Adaptation in natural and artificial systems
Adaptation in natural and artificial systems
Multiobjective evolutionary algorithms in pump scheduling optimisation
ICECT'03 Proceedings of the third international conference on Engineering computational technology
A fully adaptive forecasting model for short-term drinking water demand
Environmental Modelling & Software
| Hi-index | 0.00 |
Water utilities across the developed world have been installing and operating telemetry and SCADA (supervisory control and data acquisition) facilities for at least three decades. They have amassed substantial quantities of historical operational data held in databases called data warehouses. Significant interest exists for extracting the underlying value from this data to support decision making and performance improvement. The water industry uses approximately 3% of total electricity production in developed countries such as the United Kingdom and the United States. Up to 90% of this electrical energy is consumed by pumps. Small improvements in pump efficiency will yield significant reductions in energy consumption with consequential reductions of carbon emissions to the atmosphere. Technologies to initially select a pump to match the expected performance requirements and then maintain optimal performance through periodic refurbishment are well established. Dynamically optimizing the scheduling of pump operation to improve efficiency under changing diurnal and seasonal water demand patterns is far more complex. This paper summarizes 20 years of progress in the development of pump efficiency improvement techniques and focuses on real-time dynamic optimization technologies and data-mining techniques to improve energy efficiency. Case study results from an automated real-time commercial pump-scheduling system, Aquadapt™, are presented.