Risk-constrained stochastic power procurement of storage-based large electricity consumer. (April 2020)

• Record Type: Journal Article
• Authors: Cao, Yan; Wang, Qiangfeng; Fan, Qingming; Nojavan, Sayyad; Jermsittiparsert, Kittisak
• Is Part Of: Journal of energy storage. Volume 28(2020)
• Publication Date: 2020-04
• Subjects: Large electricity consumer -- Risk-measuring -- Energy trading and business -- Energy storage management -- Stochastic programming -- Downside risk constraints; Energy storage -- Periodicals; Energy storage -- Research -- Periodicals; 621.3126
• DOI: 10.1016/j.est.2019.101183
• Languages: English
• ISSNs: 2352-152X

Abstract

Highlights: Stochastic energy procurement of large consumers in the uncertain environment. Use of downside risk constraints method to model the imposed risks from uncertainty. Propose the risk-averse and risk-neutral strategies for large electricity consumers. Calculate of exact zero risk cost by downside risk constraints method. Abstract: Large electricity consumers can be either a large industrial consumer or a coalition of small electricity consumers. Large consumers (LCs) confront with various uncertainties due to the use of various power resources in the power procurement process, such as renewable resources, self-generation units, forward contracts, and pool market. These uncertainties can be lead to many financial risks for LCs. In this paper, the stochastic power procurement problem of large consumers is solved, and the new risk-measurement method is used to analyze the large consumer risks in power procurement process. The mentioned risk-measurement method is called downside risk constraints (DRC) method, which is used to model the financial risk imposed from uncertain parameters along with the stochastic problems. According to obtained results, it can be concluded that DRC method is a non-equilibrium method, which is applied clearly as a constraint to the optimization problem. In addition by using the DRC, LC can experience lower-risk strategy in the power procurement problem. Also, using DRC can make the total cost of large consumer independent of scenarios, which led to the lower-risk experiencing by the large consumer. Finally, results are expressed that lower-risk cost in DRC is less than the cost of the worst scenario in stochastic programming.