Keywords

waste water treatment plants, decision support systems, energy efficiency

Start Date

26-6-2018 10:40 AM

End Date

26-6-2018 12:00 PM

Abstract

The availability of real-time measurements in Wastewater Treatment Plants (WWTPs) can produce environmental and economic benefits. Since a WWTP can produce up to 300k records per day, computational analytics support is necessary for efficient decision-making. Recently a Shared Knowledge Decision Support System (SK-DSS) was presented with specific applications for energy saving in pumps and blowers. The SK-DSS is based on fuzzy analysis, identifies the operational conditions of devices and provides case-based solutions. With a large number of monitored devices, it is necessary to provide a global synthetic index, able to represent the performance of the plant and the different importance of devices. In this paper, the global index has been proposed and calculations performed with a multi-level fuzzy logic engine. In the bottom layer of this multi-level fuzzy logic engine, pumps and blowers are individually assessed. The top layer allows the calculation of a score in the range [0-100] by processing the outputs of the individual device assessments without losing the detailed information stored at the bottom level. Different weights are attributed to devices in the calibration of the top-layer fuzzification process. The output results are visualized to better identify the source of inefficiency. Results show the potential of such indicators with a larger number of plant devices and, in future, the global index will trigger an alert-system for plant manager.

Stream and Session

A3 Session

COinS
 
Jun 26th, 10:40 AM Jun 26th, 12:00 PM

A multi-level decision support system for energy optimization in WWTPs

The availability of real-time measurements in Wastewater Treatment Plants (WWTPs) can produce environmental and economic benefits. Since a WWTP can produce up to 300k records per day, computational analytics support is necessary for efficient decision-making. Recently a Shared Knowledge Decision Support System (SK-DSS) was presented with specific applications for energy saving in pumps and blowers. The SK-DSS is based on fuzzy analysis, identifies the operational conditions of devices and provides case-based solutions. With a large number of monitored devices, it is necessary to provide a global synthetic index, able to represent the performance of the plant and the different importance of devices. In this paper, the global index has been proposed and calculations performed with a multi-level fuzzy logic engine. In the bottom layer of this multi-level fuzzy logic engine, pumps and blowers are individually assessed. The top layer allows the calculation of a score in the range [0-100] by processing the outputs of the individual device assessments without losing the detailed information stored at the bottom level. Different weights are attributed to devices in the calibration of the top-layer fuzzification process. The output results are visualized to better identify the source of inefficiency. Results show the potential of such indicators with a larger number of plant devices and, in future, the global index will trigger an alert-system for plant manager.