Research On Failure Diagnosis For Electro-Hydraulic System Based On Sound Intensity

This paper researches on a novel method to measure three-dimensional sound intensity use for failure diagnosis of electro-hydraulic system and the fabrication of a miniature three-dimensional sound intensity probe by using two separate micro-machined pressure probes and one microphone frame. A three-dimensional sound intensity probe has been realized based on a three-dimensional micro-machined pressure microphone, built into a miniature pressure microphone probe.Sound intensity describes the flow of acoustic energy produced by a sound source, and it is a vector. Sound intensity measurements accurately capture only the sound produced by the source under test, eliminating interference from other sounds and realizing the location of the sound.Traditional measurement of sound power and sound pressure must be processed under a rigorous acoustic environment and required that the device be shipped to an acoustical laboratory chamber. It is very difficult to realize under industrial circumstance, but the measurement of sound intensity can be processed. Sound intensity measurement allows accurate identification, isolation and ranking of individual sound sources. For complex machines and processes, the noisy components can be identified and treated. The measurement of sound intensity can be applied to course detect, product research and development applications.Sound localization is the process of determining the position of the sound source expressed by the azimuth and the elevation angles. Failure localization is the problem of finding out the pose (position and orientation) of the hydraulic system from sensor readings. Based upon recent researches done on binaural sound localization algorithms in the Hydraulic Systems Laboratory at Hebei University of Technology, this research sought to apply localization of sound to automatically localize the hydraulic system in a small laboratory environment. By placing a failure sound source at predetermined positions and comparing its signal acquired by probes with a calibrated signal memorized in computer, letting the software we programmed decide thefailure between the two sound sources by rotating in place to orient itself to each source in succession, if the probing signal is bigger than calibrated signal, the computer will give a alert. The hydraulic system was able to be determining its current pose. The research successfully realized this model of hydraulic system localization and diagnosis.