Research On Fault Diagnosis Method Of Analog Circuits Based On Noise Theory

With the rapid development and application of modern electronic systems in major industrial fields,the demand of research on circuit fault diagnosis,such as practicality,effectiveness,accuracy of fault diagnosis,has been urgently raised.Usually the electronic system circuit is divided into analog circuits and digital circuits.In recent years,digital circuits have rapidly emerged due to advantages such as low power consumption,high anti-interference,and ease of integration,which making digital devices occupy more than eighty percent of the entire circuit system.However,many key components in the system,such as filter circuits,feedback circuits,analog-to-digital converters,operational amplifiers,phase-locked loops,etc.,still cannot be replaced by digital circuits.In some large-scale synthetic systems,the analog subsystem consisting of analog components plays a central role in the overall system.Its performance is related to the stability and reliability of the entire system.Analog circuit fault diagnosis technology has become one of the important means to assist the safe operation of the system.This article focuses on the analog circuit parameter failure.From the perspective of saving test costs,improving diagnostic efficiency,and increasing practicality for analog circuit faults.An analog circuit fault diagnosis technique based on noise information is studied.The main tasks are as follows:1)The description of the disturbances of the linear system in the traditional time and frequency domain method is analyzed.A new description of the linear system under noise information is given.The noise information of the response output by the system can quantitatively reflect the status of the system.Focus on system disturbances caused by component failures.In this way,via noise-based signal processing technology,an analog circuit fault diagnosis scheme is set up.2)The author studies the nature and characteristics of the Ramanujan Sums.We define a class of operators,based on the Ramanujan Sums,termed here as Ramanujan class of operators.We then prove that these operators possess properties of first derivative and with a particular shift,of second derivative also.This extends the noise estimation scheme based on the Ramanujan operator.It has been proved that this scheme has a very high accuracy.3)The noise information of the system uncertainty under the disturbance of the system fault is excavated,and a dynamical fault model of analog circuits is constructed.This model transforms the problem of solving analog circuit faults into noise estimation in system uncertainties,and uses the noise estimation application program based on the Ramanujan operator to complete noise estimation of the system disturbance components.This systematic noise,which is sensitive to changes in component parameters,was selected as the fault feature.Finally,it is judged whether the circuit is faulty by comparing the difference of the noise information between the fault-free and faulty circuit.4)The simulation results in the actual circuit show that the proposed diagnosis method based on noise information can significantly identify single faults and multiple faults in the circuit.Comparison with other methods shows that the method has more advantages in the accuracy of fault diagnosis and the recognition ability fault diagnosis.