SOH prediction for Lithium-Ion batteries by using historical state and future load information with an AM-seq2seq model

•A new multisource seq2seq model is developed for SOH prediction.•Information of future load is involved as additional input in SOH prediction.•Accurate long-term SOH predictions of batteries under varying loads are achieved.•The proposed model shows good robustness against different beginnings of prediction.•The proposed model shows good robustness against lengths of both two inputs.