Modelling and Uncertainty Analysis of On-site Renewable Sources for Optimal EV Charging

Road transport is the second largest dimension of carbon emission, both nationally in the UK and locally in Newcastle upon Tyne, contributing about 33% of total emission in 2020. In line with the UK’s target to reach net zero by 2050 (and the city of Newcastle upon Tyne’s ambition to do so by 2030), electric vehicles (EVs) play a critical role in meeting net zero road transportation though it does not automatically imply a reduction of overall emission nationally or globally if the electricity to charge EVs is sourced from the fossil fuels. To achieve optimal EV charging, a better understanding of the uncertainties of ORES power generation is necessary. ANN (Artificial Neural Network) and time series forecasting methods are used in this paper to model wind and solar power generation and the power generation of ORES. Such a model is able to represent the relationship between the power generation and the wind speed as well as solar irradiation, which is of significant uncertainties due to weather changes in both short-time (hourly) and long-term (seasonally). The proposed method uses historical solar irradiance and wind speed data, together with numerical weather prediction (NWP) data. The proposed neural network is verified with the historic data at Newcastle upon Tyne for the years 2020 to 2022. The proposed methods have a root mean square error (RSME) of 2.26 (m/s) in wind speed modelling, and the RSME of solar irradiance is 50.79 (W/m2 ). The uncertainties analysis shows that the uncertainties in wind speed at Newcastle upon Tyne can be modelled as a Weibull distribution with parameters A = 19.98 and B = 1.91.