• Journal of Institute of Control, Robotics and Systems
• /
• v.8 no.10
• /
• pp.851-855
• /
• 2002

This paper proposes a method of fault detection and diagnosis of agitators based on the wavelet analysis of the current and vibration signals. The wavelet transform has received considerable interest in the fields of acoustics, communication, image compression, vision. and seismic since it provides the fast and effective means of analyzing signals recorded during operation. Neural network is used to diagnose the fault. Specifically, the proposed approach consists of (i) fault detection, (ii) feature extraction, and (iii) classification of fault types. The results show an effective application of the wavelet analysis on the monitoring of an agitator.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.25 no.9
• /
• pp.1341-1349
• /
• 2001

In this paper, 3 D.O.F. actuator, which has three degrees of freedom in one joint, is proposed. The proposed 3 D.O.F. spherical actuator is composed of the rotor and atator. The upper plate of the stator supports the rotor and five electromagnets are located at the base of the stator. The rotor has two permanent magnets, and each rotational axis of the rotor gimbal system is supported by the bearing. To find out the governing equations for the torque generation, Coulombs law and Lorentz force with respect to magnetism is applied. As the experimental results, if the distance between electromagnet and permanent maget is far enough, the force between these magnets can be expressed from current of coils and z-axial distance. For the purpose of control 3 D.O.F. actuator, PID control law is applied. The experimental results are presented to show the validity of the proposed 3 D.O.F. actuator.