News

NEWS DYNAMIC

Location:Home > News > Company News > Production process sensors

Company News

Production process sensorsSource:上海郎福传感器制造中心 Time:2014-10-16 Hits:

Abstract:In order to accurately distinguish mutation causes of the sensor signal is proposed based on wavelet analysis of mathematical models of the band. For industrial process measurement and control system, analyzed the relationship between the frequency of the output signal consisting of mutation and mutation causes. , The high-frequency band signal separation using wavelet analysis techniques, and energy statistics, based on the proportion of high and low frequency signal energy changes, determine the cause of the mutation signal. Computer simulation and experimental results Water Supply System classic control system show that this method can effectively diagnose the sensor is faulty.

In the measurement and control system, the output signal of the sensor is affected by many factors, often mutated. These mutations point values contain important fault information, and to distinguish accurately capture point mutations cause these reasons generated, is the key sensor fault diagnosis.

Document depends only on the time series of sensor output to diagnose a fault of the sensor, the sensor output signals are mutations due to a sensor malfunction. Practice of literature is to control input and output signals, respectively, wavelet transform, wavelet function can be regarded as when the first derivative of a smooth function mutation of the signal corresponding to its wavelet transform modulus maxima, thus detecting point mutations, and analysis of the sequence and generates a residual sensor failure, and that mutation of the sensor output signal is input due to mutation or system failure caused by the signal of the sensor in fact, many of the sensor output signal caused by mutations in the cause, in addition to the system input outside the mutation and the sensor itself fails, there is a process disturbance, actuator failure, controller failure, the controlled object and external electromagnetic interference, etc. In practical applications, the sensor fault diagnosis method has certain limitations.

In general, industrial process control, the time constant of the controlled object is large, can not respond to the mutant signal high frequency components. Based on the frequency band of the wavelet analysis technique to explore the analysis leading to the sensor output signal mutations cause analysis provides a practical method for fault diagnosis and performance assessment online sensor.

All rights reserved©  上海郎福传感器制造中心   Copyright All rights reserved   Tel:   021—20935277   Phone:  13361825277

Address:上海市浦东新区沪南公路7519号  Website building: Trueland