| Title |
DC-Link Capacitance Estimation using Support Vector Regression in AC/DC/AC PWM Converters |
| Authors |
아보칼릴 아메드(Ahmed G. Abo-Khalil) ; 장정익(Jang, Jeong-Ik) ; 이동춘(Lee, Dong-Choon) |
| Keywords |
Electrolytic capacitor ; PWM converters ; capacitance estimation ; SVR |
| Abstract |
This paper proposes a new capacitance estimation scheme for a DC-link capacitor in a three-phase AC/DC/AC PWM converter. A controlled AC voltage with a lower frequency than the line frequency is injected into the DC-link voltage, which then causes AC power ripples at the DC side. By extracting the AC voltage and power components on the DC output side using digital filters, the capacitance can then be calculated using the Support Vector Regression (SVR). By training of SVR, a function which relates a given input (capacitor's power) and its corresponding output (capacitance value) can be derived. This function is used to predict outputs for given inputs that are not included in the training set. The proposed method does not require the information of DC-link current and can be simply implemented with only software and no additional hardware. Experimental results confirm that the estimation error is less than 0.16%. |