Desenvolvimento de ferramenta computacional para a gestão energética de consumidores do grupo A

Abstract

Nowadays, companies are always looking to become more competitive in the market, and any savings in the final cost of products or services can make the company more competitive. There are many ways to reduce costs in the business environment and issues related to energy management, and in particular, the choice of contracted demand and tariff modality are considered important in this process, especially for consumers classified in Group A (high voltage). The main objective of this work is to develop a computational energy management tool to indicate the most viable tariff option and to calculate adequate contracted demand for Consumer Units (UCs) of Group A. A genetic algorithm was developed using the Python language and considering UC invoice history data. Part of the results obtained by the proposed tool were compared with the results presented by a similar tool available in the literature considering two case studies. The difference between contracted demands calculated by both tools was not significant in percentage terms, validating the proposed algorithm. The case studies carried out resulted in an estimated annual saving of BRL 775.72 for UC 1 and a three-year saving estimated at BRL 6699.48 for UC 2, if the suggested modifications had been implemented. As main contributions of the tool developed, the following stand out: (i) the calculation of the “optimal” contracted demand, the term optimal in this work refers to adequate demand, since a genetic algorithm was used to carry out the calculations, which is based on it is a heuristic method, so it is accurate but not extremely accurate, thus always resulting in a value close to the real optimum, (ii) the indication of the most appropriate tariff modality and (iii) it considers the tariff components (with and without tax ) for the peak and off-peak period, resulting in greater accuracy regarding the predicted savings for the UC when compared to similar tools available in the literature. For future work, the possibility of taking into account an additional set of variables for calculating the 'optimal' contracted demand is highlighted, considering, for example, the inclusion of the impact of the implementation of a photovoltaic generation system at the consumer unit.