|
Modern industrial production of large rotating machinery, such as compressors, turbines, generators and other key equipment for the chemical, oil, electricity and other pillar industries of the national economy, the event of a failure, will give companies and countries enormous economic losses, but also threaten the lives and safety of the workers at the scene. Master equipment operating status, reduce or even prevent equipment accidents, the implementation of these key units online monitoring and diagnosis, with significant economic and social benefits. The pros and cons of the unit from the parking monitoring and analysis system is an important indicator to measure a set of rotating machinery condition monitoring and fault diagnosis system performance, has a very important role in monitoring and fault diagnosis system. However, the traditional unit from the parking analysis and diagnosis based mostly in the single section of the rotor single-source information, ignore the links between the various sensor information, the information obtained is not complete, accurate, precise diagnosis of the fault adversely affect . To change this situation, the whole IT transient signal processing, research and exploration of the full vector-based spectral theory of transient signal characterization and fault identification. On the unit from the parking process and from the traditional parking feature extraction method and transient analysis technology a brief introduction to seriously study the typical failure mechanism and vibration characteristics of the rotor, with examples of the traditional transient analysis, the full spectrum of transient analysis from the holographic stop technology technical characteristics, to lay a solid foundation for the full vector transient analysis method proposed and fault identification. Single cross-section of the rotor dual-channel information fusion theoretical basis, combined with the theory of vector spectrum fast algorithm to derive the full vector transient analysis of various parameters required. Two full vector transient analysis method, that is, the full vector Bode plot (FullVector Bode, FVB), ??three-dimensional full vector waterfall chart (Full Vector Cascade, FVC), and the specific unit from the parking instance and experimental its feasibility has been verified. The analysis results show that the full vector Bode plot (FVB), ??three-dimensional full vector waterfall chart (FVC) is feasible, can accurately show the critical speed of the rotor from the parking process and over-critical characteristics, you can replace the traditional Bode plot, waterfall applied to the unit from the parking analysis. According to the statistics of the fault in the unit from the parking process, combined with the existing experimental conditions, Typical Fault parking process may arise from the experimental simulation of the unit, such as rotor imbalance, Stator Rubbing Oil Whip (oil whirl ), the rotor-bearing system supporting loose. With full vector Bode plot, three-dimensional full vector waterfall plot carried out the analysis and diagnosis. The results show that: the the full vector Bode diagram, three-dimensional full vector waterfall chart for some type of fault performance-sensitive, reflect the vibration characteristics, more accurate diagnosis.
|