Gerenciamento de resposta a demanda residencial utilizando algoritmo genético

Abstract

Electricity supply and demand need to be balanced simultaneously. To ensure sustained supply, the total electricity generation capacity must be greater than the expected maximum demand. In recent years, the demand for electricity is increasing every year, and the construction of new power plants is being barred due to the economic and environmental cost. One method for delaying or even avoiding certain investments are Demand Response Programs. Its purpose is to modify energy consumption by modeling system loads to fill the valleys and reducing peak consumption loads, thus making the energy system more efficient. Thus, this paper presents the development of a genetic algorithm capable of managing a residential demand response system. Therefore, it was necessary to generate household consumption models, which are based on real data acquired through various studies of consumer usage habits, load characterization, equipment power and daily use patterns. As a demand response strategy, load reallocation within a 24-hour horizon was used, combined with dynamic energy pricing, and its values varied based on the national load curve. The work was performed using Matlab software. Given this, it was found that the algorithm was able to reduce the total cost of energy by 16,40% for 100 consumers, divided into 5 consumption ranges, which requires that the methodology used had a positive impact on the total energy cost (sic).