STOCHASTIC POSITIVE-REAL BALANCED TRUNCATION: A NOVEL MODEL ORDER REDUCTION TECHNIQUE

Abstract

In the context of the digital revolution, the development of low-order model reduction techniques for electrical and electronic circuits is imperative. In this paper, we introduce a novel algorithm called Stochastic Positive-Real Balanced Truncation (SPBT) based on the combination of two techniques: Positive-Real Balanced Reduction (PRR) and Stochastic Balanced Reduction (SBR) to reduce the model order. From the experimental results and comparisons of SPBT with PRR and SBR methods on electronic circuit models, it is evident that SPBT not only preserves the stability, passivity, and minimum phase properties of the original system but also provides higher performance and accuracy compared to PRR and SBR. With low errors and matching responses between the reduced-order system and the original system when applying SPBT in the time and frequency domains, this method holds promise as an effective tool for reducing complexity while maintaining the essential physical characteristics of the system.