Applications of the Chaotic Quantum Genetic Algorithm with Support Vector Regression in Load Forecasting

• Main Authors: Cheng-Wen Lee, Bing-Yi Lin
• Format: Article
• Language: English
• Published: MDPI AG 2017-11-01
• Series: Energies
• Subjects: chaotic mapping function; support vector regression (SVR); quantum genetic algorithm (QGA); electricity demand forecasting
• Online Access: https://www.mdpi.com/1996-1073/10/11/1832

Accurate electricity forecasting is still the critical issue in many energy management fields. The applications of hybrid novel algorithms with support vector regression (SVR) models to overcome the premature convergence problem and improve forecasting accuracy levels also deserve to be widely explored. This paper applies chaotic function and quantum computing concepts to address the embedded drawbacks including crossover and mutation operations of genetic algorithms. Then, this paper proposes a novel electricity load forecasting model by hybridizing chaotic function and quantum computing with GA in an SVR model (named SVRCQGA) to achieve more satisfactory forecasting accuracy levels. Experimental examples demonstrate that the proposed SVRCQGA model is superior to other competitive models.