Performance enhancement of various bos of solar photovoltaic system

Abstract

The Balance of System (BOS) in a Solar Photovoltaic (PV) system newlineconstitutes whole pack of blocks/modules which balances the power generated newlineby a PV array with the power consumed in the load side. The BOS includes all newlinethe main blocks PV array, Battery, Charge Controller and Inverter in a newlinestandalone PV system. The primary objective in this research is to enhance the newlineperformance of PV, battery and inverter modules in a balance of solar newlinephotovoltaic system. newlineRegarding the performance enhancement in PV module, in the first newlineproposed work voltage/current mismatch is analyzed during fault panel newlinereplacement. Barely one or two PV panels demand for replacement instead of newlinethe entire PV array because of some unexpected damages/faults occurs within newlinelesser years of installation. To compete with the emerging trends in solar market, newlinemanufacturers upgrade their products frequently results in the unavailability of newlineexact same panel for replacement. The replaced panel may vary either in the newlineratings or with the type. Based on the ratings/type the replaced panels are newlineclassified and analyzed under five cases : i) Under rated same type ii) Over rated newlinesame type iii) Under rated different type iv) Over rated different type v) Same newlinerated different type. Test results are carried out for 3*3 PV array under five cases newlinefor without and with changing in position. The performance parameters like newlinemaximum power achievement, power loss and fill factors are measured and its newlinevoltage/current mismatch is calculated to choose the better matching panel in the newlinereplacement position. The second proposed work is to detect and classify the newlinehealthy, hotspot and micro crack faults using Feed Forward Back Propagation newlineAlgorithm FFBPNN (based on Artificial Neural Network ANN) and Support newlineVector Machine SVM techniques (based on Machine Learning). newline