Feasibility of solar energy in south sweden: artificial neural network modeling
Paper i proceeding, 2016
Observations provide evidence that atmospheric greenhouse gases concentration has increased rapidly over last century. This leads to extreme climate changes such as heat waves, rising sea-levels, changes in precipitation resulting in flooding and droughts, intense hurricanes, and degraded air quality, affect directly and indirectly the physical, social, and psychological health of humans. For that reasons, and in helping achieving the EU targets for 2020 and 2050, utilizing the available local renewable energy resources is needed. Although the current efficiency of PV system (PVs) is still relatively low and the capital cost is still high, the abundance of solar energy that strikes the Earth continuously makes the photovoltaic systems viable alternative. The current work, therefore, investigate the potential of utilizing solar energy for electricity generation in Europe. For this aim, a residential building in Landskrona, Sweden, was chosen as a case study. Solar World SW325 XL, which is a monocrystalline module, was selected as PV panel. A computer model was built to simulate grid-connected rooftop PV system in which the module elements are attached to the roof of the building. Sensitive analysis was carried out to test the robustness of the simulation results. Performed calculations show that there is a big potential to use PVs with 193 kWh/y electricity can be generated per square meter of PV. The payback time of the systems is 6 years with levelized cost of electricity is 10.3 C/kWh. Finally, artificial neural network (ANN)-base model was built to generate user-friendly formula that states the relationship between the NPV (net present value) of PVs and specific factors of interest. These factors, including electricity price, real interest rate, module price and inverter price, were chosen based on sensitivity analysis results. The study leaded to create a simple formula that can easily be used to estimate the NPV of PVs without use of complicated software.