【作者】 王小进；

【导师】 李艳军；

【作者基本信息】 南京航空航天大学 ， 载运工具运用工程， 2008， 硕士

【摘要】 XRF(X射线荧光)技术利用同位素放射源辐射被测物质,通过一系列装置产生能谱,根据能谱来判别物质的元素类别。这种技术在冶金、地质勘探中已经得到广泛的应用。本文把它应用到发动机智能监测中来,致力于研制出一套新型的智能监测系统。智能分析技术,包括状态预测和故障诊断。成功的诊断推理、计算方法很多,本文重点研究了灰色系统理论的灰预测和灰色关联度分析两个分支,以及统计法、人工免疫系统。灰色预测是把被预测系统看成是一个灰色系统,利用存在的已知信息去推知含有故障模式的不可知信息及其特性、状态和发展趋势。本文系统模型用该方法预测油液金属含量和发动机状态。灰关联度分析是灰色理论的一种新方法,它用关联度大小来描述事物、因素之间关联程度的一种定量化的方法。系统模型用该方法分析油液中各元素的浓度关联度,从而判别发动机的工况和元素分组。人工免疫系统,是受生物免疫系统的启发而产生的一种新型的智能计算方法。它通过对原始数据模式进行学习记忆,以及进一步的调整,记忆了不同的数据模式,然后对这些模式进行分类和“标记”,其结果可用于模式识别和状态聚类。本文的研究工作包括两部分:同位素XRF能谱处理与分析、智能分析模型。前者的输入是能谱数字数据,输出是元素浓度,同时元素浓度又是后者的输入,后者输出是发动机当前状态和预测状态。最后,基于以上的研究工作开发了发动机智能监测软件系统。更多还原

【Abstract】 XRF (X-ray fluorescence) technology, using isotope radioactive sources to eradiate materials which to be measured, produces energy spectrum through a series of installations, which is used to determine the elemental material categories. This technology has been widely used in metallurgy industry and geological exploration. In this paper it is applied into the engine intelligent monitoring, so that a new type of intelligent monitoring system could be developed.Intelligent analysis technology includes status prediction and fault diagnosis. There have many successful diagnosis reasoning and calculation methods been excogitated. In this paper it focuses on two branches of grey system theory (Gray Forecasting and Gray Relational Analysis), as well as Statistics, Artificial Immune System.Gray Forecasting, regarding the predicted system as a grey system, makes use of information known to deduce the information unknown that contain failure mode and its specialty, status and development trends. The system model, in this paper, predicts metal content in oil and the engine status with this method. Gray Relational Analysis is a new and quantitative method of grey system theory, which describes the degree of correlation between the objects and factors with relational value. The system model, analyzes relational value of concentration of various elements in oil, which is used to conclude engine condition and elemental grouping.Artificial immune system is a new type of intelligent calculation that is inspired of biological immune system. It learns and memorizes the original data models, and then, does further adjustment to memorize the different data models. Finally the models are classified and marked, the results of which are used for pattern recognition and state clustering.This study contains two parts: XRF energy spectrum processing and analysis, intelligent analysis model. For the former, its input is spectrum digital data, while its output is element concentrations. Meanwhile element concentration is the input of the latter, and the outputs of the latter are current status and prediction status of the engine. Finally, an intelligent monitoring software system of engine is developed based on the above research.更多还原