A noninvasive on-line failure prediction technique for aluminum electrolytic capacitors in photovoltaic grid-connected inverters

Since the electrolytic capacitors are mainly responsible for breakdowns in power electronics converters, they usually determine the overall lifetime. Their failure is due to the deterioration of their dielectric, the production of gases, and, eventually, their explosion. The deterioration process leads to a decrease in its capacitance value and simultaneously, an increase in the capacitor equivalent series resistance (ESR). In this paper, a novel noninvasive technique for failure prediction of the electrolytic capacitors in grid-connected photovoltaic inverters is presented, which can easily be applied online and even in the real time. The technique is based on the DC-link voltage measurement, current calculation, and extracting the current integral and the voltage change during switching intervals. Then, the capacitance can be calculated using the recursive least squares method. Although the estimation results are influenced by the temperature at which the measurement is performed, and the results are dependent on the operating conditions, but the proposed method is modified to obviate the above distorting factors effectively. The proposed technique is simple to apply, accurate and does not need to any extra hardware. Also it can be adapted for any other inverter based applications. The performance of the proposed method is verified by the simulation and experimental results.