Vibration control is critically important for engineering equipment, and in modern industrial engineering active strategies with robust performance are often adopted. In traditional studies, a single-objective consideration is often taken into account when robust control is performed, while a simultaneous multi-criterial consideration is ignored. The study outlined in this paper focuses on typical equipment, namely machinery and sensitive equipment. Meanwhile, evaluation of robust performances based on feedback control is considered as the vibration control objective, and performance indexes using H∞ and H2 criterion are regarded as fitness functions. In addition, the latest intelligent algorithm – MOPSO (multi-objective particle swarm optimization) is used and the SPEA2 (strength Pareto evolutionary algorithm 2) is also introduced for comparison as a representative of evolution algorithm. Numerical results show that the Pareto frontier of MOPSO is much smoother and more uniformly distributed than SPEA2, and even more important is that MOPSO can obtain a unique, global and optimal solution gbest, which can avoid having to select just one from a group of equivalent solutions Finally, an analysis of factors which affect the norms is performed, and the numerical verification shows that the disturbance type (single input or multi input) can apparently affect the magnitude of norms, and this finding can provide a broader understanding of robust vibration control. This research proposes a novel multi-objective optimization strategy for robust vibration control, while the traditional approaches can and are still employed. In addition, advanced artificial intelligence plays an important role in vibration detection in engineering application.