Sigmoid-Function-Based Feedback Linearization Random Selection Control for Cascaded H-Bridge Multilevel Inverters

To improve the steady-state and dynamic performance of cascaded H-bridge multilevel inverters (CHBMIs) and achieve power balance, this paper proposes a control method based on the sigmoid function that combines feedback linearization with a random selection strategy. First, a continuous-smooth sigmoid function model that can uniformly describe all switching-state combinations is established. Subsequently, the control law is derived by applying the feedback linearization theory. The output levels are determined based on the control law and the characteristics of the sigmoid function. Finally, the desired switching-state combinations are determined through two random selections and applied directly to the switching devices. The proposed method is not only applicable to symmetric CHBMIs but also applicable to asymmetric CHBMIs. Simulations and experimental verifications are carried out on the control systems of a symmetric CHBMI with a linear load and an asymmetric CHBMI with a nonlinear load, and comparisons are made with other methods. The results show that the proposed method exhibits better steady-state and dynamic performance, achieving a higher power balance degree, and is suitable for both symmetric and asymmetric CHBMIs.