Neural Network Models for Wind Power Forecasting in Smart Cities- A review

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Urbanization’s relentless advance intensifies the quest for sustainable energy sources, with smart cities leading the shift toward sustainability. In these innovative urban landscapes, wind power is pivotal in the clean energy paradigm. Efficient wind energy utilization hinges on accurate wind power forecasting, essential for energy management and grid stability. This review explores the use of neural network models for wind power forecasting in smart cities, driven by wind power’s growing importance in urban energy strategies and the expanding role of artificial neural networks (ANNs) in wind power prediction. Wind power integration mitigates greenhouse gas emissions and enhances energy resilience in urban settings. However, wind’s inherently variable nature necessitates precise forecasting. The surge in ANN use for wind power forecasting is another key driver of this review, as ANNs excel at modelling complex relationships in data. This review highlights the synergy between wind power forecasting and neural network models, emphasizing ANNs’ vital role in enhancing the accuracy of wind power predictions in urban environments. It covers neural network fundamentals, data preprocessing, diverse neural network architectures, and their applicability in short-term and long-term wind power forecasting. It also delves into training, validation methods, performance assessment metrics, challenges, and prospects. As smart cities champion urban sustainability, neural network models for wind power forecasting are poised to revolutionize urban energy systems, making them cleaner, more efficient, and more resilient.

This publication has 21 references indexed in Scilit: