Prediction of market-clearing price using neural networks based methods and boosting algorithms

Abstract

The development of Turkey's industry is contributing to a significant rise in electrical energy demand. Also, electricity is one of the critical elements in the household sectors. Therefore, the planning and managing of electrical energy is of great importance to support economic growth. In addition, effective prediction of market-clearing prices (MCP) is critical topic to meet the increasing energy demand and provide basis for decision making process. In this paper, MCP is predicted using artificial neural network (ANN), convolutional neural network (CNN), and also three boosting algorithms including extreme gradient boosting (XGBoost), categorical boosting (CatBoost), and adaptive boosting (AdaBoost). Various performance metrics are employed to evaluate the prediction performance of proposed methods. The results showed that proposed methods provide reasonable prediction results for energy sector. Hence, producers and consumers can use these methods to determine the bidding strategies and to maximize their profits.