Distributed management of energy communities using stochastic profile steering

Author(s)

Lopez, Juan Camilo ; Pappu, Aditya ; Hoogsteen, Gerwin ; Hurink, Johann L. ; Rider, Marcos J.

• Document type: Journal article
• Source: INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS; v. 158, p. 10-pg., 2024-04-06.

Abstract

This paper presents a novel approach for distributed management of energy communities. The proposed method utilizes a stochastic profile steering algorithm as a greedy heuristic. The optimization process considers random parameters such as local forecasted demand, photovoltaic (PV) production, and the initial energy of electric vehicles (EVs) as embedded scenarios. Profile steering coordinates the flexible electricity assets within an energy community by determining the contribution of each prosumer's profile to the average value of the objective function. It iteratively selects the prosumer that contributes the most until no further improvements can be made. This process scales linearly with the number of controllable prosumers and can achieve various community -level objectives, such as maximizing self-sufficiency or minimizing aggregated cost -of -energy, even when dealing with non -convex optimization problems for modeling each prosumer's local energy management system. The outcome of the proposed method optimizes the average value of the community's objective while ensuring that grid limitations are met within a specified probability. The proposed method is evaluated through simulations involving small-scale communities (5 households) and large-scale communities (100 households). The results demonstrate the efficiency, flexibility, and scalability of the proposed method, as well as its ability to reschedule the aggregated demand to ensure that grid limits are not violated with at least a 95% probability. (AU)

• FAPESP's process: 21/11380-5 - CPTEn - São Paulo Center for the Study of Energy Transition
• Grantee: Luiz Carlos Pereira da Silva
• Support Opportunities: Research Grants - Science Centers for Development
• FAPESP's process: 19/01906-0 - Optimal distributed restoration of electrical distribution systems using alternating direction method of multipliers
• Grantee: Juan Camilo Lopez Amezquita
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• FAPESP's process: 21/11310-7 - Application of heuristic techniques to deal with binary variables in optimal distributed restoration
• Grantee: Juan Camilo Lopez Amezquita
• Support Opportunities: Scholarships abroad - Research Internship - Post-doctor