【作者】 孙德平；

【导师】 孙玉清；

【作者基本信息】 大连海事大学 ， 轮机工程， 2005， 硕士

【摘要】 本文是笔者所在实验室用功率流理论进行液压系统的状态监控和故障诊断研究的一部分,是在前人非插入式流量测量研究的基础上所展开的进一步研究。 文章中介绍了非插入式流量计的基本原理,基本设计思想。着重解决实验过程中所遇到的诸如误差问题,数据的处理以及实验中经验公式的确定等方面的困难。 分析了实验中可能产生的误差,查找产生系统误差的原因,并从实验数据处理上消除系统误差的影响;分别应用3σ准则、肖维勒准则、格拉布斯准则、t检验准则,对实验所得的数据中存在的粗大误差进行剔除,并对这四种准则进行分析、比较,并据此提出了本实验最终处理数据时所采用的准则。 在对实验数据进行相关性分析的基础上,运用图解法、平均值法、最小二乘法对处理后的数据进行曲线拟合,获取实验的经验公式。对用三种方法所得的经验公式进行回归精度分析后,选定具体处理实验数据的方法。 采用本文提出的数据处理方法及经验公式的取得方法,在实验理论与实验方法正确的前提下,能排除实验中误差数据对实验结论的影响,较为真实地反映出各检测量之间的关系。对于其它的实验进行误差分析、数据处理及经验公式的取得有一定的借鉴意义。更多还原

【Abstract】 This paper is one part of the study on condition monitoring and fault diagnosis on the basis of power flow theory. It is a step forward study on the basis of non-inserting measurement research.It introduces the basic theory and design philosophy of non-inserting flow meter, and mainly solves the problems in experiment such as error, data processing and the confirmation of experience formula.It analyzes the error that may occur in the experiment and finds out the reason, therefore reduced the influence of the systematic error form data processing of the experiment. The research succeeds in getting rid of the big errors by using 3 σ criterion, Charvent criterion, Grubbs’ criterion and T test criterion. It also makes a comparison of the four criteria and finally put forward an experiment in view of the above.On the basis of correlation analysis of the experiment data, it uses graphic method, average law, least square method to obtain the experience formula of the experiment. Then it does a regression of the experience research data and selects a concrete method to deal with the experimental data.By adopting the data processing and experience formula raised by this paper, one can get rid of the influence to experiment result and reflect the relations between every measuring researches on the basis of the correct theory and method. It is a good reference for other relevant error analysis, data processing and experience formula gaining.更多还原