An Intelligence-Based Fuzzy Inference System for Smart Home Real-Time Load Forecasting

摘要

This paper proposes an intelligence-based fuzzy inference system (FIS) for smart home real-time load forecasting. The electricity consumption of household is inherently nonlinear and heavily dependent on the habitual nature of power demand, activities of daily livings and on holidays or weekends, so it is often difficult to construct an adequate forecasting model for this type of load. To tackle with this problem, an intelligence-based FIS is proposed in this paper. In the proposed scheme, the popular particle swarm optimization (PSO) algorithm is first used to determine the locations of fuzzy membership functions. Then, the proposed FIS technique is used to develop the real-time load forecasting model. Because of the robust nature of the proposed FIS technique, which does not need to retrain or re-estimate model parameters, it is very suitable for smart home load forecasting. The proposed method was verified using the households load data. Simulation results indicate that the proposed method produces better forecasting result than existing methods.