模拟电路软故障诊断测试节点优选的仿真研究.pdfVIP

第 21 卷第 12 期 系 统 仿 真 学 报© Vol. 21 No. 12 2009 年6 月 Journal of System Simulation Jun., 2009 模拟电路软故障诊断测试节点优选的仿真研究 胡 梅，王 红，杨士元，胡 庚 (清华大学自动化系，北京 100084) 摘 要：基于斜率故障模型和线性相关系数故障模型，提出了一种模拟电路软故障诊断字典法中 的测试节点优选算法。针对模拟电路中的单故障和双故障，分别利用节点电压增量计算出的斜率和 线性相关系数作为统一故障特征，建立故障字典；给出了在字典建立过程中，利用PSPICE 仿真工 具对电路进行故障仿真，采用特征函数描述故障统一特征的差别，根据故障隔离度优选测试节点的 算法和步骤；并用仿真实例验证了该测试节点优选方法的有效性。 关键词：模拟电路；故障仿真；软故障诊断；统一故障特征；故障字典；测试节点；故障隔离 中图分类号：TP391.9 文献标识码：A 文章编号：1004-731X (2009) 12-3837-05 Simulation Studies of Test Nodes Selection for Soft Fault Diagnosis in Analog Circuits HU Mei, WANG Hong, YANG Shi-yuan, HU Geng (Department of Automation, Tsinghua University, Beijing 100084, China) Abstract: Based on the slope fault mode and the coefficient of linear relativity fault mode, a test nodes selection algorithm for soft fault diagnosis dictionary method of analog circuits was proposed. Considering the single faults and double faults in the circuits, the slopes and the coefficients of linear relativity were utilized which were calculated from node-voltage increments as unified fault features respectively to establish fault dictionary. PSPICE simulation tool was used to simulate the fault states of the circuit, and the eigenfunction was used to describe the differences between every two unified features, the algorithm and approach of test nodes selection was proposed according to the fault isolation ability. Finally, simulation results demonstrate to validate the efficiency of the test nodes selection