A Review of Bio-Inspired Computational Intelligence Algorithms in Electricity Load Forecasting

Abstract

The electricity load forecasting is an emerging research field in computer science. It plays a major role in the power system and power utility companies for balancing the power demand with the power supply. An accurate load forecasting also helps for managing the energy usage, balancing the energy consumption, and monetizing the energy in smart buildings. The machine learning methodologies achieve an accurate load forecasting by effectively analyzing the nonlinear and uncertain characteristics of the load. However, the performance of the machine learning can be improved by optimizing the network parameters, so the machine learning algorithms combined with the bio-inspired computational intelligence algorithms can greatly increase the accuracy of the load forecasting. This chapter begins with the introduction to smart building, load forecasting, machine learning, and bio-inspired computational intelligence algorithms. Then, it discusses the classification of the bio-inspired computational intelligence algorithms, namely evolution-based algorithms, swarm intelligence-based algorithms, neural system-based algorithms, and artificial immune system (AIS)-based algorithms. Finally, it reviews the bio-inspired computational intelligence algorithms utilized by various researchers in the electricity load forecasting.