On the use of measure-correlate-predict methodologies and energy demand forecasting to assess energy storage capabilities for offshore wind farms

Abstract

Energy storage is crucial for the continued penetration of renewable energy. One of the most important reasons for this is that, for a given point of time, the availability of renewable energy resources rarely matches the demand for electrical energy. The integration of offshore windfarms with energy storage facilities, requires a capital-intensive investment which can only be justified by an adequate return on investment (ROI). Currently, Measure-Correlate-Predict (MCP) analysis is used to assess the viability of offshore windfarms while energy demand forecasting is normally used to manage and plan the electricity grid infrastructure. This research combined wind energy prediction methodologies with Energy Demand Forecasting (EDF) methodologies to size the energy storage capacity for an offshore windfarm and evaluated the economic feasibility. This research analysed various regression techniques for MCP analysis. Data from a Light Detection and Ranging (LiDAR) system were utilised. The study was extended to analyse the behaviour of a hypothetical floating windfarm, situated off the Northern Coast of the Island of Malta. The effect of using the different regression techniques for MCP analysis on the power output from the windfarm could therefore be evaluated. The second part of the research used a combination of ARIMA and regression techniques to forecast the energy demand over several years. The output from the windfarm was applied to a model which integrated the said windfarm to an Energy Storage System (ESS) and the electricity grid. Measurement matrices were used to compare the behaviour of the combined windfarm, ESS and electricity grid, based on the actual and predicted data from the various regression techniques used for the MCP analysis and EDF. This created a matrix of results which was used to determine the optimal combination of regression techniques used for MCP analysis and EDF, following which, the optimal capacity of the ESS was established. The long-term behaviour of the windfarm and the of the energy storage system were also predicted. The Levelised Cost of Energy (LCOE) for the windfarm and the Levelised Cost of Storage (LCOS) for the Energy Storage System were also calculated, using different windfarm scenarios, and analysing the error due to the use of the MCP and EDF methodologies. This research therefore established a methodology for combining MCP and EDF to determine the optimal capacity of an ESS which was coupled to an offshore windfarm and the electricity grid. The error in establishing this capacity was determined. The end result was the determination of the LCOE of the windfarm and the LCOS of the ESS based on the combination of MCP analysis and EDF, together with the error introduced due to the use of the two methodologies.