Wind Speed Prediction Analysis in Pekalongan Regency Using the Decision Tree Regression Algorithm

The coastal characteristics of Pekalongan Regency indicate substantial wind potential, prompting the need for studies on renewable energy utilization. This study aims to forecast wind speed and estimate the electrical power that can be generated using a Decision Tree Regression algorithm. Eleven years of historical climate data (2013–2023) from the Indonesian Meteorology, Climatology, and Geophysics Agency (BMKG) were used to build the model. Evaluation results show a Mean Squared Error (MSE) of 4.108 and a coefficient of determination (R²) of 0.049, indicating the model has limited predictive performance. The 2024 wind speed forecast ranges from 3.8 to 7 m/s, with an average of 4.5 m/s. This wind speed translates into estimated electrical power ranging from 844 to 5,277 watts, averaging 1,599 watts per month, equivalent to a potential monthly energy output of 191.88 kWh. This study concludes that while there is potential for small-scale Wind Power Plant (PLTB) development, such as for public street lighting, the accuracy of the predictive model needs to be significantly improved for more critical applications.