作者：王力1, 周志杰1,2, 胡昌华1, 刘涛源1
作者单位：1. 火箭军工程大学控制科学与工程系, 陕西 西安 710025;
2. 西安理工大学自动化与信息工程学院, 陕西 西安 710048
关键词：置信规则库;证据理论;模式识别;超声检测
摘要：
为解决超声检测缺陷精确识别问题，综合运用检测数据和专家知识，研究一种基于置信规则库（belief-rule-base，BRB）和证据推理（evidential reasoning，ER）进行超声检测缺陷识别的方法。提出一种融合多种特征信息的BRB-ER缺陷识别模型，利用最小均方误差算法进行模型初始参数的优化，从而提高缺陷识别的准确性。通过超声检测手段获取某航空材料的缺陷数据，并对所提出识别方法进行验证。试验结果显示：该方法能够准确地进行缺陷识别，并可根据已有的产品缺陷类型进行训练，建立更加准确的缺陷识别模型。
Ultrasonic testing defect recognition based on belief-rule-base and evidential reasoning
WANG Li1, ZHOU Zhijie1,2, HU Changhua1, LIU Taoyuan1
1. Department of Control Engineering, Rocket Force University of Engineering, Xi'an 710025, China;
2. School of Automation and Information Engineering, Xi'an University of Technology, Xi'an 710048, China
Abstract: To improve accuracy in defects recognition by ultrasonic testing, a method of defects recognition based on belief-rule-base (BRB) and evidential reasoning (ER) is proposed according to test data and expert knowledge. Firstly, a new model of defects recognition integrating feature information based on belief-rule-base(BRB) and evidential reasoning(ER) is presented, and then the initial parameters of model are optimized with the minimum mean square error algorithm to improve the accuracy of defects recognition. Finally, ultrasonic testing is used to get the defect data of one aeronautical material and a case study is carried out to illustrate the ability and efficiency of the proposed method. The study results show that the method can recognize defects accurately. A more accurate defects recognition model is established based on exercises on existing product defects.
Keywords: belief-rule-base;evidential reasoning;pattern recognition;ultrasonic testing
2017, 43(4): 6-10,67 收稿日期: 2016-09-19;收到修改稿日期: 2016-11-23
基金项目: 国家自然科学基金（61370031，60736026）；飞行器海上测量与控制联合实验室开放基金（FOM2014OF14）
作者简介: 王力(1992-),男,陕西西安市人,硕士研究生,专业方向为诊断技术及自动化装置。
参考文献
[1] 美国无损检测学会. 美国无损检测手册(超声卷)[M]. 北京:世界图书出版社,1996:28-38.
[2] 张道富,曾燕,周晓车,等. 基于信息融合技术的无损检测缺陷模式识别[J]. 仪器仪表学报,2003,24(5):449-452.
[3] 韩崇朝,朱红艳,段战胜,等. 多源信息融合[M]. 北京:清华大学出版社,2006:19-44.
[4] 潘泉,王增福,梁彦,等. 信息融合理论的基本方法与进展(II)[J]. 控制理论与应用,2012,29(10):1233-1234.
[5] YANG J B. Rule and utility based evidential reasoning approach for multi-attribute decision analysis under uncertainties[J]. European Journal of Operational Research,2001,131(1):31-61.
[6] YANG J B, LIU J, WANG J. Belief rule-base inference methotology using the evidential reasoning approach-RIMER[J]. Systems, Man and Cybernetics(Part A),2006,36(2):266-285.
[7] 周志杰,杨剑波,胡昌华,等. 置信规则库专家系统与复杂系统建模[M]. 北京:科学出版社,2011:9-30.
[8] 叶钱. 噪声背景下金属材料小缺陷超声检测关键技术的研究[D]. 杭州:浙江大学,2015.
[9] CHANG L L, ZHOU Z J, YOU Y, et al. Belief rule based expert system for classification problems with new rule activation and weight calculation procedures[J]. Information Science,2016,336(1):75-91.
[10] XU D L, LIU J, YANG J B, et al. Inference and learning methodology of belief-rule-based expert system for pipeline leak detection[J]. Expert Systems with Applications,2007,32(1):103-113.
