Abstract

On-line condition monitoring system for railway hydraulic shock absorbers is presented. The shock absorbers are used for controlling the yaw movement of the bogie, highspeed ICE II trains. Uncontrolled yaw (hunting) movement can cause the instability of the bogie, which may lead to the derailment of the train. The hydraulic shock absorbers are the devices widely used in automotive and railway applications. They are the subject of regular servicing. The railway shock absorbers are tested in the special testing rigs after dis-asseblying from the train. The testing procedure is very expensive and time consuming and generates high idle cost. Application of the condition monitoring system of the anti-yaw shock absorbers can increase the safety and reduce the maintenance costs of the train. The train operator will be informed about the condition of shock absorbers, thus the shock absorbers will be replaced if needed, not according to the mileage schedule. The thesis deals with the condition monitoring system for the choice of the measured variables, which describe the properties of shock absorbers. The choice is based on numerical and laboratory experiments. The testing has been performed on the dampers with simulated failures. The performance of shock absorbers can be described using the graphs on the phase plane, combining the force acting on the piston rod and the velocity of the movement between piston and cylinder. The artificial neural networks (ANN) have been successfully applied in the inference engine. The thesis presents the results of the learning of the neural networks and the results of validation testing.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call