Study Of Analysis And Recognition Technique Of Acoustic Signals From Non-contacting Multi-sensor Testing

Rolling machinery equipment widely used and easy to damage one of the mechanical parts in machinery and equipment play a role in load to bear and pass. At the same time, due to bearing damage was also caused huge economic losses. Therefore, working conditions, bearing in real-time monitoring and fault diagnosis of more and more attention. In this paper, acoustic theory and modern signal analysis methods in-depth study of non-contact rolling bearing fault diagnosis method of multi-sensor acoustic, this rolling bearing failure of early diagnosis and prevention of accidents has important theoretical significance and practical value.In this paper, a rolling acoustic emission experiments non-contact multi-detection system completed under the conditions of non-contact single-sensor and multi-sensor bearing fault detection experiments. By comparing the experiments show that failure of single sensor acquisition impact theory of collision number and frequency of the same, non-contact acoustic emission test failures provide the basis for pattern recognition.Based on the rolling bearing fault cyclical nature of acoustic emission signals to establish based on the sampling time and signal correlation of multi-sensor acoustic emission signals cyclical time - related (R-T) reconstruction method. According to the signal to trigger the time and bearing a different fault type of the relationship between theories of collision frequency, remove duplication and ineffective part of the reconstruction of a complete failure of periodic signals. RT reconstruction method to achieve multi-sensor-based bearing fault periodic acoustic signal reconstruction, access to a mobile in the rolling bearing fault periodic complete information for further different types of rolling bearing fault diagnosis was provided.Rolling bearing fault type identification method of research is another important part of this article. By analyzing the rolling bearing fault characteristic frequency and the acoustic emission signal parameters of the relationship between the establishments of impact based on the periodic acoustic emission counts of rolling bearing fault diagnosis. Typical of the rolling bearing fault detection of non-contact acoustic emission analysis of experimental data show that the established method can accurately diagnose a single rolling bearing failure. For the rolling bearing failure of multi-sensor non-contact acoustic emission testing, based on a complete reconstruction of periodic waveforms of acoustic emission signals using the fuzzy recognition theory, a combination of different bearing fault acoustic emission signal characteristics, to establish rolling multi-sensor acoustic emission signal reconstruction wave Figure fuzzy identification method.