An artificial intelligence data-driven approach for optimizing pump system in sewage water treatment plant, its objective is to reduce the energy consumption and optimize pump configuration and pump speed settings. Under different operating scenarios (i.e. at the different flow rate, pump speed, wet well level), the energy performance of the pump can be very much different. There are many reasons, e.g. the configuration of the discharge pipework of each pump is not identical. Data-mining algorithm, multilayer perception neural network is used to build the pumping energy model. Significant energy saving can be found in the computer simulation by using the historical data.
- Through the data mining technique to find out the energy saving opportunity from historical data.
- Optimize the pump configuration and pump speed setting automatically to achieve energy saving.
- Preventive maintenance