Intelligent Management and Artificial Intelligence: Trends, Challenges, and Opportunities, Vol.2

Purpose: The purpose of this paper is to evaluate the application of advanced artificial intelligence methods in the forecasting of photovoltaic farm electricity prices and to analyse their impact on the optimisation of business decisions and operational processes. Need for the study: With the increasing volatility of energy markets due to unstable weather conditions, dynamic regulatory changes and financial risks associated with long-term PPAs, traditional forecasting methods are proving insufficient. It is necessary to develop state-of-the-art solutions that, through the integration of Big Data and advanced machine learning algorithms, enable accurate energy price forecasting. Methodology: The study is based on a two-stage analysis: a literature review and a conceptual analysis, leading to the formulation of the concept of an intelligent energy price forecasting system. This system integrates data from multiple sources, which are subjected to processes of cleaning, transformation and feature extraction. The resulting hybrid predictive model, based on the LSTM-Transformer architecture, is enriched with adaptation mechanisms that enable continuous updating of the models in the context of optimising business and operational decisions. The approach is validated through an empirical case study on a 1 MW PV farm, using historical data to train the model and evaluate its performance. Findings: The literature analysis and the conceptual analysis conducted show that the implementation of a system based on a hybrid AI model allows for a reduction in energy price forecast errors compared to traditional methods. Adaptive mechanisms allow the model to adapt to changing market conditions on an ongoing basis, resulting in higher forecast accuracy and more effective financial risk management. Furthermore, the case study results confirmed improvements in forecast accuracy, which in turn enhances operational decision-making for the PV farm. Practical Implications: The proposed AI system enables PV farm operators to make data-driven decisions based on reliable data, resulting in optimised sales strategies, energy storage management and stabilised financial flows, thus providing a competitive advantage in the renewables market.