Application of remote monitoring and automatic control system using neural network for small wastewater treatment plants in Korea

Abstract

For this study, an automatic control system has been developed by using a neural network and internet-based remote monitoring system for efficient operation of plants that have a serious variance of influent loading and have difficulties in appropriate maintenance, just like small wastewater treatment plants in Korea. In the control algorithm, ORP was used as the main sensor for control. At the point where the ORP value was judged to reach the "nitrate knee" of denitrification and phosphorus release, ORP indicated the state of lower saturation read by the neural network and then changed the operating condition from the reduction state to the oxidation state. For example, if ORP indicates the state of higher saturation at the point of "nitrogen breakpoint" or "ammonia valley" of nitrification, the neural network reads it and cuts off the oxygen supply and mixing. The dORP data have been used as one of the main input for the neural network. After the operation of lab-scale cyclic aeration process using an automatic control system, it has been found that regardless of loading variance, more than 95% of organic matters and more than 60% of nitrogen and phosphorus have been removed. Assuming that an internet-connected computer and a basic web browser are available, this study has developed a remote monitoring system that can monitor the operating status of small plants or any troubles with them.