Assessment of Aluminum Electrolytic Capacitors Health Status Through Signal-Based Techniques

Aluminum electrolytic capacitors (AECs) are essential for the operation of power converters; however, they stand out as one of the most susceptible components. Consequently, there is a pressing need to develop fault diagnosis techniques (FDTs) to assess the AECs health status during converter operation.With aging, certain AECs characteristics undergo changes resulting from electrolyte loss. These changes are evidenced by the decline in the AECs capacitance (C) and the increase in their equivalent series resistance (ESR). Therefore, ESR and C are frequently employed as indicators to assess the health condition of AECs.Several FDTs have been proposed to estimate AECs health status, many of which are invasive, requiring the introduction of sensors inside the converter. A commonly used sensor is the capacitor current sensor, which imposes practical restrictions, namely requiring space and introducing an undesirable resistive and inductive effect. It is therefore essential to propose solutions that avoid this sensor.The proposed approach eliminates the requirement for integrating sensors within the converter. It demands the measurement of the inductor currents, essential for the controller, and the converter’s output voltage. By employing two 5^th order Butterworth filters and extracting fundamental features from the preceding signals, it becomes feasible to estimate AECs equivalent series resistance value through a direct equation.