The transition from traditional energy sources to renewable alternatives has become imperative in combating climate change. Wind power has emerged as a viable solution, driven by technological advancements and its role in addressing global energy needs. This study employs eXplainable AI (XAI), specifically TimeSHAP, to enhance the interpretability of a time-series model, providing insights into the contributions of each time-event and feature to the predictions. Utilizing a long short-term memory (LSTM) model for accurate wind power forecasting, the study leverages its ability to capture temporal dependencies in data. Through TimeSHAP’s local explanations, the interpretability of the LSTM model in capturing nonlinear and dynamic wind energy patterns is showcased. This combination of LSTM model and TimeSHAP not only improves forecast accuracy but also enhances understanding of the factors influencing wind power predictions.