Linear actuator is a device that converts the rotational motion of a motor into linear motion, widely used in various automation systems. However, linear actuators may malfunction during operation, affecting the performance and safety of the system. Therefore, fault diagnosis of linear actuators is very important. This article introduces a data-based linear actuator fault diagnosis method and an adaptive observer based linear actuator fault diagnosis method.
A data-based fault diagnosis method for linear actuators
This method utilizes the current and position measurement data of linear actuators to extract features and construct fault indicators. The advantage of this method is that it does not require a physical model or additional sensors, but only utilizes the data already present in the controller itself. The specific steps are as follows:
Extract features from current and position measurement data. The characteristics include the dynamic characteristics of the system during steady-state and transient operation, such as current mean, current peak, position mean, position peak, position change rate, etc.
Combine features into fault indicators. The fault indication quantity is a numerical value that reflects whether there is a fault in the system and the severity of the fault. A commonly used method is to use Support Vector Machine (SVM) to classify features and obtain discriminant functions for normal and fault states.
Set the threshold for fault indication. The threshold is a numerical value used to determine whether the fault indication exceeds the normal range. The threshold can be determined based on historical data or experience, or dynamically adjusted based on real-time data or statistical methods.
Perform fault detection and diagnosis based on the fault indication quantity and threshold. If the fault indication exceeds the threshold, it is considered that the system has malfunctioned, and the type and location of the fault are determined based on the size and trend of the fault indication.