Digital control strategies for EMI reduced switching of smart power switches

Smart Power Switches are widely used in automotive applications. Due to their rapid and periodic switching, Smart Power Switches generate electromagnetic interferences which have to be limited to fulfill the demands on electromagnetic compatibility. One possibility to reduce these interferences is by the control of the switching transition. Therefore, this thesis deals with the development of digital control concepts for the switching transition control of Smart Power Switches. First, a mathematical model of the Smart Power Switch is developed. This model is used for simulation studies and for the controller design. Second, the influence of the switching transition on the generated electromagnetic interferences is investigated. Based on this investigation, different control strategies are introduced which all use feedforward gate current profiles as control inputs. In the first strategy, the profiles are designed to limit and control the slew rate of the switching transition and are determined by numerically solving an optimal control problem. The second design aims at tracking a desired Gaussian-shape switching transition and the third design limits and controls the first- and/or second-order derivatives of the switching transition. In the second and third design, the feedforward profiles are iteratively adapted from switching cycle to switching cycle by an Iterative Leaning Control strategy to compensate for model uncertainties and load variations. The performance of the control concepts are analyzed by test bench measurements in the time and iteration domain. Finally, the electromagnetic compatibility is tested according to the CISPR 25 standard. The test proves that the electromagnetic interferences can be reduced by the proposed control concepts.