Battery charging and discharging feature extraction method based on the best u-shapelets

摘要

With the advancement of science and technology,batteries have become an indispensable item in our daily life. At the same time, the study of the charge-discharging curve of the battery plays an important role. The problem of battery charging and discharging curve can be regarded as a time series data mining problem. We utilize the unsupervised shape u-shapelets for time series data mining, which is a newly emerging tiny local feature that has been widely used in many fields, e.g., battery grouping. Experimental results show the practicability and effectiveness of the battery charge/discharge feature extraction method using the best u-shapelets, the ability of the local characteristics of u-shapelets to provide more insights for the data, and the sensitivity to irrelevant data in the charging and discharging curve of the battery is reduced. Extracting local feature u-shapelets from battery charging and discharging curves is helpful for battery grouping.