An Optimal Simulation Configuration Method for Real-Time Simulation Based on Simulation Requirements

Abstract

An optimal configuration method of simulation parameters, including the solver and step size of each sub-model, is proposed to realize real-time virtual simulation. The content of the presented study is threefold: firstly, the simulation requirements of three simulation modes, including Model-Exchange, Co-Simulation, and Hybrid Co-Simulation, are analyzed; secondly, the mathematical descriptions of the three simulation configuration problems are constructed, which are classified as an inequality problem and two assignment problems, where the assignment problems are decoupled; finally, an optimal simulation configuration procedure is proposed, where the analysis method and corresponding design methods for configured simulation parameters of each sub-model are given. The configuration method is applied to configure the simulation parameters of an aeroengine control system model with 6 separate components, and the simulation results verify the effectiveness of the proposed method. Specifically, the feasibility analysis results of the sub-models reveal that 4 of them meet the requirements while 2 of them are numerically unstable, which are consistent with the verification results. Meanwhile, after redesigning the infeasible sub-models using the proposed design methods, all the sub-models can meet their simulation requirements. In addition, an optimal configuration of the simulation parameters as well as several comparative configurations are provided, and the simulation results show that the time consumption of the optimal configuration is minimal and it can achieve real-time performance.