Effects of switching-time uncertainties on pulsewidth-modulated power converters: modeling and analysis

This paper introduces a stochastic model for pulsewidth-modulated (PWM) switching dc-dc boost converters as an enhancement to conventional deterministic models. The important benefit of the stochastic model is that it includes effects of often ignored disturbances and their sources such as measurement error, ambient temperature, and parasitics, which are known to be present in power-electronic switching converters. In this approach, the stochastic model is based on introduction of perturbations in the duty ratio of switching converters as random-noise processes that account for these disturbances. The system dynamics are modeled as a diffusion process. Mean first-passage time (MFPT) is defined as a performance index. It quantifies the average time it takes for a state-space trajectory to evolve from a given operating point to the boundary of its domain of attraction under the influence of small perturbations. It is shown that MFPT provides further insight into converter operation under random perturbations.