A hybrid forecasting model with parameter optimization for short-term load forecasting of micro-grids

Short-term load forecasting is an important part in the energy management of micro-grid. The forecasting errors directly affect the economic efficiency of operation. Compared to larger-scale power grid, micro-grid is more difficult to realize the short-term load forecasting for its smaller capacity and higher randomness. A hybrid load forecasting model with parameter optimization is proposed for short-term load forecasting of micro-grids, being composed of Empirical Mode Decomposition (EMD), Extended Kalman Filter (EKF), Extreme Learning Machine with Kernel (KELM) and Particle Swarm Optimization (PSO). Firstly, the time-series load data are decomposed into a number of Intrinsic Mode Function (IMF) components through EMD. Two typical different forecasting algorithms (EKF and KELM) are adopted to predict different kinds of IMF components. Particle Swarm Optimization (PSO) is used to optimize the parameters in the model. Considering the limited computation resources, an implementation mode based on off-line parameter optimization, period parameters updating and on-line load forecasting is proposed. Finally, four typical micro-grids with different users and capacities are used to test the accuracy and efficiency of the forecasting model.