Model-Based Fault Detection Method for Coil Burnout in Solenoid Valves Subjected to Dynamic Thermal Loading

Abstract

Solenoid valves are widely used to control fluid flow in various mechanical systems. If the valves do not function properly, the mechanical systems can lose their ability to control the fluid flow. This paper describes a fault detection method that can monitor coil burnout under dynamic thermal loading. The method consists of three steps. First, an equivalent current model of the solenoid valves is derived from Kirchhoff's voltage law. Then, a predictive regression model is developed to describe the relationship between the electric current and the dynamic change of operating temperature of the valves. Finally, a health indicator of solenoid coil burnout is devised in conjunction with the derived model. To demonstrate the validity of the proposed fault detection method, a case study is presented with solenoid valves taken from real braking systems of urban railway vehicles. The case study confirms that the proposed method can detect the coil burnout independent of the operating temperature. We anticipate that the proposed method can be widely applicable to diagnose any electromagnetic actuator of engineered systems as well as solenoid valves in various industrial applications.

Keywords

• Electromagnetic actuators
• coil burnout
• condition-based maintenance
• fault detection
• regression