Solar panels consist of photovoltaic cells which are identical to diode semiconductor devices. Photovoltaics have a special point which is usually called MPP (Maximum Power Point). At that point, the photovoltaic is in an optimal state, both the voltage and current produced. The location of the MPP is unknown but can be found using calculations or a tracking algorithm called Maximum Power Point Tracking (MPPT). In general, MPPT uses the observe and observe algorithm which has shortcomings in determining the maximum power point because it is often trapped at the local maximum point. In this research, the MPPT algorithm used is based on a neural network with a buck-boost type converter. This method is done without requiring detailed information about photovoltaics. This MPPT will be used as a trainer kit for a simulator for identifying solar cell characteristics in the PPNS power electronics laboratory. The MPPT that has been created has an efficiency of between 80-92% and can increase the average power by 8-11.36% for monocrystalline solar panels and 1-13.05% for polycrystalline solar panels depending on the load used.
Copyrights © 2024