call the health management and control solution library to generate the health management and control strategy, where the health management and control discrimination includes consistency evaluation and control direction discrimination.

[0085] Specifically, first calculate the remaining service life according to the corrected health state of the battery pack. The remaining service life refers to the distance of the battery pack from the preset life termination (life control limit) under the current corrected health state, and is used to predict the remaining normal usage time of the battery pack. Exemplarily, compare the current corrected health state of the battery pack with the preset life control limit, and calculate the remaining health state difference. For example, if the life control limit is 80% of the rated capacity and the current corrected health state is 85% of the rated capacity, the remaining service life is 5%.

[0086] Specifically, health control discrimination refers to evaluating whether the battery pack needs to take control measures based on the remaining service life, including consistency evaluation and control direction discrimination. Among them, consistency evaluation means that the system evaluates whether the health states of the individual battery cells inside the battery pack are consistent. If the parameters such as the capacity and internal resistance of the individual battery cells vary too much, it may be necessary to adjust the balancing strategy or limit the charge and discharge rate. Control direction discrimination refers to judging whether to take measures to extend the life or measures to optimize the performance according to the overall health state of the battery pack. For example, when the life of the battery pack is approaching exhaustion, the system will tend to reduce losses, while when the battery pack is in good condition, the system can allow higher performance output.

[0087] Specifically, the health control solution library contains various control measures for different health states of the battery pack. According to the control discrimination result, the system selects a suitable strategy from the solution library and implements it. Exemplarily, the health control strategies include limiting the charge and discharge rate (limiting the rate of charging or discharging, reducing the stress on the battery cells, and delaying their capacity decay), adjusting the temperature control range (regulating the cooling system or restricting high-load operations to ensure that the battery pack operates within an appropriate temperature range, thereby extending the service life), prompting the battery life (issuing a prompt to the user to recommend replacing the battery pack or taking maintenance measures), etc. The health control strategy generated based on the corrected health state and health control discrimination helps to extend the service life and safety of the battery pack.

[0088] In summary, the powertrain health control system of the new energy vehicle provided by the present invention has the following technical effects: Through the initial state acquisition module, the initial state parameters of the target battery pack are acquired, and its initial health condition is determined. The target battery pack includes multiple battery cells; the health state prediction module collects the charge and discharge data of the battery pack, combines the initial health condition for evaluation, and generates a reference health state; the balancing simulation module, based on the charge and discharge balancing strategy of the battery pack, combines the initial health condition, establishes a balancing simulation model, and simulates through this model to generate simulation balancing data; the health state correction module evaluates and corrects the reference health state based on the simulation balancing data, and updates the corrected health state of the battery pack in real time; the health control module performs health management on the target battery pack according to the corrected health state, and generates corresponding health control strategies. Thus, the technical effects of improving monitoring reliability, reducing monitoring difficulty and control cost are achieved.

[0089] It should be understood that the embodiments and the above descriptions disclosed in the present invention enable those skilled in the art to implement the present invention using the present invention. At the same time, the present invention is not limited to the above-mentioned part of the embodiments. It should be understood that those of ordinary skill in the art can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A power-train health management and control system for a new energy vehicle, characterized in that The system includes: An initial state acquisition module, which is used to acquire the initial state parameters of the target battery pack and determine the initial health state of the battery pack, where the target battery pack includes multiple battery cells; A health state prediction module, which is used to collect the charge and discharge data of the battery pack, conduct a health state assessment in combination with the initial health state, and generate a reference health state; An equalization simulation module, which is used to establish an equalization simulation model according to the charge and discharge equalization strategy of the target battery pack in combination with the initial health state, and perform equalization simulation based on the equalization simulation model to generate simulation equalization data; A health state correction module, which is used to evaluate and correct the reference health state by using the simulation equalization data to obtain a corrected health state, where the corrected health state is updated in real time; A health management and control module, which is used to perform health management and control of the target battery pack based on the corrected health state and generate corresponding health management and control strategies.

2. The system according to claim 1, wherein The steps for acquiring the initial state parameters of the target battery pack and determining the initial health state of the battery pack include: An interactive battery management component extracts the initial state parameters of multiple battery cells in the target battery pack, where the initial state parameters at least include capacity and internal resistance; According to the initial state parameters, calculate the initial cell health states of multiple battery cells by traversing, where the initial cell health states include absolute cell states and relative cell states; Output the initial cell health states of multiple battery cells as the initial health state.

3. The system according to claim 2, wherein Before collecting the charge and discharge data of the battery pack, conducting a health state assessment in combination with the initial health state, and generating a reference health state, the steps include: Obtain the sample aging data of the battery cells in the target battery pack; Analyze the sample aging data, and extract a sample temperature sequence, a sample energy flow sequence, and a sample health state sequence, where the sequence interval of the sample health state sequence is greater than the sequence intervals of the extracted sample temperature sequence and the sample energy flow sequence; Taking the sample health state sequence as the division target, synchronously divide the sample temperature sequence and the sample energy flow sequence, and associate the sample health state sequence with the synchronous division result to obtain multiple standard sample data groups; Taking multiple standard sample data groups as training data, construct and train a health state prediction model.

4. The system according to claim 3, wherein The steps for collecting the charge and discharge data of the battery pack, conducting a health state assessment in combination with the initial health state, and generating a reference health state include: Interactively obtain the charge and discharge data, where the charge and discharge data includes a historical temperature sequence and a historical energy flow sequence; Based on the elbow method, analyze the dispersion degree of the temperature data in the historical temperature sequence, determine the number of clustering clusters, and perform clustering division on the charge and discharge data according to the number of clustering clusters to obtain multiple environmental charge and discharge data groups; Input multiple environmental charge and discharge data groups into the health state prediction model respectively, obtain multiple discrete prediction results, and fit multiple discrete prediction results to the initial health state to obtain the reference health state.

5. The system according to claim 4, wherein According to the charge-discharge equalization strategy of the target battery pack, an equalization simulation model is established in combination with the initial state of health, and equalization simulation is carried out based on the equalization simulation model to generate simulation equalization data. The execution steps include: Define an equivalent state of health by combining the initial state of health and the reference state of health, where the equivalent state of health is between the initial state of health and the reference state of health; Construct a digital model of the target battery pack, and initialize the digital model of the target battery pack according to the charge-discharge equalization strategy to obtain the equalization simulation model; Input the charge-discharge data into the equalization simulation model to perform equalization simulation of the target battery pack, and synchronously record the equalization data, and the output is the simulation equalization data, where the simulation equalization data includes an equalization temperature sequence and an equalization energy flow sequence.

6. The system according to claim 5, wherein Use the simulation equalization data to evaluate and correct the reference state of health to obtain a corrected state of health. The execution steps include: Input the simulation equalization data into the state-of-health prediction model to obtain an equalization prediction result; Fit the equalization prediction result to the reference state of health to obtain the corrected state of health, where the corrected state of health includes the states of health of multiple battery cells in the target battery pack.

7. The system according to claim 1, characterized in that, Perform health management and control of the target battery pack based on the corrected state of health, and generate corresponding health management and control strategies, including: Calculate the remaining service life, where the remaining service life is the difference between the preset life control limit and the corrected state of health; Perform health management and control discrimination according to the remaining service life, and call the health management and control solution library to generate the health management and control strategy, where the health management and control discrimination includes consistency evaluation and control direction discrimination.

8. The system according to claim 3, wherein the sample aging data includes intrinsic aging data and homologous aging data, where The homologous aging data is the aging data of the same type of battery cells of the target battery pack.

Images