CN104156558B - A kind of conforming determination methods of different chemical composition analysis method result - Google Patents

Abstract

The invention belongs to analysis of chemical elements technology, is related to a kind of conforming determination methods of different chemical composition analysis method result.The method includes test of outlier, three steps of double sample precision consistency check and double sample meansigma methodss consistency check.The present invention is the own characteristic that chemical analyses specialty is fully taken into account on the basis of Principle of Statistics, Principle of Statistics is organically combined with chemical detection data analysiss, two kinds of distinct methods analysis results of certain element-specific in material are processed, to judge that the data obtained by the two analysis methods whether there is concordance in theory.

Description

A kind of conforming determination methods of different chemical composition analysis method result

Technical field

The invention belongs to analysis of chemical elements technology, it is related to that a kind of different chemical composition analysis method result is conforming to be sentenced Disconnected method.

Background technology

During chemical composition analysis, different analysis method data consistency detection problems are particularly important.Carry out In new chemical composition analysis method formulation process, brand-new setting analysis method is judged with the concordance of existing method；Connect in task During receiving, the concordance between the analysis method used by the analysis method of customer requirement and actually detected data judges, is all Essential step.Under existence conditionses, method concordance judges to include two aspect contents：Method precision judges and method Accuracy judges.At present, method precision judges the judgment mode for typically taking relative standard deviation (RSD).Method accuracy By the way of judging the general analysis method using new formulation and two groups of detection datas carrying out directly contrast obtained by original analysis method Carry out.Specific way of contrast be the new analysis method detection data formulated is counted with the detection data of original analysis method it is flat Average ratio is compared with if the new analysis method detection data formulated judges what is newly formulated in the range of former analysis method tolerance Analysis method is consistent with former method accuracy.

The content of the invention

The purpose of the present invention is：The different analysis methods of certain element-specific in material are entered through the resulting data of test Row compares, analyzes, to judge that the data obtained by different analysis methods whether there is concordance in theory.

The technical scheme is that：A kind of conforming determination methods of different chemical composition analysis method result, which is special Levying is, determination methods include test of outlier, double sample precision consistency check and double sample meansigma methodss consistency check three Individual step, comprises the following steps that：

(1), test of outlier：

(1.1) different chemical composition analysis method result the data obtained should meet following requirements：

A) two groups of detection datas obtained by identical material, identity element, different analysis methods；

B) detection data is without technological deficiency；

C) revision of the convention principle for taking " four house six enters five Dan Shuan " is accepted or rejected at detection data end, i.e., and " four houses six enter five considerations, and five Non-zero then enters one afterwards, and after five, all zero regards odd even, before five be it is even should cast out, be very then to enter one before five " 0 be considered as even number；

D) every group of detection data unit is identical；

E) every group of detection data significant digits are identical；

F) each group of detection data number n >=3,

(1.2) one group of data in two groups of detection datas are extracted, data processing is carried out：

(1.2.1) arithmetic average of this group of data is calculated, formula is as follows:

In formula：

x₁,x₂,…x_n- measured value；

The symbol of ∑-expression summation；

- represent from x₁To x_nMeasured value is sued for peace；

(1.2.2) with arithmetic average as true value, calculate the absolute deviation of each data；

Absolute deviation=measured value-arithmetic average

(1.2.3) calculate standard deviation

Standard deviation data expression formula is：

In formula：

S-standard deviation

The arithmetic mean of instantaneous value of-n observation

x_i- any single-measurement value

N- detection data numbers

(1.3) this group of detection data inspection

By the detection data of (1.2) by arranging from small to large；

The maximum and minima of detection data are chosen respectively, are calculated as follows respectively

x_i- for the measured value of maximum or minimum

Determine that significance level is α=0.05, corresponding marginal value is found in the T value table of Grubbs Law, mark It is designated as G_Table；

Judge：Work as G_n＞ G_TableWhen, judge measured value x_iFor exceptional value；Otherwise no exceptional value；

(1.4) rejecting abnormalities value

If x_iFor exceptional value, then need to reject, if detection data number n >=3 after rejecting abnormalities value, should by remaining data after It is continuous to process by the step of (1.2)-(1.3), till being as good as constant value；After rejecting abnormalities value, if data amount check n ＜ 4, calculate Terminate, it is impossible to carry out data consistency judgement；

(1.5) inspection of another group of detection data is carried out by (1.2)-(1.4) step；

(2), double sample precision consistency check

(2.1) by two groups of data of the rejecting abnormalities value after (1) is checked as this step calculating data, and data Number n meets n >=3；

(2.2) by formula in entry (1.2.3), the standard deviation S of two groups of data is calculated respectively₁And S₂；

(2.3) compare S₁And S₂Size；

(2.4) calculate double sample precision concordance F_Meter：

If S₁＞ S₂, then calculated with equation below

If S₂＞ S₁, then calculated with equation below

(2.5) the degree of freedom V of every group of data variance is calculated respectively_Point, computing formula is：

V_Point=n-1

(2.6) according to the degree of freedom numerical value of variance, in the case of level of significance α=0.05, divide in F check tables or F F values are found in cloth table, F is labeled as_Table

(2.7) judged result is drawn according to institute's value：

If F_Meter≤F_TableDuring value, there is no significant difference, precision in two kinds of different chemical composition Standard of analytical methods deviations Unanimously；

If F_Meter＞ F_TableDuring value, there is significant difference in two kinds of different chemical composition Standard of analytical methods deviations, precision is not Unanimously, calculate and terminate, it is impossible to carry out data consistency judgement；

(3), double sample meansigma methodss consistency check；

(3.1) by by (1), (2) two groups of data after the assay was approved as this step calculating data；

(3.2) calculate the arithmetic average of two groups of data by formula in entry (1.2.1) respectivelyWith；

(3.3) calculate the standard deviation S of two groups of data by formula in entry (1.2.3) respectively₁And S₂；

(3.4) determine degree of freedom V_Always

V_Always=n₁+n₂-2

In formula：

n₁- for first group of data number；

n₂- for second group of data number；

(3.5)S_CloseCalculate

S_CloseComputing formula is：

(3.6) double sample meansigma methodss concordance T_CloseValue is calculated

T_CloseValue computing formula is：

(3.7) according to total degree of freedom V of two groups of detection datas_AlwaysNumerical value, in the case of level of significance α=0.05, in t Inspection tables of critical values or t-distribution table, are labeled as T_Table

(3.8) judged result is drawn according to institute's value：

If T_Close≥T_Table,During value, two kinds of different chemical composition analysis method average results have differences, and average result is in statistics It is inconsistent on, calculate and terminate, judge that this two kinds of analysis method results are inconsistent；

If T_Close＜ T_Table,During value, two kinds of different chemical composition analysis method average results do not have differences, and average result is in system Meter is learned has concordance, judges that this two kinds of analysis method results are consistent.

The method have the advantages that：

1) present invention considers the own characteristic of chemical analyses specialty, and Principle of Statistics is had with chemical detection data analysiss Machine is combined, and defines the judgment criterion for judging whether two kinds of analysis methods are consistent；

2) present invention adopts the treatment principle of " four houses six enter five Dan Shuan " on data are accepted or rejected, compared with " rounding up " method more Rationally, it is to avoid after the revision of the convention, the system of measured value is higher；

3) in abnormality value removing, double sample data precision consistency check, double sample statistical average consistency check In determine be suitable for significance level, so both ensured the requirement of quality control in normal chemical analysis measurement, kept away again Exempt from due to the too strict and caused phenomenon without alternative method of method Data Control；

4) two-sided test method is taken in double sample statistical average consistency check, it is to avoid due to small probability event And the misjudgment for causing；

5) present invention for the corresponding data diversity that different chemical analysis methods are obtained carry out it is directly perceived, flexibly compare and Analysis, so as to reach the purpose of concordance judgement.

Specific embodiment

A kind of conforming determination methods of different chemical composition analysis method result, is characterized in that, determination methods include different Constant value inspection, three steps of double sample precision consistency check and double sample meansigma methodss consistency check, comprise the following steps that：

(1), test of outlier：

(1.1) different chemical composition analysis method result the data obtained should meet following requirements：

A) two groups of detection datas obtained by identical material, identity element, different analysis methods；

B) detection data is without technological deficiency；

C) revision of the convention principle for taking " four house six enters five Dan Shuan " is accepted or rejected at detection data end, i.e., and " four houses six enter five considerations, and five Non-zero then enters one afterwards, and after five, all zero regards odd even, before five be it is even should cast out, be very then to enter one before five " 0 be considered as even number；

D) every group of detection data unit is identical；

E) every group of detection data significant digits are identical；

F) each group of detection data number n >=3,

(1.2) one group of data in two groups of detection datas are extracted, data processing is carried out：

(1.2.1) arithmetic average of this group of data is calculated, formula is as follows:

In formula：

x₁,x₂,…x_n- measured value；

The symbol of ∑-expression summation；

- represent from x₁To x_nMeasured value is sued for peace；

(1.2.2) with arithmetic average as true value, calculate the absolute deviation of each data；

Absolute deviation=measured value-arithmetic average

(1.2.3) calculate standard deviation

Standard deviation data expression formula is：

In formula：

S-standard deviation

The arithmetic mean of instantaneous value of-n observation

x_i- any single-measurement value

N- detection data numbers

(1.3) this group of detection data inspection

By the detection data of (1.2) by arranging from small to large；

The maximum and minima of detection data are chosen respectively, are calculated as follows respectively

x_i- for the measured value of maximum or minimum

Determine that significance level is α=0.05, corresponding marginal value is found in the T value table of Grubbs Law, mark It is designated as G_Table；

Judge：Work as G_n＞ G_TableWhen, judge measured value x_iFor exceptional value；Otherwise no exceptional value；

(1.4) rejecting abnormalities value

If x_iFor exceptional value, then need to reject, if detection data number n >=3 after rejecting abnormalities value, should by remaining data after It is continuous to process by the step of (1.2)-(1.3), till being as good as constant value；After rejecting abnormalities value, if data amount check n ＜ 4, calculate Terminate, it is impossible to carry out data consistency judgement；

(1.5) inspection of another group of detection data is carried out by (1.2)-(1.4) step；

(2), double sample precision consistency check

(2.1) by two groups of data of the rejecting abnormalities value after (1) is checked as this step calculating data, and data Number n meets n >=3；

(2.2) by formula in entry (1.2.3), the standard deviation S of two groups of data is calculated respectively₁And S₂；

(2.3) compare S₁And S₂Size；

(2.4) calculate double sample precision concordance F_Meter：

If S₁＞ S₂, then calculated with equation below

If S₂＞ S₁, then calculated with equation below

(2.5) the degree of freedom V of every group of data variance is calculated respectively_Point, computing formula is：

V_Point=n-1

(2.6) according to the degree of freedom numerical value of variance, in the case of level of significance α=0.05, divide in F check tables or F F values are found in cloth table, F is labeled as_Table

(2.7) judged result is drawn according to institute's value：

If F_Meter≤F_TableDuring value, there is no significant difference, precision in two kinds of different chemical composition Standard of analytical methods deviations Unanimously；

If F_Meter＞ F_TableDuring value, there is significant difference in two kinds of different chemical composition Standard of analytical methods deviations, precision is not Unanimously, calculate and terminate, it is impossible to carry out data consistency judgement；

(3), double sample meansigma methodss consistency check；

(3.1) by by (1), (2) two groups of data after the assay was approved as this step calculating data；

(3.2) calculate the arithmetic average of two groups of data by formula in entry (1.2.1) respectivelyWith；

(3.3) calculate the standard deviation S of two groups of data by formula in entry (1.2.3) respectively₁And S₂；

(3.4) determine degree of freedom V_Always

V_Always=n₁+n₂-2

In formula：

n₁- for first group of data number；

n₂- for second group of data number；

(3.5)S_CloseCalculate

S_CloseComputing formula is：

(3.6) double sample meansigma methodss concordance T_CloseValue is calculated

T_CloseValue computing formula is：

(3.7) according to total degree of freedom V of two groups of detection datas_AlwaysNumerical value, in the case of level of significance α=0.05, in t Inspection tables of critical values or t-distribution table, are labeled as T_Table

(3.8) judged result is drawn according to institute's value：

If T_Close≥T_Table,During value, two kinds of different chemical composition analysis method average results have differences, and average result is in statistics It is inconsistent on, calculate and terminate, judge that this two kinds of analysis method results are inconsistent；

If T_Close＜ T_Table,During value, two kinds of different chemical composition analysis method average results do not have differences, and average result is in system Meter is learned has concordance, judges that this two kinds of analysis method results are consistent.

Embodiment one

A kind of conforming determination methods of different chemical composition analysis method result, is characterized in that, determination methods include different Constant value inspection, three steps of double sample precision consistency check and double sample meansigma methodss consistency check, comprise the following steps that：

(1), test of outlier：

(1.1) test the data obtained as follows：

A) material trademark：K4169 high temperature alloys；Measurement element：Al；Measuring method：Cupferron, cupferron point EDTA holds Mensuration measure aluminium content (HB 5220.19-2008)；ICP-AES methods determine aluminium content (Q/6S in Ni-based, Fe Ni matrix high temperature alloy 2205.4-2009)；

B) without technological deficiency, order ticket requires to retain 2 significant digits significant digits detection data, and this detection data is protected 3 significant digits, data are stayed to take the revision of the convention principle of " four houses six enter five Dan Shuan ".Detection data is shown in Table 1：

Table 1：The analysis result of Al elements in K4169 materials

(1.2) one group of data obtained by HB 5220.19-2008 analysis methods are extracted, data processing is carried out；

(1.2.1) arithmetic average of this group of data is calculated, formula is as follows:

(1.2.2) with arithmetic average as true value, calculate the absolute deviation of each data；

Absolute deviation=measured value-arithmetic average

Absolute deviation is followed successively by：-0.014、0.015、-0.009、-0.003、0.008、0.005、0.001、-0.001

(1.2.3) calculate standard deviation

The date expression of standard deviation is：

(1.3) this group of detection data inspection

By the detection data of (1.2) by arranging from small to large：0.612、0.617、0.623、0.625、0.627、0.631、 0.634、0.641

Maximum 0.641 is first chosen, is calculated as follows

Determine that significance level is α=0.05, publish Yuan Guangwu chief editors' in petroleum industry publishing house《Practical credit Analysis》When finding n=8 in one book, 220 page table 9-2, corresponding marginal value G_Table=2.03；

Judge：Because 1.613 ＜ 2.03, X₈It is not exceptional value；

Minima 0.612 is chosen again, is calculated as follows

Determine that significance level is α=0.05, publish Yuan Guangwu chief editors' in petroleum industry publishing house《Practical credit Analysis》When finding n=8 in one book, 220 page table 9-2, corresponding marginal value G_Table=2.03；

Judge：Because 1.505 ＜ 2.03, X₁It is not exceptional value；

(1.4) rejecting abnormalities value

Because x₁And x₈It is not exceptional value, so being not required to reject, following calculation procedure can be carried out.

(1.5) another group of detection data is carried out by (1.2)-(1.4) step, is as a result as good as constant value, can be carried out double sample Precision consistency check.

(2), double sample precision consistency check

(2.1) by by (1) check after rejecting abnormalities value two groups of data as this step calculating data；

(2.2) by formula in entry (1.2.3), the standard deviation S of two groups of data is calculated respectively₁And S₂；

S₁=0.00926

S₂=0.01071

(2.3) compare S₁And S₂Size：S₂＞ S₁：

(2.4) calculate double sample precision concordance F_Meter：

Computing formula is as follows：

(2.5) the degree of freedom V of every group of data variance is calculated respectively_Point：

V₁=n-1=8-1=7

V₂=n-1=8-1=7

(2.6) according to the degree of freedom numerical value of variance, in the case of level of significance α=0.05, publish in petroleum industry Society publishes Yuan Guangwu chief editors'《Practical chemistry is analyzed》Corresponding F values, F are found in one book, 228 page table 9-5_Table=4.99

(2.7) judged result is drawn according to institute's value：

Because 1.3377 ＜ 4.99, there was no significant difference for two kinds of Standard of analytical methods deviations, and precision is consistent；

(3), double sample meansigma methodss consistency check；

(3.1) by by (1), (2) inspection after two groups of data as this step calculating data；

(3.2) calculate the arithmetic average of two groups of data by formula in entry (1.2.1) respectively；

(3.3) calculate the standard deviation S of two groups of data by formula in entry (1.2.3) respectively₁And S₂；

S₁=0.00926；

S₂=0.01071

(3.4) determine degree of freedom V_Always

V_Always=n₁+n₂- 2=8+8-2=14

(3.5)S_CloseCalculate

S_CloseComputing formula is：

(3.6) double sample meansigma methodss concordance T_CloseValue is calculated

T_CloseValue computing formula is：

(3.7) according to total degree of freedom V of two groups of detection datas_AlwaysNumerical value, in the feelings of two-sided test level of significance α=0.05 Under condition, publish Yuan Guangwu chief editors' in petroleum industry publishing house《Practical chemistry is analyzed》Degree of freedom is found in one book, 229 page table 9-9 For 14 when T value, T_Table=2.145

(3.8) judged result is drawn according to institute's value：

Because 2.4 ＞ 2.145, both analysis method the data obtained average results have differences, and average result exists It is statistically inconsistent, calculate and terminate, judge that this two kinds of analysis method results are inconsistent.

Embodiment two

A kind of conforming determination methods of different chemical composition analysis method result, is characterized in that, determination methods include different Constant value inspection, three steps of double sample precision consistency check and double sample meansigma methodss consistency check, comprise the following steps that：

(1), test of outlier：

(1.1) test the data obtained as follows：

A) material trademark：ZL106 aluminium alloys；Measurement element：Ti；Measuring method：Two antipyrine methane spectrophotographys Determine titanium amount (GB/T 6987.12-2001)；ICP-AES determine Cu, Mg, Zn, Cd, Fe, Mn, B, Ti, Zr, V, Ni, Cr content (HB6731.10-2005)；

B) without technological deficiency, order ticket requires to retain 2 significant digits significant digits detection data, and this detection data is protected 3 significant digits, data are stayed to take the revision of the convention principle of " four houses six enter five Dan Shuan ".Detection data is shown in Table 1：

Table 1：The analysis result of Ti elements in ZL106 aluminum alloy materials

(1.2) one group of data obtained by GB/T 6987.12-2001 analysis methods are extracted, data processing is carried out；

(1.2.1) arithmetic average of this group of data is calculated, formula is as follows:

(1.2.2) with arithmetic average as true value, calculate the absolute deviation of each data；

Absolute deviation=measured value-arithmetic average

Absolute deviation is followed successively by：0.001、0.003、0.001、-0.001、0.001、-0.002、-0.001、-0.0010

(1.2.3) calculate standard deviation

The date expression of standard deviation is：

(1.3) this group of detection data inspection

By the detection data of (1.2) by arranging from small to large：0.182、0.183、0.183、0.183、0.185、0.185、 0.185、0.187

Maximum 0.187 is first chosen, is calculated as follows

Determine that significance level is α=0.05, publish Yuan Guangwu chief editors' in petroleum industry publishing house《Practical credit Analysis》When finding n=8 in one book, 220 page table 9-2, corresponding marginal value G_Table=2.03；

Judge：Because 1.875 ＜ 2.03, X₈It is not exceptional value；

Minima 0.182 is chosen again, is calculated as follows

Determine that significance level is α=0.05, publish Yuan Guangwu chief editors' in petroleum industry publishing house《Practical credit Analysis》When finding n=8 in one book, 220 page table 9-2, corresponding marginal value G_Table=2.03；

Judge：Because 1.25 ＜ 2.03, X₁It is not exceptional value；

(1.4) rejecting abnormalities value

Because x₁And x₈It is not exceptional value, so being not required to reject, following calculation procedure can be carried out.

(1.5) another group of detection data is carried out by (1.2)-(1.4) step, is as a result as good as constant value, can be carried out double sample Precision consistency check.

(2), double sample precision consistency check

(2.1) by by (1) check after rejecting abnormalities value two groups of data as this step calculating data；

(2.2) by formula in entry (1.2.3), the standard deviation S of two groups of data is calculated respectively₁And S₂；

S₁=0.0016

S₂=0.0008

(2.3) compare S₁And S₂Size：S₁＞ S₂：

(2.4) calculate double sample precision concordance F_Meter：

Computing formula is as follows：

(2.5) the degree of freedom V of every group of data variance is calculated respectively_Point：

V₁=n-1=8-1=7

V₂=n-1=8-1=7

(2.6) according to the degree of freedom numerical value of variance, in the case of level of significance α=0.05, publish in petroleum industry Society publishes Yuan Guangwu chief editors'《Practical chemistry is analyzed》Corresponding F values, F are found in one book, 228 page table 9-5_Table=4.99

(2.7) judged result is drawn according to institute's value：

Because 4.00 ＜ 4.99, there was no significant difference for two kinds of Standard of analytical methods deviations, and precision is consistent；

(3), double sample meansigma methodss consistency check；

(3.1) by by (1), (2) inspection after two groups of data as this step calculating data；

(3.2) calculate the arithmetic average of two groups of data by formula in entry (1.2.1) respectively；

(3.3) calculate the standard deviation S of two groups of data by formula in entry (1.2.3) respectively₁And S₂；

S₁=0.0016；

S₂=0.0008

(3.4) determine degree of freedom V_Always

V_Always=n₁+n₂- 2=8+8-2=14

(3.5)S_CloseCalculate

S_CloseComputing formula is：

(3.6) double sample meansigma methodss concordance T_CloseValue is calculated

T_CloseValue computing formula is：

(3.7) according to total degree of freedom V of two groups of detection datas_AlwaysNumerical value, in the feelings of two-sided test level of significance α=0.05 Under condition, publish Yuan Guangwu chief editors' in petroleum industry publishing house《Practical chemistry is analyzed》Degree of freedom is found in one book, 229 page table 9-9 For 14 when T value, T_Table=2.145

(3.8) judged result is drawn according to institute's value：

Because 1.538 ＜ 2.145, both analysis method the data obtained average results do not have differences, average to tie Fruit is statistically consistent, judges that this two kinds of analysis method results are consistent.

Embodiment three

A kind of conforming determination methods of different chemical composition analysis method result, is characterized in that, determination methods include different Constant value inspection, three steps of double sample precision consistency check and double sample meansigma methodss consistency check, comprise the following steps that：

(1), test of outlier：

(1.1) test the data obtained as follows：

A) material trademark：K465 high temperature alloys；Measurement element：Mo；Measuring method：EDTA volumetric determination molybdenum contents (HB5220.21-2008)；Rhodanate Their Determination by Spectrophotometry molybdenum content (HB5220.22-2008)；

B) without technological deficiency, order ticket requires to retain 2 significant digits significant digits detection data, and this detection data is protected 3 significant digits, data are stayed to take the revision of the convention principle of " four houses six enter five Dan Shuan ".Detection data is shown in Table 1：

Table 1：The analysis result of Mo elements in K465 high-temperature alloy materials

(1.2) one group of data obtained by HB5220.21-2008 analysis methods are extracted, data processing is carried out；

(1.2.1) calculate the arithmetic average of this group of data:

(1.2.2) with arithmetic average as true value, calculate the absolute deviation of each data；

Absolute deviation=measured value-arithmetic average

Absolute deviation is followed successively by：-0.068、0.029、0.015、-0.006、-0.081、-0.020、0.029、0.029

(1.2.3) calculate standard deviation

The date expression of standard deviation is：

(1.3) this group of detection data inspection

By the detection data of (1.2) by arranging from small to large：1.422、1.435、1.495、1.509、1.530、1.544、 1.544、1.544

Maximum 1.544 is first chosen, is calculated as follows

Determine that significance level is α=0.05, publish Yuan Guangwu chief editors' in petroleum industry publishing house《Practical credit Analysis》When finding n=8 in one book, 220 page table 9-2, corresponding marginal value G_Table=2.03；

Judge：Because 0.829 ＜ 2.03, X₈It is not exceptional value；

Minima 1.422 is chosen again, is calculated as follows

Determine that significance level is α=0.05, publish Yuan Guangwu chief editors' in petroleum industry publishing house《Practical credit Analysis》When finding n=8 in one book, 220 page table 9-2, corresponding marginal value G_Table=2.03；

Judge：Because 1.64 ＜ 2.03, X₁It is not exceptional value；

(1.4) rejecting abnormalities value

Because x₁And x₈It is not exceptional value, so being not required to reject, following calculation procedure can be carried out.

(1.5) another group of detection data is carried out by (1.2)-(1.4) step, is as a result as good as constant value, can be carried out double sample Precision consistency check.

(2), double sample precision consistency check

(2.1) by by (1) check after rejecting abnormalities value two groups of data as this step calculating data；

(2.2) by formula in entry (1.2.3), the standard deviation S of two groups of data is calculated respectively₁And S₂；

S₁=0.0494

S₂=0.0201

(2.3) compare S₁And S₂Size：S₁＞ S₂：

(2.4) calculate double sample precision concordance F_Meter：

Computing formula is as follows：

(2.5) the degree of freedom V of every group of data variance is calculated respectively_Point：

V₁=n-1=8-1=7

V₂=n-1=8-1=7

(2.6) according to the degree of freedom numerical value of variance, in the case of level of significance α=0.05, publish in petroleum industry Society publishes Yuan Guangwu chief editors'《Practical chemistry is analyzed》Corresponding F values, F are found in one book, 228 page table 9-5_Table=4.99

(2.7) judged result is drawn according to institute's value：

Because 6.04 ＞ 4.99, two kinds of Standard of analytical methods deviations have significant difference, precision is inconsistent；This When need not carry out t-distribution inspection, calculate and terminate, illustrate this in two detection method result it is inconsistent.

Claims (4)

1. a kind of conforming determination methods of different chemical composition analysis method result, is characterized in that, determination methods include exception Value inspection, three steps of double sample precision consistency check and double sample meansigma methodss consistency check, comprise the following steps that：

(1), test of outlier：

(1.1) different chemical composition analysis method result the data obtained should meet following requirements：

A) two groups of detection datas obtained by identical material, identity element, different analysis methods；

B) detection data is without technological deficiency；

C) revision of the convention principle for taking " four houses six enter five Dan Shuan " is accepted or rejected at detection data end, i.e., " four houses six enter five considerations, non-after five Zero is entered one, and after five, all zero regards odd even, before five be it is even should cast out, be very then to enter one before five " 0 be considered as even number；

D) every group of detection data unit is identical；

E) every group of detection data significant digits are identical；

F) each group of detection data number n >=3,

(1.2) one group of data in two groups of detection datas are extracted, data processing is carried out：

(1.2.1) arithmetic average of this group of data is calculated, formula is as follows:

$\stackrel{\‾}{x}=\frac{x_1+x_2+x_3+\mathrm{...}+x_n}{n}=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}{x}_i$

In formula：

x₁,x₂,…x_n- measured value；

The symbol of ∑-expression summation；

- represent from x₁To x_nMeasured value is sued for peace；

(1.2.2) with arithmetic average as true value, calculate the absolute deviation of each data；

Absolute deviation=measured value-arithmetic average

(1.2.3) calculate standard deviation

Standard deviation data expression formula is：

$S=\sqrt{\frac{\underset{i=1}{\overset{n}{\Σ}}{(x_i-\stackrel{\‾}{x})}^2}{n-1}}$

In formula：

S-standard deviation

The arithmetic mean of instantaneous value of-n observation

x_i- any single-measurement value

N- detection data numbers

(1.3) this group of detection data inspection

By the detection data of (1.2) by arranging from small to large；

The maximum and minima of detection data are chosen respectively, are calculated as follows respectively

$G_n=\frac{|x_i-\stackrel{\‾}{x}|}{S}$

x_i- for the measured value of maximum or minimum

Determine that significance level is α=0.05, corresponding marginal value is found in the T value table of Grubbs Law, be labeled as G_Table；

Judge：Work as G_n＞ G_TableWhen, judge measured value x_iFor exceptional value；Otherwise no exceptional value；

(1.4) rejecting abnormalities value

If x_iFor exceptional value, then need to reject, if detection data number n >=3 after rejecting abnormalities value, remaining data should be continued to press (1.2)-(1.3) the step of, is processed, till being as good as constant value；After rejecting abnormalities value, if data amount check n ＜ 4, calculate eventually Only, it is impossible to carry out data consistency judgement；

(1.5) inspection of another group of detection data is carried out by (1.2)-(1.4) step；

(2), double sample precision consistency check

(2.1) by two groups of data of the rejecting abnormalities value after (1) is checked as this step calculating data, and data amount check N meets n >=3；

(2.2) by formula in entry (1.2.3), the standard deviation S of two groups of data is calculated respectively₁And S₂；

(2.3) compare S₁And S₂Size；

(2.4) calculate double sample precision concordance F_Meter：

If S₁＞ S₂, then calculated with equation below

If S₂＞ S₁, then calculated with equation below

(2.5) the degree of freedom V of every group of data variance is calculated respectively_Point, computing formula is：

V_Point=n-1

(2.6) according to the degree of freedom numerical value of variance, in the case of level of significance α=0.05, in F check tables or F-distribution table In find F values, be labeled as F_Table；

(2.7) judged result is drawn according to institute's value：

If F_Meter≤F_TableDuring value, there is no significant difference in two kinds of different chemical composition Standard of analytical methods deviations, precision is consistent；

If F_Meter＞ F_TableDuring value, there is significant difference in two kinds of different chemical composition Standard of analytical methods deviations, precision is inconsistent, Calculate and terminate, it is impossible to carry out data consistency judgement；

(3), double sample meansigma methodss consistency check；

(3.1) by by (1), (2) two groups of data after the assay was approved as this step calculating data；

(3.2) calculate the arithmetic average of two groups of data by formula in entry (1.2.1) respectivelyWith；

(3.2) calculate the standard deviation S of two groups of data by formula in entry (1.2.3) respectively₁And S₂；

(3.3) determine degree of freedom V_Always

V_Always=n₁+n₂-2

In formula：

n₁- for first group of data number；

n₂- for second group of data number；

(3.4)S_CloseCalculate

S_CloseComputing formula is：

(3.5) double sample meansigma methodss concordance T_CloseValue is calculated

T_CloseValue computing formula is：

(3.6) according to total degree of freedom V of two groups of detection datas_AlwaysNumerical value, in the case of level of significance α=0.05, checks in t T value is found in tables of critical values or t-distribution table, T is labeled as_Table

(3.7) judged result is drawn according to institute's value：

If T_Close≥T_Table, during value, two kinds of different chemical composition analysis method average results have differences, and average result is statistically It is inconsistent, calculate and terminate, judge that this two kinds of analysis method results are inconsistent；

If T_Close＜ T_Table, during value, two kinds of different chemical composition analysis method average results do not have differences, and average result is in statistics It is upper that there is concordance, judge that this two kinds of analysis method results are consistent.

2. conforming determination methods of a kind of different chemical composition analysis method result according to claim 1, its feature It is, number n >=4 of described every group of experimental data.

3. conforming determination methods of a kind of different chemical composition analysis method result according to claim 1, its feature It is, when described determination methods selection sample is more than 2, with 2 as a judging unit, in a judging unit There is exceptional value in group detection data, after the exceptional value is disallowable, number n >=3 of detection data, this group of detection data can with it is another Individual detection sample reformulates a judging unit.

4. conforming determination methods of a kind of different chemical composition analysis method result according to claim 1, its feature It is, when described determination methods selection sample is more than 2, with 2 as a judging unit, in a judging unit Group detection data exist this group of data of exceptional value it is disallowable after, another group of detection data in the judging unit is detected with another Sample reformulates a judging unit.