阅读说明：本技术 一种电池soc估算方法 (Battery SOC estimation method ) 是由 程海峰 方伟峰 马瑞军 于 2018-07-18 设计创作，主要内容包括：本发明提供一种电池SOC估算方法。包括如下步骤：A、通过电池管理系获取电池当前的SOC值SOC<Sub>BMS</Sub>,根据SOC<Sub>BMS</Sub>确定递推最小二乘法的退出条件；B、利用安时积分算法实时估算电池的SOC值；C、利用递推最小二乘法估算电池在t<Sub>A</Sub>时刻的SOC值并在符合退出条件时停止递推,得到t<Sub>A</Sub>时刻的SOC值SOC<Sub>A</Sub>；D、利用安时积分记录t<Sub>A</Sub>时刻后的SOC变化量,当电池的SOC变化量大于预设值时,再次利用递推最小二乘法估算电池此时t<Sub>B</Sub>时刻的SOC值并在符合退出条件时停止递推,得到t<Sub>B</Sub>时刻的SOC值SOC<Sub>B</Sub>；E、判断SOC<Sub>B</Sub>的有效性。利用本发明提供的SOC估算方法能有效校准安时积分累积误差,提高SOC估算精度。(The invention provides a battery SOC estimation method. The method comprises the following steps: A. obtaining the current SOC value SOC of the battery through the battery management system BMS According to SOC BMS Determining an exit condition of a recursive least square method; B. estimating the SOC value of the battery in real time by using an ampere-hour integration algorithm; C. estimation of battery at t using recursive least squares A The SOC value at the moment stops recursion when meeting the exit condition to obtain t A SOC value SOC of time A (ii) a D. Recording t using ampere-hour integration A When the SOC variation of the battery is larger than the preset value, estimating the t of the battery at the moment by using the recursive least square method again B The SOC value at the moment stops recursion when meeting the exit condition to obtain t B SOC value SOC of time B (ii) a E. Judging SOC B The effectiveness of (c). The SOC estimation method provided by the invention can effectively calibrate the ampere-hour integral accumulated error and improve the SOC estimation precision.)

1. A battery SOC estimation method, characterized by comprising the steps of:

A. acquiring the current SOC value SOC of the battery through a battery management system_BMSAccording to SOC_BMSDetermining an exit condition of a recursive least square method;

B. estimating the SOC value of the battery in real time by using an ampere-hour integration algorithm;

C. estimation of battery at t using recursive least squares_AThe SOC value at the moment stops recursion when meeting the exit condition to obtain the battery at t_ASOC value SOC of time_A；

D. Recording t using an ampere-hour integral algorithm_AWhen the SOC variation of the battery is larger than the preset value, estimating the current t of the battery by using the recursive least square method_BThe SOC value at the moment stops recursion when meeting the exit condition to obtain the battery at t_BSOC value SOC of time_B；

E. SOC (state of charge) judgment by using ampere-hour integration algorithm_BIf valid, the SOC is determined_BOutputting the value; if not, the t estimated by the ampere-hour integral algorithm is used_BSOC value SOC of time_AhAnd (6) outputting.

2. The battery SOC estimation method according to claim 1, wherein the exit condition is: in the recursive process of the recursive least square method, the SOC accumulated variation recorded by the ampere-hour integration algorithm is more than or equal to delta and the recursive times N>Lo, or number of recursions N>Hi, where δ ═ Δ Err ═ dSOC_BMS/dOCV_BMSLo and Hi are preset thresholds for recurrence times, Lo<Hi; and delta Err is the error precision of the battery management system to the voltage sampling of the single battery.

3. The battery SOC estimation method according to claim 2, wherein, when the sampling period of the cell voltage is 0.5s to 1.5s, the value range of Lo is [60,120], and the value range of Hi is [300,360 ].

4. The battery SOC estimation method according to claim 1, wherein the estimation of the battery at t using the recursive least squares method is performed_AThe SOC value at the time specifically includes: identifying t by recursive least squares_AObtaining the parameters of the equivalent circuit model of the battery at the moment until the exit condition is met, and obtaining the battery at t_AOpen circuit voltage value OCV of moment_AAccording to OCV_AInquiring an OCV-SOC table to obtain the SOC_A。

5. The battery SOC estimation method according to claim 1, wherein the estimation of the battery at t using the recursive least squares method is performed_BThe SOC value at the time specifically includes: identifying t by recursive least squares_BObtaining the parameters of the equivalent circuit model of the battery at the moment until the exit condition is met, and obtaining the battery at t_BOpen circuit voltage value OCV of moment_BAccording to OCV_BInquiring an OCV-SOC table to obtain the SOC_B。

6. The battery SOC estimation method according to claim 4 or 5, wherein the equivalent circuit model employs a first order equivalent circuit model or a second order equivalent circuit model.

7. The battery SOC estimation method according to claim 1, wherein the preset value ranges from [ 5% to 80% ].

8. The battery SOC estimation method according to claim 1, wherein the determination SOC_BThe effectiveness of (a) includes: according to SOC_B-SOC_ARelative relationship to Δ SOC and at said SOC_BOCV value of (d) and change in SOC value of (d) in the vehicle_B/dOCV_BJudging SOC_BEffectiveness of, Δ SOC represents t recorded by ampere-hour integration algorithm_ATo t_BBy the amount of change in the SOC of the battery.

9. The battery SOC estimation method according to claim 8, wherein the determination SOC_BThe effectiveness of (a) specifically includes: when SOC is reached_BConforming to dSOC_B/dOCV_B< η and abs | -SOC_B-SOC_AWhen-delta SOC | is less than or equal to epsilon, then the SOC is judged_BThe method is effective; when dSOC_B/dOCV_BAbs ≧ η_B-SOC_AWhen-delta SOC | is less than or equal to epsilon, then the SOC is judged_BAnd η and epsilon are both preset thresholds.

10. The battery SOC estimation method of claim 9, wherein the value of epsilon ranges from [ 0.5%, 1.5% ], and the value of η ranges from [ 0.05%/mV, 0.15%/mV ].

Technical Field

The invention relates to the field of battery energy management, in particular to a battery SOC estimation method.

Background

State of charge (SOC) estimation of a battery is one of key technologies in a battery management system, and is related to not only estimation of the driving mileage of an automobile but also safety of a power battery. The accurate estimation of the SOC of the battery is beneficial to accurately predicting the residual capacity of the battery, so that an effective battery management strategy is determined, the damage to the battery caused by overcharge or discharge is avoided, and the service life of the battery is prolonged

At present, a common estimation method obtains the SOC by studying characteristics of an Open Circuit Voltage (OCV) of the battery, a change rule of the Voltage during constant current discharge, an internal resistance of the battery, and the like, thereby generating some basic calculation methods of the SOC, such as a constant current Voltage method, an Open Circuit Voltage method, an ampere-hour integration method, an internal resistance method, a specific gravity method, and the like. The ampere-hour integration method is widely applied to SOC estimation due to its advantages of simplicity, easy implementation, reliability, etc., but it generates a large accumulated error after long-term operation. The recursive least square method can effectively correct the charge state estimated by the ampere-hour integral, but the recursive least square method has two problems in the application process: 1. the recursive least square method is easy to generate data saturation; 2. forgetting factors in the recursive least square method with the forgetting factors cannot be optimized in real time along with working conditions, and therefore estimation error fluctuation is large.

Therefore, it is an urgent technical problem to be solved by those skilled in the art to design an SOC estimation method that is simple and easy to implement, has high estimation accuracy, and can effectively reduce the ampere-hour integral accumulated error.

Disclosure of Invention

One of the purposes of the invention is to provide a battery SOC estimation method, which optimizes the exit condition of a recursive least square method in real time according to working conditions, avoids data saturation, reduces the accumulated error of an ampere-hour integral algorithm, and improves the estimation precision of the battery SOC.

In order to achieve the purpose, the technical scheme of the invention is as follows:

the invention provides a battery SOC estimation method, which comprises the following steps:

A. acquiring the current SOC value SOC of the battery through a battery management system_BMSAccording to SOC_BMSDetermining an exit condition of a recursive least square method; the exit condition of the recursive least square method is set by utilizing the real-time SOC value provided by the battery management system, so that the recursive least square method is realized to change the recursive times in real time along with the working condition, and the condition of recursive data saturation is effectively avoided.

B. And estimating the SOC value of the battery in real time by using an ampere-hour integration algorithm, wherein the ampere-hour integration algorithm calculates the SOC value of the battery in real time from the beginning of an estimation program until the end of the program. The execution sequence of the step A and the step B is not divided into a sequence.

C. Estimation of battery at t using recursive least squares_AThe SOC value at the moment stops recursion when meeting the exit condition to obtain the battery at t_ASOC value SOC of time_A；

D. Recording t using an ampere-hour integral algorithm_AThe variation of the SOC of the battery after the moment, when the variation of the SOC of the battery is larger than a preset value, the recursive least square method is used again to estimate the t of the battery at the moment_BThe SOC value at the moment stops recursion when meeting the exit condition to obtain the battery at t_BSOC value SOC of time_B；

E. SOC (state of charge) judgment by using ampere-hour integration algorithm_BIf valid, the SOC is determined_BOutputting the value; if not, the t estimated by the ampere-hour integral algorithm is used_BSOC value SOC of time_AhAnd (6) outputting.

After the ampere-hour integral algorithm runs for a long time, a large SOC estimation error is easy to generate, the estimation error can be effectively reduced by using the recursive least square method to carry out SOC estimation, meanwhile, the accuracy of SOC variation estimation in a short time interval of the recursive least square method is verified by using the ampere-hour integral algorithm, and the real-time SOC estimation precision is effectively improved by combining different estimation advantages of the two algorithms.

The exit conditions are as follows: in the recursion process of the recursion least square method, the SOC accumulated variation recorded by the ampere-hour integration algorithm is more than or equal to delta and the recursion times N>Lo, or number of recursions N>Hi, where δ ═ Δ Err ═ dSOC_BMS/dOCV_BMSLo and Hi are preset thresholds for recurrence times, Lo<Hi; and delta Err is the error accuracy of the battery management system to the voltage sampling of the single battery. The recursive least square method is effective only when the parameters to be calibrated of the equivalent circuit model are unchanged, and the exit condition is set, so that the recursive least square method can be effectively guaranteed to exit when the voltage of the single battery is enough to identify whether the parameters to be calibrated in the equivalent circuit model are changed, the saturation of data is avoided, and the accuracy of the algorithm is improved.

When the sampling period of the voltage of the single battery is 0.5-1.5s, the value range of Lo is [60,120], and the value range of Hi is [300,360 ].

Estimating the battery at t by using a recursive least square method_AThe SOC value at the time specifically includes: identifying t by recursive least squares_AObtaining the parameters of the equivalent circuit model of the battery at the moment until the exit condition is met, and obtaining the battery at t_AOpen circuit voltage value OCV of moment_AAccording to OCV_AInquiring an OCV-SOC table to obtain the SOC_A。

Estimating the battery at t by using a recursive least square method_BThe SOC value at the time specifically includes: identifying t by recursive least squares_BObtaining the parameters of the equivalent circuit model of the battery at the moment until the exit condition is met, and obtaining the battery at t_BOpen circuit voltage value OCV of moment_BAccording to OCV_BInquiring an OCV-SOC table to obtain the SOC_B。

The equivalent circuit model adopts a first-order equivalent circuit model or a second-order equivalent circuit model.

The range of the preset value is [ 5% -80% ]. In the set interval, the estimation of the ampere-hour integral algorithm on the SOC variation has higher accuracy, and the validity of the least square method verified by the ampere-hour integral algorithm is further ensured.

The determination SOC_BThe effectiveness of (a) includes: according to SOC_B-SOC_ARelative relationship to Δ SOC and at said SOC_BOCV value of (d) and change in SOC value of (d) in the vehicle_B/dOCV_BJudging SOC_BEffectiveness of, Δ SOC represents t recorded by ampere-hour integration algorithm_ATo t_BBy the amount of change in the SOC of the battery. The ampere-hour integration algorithm has higher accuracy on estimation of the SOC variation in a short time, and can effectively verify the accuracy of the SOC estimation by the recursive least square method.

The determination SOC_BThe effectiveness of (a) specifically includes: when SOC is reached_BConforming to dSOC_B/dOCV_B< η and abs | -SOC_B-SOC_AWhen-delta SOC | is less than or equal to epsilon, then the SOC is judged_BThe method is effective; when dSOC_B/dOCV_BAbs ≧ η_B-SOC_AWhen-delta SOC | is less than or equal to epsilon, then the SOC is judged_BInvalid, η and epsilon are both preset thresholds, will SOC_BThe correlation of the SOC value with the change in the OCV value as the validity determination condition can further ensure that the SOC value with higher accuracy and stability as t_BAnd outputting the optimal battery SOC value at the moment.

The value range of the epsilon is (0.5 percent and 1.5 percent)]The value range of the η is [ 0.05%/mV, 0.15%/mV]. The value range can further ensure the SOC of the final output_BThe values have high stability and accuracy.

The invention provides a battery SOC estimation method. Compared with the prior art, the exit condition of the recursive least square method is set by utilizing the real-time SOC value provided by the battery management system, so that the recursive least square method is realized to change the recursive times in real time along with the working condition, and the condition of recursive data saturation is effectively avoided; the long-time estimation error of the ampere-hour integral algorithm is corrected by using the recursive least square method, the accuracy of SOC variation estimation in a short time by using the ampere-hour integral algorithm is verified by using the recursive least square method, and the SOC real-time estimation precision is effectively improved by combining different calculation advantages of the two algorithms.

Drawings

FIG. 1 is a flowchart illustrating a method for estimating SOC of a battery according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a normal temperature DST operating condition in the first embodiment of the invention;

fig. 3 is a comparison graph of the SOC estimation value obtained by the method provided by the present invention under the normal temperature DST condition according to the first embodiment of the present invention and the SOC estimation value obtained by the recursive least square method based on the forgetting factor;

fig. 4 is a comparison graph of the SOC estimation error generated by the method provided by the present invention under the normal temperature DST condition according to the first embodiment of the present invention and the SOC estimation error generated by the recursive least square method based on the forgetting factor;

fig. 5 is a comparison graph of the SOC estimation error generated by the method provided by the present invention and the SOC estimation error generated by the recursive least square method based on the forgetting factor under the normal temperature DST condition in the second embodiment of the present invention.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. It is to be noted that the drawings are in simplified form and are not to precise scale, which is provided for the purpose of facilitating and distinctly claiming the embodiments of the present invention.