An Equilibrium Method Based on Outlier Detection

Abstract

This patent invented an equalization method based on abnormal point detection. The method includes data processing on the sample data set to obtain the relationship function of the parameters with respect to the distribution of data objects; collect the current moment parameters such as the voltage, SOC, and temperature of each cell in the battery pack. As the data set to be detected, the two parameters of the number of neighbors and the threshold in the field are adaptively obtained through the relationship function; abnormal point detection is performed on the battery pack, and the abnormal cell that needs to be balanced is identified, and the abnormal cell is balanced operate. The present invention comprehensively considers multiple factors such as battery voltage, SOC, temperature, etc., does not need to manually set parameters, and can adaptively obtain different parameters according to the distribution of different data objects, and perform accurate equalization detection on the battery pack, effectively improving the equalization efficiency. Shorten the equalization time, increase the actual available capacity of the battery pack, and prolong the service life.

Description

technical field [0001] The invention belongs to the technical field of battery management, in particular to an equalization method based on abnormal point detection. Background technique [0002] With the gradual consumption of non-renewable energy sources, the development of new energy sources is crucial. Transportation is the main aspect of energy consumption, and it is an inevitable trend for electric vehicles to become the main means of transportation of the new generation. Lithium battery is the main source of power for electric vehicles, and its safety, economy and power are decisive factors for the future development of electric vehicles. Due to the inconsistency of its own internal resistance, capacity, self-discharge rate, different use environments, and differences in battery aging speed, when lithium batteries are used in series in groups, there are often inconsistencies between battery cells, and the inconsistency of battery packs Sexual issues often affect the...

AI Technical Summary

Problems solved by technology

For the voltage-based equalization algorithm, because the voltage consistency after equalization cannot truly reflect the consistency of the battery pack capacity state, there are still some differences in the consistency of the battery pack after the equalization is completed; the SOC-based equalization algorithm needs to use accurate SOC Although the current research on SOC estimation has been relatively in-depth, the prediction accuracy of each method is positively correlated with the amount of calculation. Different SOC prediction methods will have different effects on the effect of the balancing strategy; the balancing algorithm based on capacity also needs to be accurate. The SOC and capacity are based on the premise, but there are still some difficulties in estimating the capacity online

[0004] In recent years, some scholars have used the distance-based outlier detection algorithm for battery balancing detection, and achieved good results. However, the traditional distance-based outlier detection algorithm needs to set at least two parameters: the number of neighbors in the field and the threshold. The setting of the two parameters will directly affect the detection results of the algorithm. If the threshold is set too large, it will easily cause missed detection, and if the threshold is set too small, it will easily cause misjudgment.

Images