A performance evaluation device and method for a medicinal material storage container
By constructing experimental datasets and conducting multi-level preprocessing and evaluation, the systematic and scientific issues of performance evaluation of containers for the transport of medicinal materials were resolved, enabling a more comprehensive performance evaluation, improving the accuracy and reliability of the evaluation results, guiding container design and optimization, and enhancing transportation and storage efficiency.
Patent Information
- Application Number
- CN202610578443.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-29
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2046-04-29
AI Technical Summary
Existing methods for evaluating the performance of containers used for transporting medicinal materials lack systematicity and scientific rigor, failing to comprehensively and accurately reflect the overall performance of containers under different test conditions. Furthermore, limitations exist in data processing and analysis, resulting in constraints on the reliability and accuracy of the evaluation results.
An experimental dataset was constructed, and multi-level preprocessing and evaluation were performed, including anomaly detection, normalization, boundary checks, multiple linear regression analysis, F-test statistic calculation, and difference fusion calculation. The fusion evaluation was performed through the significance discrimination result set, and the first and second matrices to be evaluated were constructed and weighted summation was performed to scientifically identify the factors affecting the performance of the medicinal material protection container.
This enabled a comprehensive, systematic, and scientific evaluation of the performance of containers for storing medicinal materials, improving the accuracy and reliability of the evaluation results, providing scientific basis to guide container design and optimization, and enhancing transportation and storage efficiency.
Smart Images

Figure CN122548144B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of data modeling and experimental design and evaluation, specifically to a performance evaluation device and method for a medicinal material storage container. Background Technology
[0002] The impact of transportation and storage conditions on the quality of medicinal materials is receiving increasing attention. Traditional methods of transporting and storing medicinal materials often fail to effectively guarantee their quality and stability, especially during long-distance transport and under complex environmental conditions. In recent years, medicinal material storage containers have become a new type of transportation and storage tool, and their ability to provide controlled environmental conditions has led to their widespread adoption.
[0003] However, current performance evaluation methods for containers used to transport medicinal materials still have shortcomings. Traditional evaluation methods mainly rely on empirical judgment or testing of single indicators, lacking systematicity and scientific rigor. These methods cannot comprehensively and accurately reflect the overall performance of containers under different test conditions, nor can they effectively identify factors that significantly affect the quality of medicinal materials. In addition, existing methods have limitations in data processing and analysis, such as the inability to effectively handle outlier data and the lack of scientific estimation of the number of tests, which limits the reliability and accuracy of the evaluation results.
[0004] Therefore, developing a scientific, systematic, and efficient method for evaluating the performance of containers used for transporting medicinal materials is of great significance for ensuring the quality of medicinal materials and improving transportation and storage efficiency. Summary of the Invention
[0005] This invention primarily addresses the problem of how to achieve a scientific, systematic, and efficient performance evaluation method for medicinal material supply containers. This invention discloses a performance evaluation device and method for medicinal material supply containers.
[0006] In a first aspect, the present invention discloses a method for evaluating the performance of a medicinal material supply container, comprising: S1, Obtain the test dataset for the medicinal material supply container; the test dataset includes a test condition variable dataset and a test response variable dataset; the test condition variable dataset includes a subset of test condition variable data; the test response variable dataset includes a subset of test response variable data; each subset of test response variable data corresponds to a type of test response variable data, which is the test value of a type of technical indicator obtained in each test when testing the medicinal material supply container based on the test condition variable data; the subset of test condition variable data is the collection of test condition variable data for a single test; S2, preprocess the test dataset to obtain a preprocessed test dataset; S3, perform sample evaluation processing on the preprocessed experimental dataset to obtain a set of significance discrimination results and standard values for the number of trials; S4. According to the standard value of the number of tests, conduct a corresponding number of tests on the medicinal material supply container to obtain a set of technical indicator test result information; the set of technical indicator test result information includes the test values of all types of technical indicators obtained in each test; S5. Based on the set of significance discrimination results, the set of test results of the technical indicators is fused and evaluated to obtain the performance evaluation result value of the medicinal material support container.
[0007] The preprocessing of the test dataset to obtain a preprocessed test dataset includes: S21, perform anomaly detection processing on the test dataset to obtain a first data set; S22, normalize the first data set to obtain the second data set; S23, perform boundary checks on the second dataset to obtain a preprocessed test dataset.
[0008] The step of performing sample evaluation processing on the preprocessed experimental dataset to obtain a set of significance discrimination results and standard values for the number of trials includes: S31, Perform statistical feature discrimination processing on each type of experimental response variable data in the preprocessed experimental dataset to obtain the statistical discrimination result of each type of experimental response variable data; S32, perform optimal trial number estimation processing on the preprocessed test dataset to obtain the standard value of the number of trials.
[0009] The step involves performing statistical feature discrimination processing on each type of experimental response variable data in the preprocessed experimental dataset to obtain the statistical discrimination result for each type of experimental response variable data, including: S311, take the experimental condition variable dataset in the preprocessed experimental dataset as the multivariate independent variable, and take the experimental response variable data subset corresponding to each type of experimental response variable data in the experimental response variable dataset in the preprocessed experimental dataset as the dependent variable, and perform multivariate linear regression analysis on the multivariate independent variable and dependent variable to obtain the regression analysis model of each type of experimental response variable data. S312, using the regression analysis model of each type of experimental response variable data, the experimental condition variable dataset is calculated and processed to obtain the corresponding set of output values; S313, Perform F-test statistic calculation on the output value set and the corresponding experimental response variable data subset to obtain the F-test statistic of the experimental response variable data of the class; S314, Determine the corresponding critical value of the F-distribution based on the preset significance level; S315, perform difference calculation on the subset of experimental response variable data and the corresponding set of output values to obtain the corresponding set of difference values; S316, Perform difference fusion calculation on the set of difference values of each type of experimental response variable data to obtain the difference fusion quantity; S317. For each type of experimental response variable data, determine whether the corresponding F-test statistic is greater than the critical value of the F-distribution and the corresponding difference fusion value is greater than the preset difference discrimination threshold at the same time, and obtain the statistical discrimination result of the experimental response variable data of the class.
[0010] The expression for the difference fusion calculation process is: , in, v For the difference fusion quantity, For the i-th element of the set of difference values, Let N be the mean of all elements in the set of differences, and N be the number of elements in the set of differences. It is the maximum value of all elements in the set of differences.
[0011] The step of performing optimal trial number estimation on the preprocessed test dataset to obtain a standard value for the number of trials includes: S321, using the subset of test condition variable data in the preprocessed test dataset, a test condition matrix is constructed; the row vectors of the test condition matrix represent the values of various test conditions when conducting a test on the medicinal material support container. S322, using the test response variable dataset in the preprocessed test dataset, a test response matrix is constructed; the row vectors of the test response matrix are the test values of various technical indicators measured during a test of the medicinal material support container; S323, calculate the rank of the test condition matrix and the test response matrix respectively to obtain the rank N1 of the test condition matrix and the rank N2 of the test response matrix; S324, Perform singular value solving on the test condition matrix and the test response matrix respectively to obtain the corresponding singular value set; S325, calculate the rank of the matrix and the set of singular values to obtain the standard value of the number of trials; The formula for calculating the standard value of the number of trials is: , in, N 3 is the least common multiple of N1 and N2. Indicates rounding down. and Let represent the maximum and mean values of all elements in the singular value set of the experimental condition matrix, respectively. and Let represent the maximum value and mean value of all elements in the singular value set of the experimental response matrix, respectively, and M be the standard value of the number of trials.
[0012] The performance evaluation result value of the medicinal material supply container is obtained by fusing and evaluating the set of technical indicator test results based on the set of significance discrimination results, including: Based on the significance discrimination result set, the technical indicator test results corresponding to all experimental response variable data types with a significance discrimination result of "yes" are collected from the technical indicator test result information set. Using all the corresponding technical indicator test results, the first matrix to be evaluated is constructed. Based on the significance discrimination result set, the test results of technical indicators corresponding to all experimental response variable data types with no significance discrimination result are collected from the technical indicator test result information set. Using all the corresponding technical indicator test results, a second matrix to be evaluated is constructed. The first matrix to be evaluated is subjected to a first calculation process to obtain a first evaluation value and a first weighted value; The second matrix to be evaluated is subjected to a second calculation process to obtain a second evaluation value and a second weighted value; Using the first weighted value and the second weighted value, the first evaluation value and the second evaluation value are weighted and summed to obtain the performance evaluation result value of the medicinal material support container.
[0013] According to a second aspect of the present invention, a performance evaluation device for a medicinal material transport container is disclosed, the device comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the performance evaluation method for the medicinal material supply container.
[0014] In a third aspect, the present invention discloses a computer-storeable medium storing computer instructions, which, when invoked by a computer, are used to execute the performance evaluation method for the medicinal material supply container.
[0015] In a fourth aspect, the present invention discloses an information data processing terminal, which is used to implement the performance evaluation method for the medicinal material supply container.
[0016] The beneficial effects of this invention are as follows: This invention constructs an experimental dataset encompassing both experimental condition variables and experimental response variables, and performs multi-level preprocessing and evaluation of the data. This allows for a comprehensive and systematic reflection of the performance of medicinal material transport containers under different conditions. Compared to traditional methods, this invention can simultaneously consider multiple technical indicators and experimental conditions, providing a more comprehensive performance evaluation.
[0017] This invention introduces statistical methods such as multiple linear regression analysis, F-test statistic calculation, and difference fusion calculation to scientifically identify factors that significantly affect the performance of medicinal material storage containers. It then uses a set of significance judgment results to fuse and evaluate the test results of technical indicators. This method improves the accuracy and reliability of the evaluation results and avoids the subjectivity and bias of traditional experience-based judgments.
[0018] This invention constructs a first and a second evaluation matrix, and combines them with weighted summation to comprehensively consider the impact of significant and non-significant factors on the performance of medicinal material storage containers. This comprehensive evaluation method provides more complete and objective performance evaluation results, offering a scientific basis for the design, optimization, and application of medicinal material storage containers.
[0019] The performance evaluation method of this invention can effectively guide the design and improvement of containers for transporting medicinal materials, enhancing their adaptability and reliability under different environmental conditions. Simultaneously, this method can also provide a standardized evaluation process for the transportation and storage of Chinese medicinal materials, promoting the modernization of the traditional Chinese medicine industry. Attached Figure Description
[0020] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation
[0021] To better understand the content of this invention, an embodiment is provided here.
[0022] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention.
[0023] In a first aspect, the present invention discloses a method for evaluating the performance of a medicinal material supply container, comprising: S1, Obtain the test dataset for the medicinal material supply container; the test dataset includes a test condition variable dataset and a test response variable dataset; the test condition variable dataset includes a subset of test condition variable data; the test response variable dataset includes a subset of test response variable data; the subset of test response variable data corresponds to a type of test response variable data, which is the test value of a type of technical indicator obtained in all tests when testing the medicinal material supply container based on the test condition variable data; the subset of test condition variable data is the collection of test condition variable data for a single test; The data in the experimental condition variable dataset and the data in the experimental response variable dataset are in one-to-one correspondence. S2, preprocess the test dataset to obtain a preprocessed test dataset; S3, perform sample evaluation processing on the preprocessed experimental dataset to obtain a set of significance discrimination results and standard values for the number of trials; S4. According to the standard value of the number of tests, conduct a corresponding number of tests on the medicinal material supply container to obtain a set of technical indicator test result information; the set of technical indicator test result information includes the test values of all types of technical indicators obtained in each test; S5. Based on the set of significance discrimination results, the set of test results of the technical indicators is fused and evaluated to obtain the performance evaluation result value of the medicinal material support container.
[0024] The preprocessing of the test dataset to obtain a preprocessed test dataset includes: S21, perform anomaly detection processing on the test dataset to obtain a first data set; S22, normalize the first data set to obtain the second data set; S23, perform boundary checks on the second dataset to obtain a preprocessed test dataset; In the data preprocessing stage, this invention effectively handles outliers and boundary issues in experimental data through steps such as anomaly detection, normalization, and boundary checks, ensuring data quality and consistency. This data processing method improves the accuracy of subsequent analysis and avoids misjudgments caused by data quality problems.
[0025] The step of performing sample evaluation processing on the preprocessed experimental dataset to obtain a set of significance discrimination results and standard values for the number of trials includes: S31, Perform statistical feature discrimination processing on each type of experimental response variable data in the preprocessed experimental dataset to obtain the statistical discrimination result of each type of experimental response variable data; S32, perform optimal trial number estimation processing on the preprocessed test dataset to obtain the standard value of the number of trials; The step involves performing statistical feature discrimination processing on each type of experimental response variable data in the preprocessed experimental dataset to obtain the statistical discrimination result for each type of experimental response variable data, including: S311, take the experimental condition variable dataset in the preprocessed experimental dataset as the multivariate independent variable, and take the experimental response variable data subset corresponding to each type of experimental response variable data in the experimental response variable dataset in the preprocessed experimental dataset as the dependent variable, and perform multivariate linear regression analysis on the multivariate independent variable and dependent variable to obtain the regression analysis model of each type of experimental response variable data. S312, using the regression analysis model of each type of experimental response variable data, the experimental condition variable dataset is calculated and processed to obtain the corresponding set of output values; S313, Perform F-test statistic calculation on the output value set and the corresponding experimental response variable data subset to obtain the F-test statistic of the experimental response variable data of the class; S314, Determine the corresponding critical value of the F-distribution based on the preset significance level; S315, perform difference calculation on the subset of experimental response variable data and the corresponding set of output values to obtain the corresponding set of difference values; The difference values in the set of difference values are obtained by subtracting the output value in the corresponding set of output values from each data point in the subset of experimental response variable data. S316, Perform difference fusion calculation on the set of difference values of each type of experimental response variable data to obtain the difference fusion quantity; S317. For each type of experimental response variable data, determine whether the corresponding F-test statistic is greater than the critical value of the F-distribution and the corresponding difference fusion value is greater than the preset difference discrimination threshold at the same time, and obtain the statistical discrimination result of the experimental response variable data of the class.
[0026] The step of performing difference calculation on the subset of experimental response variable data and the corresponding set of output values involves subtracting each data point in the subset of experimental response variable data from its corresponding data point in the set of output values to obtain the corresponding difference; and using all the differences, a set of difference values is constructed. The calculation and processing of the F-test statistic shall be carried out in accordance with the calculation formula of the F-test statistic. The difference discrimination threshold can be 20; Using the experimental condition variable dataset as a multivariate independent variable means taking a subset of each type of data in the experimental condition variable dataset as an independent variable to construct a multivariate independent variable. The critical value of the F-distribution is determined according to the preset significance level, which can be obtained by consulting the F-function numerical table.
[0027] The regression analysis model is a multinomial model; the linear regression analysis can be performed by using the least squares method to solve for the coefficients of the regression equation. The expression for the difference fusion calculation process is: , in, v For the difference fusion quantity, For the i-th element of the set of difference values, Let N be the mean of all elements in the set of differences, and N be the number of elements in the set of differences. The maximum value among all elements of the set of differences; The expression used in the difference fusion calculation comprehensively integrates multi-dimensional information such as the dispersion, mean, and maximum value of the difference between the experimental response variable and the regression model output value. It does not rely solely on a single dimension (such as the mean) to assess the difference level, but rather comprehensively and accurately reflects the overall degree of difference between the actual value of the experimental response variable and the predicted value of the regression model. This multi-dimensional fusion calculation effectively avoids the one-sidedness of single-indicator evaluation, providing a reliable quantitative basis for subsequent judgment of the significance of the experimental response variable data. This enhances the scientific rigor of judging the response patterns of the technical indicators of the pharmaceutical material supply container, ensuring that the statistical feature discrimination results truly reflect the degree of fit between the experimental data and the model.
[0028] The step of performing optimal trial number estimation on the preprocessed test dataset to obtain a standard value for the number of trials includes: A test condition matrix is constructed using a subset of test condition variable data from the preprocessed test dataset. The row vectors of the test condition matrix represent the values of various test conditions when conducting a test on the medicinal material support container, which is a subset of test condition variable data from the test condition variable dataset. Using the test response variable dataset in the preprocessed test dataset, a test response matrix is constructed; the row vectors of the test response matrix are the test values of various technical indicators measured during a test of the medicinal material supply container, which is a subset of test response variable data. The rank of the test condition matrix and the test response matrix are calculated respectively to obtain the rank N1 of the test condition matrix and the rank N2 of the test response matrix; The singular value set of the test condition matrix and the test response matrix is obtained by solving the matrix singular value problem. The standard value for the number of trials is obtained by calculating the rank of the matrix and the set of singular values; The formula for calculating the standard value of the number of trials is: , in, N 3 is the least common multiple of N1 and N2. Indicates rounding down. and Let represent the maximum and mean values of all elements in the singular value set of the experimental condition matrix, respectively. and Let represent the maximum value and mean value of all elements in the singular value set of the experimental response matrix, respectively, and M be the standard value of the number of trials.
[0029] The formula for calculating the standard number of trials deeply integrates the core features of the experimental condition matrix and the experimental response matrix. On the one hand, it takes into account the dimensional correlation between experimental conditions and experimental responses by using the least common multiple of the matrix's rank (reflecting the core feature of data dimensions). On the other hand, it introduces the extreme values and mean values of the matrix's singular values (reflecting data distribution characteristics), amplifies the influence of core data features through square root operations, maps the values to a reasonable range using the arctangent function, and finally transforms the abstract matrix features into an integer number of trials that meets practical operational requirements through floor division. This formula scientifically avoids the problem of blindly selecting the number of trials: it avoids both the increased testing costs and reduced efficiency caused by too many trials, and the insufficient representativeness of the data caused by too few trials. It ensures that the determined number of trials meets the sample size requirements for statistical analysis and fits the actual scenario of testing the performance of medicinal material protection containers, balancing the economy of the experiment and the reliability of the results.
[0030] This invention, through optimal experiment number estimation, scientifically determines the standard value of the number of experiments, avoiding the blind selection of the number of experiments in traditional methods. This not only improves experimental efficiency and reduces experimental costs, but also ensures the representativeness of the experimental results.
[0031] The performance evaluation result value of the medicinal material supply container is obtained by fusing and evaluating the set of technical indicator test results based on the set of significance discrimination results, including: Based on the significance discrimination result set, the test results of the technical indicators corresponding to all experimental response variable data types with a significance discrimination result of "yes" are collected from the technical indicator test result information set. Using all the corresponding technical indicator test results, the first matrix to be evaluated is constructed. Based on the significance discrimination result set, the test results of technical indicators corresponding to all experimental response variable data types with no significance discrimination result are collected from the technical indicator test result information set. Using all the corresponding technical indicator test results, a second matrix to be evaluated is constructed. The first matrix to be evaluated is subjected to a first calculation process to obtain a first evaluation value and a first weighted value; The second matrix to be evaluated is subjected to a second calculation process to obtain a second evaluation value and a second weighted value; Using the first weighted value and the second weighted value, the first evaluation value and the second evaluation value are weighted and summed to obtain the performance evaluation result value of the medicinal material support container.
[0032] The first calculation process includes: Principal component analysis is performed on the first matrix to be evaluated to obtain the principal component matrix. The principal component matrix is subjected to singular value calculation to obtain a set of singular values; All non-zero singular values in the singular value set are arranged in descending order of value to obtain a sequence of non-zero singular values; The geometric mean of the non-zero singular value sequence is calculated to obtain the first evaluation value; The principal component matrix is weighted to obtain a first weight value; The expression for calculating the geometric mean is: , in, Let be the j-th element of the non-zero singular value sequence, and n be the length of the non-zero singular value sequence. The first evaluation value; Compared to the arithmetic mean, the geometric mean is better suited to the characteristics of non-zero singular value sequences. It can effectively weaken the excessive influence of extreme singular values, while fully integrating the core features represented by each non-zero singular value after principal component analysis (the variance contribution of the singular value to the principal component reflects the importance of the core technical indicators of the pharmaceutical material support container). By calculating the geometric mean of the non-zero singular value sequence, the core performance level of the pharmaceutical material support container reflected by the first evaluation matrix (the technical indicator data with significance) can be accurately quantified. This avoids evaluation bias caused by an excessively high proportion of a single principal component, making the evaluation results of the core technical indicators more stable and representative, and truly reflecting the actual state of the container's core performance.
[0033] The expression for weight calculation is: , in, As the first weight value, Let be the mean of the i-th row of the first matrix to be evaluated. The mean of the vectors in the i-th row of the principal component matrix. The mean of the j-th column vector of the principal component matrix. The elements in the i-th row and j-th column of the principal component matrix, where I and J are the row and column dimensions of the principal component matrix, respectively, are used in the weight calculation expression. This expression integrates information such as the row mean of the first matrix to be evaluated, the row mean, column mean, and element values of the principal component matrix from multiple dimensions. It reflects both the basic characteristics of the original data to be evaluated and the core patterns of the data after principal component analysis, thus eliminating the subjective arbitrariness of weight allocation. Through this comprehensive calculation of multi-dimensional information, the weight ratio of the first evaluation value in the overall performance evaluation can be accurately determined, ensuring that the influence of the core significant technical indicators of the pharmaceutical material supply container is scientifically reflected. This avoids evaluation bias caused by unreasonable weight allocation of core indicators and lays a reasonable weight foundation for the subsequent weighted summation to obtain the overall evaluation result.
[0034] The expression for the principal component analysis processing is: Y=CX, In this model, Y is the principal component matrix, C is the coefficient matrix, and X is the first matrix to be evaluated. Principal component analysis (PCA) is performed using the PCA algorithm, and both the principal component matrix and the coefficient matrix are determined through PCA. This formula transforms the high-dimensional first matrix to be evaluated (core significant technical indicator data) into a principal component matrix using the coefficient matrix, achieving dimensionality reduction of high-dimensional data. On the one hand, it retains the key information reflecting the core performance of the medicinal material support container in the original data to the greatest extent possible; on the other hand, it eliminates redundant data and noise interference, significantly simplifying the complexity of subsequent singular value calculation, geometric mean evaluation, and weight calculation. The principal component matrix can clearly show the inherent correlation and primary-secondary relationship between each core technical indicator, focusing on core performance characteristics and improving the efficiency and accuracy of subsequent calculations of the first evaluation value and the first weight value, ensuring that the evaluation of the core significant technical indicators is focused and accurate.
[0035] The second calculation process includes: , , , in, This is the second evaluation value. The second weight value is given by m and n, which are the row and column dimensions of the second matrix to be evaluated, respectively. To evaluate the i-th element of the vector, Let be the mean of the i-th column of the second matrix to be evaluated. The element in the k-th row and i-th column of the second matrix to be evaluated; the beneficial effect of the second calculation processing related formula (evaluation vector, second evaluation value, second weight value).
[0036] The evaluation vector calculation expression in the second calculation process integrates the difference between the column mean and each element of the second matrix to be evaluated, accurately quantifying the degree to which each column of insignificant technical indicator data deviates from its column mean, and clearly reflecting the overall fluctuation characteristics of the insignificant technical indicator data. This quantification result provides a basic and reliable quantitative basis for the subsequent calculation of the second evaluation value and the second weight value, ensuring that the evaluation of insignificant technical indicators has clear data support.
[0037] The second evaluation value calculation formula, based on the core features of the evaluation vector, comprehensively considers the row and column dimension information of the second matrix to be evaluated. This not only reflects the fluctuation degree of each column's non-significant technical indicators but also avoids the one-sidedness of a single-dimensional evaluation. This formula can accurately quantify the auxiliary performance level of the medicinal material support container reflected by non-significant technical indicators, fully demonstrating the contribution of non-significant indicators to overall performance and compensating for the shortcomings of evaluating only the core significant indicators.
[0038] The second weighting formula integrates the row and column dimension characteristics of the second matrix to be evaluated and the core information of the evaluation vector, scientifically defining the weight ratio of non-significant technical indicators in the overall performance evaluation: it neither ignores the fundamental supporting role of non-significant indicators in the overall performance of the container, nor allows them to excessively interfere with the evaluation results of core significant indicators. This weighting value ensures that, during weighted summation, the proportion of core significant indicators and non-significant indicators conforms to the actual logic of performance evaluation of pharmaceutical material supply containers, improving the comprehensiveness and accuracy of the overall performance evaluation results.
[0039] The boundary check process performed on the second dataset to obtain the preprocessed test dataset includes: S231, for each data attribute in the second data set, cluster analysis is performed using the data collection information of the data as the independent variable and the data value of the data as the dependent variable to obtain the clustering result information of the data attribute; the clustering result information includes the cluster category to which all data of the data attribute belongs; the class information of the data attribute is the specific category of the experimental condition variable and the experimental response variable; the data collection information includes the time information, spatial information or technical indicator type information corresponding to the data collection; S232, determine the number of data included in each cluster category; set a data volume threshold; delete the data included in cluster categories whose data volume is less than the data volume threshold from the second data set; S233, For all data in the second dataset, execute S231 and S232 to obtain the preprocessed experimental dataset.
[0040] The technical specifications of the medicinal material support container include medicinal material storage and transportation capacity, medicinal material preparation and distribution capacity, medical oxygen supply capacity, medical water supply capacity, and information technology capability. The medicinal material storage and transportation capacity includes pre-cooling / preheating time, quantity of medicinal materials stored, and integrity rate of medicinal materials; The medicinal material preparation and distribution capability includes the ability to monitor the medicinal material storage environment and the satisfaction level of medicinal material preparation and distribution. The medical oxygen supply capability includes the satisfaction of the pressurized oxygen production capacity and the compatibility of the oxygen supply interface; The anomaly detection and processing includes filling missing values, smoothing noisy data, and smoothing or deleting outliers. Smoothing noisy data involves first identifying the noisy data, and then smoothing it based on the data preceding and following it. The noisy data refers to values whose values are less than the sensor's detection sensitivity for the observed data, or greater than the sensor's measurement upper limit for the observed data. Outlier identification can be performed using a Kalman filter. The filling value for missing values can be determined by averaging the measurements within a certain sampling interval before and after the missing value.
[0041] The experimental dataset can be obtained by searching a database, or by collecting experimental condition variable datasets and experimental response variable datasets through sensors installed in the workplace, or by conducting virtual simulation experiments on the digital model of the medicinal material supply container.
[0042] In all embodiments of the present invention, the variables involved in all computational expressions or mathematical functions have been dimensionlessized before computation.
[0043] In all embodiments of the present invention, the values of the independent variables in the input of all computational expressions or mathematical functions meet the reasonable requirements of the input range of the computational expressions or mathematical functions, and can ensure that the computational expressions or mathematical functions can be calculated smoothly without violating physical laws or mathematical rules.
[0044] According to a second aspect of the present invention, a performance evaluation device for a medicinal material transport container is disclosed, the device comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the performance evaluation method for the medicinal material supply container.
[0045] In a third aspect, the present invention discloses a computer-storeable medium storing computer instructions, which, when invoked by a computer, are used to execute the performance evaluation method for the medicinal material supply container.
[0046] In a fourth aspect, the present invention discloses an information data processing terminal, which is used to implement the performance evaluation method for the medicinal material supply container.
[0047] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.
Claims
1. A performance evaluation method for a medicinal material transport container, characterized in that, include: S1, Obtain the experimental dataset of the medicinal material supply container; The experimental dataset includes an experimental condition variable dataset and an experimental response variable dataset; the experimental condition variable dataset includes a subset of experimental condition variable data; the experimental response variable dataset includes a subset of experimental response variable data; each subset of experimental response variable data corresponds to a type of experimental response variable data, which is the test value of a type of technical indicator obtained in each test when testing the medicinal material supply container based on the experimental condition variable data; the subset of experimental condition variable data is the collection of experimental condition variable data for a single test. S2, preprocess the test dataset to obtain a preprocessed test dataset; S3, perform sample evaluation processing on the preprocessed experimental dataset to obtain a set of significance discrimination results and standard values for the number of trials, including: S31, Perform statistical feature discrimination processing on each type of experimental response variable data in the preprocessed experimental dataset to obtain the statistical discrimination result of each type of experimental response variable data; S32, perform optimal trial number estimation processing on the preprocessed experimental dataset to obtain a standard value for the number of trials, including: S321, using the subset of test condition variable data in the preprocessed test dataset, a test condition matrix is constructed; the row vectors of the test condition matrix represent the values of various test conditions when conducting a test on the medicinal material support container. S322, using the test response variable dataset in the preprocessed test dataset, a test response matrix is constructed; the row vectors of the test response matrix are the test values of various technical indicators measured during a test of the medicinal material support container; S323, calculate the rank of the test condition matrix and the test response matrix respectively to obtain the rank N1 of the test condition matrix and the rank N2 of the test response matrix; S324, Perform singular value solving on the test condition matrix and the test response matrix respectively to obtain the corresponding singular value set; S325, calculate the rank of the matrix and the set of singular values to obtain the standard value of the number of trials; The formula for calculating the standard value of the number of trials is: , Where N3 is the least common multiple of N1 and N2, Indicates rounding down. and Let represent the maximum and mean values of all elements in the singular value set of the experimental condition matrix, respectively. and Let represent the maximum value and mean value of all elements in the singular value set of the experimental response matrix, respectively, and M be the standard value of the number of trials; S4. According to the standard value of the number of tests, conduct a corresponding number of tests on the medicinal material supply container to obtain a set of technical indicator test result information; the set of technical indicator test result information includes the test values of all types of technical indicators obtained in each test; S5. Based on the set of significance discrimination results, the set of test results of the technical indicators is fused and evaluated to obtain the performance evaluation result value of the medicinal material support container.
2. The performance evaluation method for medicinal material supply containers as described in claim 1, characterized in that, The preprocessing of the test dataset to obtain a preprocessed test dataset includes: S21, perform anomaly detection processing on the test dataset to obtain a first data set; S22, normalize the first data set to obtain the second data set; S23, perform boundary checks on the second dataset to obtain a preprocessed test dataset.
3. The performance evaluation method for medicinal material supply containers as described in claim 1, characterized in that, The step involves performing statistical feature discrimination processing on each type of experimental response variable data in the preprocessed experimental dataset to obtain the statistical discrimination result for each type of experimental response variable data, including: S311, take the experimental condition variable dataset in the preprocessed experimental dataset as the multivariate independent variable, and take the experimental response variable data subset corresponding to each type of experimental response variable data in the experimental response variable dataset in the preprocessed experimental dataset as the dependent variable, and perform multivariate linear regression analysis on the multivariate independent variable and dependent variable to obtain the regression analysis model of each type of experimental response variable data. S312, using the regression analysis model of each type of experimental response variable data, the experimental condition variable dataset is calculated and processed to obtain the corresponding set of output values; S313, Perform F-test statistic calculation on the output value set and the corresponding experimental response variable data subset to obtain the F-test statistic of the experimental response variable data of the class; S314, Determine the corresponding critical value of the F-distribution based on the preset significance level; S315, perform difference calculation on the subset of experimental response variable data and the corresponding set of output values to obtain the corresponding set of difference values; S316, Perform difference fusion calculation on the set of difference values of each type of experimental response variable data to obtain the difference fusion quantity; S317. For each type of experimental response variable data, determine whether the corresponding F-test statistic is greater than the critical value of the F-distribution and the corresponding difference fusion value is greater than the preset difference discrimination threshold at the same time, and obtain the statistical discrimination result of the experimental response variable data of the class.
4. The performance evaluation method for medicinal material supply containers as described in claim 3, characterized in that, The expression for the difference fusion calculation process is: , Where v is the difference fusion quantity. For the i-th element of the set of difference values, Let N be the mean of all elements in the set of differences, and N be the number of elements in the set of differences. It is the maximum value of all elements in the set of differences.
5. The performance evaluation method for medicinal material supply containers as described in claim 1, characterized in that, The performance evaluation result value of the medicinal material supply container is obtained by fusing and evaluating the set of technical indicator test results based on the set of significance discrimination results, including: Based on the significance discrimination result set, the technical indicator test results corresponding to all experimental response variable data types with a significance discrimination result of "yes" are collected from the technical indicator test result information set. Using all the corresponding technical indicator test results, the first matrix to be evaluated is constructed. Based on the significance discrimination result set, the test results of technical indicators corresponding to all experimental response variable data types with no significance discrimination result are collected from the technical indicator test result information set. Using all the corresponding technical indicator test results, a second matrix to be evaluated is constructed. The first matrix to be evaluated is subjected to a first calculation process to obtain a first evaluation value and a first weighted value; The second matrix to be evaluated is subjected to a second calculation process to obtain a second evaluation value and a second weighted value; Using the first weighted value and the second weighted value, the first evaluation value and the second evaluation value are weighted and summed to obtain the performance evaluation result value of the medicinal material support container.
6. A performance evaluation device for a medicinal material transport container, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the performance evaluation method for the medicinal material supply container as described in any one of claims 1 to 5.
7. A computer-storable medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked by the computer, are used to execute the performance evaluation method for the medicinal material supply container as described in any one of claims 1 to 5.
8. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the performance evaluation method for the medicinal material supply container as described in any one of claims 1 to 5.
Citation Information
Patent Citations
Kernel density estimation-based load spectrum compilation method for transmission shaft of tracked vehicle
CN106886638A
Dam deformation abnormal value identification method and system
CN117272202A
Technical performance evaluation method and device for field support safeguard system
CN121365909A
Product design parameter decision method and support system for it
JP2006344200A
