Virtual storage-based DSM with error-driven prediction modulation for microgrids

Abstract

Microgrids consider adjustable loads in demand-side management (DSM), which respond to dynamic market prices. A reliable DSM strategy relies on load forecasting techniques in day-ahead (DA) scheduling. This paper applies an error-driven prediction modulation to evaluate these differences. In addition, this paper creates two new DSM methods with an evaluation environment to utilize this modulation. The first method adds this modulation directly to traditional microgrid DSM with electrical storage. The second method creates two virtual sub-storages for behavior adjustment in both DA and real-time (RT) markets. The results of numerical studies indicate that the new DSM methods can reduce microgrid operation costs.