Hydropower energy recovery form the water supply network : feasibility, risk analysis, optimisation and implementation

Abstract

The provision of clean and reliable water supply is highly energy intensive. Recent research has focused on the development and application of energy and hydraulic efficiency measures to streamline water usage and improve energy security. This thesis presents the findings of an investigation of the potential for hydropower energy recovery within water supply networks (WSNs). Four core themes are presented and discussed; Feasibility, Risk Analysis, Optimisation and Implementation. Firstly, a feasibility study was undertaken of the potential for hydropower energy recovery at existing locations within the WSNs of Ireland and Wales. Locations investigated included pressure reducing valves (PRVs), break pressure tanks and reservoirs. The total power generation available at the sites investigated was found to be 1.3MW. Flow rate variation, turbine selection and turbine costs were identified as key factors when determining the investment payback periods of potential projects. Secondly, through an analysis of ten years of long term flow rate data, a methodology for future forecasting of flow rates at potential hydropower locations was developed. Long term flow variation due to changes in water demand will impact upon power generation capacity. Large changes in demand could render a turbine unsuitable in the future. Multiple linear regression and artificial neural network forecasting techniques were applied and compared as potential methods for long term forecasting of flow rates at hydropower locations. Using these models, future scenarios for flow rates and hence power outputs were forecast.