Open AccessGenetic Algorithm Based Optimum Solar Power Prediction by Environmental Features for Indian Railway Stations
Author(s) -
Sanjeev Kumar Sukalikar*,
Dr S.R Awasthi
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.d7278.118419
Subject(s) - genetic algorithm , particle swarm optimization , solar energy , solar power , power (physics) , correlation coefficient , computer science , algorithm , photovoltaic system , roof , mathematical optimization , statistics , engineering , mathematics , machine learning , civil engineering , electrical engineering , physics , quantum mechanics
Electric Power requirement is increasing with every new day. Bulk of electric power in India is consumed by Indian railways, hence green energy resources are required to reduce this load. Out of many available green energy resources solar energy is playing an important role. Hence desired optimum power output from solar plant is always of area of research. This work focuses on estimation of solar plant output with the affecting environmental variables by using genetic algorithm. Genetic algorithm predicts the ratio of environmental variables that directly increase or decrease the solar power output. Work will determine this ratio by using modified butterfly particle swarm optimization algorithm and Teacher Learning Based Optimization. Here it was obtained that proposed TLBO model is good for estimating monthly power prediction while MBAPSO estimate accurate values for daily power prediction. Proposed model experiment was done on dataset of Raipur city in Chhattisgarh state of India and the results show reduction in MAE, RMSE and improvement in Correlation Coefficient thereby accuracy in power prediction when compared with ground truth values taken from Indian railway top roof solar installation.