A Robust Inductance Estimation Method for Model Predictive Control of Grid-Connected Inverters

The inductance parameter is crucial to realize high-precision model predictive control (MPC) for grid-connected inverter (GCI), while the traditional inductance estimation method is sensitive to the grid frequency deviation and cannot work normally when the active power is zero. To solve these problems, a new inductance online estimation method based on a full order sliding mode voltage observer (FSMVO) is proposed. First, the drawbacks of the traditional FSMVO-based inductance estimation method are analyzed. Second, a new inductance estimation method is proposed, which uses the error between the observed grid voltage and the actual one passed through a filter, whose transfer function is the same as the one of FSMVO. As a result, the new inductance estimation method is immune to the grid frequency deviation. Moreover, a new Lyapunov function is also designed in this article to deduce the new inductance estimation method, which can work well, no matter the active power is zero or not. Finally, by substituting the estimated inductance parameter into the MPC algorithm, the control performance of GCI can be improved. The effectiveness and accuracy of the proposed inductance estimation method are verified by experimental researches.