Short-term prediction of dissolved oxygen and water temperature using deep learning with dual proportional-integral-derivative error corrector in pond culture

Xinhui Zhou, Yinfeng Hao, Yang Liu, Lanxue Dang, Baojun Qiao, Xianyu Zuo

Abstract: Highlights•The combination of PID control principle and deep learning is used for time series prediction.•The trapezoidal error integration method is used to replace the integral term of traditional PID controllers.•The PID error corrector greatly improves prediction accuracy without increasing computational complexity.•The proposed method can achieve accurate prediction of both dissolved oxygen and water temperature simultaneously.