PROJECT TITLE:

Neural Network-Based Model Design for Short-Term Load Forecast in Distribution Systems - 2015

ABSTRACT:

Accurate forecasts of electrical substations are obligatory for the efficiency of the Advanced Distribution Automation functions in distribution systems. The paper describes the planning of a category of machine-learning models, specifically neural networks, for the load forecasts of medium-voltage/low-voltage substations. We tend to focus on the methodology of neural network model style so as to obtain a model that has the best achievable predictive ability given the obtainable information. Variable selection and model selection are applied to electrical load forecasts to confirm an optimal generalization capability of the neural network model. Real measurements collected in French distribution systems are used to validate our study. The results show that the neural network-primarily based models outperform the time series models and that the look methodology guarantees the best generalization ability of the neural network model for the load forecasting purpose based mostly on the identical data.