Li-ion battery capacity prediction based on artificial intelligence for production line

A capacity prediction method is proposed for a production line to reduce the battery production cost, which can reduce the capacity measurement time by half. The artificial intelligence algorithm predicts the capacity based on the features extracted from the partial charge-discharge data. The neural network performs best among common algorithms due to its nonlinear fitting ability. The prediction error of the testing set is 0.19%. In addition, a multi-model parallel prediction method is proposed, which can effectively solve the problem of maximum error.