No abstract

<p>The increasing integration of distributed energy resources (DERs) into power grid makes it significant to forecast solar irradiance for power system planning. With the advent of deep learning techniques, it is possible to forecast solar irradiance accurately for a longer time. In this paper, day-ahead solar irradiance is forecasted using encoder-decoder sequence-to-sequence models with attention mechanism. This study formulates the problem as structured multivariate forecasting and comprehensive experiments are made with the data collected from National Solar Radiation Database (NSRDB). Two error metrics are adopted to measure the errors of encoder-decoder sequence-to-sequence model and compared with smart persistence (SP), back propagation neural network (BPNN), recurrent neural network (RNN), long short term memory (LSTM) and encoder-decoder sequence-to-sequence LSTM with attention mechanism (Enc-Dec-LSTM). Compared with SP, BPNN and RNN, Enc-Dec-LSTM is more accurate and has reduced forecast error of 31.1%, 19.3% and 8.5% respectively for day-ahead solar irradiance forecast with 31.07% as forecast skill.</p>

Summary Microgrid with distributed energy resources and energy storage system provides sustainability and resiliency. In this research, residential community microgrid is examined with responsive loads that create flexible generation‐demand model. An optimization algorithm using mixed integer linear programming (MILP) has been formulated to minimize the operating cost and emission of dispatchable power generation, with the help of demand response. Usually, in renewable energy–based grid‐connected microgrid, the batteries are managed under partial state of charge (SoC) conditions due to the limit of power imported from grid. The proposed MILP model ensures full SoC operation and safe charging or discharging dynamics of the battery in order to enhance its lifespan. Moreover, the day‐ahead scheduling of household appliances is carried out using a novel hybrid knapsack method, which combines binary and fractional knapsack algorithms. An electric vehicle battery is considered as a flexible power load, which offers an unique way of approach in scheduling of appliances. The results confirm that the power demanded by the appliances is fulfilled at the user‐specified hour for maximum comfort along with minimum operating cost of microgrid. Generic algebraic modeling system (GAMS) tool is used to run the proposed algorithms.