Pemodelan dan Prediksi Daya Ouput Photovoltaic secara Real Time Berbasis Mikrokontroler

Abstract

The electrical energy generated by the photovoltaic (PV) as a renewable energy source highly affected by environmental conditions such as intensity of sunlight irradiance, temperature, geographic location and tilt angle of PV itself. How much power should be generated by the PV for every times and anywhere will be discussed in this paper. This paper are implemented models of clear sky solar irradiance, solar position and PV temperature to predict the power output should be generated by PV. The model is implemented on the ARM Cortex M4F microcontroller STM32F407 which is a 32bit microcontroller and equipped with DSP, so the prediction of PV power output can be done online and in real time. To be able to predict the PV power output online, at any time and wherever they are, in this study microcontroller equipped with temperature sensors and input geographical information (latitude-longitude) and also equipped with a memory card for data logger between the predictions and field measurement. Results have been obtained by field experiments, measurements test for PV is very close to predictions with an average error 4.72% and computation time for all models by microcontroller with DSP instruction 33.64% faster compare to without DSP instruction.

Keywords

• Photovoltaic
• Real time power prediction and Microcontroller