盾构机推进系统故障预测研究

发布时间：2018-06-07 15:50

本文选题：盾构机推进系统 + 故障预测　；参考：《南京理工大学》2014年硕士论文

【摘要】：盾构机作为一种被广泛应用于城市地铁建设的大型工程机械,其工作条件受到多种自然环境的影响,容易发生故障,因此对其故障预测技术的研究有十分重要的意义,但是采用传统故障预测技术很难满足要求,而随着人工智能故障预测技术的出现及其在实际工程应用中取得了很好的预测效果,所以对盾构机的故障采用智能预测方法变得现实可行。本文主要是通过对专家系统理论知识的分析,并结合模糊逻辑理论和神经网络知识的技术优势,对盾构机推进系统的故障预测进行了初步的探讨,完成了如下几个方面的工作： (1)建立了盾构机推进系统的故障知识库。对盾构机推进系统的故障产生机理进行了分析,将其故障分为了浅层故障知识和深层故障知识,并对与盾构机推进系统相关的故障征兆参数进行了选取,同时引入数据库技术对故障知识库进行了设计和处理。 (2)研究了盾构机推进系统的故障预测推理机的算法。针对盾构机推进系统故障的复杂性和不确定性,引入模糊逻辑理论和神经网络知识对其故障预测推理机分别进行设计与仿真,在对比分析了它们的优缺点与精确度之后,提出将模糊神经网络运用于故障预测推理机的设计之中,并在MATLAB软件中对模糊神经网络故障预测算法进行了实验仿真,其仿真结果具有更高的精度,证明了其在盾构机推进系统故障预测中的有效性和准确性。 (3)设计了盾构机推进系统的故障预测软件。结合软件设计原则,本文选择VisualC++6.0软件作为专家系统的软件设计平台,并通过OPC技术完成了VC与WinCC软件的数据交换,采用COM组件技术实现了VC对MATLAB编写的神经网络和模糊神经网络故障预测算法的调用,同时在开发过程中采用界面化和模块化设计方式,使得对系统软件功能模块的扩充更加方便,也更加符合整个系统软件的设计要求。
[Abstract]:As a kind of large-scale construction machinery widely used in urban subway construction, the working conditions of shield machine are affected by many kinds of natural environment and are prone to failure. Therefore, it is of great significance to study the fault prediction technology of shield machine. However, it is difficult to meet the requirements by using the traditional fault prediction technology, and with the emergence of artificial intelligence fault prediction technology and its application in practical engineering has achieved a very good prediction effect. So it is feasible to apply intelligent prediction method to shield machine fault. Based on the analysis of expert system theory knowledge and the technical advantages of fuzzy logic theory and neural network knowledge, this paper makes a preliminary discussion on the fault prediction of shield machine propulsion system, and accomplishes the following work: The fault knowledge base of shield machine propulsion system is established. The fault generation mechanism of shield machine propulsion system is analyzed, the fault is divided into shallow fault knowledge and deep fault knowledge, and the fault symptom parameters related to shield machine propulsion system are selected. At the same time, the database technology is introduced to design and deal with the fault knowledge base. The algorithm of fault prediction inference machine for shield machine propulsion system is studied. In view of the complexity and uncertainty of the fault of shield machine propulsion system, the fuzzy logic theory and neural network knowledge are introduced to design and simulate the fault prediction inference machine respectively. The fuzzy neural network is applied to the design of the fault prediction inference machine, and the simulation of the fuzzy neural network fault prediction algorithm is carried out in the MATLAB software. The simulation results show that the simulation results have higher accuracy. The validity and accuracy of this method in fault prediction of shield machine propulsion system are proved. The software of fault prediction for shield machine propulsion system is designed. Combined with the principle of software design, this paper chooses VisualC 6.0 software as the software design platform of expert system, and completes the data exchange between VC and WinCC software through OPC technology. The COM component technology is used to realize the call of the neural network and fuzzy neural network fault prediction algorithm written by MATLAB by VC. At the same time, the interface and modularization design method are adopted in the development process. It makes it more convenient to expand the function module of the system software and meets the design requirements of the whole system software.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U455.39

【参考文献】