ABSTRACT
Global energy crises has been a huge challenge which causes the need for diversification in order to be able to satisfy the increasing demand. In this work, we developed optimal production and maintenance planning strategies on two (2) techniques to; the support vector regression (SVR) and the theoretical techniques for a photovoltaic (PV) system. Optimization of the production planning was performed to determine the minimal number of PV panels and battery storage required to satisfy the demand. After successful production planning, an integrated maintenance model was the developed to determine the cost for optimal number of maintenance actions to be performed on the system over a production horizon. A case study of Sokoto power plant in Nigeria was considered to validate the algorithm. The two results were compared to determine the best technique for the PV system management. The results obtained presents the machine learning (SVR) approach as the most suitable technique of the two considered.
