Knowledge-based Bi-correction model for achieving effective lag-free characteristic on daily urban water demand forecasting

Shunyu Wu, Jingcheng Wang, Haotian Xu, Shangwei Zhao, Jiahui Xu

Abstract: Highlights•A bi-correction model is proposed for daily urban water demand forecasting.•Priori and posteriori knowledge are utilized to attenuate lag forecasting problem.•We propose the novel lag penalty attention to extract effective priori knowledge.•We modify the weighted Markov chain to make it suitable for application here.•The proposed model performs superior on lag-free forecasting with baseline methods.