Diagnosis of Incipient Faults in Nonlinear Analog Circuits Based on High Order Moment Fractional Transform
- PDF / 1,483,989 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 38 Downloads / 194 Views
Diagnosis of Incipient Faults in Nonlinear Analog Circuits Based on High Order Moment Fractional Transform Yong Deng 1
&
Ting Chen 1 & Di Zhang 1
Received: 14 December 2019 / Accepted: 1 June 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Considering the problem to diagnose incipient faults in nonlinear analog circuits, a novel approach of fault feature-extracting based on HMFT (high order moment fractional transform) of the component parameter change is proposed. Firstly, the equivalence circuit model is established according to the topological structure and the voltage sensitivity of the circuit under test (CUT). Then the HMFT of component parameter changes are derived, which are used as fault features. Finally, using the extracted features, incipient fault diagnosis is accomplished. The simulations illustrate the proposed method and show its effectiveness in the incipient fault recognition capability. Keywords Nonlinear circuits . Incipient fault diagnosis . High order moment . Fractional transform
1 Introduction Analog integrated circuits and mix-signal circuits play important roles in electronic systems and have got more and more attentions in recent years. However, it is reported that although only 20 percentages of electronic components are analog, they results in around 80% of faults in the system [17]. Therefore, detection and isolation of faults in analog circuits is of major importance and challenging. The accurate models of most analog circuits are difficult to establish due to the following factors: (1) There exist tolerances for the analog parts; (2) Status and all the parameters of the circuit are continuous, which makes the possibility of status of the circuit is infinite;(3) most nodes of the circuits can not be tested. Some authors use the fuzzy model to test faults and diagnose soft faults in analog circuits [10, 38, 46]. Other authors study special signals [41]. They try to select optimal frequency to extract the fault features of analog circuits. Test point selection method for analog circuit fault diagnosis is also deeply studied by researchers [25, 26]. The authors try to find more Responsible Editor: S. Sindia * Yong Deng [email protected] 1
School of Mechatronic Engineering, Southwest Petroleum University, Chengdu 610500, China
test nodes instead of the output node to solve the problem of unique testing point. The entropy method is also introduced to diagnose faults in analog circuits [16]. An approach based on the ratio of normal variables and the slope fault model is proposed in reference [2]. By polynomial approximation, fault detection method is proposed [22]. It should be noted that the methods in [2, 16, 22] belong to statistic analysis methods, which diagnose faults in analog faults using statistic features and analyze the possibility of the circuit faulty states. The use of probability moments is proposed to test and diagnose analog circuits, which reduces complexity of input signal design, increases resolution of fault detection, and reduce
Data Loading...