Machine Learning Classification and Prediction of Wind Estimation Using Artificial Intelligence Techniques and Normal PDF

Estimating wind energy at a specific site depends on how well the real data in that area can be represented using an appropriate distribution function. In fact, sites differ extent to which their from one region another, despite widespread use of Weibull function representing speed various locations world. this study, new probability model (normal PDF) was tested implement several Jordan. The results show high compatibility between and resources Therefore, used estimate values extracted turbines compared those obtained by PDF. Several artificial intelligence techniques were (GA, BFOA, SA, neuro-fuzzy method) predict parameters both normal PDFs reflected conjunction with actual observed probabilities. Afterward, goodness fit decided aid two performance indicators (RMSE MAE). Surprisingly, (PDF) outstripped PDF, interestingly, BFOA SA most accurate methods. last stage, machine learning classify error level estimated based trained PDF parameters. proposed novel methodology aims parameters, as subsequent calculation phases depend proper selection these Hence, 24 classifier algorithms study. medium tree shows best accuracy training time points view, while ensemble-boosted trees poor regarding providing correct predictions.