Main performance indicators

I. Overview

Instrumental performance indicators are often described in terms of accuracy (also known as accuracy), variation, and sensitivity. Meter calibration instruments are usually calibrated for accuracy, variation, and sensitivity. Variation refers to the maximum difference between the indicator values ​​when the meter's measured variable (which can be understood as the input signal) reaches the same value from different directions multiple times, or the meter is tested under the same external conditions. Change the parameter from small to large (positive characteristics) and measured parameters from large to small changes (reverse characteristics) degree of inconsistency, the difference between the two is the instrument deterioration, as shown in Figure 1-1-1. The percentage of variation in the ratio of the maximum absolute error to the scale range of the meter:

The main cause of the variation is the gap between the instrument's vibratory motion mechanism, the friction of moving parts, and the delay of elastic elements. Winning the continuous improvement of the instrument manufacturing technology, especially the introduction of microelectronics technology, many of the instruments are all electronic, no moving parts, analog instruments are changed to digital instruments, etc., so the indicator of deterioration is not seen in smart meters. So important and outstanding.

Sensitivity is the sensitivity of the instrument to the change of the measured parameter, or the ability to respond to the change in the measured quantity. It is the ratio of the output change increment to the input change increment in the steady state:

The sensitivity is sometimes referred to as the "magnification ratio," which is also the slope of the points on the instrument's static properties. Increasing the magnification can improve the sensitivity of the instrument. Simply increasing the sensitivity does not change the basic performance of the instrument. That is, the accuracy of the instrument does not increase. On the contrary, the phenomenon of oscillation sometimes occurs and the output is unstable. The sensitivity of the meter should be maintained in an appropriate amount.

However, for instrument users, such as chemical industry instrumentation workers, instrument accuracy is an important indicator, but in actual use, it often emphasizes the stability and reliability of the instrument, because chemical industry testing and process control instrumentation for measurement There are few, and a large number are used for testing. In addition, the stability and reliability of detection instruments used in process control systems are more important than accuracy.

Second, accuracy

Instrument accuracy is called precision, also known as accuracy. Accuracy and error can be said to be twin brothers, because there is an error, there is the concept of accuracy. Instrument accuracy is, in a nutshell, the degree to which the instrument's measured value is close to the true value, usually expressed as a relative percentage error (also known as a relative compromise error). The relative percentage error formula is as follows:

(1-1-3)

The relative percentage error in the δ-detection process;

(Scale upper limit - lower limit of the scale)--meter measurement range;

The Δx-absolute error is the difference between the measured value of the measured parameter x1 and the measured value of the parameter x0.

The so-called standard value is the value measured by a standard table that is 3 to 5 times more accurate than the meter under test.

From the equation (1-1-3), it can be seen that the meter accuracy is not only related to the absolute error, but also related to the measuring range of the meter. The absolute error is large, the relative percentage error is large, and the accuracy of the meter is low. If the two instruments with the same absolute error have different measurement ranges, the relative percentage error of the instrument with a large measurement range is small and the instrument accuracy is high. Accuracy is a very important quality indicator of the instrument. It is commonly used to standardize and represent accuracy levels. The accuracy level is the maximum relative percent error minus the sign and %. According to the unified national regulations, there are 0.005, 0.02, 0.05, 0.1, 0.2, 0.35, 1.0 and 1.5.

2.5, 4 and so on, the instrument accuracy grade is generally marked on the instrument scale or signage, such as,, 0.5, etc., the smaller the number, the higher the instrument accuracy.

To improve the accuracy of the instrument, error analysis must be performed. Errors can often be classified as careless errors, slow errors, system errors, and random errors. Negligence error refers to the artificial error caused by the measurement process, one can be overcome, and the second one has nothing to do with the instrument itself. The gradual error is caused by the aging process of the internal components of the instrument. It can be overcome and eliminated by replacing the components and parts or through constant correction. Systematic error refers to the error that the magnitude or sign of the same value appears when the same measured parameter is repeatedly measured multiple times, or the error that changes according to a certain rule, but it has not yet been caused by accidental factors recognized by people. The size and nature are not fixed and difficult to estimate. However, statistical methods can be used to theoretically estimate the impact on the test results. Sources of error mainly refer to systematic and random errors. When error is used to represent accuracy, it refers to the sum of random error and system error.

Third, reproducibility (repeatability)

Measurement reproducibility is the extent to which the measurement results are consistent under different measurement conditions, such as different methods, different observers, and when the same detected quantity is detected in different detection environments. Measurement reproducibility will become an important performance indicator of the instrument.

The accuracy of the measurement is not only the accuracy of the instrument, it also includes the influence of various factors on the measurement parameters, and is a comprehensive error. Take the electric type III differential pressure transmitter as an example, the overall error is as follows:

(1-1-4)

The reference accuracy in the state of e0-(25±1)°C, ±0.25% or ±0.5%;

e1-Effect of ambient temperature on zero (4mA), ±1.75%;

e2--Effect of ambient temperature on full scale (20mA), ±0.5%;

E3 - Effect of working pressure on zero (4mA), ± 0.25%;

e4--Effect of working pressure on full scale (20mA), ±0.25%;

Substituting the values ​​of e0, e1, e2, e3, and e4 into equation (1-1-4) results in:

This shows that the measurement accuracy of the 0.25-level Electric III transmitter was reduced from the original 0.25 level to 1.87 due to changes in the temperature and working pressure, indicating that this meter has poor reproducibility. It also shows that when measuring the same measured quantity, due to different measurement conditions, affected by the ambient temperature and working pressure, the measurement results are inconsistent.

If a full smart differential transmitter is used to replace the electric type III differential pressure transmitter in the above example, e0=±0.0625% in the corresponding equation (1-1-4), e1+e2=±0.075%, e3+e4=±0.15 %, substituting equation (1-1-4) to obtain e = ±0.18%, which is much smaller than the electric type III differential pressure transmitter e = ±1.87%, indicating that the full-intelligent differential pressure transmitter is Pressure compensation, resistance to ambient temperature and working pressure. Instrument reproducibility can be used to describe the instrument's ability to resist interference.

Measurement reproducibility is usually estimated using uncertainty. Uncertainty is the degree of uncertainty about the measured value due to the presence of a measurement error, which can be expressed as a variance or standard deviation (positive square root of Deng's variance). All components of uncertainty are divided into two categories:

Class A: Component B determined by statistical methods: Components determined by non-statistical methods: Set the variance of class A uncertainty to si2 (standard deviation is si), and the corresponding approximate variance of class B holiday uncertainty to ui2 ( If the standard deviation is (ui), the combined uncertainty is:

(1-1-5)

Fourth, stability

Within the specified operating conditions, the ability of the instrument to maintain certain performance over time is called stability (degrees). Instrument stability is a performance indicator that chemical industry instrumentation workers are very concerned about. Because the environment in which chemical companies use meters is relatively poor, the measured temperature and pressure of medium are also relatively large. In this environment, the use of meters will reduce the ability of certain parts of the meter to remain constant over time. Stability will decline.徇 Or characterization instrumentation stability has not yet had a quantitative value, chemical companies usually use instrument zero drift to measure the stability of the instrument. During the one-year operation of the instrument, there is no drift in the zero position. On the contrary, when the instrument is put into operation for less than three months, the zero point of the instrument is changed, indicating that the instrument is not stable. Instrument stability is directly related to the scope of use of the instrument, and sometimes directly affect the chemical production, the impact of instrument stability caused by the impact of the decline in the accuracy of dual instruments will have greater impact on chemical production. Poor instrument stability The instrument maintenance is also large, and it is the most undesirable thing for the instrumentation worker.

Fifth, reliability

Instrument reliability is another important performance indicator pursued by the instrumentation industry of chemical companies. The reliability and maintenance of the instrument are inversely related to each other. The high reliability of the instrument indicates that the maintenance of the instrument is small. On the contrary, the reliability of the instrument is poor, and the maintenance of the instrument is large. For the detection and process control instruments of chemical companies, most of them are installed on process pipelines, various types of towers, tanks, tanks, and vessels, and the continuity of chemical production, most toxic, inflammable and explosive environments, and these bad conditions give the instrument maintenance. Many difficulties have been added. One is to consider the safety of chemical production, and the other is to relate to the personal safety of the instrument maintenance personnel. Therefore, the use of inspection and process control instruments by chemical companies requires that the maintenance quantity is as small as possible, which means that the instrument reliability is required to be as high as possible.

With the upgrading of the instrument, especially the introduction of microelectronics technology into the instrument manufacturing industry, the instrument's noticeability has been greatly improved. Instrument manufacturers are also paying more and more attention to this performance indicator. MTBF is usually used to describe the reliability of the instrument. The MTBF of an all-intelligent transmitter is about 10 times higher than that of non-intelligent instruments such as the Electric III transmitter, and it can be as high as 100 to 390 years.