WO2022088591A1 - 一种估计新能源汽车电池包均衡状态的方法及系统 - Google Patents

一种估计新能源汽车电池包均衡状态的方法及系统 Download PDF

Info

Publication number
WO2022088591A1
WO2022088591A1 PCT/CN2021/081745 CN2021081745W WO2022088591A1 WO 2022088591 A1 WO2022088591 A1 WO 2022088591A1 CN 2021081745 W CN2021081745 W CN 2021081745W WO 2022088591 A1 WO2022088591 A1 WO 2022088591A1
Authority
WO
WIPO (PCT)
Prior art keywords
cell voltage
voltage difference
compensated
battery
new energy
Prior art date
Application number
PCT/CN2021/081745
Other languages
English (en)
French (fr)
Inventor
张轶珍
尚进
张�雄
许永刚
Original Assignee
广州汽车集团股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 广州汽车集团股份有限公司 filed Critical 广州汽车集团股份有限公司
Priority to CN202180004444.0A priority Critical patent/CN114982089A/zh
Publication of WO2022088591A1 publication Critical patent/WO2022088591A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/0047—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with monitoring or indicating devices or circuits
    • H02J7/0048—Detection of remaining charge capacity or state of charge [SOC]
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L58/00—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles
    • B60L58/10—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries
    • B60L58/12—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries responding to state of charge [SoC]
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L58/00—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles
    • B60L58/10—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries
    • B60L58/18—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries of two or more battery modules
    • B60L58/22—Balancing the charge of battery modules
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/3644—Constructional arrangements
    • G01R31/3648—Constructional arrangements comprising digital calculation means, e.g. for performing an algorithm
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
    • G01R31/3835—Arrangements for monitoring battery or accumulator variables, e.g. SoC involving only voltage measurements
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/392—Determining battery ageing or deterioration, e.g. state of health
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/396—Acquisition or processing of data for testing or for monitoring individual cells or groups of cells within a battery
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L2240/00—Control parameters of input or output; Target parameters
    • B60L2240/10—Vehicle control parameters
    • B60L2240/12—Speed
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L2240/00—Control parameters of input or output; Target parameters
    • B60L2240/40—Drive Train control parameters
    • B60L2240/54—Drive Train control parameters related to batteries
    • B60L2240/545—Temperature
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L2240/00—Control parameters of input or output; Target parameters
    • B60L2240/40—Drive Train control parameters
    • B60L2240/54—Drive Train control parameters related to batteries
    • B60L2240/547—Voltage
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L2260/00—Operating Modes
    • B60L2260/40—Control modes
    • B60L2260/44—Control modes by parameter estimation
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L2260/00—Operating Modes
    • B60L2260/40—Control modes
    • B60L2260/46—Control modes by self learning
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/0013—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries acting upon several batteries simultaneously or sequentially
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/60—Other road transportation technologies with climate change mitigation effect
    • Y02T10/70—Energy storage systems for electromobility, e.g. batteries

Definitions

  • the invention relates to the technical field of battery safety, in particular to a method and system for estimating the equilibrium state of a battery pack of a new energy vehicle.
  • New energy vehicles include electric vehicles (EVs), hybrid electric vehicles (HEVs), plug-in hybrid electric vehicles (PHEVs), etc.
  • New energy vehicles can transmit real-time vehicle data to an internet cloud server for remote monitoring and data collection. As a result, a large amount of data on new energy vehicles has accumulated over time. The data hides valuable clues about the performance and health of NEVs, especially battery packs that are a key component of NEVs.
  • the present invention provides a method for estimating the equilibrium state of a battery pack of a new energy vehicle.
  • the method includes: acquiring an actual battery cell voltage difference measured for a battery pack of a new energy vehicle, and acquiring one or more operating condition parameters for a working condition for measuring the actual battery cell voltage difference, wherein the actual battery cell voltage difference
  • the battery cell voltage difference is the difference between the maximum battery cell voltage and the minimum battery supply voltage in the battery pack
  • the desired battery cell voltage difference is determined according to the one or more operating condition parameters
  • according to the actual battery cell The voltage difference and the expected battery cell voltage difference are determined to determine the compensated battery cell voltage difference
  • the equilibrium state of the new energy vehicle battery pack is estimated according to the compensated battery cell voltage difference.
  • the one or more operating condition parameters of the operating conditions include at least one of: vehicle current, state of charge, battery temperature, vehicle speed, or odometer value.
  • determining the desired battery cell voltage difference based on the one or more operating condition parameters includes: determining the desired battery cell voltage difference based on the one or more operating condition parameters by using a machine learning model Cell voltage difference.
  • the machine learning model is a linear model
  • the weights w 0 , w 1 , . . . , w M of the linear model are obtained through training based on first training data, wherein the first training data includes data for a plurality of different working conditions One or more operating condition parameters of , and the actual battery cell voltage difference measured for multiple new energy vehicle battery packs under multiple different working conditions.
  • the preset conditions include: a new energy vehicle with abnormal performance, and a new energy vehicle with the highest or lowest performance among the multiple new energy vehicles.
  • the machine learning model is a nonlinear model
  • determining the desired battery cell voltage difference according to the one or more operating condition parameters includes: by using With one or more operating condition parameters as input, the desired battery voltage difference ⁇ ref is determined based on the nonlinear model, including a neural network based model or a random forest based model.
  • the weights of the nonlinear model are obtained by training based on first training data, wherein the first training data includes one or more operating condition parameters for a plurality of different operating conditions , and the actual cell voltage difference measured for multiple NEV battery packs under multiple different operating conditions.
  • the actual battery cell voltage difference measured for the battery pack of the new energy vehicle that meets the preset condition is changed from the first Removed from the training data, wherein the preset conditions include: new energy vehicles with abnormal performance, and new energy vehicles with the highest or lowest performance among multiple new energy vehicles.
  • the actual battery cell voltage difference measured for the new energy vehicle battery pack includes the measured actual battery cell voltage difference recorded for each time; and according to the actual battery cell voltage difference
  • the voltage difference and the desired battery cell voltage difference determine the compensated battery cell voltage difference, including: determining the compensated battery cell voltage difference recorded at each time, and comparing the compensated battery cell voltage recorded at each time within a preset time unit The difference is summed to obtain the average compensated cell voltage difference; or the compensated cell voltage difference recorded at each time is determined and the compensated cell voltage difference recorded at each time is summed on a battery cycle cycle basis to obtain Average the compensated cell voltage differences; or determine the compensated cell voltage differences recorded at each time and sum the compensated cell voltage differences recorded at each time on a charge cycle basis to obtain the average compensated cell voltage differences .
  • estimating the equilibrium state of the battery pack of the new energy vehicle according to the compensated battery cell voltage difference includes: when the average compensated battery cell voltage difference is greater than a first threshold, or a plurality of preset time units or more In the case that the average compensated battery cell voltage difference in a number of battery cycles or a plurality of charging cycles shows a trend of degradation, it is determined that the battery pack of the new energy vehicle is in an unbalanced state.
  • the method before estimating the equilibrium state of the battery pack of the new energy vehicle according to the compensated battery cell voltage difference, the method further includes: determining a normalized weighted compensated battery cell voltage difference within a preset time period; And estimating the balance state of the battery pack of the new energy vehicle according to the compensated battery cell voltage difference includes: estimating the balance state of the battery pack of the new energy vehicle according to the normalized weighted compensated battery cell voltage difference within a preset time period.
  • determining the normalized weighted compensated battery cell voltage difference within a preset time period includes: determining the normalized weighted compensated battery cell voltage difference based on the following formula: Among them, ⁇ 1 , ⁇ 2 ,..., ⁇ N are the respective instances of the compensated battery cell voltage differences within the preset time period, and ⁇ 1 , ⁇ 2 ,..., ⁇ N are the compensated battery cells within the preset time period Weight of each instance of cell voltage difference, N being the number of instances of compensated cell voltage difference within a preset time period.
  • the weight is equal to the preset value, and the weight of earlier instances of the compensated battery voltage difference over the preset time period decays exponentially.
  • the second training data includes actual vehicle records with abnormal pack battery imbalances.
  • estimating the equilibrium state of the battery pack of the new energy vehicle according to the normalized weighted compensated battery cell voltage difference within a preset time period includes: in the normalized weighted compensated battery cell voltage difference When the voltage difference between the battery cells that is greater than the second threshold value or the normalized weighted compensation within a plurality of preset time periods shows a trend of degradation, it is determined that the battery pack of the new energy vehicle is in an unbalanced state.
  • the method further includes: when the result of estimating the equilibrium state of the battery pack of the new energy vehicle according to the compensated battery cell voltage difference indicates that the battery pack of the new energy vehicle is in an unbalanced state, sending a maintenance request warning.
  • the present invention provides a system for estimating the equilibrium state of a battery pack of a new energy vehicle, the system is located in a server or an onboard computing device, the system includes a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the computer program to implement any one or more of the methods as in the preceding embodiments.
  • the present invention provides a non-transitory computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor At the time, the processor is caused to execute any one or more of the methods in the foregoing method embodiments.
  • Figure 1(a) is a schematic diagram of a time series of average cell voltage differences per charge for an electric vehicle (EV);
  • Figure 1(b) is a schematic diagram of the average cell voltage difference and the average charging current of the same electric vehicle (EV).
  • FIG. 2 is a flowchart of a method for estimating the balance state of a battery pack of a new energy vehicle according to an embodiment of the present invention.
  • FIG. 3 is a flowchart of a method for estimating the balance state of a battery pack of a new energy vehicle according to an exemplary embodiment of the present invention.
  • FIG. 4 is a schematic diagram of a system for estimating the balance state of a battery pack of a new energy vehicle according to an embodiment of the present invention.
  • the difference ⁇ between the maximum battery voltage and the minimum battery voltage within the battery pack is the battery pack Important indicator of health status.
  • the voltage difference between the cells must be kept within a very small range most of the time.
  • Figure 1(a) is a schematic diagram of a time series of average cell voltage differences per charge for an electric vehicle (EV)
  • Figure 1(b) is a historical record of average cell voltages for the same electric vehicle (EV).
  • Schematic diagram of voltage difference versus average charge current In Figures 1(a) and 1(b), darker points correspond to higher average charging currents (fast charging), while brighter points correspond to lower average charging currents (slow charging).
  • the average cell voltage difference during fast charging is much higher than that during slow charging.
  • the average cell voltage difference may fluctuate up and down.
  • the cell voltage difference ⁇ may also vary with other operating condition parameters, such as state of charge, temperature, vehicle speed, odometer value, etc.
  • embodiments of the present invention propose a solution that can monitor and analyze the battery pack balance of all actively running new energy vehicles (NEVs), detect any abnormal battery pack degradation trends, and report to the entire vehicle Factory, dealers and end customers provide early warning. Plus, troubleshooting and predictive maintenance can be done at the dealership as soon as needed to extend battery life and reduce warranty costs.
  • NEVs actively running new energy vehicles
  • a method for estimating the equilibrium state of a battery pack of a new energy vehicle is provided.
  • the method may be performed by a server (eg, a cloud-based server) or an onboard computing device or any device with computing capabilities.
  • FIG. 2 shows a flowchart of a method for estimating the equilibrium state of a battery pack of a new energy vehicle according to an embodiment of the present invention.
  • the method for estimating the equilibrium state of a battery pack of a new energy vehicle includes the following steps S202 to S208.
  • step S202 the actual battery cell voltage difference measured for the battery pack of the new energy vehicle is obtained, and one or more operating condition parameters under the operating conditions under which the actual battery cell voltage difference is measured are obtained, wherein the actual battery cell voltage difference is measured.
  • the cell voltage difference is the difference between the maximum cell voltage and the minimum cell voltage in the pack.
  • the one or more operating condition parameters for operating conditions may include one or more operating condition parameters that affect the value of the measured actual cell voltage difference.
  • the measured actual cell voltage difference is a variable that depends on one or more operating condition parameters corresponding to the operating conditions under which the actual cell voltage difference is measured.
  • the one or more operating condition parameters for operating conditions may include at least one of: vehicle current, state of charge, battery temperature, vehicle speed, or odometer value. When using these operating condition parameters to determine the desired cell voltage difference, either lower or higher order versions of these operating condition parameters may be used.
  • the acquisition of the actual battery cell voltage difference and the acquisition of one or more working condition parameters are independent operations, and there is no particular restriction on the execution sequence between the two acquisition actions.
  • the server or device located outside the vehicle can obtain the measured actual battery cell voltage difference and one or more operating condition parameters from the vehicle based on an active reporting approach or a request-response approach .
  • the onboard computing device may obtain the measured actual cell voltage difference and one or more operating condition parameters in a data record stored in local memory after measurements and data collection have been performed.
  • step S204 a desired battery cell voltage difference is determined according to one or more operating condition parameters.
  • step S204 may be implemented in the following manner:
  • the desired cell voltage difference is determined based on one or more operating condition parameters by using a machine learning model.
  • Machine learning models come in many forms and can accurately estimate expected cell voltage differences based on one or more operating condition parameters.
  • a machine learning model of the following form can be used.
  • the machine learning model may be a linear model.
  • the weights w0 , w1, are selected from the weights w0 , w1, .
  • the actual battery cell voltage difference is removed from the first training data, wherein the preset conditions include: a new energy vehicle with abnormal performance, and a new energy vehicle with the highest or lowest performance among the multiple new energy vehicles. That is to say, the actual battery cell voltage difference measured by an abnormal or obviously abnormal new energy vehicle battery pack can be removed from the first training data to ensure the accuracy of the training model.
  • the machine learning model may be a nonlinear model.
  • nonlinear models may include neural network-based models or random forest-based models.
  • a neural network-based model can have a network structure that includes an input layer, one or more hidden layers, and an output layer.
  • Step S204 may include determining the desired battery voltage difference ⁇ ref based on the nonlinear model by using one or more operating condition parameters as input.
  • the weights in the nonlinear model are obtained by training based on first training data, wherein the first training data includes one or more operating conditions for a plurality of different operating conditions parameters, and the actual cell voltage difference measured for multiple NEV battery packs under multiple different operating conditions.
  • the actual battery cell voltage difference measured for the battery pack of the new energy vehicle that meets the preset condition is obtained from the first training Removed from the data, where the preset conditions include: new energy vehicles with abnormal performance, and new energy vehicles with the highest or lowest performance among multiple new energy vehicles. That is to say, the actual battery cell voltage difference measured by an abnormal or obviously abnormal new energy vehicle battery pack can be removed from the first training data to ensure the accuracy of the training model.
  • step S206 the compensated battery cell voltage difference is determined according to the actual battery cell voltage difference and the expected battery cell voltage difference.
  • the compensated cell voltage difference ⁇ can be calculated for each time record and totaled daily (or every 12 hours, or weekly or monthly, etc.) (the total time unit can be any preset time unit , the length can be specified according to actual requirements), or it can be determined according to each battery cycle (trip-on-battery)/charge cycle.
  • the actual battery cell voltage difference measured for the new energy vehicle battery pack includes the measured actual battery cell voltage difference recorded for each time (the measurement may be in a preset time intervals, either according to the occurrence of certain events, or according to a combination of the two).
  • Step S206 may include one of the following:
  • step S208 the equilibrium state of the battery pack of the new energy vehicle is estimated according to the compensated battery cell voltage difference.
  • step S208 may include: when the average compensated battery cell voltage difference is greater than a first threshold, or within a plurality of preset time units or a plurality of battery cycle cycles or a plurality of charging cycles When the average compensated battery cell voltage difference shows a trend of degradation, it is determined that the battery pack of the new energy vehicle is in an unbalanced state.
  • the method may further include: determining a normalized weighted compensated battery cell voltage difference within a preset time period.
  • Step S208 may include: estimating the equilibrium state of the battery pack of the new energy vehicle according to the normalized weighted compensated battery cell voltage difference within a preset time period.
  • the step of determining the normalized weighted compensated battery cell voltage difference within a preset time period may include: determining the normalized weighted compensated battery cell voltage difference based on the following formula: where ⁇ 1 , ⁇ 2 ,..., ⁇ N are the respective instances of the compensated battery cell voltage differences within the preset time period, and ⁇ 1 , ⁇ 2 ,..., ⁇ N are the compensated battery cells within the preset time period The weight of each instance of the voltage difference, N being the number of instances of the compensated cell voltage difference within a preset time period.
  • the weight of the instance may be equal to a preset value (eg, the weight of the latest instance of the compensated cell voltage difference may be equal to 1), and the weight of the earlier instance of the compensated cell voltage difference over a preset time period decays exponentially.
  • the latest compensated cell voltage difference instance has the greatest influence on the normalized weighted compensated cell voltage difference, while the earlier compensated cell voltage difference instance has the greatest effect on the normalized weighted compensated cell voltage difference. minor impact.
  • the weights ⁇ 1 , ⁇ 2 , . . . , ⁇ N of each instance of the compensated cell voltage difference within a preset time period may be obtained by performing training based on the second training data , wherein the second training data includes actual vehicle records with abnormal pack-to-battery imbalances.
  • step S208 may include: when the normalized weighted compensated battery cell voltage difference is greater than a second threshold or within a plurality of preset time periods, the normalized weighted compensated battery cell voltage difference In the case of a degradation trend, it is determined that the battery pack of the new energy vehicle is in an unbalanced state.
  • FIG. 3 shows a flowchart of a method for estimating the equilibrium state of a battery pack of a new energy vehicle according to an exemplary embodiment of the present invention. As shown in FIG. 3, in at least one exemplary embodiment of the present invention, the method may further include step S302.
  • step S302 when the result of estimating the equilibrium state of the battery pack of the new energy vehicle according to the compensated battery cell voltage difference indicates that the battery pack of the new energy vehicle is in an unbalanced state, a maintenance-required warning is sent.
  • the method described in this embodiment can be applied to analyze the relevant data stored in the cloud server and effectively estimate the current balance state of the battery pack. Degradation trends of battery pack equalization conditions can be detected, allowing automatic predictive maintenance notifications to be generated for battery packs of different types of NEVs (eg EV, HEV, PHEV, etc.). If troubleshooting and predictive maintenance actions are carried out shortly after an early warning notification, battery life can be extended for a better customer experience and warranty costs can be significantly reduced.
  • the proposed method enables high levels of predictive maintenance and customer satisfaction, rather than passive maintenance that leads to poor customer experience and higher warranty costs for OEMs.
  • FIG. 4 shows a schematic diagram of a system for estimating the equilibrium state of a battery pack of a new energy vehicle according to an embodiment of the present invention.
  • the system includes a memory 42 and a processor 44, wherein a computer program is stored in the memory 42, and the processor 44 is configured to execute the computer program to implement any one or more of the foregoing method embodiments method.
  • a non-transitory computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to perform any one or more of the methods as in the foregoing embodiments.
  • the method proposed in this embodiment uses a Machine Learning (ML) model to estimate under different operating conditions such as vehicle current i, battery pack state of charge SOC, battery temperature T, vehicle speed V , the desired cell voltage difference (ie, ⁇ ref ) for the odometer value, etc.
  • ML Machine Learning
  • the difference ( ⁇ ) between the actual cell voltage difference ( ⁇ act ) measured for each individual vehicle and the cell voltage difference reference value ( ⁇ ref ) estimated by the model is Referred to in this example as the compensated cell voltage difference ( ⁇ ), it can be defined as a metric that quantifies the equilibrium state of the cells for each pack in real time.
  • Equation 1 The above machine learning model shown in Equation 1 can be implemented with a linear or nonlinear model function f.
  • a linear model function as shown in Equation 3 can be defined. As shown in Figure 3, where w 0 , w 1 and w 2 are the weights to be trained and optimized.
  • Equation 3 In another embodiment, more features (eg, temperature, speed, odometer, etc.) and their higher order terms can be added to Equation 3 for better model fit.
  • nonlinear models such as neural networks (NN) and random forests (RF) may be employed.
  • NN neural networks
  • RF random forests
  • NN model with two hidden layers can achieve better model fit than a linear model.
  • the above ML model defines the expected cell voltage reference value ( ⁇ ref ) for a given operating condition based on normal fleet performance as training data, where some obviously abnormal vehicle data (e.g., the top 1%) can be found in the data. Filtered out during cleanup. Assuming that most of the vehicles in the fleet training data are operating normally, the expected cell voltage difference reference value output by the ML model will be very close to the average or median value of the normal fleet cell voltage difference under different operating conditions. Then, the compensated cell voltage difference ( ⁇ ), as defined in Equation 2, can be used as a metric to quantify the cell equilibrium state for each pack. This compensated cell voltage difference ( ⁇ ) can be calculated for each time record and aggregated daily or by each battery trip/charge cycle. Then, high-risk vehicles can be identified when ⁇ is above a certain threshold and/or its time series shows a certain increasing trend (ie, degradation).
  • the cell voltage difference reference value ⁇ ref is given by a NN model with two layers trained on samples from one day's data. The latter is more concentrated and closer to the normal distribution with smaller standard deviation, as shown in Fig. 5(b). Therefore, the proposed metric can better describe the battery cell equilibrium state of each vehicle in real time despite different operating conditions.
  • its current cell imbalance state may be defined as its normalized weighted average of its most recently compensated cell voltage difference delta values over a period of time (eg, the last N days). For example, the last day has a weight of 1, and the weight of days earlier than the last day decays exponentially, i.e.
  • All vehicles can be ranked by the cell imbalance status assessed using the above metrics.
  • a battery imbalance above a certain threshold or showing a trend of degradation over a time series can be defined as a high-risk vehicle, and immediate attention and preventive maintenance actions are recommended at the dealership.
  • parameters are selected to minimize model fit error based on training data for normal vehicles.
  • Products with proposed solutions can be implemented in cloud-based servers or on-board computing devices. It can analyze the time series of relevant data provided by on-board sensors and/or CAN bus, identify all vehicles with unhealthy degradation trends, and generate automatic predictive forecasts for various NEVs (e.g. EV, HEV, PHEV, etc.) Maintenance warning notice. It will ensure that all running packs are operating in healthy and balanced conditions and detect any potential pack imbalance issues or unhealthy degradation trends at an early stage. Once a reasonable alarm threshold has been determined for a vehicle model, "Maintenance Required" warnings can be sent directly to OEMs, dealers and customers in different formats so that early troubleshooting and predictive maintenance actions can be taken to extend the battery life, avoid costly warranty losses for OEMs, and greatly improve customer satisfaction.
  • NEVs e.g. EV, HEV, PHEV, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Power Engineering (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Sustainable Development (AREA)
  • Sustainable Energy (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Secondary Cells (AREA)
  • Charge And Discharge Circuits For Batteries Or The Like (AREA)
  • Electric Propulsion And Braking For Vehicles (AREA)

Abstract

提供一种估计新能源汽车电池包均衡状态的方法和系统。该方法包括：获取为新能源汽车电池包测量的实际的电池单元电压差，并获取用于测量实际的电池单元电压差的工作条件的一个或多个工作条件参数(S202)，其中，实际的电池单元电压差是所述电池包中最大电池单元电压与最小电池电源电压之间的差；根据基于一个或多个工作条件参数的模型确定期望的电池单元电压差(S204)；根据实际的电池单元电压差和期望的电池单元电压差，确定补偿的电池单元电压差(S206)；以及根据补偿的电池单元电压差估计新能源汽车电池包均衡状态(S208)。

Description

一种估计新能源汽车电池包均衡状态的方法及系统
相关申请
本申请要求于2020年10月28日提交美国专利商标局、申请号为17082051、发明名称为“一种估计新能源汽车电池包均衡状态的方法及系统”的美国专利申请的优先权，上述专利的全部内容通过引用结合在本申请中。
技术领域
本发明涉及电池安全技术领域，尤其涉及一种估计新能源汽车电池包均衡状态的方法及系统。
背景技术
新能源汽车(NEV)包括电动汽车(EV)，混合动力电动汽车(HEV)，插电式混合动力电动汽车(PHEV)等。新能源汽车可以将实时车辆数据传输到互联网云服务器以进行远程监控和数据收集。结果，随着时间的推移，新能源车型的大量数据已经积累。数据中隐藏着有关新能源汽车(尤其是作为是新能源汽车的关键组成部分的电池包)的性能和运行状况的有价值的线索。
需要建立一种有效的数据驱动的电池包健康状况监测方法，电池包的各电池单元之间的均衡状态对于电池包的健康状况至关重要。
发明内容
以下呈现简化的本发明概述，以提供对本发明的一些方面的基本理解。该概述不是本发明的广泛说明。它并不旨在确定本发明的关键或重要元素。以下概述仅以简化形式呈现本发明的一些概念，作为以下具体描述的序言。
根据本发明实施例的一个方面，本发明提供了一种估计新能源汽车电池包均衡状态的方法。该方法包括：获取为新能源汽车电池包测量的实际的电池单元电压差，并获取用于测量所述实际的电池单元电压差的工作条件的一个或多个工作条件参数，其中，所述实际的电池单元电压差是所述电池包中最大电池单元电压与最小电池电源电压之间的差；根据所述一个或多个工作条件参数确定期望的电池单元电压差；根据所述实际的电池单元电压差和期望的电池单元电压差，确定补偿的电池单元电压差；以及根据补偿的电池单 元电压差估计新能源汽车电池包均衡状态。
在本发明至少一个示例性实施例中，所述工作条件的一个或多个工作条件参数包括以下至少之一：车辆电流，荷电状态，电池温度，车辆速度或里程表值。
在本发明至少一个示例性实施例中，根据所述一个或多个工作条件参数确定期望的电池单元电压差包括：通过使用机器学习模型，根据所述一个或多个工作条件参数确定期望的电池单元电压差。
在本发明至少一个示例性实施例中，所述机器学习模型为线性模型，通过使用所述机器学习模型，根据所述一个或多个工作条件参数确定期望的电池单元电压差包括：基于以下公式确定期望的电池单元电压差δ ref：δ ref＝w 0+w 1*P 1+…+w M*P M，其中，w 0,w 1,…,w M是线性模型的权重，P 1,…,P M为一个或多个工作条件参数，M为一个或多个工作条件参数的数量。
在本发明至少一个示例性实施例中，所述线性模型的权重w 0,w 1,…,w M基于第一训练数据通过训练获得，其中，第一训练数据包括用于多个不同工作条件的一个或多个工作条件参数，以及在多个不同的工作条件下为多个新能源汽车电池包测量的实际电池单元电压差。
在本发明至少一个示例性实施例中，在基于第一训练数据通过训练获得线性模型的权重w 0,w 1,…,w M之前，为满足预设条件的新能源汽车电池包测量的实际电池单元电压差从第一训练数据中去除，其中，预设条件包括：性能异常的新能源汽车，多辆新能源汽车中性能最高或最低的新能源汽车。
在本发明至少一个示例性实施例中，所述机器学习模型为非线性模型，通过使用所述机器学习模型，根据所述一个或多个工作条件参数确定期望的电池单元电压差包括：通过使用一个或多个工作条件参数作为输入，基于所述非线性模型确定期望的电池电压差δ ref，所述非线性模型包括基于神经网络的模型或基于随机森林的模型。
在本发明至少一个示例性实施例中，所述非线性模型的权重通过基于第一训练数据的训练获得，其中，第一训练数据包括用于多个不同工作条件的一个或多个工作条件参数，以及在多个不同的工作条件下为多个新能源汽车电池包测量的实际电池单元电压差。
在本发明至少一个示例性实施例中，在基于第一训练数据通过训练获得所述非线性模型的权重之前，为满足预设条件的新能源汽车电池包测量的实际电池单元电压差从第一训练数据中去除，其中，预设条件包括：性能异常的新能源汽车，多辆新能源汽车中性能最高或最低的新能源汽车。
在本发明至少一个示例性实施例中，根据所述实际的电池单元电压差和期望的电池单元电压差确定补偿的电池单元电压差，包括：根据以下公式确定补偿的电池单元电压差Δ：Δ＝δ act-δ ref，其中，δ act是实际的电池单元电压差，δ ref是期望的电池单元电压差。
在本发明至少一个示例性实施例中，为新能源汽车电池包测得的实际的电池单元电压差包括针对每个时间记录的测量的实际的电池单元电压差；以及根据所述实际的电池单元电压差和期望的电池单元电压差确定补偿的电池单元电压差，包括：确定每个时间记录的补偿的电池单元电压差，并在预设时间单位内对每个时间记录的补偿的电池单元电压差进行合计，以获得平均补偿的电池单元电压差；或者确定每个时间记录的补偿的电池单元电压差，并以电池循环周期为基础合计每个时间记录的补偿的电池单元电压差，以获得平均补偿的电池单元电压差；或者确定每个时间记录的补偿的电池单元电压差，并以充电周期为基础合计每个时间记录的补偿的电池单元电压差，以获得平均补偿的电池单元电压差。
在本发明至少一个示例性实施例中，根据补偿的电池单元电压差估计新能源汽车电池包均衡状态包括：在平均补偿的电池单元电压差大于第一阈值，或者多个预设时间单位或多个电池循环周期或多个充电周期中的平均补偿的电池单元电压差呈现退化趋势的情况下，确定新能源汽车电池包处于不均衡状态。
在本发明至少一个示例性实施例中，在根据补偿的电池单元电压差估计新能源汽车电池包均衡状态之前，还包括：确定预设时间段内的归一化加权补偿的电池单元电压差；以及根据补偿的电池单元电压差估计新能源汽车电池包均衡状态包括：根据预设时间段内的归一化加权补偿的电池单元电压差，估计新能源汽车电池包均衡状态。
在本发明至少一个示例性实施例中，确定预设时间段内的归一化加权补 偿的电池单元电压差包括：基于以下公式确定归一化加权后补偿的电池单元电压差：
Figure PCTCN2021081745-appb-000001
其中，Δ 1,Δ 2,…,Δ N分别是预设时间段内补偿的电池单元电压差的各个实例，ω 1,ω 2,…,ω N是在预设时间段内的补偿的电池单元电压差的各个实例的权重，N是在预设时间段内的补偿的电池单元电压差的实例的数量。
在本发明至少一个示例性实施例中，在预设时间段内的补偿的电池单元电压差的各个实例的权重ω 1,ω 2,…,ω N中，补偿的电池单元电压差的最新实例的权重等于预设值，并且在预设时间段内的补偿的电池电压差的较早实例的权重呈指数衰减。
在本发明至少一个示例性实施例中，预设时间段内的补偿的电池单元电压差的各个实例的权重ω 1,ω 2,…,ω N通过基于第二训练数据进行训练获得，其中，第二训练数据包括具有异常电池包电池失衡的实际车辆记录。
在本发明至少一个示例性实施例中，根据预设时间段内的归一化加权补偿的电池单元电压差，估计新能源汽车电池包均衡状态包括：在归一化加权补偿的电池单元电压差大于第二阈值或者多个预设时间段内归一化加权补偿的电池单元电压差呈现退化趋势的情况下，确定新能源汽车电池包处于不均衡状态。
在本发明至少一个示例性实施例中，所述方法还包括：当根据补偿的电池单元电压差估计新能源汽车电池包均衡状态的结果指示新能源汽车电池包处于不均衡状态时，发送需要维护的警告。
根据本发明实施例的一个方面，本发明提供了一种估计新能源汽车电池包均衡状态的系统，所述系统位于服务器或机载计算设备中，所述系统包括存储器和处理器，其中计算机程序被存储在所述存储器中，并且所述处理器被配置为执行所述计算机程序以实施如前述实施例中任一或多个方法。
根据本发明实施例的一个方面，本发明提供了一种非暂时性计算机可读存储介质，所述非暂时性计算机可读存储介质存储有计算机程序，其中，当所述计算机程序被处理器执行时，使所述处理器执行如前述方法实施例中任一或多个方法。
附图说明
这里描述的附图用于提供对本发明的更深入的理解，并构成本发明的一部分。示意性的实施例及其描述用于举例说明本发明，和说明用于解释本发明，而不意图对本发明构成不适当的限制。在附图中：
图1(a)是一辆电动汽车(EV)的每次充电的平均电池单元电压差的时间序列的示意图；
图1(b)是同一电动汽车(EV)的平均电池单元电压差与平均充电电流的示意图。
图2是根据本发明实施例的一种估计新能源汽车电池包平衡状态的方法的流程图。
图3根据本发明示例性实施例的一种估计新能源汽车电池包平衡状态的方法的流程图。
图4是根据本发明实施例的一种估计新能源汽车电池包平衡状态的系统的示意图。
图5(a)是在一天中(标准差std＝0.0062)对一类电动汽车(EV)充电期间每辆车的平均电池单元电压差δ(原始数据)的柱形图。
图5(b)是对于上述相同数据(标准差std＝0.0037)的每辆车的平均补偿电池电压差Δ值的柱形图。
具体实施方式
根据对多个已报告电池包故障的一种类型电动汽车的单个电池单元电压的时间序列进行的历史数据分析，可以确定电池包内最大电池电压和最小电池电压之间的差值δ是电池包健康状况的重要指标。为了使电池包正常工作，大多数时候必须将电池单元的电压差保持在很小的范围内。当电池电压差稳定增加时，在某些车辆上会观察到异常的电池包退化，并且电池容量可能同时异常下降。事实证明，异常的电池组退化趋势可能表明存在严重问题，应立即采取措施予以纠正。
然而，还观察到，电池单元电压差δ是取决于某些工作条件的变量，例如电流，荷电状态，温度，车速，里程表值等。例如，图1(a)是一辆电动汽车(EV)的每次充电的平均电池单元电压差的时间序列的示意图，图1 (b)是是同一电动汽车(EV)历史记录的平均电池单元电压差与平均充电电流的示意图。在图1(a)和1(b)中，较暗的点对应于较高的平均充电电流(快速充电)，而较亮的点对应于较低的平均充电电流(缓慢充电)。如图1(a)和1(b)所示，对于一种类型的电动汽车，快速充电过程中的平均电池单元电压差要比慢速充电过程中的平均电池单元电压差高得多。结果，如图1(a)所示，当用户在不同充电周期的快速充电模式和慢速充电模式之间切换时，平均电池单元电压差可能会忽高忽低。在这种情况下，由于由不同水平的充电模式引起的变化可能比在相同水平的充电模式下观察到的较小的退化趋势大得多，对于真实的电池单元失衡状态，很难确定任何缓慢的退化趋势。电池单元电压差δ也可能随其他工作条件参数而变化，例如荷电状态，温度，车速，里程表值等。
因此，需要建立一个模型来描述不同工作条件下电池单元电压差的预期正常范围，并且需要针对不同工作条件识别异常情况。鉴于此，本发明实施例提出了一种解决方案，该解决方案可以监视和分析所有活跃运行的新能源汽车(NEV)的电池包平衡状况，检测任何异常的电池组退化趋势，并向整车厂，经销商和终端客户提供预警。此外，可以在需要时尽快在经销商处进行故障排除和预测性维护工作，以延长电池寿命并降低保修成本。
为了使本领域技术人员更加清楚地理解本发明的方案，下面结合附图对本发明实施例中的技术方案进行清楚和完整地阐述。显然，所描述的实施例仅仅是本发明实施例的一部分，而不是全部。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
需要指出的是，本发明的说明书和权利要求以及附图中的术语“第一”、“第二”等旨在区分类似的对象，并且不需要描述特定的顺序或优先顺序。应当理解的是，可以在适当的条件下交换以这种方式使用的数据，以便这里描述的本公开的实施例可以以除了此处附图示出或描述的顺序之外的顺序来实现。另外，术语“包括”、“包含”及其变体旨在涵盖非排他性的内容。例如，包含一系列步骤或单元的过程、方法、系统、产品或设备不必限于那些明确列出的步骤或单元，并且可以包括没有清楚地显示的这些过程、方法、 产品或设备的其他固有步骤或单元。
根据本发明实施例，提供了一种估计新能源汽车电池包均衡状态的方法。该方法可以由服务器(例如，基于云的服务器)或机载计算设备或具有计算能力的任何设备执行。图2示出了根据本发明实施例的估计新能源汽车电池包均衡状态的方法的流程图。如图2所示，估计新能源汽车电池包均衡状态的方法包括以下步骤S202至S208。
在步骤S202中，为新能源汽车电池包测量的实际电池单元电压差被获取，以及在该实际电池单元电压差被测量的工作条件下的一个或多个工作条件参数被获取，其中，该实际电池单元电压差是该电池包中最大电池单元电压和最小电池单元电压之间的差。
在本发明的至少一个示例性实施例中，用于工作条件的一个或多个工作条件参数可以包括影响所测量的实际电池单元电压差的值的一个或多个工作条件参数。换言之，所测量的实际电池单元电压差是一个取决于与测量实际电池单元电压差的工作条件相对应的一个或多个工作条件参数的变量。例如，用于工作条件的一个或多个工作条件参数可以包括以下各项中的至少一项：车辆电流，充电状态，电池温度，车辆速度或里程表值。当使用这些工作条件参数来确定期望的电池单元电压差时，可以使用这些工作条件参数的低阶或高阶形式。
在实际情况中，实际电池单元电压差的获取和一个或多个工作条件参数的获取是彼此独立的操作，并且对两个获取动作之间的执行顺序没有特别的限制。当该方法由位于车辆外部的服务器或设备执行时，位于车辆外部的服务器或设备基于主动报告方式或请求-响应方式，可以从车辆获取测量的实际电池单元电压差和一个或多个工作条件参数。当该方法由车载计算设备执行时，车载计算设备可以在已经进行了测量和数据收集之后存储在本地存储器中的数据记录中获取测量的实际电池单元电压差以及一个或多个工作条件参数。
在步骤S204中，根据一个或多个工作条件参数确定期望的电池单元电压差。
在本发明的至少一个示例性实施例中，可以通过以下方式实施步骤S204：
通过使用机器学习模型，根据一个或多个工作条件参数确定期望的电池单元电压差。
机器学习模型有多种形式，能够根据一个或多个工作条件参数准确估计预期的电池单元电压差。例如，可以采用以下形式的机器学习模型。
(1)在本发明的至少一个示例性实施例中，机器学习模型可以是线性模型。
步骤S204可以包括：基于以下公式确定期望的电池单元电压差δ ref：δ ref＝w 0+w 1*P 1+…+w M*P M，其中w 0,w 1,…,w M是线性模型的权重，P 1,…,P M为一个或多个工作条件参数，M为一个或多个工作条件参数的数量。
在本发明的至少一个示例性实施例中，线性模型的权重w 0,w 1,…,w M是基于第一训练数据通过训练获得的，其中，第一训练数据包括用于多个不同工作条件的一个或多个工作条件参数，以及在多个不同的工作条件下为多个新能源汽车电池包测量的实际电池单元电压差。
在本发明的至少一个示例性实施例中，在基于第一训练数据通过训练获得线性模型的权重w 0,w 1,…,w M之前，为满足预设条件的新能源汽车电池包测量的实际电池单元电压差从第一训练数据中去除，其中，预设条件包括：性能异常的新能源汽车，多辆新能源汽车中性能最高或最低的新能源汽车。也就是说，可以从第一训练数据中去除异常或明显异常的新能源汽车电池包测得的实际电池单元电压差，以确保训练模型的准确性。
(2)在本发明的至少一个示例性实施例中，机器学习模型可以是非线性模型。例如，非线性模型可以包括基于神经网络的模型或基于随机森林的模型。本领域技术人员应该理解，其他类型的非线性模型也可以应用在本实施例的解决方案中。基于神经网络的模型可以具有包括输入层，一个或多个隐藏层以及输出层的网络结构。
步骤S204可以包括：通过使用一个或多个工作条件参数作为输入，基于非线性模型确定期望的电池电压差δ ref。
在本发明的至少一个示例性实施例中，通过基于第一训练数据的训练来获得非线性模型中的权重，其中，第一训练数据包括用于多个不同工作条件的一个或多个工作条件参数，以及在多个不同的工作条件下为多个新能源汽 车电池包测量的实际电池单元电压差。
在本发明的至少一个示例性实施例中，在基于第一训练数据通过训练获得非线性模型的权重之前，为满足预设条件的新能源汽车电池包测量的实际电池单元电压差从第一训练数据中去除，其中，预设条件包括：性能异常的新能源汽车，多辆新能源汽车中性能最高或最低的新能源汽车。也就是说，可以从第一训练数据中去除异常或明显异常的新能源汽车电池包测得的实际电池单元电压差，以确保训练模型的准确性。
在步骤S206中，根据实际的电池单元电压差和预期的电池单元电压差确定补偿的电池单元电压差。
在本发明的至少一个示例性实施例中，步骤S206可以包括：基于以下公式确定补偿的电池单元电压差Δ：Δ＝δ act-δ ref，其中δ act是实际的电池单元电压差，而δ ref是期望的电池电压差。
在实际应用中，可以为每个时间记录计算补偿的电池单元电压差Δ，并每天(或每12小时，或每周或每月等)进行合计(合计的时间单位可以是任何预设时间单位，长度可以根据实际要求进行指定)，也可以根据每次电池循环周期(trip-on-battery)/充电周期确定。在本发明的至少一个示例性实施例中，为新能源汽车电池包测得的实际的电池单元电压差包括针对每个时间记录的测量的实际的电池单元电压差(该测量可以以预设的时间间隔进行，或者根据某些事件的发生，或根据两者的结合)。步骤S206可以包括以下之一：
(1)确定每个时间记录的补偿的电池单元电压差，并在预设时间单位内对每个时间记录的补偿的电池单元电压差进行合计，以获得平均补偿的电池单元电压差(对于每个预设时间单位)。
(2)确定每个时间记录的补偿的电池单元电压差，并以电池循环周期为基础合计每个时间记录的补偿的电池单元电压差，以获得平均补偿的电池单元电压差(对于每个电池循环周期)。
(3)确定每个时间记录的补偿的电池单元电压差，并以充电周期为基础合计每个时间记录的补偿的电池单元电压差，以获得平均补偿的电池单元电压差(对于每个充电周期)。
在步骤S208中，根据补偿的电池单元电压差来估计新能源车辆电池包均衡状态。
基于上述平均补偿的电池单元电压差(对于每个预设时间单位，或对于每个电池循环周期，或对于每个充电周期)，可以实时估计新能源汽车电池包均衡状态。在本发明的至少一个示例性实施例中，步骤S208可以包括：在平均补偿的电池单元电压差大于第一阈值，或者多个预设时间单位或多个电池循环周期或多个充电周期中的平均补偿的电池单元电压差呈现退化趋势的情况下，确定新能源汽车电池包处于不均衡状态。
当需要根据预设时间段的数据记录来评估电池包的均衡状态和退化趋势时，可以基于预设时间段内补偿的电池单元电压差确定预设时间段内归一化加权补偿的电池单元电压差，从而可以基于归一化加权补偿的电池单元电压差来估计新能源汽车电池包均衡状态。在本发明的至少一个示例性实施例中，在步骤S208之前，该方法还可以包括：确定预设时间段内的归一化加权补偿的电池单元电压差。步骤S208可以包括：根据预设时间段内的归一化加权补偿的电池单元电压差，估计新能源汽车电池包均衡状态。
在本发明的至少一个示例性实施例中，确定预设时间段内的归一化加权补偿的电池单元电压差的步骤可以包括：基于以下公式确定归一化加权后补偿的电池单元电压差：
Figure PCTCN2021081745-appb-000002
其中Δ 1,Δ 2,…,Δ N分别是预设时间段内补偿的电池单元电压差的各个实例，ω 1,ω 2,…,ω N是在预设时间段内的补偿的电池单元电压差的各个实例的权重，N是在预设时间段内的补偿的电池单元电压差的实例的数量。
在本发明的至少一个示例性实施例中，在预设时间段内的补偿的电池单元电压差的各个实例的权重ω 1,ω 2,…,ω N中，补偿的电池单元电压差的最新实例的权重可以等于预设值(例如，补偿的电池单元电压差的最新实例的权重可以等于1)，并且在预设时间段内的补偿的电池电压差的较早实例的权重呈指数衰减。这样，最新的补偿后电池单元电压差实例对归一化加权补偿后电池单元电压差具有最大的影响，而较早的补偿后电池单元电压差实例对归一 化加权补偿的电池单元电压差具有较小的影响。
在本发明的至少一个示例性实施例中，可以通过基于第二训练数据进行训练来获得预设时间段内的补偿的电池单元电压差的各个实例的权重ω 1,ω 2,…,ω N，其中，第二训练数据包括具有异常电池包电池失衡的实际车辆记录。
基于上述归一化加权补偿的电池单元电压差，可以实时估计新能源汽车电池包均衡状态。在本发明的至少一个示例性实施例中，步骤S208可以包括：在归一化加权补偿的电池单元电压差大于第二阈值或者多个预设时间段内归一化加权补偿的电池单元电压差呈现退化趋势的情况下，确定新能源汽车电池包处于不均衡状态。
当补偿的电池单元电压差(平均补偿的电池电压差或归一化加权补偿的电池电压差)表明电池组处于不均衡状态或电池包均衡条件出现退化趋势时，可以为不同类型新能源汽车(例如，EV，HEV，PHEV等)的电池包生成自动预测性维护通知。图3示出了根据本发明的示例性实施例的估计新能源汽车电池包均衡状态的方法的流程图。如图3所示，在本发明的至少一个示例性实施例中，该方法可以进一步包括步骤S302。
在步骤S302中，当根据补偿的电池单元电压差估计新能源汽车电池包均衡状态的结果指示新能源汽车电池包处于不均衡状态时，发送需要维护的警告。
本实施例中描述的方法可以应用于分析存储在云服务器中的相关数据并有效地估计当前电池包均衡状态。可以检测到电池包均衡条件的退化趋势，从而可以为不同类型的新能源汽车NEV(例如EV，HEV，PHEV等)的电池包生成自动的预测性维护通知。如果在发出预警通知后不久进行故障排除和预测性维护措施，则可以延长电池寿命以创造更好的客户体验，并可以大大降低保修成本。所提出的方法可以实现高水平的预测性维护和客户满意度，而不是会导致不良的客户体验和整车厂更高的保修费用的被动维护。
要注意的是，为了简单说明起见，该方法的每个前述实施例被描述为一系列动作组合。但是本领域技术人员应该知道，本发明不限于所描述的步骤的顺序，这是因为根据本发明，某些步骤可以以其他顺序执行或者同时执行。 此外，本领域技术人员还应该知道，说明书中所描述的所有实施例都是优选实施例，所涉及的动作和模块可能不是必需的。
根据本发明的另一方面，提供了一种估计新能源汽车电池包均衡状态的系统。该系统可以位于服务器或机载计算设备或具有计算能力的任何设备中。图4示出了根据本发明实施例的估计新能源汽车电池包均衡状态的系统的示意图。如图4所示，该系统包括存储器42和处理器44，其中计算机程序被存储在存储器42中，并且处理器44被配置为执行该计算机程序以实施如前述方法实施例中任一或多个方法。
基于前述实施例中的描述，可以得到估计新能源汽车电池包均衡状态的系统的其他方面，在此不再赘述。
根据本发明的另一方面，提供了一种非暂时性计算机可读存储介质。非暂时性计算机可读存储介质存储有计算机程序，其中，当该计算机程序被处理器执行时，使处理器执行如前述实施例中的任一或多个方法。
基于前述实施例中的描述，可以得到估计新能源汽车电池包均衡状态的系统的其他方面，在此不再赘述。
根据本发明的又一个实施例，描述了一种估计新能源汽车电池包均衡状态的详细方法。
如以下等式1所示，本实施例中提出的方法使用机器学习(Machine Learning，ML)模型来估计在不同工作条件下诸如车辆电流i，电池包荷电状态SOC，电池温度T，车速V，里程表值等的期望的电池单元电压差值(即，δ ref)。
然后如等式2所示，为每个单独的车辆测量的实际电池单元电压差(δ act)与由模型估计的电池单元电压差参考值(δ ref)之间的差(Δ)在本实施例中被称为补偿的电池单元电压差(Δ)，它可以被定义为实时量化每个电池包的电池单元均衡状态的度量标准。
δ ref＝f(i,SOC,T,…)   (等式1)
Δ＝δ act-δ ref   (等式2)
等式1所示的上述机器学习模型可以用线性或非线性模型函数f来实现。
(1)用线性函数f实现的机器学习模型
在一种实施方式中，可以定义如等式3中所示的线性模型函数。如图3所示，其中w 0,w 1和w 2是要训练和优化的权重。
δ ref＝w 0+w 1*i+w 2*SOC   (等式3)
在另一种实施方式中，可以将更多特征(例如温度，速度，里程表等)及其更高阶项添加到等式3中，以实现更好的模型拟合。
(2)用非线性函数f实现的机器学习模型
在一种实施方式中，可以采用非线性模型，例如神经网络(NN)和随机森林(RF)。本领域技术人员应该理解，在该解决方案中也可以采用其他类型的非线性模型。例如，具有两个隐藏层的NN模型比线性模型可以实现更好的模型拟合。
上面的ML模型基于正常车队性能定义了在给定工作条件下的期望的电池单元电压参考值(δ ref)作为训练数据，其中一些明显异常的车辆数据(例如，最高的1％)可以在数据清理期间被滤除。假设车队训练数据中的大多数车辆运行正常，则ML模型输出的期望的电池单元电压差参考值将非常接近在不同工作条件下正常车队电池单元电压差的平均值或中间值。然后，如等式2所定义的补偿的电池单元电压差(Δ)，可以用作度量标准来量化每个电池包的电池单元均衡状态。可以为每个时间记录计算该补偿的电池单元电压差(Δ)，并每天或通过每次电池跳闸/充电周期进行合计。然后，当Δ高于某个阈值和/或其时间序列显示出一定的增长趋势(即退化)时，便可以识别出高风险车辆。
图5(a)是在一天中对一种类型的电动汽车(EV)充电期间每辆车的平均电池单元电压差δ(原始数据)的柱形图(标准差std＝0.0062)。图5(b)是对于以上相同数据(标准差std＝0.0037)的每辆车的平均补偿的电池单元电压差Δ值的柱形图。如图5(a)和5(b)所示，对于研究的一种类型的电动汽车，可以观察到对于原始电池单元电压差(δ)数据和补偿的电池单元电压差(Δ)数据，车队分布明显不同，其中电池单元电压差参考值δ ref由具有从一天的数据的样本进行训练的两层的NN模型给定。后者更集中并且更接近于具有较小标准偏差的正态分布，如图5(b)所示。因此，尽管工作条件不同，但提出的度量标准仍可以更好地实时描述每辆车的电池 单元均衡状态。
对于每辆车，其当前电池单元不均衡状态可以被定义为其一段时间(例如最近N天)的其最近补偿的电池单元电压差Δ值的归一化加权平均值。例如，最后一天的权重为1，并且比最后一天早几天的权重呈指数衰减，即
Figure PCTCN2021081745-appb-000003
所有车辆(在最近N天内运行过的车辆)都可以通过使用上述度量标准评估的电池单元失衡状态进行排名。高于某个阈值或在时间序列上显示出退化趋势的电池失衡可被定义为高风险车辆，建议在经销商处立即注意并采取预防性维护措施。
线性(例如等式3)和/或非线性(例如NN)ML模型中的权重要训练和优化。等式4中权重的衰减方案也是如此。在一种实施方式中，基于正常车辆的训练数据，选择参数以使模型拟合误差最小。
带有建议解决方案的产品可以在基于云的服务器或机载计算设备中实现。它可以分析由车载传感器和/或CAN总线提供的相关数据的时间序列，识别所有具有不健康退化趋势的车辆，并为各种新能源汽车NEV(例如EV，HEV，PHEV等)生成自动的预测性维护警告通知。它将确保所有正在运行的电池包在健康均衡的条件下工作，并在早期阶段检测到任何潜在的电池包不均衡问题或不健康的退化趋势。一旦为一种车型确定了合理的警报阈值，就可以将“需要维护”警告直接以不同的格式发送给整车厂，经销商和客户，以便可以及早采取故障排除和预测性维护措施，以延长电池寿命，避免为整车厂造成高昂的保修损失，并极大地提高客户满意度。
以上仅是本发明的示例性实施方式。应当指出，在不脱离本发明原理的前提下，本领域普通技术人员也可以做出许多改进和补充，这些改进和补充应属于本发明的保护范围。

Claims (20)

  1. 一种估计新能源汽车电池包均衡状态的方法，包括：
    获取为新能源汽车电池包测量的实际的电池单元电压差，并获取用于测量所述实际的电池单元电压差的工作条件的一个或多个工作条件参数，其中，所述实际的电池单元电压差是所述电池包中最大电池单元电压与最小电池电源电压之间的差；
    根据所述一个或多个工作条件参数确定期望的电池单元电压差；
    根据所述实际的电池单元电压差和期望的电池单元电压差，确定补偿的电池单元电压差；以及
    根据补偿的电池单元电压差估计新能源汽车电池包均衡状态。
  2. 根据权利要求1所述的方法，其中，所述工作条件的一个或多个工作条件参数包括以下至少之一：
    车辆电流，荷电状态，电池温度，车辆速度或里程表值。
  3. 根据权利要求1所述的方法，其中，根据所述一个或多个工作条件参数确定期望的电池单元电压差包括：
    通过使用机器学习模型，根据所述一个或多个工作条件参数确定期望的电池单元电压差。
  4. 根据权利要求3所述的方法，其中，所述机器学习模型为线性模型，通过使用所述机器学习模型，根据所述一个或多个工作条件参数确定期望的电池单元电压差包括：
    基于以下公式确定期望的电池单元电压差δ ref：
    δ ref＝w 0+w 1*P 1+…+w M*P M
    其中，w 0,w 1,…,w M是线性模型的权重，P 1,…,P M为一个或多个工作条件参数，M为一个或多个工作条件参数的数量。
  5. 根据权利要求4所述的方法，其中，所述线性模型的权重w 0,w 1,…,w M基于第一训练数据通过训练获得，其中，第一训练数据包括用于多个不同工作条件的一个或多个工作条件参数，以及在多个不同的工作条件下为多个新能源汽车电池包测量的实际电池单元电压差。
  6. 根据权利要求5所述的方法，其中，在基于第一训练数据通过训练 获得线性模型的权重w 0,w 1,…,w M之前，为满足预设条件的新能源汽车电池包测量的实际电池单元电压差从第一训练数据中去除，其中，预设条件包括：性能异常的新能源汽车，多辆新能源汽车中性能最高或最低的新能源汽车。
  7. 根据权利要求3所述的方法，其中，所述机器学习模型为非线性模型，通过使用所述机器学习模型，根据所述一个或多个工作条件参数确定期望的电池单元电压差包括：
    通过使用一个或多个工作条件参数作为输入，基于所述非线性模型确定期望的电池电压差δ ref，所述非线性模型包括基于神经网络的模型或基于随机森林的模型。
  8. 根据权利要求7所述的方法，其中，所述非线性模型的权重通过基于第一训练数据的训练获得，其中，第一训练数据包括用于多个不同工作条件的一个或多个工作条件参数，以及在多个不同的工作条件下为多个新能源汽车电池包测量的实际电池单元电压差。
  9. 根据权利要求8所述的方法，其中，在基于第一训练数据通过训练获得所述非线性模型的权重之前，为满足预设条件的新能源汽车电池包测量的实际电池单元电压差从第一训练数据中去除，其中，预设条件包括：性能异常的新能源汽车，多辆新能源汽车中性能最高或最低的新能源汽车。
  10. 根据权利要求1所述的方法，其中，根据所述实际的电池单元电压差和期望的电池单元电压差确定补偿的电池单元电压差，包括：
    根据以下公式确定补偿的电池单元电压差Δ：
    Δ＝δ act-δ ref
    其中，δ act是实际的电池单元电压差，δ ref是期望的电池单元电压差。
  11. 根据权利要求1所述的方法，其中，为新能源汽车电池包测得的实际的电池单元电压差包括针对每个时间记录的测量的实际的电池单元电压差；以及
    根据所述实际的电池单元电压差和期望的电池单元电压差确定补偿的电池单元电压差，包括：
    确定每个时间记录的补偿的电池单元电压差，并在预设时间单位内对每个时间记录的补偿的电池单元电压差进行合计，以获得平均补偿的电池单元 电压差；或者
    确定每个时间记录的补偿的电池单元电压差，并以电池循环周期为基础合计每个时间记录的补偿的电池单元电压差，以获得平均补偿的电池单元电压差；或者
    确定每个时间记录的补偿的电池单元电压差，并以充电周期为基础合计每个时间记录的补偿的电池单元电压差，以获得平均补偿的电池单元电压差。
  12. 根据权利要求11所述的方法，其中，根据补偿的电池单元电压差估计新能源汽车电池包均衡状态包括：
    在平均补偿的电池单元电压差大于第一阈值，或者多个预设时间单位或多个电池循环周期或多个充电周期中的平均补偿的电池单元电压差呈现退化趋势的情况下，确定新能源汽车电池包处于不均衡状态。
  13. 根据权利要求1所述的方法，其中，在根据补偿的电池单元电压差估计新能源汽车电池包均衡状态之前，还包括：确定预设时间段内的归一化加权补偿的电池单元电压差；以及
    根据补偿的电池单元电压差估计新能源汽车电池包均衡状态包括：根据预设时间段内的归一化加权补偿的电池单元电压差，估计新能源汽车电池包均衡状态。
  14. 根据权利要求13所述的方法，其中，确定预设时间段内的归一化加权补偿的电池单元电压差包括：
    基于以下公式确定归一化加权后补偿的电池单元电压差：
    Figure PCTCN2021081745-appb-100001
    其中，Δ 1,Δ 2,…,Δ N分别是预设时间段内补偿的电池单元电压差的各个实例，ω 1,ω 2,…,ω N是在预设时间段内的补偿的电池单元电压差的各个实例的权重，N是在预设时间段内的补偿的电池单元电压差的实例的数量。
  15. 根据权利要求14所述的方法，其中，在预设时间段内的补偿的电池单元电压差的各个实例的权重ω 1,ω 2,…,ω N中，补偿的电池单元电压差的最新实例的权重等于预设值，并且在预设时间段内的补偿的电池电压差的较早实例的权重呈指数衰减。
  16. 根据权利要求14所述的方法，其中，预设时间段内的补偿的电池单元电压差的各个实例的权重ω 1,ω 2,…,ω N通过基于第二训练数据进行训练获得，其中，第二训练数据包括具有异常电池包电池失衡的实际车辆记录。
  17. 根据权利要求13所述的方法，其中，根据预设时间段内的归一化加权补偿的电池单元电压差，估计新能源汽车电池包均衡状态包括：
    在归一化加权补偿的电池单元电压差大于第二阈值或者多个预设时间段内归一化加权补偿的电池单元电压差呈现退化趋势的情况下，确定新能源汽车电池包处于不均衡状态。
  18. 根据权利要求1所述的方法，其中，还包括：
    当根据补偿的电池单元电压差估计新能源汽车电池包均衡状态的结果指示新能源汽车电池包处于不均衡状态时，发送需要维护的警告。
  19. 一种估计新能源汽车电池包均衡状态的系统，所述系统位于服务器或机载计算设备中，所述系统包括存储器和处理器，其中计算机程序被存储在所述存储器中，并且所述处理器被配置为执行所述计算机程序以实施如权利要求1所述的方法。
  20. 一种非暂时性计算机可读存储介质，所述非暂时性计算机可读存储介质存储有计算机程序，其中，当所述计算机程序被处理器执行时，使所述处理器执行如权利要求1所述的方法。
PCT/CN2021/081745 2020-10-28 2021-03-19 一种估计新能源汽车电池包均衡状态的方法及系统 WO2022088591A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202180004444.0A CN114982089A (zh) 2020-10-28 2021-03-19 一种估计新能源汽车电池包均衡状态的方法及系统

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US17/082,051 2020-10-28
US17/082,051 US11513164B2 (en) 2020-10-28 2020-10-28 Method and system for estimating battery pack balance state of new energy vehicle

Publications (1)

Publication Number Publication Date
WO2022088591A1 true WO2022088591A1 (zh) 2022-05-05

Family

ID=81256913

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/081745 WO2022088591A1 (zh) 2020-10-28 2021-03-19 一种估计新能源汽车电池包均衡状态的方法及系统

Country Status (3)

Country Link
US (1) US11513164B2 (zh)
CN (1) CN114982089A (zh)
WO (1) WO2022088591A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115312895B (zh) * 2022-10-12 2022-12-20 启东市航新实用技术研究所 一种监测新能源车辆的电池包的稳态方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101262138A (zh) * 2007-12-05 2008-09-10 奇瑞汽车股份有限公司 锂离子动力电池的单体均衡化系统
CN103779938A (zh) * 2014-02-11 2014-05-07 湖南江麓容大车辆传动股份有限公司 电池组用充电均衡电路、方法及放电均衡电路、方法
US10059217B2 (en) * 2014-08-12 2018-08-28 Hyundai Motor Company System and method for controlling battery switching serial/parallel connection of battery modules due to accelerator operation
US10416238B2 (en) * 2010-12-06 2019-09-17 Exergonix, Inc. Electrochemical cell monitoring and balancing circuit with self-diagnostic feature
CN110293881A (zh) * 2019-07-15 2019-10-01 西南交通大学 一种基于反激式变压器的锂电池双向均衡系统及方法
CN111049153A (zh) * 2019-11-21 2020-04-21 西安工程大学 一种基于模糊神经网络的电压无功优化控制方法
CN111469713A (zh) * 2020-04-28 2020-07-31 东风汽车集团有限公司 一种新能源汽车动力电池被动均衡控制方法

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1814206A1 (en) * 2006-01-27 2007-08-01 Berner Fachhochschule Hochschule für Technik und Architektur Biel Battery balancing apparatus
JP2009071936A (ja) * 2007-09-11 2009-04-02 Fuji Heavy Ind Ltd 組電池の電圧均等化システム
US8552693B2 (en) * 2009-07-17 2013-10-08 Tesla Motors, Inc. Low temperature charging of Li-ion cells
JP2014060868A (ja) * 2012-09-18 2014-04-03 Suzuki Motor Corp 組電池均等化装置
US9091738B2 (en) * 2012-10-10 2015-07-28 GM Global Technology Operations LLC Vehicle battery pack cell voltage determination
KR101967863B1 (ko) 2012-11-14 2019-04-11 현대모비스 주식회사 고전압 셀 밸런싱의 밸런싱 필요 시간 추정 장치 및 방법
US9213068B2 (en) * 2013-03-14 2015-12-15 GM Global Technology Operations LLC Method and system for determining a battery element voltage
US9770997B2 (en) * 2013-06-11 2017-09-26 Ford Global Technologies, Llc Detection of imbalance across multiple battery cells measured by the same voltage sensor
GB2528290A (en) * 2014-07-16 2016-01-20 John Leslie Gordon Hardy Battery management
CN105449740B (zh) * 2015-11-13 2018-09-04 中国东方电气集团有限公司 一种储能锂电池主动均衡控制系统及控制方法
US11437827B2 (en) * 2016-03-01 2022-09-06 Volvo Truck Corporation Control of a relatively low current fed to a battery pack
DE102017201622A1 (de) * 2017-02-01 2018-08-02 Bayerische Motoren Werke Aktiengesellschaft Verfahren zum Betrieb eines Energiespeichersystems und Energiespeichersystem
CN112952224B (zh) * 2019-12-11 2022-12-20 南京泉峰科技有限公司 一种电池包的充电平衡方法、系统和电池包

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101262138A (zh) * 2007-12-05 2008-09-10 奇瑞汽车股份有限公司 锂离子动力电池的单体均衡化系统
US10416238B2 (en) * 2010-12-06 2019-09-17 Exergonix, Inc. Electrochemical cell monitoring and balancing circuit with self-diagnostic feature
CN103779938A (zh) * 2014-02-11 2014-05-07 湖南江麓容大车辆传动股份有限公司 电池组用充电均衡电路、方法及放电均衡电路、方法
US10059217B2 (en) * 2014-08-12 2018-08-28 Hyundai Motor Company System and method for controlling battery switching serial/parallel connection of battery modules due to accelerator operation
CN110293881A (zh) * 2019-07-15 2019-10-01 西南交通大学 一种基于反激式变压器的锂电池双向均衡系统及方法
CN111049153A (zh) * 2019-11-21 2020-04-21 西安工程大学 一种基于模糊神经网络的电压无功优化控制方法
CN111469713A (zh) * 2020-04-28 2020-07-31 东风汽车集团有限公司 一种新能源汽车动力电池被动均衡控制方法

Also Published As

Publication number Publication date
US11513164B2 (en) 2022-11-29
US20220128631A1 (en) 2022-04-28
CN114982089A (zh) 2022-08-30

Similar Documents

Publication Publication Date Title
Ning et al. Adaptive sliding mode observers for lithium-ion battery state estimation based on parameters identified online
JP6588632B2 (ja) 電池制御装置
US9533597B2 (en) Parameter identification offloading using cloud computing resources
JP5400732B2 (ja) パラメータ推定装置
US20210181263A1 (en) Method and battery management system for ascertaining a state of health of a secondary battery
CN112068004A (zh) 一种电池异常、电池充电剩余时间的确定方法及装置
US11385631B2 (en) Method and system for detecting faults in a charging infrastructure system for electric vehicles
JP7016704B2 (ja) 二次電池システム
AU2021315194A1 (en) Battery pack diagnosing method, cell diagnosing method, battery pack diagnosing device, and cell diagnosing device
WO2022088591A1 (zh) 一种估计新能源汽车电池包均衡状态的方法及系统
He et al. Voltage abnormality-based fault diagnosis for batteries in electric buses with a self-adapting update model
WO2022016862A1 (zh) 用于监控电池组健康状况的方法和系统
Vennam et al. A novel coupled electro-thermal-aging model for simultaneous SOC, SOH, and parameter estimation of lithium-ion batteries
Lai et al. A quantitative method for early-stage detection of the internal-short-circuit in Lithium-ion battery pack under float-charging conditions
CN116893366A (zh) 用于运行系统以识别设备的电蓄能器的异常的方法和装置
CN116699435A (zh) 基于机器学习方法识别电池组中的临界异常的方法和装置
JP2019152657A (ja) 電池監視システム
CN115219932A (zh) 用于评价设备电池组的相对老化状态的方法和装置
Gholizadeh et al. SOC estimation for a lithium-ion battery by designing a nonlinear observer based on an equivalent circuit model
Li et al. Robust state of charge estimation of lithium-ion batteries via an iterative learning observer
CN113009363A (zh) 用于测量电池欧姆电阻的方法和装置
JP2021145508A (ja) 充放電制御方法、電池搭載機器、管理システム、充放電制御プログラム、管理方法、管理サーバ及び管理プログラム
Wang et al. Voltage measurement-based recursive adaptive method for internal short circuit fault diagnosis in lithium-ion battery packs
US20230366937A1 (en) Computer-implemented method for diagnosing states of a battery
US20230213587A1 (en) Method and System for Efficiently Monitoring Battery Cells of a Device Battery in an External Central Processing Unit Using a Digital Twin

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21884335

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21884335

Country of ref document: EP

Kind code of ref document: A1