The wind has exploded in popularity in recent years, and it is expected to continue to do so in the future. To efficiently schedule and utilize that source of energy, better forecasting methodologies are required. In the recent decade, numerous studies on forecasting wind speed generation on timescales of minutes, days, months, and years have been conducted. According to a comprehensive set of forecasting methodologies, physical approaches, statistical or hybrid methods, such as neural networks, are the most widely used tactics for predicting wind speed day-ahead. The goal of this paper is to keep prediction error to a minimum. Plotting and predicting the speed of wind in Dohuk city, KRG/Iraq, using a recurrent neural network model. The LSTM architecture is the type of artificial recurrent neural network used in deep learning. Based on the dataset, the approach plots the predicted wind speed and forecasts the future dispersion. Data centers were suggested for Dhouk as a way to utilize the electricity generated by wind turbines and integrate it with other sources of renewable energy and the electrical grid. The city accepted the proposal. With future implementations, it is possible to accurately quantify how much energy is being created as well as how much money is spent on operations and maintenance.
- Zozan Saadallah Hussain, Najat Yohana Danha , Karwan Muhammed Muheden, Shahab Wahhab Kareem
- International Journal of Intelligent Systems and Applications in Engineering (IJISAE)