Servo motor fault detection for construction robot base on observer

In this paper, a fault diagnosis scheme based on observer is proposed to solve the problem of servo motor fault in construction robot actuator system. Firstly, the mathematical model is established in abc coordinate system, and Park transformation is introduced to transform abc three-phase symmetry variable into dq coordinate DC through Park transformation, which greatly reduces the modeling complexity in normal state of motor, eliminate redundant motor models. Based on this, the state space equations are constructed, and an interval observer is designed. Compared with the traditional Luenberger interval observer, the interval observer proposed in this paper has better decoupling performance in solving the gain matrix, at the same time, it directly improves the robustness of the interval observer and the sensitivity to the fault signal under the electromagnetic interference. Finally, the method is verified on the fault simulation testbed, which proves the feasibility of the proposed method, and further improves the timeliness of fault detection of servo motor in construction robot actuator system.