We are happy to announce that our paper got accepted and published at the 10th International Workshop on Energy Harvesting & Energy-Neutral Sensing Systems (ENSsys 2022)
- Title: PROGNOES: Prediction of haRvestable sOlar enerGy based on suN irradiatiOn and wEather conditionS
- Authors: Sayedsepehr Mosavat, Pedram Golkar, Matteo Zella, Pedro José Marrón
- Abstract: Energy harvesting, batteryless devices can mitigate many shortcomings of battery-operated devices in the coming years, particularly in the area of the Internet of Things. Solar energy is an accessible and effective source of power for such devices. However, the uncontrollable nature of solar energy makes the task of designing and evaluating energy harvesting devices a challenging one. In this work, we propose PROGNOES, a tool that allows the user to forecast the harvestable solar energy by a particular solar cell at a specific time and location. To carry out the forecasts, various algorithms and models are used not only to provide the theoretical solar irradiation but also to provide the weather condition data
corresponding to the time and location of the forecasting. PROGNOES can generate forecasts in the form of IV curves, similar to those generated by physical solar cells. Such IV curves can then be emulated with suitable tools to provide energy traces to the energy harvesting devices at design time, therefore facilitating the task of designing robust, reliable energy harvesting, batteryless devices. In our experiments, we find that the energy forecasts carried out by PROGNOES can reach a mean absolute error of 7.16 mW.