Principle and Method for Determination Verification Period of Measuring Instruments
Some standard content:
National Metrology Technical Specification of the People's Republic of China JJF 1139—2005 Principle and Method for Determinatlon Verification Period of Measuring Instruments 2005-12-20 Issued 2006-03-20 Issued by the General Administration of Quality Supervision, Inspection and Quarantine JJF 1139—2005 Principle and Method for Determination Verification Period of Measuring Instruments JJF 1139—2005
This specification was approved by the General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China on December 20, 2005, and will be implemented from March 20, 2006.
Responsible unit:
Main drafting unit:
Participating drafting unit:
National Legal Metrology Management and Planning Technical Committee Guangzhou Metrology and Testing Institute
The First Metrology and Testing Research Institute of the National Defense Science and Technology Commission China National Institute of Metrology
Jiangxi Jijing Testing Institute
This specification is responsible for the interpretation of the National Legal Metrology Pipeline Burial Technical Committee. The main drafters of this specification are:
Zhou Lunbin
Shi Changyan
Dai Runsheng
Participating drafters:
Hong Baolin
Bai Zhongyuan
Li Weiming
Xia Pingyi
JJF 11392005
(Guangzhou Metrology and Testing Institute)
(The First Measurement and Testing Research Center of the National Defense Science and Technology Commission) (China Institute of Metrology)
(Guangzhou Metrology and Testing Institute)
(Jiangxi Metrology and Measurement Institute)
References
3 Terms and definitions
Measuring instruments
[Measuring instruments] Verification
First verification
Subsequent verification
Periodic verification
Verification period
Principles for determining the verification period
Determination principle
Determination principle 1
Determination principle 3
5 Methods for determining the verification period
Reaction method
Maximum likelihood method
JJF 1139—2005
Appendix A Calculation Example of the Incremental Response Adjustment Method. Appendix
JJF 11392005
Principles and Methods for Determining the Verification Period of Measuring Instruments This specification is formulated with reference to the international document 0IM1.1310:1984 "Criteria for Determining the Recalibration Interval of Measuring Equipment Used in Testing Laboratories" published by the International Organization for Legal Metrology and the industry SERP-1!1996 "Confirmation and Adjustment of Calibration Intervals" published by the American National Standards Laboratory Association. The purpose is to scientifically and reasonably determine the verification period of measuring instruments to ensure that the measuring instruments are accurate and reliable within the specified verification period. 1 Scope
This specification specifies the basic principles and methods for determining the verification period of measuring instruments. This specification is applicable to the determination of the verification period of applicable measuring instruments when formulating or revising detailed measurement procedures. It can also be used as a reference for the adjustment of the verification interval of in-use measuring instruments and the confirmation of the calibration time interval of in-use measuring instruments. 2 References
JJF10N01--1998 General metrological terms and definitions JJF1033-2001 Metrological standards assessment specification JJF1071-2000D National metrological calibration specification compilation rules GB/T19022-2004 Measurement management system requirements for measurement processes and measuring equipment ISO010012:2003 Measurement management system requirements for measurement processes and measuring equipment 0IMLD10:1984 Guidelines for Determination of Recalibration Intervals of Measuring Equipment Used in Testing Laboratories
NCSL RP-1:1996 Confirmation and adjustment of calibration intervals When implementing this specification, the relevant parties shall pay attention to the current effective versions of the above-mentioned documents. 3 Terms and Definitions
3.1 Measuring instruments Instruments used alone or together with auxiliary equipment to perform measurements 3.2 [Verification of measuring instruments] Procedures for verifying and confirming whether a measuring instrument meets the statutory requirements, including inspection, marking and issuing a verification certificate.
3.3 Initial verification A verification of a new measuring instrument that has been verified before. 3.4 Subsequent verification Any type of verification after the initial verification of a measuring instrument: 1
1) Mandatory periodic verification:
2) Verification after repair:
JJF 113↓2005
3) Verification within the validity period of periodic verification, whether it is requested by the user or due to some reason that the seal of the validity period is invalid, 3.5 Periodic verification periodic verification The periodic verification of a measuring instrument carried out at regular intervals according to the prescribed procedures. 3.6 Verification period The time interval for periodic verification of a measuring instrument according to the prescribed procedures. 4 Principles for determining the verification period
When formulating or revising the verification procedures for measuring instruments, the verification period of the applicable measuring instruments shall be determined in accordance with the following basic principles; the adjustment of the verification time interval or the confirmation of the calibration time interval of the in-use instruments shall also be carried out in accordance with the following basic principles.
4. Determination principle
When measuring instruments through verification, the verification period should be determined according to the characteristics of the applicable measuring instruments (such as the working principle, structural type and materials used for measuring instruments), the performance requirements of the measuring instruments (such as the maximum allowable error, measurement repeatability and measurement stability) and the use of the measuring instruments (such as environmental conditions, use frequency and maintenance status).
4.2 Determination principle:
When determining the verification period of measuring instruments, first clarify the measurement reliability of the applicable measuring instruments, that is, the target R: the measurement reliability target R% of general measuring instruments (as shown in Figure 1).汪
1 Reliability of measurement is the main indicator of the change of the overall performance of a measuring instrument over time. 2 Reliability target refers to the rate at which the overall performance of a measuring instrument remains within the expected qualified range when it is reconfirmed (re-calibrated after completion).
Test reliability
Reliability target before test R=90%
Measurement reliability change curve
0 Selected calibration time
Time after calibration
Figure 1 Schematic diagram of measurement reliability R () change 2
4.3 Determination principle III
JJF 1139—-2005
The determination of the calibration period of a measuring instrument should be based on the appropriate selection of one or several appropriate methods from the following reaction method or maximum likelihood estimation method for analysis and measurement. Note: The determination of the verification period of the measuring instrument can also refer to the management chart method or the verification mark method ("black" verification method). The determination of the verification period of the measuring instrument should be carried out by appropriately selecting the following appropriate methods for feasibility analysis and mathematical calculation. 5.1 Reaction method The method of adjusting the time interval of the measuring instrument by responding to the most recent verification results and using a simple and direct formula or the simplest calculation method is called the reaction method. The reaction method includes several specific methods such as the step adjustment method, the incremental reaction adjustment method and the interval test method. 5.1.1-week step adjustment method
When a certain type of measuring instrument has been put into use for a clear initial time interval (or a subsequent calibration after a time interval), if its overall performance exceeds the specified measurement reliability target R after reconfirmation, consideration should be given to appropriately shortening the calibration time interval for this type of measuring instrument: if its overall performance exceeds the specified measurement reliability target after reconfirmation, consideration can be given to appropriately extending the calibration time interval for the measuring instrument, or the original calibration time interval can be kept unchanged. When this method is used for adjustment, the increment (extension) or reduction (shortening) of the calibration time is generally a fixed number of months, which is gradually increased or reduced in steps; and the time interval increment coefficient is generally smaller than the time interval reduction coefficient 6
1 The determination of the initial interval can refer to the calibration period determined by similar measuring instruments, and compare the measurement availability objectives, performance requirements, usage conditions, environmental conditions and calibration methods of the measuring instruments; it can also be determined by analyzing the design structure, energy requirements and usage of the measuring instruments, and selecting engineering analysis and confirmation after listening to the suggestions of the manufacturer.
2 Once the alarm is turned on, the time interval increase coefficient is smaller than the time interval reduction coefficient: the time interval to be adjusted (or the interval to be reduced)
△(or)x
where is the opposite time before adjustment:
For example: take = 0.10. =0.50, 1. =6 and, then the time interval for adjustment is 4=×. =0.6 month 1 month, the time interval to be adjusted is reduced by =b×, =3 complete
3 The correct step adjustment method has a faster response speed and lower cost. It is easier to implement the required adjustment rate for the verification time of the measuring instrument: However, this method requires multiple adjustments before it can stabilize to the expected reliability standard R
For example: After confirming that the initial time interval of a certain measuring instrument is s=3, 6, 12, 18, 60 months, the measurement reliability scale H is more than 0%, through the evaluation of the test sample periodic inspection results, the adjustment of the verification time interval is shown in Table 1,
5.1.2 Incremental response adjustment method
JJF 1139—2005
The basic procedures and methods of the incremental response adjustment method are similar to those of the fixed step adjustment method, except that the confirmation of the test time interval and the confirmation of the time increment required for adjustment are closely related to the selection and change of the relevant parameters before adjustment.
Test time interval adjustment table
Test pass rate
%~ 95%
Time interval adjustment
Initial time interval
Interval shortened to
Interval unchanged
Interval extended to
Test interval performed/month
The relationship between the adjusted test time interval I and the time interval before adjustment is: I =IL [1+A. (-R)l-. (R)r.]
The relationship between the time increment △ required for adjustment and the time increment A before adjustment is: Ax-1
A. = 2y.*.
Wherein: I-mth time interval to be determined;
4.-time increment/decrement for the mth adjustment; R——measurement reliability target;
4=1, r= 1
(5 - 1)
(5 - 2)
Ya calculation factor, when the mth verification is qualified, y=1; when the mth verification is unqualified, ym=0;-verification or time interval adjustment sequence number.
Note: The calculation procedure of the incremental response adjustment method is simple, and it is also relatively easy to implement for determining the verification period of the meter and adjusting the verification time interval of the meter in use; the adjustment of the time interval is different from the step adjustment method, and it is determined based on isolated verification results: the basis is insufficient, and it takes a long time to adjust the correct verification. 5.1.3 Interval test method The difference between the interval test method and the fixed step adjustment method and the incremental response adjustment method is that it does not make a retrograde adjustment to the time interval based on the result of a single test, but calculates a level limit to determine whether the test result is significantly higher or lower than the measurement reliability standard R. If yes, the time interval is adjusted; if not, no adjustment is made. The level limit of the interval test method can be obtained by the following formula: 71.1
R(1 - Rn)hR = α
.th!(n- h!
Wherein: R
K(n- A)Ri(1 - Rt)\* = u
R's significance level limit;
R's significance level lower limit;
the number of samples of the measurement group tested in the interval; (5- 3)
..[F 1139—2(05
e——the number of qualified samples observed in 1 interval; a——the significance level of the interval test (α is generally taken as 30%). If the measurement reliability standard R is not between R and R, the time interval needs to be adjusted; the adjustment can be carried out according to the interval collapse extension method and the interval interpolation method respectively. The interval interpolation method is to first obtain the interval 1 by formula (5-5). The reliability R at the time of measurement is calculated, and then the time interval to be adjusted is obtained by using the exponential reliability model formula (5-7) from R: the qualified number of tests in the interval
R. = - the total number of tests in the interval
1, = R
(5-6)
If R. is lower than the measurement reliability target, then 1<1., and the immediate interval is shortened: if R. is higher than R, then 1>1, and the immediate interval is extended. If the time interval is adjusted too much according to formula (5-5) and formula (5-6), that is, the reliability observed under the extended interval is significantly lower than the standard, then the interval needs to be adjusted to the previous time interval and the new time interval according to the interval interpolation method! ! Point, that is: lo+1,
If the adjusted time interval is still too long or too regular, the new interval is: Ia + 12
(5 - 7)
Note: The time interval adjustment by the test method has some advantages of mathematical statistics, and the operating cost is also low; however, the verification time interval needs to be strictly controlled, and the correctness of the initial calculation is still under discussion. 5.2 Maximum Likelihood Estimation Method
The maximum likelihood estimation method is to study and evaluate the state of the measured instrument exceeding the allowable error through the probability distribution of the likelihood function, and finally determine the verification time interval of the measuring instrument. The maximum likelihood estimation method is based on mathematical statistics and a large amount of data analysis
Since this method is based on data analysis, when using the maximum likelihood estimation method to confirm and adjust the time interval, attention should be paid to the validity, consistency and continuity of the data used. There are three specific algorithms for the maximum likelihood estimation method: the classical method. Binomial method New time method: 5.2.1 Classical method
The classical method is based on the classical reliability analysis to construct the likelihood function: Assuming that the measurement uncertainty of the measuring instrument changes according to the exponential function, according to the out-of-tolerance situation of the measuring instrument and the assumption that its measurement uncertainty grows exponentially, first infer the time when the instrument begins to overtime, thereby constructing its likelihood function: L = If(L12)]*[R(L)] x
[1, the first verification failed
In the formula:
=lo, the first verification passed
n——the number of observed verification samples;
1—the first observed verification time.
(5 - 9)
JJF 1139—2005
According to the assumption that the reliability function R (1.) is an exponential function: R()=e
According to the assumption that the failure time probability distribution function is also an exponential function: (112) =Ae-uy2
After taking the logarithm of the quasi-function:
After taking partial derivatives on both sides;
When L=0:
In the formula: =
x,nf(1/2):+
>x,Ina +
(1 - X)In[R()
xx,-x1
, the total number of observed unqualified tests; "=(5 -10)
(5- 11)
(5 - 12)
,,, the total number of observed test times
is obtained by the above method. The coefficients of the reliability function and the reliability function are obtained, and then the required test interval can be determined by determining the reliability target. Note: The use of the classical method for statistical calculation can achieve a better reliability target, and the execution is simple and the cost is low. However, the weakness of this method is that the reliability model described is limited to the exponential model. 5.2.2 Binomial method
Unlike the classical method, the binomial method is not limited to It is not limited to a single reliability model, nor is it hindered by unknown out-of-tolerance time! However, it requires a large-scale time analysis system. The term method is used for the adjustment and determination of various reliability models and various types of meter verification time intervals, but its operating cost is high. It not only requires a high level of system analysis, but also requires statistical expertise. For specific operation analysis, please refer to NCSI.RP II 1: 1996 Appendix C_5.2.3 Update Time Method
The new time method is a refinement of the term method. Its specific method can be referred to NCSL for details. RP-1:1996 Appendix D. Note: The maximum likelihood estimation based on the statistical analysis of large data is the most advanced technology for determining the time interval between calibrations. However, to operate this technology well, it is necessary not only to consider the collection, storage and analysis of a large amount of data, but also to seriously consider the input and output of the final cost, as well as the requirements for the technical requirements of the operating personnel.
Appendix A
JJF 1139-205
Calculation example using the incremental response adjustment method The initial calibration time interval of a certain measuring instrument is 45 days, the measurement reliability target is 90%, and the calibration results are:
Calibration order
Calibration results
Out of tolerance failure
Out of tolerance failure
After the first unsatisfactory calibration, the incremental response adjustment method is used for adjustment: y1 = 0, 4,
216-1 =0.5
f=45×[1+0.5 ×(-0.9)-0 × (0.9)\1 =24.75==25 (days) The second test passed
Y=, A, =
-21l-m
1, =25×[1 +0.25×(-0.9)-1×(0.9)]=30.625±31 () The third test passed
Y3 =1, A, :
1, = 31 × [1 +0.25 × ( -0.9)1-t × (0.9)[1 = 37.975 -38 (days) The fourth test passed
20:2=0.25
y. =1, 4 =
14 =38 × [1+0.25× (- 0.9)*-1 × (0.9)] - 46.55-47 (±)Fifth test passed
1, =47×[1+0.25 ×(-0.9)1-1 × (0.9) = 57.575-58 (days)Sixth test failedWww.bzxZ.net
JF 1139—2005
y =0, A. = 2m-1 ?
1g=58 × [1+0.125× (-0.9)-0× (0.9)° =51.475=51 (days)Sixth test passed
J, = 1, 3, =
(-0.9)*-1×(0.9)1=53.869-54(days)1 =51 × 1+0.0625 x
The eighth inspection passed
Yr =1. A. =
1,-54 × _1+0.0625×(-0.9)× (0.9)-- 57.038-57 (days)
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.