A dynamic risk aversion model for virtual energy plant considering uncertainties and demand response. (15th January 2019)

• Record Type: Journal Article
• Authors: Ju, Liwei; Li, Peng; Tan, Zhongfu; Wang, Wei
• Is Part Of: International journal of energy research. Volume 43:Number 3(2019)
• Page Start: 1272
• Page End: 1293
• Publication Date: 2019-01-15
• Subjects: demand response -- risk aversion -- uncertainty -- virtual energy plant; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Power resources -- Research -- Periodicals; 621.042
• DOI: 10.1002/er.4366
• Languages: English
• ISSNs: 0363-907X

Abstract

Summary: To full use clean energy to meet load demand of electrical and thermal, the paper proposed a novel concept of virtual energy plant (VEP) including wind power plant (WPP), photovoltaic power generation (PV), combined heat and power generation (CHP), solar collectors (SC), electric boiler (EB), heat storage tank (HSK), and incentive‐based demand response (IBDR). Firstly, the basic structure of VEP is designed, including three subsystems, namely, electricity, heating, and energy storage. Then, a basic scheduling model is constructed under the objective of maximizing operating revenue without considering uncertainty. Thirdly, the conditional value at risk (CVaR) method and the robust optimization theory are used to handle the uncertainty factors in objective functions and condition constraints, and the risk aversion scheduling model is proposed. Finally, industrial park group in northern China are chosen, for example, analysis results show (1) VEP could convert the abandoned clean energy, use HSK to store heating energy during the valley load period, and supply heating energy in the peak period to obtain the excess economic benefits. (2) Lower‐prediction accuracy will amplify the uncertainty risk, when the robust β∈[0.8, 0.825]&(0.925, 1], the increase of confidence level β will lead to larger increase in CVaR. Especially when β∈(0.925, 1), decision makers are extremely disgusted with the risks brought by the uncertainty factors, and correspondingly, the output of clean energy becomes minimum. (3) When the capacity ratio of HSK, EB and the electricity price of peak, valley are lower than 3, the values of revenue, VaR and CVaR change faster, but the ratios are larger than 3, the values change slower, which indicates that the scale of HSK capacity needs to be properly controlled to optimize the use of clean energy, and price‐based demand response could improve the operation profit while controlling risk properly. In general, the proposed scheduling model can maximize the use of clean energy to obtain economic benefits while rationally controlling risks.