Co-simulation is widely used in the industry for the simulation of multidomain systems. Because the coupling variables cannot be communicated continuously, the co-simulation results can be unstable and inaccurate, especially when an explicit parallel approach is applied. To address this issue, new coupling methods to improve the stability and accuracy have been developed in recent years. However, the assessment of their performance is sometimes not straightforward or is even impossible owing to the case-dependent effect. The selection of the coupling method and its tuning cannot be performed before running the co-simulation, especially with a time-varying system.

In this work, the co-simulation system is analyzed in the frequency domain as a sampled-data interconnection. Then a new coupling method based on the H-infinity synthesis is developed. The method intends to reconstruct the coupling variable by adding a compensator and smoother at the interface and to minimize the error from the sample-hold process. A convergence analysis in the frequency domain shows that the coupling error can be reduced in a wide frequency range, which implies good robustness. The new method is verified using two co-simulation cases. The first case is a dual mass–spring–damper system with random parameters and the second case is a co-simulation of a multibody dynamic (MBD) vehicle model and an electric power-assisted steering (EPAS) system model. Experimental results show that the method can improve the stability and accuracy, which enables a larger communication step to speed up the explicit parallel co-simulation.

协同仿真在业界广泛用于多域系统的仿真。由于耦合变量无法连续通信，因此联合仿真结果可能不稳定且不准确，尤其是在应用显式并行方法时。为了解决这个问题，近年来已经开发了用于提高稳定性和准确性的新的耦合方法。但是，由于案例相关性的影响，有时对它们的性能进行评估不是很简单，甚至是不可能的。在运行协同仿真之前，尤其是在时变系统中，无法进行耦合方法的选择及其调整。

在这项工作中，协同仿真系统在频域中作为采样数据互连进行了分析。然后，开发了一种基于H-无穷大合成的新耦合方法。该方法旨在通过在接口处添加补偿器和平滑器来重构耦合变量，并最大程度地减少来自采样保持过程的误差。频域上的收敛分析表明，耦合误差可以在较宽的频率范围内减小，这意味着良好的鲁棒性。使用两个联合仿真案例验证了该新方法。第一种情况是带有随机参数的双质量弹簧减振器系统，第二种情况是多体动态（MBD）车辆模型和电动助力转向（EPAS）系统模型的共同仿真。实验结果表明，该方法可以提高稳定性和准确性，