Standard substance quality evaluation method and calculation equipment

Abstract

The embodiment of the invention provides a standard substance quality evaluation method and computing equipment. The standard substance quality evaluation method comprises the following steps: taking actually measured concentration values and nominal concentration values of target components in each to-be-evaluated standard substance as a plurality of coordinate points; taking the minimum loss residual error as a target, and carrying out analysis based on a plurality of coordinate points until an optimal linear regression equation is determined; under the condition that the loss residual error determined based on the optimal linear regression equation is smaller than a residual error threshold value, determining the linear regression fitting degree of each standard substance to be evaluated; and under the condition that the linear regression fitting degree is greater than a set fitting degree threshold value, determining that the corresponding to-be-evaluated standard substance is a non-equivalent standard substance relative to other to-be-evaluated standard substances. According to the scheme, an applicable result evaluation method is provided for a project with an appropriate reference standard, and the problem that objective evaluation cannot be carried out on certain types of existing standard substances without a reference standard substance is solved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of standardization monitoring, and in particular to a method and a computing device for evaluating the quality of a standard substance. Background Art

[0002] A standard substance is a substance whose one or several properties have been determined and is used to calibrate measuring instruments, evaluate measurement methods, or determine the value of material properties. According to the existing classification standards, standard substances are divided into primary standard substances and secondary standard substances. Primary standard substances are substances determined by absolute measurement methods or accurate and reliable methods based on two or more different principles, and are used to evaluate secondary standard substances; secondary standard substances are standard substances that meet the needs of general measurement calibration. Under normal circumstances, secondary standard substances need to use primary standard substances as reference standard substances for quality verification.

[0003] However, due to limited conditions, in some scenarios, primary standard materials are not easily available or are not available, and the quality verification of secondary standard materials cannot be carried out according to the above method. For example, there is currently no corresponding national primary standard material as a reference standard material for multi-component aldehyde and ketone mixtures, and it is impossible to conduct quality verification and evaluation on the existing commercially available secondary standard materials used for measurement and calibration. Summary of the invention

[0004] In order to solve the problem that the quality of commercially available secondary standard substances cannot be evaluated due to the lack of corresponding primary standard substances, the embodiments of the present disclosure provide a standard substance quality evaluation method and a computing device.

[0005] In a first aspect, the present disclosure provides a method for evaluating the quality of a reference material, comprising:

[0006] Test a variety of similar standard substances to be evaluated respectively to determine the actual concentration values ​​of the target components in the various standard substances to be evaluated, and determine the standard uncertainty of the actual concentration values ​​of the target components in the various standard substances to be evaluated;

[0007] Obtain the nominal concentration values ​​and nominal standard uncertainties of the target components in various standard substances to be evaluated;

[0008] Determine multiple coordinate points in a two-dimensional coordinate system based on the measured concentration values ​​and nominal concentration values ​​of the target components in each of the standard substances to be evaluated;

[0009] With the goal of minimizing the loss residual, a linear regression analysis is performed based on the multiple coordinate points until an optimal linear regression equation is determined, wherein the loss residual is calculated and determined based on the coordinate points corresponding to each standard substance to be evaluated, the projection points of the corresponding coordinate points on the linear regression equation, and the standard uncertainty and nominal standard uncertainty of the actual measured concentration value of the target component in each standard substance to be evaluated;

[0010] Determining whether the loss residual determined based on the optimal linear regression equation is less than a residual threshold determined based on the amount of the standard substance to be evaluated;

[0011] When the loss residual determined based on the optimal linear regression equation is less than the residual threshold, determining the linear regression fit of various standard substances to be evaluated;

[0012] When the linear regression fit is greater than a set fit threshold, the corresponding standard substance to be evaluated is determined to be a non-equivalent standard substance relative to other standard substances to be evaluated.

[0013] Optionally, the loss residual is or Where n is the number of standard substances to be evaluated, x i is the nominal concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the x-axis, u(x i ) is the nominal standard uncertainty of the target component in the i-th standard substance to be evaluated, y i is the measured concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the y-axis, u(y i ) is the standard uncertainty of the measured concentration value of the target component in the i-th standard substance to be evaluated.

[0014] Optionally, the determination of the linear regression fit of various types of standard substances includes:

[0015] use or Determine the linear regression fit of the i-th standard substance to be evaluated;

[0016] where x i is the nominal concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the x-axis, u(x i ) is the nominal standard uncertainty of the target component in the i-th standard substance to be evaluated, y i is the measured concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the y-axis, u(y i ) is the standard uncertainty of the measured concentration value of the target component in the i-th standard substance to be evaluated.

[0017] Optionally, the standard substance to be evaluated is a liquid substance, and the nominal concentration value of the target component in the standard substance to be evaluated is the concentration after dilution;

[0018] The method of testing a plurality of the same type of standard substances to be evaluated to determine the actual concentration values ​​of the target components in the various standard substances to be evaluated includes:

[0019] According to the dilution analysis model, the standard substance to be evaluated is diluted to the specified concentration using the calibration standard solution. or Obtaining multiple measured concentration values ​​of target components in the standard substance to be evaluated;

[0020] Calculating an average value of a plurality of measured concentration values ​​as the measured concentration value;

[0021] Where X meas is the measured concentration value of the target component in the standard substance to be evaluated, V1 is the volume of the standard solution to be determined, V2 is the fixed volume of the labeled standard solution, V3 is the volume of the standard substance to be evaluated, V4 is the fixed volume of the standard substance to be evaluated, V5 is the volume of the standard substance to be evaluated during the secondary dilution, V6 is the fixed volume of the standard substance to be evaluated during the secondary dilution, A sample is the peak area of ​​the target component in the standard substance to be evaluated, A stan is the peak area of ​​the target component in the reference standard solution, X crm is the reference value of the target component in the calibration standard solution.

[0022] Optionally, the determination of the standard uncertainty of the measured concentration values ​​of various standard substances to be evaluated includes:

[0023] Determine the uncertainty component u introduced by sample dilution corresponding to each standard substance to be evaluated d , the uncertainty component u introduced by repeated sample injection s , uncertainty component u introduced by analytical instrument drift I and the uncertainty component u introduced by the standard solution r ;

[0024] use The standard uncertainty of the measured concentration value of the target component in the standard substance to be evaluated is calculated.

[0025] Optionally, relative standard uncertainty introduced by sample dilution Where X meas is the detection value of the target component in the standard substance to be evaluated, A1 is the maximum capacity tolerance of the pipette used in the sample dilution process, A2 is the maximum capacity tolerance of the volumetric flask used in the sample dilution process, p is the number of times the pipette is used, m is the number of times the volumetric flask is used, and X measis the measured concentration value of the target component in the standard substance to be evaluated.

[0026] Optionally, uncertainty component introduced by sample replicate injections RSD stan The relative standard deviation of the peak area of ​​the target component when the standard solution is repeatedly measured q times, RSD sample is the relative standard deviation of the peak area of ​​the target component when the standard substance to be evaluated is measured q times repeatedly, X meas is the measured concentration value of the target component in the standard substance to be evaluated.

[0027] Optionally, analyze the uncertainty component introduced by instrument drift RSD re is the relative standard deviation of the peak area of ​​the standard solution during the instrument precision test, ρ is the number of repeated measurements of the standard solution during the instrument precision test, and X meas is the measured concentration value of the target component in the standard substance to be evaluated.

[0028] Optional, uncertainty introduced by standard solution Among them U r is the relative expanded uncertainty of the standard solution, k is the coverage factor, X meas is the measured concentration value of the target component in the standard substance to be evaluated.

[0029] In a second aspect, an embodiment of the present disclosure provides a computing device, including a processor and a memory, wherein the memory is used to store a computer program; when the computer program is loaded by the processor, the processor executes the following method:

[0030] Based on the results of sample testing of multiple similar standard substances to be evaluated, determine the measured concentration values ​​of the target components in various standard substances to be evaluated, and determine the standard uncertainty of the measured concentration values ​​of the target components in various standard substances to be evaluated;

[0031] Obtain the nominal concentration values ​​and nominal standard uncertainties of the target components in various standard substances to be evaluated;

[0032] Determine multiple coordinate points in a two-dimensional coordinate system based on the measured concentration values ​​and nominal concentration values ​​of the target components in each of the standard substances to be evaluated;

[0033] With the goal of minimizing the loss residual, a linear regression analysis is performed based on the multiple coordinate points until an optimal linear regression equation is determined, wherein the loss residual is constructed based on the coordinate points corresponding to each standard substance to be evaluated, the projection points of the corresponding coordinate points on the linear regression equation, and the standard uncertainty and nominal standard uncertainty of the measured concentration value of the target component in each standard substance to be evaluated;

[0034] Determining whether the loss residual determined based on the optimal linear regression equation is a residual threshold determined based on the amount of the standard substance to be evaluated;

[0035] When the loss residual determined based on the optimal linear regression equation is less than the residual threshold, determining the linear regression fit of various types of standard substances;

[0036] When the linear regression fit is greater than a set fit threshold, the corresponding standard substance to be evaluated is determined to be a non-equivalent standard substance relative to other standard substances.

[0037] In the absence of a primary standard substance as a reference, the embodiment of the present disclosure performs a linear regression analysis based on the measured concentration value of the target component in various standard substances to be evaluated, the standard uncertainty of the measured concentration value, the nominal concentration value and the standard uncertainty. When it is determined that the loss residual corresponding to the optimal linear regression curve is less than the residual threshold, it is determined based on the linear regression fit and the fit threshold of each standard substance to be evaluated whether the corresponding standard substance to be evaluated is a non-equivalent standard substance relative to other standard substances to be evaluated. That is, the present solution provides an applicable result evaluation method for projects with appropriate reference standards, avoiding the problem that some existing types of standard substances cannot be objectively evaluated without reference standards. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art description. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor, including

[0040] Figure 1 is a flow chart of a method for evaluating the quality of a standard substance provided in an embodiment of the present disclosure;

[0041] Figure 2 It is a schematic diagram of the structure of a computing device provided in some embodiments of the present disclosure. DETAILED DESCRIPTION

[0042] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0043] The term "including" and its variations used in this document are open-ended, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description. In this document, relational terms such as "first" and "second" are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that these entities or operations are the same.

[0044] In order to solve the problem that secondary standard substances or similar substances cannot be reasonably evaluated due to the lack of primary standard substances, the present disclosure implements a method for evaluating the quality of standard substances without reference to standard substances. The method provided in this scheme determines whether the quality of each standard substance to be evaluated is reasonable by horizontally comparing a variety of standard substances to be evaluated. The specific evaluation principle will be further analyzed in the following text. The standard substance quality evaluation method provided in the embodiment of the present disclosure is executed by a computing device.

[0045] Figure 1 : is a flow chart of the method for evaluating the quality of a standard substance provided by the embodiment of the present disclosure. Figure 1 As shown, the standard substance quality evaluation method provided in the embodiment of the present disclosure includes S110-S170.

[0046] S110: testing a plurality of the same type of standard substances to be evaluated respectively, determining the actual measured concentration values ​​of the target components in the various standard substances to be evaluated, and determining the standard uncertainty of the actual measured concentration values ​​of the target components in the various standard substances to be evaluated.

[0047] The aforementioned multiple same type of standard substances to be evaluated are standard substances with the same main components and used for the same test calibration or test benchmark determination. For example, the multiple same type of standard substances to be evaluated may be aldehydes and ketones used for calibration of air pollution test equipment.

[0048] In specific implementation, appropriate analytical instruments can be selected according to the concentration range, determination method, etc. of various standard substances to be evaluated, the measurement range of the analytical instrument can be clarified, the standard substances to be evaluated can be diluted to the aforementioned concentration range (if the standard substances to be evaluated are already within the aforementioned concentration range, no further dilution should be performed), and data measurement can be performed to determine the actual measured concentration value of the target component in the standard to be measured and the standard uncertainty of the actual measured concentration value of the target component.

[0049] During the test, the standard substance to be evaluated or the diluted standard substance to be evaluated can be repeatedly injected into the analytical instrument for multiple times to obtain multiple measured concentration values ​​of the target component, and then the arithmetic mean of the aforementioned multiple measured concentration values ​​is calculated as the actual measured concentration value of the target component in the standard substance to be evaluated.

[0050] In the specific implementation, the standard substance to be evaluated is a liquid substance, and the nominal concentration value of the Mubao component in the standard substance to be evaluated is the concentration after dilution. Accordingly, the computing device determines the measured concentration value of the target component in the standard substance to be evaluated according to the corresponding calculation formula through various parameters in the dilution process and parameters measured by the analytical instrument.

[0051] In some embodiments, when the calibration standard solution and the standard substance to be evaluated are diluted to the aforementioned concentration range by only one dilution, the computing device may use Get the measured concentration value, where X meas is the measured concentration value of the target component in the standard substance to be evaluated, V1 is the volume of the standard solution, V2 is the fixed volume of the labeled standard solution, V3 is the volume of the standard substance to be evaluated, V4 is the fixed volume of the standard substance to be evaluated, A sample is the peak area of ​​the target component in the standard substance to be evaluated, A stan is the peak area of ​​the target component in the reference standard solution, X crm is the reference value of the target component in the calibration standard solution.

[0052] In some other embodiments, after the calibration standard solution and the standard substance to be evaluated are diluted to the aforementioned concentration range by primary and secondary dilution, the computing device may use The measured concentration value of the target component in the standard substance to be evaluated is obtained, wherein V5 is the liquid volume taken during the secondary dilution of the standard substance to be evaluated, and V6 is the fixed volume during the secondary dilution of the standard substance to be evaluated.

[0053] The standard uncertainty of the measured concentration value of the target component is used to reflect the possible fluctuation range of the measured concentration value. According to the principle of chemical experiments, each operation process in the aforementioned test process may introduce errors. Correspondingly, the errors introduced by each operation process need to be integrated to determine the standard uncertainty of the measured concentration value of the target component.

[0054] According to the analysis above, the sample dilution process, the sample injection process in the analytical instrument, the measurement drift of the analytical instrument itself, and the concentration of the standard solution itself will all introduce errors, that is, they will all affect the standard uncertainty of the measured concentration value of the target component. Therefore, the computing device first determines the uncertainty component u introduced by the sample dilution corresponding to each standard substance to be evaluated. d , the uncertainty component u introduced by repeated sample injection s , uncertainty component u introduced by analytical instrument drift I and the uncertainty component u introduced by the standard solution r , and then the aforementioned uncertain components are integrated to obtain the standard uncertainty of the measured concentration value of the target component in the standard substance to be evaluated.

[0055] In the specific implementation, the computing device adopts The standard uncertainty of the measured concentration value of the target component is obtained. In practical applications, if the standard substance to be evaluated is not diluted before measurement, the uncertainty component u introduced by the sample dilution in the above formula is d and the uncertainty component u introduced by the standard solution r is 0.

[0056] It should be noted that although the standard uncertainty mentioned here is calculated, considering that relative uncertainty or expanded uncertainty may be obtained in practical applications, the above two uncertainties can be converted into standard uncertainty through conversion, so even if relative uncertainty or expanded uncertainty is obtained, it is equivalent to the standard uncertainty mentioned in the embodiments of the present disclosure. For the sake of simplicity and avoiding technical terminology errors, the embodiments of the present disclosure no longer describe the relative uncertainty or expanded uncertainty of the target component in the standard substance to be evaluated.

[0057] In a specific implementation, the relative standard uncertainty introduced by the aforementioned sample dilution is determined based on the operation process during the sample dilution process and the error characteristics of the experimental instruments used in the operation process. Considering that the experimental instruments used in the sample dilution process include pipettes and volumetric flasks, in a specific implementation, the computing device can use Calculate the relative standard uncertainty introduced by sample dilution. In the above formula, X meas is the detection value of the target component in the standard substance to be evaluated, A1 is the maximum capacity tolerance of the pipette used in the sample dilution process, A2 is the maximum capacity tolerance of the volumetric flask used in the sample dilution process, p is the number of times the pipette is used, m is the number of times the volumetric flask is used, and X meas is the measured concentration value of the target component in the standard substance to be evaluated.

[0058] The uncertainty component introduced by repeated sample injection represents the error introduced during the test. According to the test principle and statistical analysis principle, the uncertainty component introduced by repeated sample injection is RSD stan The relative standard deviation of the peak area of ​​the target component when the standard solution is repeatedly measured q times, RSD sample is the relative standard deviation of the peak area of ​​the target component when the standard substance to be evaluated is measured q times repeatedly, X meas is the measured concentration value of the target component in the standard substance to be evaluated.

[0059] The uncertainty introduced by the drift of the analytical instrument characterizes the error caused by the system characteristics of the analytical instrument. RSD re is the relative standard deviation of the peak area of ​​the standard solution during the instrument precision test, ρ is the number of repeated measurements of the standard solution during the instrument precision test, and X meas is the measured concentration value of the target component in the standard substance to be evaluated.

[0060] The uncertainty introduced by the standard solution represents the error introduced by the standard solution. Among them U r is the relative expanded uncertainty of the standard solution, and k is the coverage factor.

[0061] S120: Obtain the nominal concentration value and nominal standard uncertainty of the target component in various standard substances to be evaluated.

[0062] In a specific implementation, the standard concentration value and standard concentration uncertainty provided in the certificate of the standard substance to be evaluated can be written into the computing device, so that the computing device obtains the nominal concentration value and nominal standard uncertainty of the target component in the standard substance to be evaluated.

[0063] S130: Determine a plurality of coordinate points in a two-dimensional coordinate system based on the measured concentration values ​​and nominal concentration values ​​of the target components in each of the standard substances to be evaluated.

[0064] After obtaining the measured concentration value and the nominal concentration value of the target component in the standard substance to be evaluated, the two are combined and bound, specifically, one concentration value is used as the x-axis coordinate of the two-dimensional coordinate system, and the other concentration value is used as the y-axis coordinate of the two-dimensional coordinate system, so that a coordinate point in the two-dimensional coordinate system can be determined. Correspondingly, for each standard substance to be evaluated, multiple coordinate points can be obtained.

[0065] For convenience, the nominal concentration value of the target component of the standard substance to be evaluated is used as the x-axis coordinate of the two-dimensional coordinate system, and the measured concentration value of the target component of the standard substance to be evaluated is used as the y-axis coordinate of the two-dimensional coordinate system. Of course, the corresponding relationship between the two values ​​and the coordinate axis is swapped, and the essential content is still the same, so the embodiment of the present disclosure will not be further analyzed for this, and you can just participate in the analysis below.

[0066] S140: With the goal of minimizing the loss residual, a linear regression analysis is performed based on multiple coordinate points until an optimal linear regression equation is determined.

[0067] After obtaining the coordinates of the aforementioned multiple coordinate points, the computing device can perform linear regression analysis based on the multiple coordinate points to determine a straight line that meets the set requirements determined by each coordinate point. The aforementioned straight line is represented by the optimal linear regression equation, using the formula y ^ =b0+b1x ^ , where b0 is the intercept of the optimal linear regression equation and b1 is the slope of the optimal linear regression equation.

[0068] In the disclosed embodiment, the aforementioned setting requires the use of loss residual representation, which represents the deviation of each determined coordinate point relative to the determined linear regression equation (including the optimal linear regression equation). In a specific implementation, the aforementioned loss residual is calculated and determined based on the coordinate points corresponding to the standard substance to be evaluated, the projection points of the corresponding coordinate points on the linear regression equation, and the standard uncertainty and nominal standard uncertainty of the measured concentration value of the target component in each standard substance to be evaluated.

[0069] The projection point of the aforementioned corresponding coordinate point on the linear regression equation is the intersection point of the perpendicular line from the coordinate point to the straight line corresponding to the linear regression equation and the straight line after the linear regression equation is determined.

[0070] In a specific implementation, in order to obtain the optimal linear regression equation, the computing device can be implemented using a gradient descent algorithm. Specifically, first randomly determine the initial intercept and initial slope of the linear regression equation, calculate the projection points corresponding to each coordinate point according to the initial intercept and initial slope, calculate the loss residual based on the coordinates of each coordinate point and the corresponding projection point, and the standard uncertainty and nominal uncertainty of the measured concentration value of the target component in each standard substance to be evaluated, and adjust the initial intercept and initial slope according to the set step size until the loss residual is determined to be the minimum, and use the linear regression equation corresponding to the aforementioned minimum loss residual as the optimal linear regression equation.

[0071] In one embodiment, the loss residual can be expressed as or

[0072] It is preferred to use express.

[0073] In the above formula, n is the number of standard substances to be evaluated, x i is the nominal concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the x-axis, u(x i ) is the nominal standard uncertainty of the target component in the i-th standard substance to be evaluated, y i is the measured concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the y-axis, u(y i ) is the standard uncertainty of the measured concentration value of the target component in the i-th standard substance to be evaluated.

[0074] It should be noted that, according to the concept of loss residual, the aforementioned loss residual formula can also be represented by higher-order terms, which can also reflect the concept of loss residual. Therefore, even if the loss residual is represented by other formulas similar to the previous loss residual, it is also acceptable.

[0075] Here we analyze the meaning of the above loss residual representation. According to the linear regression fitting principle, if the distance between the optimal linear regression equation and each coordinate point is smaller, the corresponding and The smaller it is, the higher the linear fit between the measured concentration value of the standard substance to be evaluated and the nominal concentration value is. On the contrary, and The larger the value, the worse the linear fit between the measured concentration range and the nominal concentration of the corresponding standard substance to be evaluated. and The smaller the values ​​are, the higher the consistency of the proportional relationship between the measured concentration values ​​and the nominal concentration values ​​of the various standard substances to be evaluated.

[0076] However, only comparing the above and The difference only characterizes whether the measured concentration values ​​of the standard substance to be evaluated are uniform, and does not consider whether the standard uncertainty of the standard substance to be evaluated is reasonable.

[0077] Based on the above considerations, in the embodiment of the present disclosure, the nominal standard uncertainty u(x i ) and the standard uncertainty of the measured concentration value u(y i ),use with u(x i ) to obtain the first ratio, using and u(y i) obtains the second ratio, and adds the power sum of the first ratio and the second ratio, introduces the uncertainty of the target component of the standard substance to be evaluated into the evaluation of the optimal linear regression equation, and finally makes the optimal linear regression equation take into account the factors of the concentration value and the uncertainty corresponding to the concentration value.

[0078] Here we analyze the physical principle expressed by the above formula. As analyzed above, the horizontal and vertical coordinates of the coordinate point are the measured concentration value and the nominal concentration value of the target component of the standard substance to be evaluated, respectively, and the projection point represents the corrected measured concentration value and the corrected nominal concentration value determined based on the optimal linear regression equation. When the measured concentration value and the nominal concentration value of each standard substance to be evaluated have a consistent proportional relationship, the difference between the corrected measured concentration value and the measured concentration value will not exceed the standard uncertainty of the measured concentration value, and the difference between the corrected nominal concentration value and the nominal concentration value will not exceed the nominal standard uncertainty. That is, under normal circumstances, for the loss residual reflected in the above formula, when the nominal concentration value and the nominal standard uncertainty of each standard substance to be evaluated are consistent with the actual situation of its product, the loss residual will not be too large, and theoretically will not be greater than 2n.

[0079] Based on the above ideas, the embodiment of the present disclosure executes the following S150.

[0080] S150: Determine whether the loss residual determined based on the optimal linear regression equation is less than the residual threshold determined based on the amount of the standard substance to be evaluated; if so; execute S160.

[0081] The residual threshold determined based on the number of standard substances to be evaluated is a predetermined residual threshold. As analyzed above, the maximum setting of the residual threshold is 2n. In specific applications, in order to exclude standard substances to be evaluated that have numerical characteristics that are different from other standard substances to be evaluated (in other words, the nominal concentration value and the nominal standard uncertainty are not qualified), in specific implementations, the aforementioned residual threshold should be reduced. For example, in a specific implementation, the residual threshold is determined based on the number of standard substances to be evaluated minus the number of parameters of the linear regression equation, that is, the residual threshold is 2n-4. Of course, in specific implementations, the residual threshold can also be reasonably determined based on actual usage experience, and is not limited to the aforementioned 2n-4.

[0082] If the loss residual determined based on the optimal linear regression equation is less than the residual threshold determined by the number of standard substances to be evaluated, it is determined that the data determined by the aforementioned linear regression equation is basically credible, that is, the nominal concentration value and nominal standard uncertainty of the majority of the standard substances to be evaluated are reasonable. Subsequently, the evaluation of a single standard substance to be evaluated can be performed.

[0083] S160: Determine the linear regression fit of various standard substances to be evaluated.

[0084] The linear regression fit is a numerical value used to evaluate the degree of fit between the aforementioned coordinate point and the aforementioned optimal linear regression equation, and it is correlated with the distance from the coordinate point to the optimal linear regression equation. In specific implementations, the aforementioned correlation is mostly a positive correlation, that is, if the coordinate point corresponding to the standard substance to be evaluated is farther away from the optimal linear regression equation, the corresponding linear regression fit is greater.

[0085] In a specific implementation, the computing device may use the following computing method to determine the linear regression fit of various standard substances to be evaluated: or According to the above formula, it can be imagined that the linear regression fit is normalized using standard uncertainty to eliminate subsequent problems that cannot be judged due to inconsistent comparison benchmarks.

[0086] S170: When the linear regression fit is greater than a set fit threshold, determining that the corresponding standard substance to be evaluated is a non-equivalent standard substance relative to other standard substances to be evaluated.

[0087] The fitting threshold is a predetermined threshold for whether the quality of a single standard substance to be evaluated is reliable. According to the analysis in the previous article, in the case of the above calculation of the linear regression fitting, it is more reasonable to set the fitting threshold to 2. If the linear regression fitting is greater than the set fitting threshold, it is determined that the corresponding standard substance to be evaluated has a large loss residual, and its quality homogeneity similarity with other standard substances to be evaluated is quite different. Therefore, it is regarded as a non-equivalent standard substance relative to other standard substances to be evaluated, and other standard substances to be evaluated whose linear regression fitting is less than the set fitting threshold are also equivalent standard substances to each other.

[0088] As analyzed above, in the absence of a primary standard substance as a reference, the embodiment of the present disclosure performs a linear regression analysis based on the measured concentration value of the target component in various standard substances to be evaluated, the standard uncertainty of the measured concentration value, the nominal concentration value and the standard uncertainty. When it is determined that the loss residual corresponding to the optimal linear regression curve is less than the residual threshold, it is determined based on the linear regression fit and fit threshold of each standard substance to be evaluated whether the corresponding standard substance to be evaluated is a non-equivalent standard substance relative to other standard substances to be evaluated. It provides an applicable result evaluation method for projects with appropriate reference standards, avoiding the problem that some existing types of standard substances cannot be objectively evaluated.

[0089] The above method has been applied to the quality value consistency evaluation of aldehyde and ketone standard substances in the field of environmental monitoring. The following table is the data obtained by evaluating 10 propionaldehyde standard substances to be evaluated using the above method.

[0090] Table 1 Standard uncertainty introduced by the dilution process

[0091]

[0092]

[0093] Table 2 Standard uncertainty introduced by repeated sample injections

[0094]

[0095] Table 3 Relative standard uncertainty introduced by instrument drift

[0096]

[0097] Table 4 Relative standard uncertainty introduced by calibration solutions

[0098]

[0099]

[0100] Table 5. Example of standard uncertainty and its components of the measured values ​​of the reference material to be evaluated (propionaldehyde)

[0101]

[0102] Table 6 Linear regression fitting data and results of the standard substances to be evaluated

[0103]

[0104]

[0105] In addition to providing the aforementioned labeled substance quality evaluation method, the embodiment of the present disclosure also provides a computing device. Figure 2 is a schematic diagram of the structure of a computing device provided by some embodiments of the present disclosure. Figure 2 As shown, the computing device 200 includes a processor 201 and a memory 202, wherein the memory 202 is used to store a computer program; when the computer program is loaded by the processor 201, the processor 201 executes the aforementioned standard substance quality evaluation method.

[0106] The above are only specific embodiments of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating the quality of a reference material, characterized in that: include: Test a variety of similar standard substances to be evaluated respectively to determine the actual concentration values ​​of the target components in the various standard substances to be evaluated, and determine the standard uncertainty of the actual concentration values ​​of the target components in the various standard substances to be evaluated; Obtain the nominal concentration values ​​and nominal standard uncertainties of the target components in various standard substances to be evaluated; Determine multiple coordinate points in a two-dimensional coordinate system based on the measured concentration values ​​and nominal concentration values ​​of the target components in each of the standard substances to be evaluated; With the goal of minimizing the loss residual, a linear regression analysis is performed based on the multiple coordinate points until an optimal linear regression equation is determined, wherein the loss residual is calculated and determined based on the coordinate points corresponding to each standard substance to be evaluated, the projection points of the corresponding coordinate points on the linear regression equation, and the standard uncertainty and nominal standard uncertainty of the actual measured concentration value of the target component in each standard substance to be evaluated; Determining whether the loss residual determined based on the optimal linear regression equation is less than a residual threshold determined based on the amount of the standard substance to be evaluated; When the loss residual determined based on the optimal linear regression equation is less than the residual threshold, determining the linear regression fit of various standard substances to be evaluated; When the linear regression fit is greater than a set fit threshold, the corresponding standard substance to be evaluated is determined to be a non-equivalent standard substance relative to other standard substances to be evaluated.

2. The method according to claim 1, characterized in that: The loss residual is taken as or Where n is the number of standard substances to be evaluated, x i is the nominal concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the x-axis, u(x i ) is the nominal standard uncertainty of the target component in the i-th standard substance to be evaluated, y i is the measured concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the y-axis, u(y i ) is the standard uncertainty of the measured concentration value of the target component in the i-th standard substance to be evaluated.

3. The method according to claim 1, characterized in that The linear regression fitting of various types of standard substances is determined, including: use or Determine the linear regression fit of the i-th standard substance to be evaluated; where x i is the nominal concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the x-axis, u(x i ) is the nominal standard uncertainty of the target component in the i-th standard substance to be evaluated, y i is the measured concentration value of the target component in the i-th standard substance to be evaluated, is the coordinate value of the projection point corresponding to the i-th standard substance to be evaluated on the y-axis, u(y i ) is the standard uncertainty of the measured concentration value of the target component in the i-th standard substance to be evaluated.

4. The method according to any one of claims 1 to 3, characterized in that: The standard substance to be evaluated is a liquid substance, and the nominal concentration value of the target component in the standard substance to be evaluated is the concentration after dilution; The method of testing a plurality of the same type of standard substances to be evaluated to determine the actual concentration values ​​of the target components in the various standard substances to be evaluated includes: According to the dilution analysis model, the standard substance to be evaluated is diluted to the specified concentration using the calibration standard solution. or Obtaining multiple measured concentration values ​​of target components in the standard substance to be evaluated; Calculating an average value of a plurality of measured concentration values ​​as the measured concentration value; Where X meas is the measured concentration value of the target component in the standard substance to be evaluated, V1 is the volume of the standard solution to be determined, V2 is the fixed volume of the labeled standard solution, V3 is the volume of the standard substance to be evaluated, V4 is the fixed volume of the standard substance to be evaluated, V5 is the volume of the standard substance to be evaluated during the secondary dilution, V6 is the fixed volume of the standard substance to be evaluated during the secondary dilution, A sample is the peak area of ​​the target component in the standard substance to be evaluated, A stan is the peak area of ​​the target component in the reference standard solution, X crm is the reference value of the target component in the calibration standard solution.

5. The method according to any one of claims 1 to 3, characterized in that: The standard uncertainty of the measured concentration values ​​of various standard substances to be evaluated is determined, including: Determine the uncertainty component u introduced by sample dilution corresponding to each standard substance to be evaluated d , the uncertainty component u introduced by repeated sample injection s , uncertainty component u introduced by analytical instrument drift I and the uncertainty component u introduced by the standard solution r ; use The standard uncertainty of the measured concentration value of the target component in the standard substance to be evaluated is calculated.

6. The method according to claim 5, characterized in that Relative standard uncertainty introduced by sample dilution Where X meas is the detection value of the target component in the standard substance to be evaluated, A1 is the maximum capacity tolerance of the pipette used in the sample dilution process, A2 is the maximum capacity tolerance of the volumetric flask used in the sample dilution process, p is the number of times the pipette is used, m is the number of times the volumetric flask is used, and X meas is the measured concentration value of the target component in the standard substance to be evaluated.

7. The method according to claim 5, characterized in that Uncertainty components introduced by repeated sample injections RSD stan The relative standard deviation of the peak area of ​​the target component when the standard solution is repeatedly measured q times, RSD sample is the relative standard deviation of the peak area of ​​the target component when the standard substance to be evaluated is measured q times repeatedly, X meas is the measured concentration value of the target component in the standard substance to be evaluated.

8. The method according to claim 5, characterized in that Uncertainty components introduced by analytical instrument drift RSD re is the relative standard deviation of the peak area of ​​the standard solution during the instrument precision test, ρ is the number of repeated measurements of the standard solution during the instrument precision test, and X meas is the measured concentration value of the target component in the standard substance to be evaluated.

9. The method according to claim 5, characterized in that Uncertainty introduced by standard solution Among them U r is the relative expanded uncertainty of the standard solution, k is the coverage factor, X meas is the measured concentration value of the target component in the standard substance to be evaluated.

10. A computing device, characterized in that The invention comprises a processor and a memory, wherein the memory is used to store a computer program; when the computer program is loaded by the processor, the processor executes the following method: Based on the results of sample testing of multiple similar standard substances to be evaluated, determine the measured concentration values ​​of the target components in various standard substances to be evaluated, and determine the standard uncertainty of the measured concentration values ​​of the target components in various standard substances to be evaluated; Obtain the nominal concentration values ​​and nominal standard uncertainties of the target components in various standard substances to be evaluated; Determine multiple coordinate points in a two-dimensional coordinate system based on the measured concentration values ​​and nominal concentration values ​​of the target components in each of the standard substances to be evaluated; With the goal of minimizing the loss residual, a linear regression analysis is performed based on the multiple coordinate points until an optimal linear regression equation is determined, wherein the loss residual is constructed based on the coordinate points corresponding to each standard substance to be evaluated, the projection points of the corresponding coordinate points on the linear regression equation, and the standard uncertainty and nominal standard uncertainty of the measured concentration value of the target component in each standard substance to be evaluated; Determining whether the loss residual determined based on the optimal linear regression equation is a residual threshold determined based on the amount of the standard substance to be evaluated; When the loss residual determined based on the optimal linear regression equation is less than the residual threshold, determining the linear regression fit of various types of standard substances; When the linear regression fit is greater than a set fit threshold, the corresponding standard substance to be evaluated is determined to be a non-equivalent standard substance relative to other standard substances.

Images