Research On Service-Oriented Lifecycle Health Management Technologies For Tunnel Boring Machine

As typical complex equipment, Tunnel Boring Machine(TBM) occupy a irreplaceable position in china’s infrastructure construction, city modernization and so on, in especial urban subway and subsea tunnel. Researching on the health management technologies of TBM is of great significance to the TBM’s safety, reliability and availability, reducing construction cost, It will also help to reduce the economic loss, ensure the safety of the lives and property of the people and elevate China’s international standing. But the increasingly complicated components, structure and work environment make it more easily fault. Traditional health management technologies and service mode are more and more difficult to cope with the requirements of equipment diagnosis and maintenance. How to improve and ensure the reliability and availability of equipment system becomes a burning question for customers. For the TBM manufacturing enterprises, product competition mainly focuses on the development stage and after-sale stage. They are facing a common problem:producing high reliable products and providing prompt and efficient health management service for customers.Aim at the above-mentioned problems, this paper considered the health management knowledge acquisition problems of TBM from the point of view of product lifecycle. Be set in state key laboratory of tunnel boring machine construction project, this thesis studied the service-oriented health management technologies and applications for TBM’s lifecycle. The main outcomes and innovations of the thesis are as follows:(1) The changes of equipment health management service pattern and the roles of health management in TBM lifecycle were analyzed. Based on these, the service-oriented TBM lifecycle health management philosophy was proposed. Follow these guidances, we studied the architecture of the service-oriented TBM lifecycle health management system in terms of conceptual model, general construction and logical structure.(2) It reviewed main methods of condition monitoring and fault diagnosis for TBM and designed a VI (Virtual Instrument)-based TBM condition monitoring system. Prediction assessment model was established based on gray prediction and neural network for TBM state, which fundamentally achieved on aim of evaluation for TBM key components running status, by hydraulic oil temperature forecast and main bearing vibration amplitude changes forecast of the propulsion system.(3) Rough set and decision tree technology were used to establish fault diagnosis model of TBM based on data mining. Fault characteristic parameter of TBM were chose by rough set theory, after data preprocessing, faults were decided by the improved C4.5 algorithm. Finally, the diagnosis model was realized using the technology of Clementine.(4) It analyzed wear form and wear factors of the TBM tools, and built tools health management model based on the Markov chain. By actual tunnelling data, the judgement basis of tool wear identification was studied, and the relationship among the working parameters of TBM how to improve the tunneling efficiency was explored through the industrial experiment means.(5) Classification of maintenance task assignment problem of TBM, characteristic and performance index classification method were analyzed. It built fault task assignment model of TBM based on FAHP(fuzzy analytical hierarchy process)-Hungary hybrid algorithm, and the result was both considered the subjective factors of decision makers, and made the machine maintenance time, maintenance cost and after maintenance cycle optimal.(6) According to the theoretical studies above and relying on state key laboratory of tunnel boring machine construction project, this thesis studied the implementation strategy of the service-oriented lifecycle health management system for TBM and built a SOA(Service-Oriented Architecture)-based TBM health management prototype system. We designed the TBM health management service system, developed the key functional modules of the prototype system and gave the running examples, which shown the correctness of the theoretical studies and feasibility of the proposed technologies.