Electric power management in a microgrid analyzing photovoltaic arrays and a turbine-generator system

Abstract

One of the priority objectives of microgrids is to achieve energy self-sufficiency, generally resorting to distributed generation sources and backup systems; however, they are usually connected to conventional electrical networks that ensure supply to the loads. Addressing this problem, this work presents a proposal (managing elements) to minimize the dependence of power from the external electrical system in the months of greatest demand and thus guarantee the supply of the other months. The proposed methodology compares two statistical techniques: central composition design with 20 simulated experimental replicas and regression with 28. In both cases, the monthly average purchased power is analyzed as a primary response and its standard deviation as a secondary. The study variables are seven photovoltaic arrays and the feed characteristics of the turbine-generator storage of the microgrid of the Center for Development of Renewable Energies (CEDER), belonging to the Center for Energy Research, Environmental and Technological (CIEMAT). The results, with high predictive quality supported by indexes of approach to the real values of solar radiation and the operation of the turbine-generator binomial, provide regions where CEDER has the possibility of increasing the capacities of solar systems and/or modifying the geometry of the mini-hydraulics supply according to your specific conditions.

Hybrid Driving Training and Particle Swarm Optimization Algorithm-Based Optimal Control for Performance Improvement of Microgrids

“…In this paper, the authors use the central composite response surface methodology (CCRSM) to model the PI and PIA controllers. The CCRSM method is a statistical tool that can be utilized for analyzing and modeling power systems [41,42]. It involves modeling the relationship between response and design variables, using a fitting technique.…”