Optimal coordinated generation scheduling considering day‐ahead PV and wind power forecast uncertainty

Abstract

AbstractEconomic operation and reliable supply‐demand balance are problems of paramount importance in power grids with a massive share of intermittent renewable energy sources (RESs) of great interest. This paper sought an optimal coordinated generation scheduling for day‐ahead power system operation considering RESs and energy storage units. Renewable power generation, particularly, wind and photovoltaic are uncontrollable, whereas can be predicted using forecasting models. Within the proposed framework, a hyperparameter‐optimized long short‐term memory (LSTM) regression model is employed to forecast the day‐ahead weather from the historical time‐series weather data. Eventually, an empirical formula is used to estimate the power conversion from the day‐ahead weather forecasts for a selected PV module and wind turbine. The objective of the scheduling framework is to keep a delicate supply‐demand balance at the lowest possible cost of generation while maintaining the prevailing generation and system constraints. A variance measure uncertainty handling‐based grey wolf optimizer (GWO) technique is used to find the optimal day‐ahead generation schedules and dispatches under RESs forecast uncertainty. The proposed generation scheduling framework is examined on the IEEE 6 and 30‐bus systems. In the studied scenarios, the coordinated operation of generators can decrease the total day‐ahead operating cost for the modified IEEE 6‐bus system by 2.57% compared to supplying electricity generation with conventional generators alone. Likewise, the total operating cost from the coordinated operation of all generation portfolios was reduced by 6.93% from the operating cost of generation during base case simulation (supply only from dispatchable thermal units) on the modified IEEE 30‐bus system. Moreover, the case studies show that coordinated generation scheduling can mitigate the RESs power variability problem, provide secure supply‐demand operation, and minimize the operating cost of electricity generation.