收录:
摘要:
This paper proposes an optimal control solution for regulating the dissolved oxygen (DO) level in wastewater treatment processes (WWTPs). Our method integrates the Echo State Network (ESN) with online Adaptive Dynamic Programming (ADP) to develop an ESN-ADP controller that adeptly handles the dynamic and nonlinear properties of WWTPs. Compared to the traditional Backpropagation (BP) neural network, which may not be optimal for handling the nonlinear and time-varying dynamics of WWTPs, ESN provides better control performance. By utilizing an online learning approach, our algorithm guarantees that the ESN-ADP controller is both adaptive and convergent in the face of changing conditions. Results of experimental tests show that the proposed ESN-ADP controller outperforms other existing control methods in terms of regulatory performance. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
关键词:
通讯作者信息:
电子邮件地址:
来源 :
ISSN: 1865-0929
年份: 2023
卷: 1869 CCIS
页码: 47-61
语种: 英文
归属院系: