article · IET Generation Transmission & Distribution
Load forecasting models based on Gaussian Process Regression apply a Bayesian framework to predict electrical power demand while handling measurement uncertainty. This non-parametric, kernel-based approach generates hourly and monthly electricity load forecasts using historical hourly demand and environmental information. Twelve specific models were evaluated on real data from an Australian city as well as four Indian cities located in Maharashtra state: Nasik, Bhusawal, Kolhapur, and Aurangabad. Performance assessments using mean average percentage error demonstrated strong accuracy, achieving error rates of 0.15 percent for the Australian location and between 0.002 percent and 0.209 percent across the Indian cities. Among the tested variations, configurations utilising an exponential kernel function consistently achieved the lowest error margins. Comparative evaluations confirmed that this method outperforms existing predictive models across the analysed locations.
Accurate electricity demand forecasting is vital for grid operators and power utilities to balance energy generation, avoid blackouts, and manage costs. By reliably predicting power usage alongside environmental conditions and accounting for measurement uncertainties, this approach provides a highly precise tool that can help utility planners maintain electrical network stability and plan resource distribution more effectively across diverse urban regions.
This predictive tool could assist electrical utilities and grid operators seeking to optimise supply dispatch and power management. The model has reached an applied and tested stage using historical municipal and weather datasets from Australia and India. Commercial deployment would likely require integration into utility control systems or energy management software platforms to evaluate real-time grid conditions and operational workflows.
AI-generated from the published abstract. Always read the original work before citing.
Abstract In this paper, Gaussian Process Regression (GPR)‐based models which use the Bayesian approach to regression analysis problem such as load forecasting (LF) are proposed. The GPR is a non‐parametric kernel‐based learning method having the ability to provide correct predictions with uncertainty in measurements. The proposed model provides an hourly and monthly load forecast for an Australian city and four Indian cities in the Maharashtra state. Twelve GPR models are trained with historical datasets including hourly load and environmental data. To evaluate the trained model, the actual and predicted load demand curve is plotted and mean average percentage error (MAPE) is calculated corresponding to different kernel functions of the GPR model. To the best of the author's knowledge, the prediction of load demand using GPR for Indian cities of Maharashtra state has been made for the first time. The calculated MAPE in LF is 0.15% for Australia and 0.002%, 0.209%, 0.077%, and 0.140% for Indian cities viz. Nasik, Bhusawal, Kolhapur, and Aurangabad, respectively. The test results illustrate that minimum MAPE in load prediction is obtained using the proposed model that is GPR with ‘Exponential’ kernel functions. Furthermore, the comparative analysis with the existing approaches confirms the dominance of the proposed model.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1049/gtd2.12926
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
