Intelligent Generation Scheduler for a Smart Micro Grid

The exponential increase in the electricity demand, has caused a phenomenal growth of renewable energy sources, fast aiding or replacing the conventional sources. The fluctuating nature of renewable energy sources present as a challenge to power generation scheduling in a renewable Micro-grid. To address these challenges associated with unpredictability of load and renewable energy sources, this paper proposes a novel generalized algorithm for optimal day ahead power scheduling, considering the economic cost and the meteorological factors affecting forecasting of renewable energy sources. The paper contains two levels, the lower level includes 24-hour forecasting of the factors affecting generation, such as solar irradiance for solar power forecasting, etc. The higher level includes optimization for economic generation plan for the microgrid. For forecasting, Artificial Neural Networks is used, which reduces the uncertainty factor in the scheduling process. Multi-Objective Genetic Algorithm is used for optimization of the generation source with respect to demand and cost. This paper highlights the efficiency of the proposed algorithm with the help of a case study considering solar and wind as generation sources as compared to the existing MILP based algorithm in MATLAB environment