PROJECT TITLE :

Robust Big Data Analytics for Electricity Price Forecasting in the Smart Grid - 2017

ABSTRACT:

Electricity value forecasting is a vital part of sensible grid as a result of it makes good grid cost economical. Nevertheless, existing ways for price forecasting could be tough to handle with huge price data in the grid, since the redundancy from feature selection can not be averted and an integrated infrastructure is also lacked for coordinating the procedures in electricity price forecasting. To solve such a downside, a novel electricity price forecasting model is developed. Specifically, three modules are integrated in the proposed model. Initial, by merging of Random Forest (RF) and Relief-F algorithm, we have a tendency to propose a hybrid feature selector based mostly on Grey Correlation Analysis (GCA) to eliminate the feature redundancy. Second, an integration of Kernel operate and Principle Element Analysis (KPCA) is utilized in feature extraction method to appreciate the dimensionality reduction. Finally, to forecast value classification, we place forward a differential evolution (DE) primarily based Support Vector Machine (SVM) classifier. Our proposed electricity value forecasting model is realized via these 3 components. Numerical results show that our proposal has superior performance than alternative ways.