Research On Large Compressor Unit Simulation Training System And AE Signal Feature Identification Method In Offshore Platform

This thesis is divided into two parts, one is research on large compressor unit simulation training system and the other is research on AE signal feature recognition method in offshore platform.The compressor is a kind of important technology-intensive equipment that is widely used in petrochemical industry. For good daily operation and maintenance of the compressor, technical people on-site are increasingly requested for a high degree of professional skill levels, which requires the scientific and efficient training. The traditional training method is not only time-consuming and laborious, but also the training effect is not very ideal. As the problems such as the brain drain and the poor training of the new staff have become more and more prominent, so the development of new training methods is in urgent need. With the rapid development of the computer simulation technology, simulation and training technology of the compressor has achieved a constant development. And be compared with traditional training method, it shows more advantages, such as the short training cycle, low training cost and flexible training ways, etc.Under such a background, this thesis carries out the research on large compressor unit simulation training system, in order to improve the training efficiency of the enterprise. And the large centrifugal compressor unit simulation training system that this paper has developed for a certain compressor station also has a very good reference for study on the simulation training systems of other types of centrifugal compressor units.The offshore platform is the key production and living base that the humanity realizes marine oil and gas resources exploration. This thesis designs different experiment schemes, studies the acoustic emission wave propagation characteristics in the basic offshore platform structure like the metal plate and metal circular pipe. It is concluded that the law and differences of propagation characteristics in the metal plate and metal circular pipe. On this basis, the following research is mainly focused on the acoustic emission signal features of the offshore platform common fault like the crack, corrosion, contact-rubbing and lead-break (simulate the impact), etc., which includes the features in the time, frequency and time-frequency domain. And the exploration and research on fault acoustic emission signal features identification are conducted by applying the signal correlation analysis technology including two methods of the signal correlation in the time domain and time-frequency domain. The result shows that the features identification of fault acoustic emission signals based on the signal correlation method has certain feasibility. To a certain extent, both the time and time-frequency domain signal correlation methods can identify four types of faults in roughly the same accuracy, although the fault identification effect is not very good. Therefore, the fault acoustic emission signal feature identification method based on the signal correlation analysis technology proposed in this thesis has an important reference value in the field of the offshore platform acoustic emission feature identification.