Assessment of SPV system using ANN and VHDL

The number of engineering applications using artificial neural network has increased considerably in the last few years. However, the ANN application in photovoltaic systems is very limited. This paper presents the SPV (solar photovoltaic) system assessment based on ANN and VHDL. Experimental database of meteorological parameters like temperature and daily global solar irradiance for various Indian stations are used to assess the SPV system performance. The inputs of the ANN are the daily total irradiance and mean average temperature while the outputs are the current and voltage generated from the system. Subsequently, the neural network corresponding to SPV system is implemented using VHDL language based on the saved weights and bias of the network. The given model based on ANN and VHDL permit to evaluate the performance of SPV system using only the meteorological parameters and involves less computational efforts, and it can be used for predicting the output electrical energy from the system.