Abnormal Event Detection in Electric Servo Press Systems through Integrated Motion Control and Sensorless Force Estimation

Abstract

This paper presents a hybrid motion-control framework for electric servo presses that integrates robust position control, sensorless force estimation, and abnormal-event detection. While electric servo presses offer superior motion programmability and energy efficiency compared to conventional hydraulic presses, their safe operation under unexpected contact events and large parameter variations remains a critical challenge. To address this issue, a disturbance-observer-based control architecture is developed, in which a disturbance observer (DOB) is employed to compensate for model uncertainties and friction effects, and a sensorless force observer (FOB) is designed to estimate the external force acting on the punch using motor-side signals only. The proposed framework provides a unified control structure for press operations under both nominal and abnormal conditions without relying on dedicated force sensors. Robust stability of the closed-loop system under contact-induced damping variations is analytically examined using the small-gain theorem. The validity of the proposed system is demonstrated through experimental studies conducted on a prototype servo-press platform.