Monitoring PWM signals in stand-alone photovoltaic systems

Abstract

The performance of stand-alone photovoltaic (SAPV) systems can be evaluated by monitoring them in the field using data acquisition systems (DASs). Most SAPV systems use battery charge controllers with pulse width modulation (PWM) to regulate the current into the battery. The PWM signals generated by battery charge controllers imply monitoring challenges due to the complexity of this type of signal. In this sense, the aim of this paper is to develop a new and simple monitoring technique for SAPV systems which can estimate the signals provided by a PWM battery charge controller, thus avoiding expensive DASs, simultaneous sampling and the huge amount of collected data. The estimation of PWM signal parameters, such as the duty factor (df) or high and low states, shows high accuracy, with the mean absolute percentage error lower than 1.4%, a mean relative error within 1.4%, and the coefficient of determination higher than 0.9. Furthermore, the proposed technique may easily be used for other electrical devices where PWM is employed.