Method for predicting retired battery capacity based on XGBoost algorithm

• Main Authors: BAO JUNXIN, LUO GANG, MA ZHUANG
• Format: Patent
• Language: Chinese; English
• Published: 15.08.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; MEASURING; MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES; PHYSICS; TESTING

The invention relates to a method for predicting the capacity of a decommissioned battery based on an XGBoost algorithm, and the method comprises the following steps: S1, collecting the real-time current, voltage and capacity data of a plurality of decommissioned battery cells in the charging and discharging process, carrying out the feature construction, obtaining a feature data set after screening, and constructing a training set and a verification set; s2, constructing a battery capacity prediction model based on an improved XGBoost algorithm, and training the constructed battery capacity prediction model according to the training set and the verification set to obtain a trained model; and S3, for a to-be-predicted retired battery cell, predicting the battery capacity of the retired battery cell by using the trained model. Characteristic extraction is carried out by collecting part of charging data, the capacity grading time is greatly shortened through model training, and the average error of capacity pre