Shoji Hayashi

Study of machine fault diagnosis system using neural networks

Develops a machine fault diagnosis system using neural networks and spectral analysis. A neural network is applied to the fault diagnosis of the machine. The neural network has learning and memory capability. By the learning of normal and abnormal states of the object system, a method with neural networks is proposed which can diagnose a fault of the machine. The proposed fault diagnosis system is based on the spectrum of vibrations or sounds obtained from the operating machine. The difference between normal and abnormal data becomes clearer when comparing time series data. It is suitable for the detection of the fault to utilize changes of spectral data. Using this method, it is shown that it can detect unknown fault patterns. Fault diagnosis experiments are performed on both a wood slicing machine and an electromagnetic valve. The possibility of an online fault diagnosis system is examined through the construction of an online data processing system for an electromagnetic valve and it is shown that the fault diagnosis can be performed in real time. Through these results, the effectiveness of the proposed fault diagnosis system is verified.

Fault diagnosis system using vector quantization and its application to electromagnetic valve

This paper is concerned with fault diagnosis using vector quantization of sound signals and its application to electromagnetic valves. Instead of establishing a mathematical model, the vector quantization directly extracts characteristic vectors from sample data by optimized algorithms and uses these characteristic vectors as standard patterns to make fault diagnosis. An application to electromagnetic valve verified the effectiveness of this new method.

Trial production of a vane-tipe pneumatic linear actuator.

A specially constructed pneumatic actuator has been manufactured on a trial basis. This actuator is meritorious in that it does not have a rod such as an air cylinder and has a considerably long stroke. The principle of this actuator is similar in construction to air motors, and the driving object mounted on a vane moves along a grooved bench. The groove base is n the shape of a rack with wide pressure angles. The air pressure is supplied to the space which is composed of the vane and rack, and then the driving object is moved by vane-applied air pressure. The air leakage passing through the narrow slits of the actuator must be avoided to obtain good moving performance. This report has investigated both theoretically and experimentally the processing accuracy of the device to obtain excellent performance considering the air leakage from the narrow slits.

Bearingless water mill-type liquid flowmeter.

This paper is concerned with a study of a liquid flowmeter by applying the principle of the bearingless windmill-type gas flowmeter already developed by the authors. The float equipped with a water mill is floating on the surface of liquid in a bell-type enclosure, and the position of the float is held on the center by a magnetic attraction. Since the liquid is Newtonian, the friction between the float and the liquid is expected to be negligible when the relative speed is near zero. Thus the flowmeter enables a extremely flow flow rate measurement. The rotating speed of the water mill by a liquid flow is detected by means of opto-electronic counting procedures without any friction on the rotation, and from which the flow rate of a liquid is obtained. The flowmeter proposed here shows a linearity over the wide range of the flow rate from 1×10-7 to 5×10-5m3/s.