Sensor failures mainly include four types: complete failure failure, fixed deviation failure, drift deviation failure and accuracy drop.
Failure failure refers to the sudden failure of the sensor measurement, the measured value has been a certain constant;
Deviation fault mainly refers to a type of fault in which the measured value of the sensor is different from the true value by a constant constant. The faulty measurement and the non-faulty measurement are parallel;
Drift fault refers to a type of fault in which the difference between the measured value of the sensor and the true value increases with time;
Decrease in accuracy means that the measurement capability of the sensor becomes worse and the accuracy becomes lower. When the accuracy level decreases, the measured average value does not change, but the measured variance changes.
Both fixed deviation faults and drift faults are not easy to find faults, and will cause a series of unpredictable problems in the process of fault occurrence, making the control system unable to function normally for a long time.
Sensor fault classification method
- Classified according to the degree of sensor failure
According to the degree of sensor failure, it can be divided into hard failure and soft failure.
Hard faults generally refer to faults caused by structural damage, with large amplitudes and sudden changes; soft faults generally refer to characteristics variation, with small amplitudes and slow changes.
Hard failure is also called complete failure. When a complete failure occurs, the measured value does not change with actual changes, and a certain reading is always maintained. Usually this constant value is generally zero or the maximum reading. The fault measurement is roughly a horizontal straight line.
Soft faults include data deviation, drift, and drop in accuracy level. Soft faults are relatively small and difficult to be discovered. Therefore, in a sense, soft faults are more harmful than hard faults, and their harm has gradually attracted people’s attention.
- Classified by the performance of the fault
According to the performance of the fault, it can be divided into intermittent fault and permanent fault.
Intermittent failures are good and sometimes bad; after a permanent failure, it cannot be restored to normal.
- Classification according to the process of failure occurrence and development
According to the fault occurrence and development process, it can be divided into sudden fault and slow-change fault.
The rate of change of sudden fault signal is large; the rate of change of slowly changing fault signal is small.
- Classification according to the cause of the failure
According to the cause of the fault, it can be divided into deviation fault, impact fault, open circuit fault, drift fault, short circuit fault, periodic interference, and nonlinear dead zone fault.
The fault cause of deviation fault is: bias current or bias voltage, etc.;
The cause of the impact fault is: random interference in the power supply and ground wire, surge, spark discharge, burr in the D/A converter, etc.;
Causes of open circuit failure: broken signal wire, chip pin not connected, etc.;
The cause of the drift failure: warm;
Causes of short-circuit failure: bridge corrosion caused by pollution, short circuit connection, etc.;
Causes of periodic interference failure: power supply 50Hz interference, etc.;
Causes of nonlinear dead zone faults: amplifier saturation, nonlinear links, etc.
In addition, from the perspective of modeling and simulation, it can be divided into multiplicative faults and false faults. For bias faults, a constant or random small signal is added to the original signal; for impact interference, a pulse signal can be superimposed on the original signal; for short-circuit faults, the signal is close to zero; for open-circuit faults, the signal is close to the maximum output of the sensor; Drift failure, the signal deviates from the original signal at a certain rate; periodic interference failure, the original signal is superimposed with a signal of a certain frequency.
Diagnosis method of sensor failure
From different perspectives, the classification of fault diagnosis methods is not exactly the same. The fault diagnosis methods are simply divided into: methods based on analytical mathematical models and methods that do not rely on mathematical models.
- Method based on analytical mathematical model
According to the different forms of residuals, the methods based on analytical mathematical models can be further divided into: parameter estimation method, state estimation method and equivalent space method.
Model-based fault diagnosis method is one of the earliest developed diagnosis methods, and it is also one of the most widely studied and applied diagnosis methods.
The advantages are clear model mechanism, simple structure, easy implementation, easy analysis, and real-time diagnosis. It has an important position in the field of fault diagnosis, and will still be the main research direction of sensor fault diagnosis methods in the future development.
The disadvantages are the large amount of calculation and the complex system; there are modeling errors and the adaptability of the model is poor; the reliability is poor, prone to false alarms, false alarms, etc.; the robustness of external disturbances, the system is not sensitive to noise and interference.
At present, the research results of this diagnosis method are still mainly focused on linear systems, which is of great significance for the in-depth study of general fault diagnosis techniques for nonlinear systems. At the same time, robustness issues also have high research value.
- A fault diagnosis method that does not rely on mathematical models
At present, the control system is becoming more and more complex. In practice, it is difficult to establish a precise analytical mathematical model of the control system. When there is a modeling error, the fault diagnosis method based on the model will cause false alarms and omissions, so The fault diagnosis method that does not depend on the model is highly valued.
The advantage of the method that does not rely on mathematical models is that it does not require an accurate model of the object and is highly adaptable. The disadvantage is that the structure is complex and difficult to implement.
This fault diagnosis method that does not depend on the system model can be divided into fault diagnosis methods based on data-driven methods, fault diagnosis methods based on knowledge, and methods based on discrete events.
2.1 Data-driven approach
There are two categories of data-driven methods: signal processing methods and statistical methods.
Some commonly used fault diagnosis methods based on signal processing are: absolute value test and trend test, fault detection using Kullback information criterion, fault detection method based on adaptive sliding LatTIce filter, and related analysis of fault detection method based on signal modal estimation Methods, wavelet analysis methods and information fusion methods.
2.2 Knowledge-based approach
Knowledge-based fault diagnosis methods can be divided into two types: symptom-based fault diagnosis methods and qualitative model-based fault diagnosis methods.
2.3 Method based on discrete events
The fault diagnosis method based on discrete events is a new type of fault diagnosis method developed in recent years. The basic idea is: the state of the discrete event model reflects both the normal state and the fault state of the system.
With the progress of theoretical research and the continuous improvement of technical level, the research of sensor fault diagnosis will become more practical, and some problems encountered in practice will be gradually solved.
View more quality oxygen analyzers with low price and free shipping all over the world.