A DEEPSO-GMDH Model for Supporting the Battery Energy Storage Investment Planning Decision-Making

Abstract

This paper presents a model for supporting the investment planning decision-making from the perspective of an independent energy provider that wants to integrate Battery Energy Storage Systems (BESS) in distribution networks. For supporting the decision, a conditional set of economically viable optimal solutions for the business model of buying and selling energy is identified in order to allow other decision criteria (e.g. loss reduction, reliability, ancillary services, etc.) to be evaluated to enhance the economic benefits as the result of the synergy between different applications of BESS. For this, a novel approach optimization model based on the metaheuristic Differential Evolutionary Particle Swarm optimization (DEEPSO) and the Group Method Data Handling (GMDH) neural network is proposed for sizing, location, and BESS operation schedule. Experimental results indicate that after identifying the breakeven cost of the business model, a good conditional decision set can be obtained for assessing then other business alternatives.