Anish Sojan's repositories
bit-coin-prediction
Prediction of Bitcoin Close Price using machine learning algorithms in R.
anime-series-data-warehouse
R-script codes to extract, transform and create comma separated files for data extracted from three different sources.
hello-world
My first git-hub project
Ireland-Solar-Irradiance
According to the Irish Solar Energy Association (ISEA) around two hundred large-scale solar farms have received planning permissions across the country with investments exceeding a billion euros. Ireland, however, is trailing in its target to meet the carbon reduction standards set by the European Union that must be met by 2020. Ideal locations to set up solar farms is a boon when calculating the amount of investment required to harness the photovoltaic potential of the area. This project utilized CAMS-Radiation Service solar irradiance measures (global, direct, and diffuse) under normal and clear-sky conditions to detect ideal locations with photovoltaic potential for 26 counties in Ireland. A combination of k-means clustering with support vector machine(SVM) and a gradient boosted decision tree (GBDT) was used to detect ideal locations, which resulted in a detection rate of 99%. The seasonal approach towards location detection provided better location detection rates in summer (SVM-95% and GBDT-98%) with autumn (SVM-94%, GBDT-96%) providing the lowest. This can help electricity firms in Ireland, such as the Electricity Supply Board, to plan, invest and strategize immeasurably while setting up solar farms or domestic installations.