Premium
Residential virtual power plant with photovoltaic output forecasting and demand response
Author(s) -
Cui Shichang,
Wang YanWu,
Lin Xiangning,
Xiao JiangWen
Publication year - 2019
Publication title -
asian journal of control
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.769
H-Index - 53
eISSN - 1934-6093
pISSN - 1561-8625
DOI - 10.1002/asjc.2041
Subject(s) - virtual power plant , photovoltaic system , renewable energy , demand response , computer science , grid , time horizon , mathematical optimization , linear programming , wind power , electricity , distributed generation , engineering , electrical engineering , algorithm , mathematics , geometry
Decreasing conventional power supply is promoting the development of the distributed renewable energy sources, such as solar power and wind power. Recently rooftop photovoltaic has been widely applied, and accordingly efficient energy management is getting increasingly important for fully use of renewable energy and the peak shaving of the main grid. This paper investigates the residential energy management as a small‐scale virtual power plant (VPP) connected to the main grid includes distributed energy resources, energy storages and residential loads. The self‐organizing map (SOM) and the radical basis function (RBF) networks are adopted to classify the weather types and predict hourly photovoltaic output precisely. In a time‐of‐use electricity market, price‐based demand response is applied to adjust the demand. The residential VPP has two goals: maximum profit by selling surplus power to grid and minimum power purchased from grid. The two goals are integrated as an optimization object by introducing a weight parameter. The algorithm combining receding horizon optimization and linear programming is proposed to solve the optimization problem in residential VPP. Numerical simulation tests can help to find the most suitable value of the weight parameter. Different scenarios are simulated and discussed to demonstrate the performance of the VPP and the proposed algorithm.