Abstract With the automation development of excavator, fault detection for hydraulic system has become one of the key techniques of modern excavator. By analyzing the principal of hydraulic system, this paper presents an approach to detect faults of the excavator’s hydraulic system. This approach consists of three steps: process modeling, principal components and residual generation, and fault detection. Several typical faults of hydraulic system were considered in this paper and experimental data are used to validate the approach. The experimental results show that the proposed fault detection approach is high-efficiency and reliable estimator of hydraulic system and could effectively applied to the excavator’s hydraulic system.