Estimating Urban Wind Speeds and Wind Power Potentials Based on Machine Learning with City Fast Fluid Dynamics Training Data

Abstract

Wind power is known as a major renewable and eco-friendly power generation source. As a clean and cost-effective energy source, wind power utilization has grown rapidly worldwide. A roof-mounted wind turbine is a wind power system that lowers energy transmission costs and benefits from wind power potential in urban areas. However, predicting wind power potential is a complex problem because of unpredictable wind patterns, particularly in urban areas. In this study, by using computational fluid dynamics (CFD) and the concept of nondimensionality, with the help of machine learning techniques, we demonstrate a new method for predicting the wind power potential of a cluster of roof-mounted wind turbines over an actual urban area in Montreal, Canada. CFD simulations are achieved using city fast fluid dynamics (CityFFD), developed for urban microclimate simulations. The random forest model trains data generated by CityFFD for wind prediction. The accuracy of CityFFD is investigated by modeling an actual urban area and comparing the numerical data with measured data from a local weather station. The proposed technique is demonstrated by estimating the wind power potential in the downtown area with more than 250 buildings for a long-term period (2020–2049).

Machine Learning for Pedestrian-Level Wind Comfort Analysis

“…Refs. [74,75] emphasized the efficacy of ML in forecasting wind power. Considering the potential of ML to predict wind power in novel regions utilizing daily wind speed data, Demolli et al demonstrated how ML can predict wind power in various locations.…”

“…The results of their study verified the precision of ML in urban environments and emphasized the impact of neighborhood attributes, rather than the height of buildings, on the wind energy potential. They recommended that future wind energy research consider the urban morphology [74,75]. Ref.…”

Modeling the Normalized Urban Heat Island for the City of Karlsruhe by Linking Urban Morphology and Green Infrastructure

“…The spatial thermal impact of individual urban structures is initially unknown and dependent on the dynamics of the urban atmosphere. Many models tried to reproduce the dynamics with the help of computational fluid dynamics (CFD) [47][48][49][50][51][52][53]. For this study, the assumption was made that there is no horizontal air mass exchange, so that the UHI forms a typical spatial pattern.…”

Using Machine Learning to Predict Wind Flow in Urban Areas

“…Previous studies in atmospheric science utilized machine learning to characterize the flow in simple structures such as a duct, a single rectangular body, or a blade [15][16][17][18]. However, to the best of our knowledge, the only study that considered the complexities of urban flows utilized it only above roof tops [19].…”