LAUSR

The performance of a sensorless predictive current controlled boost converter is usually susceptible to the accuracy of the employed converter and current estimator models. Since the accuracy of conventional current estimator would be degraded due to parasitic parameter variations, a charge balance current estimator (CBCE) is proposed to overcome this issue. Its estimation of peak inductor current is based on the charge balance principle of output capacitor and characteristics of inductor current ripple. Only one parasitic parameter - capacitor equivalent series resistance (ESR) is used in CBCE. Furthermore, the capacitor ESR is also estimated together with output load by a proposed equivalent load estimation scheme which can improve the current estimation accuracy. Besides, to achieve a high dynamic performance digital control, a discrete-time model considering current ripple is developed, in which the ripple is taken into account rather than treats inductor current as a constant in each switching cycle. It can achieve a higher accuracy than conventional averaged model, especially in the high-frequency region. Finally, its effectiveness is verified by simulation and experiment results.