Artificial Intelligence and Big Data for Dynamic Pricing in Renewable Energy Markets: A Review of Forecasting and Optimization Approaches

Intelligent Automation and BioMed Genomics Laboratory, FST of Tangier, Abdelmalek Essaadi University, Tetouan, Morocco

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

The use of artificial intelligence (AI) and big data technologies has increased the usage of dynamic pricing mechanisms in renewable energy markets. The increasing amount of renewable energy being installed into the electric grid i.e. solar and wind is causing variable amounts of electricity to be produced, which is causing price volatility in the wholesale electricity market. AI-based forecasting and optimization techniques can effectively utilize large energy datasets to improve pricing methodologies.

This paper provides an overview of current research on the applications of artificial intelligence and big data for dynamic pricing in renewable energy markets. The review will discuss machine learning (ML), deep learning (DL), and reinforcement learning (RL) techniques, as well as big data platforms that are being used to process information obtained from smart meters, weather stations, and wholesale electricity market prices. Research demonstrates that AI models outperform traditional methods for forecasting; specifically, hybrid deep learning frameworks have provided very good results with mean absolute errors (MAEs) of as low as 0.138, root mean square error (RMSEs) of as low as 0.166 for various electricity markets. A comparison of CNN-BiLSTM models using hyperparameter optimization to traditional approaches resulted in an average reduction in RMSE of 16.7%, and an average reduction in MAE of 23.46%. In addition to achieving high levels of accuracy, deep learning-based forecasting tools also provide significant computational advantages. Specifically, deep learning-based forecasting tools run at least five times faster than traditional benchmarks for multiple markets. These results clearly indicate that AI-based forecasting models produce superior results relative to traditional forecasting methods as measured by metrics such as mean absolute percentage error (MAPE) and root mean square error (RMSE). Finally, this review identifies several critical challenges facing the use of AI-driven dynamic pricing in smart grid systems. These challenges include, but are not limited to, data availability, model interpretability, and regulatory constraints.