Dispatchable Virtual Power Plants with forecasting and decentralized control, for high levels of distributed energy resources grid penetration

Abstract

A Virtual Power Plant (VPP) composed of distributed energy resources (DER) with load and generation forecasting is presented in this paper. The large-scale integration of renewable energy and energy storage systems have introduced many technical and economic challenges to both customers and utilities. High levels of renewable power injection might reduce grid stability due to intermittencies. The proposed VPP mitigates these issues and can function as a major component of a utility grid with high levels of DER penetration. The VPP has two components: a forecast and analytics component, and an electrical systems control component. The forecaster generates the dispatch schedule for each player in the VPP e.g. a PV station, and the loads. This is done by taking historical load and weather data, models of the VPP players, and the weather forecast as inputs. The electrical systems controllers use the dispatch schedule to regulate power flows from each individual player and to the loads. Grid frequency and voltage support are also provided by the control, thereby improving system stability. An example VPP with three solar photovoltaic and two battery energy storage systems is presented and implemented using a real-time digital simulator. The results validate the VPP performance in load forecasting, power flow control and grid support. A dispatch schedule is generated for a 24-hour period and the VPP controller operation to service the loads and supply the power sold on the day-ahead market was verified.