Optimized Real-Time Stochastic Model of Power Electronic Converters based on FPGA

Shinan Wang,Xizheng Guo,Zonghui Sun,Yule Wang,Xiaojie You
DOI: https://doi.org/10.1145/3678174
IF: 0.851
2024-07-17
ACM Transactions on Modeling and Computer Simulation
Abstract:Stochastic models can effectively describe the operating characteristics of power electronic converters with stochastic parameters. However, it is difficult to implement the models in FPGA-based real-time simulation (RTS) because their high order leads to a large calculation. This paper proposes an optimized real-time stochastic modeling method for power electronic converters based on generalized polynomial chaos (gPC). Firstly, an orthogonal polynomials construction method is used based on Schmidt orthogonalization to describe stochastic variables with atypical probability distributions and provide conditions for simplifying the system model. Secondly, the method of probability space transformation is adopted to divide the system model into multiple sub-models to suppress the exponential growth of the model order while maintaining the statistical properties. This method has performed over the traditional stochastic modeling method. The proposed model is built on FPGA-based hardware-in-the-loop (HIL) experiment platform with 1us simulation step. The optimized model uses approximately 37% fewer resources than the traditional stochastic model while maintaining the same level of accuracy.
computer science, interdisciplinary applications,mathematics, applied
What problem does this paper attempt to address?