WO2023184824A1 - 电池簇荷电状态的估计方法及系统、电子设备及存储介质 - Google Patents

电池簇荷电状态的估计方法及系统、电子设备及存储介质 Download PDF

Info

Publication number
WO2023184824A1
WO2023184824A1 PCT/CN2022/112838 CN2022112838W WO2023184824A1 WO 2023184824 A1 WO2023184824 A1 WO 2023184824A1 CN 2022112838 W CN2022112838 W CN 2022112838W WO 2023184824 A1 WO2023184824 A1 WO 2023184824A1
Authority
WO
WIPO (PCT)
Prior art keywords
charge
state
estimated value
battery cluster
sample data
Prior art date
Application number
PCT/CN2022/112838
Other languages
English (en)
French (fr)
Inventor
丁鹏
赵恩海
吴炜坤
顾单飞
郝平超
宋佩
严晓
张�杰
陈晓华
Original Assignee
上海玫克生储能科技有限公司
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 上海玫克生储能科技有限公司 filed Critical 上海玫克生储能科技有限公司
Publication of WO2023184824A1 publication Critical patent/WO2023184824A1/zh

Links

Images

Classifications

    • 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]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • 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
    • 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

Definitions

  • the present invention relates to the field of battery technology, and in particular to a method and system for estimating the state of charge of a battery cluster, electronic equipment and storage media.
  • SOC state of charge
  • SOH state of health
  • SOE state of energy
  • Battery output battery health state
  • Power SOP state of power
  • the main methods for estimating SOC include: discharge experimental method, ampere-hour integration method, open circuit voltage method, Kalman filter method, combined voltage correction method, etc.
  • Discharge experiment method This method is a relatively accurate prediction method. It uses constant current continuous discharge to obtain the discharged electricity. The discharge experiment method is often used to calibrate battery capacity. This method is suitable for all batteries, but it also has obvious shortcomings: first, the charge and discharge test takes a lot of time; second, the discharge experiment method cannot be used for working batteries.
  • Ampere-hour integration method is the most commonly used SOC estimation method.
  • the principle of the ampere-hour integration method is to equate the discharge capacity of the battery at different currents to the discharge capacity at a specific current.
  • the accuracy of this method will be affected by the accuracy of the current sensor, and there is a cumulative error.
  • Open circuit voltage method Utilize the corresponding relationship between the battery OCV (Open Circuit Voltage, open circuit voltage) and the battery SOC, and estimate the SOC by measuring the battery's open circuit voltage. This method can obtain the battery SOC more directly.
  • the basic principle of the open circuit voltage method is to let the battery stand still to restore the battery terminal voltage to the circuit voltage, that is, to eliminate the influence of the polarization voltage, the standing time generally takes more than 2 hours, so this method is not suitable for real-time online monitoring.
  • battery OCV measurement is complicated, and as the battery ages, the battery OCV will undergo slight changes, causing SOC errors.
  • Kalman filter method This method is based on the ampere-hour integration method and is the optimal estimate of the state of the dynamic system in the sense of minimum variance.
  • the core idea is a recursive equation that includes the state of charge estimate and the covariance matrix that reflects the estimate error.
  • the covariance matrix is used to give the estimate error range.
  • the Kalman filter method requires a large amount of matrix operations and requires a microcontroller with high computing power.
  • the accuracy of the Kalman filter method depends on the establishment of an equivalent model. Due to the aging effect of the battery itself, it is difficult to establish an equivalent battery model that is accurate throughout its life.
  • Combined voltage correction method If the energy storage battery has constant current charging conditions and the charging conditions are stable, using ampere-hour integration combined with the charging curve to correct the SOC is an algorithm often used by most manufacturers. This algorithm has high stability, simple calculation and strong stability, and is suitable for embedded environments. However, the accuracy of this algorithm is affected by the accuracy of the charging curve, and the charging curve usually uses the battery charging curve tested at the factory. As the battery ages, the battery curve will gradually change, and the initial test curve does not meet the characteristics of the aging battery. , at this time, using the initial charging curve to correct the SOC will cause unpredictable errors. At the same time, when encountering frequency modulation power stations and frequent changes in current, it is difficult to extract the best charging and discharging parameters.
  • the technical problem to be solved by the present invention is to overcome the defects existing in the SOC estimation method in the prior art and provide a battery cluster state-of-charge estimation method and system, electronic equipment and storage media.
  • a first aspect of the present invention provides a method for estimating the state of charge of a battery cluster, which includes the following steps:
  • a final estimate of the state of charge is determined based on the first estimate, the second estimate, and the distance between the target data and the sample data.
  • the step of determining the final estimated value of the state of charge based on the first estimated value, the second estimated value and the distance between the target data and the sample data specifically includes:
  • the weight of the first estimated value and the weight of the second estimated value are determined according to the distance between the target data and the sample data.
  • the step of performing a weighted sum of the first estimated value and the second estimated value specifically includes:
  • the weight of the first estimated value is set to be smaller than the weight of the second estimated value.
  • the weight K of the first estimated value is set according to the following formula:
  • D is the distance between the target data and the sample data
  • D 1 is the maximum distance between the sample data
  • n is a hyperparameter used to represent the speed of K convergence
  • the weight is 1-K.
  • the target data input to the state-of-charge prediction model includes at least one of the following: the maximum cell voltage, the minimum cell voltage, the average cell voltage, the total voltage, the maximum temperature, the minimum voltage of the battery cluster. Temperature, current, charge and discharge status, voltage standard deviation, temperature standard deviation, voltage-temperature covariance.
  • the method for estimating the state of charge of the battery cluster further includes the following steps:
  • the target data is added to the sample data to obtain updated sample data; wherein, the second preset value Determined based on the maximum distance between the sample data;
  • the state of charge prediction model is retrained using the updated sample data.
  • the step of retraining the state of charge prediction model using updated sample data specifically includes:
  • the state-of-charge prediction model is retrained using the partial sample data.
  • a second aspect of the present invention provides a system for estimating the state of charge of a battery cluster, including:
  • a data acquisition module used to acquire target data related to the state of charge of the battery cluster
  • a first estimation module configured to estimate the state of charge of the battery cluster based on the target data using the ampere-hour integration method to obtain a first estimated value
  • the second estimation module is used to input the target data into the state of charge prediction model to estimate the state of charge of the battery cluster and obtain a second estimated value; wherein the state of charge prediction model is trained based on sample data. ;
  • a charge determination module configured to determine a final estimated value of the state of charge based on the first estimated value, the second estimated value, and the distance between the target data and the sample data.
  • a third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor.
  • the processor executes the computer program, the computer program is implemented as described in the first aspect.
  • a fourth aspect of the present invention provides a computer-readable storage medium on which a computer program is stored.
  • the computer program is executed by a processor, the method for estimating the state of charge of a battery cluster as described in the first aspect is implemented.
  • the positive and progressive effect of the present invention is that the ampere-hour integration method and the state-of-charge prediction model are combined to jointly estimate the state of charge of the battery cluster.
  • the ampere-hour integration method is used to obtain the first estimate of the state of charge of the battery cluster, and the state of charge of the battery cluster is obtained using the ampere-hour integration method.
  • the state of charge prediction model obtains a second estimate of the state of charge of the battery cluster.
  • the distance between the target data and the sample data for training the state of charge prediction model can reflect the accuracy of the state of charge prediction model in estimating the state of charge. According to The distance determines the respective proportions of the first estimated value and the second estimated value in the final estimated value, which can effectively improve the accuracy of battery cluster state-of-charge estimation.
  • the present invention does not require in-depth analysis of the reaction mechanism inside the battery cluster, nor does it need to identify the parameters of the equivalent circuit of the battery cluster, nor does it need to allow the battery cluster to stand still. It improves the accuracy of state-of-charge estimation while also reducing the cumulative error. .
  • FIG. 1 is a flow chart of a method for estimating the state of charge of a battery cluster provided in Embodiment 1 of the present invention.
  • FIG. 2 is a detailed flow chart of step S41 provided in Embodiment 1 of the present invention.
  • Figure 3 is a flow chart for updating the state of charge prediction model provided in Embodiment 1 of the present invention.
  • FIG. 4 is a schematic diagram of the estimation effect of the state of charge of a battery cluster provided in Embodiment 1 of the present invention.
  • FIG. 5 is a structural block diagram of a battery cluster state-of-charge estimation system provided in Embodiment 1 of the present invention.
  • FIG. 6 is a schematic structural diagram of an electronic device provided in Embodiment 2 of the present invention.
  • FIG. 1 is a schematic flow chart of a battery cluster state of charge estimation method provided in this embodiment.
  • the battery cluster state of charge estimation method can be executed by a battery cluster state of charge estimation system.
  • the battery cluster state of charge estimation system can be implemented by software and/or hardware, and the battery cluster state-of-charge estimation system can be part or all of the electronic device.
  • the electronic device in this embodiment can be a personal computer (PC), such as a desktop computer, an all-in-one computer, a notebook computer, a tablet computer, etc., or it can also be a mobile phone, a wearable device, a handheld computer (Personal Digital Assistant, PDA) and other terminal equipment.
  • PC personal computer
  • PDA Personal Digital Assistant
  • the method for estimating the state of charge of a battery cluster may include the following steps S1 to S4:
  • Step S1 Obtain target data related to the state of charge of the battery cluster.
  • the target data related to the state of charge of the battery cluster can also be called data that affects the state of charge of the battery cluster.
  • the battery cluster may include multiple battery boxes, and each battery box may include multiple cells.
  • Step S2 Use the ampere-hour integration method to estimate the state of charge of the battery cluster based on the target data to obtain a first estimated value.
  • step S2 the current I, rated capacity Capacity and health state SOH of the battery cluster in the target data can be substituted into the following formula to calculate the first estimated value SOC Ah :
  • Step S3 Input the target data into the state of charge prediction model to estimate the state of charge of the battery cluster to obtain a second estimated value.
  • the state of charge prediction model is trained based on sample data.
  • the state of charge prediction model can use GBDT (Gradient Boosting Decision Tree), and the decision tree used by GBDT is a CART regression tree.
  • GBDT Gradient Boosting Decision Tree
  • the decision tree used by GBDT is a CART regression tree.
  • the target data input to the state of charge prediction model may include basic information of the battery cluster, such as the maximum cell voltage V max , the minimum cell voltage V min , and the average cell voltage V of the battery cluster. ave , total voltage V total , maximum temperature T max , minimum temperature T min , average temperature T ave , current I, charge and discharge state Charge_state, etc.
  • the target data input to the state of charge prediction model may also include statistical information of the battery cluster, such as the voltage standard deviation ⁇ v , temperature standard deviation ⁇ T , voltage-temperature covariance ⁇ ( x m ,x k ) etc.
  • ⁇ V is the average voltage of each cell in the battery cluster
  • ⁇ T is the average value of each temperature measurement point in the battery cluster.
  • Step S4 Determine the final estimated value of the state of charge according to the first estimated value, the second estimated value and the distance between the target data and the sample data. Specifically, the respective proportions of the first estimated value and the second estimated value in the final estimated value may be determined according to the distance between the target data and the sample data.
  • the distance between the target data and the sample data may be calculated based on a metric matrix.
  • the ampere-hour integration method and the state of charge prediction model are combined to jointly estimate the state of charge of the battery cluster.
  • the ampere-hour integration method is used to obtain the first estimate of the battery cluster's state of charge
  • the state of charge prediction is used
  • the model obtains a second estimate of the state of charge of the battery cluster.
  • the distance between the target data and the sample data for training the state of charge prediction model can reflect the accuracy of the state of charge prediction model in estimating the state of charge.
  • the second estimate is determined based on the distance.
  • the respective proportions of the first estimated value and the second estimated value in the final estimated value can effectively improve the accuracy of battery cluster state-of-charge estimation.
  • step S4 specifically includes the following step S41:
  • Step S41 Perform a weighted sum of the first estimated value and the second estimated value to obtain a final estimated value of the state of charge.
  • the weight of the first estimated value and the weight of the second estimated value are determined according to the distance between the target data and the sample data.
  • the final estimate SOC is calculated according to the following formula:
  • SOC GBDT is the second estimated value
  • K is the weight of the first estimated value
  • 1-K is the weight of the second estimated value
  • step S41 includes the following steps S411 to S413:
  • Step S411 Determine whether the distance is greater than the first preset value. If yes, step S412 is executed. If not, step S413 is executed.
  • the first preset value may be determined according to the maximum distance between the sample data.
  • Step S412 Set the weight of the first estimated value to be greater than or equal to the weight of the second estimated value.
  • Step S413 Set the weight of the first estimated value to be smaller than the weight of the second estimated value.
  • the ampere-hour integration method is used to obtain the proportion of the first estimated value in the final estimated value. than higher. If the distance is less than or equal to the first preset value, it means that the target data is included in the sample data. At this time, the second estimated value obtained by using the state of charge prediction model accounts for a higher proportion of the final estimated value. .
  • the weight K of the first estimated value is set according to the following formula:
  • D is the distance between the target data and the sample data
  • D 1 is the maximum distance between the sample data
  • n is a hyperparameter used to represent the speed of K convergence, which can be adjusted according to the actual situation.
  • the weight of the second estimated value is 1-K.
  • D_gain 0
  • D_gain>0 it means that the target data is not included in the sample data, and the larger the D_gain is, the farther the distance is, and the closer K is to 1.
  • the first estimated value is the load estimated using the ampere-hour integration method. The electrical state accounts for a higher proportion of the final estimate.
  • Energy storage power stations are equipped with multiple battery clusters. These battery clusters generate a large amount of historical data every day. Sample data for training the state-of-charge prediction model and the corresponding state-of-charge can be selected from the historical data. Assume there are sample data for N battery clusters: The corresponding real state of charge is ⁇ y 1 ,y 2 ...y N ⁇ , the loss function is L(y,f(x)), and the number of iterations is M.
  • a strong learner for constructing a state of charge prediction model Specifically, it may include the following steps (1) to (3):
  • c usually takes the average value of all sample data corresponding to the true state of charge.
  • the sample data can be updated based on the acquired target data, and the updated sample data can be used to retrain the above state of charge prediction model.
  • the second preset value is determined according to the maximum distance between the sample data.
  • the second preset value may be the same as the above-mentioned first preset value, or may be greater than the above-mentioned first preset value.
  • the updated sample data includes original sample data and qualified target data.
  • the target data whose distance from the sample data is greater than the second preset value is the target data that meets the conditions.
  • the sample data can be reconstructed and the state of charge prediction model can be retrained.
  • the above-mentioned step of retraining the state-of-charge prediction model using the updated sample data specifically includes: extracting part of the sample data from the updated sample data through unilateral gradient sampling, And use the partial sample data to retrain the state-of-charge prediction model.
  • the sample data used to retrain the state of charge prediction model is first extracted through unilateral gradient sampling, then a new tree is obtained by fitting the residual value of the sample data, and finally the previous charge is updated. Electrical state prediction model, get the latest strong learner.
  • the negative gradient of the updated sample data is calculated to obtain:
  • FIG 4 is a schematic diagram illustrating the effect of estimating the state of charge of a battery cluster.
  • the battery cluster state of charge estimated using the ampere-hour integration method has a cumulative error, which is quite different from the real battery cluster state of charge.
  • the battery cluster state of charge estimated using the method provided in this embodiment It is less different from the real battery cluster state of charge and has higher accuracy.
  • This embodiment also provides a battery cluster state-of-charge estimation system, as shown in FIG. 5 , including a data acquisition module 40 , a first estimation module 41 , a second estimation module 42 and a charge determination module 43 .
  • the data acquisition module 40 is used to acquire target data related to the state of charge of the battery cluster.
  • the first estimation module 41 is configured to estimate the state of charge of the battery cluster according to the target data using the ampere-hour integration method to obtain a first estimated value.
  • the second estimation module 42 is used to input the target data into the state of charge prediction model to estimate the state of charge of the battery cluster to obtain a second estimated value; wherein the state of charge prediction model is trained based on sample data. .
  • the charge determination module 43 is configured to determine the final estimated value of the state of charge according to the first estimated value, the second estimated value, and the distance between the target data and the sample data.
  • the charge determination module is specifically configured to perform a weighted sum of the first estimated value and the second estimated value to obtain a final estimated value of the state of charge; wherein, The weight of the first estimated value and the weight of the second estimated value are determined according to the distance between the target data and the sample data.
  • the charge determination module is specifically configured to determine whether the distance is greater than a first preset value; wherein the first preset value is based on the maximum distance between the sample data Determine; and if yes, set the weight of the first estimated value to be greater than or equal to the weight of the second estimated value; and if no, set the weight of the first estimated value to be less than the second estimated value the weight of.
  • the target data input to the state of charge prediction model includes at least one of the following: the maximum cell voltage, the minimum cell voltage, the average cell voltage, the total cell voltage of the battery cluster, Voltage, maximum temperature, minimum temperature, average temperature, current, charge and discharge status, voltage standard deviation, temperature standard deviation, voltage-temperature covariance.
  • the battery cluster state-of-charge estimation system further includes a model training module for when the distance between the target data and the sample data is greater than a second preset value. , add the target data to the sample data to obtain updated sample data; wherein the second preset value is determined according to the maximum distance between the sample data; and retrain using the updated sample data The state of charge prediction model.
  • the model training module is specifically configured to extract partial sample data from the updated sample data through unilateral gradient sampling; and use the partial sample data to retrain the charge State prediction model.
  • the battery cluster state-of-charge estimation system in this embodiment may be a separate chip, chip module or electronic device, or may be a chip or chip module integrated in the electronic device.
  • modules/units included in the battery cluster state-of-charge estimation system described in this embodiment may be software modules/units or hardware modules/units, or they may be partly software modules/units and partly Hardware modules/units.
  • FIG. 6 is a schematic structural diagram of an electronic device provided in this embodiment.
  • the electronic device includes at least one processor and a memory communicatively connected with the at least one processor.
  • the memory stores a computer program that can be run by the at least one processor, and the computer program is executed by the at least one processor, so that the at least one processor can execute the battery cluster charge of Embodiment 1.
  • the electronic device provided in this embodiment can be a personal computer, such as a desktop computer, an all-in-one computer, a notebook computer, a tablet computer, etc., or it can also be a terminal device such as a mobile phone, a wearable device, or a handheld computer.
  • the electronic device 3 shown in FIG. 6 is only an example and should not impose any restrictions on the functions and scope of use of the embodiments of the present invention.
  • the components of the electronic device 3 may include, but are not limited to: the above-mentioned at least one processor 4, the above-mentioned at least one memory 5, and a bus 6 connecting different system components (including the memory 5 and the processor 4).
  • Bus 6 includes a data bus, an address bus and a control bus.
  • the memory 5 may include volatile memory, such as a random access memory (RAM) 51 and/or a cache memory 52 , and may further include a read-only memory (ROM) 53 .
  • RAM random access memory
  • ROM read-only memory
  • the memory 5 may also include a program/utility 55 having a set of (at least one) program modules 54 including, but not limited to: an operating system, one or more application programs, other program modules, and program data. Each of the examples, or some combination thereof, may include the implementation of a network environment.
  • the processor 4 executes a computer program stored in the memory 5 to perform various functional applications and data processing, such as the above-mentioned battery cluster state-of-charge estimation method.
  • Electronic device 3 may also communicate with one or more external devices 7 (eg keyboard, pointing device, etc.). This communication can take place via the input/output (I/O) interface 8. Moreover, the electronic device 3 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and/or a public network such as the Internet) through the network adapter 9 . As shown in FIG. 6 , the network adapter 9 communicates with other modules of the electronic device 3 through the bus 6 .
  • LAN local area network
  • WAN wide area network
  • public network such as the Internet
  • This embodiment provides a computer-readable storage medium storing a computer program, which implements the battery cluster state-of-charge estimation method of Embodiment 1 when the computer program is executed by a processor.
  • the readable storage medium that can be used may more specifically include but is not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device or any of the above. The right combination.
  • the present invention can also be implemented in the form of a program product, which includes program code.
  • program product which includes program code.
  • the program code is used to cause the electronic device to execute the implementation.
  • program code for executing the present invention can be written in any combination of one or more programming languages, and the program code can be completely executed on the electronic device, partially executed on the electronic device, as an independent
  • the software package executes partially on the electronic device and partially on the remote device or entirely on the remote device.

Abstract

一种电池簇荷电状态的估计方法及系统、电子设备及存储介质，估计方法包括以下步骤：获取与电池簇的荷电状态相关的目标数据（S1）；利用安时积分法根据目标数据对电池簇的荷电状态进行估计，得到第一估计值（S2）；将目标数据输入荷电状态预测模型对电池簇的荷电状态进行估计，得到第二估计值（S3）；其中，荷电状态预测模型基于样本数据训练得到；根据第一估计值、第二估计值以及目标数据与样本数据之间的距离确定荷电状态的最终估计值（S4）。结合安时积分法和荷电状态预测模型共同估计电池簇的荷电状态，能够有效提高电池簇荷电状态估计的准确性。

Description

电池簇荷电状态的估计方法及系统、电子设备及存储介质 技术领域
本发明涉及电池技术领域，特别涉及一种电池簇荷电状态的估计方法及系统、电子设备及存储介质。
背景技术
SOC(state of charge)是电池的荷电状态，在储能的电池管理系统中，电池SOC是核心，影响着电池健康状态SOH(state of health)、剩余能量SOE(state of energy)以及电池输出功率SOP(state of power)，甚至影响着电池安全。但是，由于电池表现为非线性特征，受温度、使用时间、倍率等各种因素影响，因此很难对电池SOC进行准确地预估。国标中，电池SOC的预估准确度要求为5％。
目前对荷电状态的研究，大多通过测量电池的电流、电压、内阻等相关特征参数，建立特征参数与电池SOC的对应函数关系，利用这些函数关系修正SOC，因此电池特征参数的准确性非常重要。目前对SOC估计的主要方法有：放电实验法、安时积分法、开路电压法、卡尔曼滤波法、组合电压修正方法等。
放电实验法：该方法是比较准确的预估方法，它采用恒流持续放电获取其放出电量。放电实验法常常被使用来标定电池的容量，该方法适用于所有电池，但也存在明显的缺点：首先，充放电试验需要花费大量时间；其次，放电实验法不能用于工作中的电池。
安时(Ah)积分法：安时积分法是最常用的SOC估计方法，安时积分法的原理是将电池在不同电流下的放电电量等价为某个具体电流下的放电电量。但是该方法精度会受电流传感器的精度影响，而且存在着累计误差。
开路电压法：利用电池OCV(Open Circuit Voltage，开路电压)与电池SOC的对应关系，通过测量电池的开路电压来估算SOC，用这种方法较为 直接地得到电池SOC。但是，由于开路电压法的基本原理是将电池静置，使电池端电压恢复至电路电压，即要消除极化电压的影响，静置时间一般需要2小时以上，所以该方法不适合实时在线监测，另外电池OCV测量复杂，且随着电池老化，电池OCV会发生微小变化造成SOC出现误差。
卡尔曼滤波法：该方法建立在安时积分法的基础之上，是对动力系统的状态做出最小方差意义上的最优估计。核心思想是包括荷电状态估计值和反映估计误差的、协方差矩阵的递归方程，协方差矩阵用来给出估算误差范围。卡尔曼滤波法在实际运用中矩阵运算量大，需要高运算能力的单片机。卡尔曼滤波法的精度取决于等价模型的建立，由于电池自身老化影响，很难建立一个整个生命内都准确的等价电池模型。
组合电压修正方法：储能电池如果有恒流充电工况，充电工况稳定，利用安时积分结合充电曲线修正SOC是大多数厂家经常用到的算法。该算法稳定性较高、计算简单、稳定性强适用于嵌入式环境。但是，该算法的精度受充电曲线精度的影响，而充电曲线通常采用的是出厂测试的电池充电曲线，随着电池的老化，电池曲线会逐渐变化，初始测试的曲线不符合老化后的电池特征，这时采用初始充电曲线修正SOC会造成不可预测的误差，同时遇到调频电站，电流频繁变化的场景，很难提取最佳的充放电参数。
发明内容
本发明要解决的技术问题是为了克服现有技术中SOC估计方法中存在的缺陷，提供一种电池簇荷电状态的估计方法及系统、电子设备及存储介质。
本发明是通过下述技术方案来解决上述技术问题：
本发明的第一方面提供一种电池簇荷电状态的估计方法，包括以下步骤：
获取与电池簇的荷电状态相关的目标数据；
利用安时积分法根据所述目标数据对所述电池簇的荷电状态进行估计，得到第一估计值；
将所述目标数据输入荷电状态预测模型对所述电池簇的荷电状态进行 估计，得到第二估计值；其中，所述荷电状态预测模型基于样本数据训练得到；
根据所述第一估计值、所述第二估计值以及所述目标数据与所述样本数据之间的距离确定荷电状态的最终估计值。
可选地，所述根据所述第一估计值、所述第二估计值以及所述目标数据与所述样本数据之间的距离确定荷电状态的最终估计值的步骤具体包括：
对所述第一估计值和所述第二估计值进行加权求和，得到荷电状态的最终估计值；
其中，所述第一估计值的权重和所述第二估计值的权重根据所述目标数据与所述样本数据之间的距离确定。
可选地，所述对所述第一估计值和所述第二估计值进行加权求和的步骤具体包括：
判断所述距离是否大于第一预设值；其中，所述第一预设值根据所述样本数据之间的最大距离确定；
若是，则设置所述第一估计值的权重大于等于所述第二估计值的权重；
若否，则设置所述第一估计值的权重小于所述第二估计值的权重。
可选地，根据以下公式设置所述第一估计值的权重K：
Figure PCTCN2022112838-appb-000001
其中，
Figure PCTCN2022112838-appb-000002
其中，D为所述目标数据与所述样本数据之间的距离，D 1为所述样本数据之间的最大距离，n为超参数，用于表示K收敛的速度，所述第二估计值的权重为1-K。
可选地，输入所述荷电状态预测模型的目标数据包括以下中的至少一种：所述电池簇的最大单体电压、最小单体电压、单体平均电压、总电压、最高温度、最低温度、电流、充放电状态、电压标准差、温度标准差、电压温度协方差。
可选地，所述电池簇荷电状态的估计方法还包括以下步骤：
若所述目标数据与所述样本数据之间的距离大于第二预设值，则将所述目标数据加入所述样本数据中，得到更新后的样本数据；其中，所述第二预设值根据所述样本数据之间的最大距离确定；
利用更新后的样本数据重新训练所述荷电状态预测模型。
可选地，所述利用更新后的样本数据重新训练所述荷电状态预测模型的步骤具体包括：
通过单边梯度采样的方式从更新后的样本数据中提取部分样本数据；
利用所述部分样本数据重新训练所述荷电状态预测模型。
本发明的第二方面提供一种电池簇荷电状态的估计系统，包括：
数据获取模块，用于获取与电池簇的荷电状态相关的目标数据；
第一估计模块，用于利用安时积分法根据所述目标数据对所述电池簇的荷电状态进行估计，得到第一估计值；
第二估计模块，用于将所述目标数据输入荷电状态预测模型对所述电池簇的荷电状态进行估计，得到第二估计值；其中，所述荷电状态预测模型基于样本数据训练得到；
荷电确定模块，用于根据所述第一估计值、所述第二估计值以及所述目标数据与所述样本数据之间的距离确定荷电状态的最终估计值。
本发明的第三方面提供一种电子设备，包括存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如第一方面所述的电池簇荷电状态的估计方法。
本发明的第四方面提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现如第一方面所述的电池簇荷电状态的估计方法。
在符合本领域常识的基础上，上述各优选条件，可任意组合，即得本发明各较佳实例。
本发明的积极进步效果在于：结合安时积分法和荷电状态预测模型共同 估计电池簇的荷电状态，具体地，利用安时积分法得到电池簇荷电状态的第一估计值，利用荷电状态预测模型得到电池簇荷电状态的第二估计值，目标数据与训练荷电状态预测模型的样本数据之间的距离可以反映出荷电状态预测模型估计荷电状态的准确性，根据所述距离确定第一估计值和第二估计值分别在最终估计值中的占比，能够有效提高电池簇荷电状态估计的准确性。
另外，本发明无需深入分析电池簇内部的反应机理，也无需辨识电池簇等效电路的参数，也无需对电池簇进行静置处理，在提高荷电状态估计准确性的同时还降低了累计误差。
附图说明
图1为本发明实施例1提供的一种电池簇荷电状态的估计方法的流程图。
图2为本发明实施例1提供的一种步骤S41的详细流程图。
图3为本发明实施例1提供的一种更新荷电状态预测模型的流程图。
图4为本发明实施例1提供的一种电池簇荷电状态的估计效果示意图。
图5为本发明实施例1提供的一种电池簇荷电状态的估计系统的结构框图。
图6为本发明实施例2提供的一种电子设备的结构示意图。
具体实施方式
下面通过实施例的方式进一步说明本发明，但并不因此将本发明限制在所述的实施例范围之中。
实施例1
图1为本实施例提供的一种电池簇荷电状态的估计方法的流程示意图，该电池簇荷电状态的估计方法可以由电池簇荷电状态的估计系统执行，该电池簇荷电状态的估计系统可以通过软件和/或硬件的方式实现，该电池簇荷电状态的估计系统可以为电子设备的部分或全部。其中，本实施例中的电子设备可以为个人计算机(Personal Computer，PC)，例如台式机、一体机、笔 记本电脑、平板电脑等，还可以为手机、可穿戴设备、掌上电脑(Personal Digital Assistant，PDA)等终端设备。下面以电子设备为执行主体介绍本实施例提供的电池簇荷电状态的估计方法。
如图1所示，本实施例提供的电池簇荷电状态的估计方法可以包括以下步骤S1～S4：
步骤S1、获取与电池簇的荷电状态相关的目标数据。
其中，与电池簇的荷电状态相关的目标数据也可以称为影响电池簇荷电状态的数据。为了提高电池簇荷电状态估计的准确性，可以尽可能多地获取目标数据。电池簇可以包括多个电池箱，每个电池箱可以包括多个电芯。
步骤S2、利用安时积分法根据所述目标数据对所述电池簇的荷电状态进行估计，得到第一估计值。
在步骤S2的具体实施中，可以将所述目标数据中电池簇的电流I、额定容量Capacity以及健康状态SOH代入以下公式计算第一估计值SOC Ah：
Figure PCTCN2022112838-appb-000003
步骤S3、将所述目标数据输入荷电状态预测模型对所述电池簇的荷电状态进行估计，得到第二估计值。
其中，所述荷电状态预测模型基于样本数据训练得到。在具体实施中，所述荷电状态预测模型可以采用GBDT(Gradient Boosting Decision Tree，梯度提升树)，GBDT使用的决策树是CART回归树。采用GBDT对电池簇的荷电状态进行估计，具有运行速度快和运行结果稳定的优点，第二估计值的准确度可以得到保证。
在步骤S3的具体实施中，输入所述荷电状态预测模型的目标数据可以包括电池簇的基本信息，例如电池簇的最大单体电压V max、最小单体电压V min、单体平均电压V ave、总电压V total、最高温度T max、最低温度T min、平均温度T ave、电流I、充放电状态Charge_state等。
在步骤S3的具体实施中，输入所述荷电状态预测模型的目标数据还可 以包括电池簇的统计信息，例如电池簇的电压标准差σ v、温度标准差σ T、电压温度协方差σ(x m,x k)等。
其中，
Figure PCTCN2022112838-appb-000004
μ V为电池簇内各单体的电压平均值；
Figure PCTCN2022112838-appb-000005
μ T为电池簇内各温度测点的平均值。
步骤S4、根据所述第一估计值、所述第二估计值以及所述目标数据与所述样本数据之间的距离确定荷电状态的最终估计值。具体地，可以根据所述目标数据与所述样本数据之间的距离确定所述第一估计值和所述第二估计值分别在所述最终估计值中的占比。
在具体实施中，可以基于度量矩阵计算所述目标数据与所述样本数据之间的距离。
本实施方式中，结合安时积分法和荷电状态预测模型共同估计电池簇的荷电状态，具体地，利用安时积分法得到电池簇荷电状态的第一估计值，利用荷电状态预测模型得到电池簇荷电状态的第二估计值，目标数据与训练荷电状态预测模型的样本数据之间的距离可以反映出荷电状态预测模型估计荷电状态的准确性，根据所述距离确定第一估计值和第二估计值分别在最终估计值中的占比，能够有效提高电池簇荷电状态估计的准确性。
在可选的一种实施方式中，步骤S4具体包括以下步骤S41：
步骤S41、对所述第一估计值和所述第二估计值进行加权求和，得到荷电状态的最终估计值。
其中，所述第一估计值的权重和所述第二估计值的权重根据所述目标数据与所述样本数据之间的距离确定。
在一个具体的例子中，根据以下公式计算最终估计值SOC：
SOC＝SOC GBDT+K*(SOC Ah-SOC GBDT)＝K*SOC Ah+(1-K)SOC GBDT。
其中，SOC GBDT为第二估计值，K为第一估计值的权重，1-K为第二估计 值的权重。
在可选的一种实施方式中，如图2所示，上述步骤S41包括以下步骤S411～S413：
步骤S411、判断所述距离是否大于第一预设值，若是，则执行步骤S412，若否，则执行步骤S413。
其中，所述第一预设值可以根据所述样本数据之间的最大距离确定。
步骤S412、设置所述第一估计值的权重大于等于所述第二估计值的权重。
步骤S413、设置所述第一估计值的权重小于所述第二估计值的权重。
本实施方式中，若所述距离大于第一预设值，说明所述目标数据未包含在所述样本数据中，此时，利用安时积分法得到第一估计值在最终估计值中的占比更高。若所述距离小于等于第一预设值，说明所述目标数据包含在所述样本数据中，此时，利用荷电状态预测模型得到的第二估计值在最终估计值中的占比更高。
在可选的一种实施方式中，根据以下公式设置所述第一估计值的权重K：
Figure PCTCN2022112838-appb-000006
其中，
Figure PCTCN2022112838-appb-000007
其中，D为所述目标数据与所述样本数据之间的距离；D 1为所述样本数据之间的最大距离；n为超参数，用于表示K收敛的速度，可以根据实际情况进行调整；所述第二估计值的权重为1-K。
本实施方式中，若D_gain≤0，说明所述目标数据包含在所述样本数据中，此时，设置K＝0，第二估计值即荷电状态预测模型估计的荷电状态在最终估计值中的占比更高。若D_gain＞0，说明所述目标数据未包含在所述样本数据中，且D_gain越大表示距离越远，K越接近于1，此时，第一估计值即利用安时积分法估计的荷电状态在最终估计值中的占比更高。
以下针对上述荷电状态预测模型的训练过程进行详细介绍。
储能电站设有多个电池簇，这些电池簇每天会产生大量的历史数据，可以从历史数据中选取训练荷电状态预测模型的样本数据，以及对应的荷电状态。假设共有N个电池簇的样本数据：
Figure PCTCN2022112838-appb-000008
对应的真实荷电状态为{y 1,y 2...y N}，损失函数为L(y,f(x))，迭代次数为M，构建荷电状态预测模型的强学习器
Figure PCTCN2022112838-appb-000009
具体可以包括以下步骤(1)～(3)：
(1)初始化弱学习器
Figure PCTCN2022112838-appb-000010
其中，c通常取所有样本数据对应真实荷电状态的平均值。
(2)对于迭代轮数m＝1,2,…,M有：
a.对每个样本数据i＝1，2，…，N计算负梯度，即残差：
Figure PCTCN2022112838-appb-000011
b.将上面得到的残差作为样本数据的新的真实荷电状态，并将数据(x i,g mi)(i＝1,2,...N)作为下棵树的训练数据，得到一颗树回归树R mj,j＝1，2...,J。其中，J为回归树的叶子节点的个数。
c.对叶子区域j＝1，2...,J计算最佳拟合值：
Figure PCTCN2022112838-appb-000012
d.更新强学习器
Figure PCTCN2022112838-appb-000013
(3)得到最终学习器：
Figure PCTCN2022112838-appb-000014
为了进一步提高上述荷电状态预测模型对电池簇荷电状态估计的准确性，可以根据获取的目标数据对样本数据进行更新，并利用更新后的样本数据重新训练上述荷电状态预测模型。在可选的一种实施方式中，如图3所示，若所述目标数据与所述样本数据之间的距离大于第二预设值，则将所述目标数据加入所述样本数据中，得到更新后的样本数据，并利用更新后的样本数 据重新训练所述荷电状态预测模型。其中，所述第二预设值根据所述样本数据之间的最大距离确定。在具体实施中，所述第二预设值可以与上述第一预设值相同，也可以大于上述第一预设值。
本实施方式中，更新后的样本数据包括原始的样本数据以及符合条件的目标数据。其中，与所述样本数据之间的距离大于第二预设值的目标数据即为符合条件的目标数据。
在具体实施中，为了避免频繁地训练上述荷电状态预测模型，可以在符合条件的目标数据达到一定数量的情况下，重新构造样本数据，并重新训练荷电状态预测模型。
在可选的一种实施方式中，上述利用更新后的样本数据重新训练所述荷电状态预测模型的步骤具体包括：通过单边梯度采样的方式从更新后的样本数据中提取部分样本数据，并利用所述部分样本数据重新训练所述荷电状态预测模型。本实施方式中，首先通过单边梯度采样的方式提取重新训练荷电状态预测模型所使用的样本数据，然后通过提取样本数据的残差值拟合得到一颗新的树，最后更新之前的荷电状态预测模型，得到最新的强学习器。
在具体实施中，对更新后的样本数据计算其负梯度，得到：
Figure PCTCN2022112838-appb-000015
根据不同样本数据的负梯度绝对值进行降序排列，提取其中的前A个样本数据，并在其余的样本数据中随机选取B个样本数据，得到(A+B)个样本数据。为了使得这(A+B)个样本数据与原始样本数据的分布空间一致，在样本数据B计算残差时乘以一个系数(1-a)/b，其中，a为A占总样本数据的百分比，b为样本数据B占总样本的百分比。
需要说明的是，在更新荷电状态预测模型之后，还需要更新样本数据之间的最大距离D 1。
图4用于示出一种电池簇荷电状态的估计效果示意图。从图3中可以看出，利用安时积分法估计的电池簇荷电状态存在累计误差，与真实的电池簇 荷电状态相差较多，利用本实施例提供的方法估计的电池簇荷电状态与真实的电池簇荷电状态相差较少，准确性更高。
本实施例还提供一种电池簇荷电状态的估计系统，如图5所示，包括数据获取模块40、第一估计模块41、第二估计模块42以及荷电确定模块43。
数据获取模块40用于获取与电池簇的荷电状态相关的目标数据。
第一估计模块41用于利用安时积分法根据所述目标数据对所述电池簇的荷电状态进行估计，得到第一估计值。
第二估计模块42用于将所述目标数据输入荷电状态预测模型对所述电池簇的荷电状态进行估计，得到第二估计值；其中，所述荷电状态预测模型基于样本数据训练得到。
荷电确定模块43用于根据所述第一估计值、所述第二估计值以及所述目标数据与所述样本数据之间的距离确定荷电状态的最终估计值。
在可选的一种实施方式中，所述荷电确定模块具体用于对所述第一估计值和所述第二估计值进行加权求和，得到荷电状态的最终估计值；其中，所述第一估计值的权重和所述第二估计值的权重根据所述目标数据与所述样本数据之间的距离确定。
在可选的一种实施方式中，所述荷电确定模块具体用于判断所述距离是否大于第一预设值；其中，所述第一预设值根据所述样本数据之间的最大距离确定；并在是的情况下设置所述第一估计值的权重大于等于所述第二估计值的权重；以及在否的情况下设置所述第一估计值的权重小于所述第二估计值的权重。
在可选的一种实施方式中，输入所述荷电状态预测模型的目标数据包括以下中的至少一种：所述电池簇的最大单体电压、最小单体电压、单体平均电压、总电压、最高温度、最低温度、平均温度、电流、充放电状态、电压标准差、温度标准差、电压温度协方差。
在可选的一种实施方式中，上述电池簇荷电状态的估计系统还包括模型训练模块，用于在所述目标数据与所述样本数据之间的距离大于第二预设值 的情况下，将所述目标数据加入所述样本数据中，得到更新后的样本数据；其中，所述第二预设值根据所述样本数据之间的最大距离确定；并利用更新后的样本数据重新训练所述荷电状态预测模型。
在可选的一种实施方式中，所述模型训练模块具体用于通过单边梯度采样的方式从更新后的样本数据中提取部分样本数据；并利用所述部分样本数据重新训练所述荷电状态预测模型。
需要说明的是，本实施例中电池簇荷电状态的估计系统具体可以是单独的芯片、芯片模组或电子设备，也可以是集成于电子设备内的芯片或者芯片模组。
关于本实施例中描述的电池簇荷电状态的估计系统包含的各个模块/单元，其可以是软件模块/单元，也可以是硬件模块/单元，或者也可以部分是软件模块/单元，部分是硬件模块/单元。
实施例2
图6为本实施例提供的一种电子设备的结构示意图。所述电子设备包括至少一个处理器以及与所述至少一个处理器通信连接的存储器。其中，所述存储器存储有可被所述至少一个处理器运行的计算机程序，所述计算机程序被所述至少一个处理器执行，以使所述至少一个处理器能够执行实施例1的电池簇荷电状态的估计方法。本实施例提供的电子设备可以为个人计算机，例如台式机、一体机、笔记本电脑、平板电脑等，还可以为手机、可穿戴设备、掌上电脑等终端设备。图6显示的电子设备3仅仅是一个示例，不应对本发明实施例的功能和使用范围带来任何限制。
电子设备3的组件可以包括但不限于：上述至少一个处理器4、上述至少一个存储器5、连接不同系统组件(包括存储器5和处理器4)的总线6。
总线6包括数据总线、地址总线和控制总线。
存储器5可以包括易失性存储器，例如随机存取存储器(RAM)51和/或高速缓存存储器52，还可以进一步包括只读存储器(ROM)53。
存储器5还可以包括具有一组(至少一个)程序模块54的程序/实用工 具55，这样的程序模块54包括但不限于：操作系统、一个或者多个应用程序、其它程序模块以及程序数据，这些示例中的每一个或某种组合中可能包括网络环境的实现。
处理器4通过运行存储在存储器5中的计算机程序，从而执行各种功能应用以及数据处理，例如上述电池簇荷电状态的估计方法。
电子设备3也可以与一个或多个外部设备7(例如键盘、指向设备等)通信。这种通信可以通过输入/输出(I/O)接口8进行。并且，电子设备3还可以通过网络适配器9与一个或者多个网络(例如局域网(LAN)，广域网(WAN)和/或公共网络，例如因特网)通信。如图6所示，网络适配器9通过总线6与电子设备3的其它模块通信。应当明白，尽管图6中未示出，可以结合电子设备3使用其它硬件和/或软件模块，包括但不限于：微代码、设备驱动器、冗余处理器、外部磁盘驱动阵列、RAID(磁盘阵列)系统、磁带驱动器以及数据备份存储系统等。
应当注意，尽管在上文详细描述中提及了电子设备的若干单元/模块或子单元/模块，但是这种划分仅仅是示例性的并非强制性的。实际上，根据本发明的实施方式，上文描述的两个或更多单元/模块的特征和功能可以在一个单元/模块中具体化。反之，上文描述的一个单元/模块的特征和功能可以进一步划分为由多个单元/模块来具体化。
实施例3
本实施例提供一种存储有计算机程序的计算机可读存储介质，所述计算机程序被处理器执行时实现实施例1的电池簇荷电状态的估计方法。
其中，可读存储介质可以采用的更具体可以包括但不限于：便携式盘、硬盘、随机存取存储器、只读存储器、可擦拭可编程只读存储器、光存储器件、磁存储器件或上述的任意合适的组合。
在可能的实施方式中，本发明还可以实现为一种程序产品的形式，其包括程序代码，当所述程序产品在电子设备上运行时，所述程序代码用于使所述电子设备执行实现实施例1的电池簇荷电状态的估计方法。
其中，可以以一种或多种程序设计语言的任意组合来编写用于执行本发明的程序代码，所述程序代码可以完全地在电子设备上执行、部分地在电子设备上执行、作为一个独立的软件包执行、部分在电子设备上部分在远程设备上执行或完全在远程设备上执行。
虽然以上描述了本发明的具体实施方式，但是本领域的技术人员应当理解，这仅是举例说明，本发明的保护范围是由所附权利要求书限定的。本领域的技术人员在不背离本发明的原理和实质的前提下，可以对这些实施方式做出多种变更或修改，但这些变更和修改均落入本发明的保护范围。

Claims (10)

  1. 一种电池簇荷电状态的估计方法，其特征在于，包括以下步骤：
    获取与电池簇的荷电状态相关的目标数据；
    利用安时积分法根据所述目标数据对所述电池簇的荷电状态进行估计，得到第一估计值；
    将所述目标数据输入荷电状态预测模型对所述电池簇的荷电状态进行估计，得到第二估计值；其中，所述荷电状态预测模型基于样本数据训练得到；
    根据所述第一估计值、所述第二估计值以及所述目标数据与所述样本数据之间的距离确定荷电状态的最终估计值。
  2. 如权利要求1所述的电池簇荷电状态的估计方法，其特征在于，所述根据所述第一估计值、所述第二估计值以及所述目标数据与所述样本数据之间的距离确定荷电状态的最终估计值的步骤具体包括：
    对所述第一估计值和所述第二估计值进行加权求和，得到荷电状态的最终估计值；
    其中，所述第一估计值的权重和所述第二估计值的权重根据所述目标数据与所述样本数据之间的距离确定。
  3. 如权利要求2所述的电池簇荷电状态的估计方法，其特征在于，所述对所述第一估计值和所述第二估计值进行加权求和的步骤具体包括：
    判断所述距离是否大于第一预设值；其中，所述第一预设值根据所述样本数据之间的最大距离确定；
    若是，则设置所述第一估计值的权重大于等于所述第二估计值的权重；
    若否，则设置所述第一估计值的权重小于所述第二估计值的权重。
  4. 如权利要求3所述的电池簇荷电状态的估计方法，其特征在于，根据以下公式设置所述第一估计值的权重K：
    Figure PCTCN2022112838-appb-100001
    其中，
    Figure PCTCN2022112838-appb-100002
    其中，D为所述目标数据与所述样本数据之间的距离，D 1为所述样本数据之间的最大距离，n为超参数，用于表示K收敛的速度，所述第二估计值的权重为1-K。
  5. 如权利要求1-4中任一项所述的电池簇荷电状态的估计方法，其特征在于，输入所述荷电状态预测模型的目标数据包括以下中的至少一种：所述电池簇的最大单体电压、最小单体电压、单体平均电压、总电压、最高温度、最低温度、平均温度、电流、充放电状态、电压标准差、温度标准差、电压温度协方差。
  6. 如权利要求1所述的电池簇荷电状态的估计方法，其特征在于，所述电池簇荷电状态的估计方法还包括以下步骤：
    若所述目标数据与所述样本数据之间的距离大于第二预设值，则将所述目标数据加入所述样本数据中，得到更新后的样本数据；其中，所述第二预设值根据所述样本数据之间的最大距离确定；
    利用更新后的样本数据重新训练所述荷电状态预测模型。
  7. 如权利要求6所述的电池簇荷电状态的估计方法，其特征在于，所述利用更新后的样本数据重新训练所述荷电状态预测模型的步骤具体包括：
    通过单边梯度采样的方式从更新后的样本数据中提取部分样本数据；
    利用所述部分样本数据重新训练所述荷电状态预测模型。
  8. 一种电池簇荷电状态的估计系统，其特征在于，包括：
    数据获取模块，用于获取与电池簇的荷电状态相关的目标数据；
    第一估计模块，用于利用安时积分法根据所述目标数据对所述电池簇的荷电状态进行估计，得到第一估计值；
    第二估计模块，用于将所述目标数据输入荷电状态预测模型对所述电池簇的荷电状态进行估计，得到第二估计值；其中，所述荷电状态预测模型基于样本数据训练得到；
    荷电确定模块，用于根据所述第一估计值、所述第二估计值以及所述目标数据与所述样本数据之间的距离确定荷电状态的最终估计值。
  9. 一种电子设备，包括存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现如权利要求1-7中任一项所述的电池簇荷电状态的估计方法。
  10. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1-7中任一项所述的电池簇荷电状态的估计方法。
PCT/CN2022/112838 2022-03-31 2022-08-16 电池簇荷电状态的估计方法及系统、电子设备及存储介质 WO2023184824A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210343744.8A CN114705990B (zh) 2022-03-31 2022-03-31 电池簇荷电状态的估计方法及系统、电子设备及存储介质
CN202210343744.8 2022-03-31

Publications (1)

Publication Number Publication Date
WO2023184824A1 true WO2023184824A1 (zh) 2023-10-05

Family

ID=82171771

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/112838 WO2023184824A1 (zh) 2022-03-31 2022-08-16 电池簇荷电状态的估计方法及系统、电子设备及存储介质

Country Status (2)

Country Link
CN (1) CN114705990B (zh)
WO (1) WO2023184824A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117060553A (zh) * 2023-10-13 2023-11-14 快电动力(北京)新能源科技有限公司 储能系统的电池管理方法、装置、系统和部件
CN117148168A (zh) * 2023-10-27 2023-12-01 宁德时代新能源科技股份有限公司 训练模型的方法、预测电池容量的方法、装置及介质

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114705990B (zh) * 2022-03-31 2023-10-20 上海玫克生储能科技有限公司 电池簇荷电状态的估计方法及系统、电子设备及存储介质
CN117590260B (zh) * 2024-01-18 2024-04-16 武汉船用电力推进装置研究所(中国船舶集团有限公司第七一二研究所) 船用锂离子动力电池荷电状态估计方法、装置及电子设备

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012063251A (ja) * 2010-09-16 2012-03-29 Calsonic Kansei Corp フィルタによるパラメータ推定装置
CN102930173A (zh) * 2012-11-16 2013-02-13 重庆长安汽车股份有限公司 一种锂离子电池荷电状态在线估算方法
CN104678316A (zh) * 2015-02-28 2015-06-03 北京交通大学 锂离子电池荷电状态估算方法和装置
CN105021996A (zh) * 2015-08-04 2015-11-04 深圳拓普科新能源科技有限公司 储能电站bms的电池soh估算方法
CN105425153A (zh) * 2015-11-02 2016-03-23 北京理工大学 一种估计电动车辆的动力电池的荷电状态的方法
CN105425154A (zh) * 2015-11-02 2016-03-23 北京理工大学 一种估计电动汽车的动力电池组的荷电状态的方法
CN114705990A (zh) * 2022-03-31 2022-07-05 上海玫克生储能科技有限公司 电池簇荷电状态的估计方法及系统、电子设备及存储介质

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102009045526A1 (de) * 2009-10-09 2011-04-14 SB LiMotive Company Ltd., Suwon Verfahren zur Initialisierung und des Betriebs eines Batteriemanagementsystems
CN106405421A (zh) * 2015-07-27 2017-02-15 中兴通讯股份有限公司 电池荷电状态soc值的估算方法及装置
CN108872866B (zh) * 2018-06-04 2021-02-05 桂林电子科技大学 一种锂离子电池荷电状态动态评估与长效预测融合方法
CN110286324B (zh) * 2019-07-18 2021-07-09 北京碧水润城水务咨询有限公司 一种电池荷电状态估算方法及电池健康状态估算方法
CN113131012B (zh) * 2019-12-30 2023-04-28 杭州海康机器人股份有限公司 Soc估算精度的确定方法、装置及存储介质
CN113466723B (zh) * 2020-03-31 2022-09-09 比亚迪股份有限公司 确定电池荷电状态的方法及装置，电池管理系统
KR102572652B1 (ko) * 2020-09-11 2023-08-31 삼성에스디아이 주식회사 배터리의 충전상태를 추정하는 방법
CN112305441B (zh) * 2020-10-14 2023-06-16 北方工业大学 一种集成式聚类下的动力电池健康状态评估方法
CN112630661B (zh) * 2020-12-28 2022-12-09 广州橙行智动汽车科技有限公司 一种电池荷电状态soc估算方法和装置
CN114184962B (zh) * 2021-10-19 2022-12-13 北京理工大学 一种多算法融合的锂离子电池soc和soh联合估算方法
CN114035072B (zh) * 2021-11-11 2023-10-03 重庆大学 一种基于云边协同的电池组多状态联合估计方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012063251A (ja) * 2010-09-16 2012-03-29 Calsonic Kansei Corp フィルタによるパラメータ推定装置
CN102930173A (zh) * 2012-11-16 2013-02-13 重庆长安汽车股份有限公司 一种锂离子电池荷电状态在线估算方法
CN104678316A (zh) * 2015-02-28 2015-06-03 北京交通大学 锂离子电池荷电状态估算方法和装置
CN105021996A (zh) * 2015-08-04 2015-11-04 深圳拓普科新能源科技有限公司 储能电站bms的电池soh估算方法
CN105425153A (zh) * 2015-11-02 2016-03-23 北京理工大学 一种估计电动车辆的动力电池的荷电状态的方法
CN105425154A (zh) * 2015-11-02 2016-03-23 北京理工大学 一种估计电动汽车的动力电池组的荷电状态的方法
CN114705990A (zh) * 2022-03-31 2022-07-05 上海玫克生储能科技有限公司 电池簇荷电状态的估计方法及系统、电子设备及存储介质

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117060553A (zh) * 2023-10-13 2023-11-14 快电动力(北京)新能源科技有限公司 储能系统的电池管理方法、装置、系统和部件
CN117060553B (zh) * 2023-10-13 2024-01-02 快电动力(北京)新能源科技有限公司 储能系统的电池管理方法、装置、系统和部件
CN117148168A (zh) * 2023-10-27 2023-12-01 宁德时代新能源科技股份有限公司 训练模型的方法、预测电池容量的方法、装置及介质
CN117148168B (zh) * 2023-10-27 2024-03-29 宁德时代新能源科技股份有限公司 训练模型的方法、预测电池容量的方法、装置及介质

Also Published As

Publication number Publication date
CN114705990B (zh) 2023-10-20
CN114705990A (zh) 2022-07-05

Similar Documents

Publication Publication Date Title
WO2023184824A1 (zh) 电池簇荷电状态的估计方法及系统、电子设备及存储介质
Shen et al. A novel online method for predicting the remaining useful life of lithium-ion batteries considering random variable discharge current
CN109196366B (zh) 使用高斯过程回归估计电池充电状态的方法和系统
Zhao et al. Elman neural network using ant colony optimization algorithm for estimating of state of charge of lithium-ion battery
JP7199568B2 (ja) バッテリ診断システム、バッテリ診断方法、及び記憶媒体
CN110488194B (zh) 一种基于电化学阻抗模型的锂电池soc估算方法及其系统
CN204269785U (zh) 用于计算电池电荷状态的系统
WO2021208079A1 (zh) 获取动力电池寿命数据的方法、装置、计算机设备及介质
CN108089135B (zh) 一种基于极限学习模型的电池状态预测系统及其实现方法
CN113219343A (zh) 基于弹性网的锂电池健康状态预测方法、系统、设备及介质
CN114371409B (zh) 电池状态预测模型的训练方法、电池状态预测方法及装置
Yu et al. Fractional-order modeling of lithium-ion batteries using additive noise assisted modeling and correlative information criterion
CN116087787A (zh) 一种基于主成分分析法的电池故障判断方法和系统
CN117110891A (zh) 锂离子电池荷电状态估计值的计算方法和计算装置
Xing et al. Remaining useful life prediction of–Lithium batteries based on principal component analysis and improved Gaussian process regression
CN114609530A (zh) 修正电池荷电状态的方法、装置、设备及介质
CN116381494A (zh) 电池最大可用容量预测方法、装置、电子设备及存储介质
CN114371408B (zh) 电池荷电状态的估算方法、充电曲线的提取方法及装置
CN110850315A (zh) 一种电池荷电状态的估算方法及装置
Chen et al. State of health estimation for lithium-ion battery based on particle swarm optimization algorithm and extreme learning machine
CN116299116A (zh) 一种储能系统ocv-soc校准方法及电子设备
CN115575835A (zh) 一种锂离子电池峰值功率预估方法
CN113093014B (zh) 一种基于阻抗参数的soh与soc的在线协同估计方法及系统
CN113933725A (zh) 一种基于数据驱动确定动力电池荷电状态的方法
CN111679196A (zh) 一种pngv模型参数的在线辨识方法、装置、计算机设备

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: 22934644

Country of ref document: EP

Kind code of ref document: A1