Wind energy plays a crucial component in the contest to fulfill environmental control objectives. Wind energy, on the other hand, will only be able to fulfill its essential importance if the wind turbines work efficiently. The paper aims to analyze the application of artificial intelligence (AI) algorithms in wind speed.  In this paper, three network parameter optimization algorithms, AdaGrad, RMSprop, and Adam, are implemented and compared in the context of wind speed forecasting. This paper employs wind speed data obtained from the Jeddah Al Jazeera database in Saudi Arabia. Mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE), and R-squared are the four metrics used to assess performance. The experiment results show that the Adam algorithm outperforms the other optimization algorithms regarding forecasting accuracy and training time. Therefore, researchers can use this study to help them choose optimization algorithms for wind energy forecasting.