Design of intelligent control system for printing and dyeing wastewater treatment under internet of things and deep learning
Tong Xu,Changrong Zhao,Guancheng Song,Zhaoyang You
DOI: https://doi.org/10.1007/s11227-022-04524-1
2022-10-16
Abstract:The study aims to improve the utilization rate of water resources and meet the new national requirements for wastewater discharge indexes. The study proposes an intelligent control system for printing and dyeing wastewater treatment based on the internet of things (IoT) and deep learning (DL). Specifically, IoT is used to accurately monitor the real-time data of sewage treatment equipment and quality, and a DL structure with multi-layer nonlinear mapping is constructed. The wastewater in a printing and dyeing factory is taken as the research object, and it is treated by the printing and dyeing wastewater treatment system with "iron–carbon micro electrolysis, biological contact oxidation (BCO), and advanced oxidation process (AOP)" as the main processes. IoT and DL are combined to establish an intelligent wastewater control system for the factory. The sewage treatment system is optimized using the back propagation neural network (BPNN) in DL. And then, the intelligent operation module based on DL is designed and implemented to manage the sewage treatment system. The results show that the removal rates of chemical oxygen demand (COD) and chromaticity by iron–carbon micro electrolysis process are 70.19% and 91.32%; those of COD and ammonia nitrogen by BCO are more than 80%; and those of COD and ammonia nitrogen by AOP are 62.2% and 25.6%. This shows that through DL of the collected data and analysis of the wastewater treatment law, predicting and controlling wastewater discharge can improve the accuracy rate and efficiency of wastewater treatment and provide technical support for wastewater treatment and control.
computer science, theory & methods,engineering, electrical & electronic, hardware & architecture