Akademik Veri Yönetim Sistemi
International Conference on Artificial Intelligence and Applied Mathematics in Engineering 2023 (ICAIAME 2023), Antalya, Türkiye, 3 - 05 Kasım 2023, ss.10
In today’s world, where demands are constantly increasing, energy has become an essential requirement for modern life. The increasing demand for energy has led to the growing importance of electricity markets. To contribute to the overall efficiency and reliability of these markets, Electricity Price Forecasting (EPF) helps market participants optimize their approaches and reduce risks by providing valuable insights. Therefore, EPF plays a vital role in the decision-making processes of market participants. Over time, various methods have been tested to improve the accuracy of EPF. Automatic Programming (AP) methods propose models with high prediction accuracy in solving many complex engineering problems. In this study, different AP methods are compared for the first time for EPF. The first method is Artificial Bee Colony Programming (ABCP), which is based on the search for food sources by honeybees. Other methods are and Genetic Programming (GP), which is developed by genetic operators such as crossover and mutation, Immune Plasma Programming (IPP) inspired by plasma therapy. The main findings of the study show that AP methods can be effectively applied to complex problems such as EPF and that GP produces models with higher predictive accuracy compared to other methods.