Intelligent Collaborative Optimal Scheduling for Water Intake-Supply Pump Groups in Drinking Water Treatment Plants

Up to 80% of the electrical energy is consumed by pump groups in water purification plants. Optimizing the scheduling of water intake-supply pump groups is crucial for saving electrical energy and reducing carbon dioxide emissions while ensuring water supply and security requirements are met. Herein, an intelligent collaborative optimal scheduling method is proposed for the water intake pump groups, clean-water reservoirs, and water supply pump groups. A long short-term memory (LSTM) network is applied to predict the flow of the water supply pump group, and a data-driven approach is used to plan the flow of the water intake pump group and model the working characteristics of working pump configurations. Furthermore, based on the dynamic programming (DP) algorithm, the optimal scheduling scheme of the water intake-supply pump groups as well as the working pump configuration at each moment can be obtained. The proposed approach dynamically updates data on an hourly basis to enhance the precision of collaborative optimal scheduling outcomes. The experimental results of long-term operation showed that pump efficiency was increased by 10.68% and 10.70% of electric energy was effectively saved as compared to the results of previous manual scheduling. This study provides a solution for energy conservation and efficiency enhancement of multiple energy-consuming equipment and carbon emission reduction.