Machine learning based solar energy forecasting and wind-solar based hybrid grid arrangement at Patenga coastal area, Bangladesh

Rapid developing country like Bangladesh will encounter a major problem in finding the sources for electricity generation as conventional sources of power are getting diminished due to shortage of fossil fuels. Renewable energy systems (RES) are the ultimate choice of future energy meet. RES like wind turbine generator (WTG) and solar photovoltaic generator (SPVG) are weather dependent and weather is a variable parameter. Prediction of weather in quick time is needed to study the nature of hybrid system. Present paper targets a potential location called Patenga, which is one of the coastal areas of Bangladesh. The annual average wind speed at 20 meter height is 7.48 ms
⁻¹
and solar radiation is 5.85 kWh/m2/day. This paper will work on modelling of WTG, SPVG and analyzes the output of generators with the variation of weather as their respective source of energy. Weather of a particular day is extrapolated for solar irradiation using common machine learning approaches. The grid power response is also observed to find the overall response with RES of a day. All the simulations of the work are carried out by NEPLAN software.