CN120197082B

CN120197082B - Confidence sorting method and system for salt iodine content detection data

Info

Publication number: CN120197082B
Application number: CN202510685463.4A
Authority: CN
Inventors: 辛刚; 左宜喜; 刘昊彤
Original assignee: Hebei Yongda Salt Co ltd
Current assignee: Hebei Yongda Salt Co ltd
Priority date: 2025-05-27
Filing date: 2025-05-27
Publication date: 2025-09-23
Anticipated expiration: 2045-05-27
Also published as: CN120197082A

Abstract

The application relates to the field of food safety monitoring, in particular to a confidence sorting method and a system for salt iodine content detection data. The method comprises the steps of performing correlation analysis on a preset influence index set according to a detection process type to obtain a correlation influence index set, receiving a correlation influence index constraint interval, collecting time sequence information of a correlation influence index monitoring value and comparing the constraint interval to obtain deviation vector time sequence information when salt samples are detected, counting error count values of a detection sample set meeting the deviation vector time sequence information, performing trusted identification on the detection value when the error count values are smaller than an error threshold, counting the detection value set with trusted identification, and obtaining an iodine content confidence value and sending the iodine content confidence value to a user side. According to the application, the multidimensional associated influence indexes are comprehensively considered, a systematic data confidence sorting mechanism is established, and the reliability and the accuracy of the detection result of the salt iodine content are effectively improved.

Description

Confidence sorting method and system for salt iodine content detection data

Technical Field

The invention relates to the field of food safety monitoring, in particular to a confidence sorting method and a system for salt iodine content detection data.

Background

Currently, the detection of the iodine content of salt mainly adopts a spectrophotometry method, and the iodine content is determined by measuring the absorbance value of a sample under a specific wavelength and combining a standard curve. However, in the actual detection process, due to the influence of various factors such as instrument performance, sample characteristics, environmental conditions, operation specifications and the like of the spectrophotometer, the detection result of the iodine content of the salt often has certain fluctuation and error, so that the reliability and the precision of the detection result are not high.

Disclosure of Invention

Aiming at the technical problem that the reliability and the precision of the detection result of the salt iodine content are not high in the prior art, the invention achieves the technical effect of improving the reliability and the precision of the detection result of the salt iodine content, and provides the confidence sorting method and the system for the detection data of the salt iodine content to solve the problem.

The technical scheme for solving the technical problems is as follows:

The invention provides a confidence sorting method of salt iodine content detection data, which comprises the steps of performing correlation analysis on a preset influence index set according to a salt iodine content detection process type to obtain an associated influence index set, receiving an associated influence index constraint interval of the associated influence index set, collecting time sequence information of an associated influence index monitoring value when a salt sample is detected, comparing the time sequence information with the associated influence index constraint interval to obtain time sequence information of an associated influence index deviation vector, counting iodine content detection error count values of an iodine content detection sample set meeting the time sequence information of the associated influence index deviation vector, carrying out trusted identification on the salt sample iodine content detection value when the iodine content detection error count values are smaller than an error threshold, carrying out statistics on the salt sample iodine content detection value set with trusted identification to obtain a salt iodine content confidence value, and sending the salt iodine content confidence value to a user side.

The invention provides a confidence sorting system of salt iodine content detection data, which comprises a correlation analysis module, a constraint interval receiving module, a deviation vector analysis module, an error mode counting module and a confidence value counting module, wherein the correlation analysis module is used for performing correlation analysis on a preset influence index set according to the salt iodine content detection process type to obtain a correlation influence index set, the constraint interval receiving module is used for receiving a correlation influence index constraint interval of the correlation influence index set, the deviation vector analysis module is used for collecting correlation influence index monitoring value time sequence information when salt samples are detected and comparing the correlation influence index monitoring value time sequence information with the correlation influence index constraint interval to obtain correlation influence index deviation vector time sequence information, the error mode counting module is used for counting iodine content detection error mode values of iodine content detection sample sets meeting the correlation influence index deviation vector time sequence information, the detection value identification module is used for carrying out confidence identification on the iodine content detection values of the salt samples when the iodine content detection error mode values are smaller than an error threshold value, and the confidence value counting module is used for counting the iodine content detection value sets of the salt samples with the confidence identification to obtain iodine content confidence values and sending the iodine content confidence values to a user.

The beneficial effects of the invention are as follows:

according to the type of the salt iodine content detection process, performing correlation analysis on a preset influence index set to obtain a correlation influence index set, and analyzing various factors possibly influencing detection results, such as instrument factors (light source stability, detector performance, monochromator performance and wavelength accuracy), sample factors (uniformity, stability and concentration range), environmental factors (temperature stability, humidity stability and electromagnetic interference), operating factors (measurement parameter setting, sample placement and operation specification and a data processing method), and the like, for specific detection process types, screening indexes with obvious correlation with the detection results through the correlation analysis to form the correlation influence index set, and laying a foundation for subsequent data processing. And receiving the association influence index constraint interval of the association influence index set. After the set of associated influence indicators is determined, constraint intervals are set for each associated influence indicator, and the constraint intervals represent reasonable fluctuation ranges of the indicators under normal detection conditions. The setting of the constraint interval can be based on theoretical analysis, historical data statistics or expert experience, so that a basis is provided for the subsequent judgment of whether the detection process is stable. In the actual salt sample detection process, the values of all the relevant influence indexes are monitored in real time, time sequence information of the relevant influence indexes changing along with time is recorded, time sequence information of the relevant influence index monitoring values is obtained, the time sequence information of the relevant influence index monitoring values is compared with the relevant influence index constraint interval, the deviation degree and direction of all the indexes are calculated, deviation vector time sequence information is formed, and the stability condition of all influence factors in the detection process is reflected.

Based on the obtained deviation vector time sequence information, a detection sample set in a specific mode of the associated influence index deviation vector time sequence information is screened out, detection errors of the samples are subjected to statistical analysis, and a crowd value of error distribution is obtained, so that a typical error mode under a specific disturbance condition is identified, and a quantization index is provided for evaluating the reliability of a detection result. And comparing the obtained iodine content detection error value with a preset error threshold value, and judging whether the detection result is reliable. When the iodine content detection error score value is smaller than the error threshold value, the detection result still has enough accuracy in spite of fluctuation of various influencing factors under the current detection condition, and the detection value is marked as trusted data. And carrying out statistical treatment on the salt sample iodine content detection value set which is evaluated by the credibility, calculating to obtain a salt iodine content confidence value with high reliability, and transmitting the result to a user side. The statistical method based on confidence sorting effectively eliminates the interference of unreliable data and improves the accuracy and reliability of a final result.

Through the technical scheme, confidence sorting of the salt iodine content detection data is achieved, the problem that detection accuracy and reliability are insufficient due to neglect of quality evaluation in the detection process in the traditional method is effectively solved, and reliability and accuracy of salt iodine content detection results are improved.

Drawings

FIG. 1 is a flow chart of a confidence sorting method for salt iodine content detection data provided by the invention;

fig. 2 is a schematic structural diagram of a confidence sorting system for salt iodine content detection data provided by the invention.

In the drawings, the components represented by the respective reference numerals are as follows:

The system comprises a correlation analysis module 11, a constraint interval receiving module 12, a deviation vector analysis module 13, an error mode statistics module 14, a detection value identification module 15 and a confidence value statistics module 16.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

In the description of the present invention, the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.

In the description of the present invention, the term "for example" is used to mean "serving as an example, instance, or illustration. Any embodiment described as "for example" in this disclosure is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the invention. In the following description, details are set forth for purposes of explanation. It will be apparent to one of ordinary skill in the art that the present invention may be practiced without these specific details. In other instances, well-known structures and processes have not been described in detail so as not to obscure the description of the invention with unnecessary detail. Thus, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

In a first embodiment, as shown in fig. 1, the embodiment of the invention provides a confidence sorting method for salt iodine content detection data, which includes:

s1, performing correlation analysis on a preset influence index set according to the type of the salt iodine content detection process to obtain a correlation influence index set.

Specifically, in the process of detecting the iodine content of table salt, the detection result is affected by various factors. The preset influence index set comprises indexes which possibly influence the detection result of the iodine content of the salt, and the indexes can be divided into four types of instrument factors, sample factors, environment factors and operation factors. The instrument factors comprise light source stability, detector performance, monochromator performance, instrument wavelength accuracy and the like, the sample factors comprise sample uniformity, sample stability, sample concentration range and the like, the environment factors comprise temperature stability, humidity stability, electromagnetic interference and the like, and the operation factors comprise measurement parameter setting, sample placement, operation specification, data processing method and the like. Different salt iodine content detection process types (such as spectrophotometry, potentiometric titration and the like) have different interference degrees with different influence indexes. Therefore, for a specific process type for detecting the iodine content of salt, a correlation analysis is performed on a preset influence index set, so as to screen out indexes with high correlation with the detection result of the specific process type, and form a correlation influence index set.

And the correlation analysis evaluates the correlation degree between each preset influence index and the detection result by a statistical method, and only the index with the correlation reaching a certain threshold value is reserved as the correlation influence index. The screening method can effectively reduce the complexity of subsequent analysis, avoid interference caused by irrelevant indexes, and improve the analysis efficiency and accuracy. The specific correlation analysis method is described in detail later, so that the influence degree of each index can be comprehensively and systematically estimated, and the reliability of the obtained associated influence index set is ensured.

By acquiring the associated influence index set, a foundation is laid for the follow-up table salt iodine content detection data confidence sorting flow, influence factors needing key monitoring are clarified, and the accuracy and reliability of table salt iodine content detection are effectively improved.

S2, receiving the association influence index constraint interval of the association influence index set.

Specifically, the association influence index constraint interval is comprehensively determined by an expert group according to the standard of salt iodine content detection, industry specifications, technical parameters of instruments and equipment and statistical analysis of a large amount of historical detection data. The expert group consists of a senior expert in the food detection field, an analytical instrument professional technician and a statistics expert, and ensures that the setting of the constraint interval has authority and scientificity. Specifically, the determining process of the associated influence index constraint interval comprises the following steps of firstly referring to the specified requirements on detection environment, instrument state and sample treatment in the relevant standard, secondly considering technical parameters of detection equipment such as a spectrophotometer and stability indexes thereof, thirdly determining the normal distribution interval of each associated influence index by carrying out statistical analysis on a large amount of historical detection data, and then comprehensively evaluating the reasonable constraint interval of each index by an expert group in combination with the actual operation condition and technical feasibility of a laboratory. For example, for the related influence index of the light source stability, the expert group determines that the constraint interval is 20+/-2 ℃ according to the technical specification of a spectrophotometer and the analysis of historical data, and for the related influence index of the temperature stability, the expert group determines that the constraint interval is 20+/-2 ℃ according to the control capability of a laboratory environment and the requirements of a detection method.

The received association influence index constraint interval is a key judgment standard for confidence sorting of the salt iodine content detection data. Reasonable setting of the constraint interval influences the recognition capability of the abnormal state, and further influences the credibility evaluation result of the detection data. The scientific basis for judging whether the index is in the normal state is obtained by receiving the association influence index constraint interval of the association influence index set determined by the expert group, and a reliable reference standard is provided for the deviation state analysis in the subsequent step.

And by receiving the association influence index constraint interval determined by the expert group, the definition of the normal state of the association influence index is established, and a foundation is laid for the follow-up table salt iodine content detection data confidence sorting flow.

And S3, when the salt sample is detected, acquiring time sequence information of the monitoring value of the associated influence index, and comparing the time sequence information with the constraint interval of the associated influence index to obtain time sequence information of the deviation vector of the associated influence index.

Specifically, on the basis of obtaining the associated influence index set and the associated influence index constraint interval thereof, the state of each associated influence index in the actual detection process is monitored and evaluated in real time, and the degree of deviation of the index from the normal state is quantified.

In the salt iodine content detection process, each associated influence index in the determined associated influence index set is monitored in real time through a sensor network, and the monitoring value of the associated influence index is recorded along with the time change to form the time sequence information of the monitoring value of the associated influence index. The acquisition frequency of the time sequence information of the monitoring value of the associated influence index is set according to the characteristics of the detection equipment and the requirements of the detection process, so that the dynamic change process of the associated influence index can be ensured to be captured. And after the time sequence information of the associated influence index monitoring value is acquired, comparing and analyzing the time sequence information with the received associated influence index constraint interval. The comparison process calculates the deviation degree of the monitoring value and the constraint interval at each monitoring time according to each associated influence index, and generates a deviation vector. The deviation vector not only contains the direction of deviation (above the upper constraint limit or below the lower constraint limit), but also quantifies the degree of deviation, providing more information for subsequent analysis. For example, if the temperature stability is set to a constraint interval of 20±2 ℃, the deviation vector at a certain time is +1.5 ℃ when the monitored value is 23.5 ℃, which means that the upper constraint limit is exceeded, the deviation vector is-0.2 ℃ when the monitored value is 17.8 ℃, which means that the temperature stability is lower than the lower constraint limit by 0.2 ℃, and the deviation vector is 0 when the monitored value is 21 ℃. The offset vectors at each time are arranged in time sequence to form the associated influence index offset vector time sequence information. The time sequence information not only reflects whether each associated influence index deviates from the constraint interval, but also reflects the deviated time mode and dynamic change trend, and provides an important basis for the subsequent evaluation of the reliability of the detection result.

By acquiring the time sequence information of the deviation vector of the associated influence indexes, the actual state of each associated influence index in the salt iodine content detection process can be comprehensively mastered, a foundation is laid for subsequent result analysis and credibility evaluation based on similar detection conditions, and the accuracy and reliability of confidence sorting of salt iodine content detection data are effectively improved.

And S4, counting the iodine content detection error crowd value of the iodine content detection sample set which meets the correlation influence index deviation vector time sequence information.

Specifically, on the basis of obtaining the time sequence information of the associated influence index deviation vector, iodine content detection samples under the condition of similar influence factors are screened out, and detection error distribution characteristics of the samples are analyzed, so that a basis is provided for evaluating the credibility of the current detection result.

First, based on the obtained correlation impact index deviation vector time sequence information, an iodine content detection sample set satisfying the correlation impact index deviation vector time sequence information is screened out from a history detection database. The satisfying the correlation impact indicator deviation vector time sequence information refers to a sample with high similarity between the deviation state of the correlation impact indicator and the deviation state of the current detection sample (namely, the correlation impact indicator deviation vector time sequence information) in the history detection process. This similarity considers not only the direction and extent of deviation of each associated impact indicator, but also the temporal pattern and dynamic change characteristics of the deviation.

And (3) calculating the iodine content detection error of each iodine content detection sample, namely the difference between the actual detection value and the true value of the sample, for the screened iodine content detection sample set. And then carrying out statistical analysis on the error values to determine the distribution characteristics of the error values, and particularly finding out the mode of the error values, namely the error value with the highest occurrence frequency, which is defined as the iodine content detection error mode. The iodine content detection error magnitude value reflects systematic errors which are most likely to be generated by the detection result under the specific association influence index deviation state (namely association influence index deviation vector time sequence information). Such systematic errors are mainly caused by deviations of associated impact indicators, and have certain regularity and predictability. Through statistical analysis of the concentrated trend of the errors, the reliability of the results under the current detection condition can be evaluated, and a quantization basis is provided for the subsequent detection value reliability judgment.

The iodine content detection error crowd value obtained through statistical analysis reflects the error characteristic of the detection system under the condition of specific influence factors, and provides a reference for the reliability of the subsequent evaluation detection result. The method based on historical data and similar condition analysis improves the scientificity and accuracy of confidence sorting of the salt iodine content detection data.

S5, when the iodine content detection error public value is smaller than an error threshold value, performing credible identification on the iodine content detection value of the salt sample.

Specifically, based on the obtained iodine content detection error value, the credibility of the current iodine content detection value of the salt sample is judged and marked, and a foundation is provided for realizing confidence sorting of the salt iodine content detection data.

Firstly, comparing the counted iodine content detection error value with a preset error threshold value. The error threshold is a preset maximum acceptable error limit, and the error threshold is set according to factors such as standard requirements including detection of the iodine content of salt, quality control requirements, accuracy requirements of actual application scenes and the like. When the iodine content detection error number value is smaller than the error threshold value, the systematic error generated by the detection system is in an acceptable range under the current association influence index deviation state, and the detection result has higher reliability. At this time, the system performs a trusted identification on the iodine content detection value of the salt sample, and marks the iodine content detection value as a trusted result. The trusted identification may take the form of a data flag bit, a quality level, a confidence score, etc. to distinguish between detection results of different confidence levels in subsequent data processing and applications. In contrast, if the iodine content detection error number is greater than or equal to the error threshold, it indicates that the detection system may generate a larger systematic error in the current association influence index deviation state, and the reliability of the detection result is affected. At this time, the iodine content detection value of the salt sample is not identified with a confidence, and may be marked as a result to be verified or a low confidence result, which indicates that re-detection or other corrective measures may be required.

By the reliability judging method based on error analysis, detection results with higher reliability under various detection conditions can be effectively identified and screened, and the influence of detection errors caused by associated influence index deviation on the final result is avoided. The method is different from the traditional method of directly counting the multiple detection results, considers the actual influence of various disturbance factors in the detection process, and can evaluate the reliability of the detection results more accurately. Through the identification process, the effective screening and grading of the detection result of the salt iodine content are realized, support is provided for obtaining more accurate and reliable detection results, and the precision and reliability of the detection of the salt iodine content are effectively improved.

And S6, counting the salt sample iodine content detection value set with the trusted mark, obtaining a salt iodine content confidence value and sending the salt iodine content confidence value to the user side.

Specifically, first, salt samples marked with a trusted identifier are selected from all iodine content detection samples to form a salt sample iodine content detection value set. The salt sample iodine content detection value in the salt sample iodine content detection value set has higher reliability due to meeting the error control requirement, and is suitable for being used as the calculation basis of the final result. Through the screening, the detection result which is influenced by excessive disturbance and generates larger error is effectively eliminated, and the accuracy of the final result is improved. And carrying out average value calculation on the salt sample iodine content detection values in the screened salt sample iodine content detection value set, and taking the average value calculation result as a salt iodine content confidence value. And then, the confidence value of the iodine content of the salt is sent to the user side through a communication network. The user side can be a data management system of a detection laboratory, a quality control system of a salt production enterprise or a supervision platform of a supervision department, and the like. The user can obtain a high-reliability detection result through the received confidence value of the iodine content of the salt, and provides a basis for subsequent quality evaluation, production adjustment or supervision decision.

Through statistical analysis and result feedback, the whole process of confidence sorting of the salt iodine content detection data is completed, and the conversion from the original detection data to the high-reliability final result is realized. Compared with the traditional method for directly counting multiple detection results, the method considers various influencing factors in the detection process, effectively improves the accuracy and reliability of the detection result of the salt iodine content through analysis and screening, and provides support for the enhanced supervision and management of the salt iodine.

Further, according to the type of the salt iodine content detection process, performing correlation analysis on a preset influence index set to obtain a correlation influence index set, including:

s11, sending the salt iodine content detection process type and the preset influence index set to a first distributed node, and performing correlation analysis to obtain a first correlation influence index set;

S12, until the salt iodine content detection process type and the preset influence index set are sent to an N distributed node, performing correlation analysis to obtain an N associated influence index set;

s13, sorting the trigger frequency based on the first association influence index set to the Nth association influence index set to obtain the association influence index set.

In one possible implementation mode, a correlation analysis method based on distributed computation is provided, and accuracy and reliability of correlation analysis are improved through multi-node collaborative analysis.

Firstly, the type of the salt iodine content detection process and a preset influence index set are sent to a first distributed node, and correlation analysis is performed to obtain a first correlation influence index set. The distributed nodes are processing units scattered on different geographic positions or different computing resources, each distributed node has independent computing power and data resources, and the first distributed node is any one of N distributed nodes. After the first distributed node receives the salt iodine content detection process type and the preset influence index set, based on the locally stored historical detection data and analysis model, correlation analysis is carried out, the correlation degree between each preset influence index and the detection result is evaluated, and indexes with obvious correlation are screened out to form a first correlation influence index set.

And similarly, the salt iodine content detection process type and the preset influence index set are transmitted to the second distributed node to the Nth distributed node in a similar mode, and each node independently executes correlation analysis to respectively obtain a second association influence index set to the Nth association influence index set. Wherein N represents the total number of nodes participating in the computation in the distributed system, and the specific numerical value can be configured according to the data scale and the computation requirement. Through parallel computation of a plurality of distributed nodes, the analysis efficiency is effectively improved, and meanwhile, the comprehensiveness and representativeness of analysis results are enhanced by utilizing data resources of different nodes and analysis view angles. And then, performing trigger frequency sorting based on the first association influence index set to the Nth association influence index set to obtain the association influence index set. The triggering frequency sorting refers to counting the occurrence frequency of each preset influence index in N associated influence index sets, and screening out the index with higher occurrence frequency as the final associated influence index. The analysis method based on multi-node cooperation and trigger frequency fully utilizes the data and calculation advantages of the distributed system, and can more comprehensively and accurately identify indexes which have obvious influence on a specific salt iodine content detection process.

Through distributed correlation analysis, the accuracy and reliability of correlation influence index identification are effectively improved, a solid foundation is laid for confidence sorting of follow-up salt iodine content detection data, the method is suitable for analysis processing of large-scale detection data, regular information contained in historical data can be fully mined, and the accuracy and stability of salt iodine content detection are improved.

Further, performing trigger frequency sorting based on the first association influence index set up to the nth association influence index set to obtain the association influence index set, including:

s131, counting the same attribute triggering frequencies from the first association influence index set to the Nth association influence index set to obtain a first index triggering frequency to a Q index triggering frequency;

S132, calculating the ratio of the first index trigger frequency to N, and setting the ratio as a first index association degree;

S133, setting the ratio of the Q index trigger frequency to N as the Q index association degree until the ratio of the Q index trigger frequency to N is calculated;

S134, extracting the first index association degree until indexes which are larger than or equal to an association degree threshold value in the Q index association degree are added into the association influence index set.

In a preferred embodiment, first, the same attribute trigger frequency statistics are performed on the first association influence index set to the nth association influence index set, so as to obtain the first index trigger frequency to the Q index trigger frequency. The statistics of the same attribute triggering frequency refers to statistics of the occurrence times of pointers in the associated influence index sets generated by N distributed nodes for each index in the preset influence index sets. Q represents the total number of indexes in a preset influence index set, each index has a corresponding trigger frequency, and the consistency degree of the relevance judgment of different distributed nodes on the indexes is reflected. For example, assuming that a certain preset impact index "light source stability" appears m times in N associated impact index sets, its trigger frequency is m, and assuming that another preset impact index "sample uniformity" appears N times in N associated impact index sets, its trigger frequency is N. Through the statistics, the first index trigger frequency to the Q index trigger frequency are obtained, and the selected condition of each index in the multi-node analysis is comprehensively reflected. Then, the ratio of the first index trigger frequency to N is calculated and set as a first index association degree. The degree of association is a value ranging from 0 to 1, representing the probability or confidence that the indicator is determined to be a relevant indicator. For example, if the triggering frequency of the first index is m and the total number of distributed nodes is N, the association degree of the first index is m/N. The higher the association degree is, the more distributed nodes are indicated to consider that the index has obvious correlation with the detection result, and the higher the reliability of the judgment is.

And then, in a similar manner, sequentially calculating the ratio of the triggering frequency to N from the second index to the Q index, and setting the ratio as the second index association degree to the Q index association degree respectively. Through the calculation, the association degree of all indexes in the preset influence index set is obtained, and a quantification basis is provided for subsequent screening. And then extracting the first index association degree until indexes which are larger than or equal to the association degree threshold value in the Q index association degree, and adding the indexes into the association influence index set. The association degree threshold is a preset judgment standard, and represents the minimum association degree required for accepting an index as an association influence index. The association threshold is set by balancing analysis precision and screening severity, and can be adjusted according to actual application requirements and system performance. For example, if the association threshold is set to 0.7, only indexes with trigger frequencies exceeding 70% of the total number of nodes are included in the final association-affecting index set.

Through the quantitative screening method based on the trigger frequency and the association degree, the association influence index with high consistency acceptance can be extracted from the multi-node analysis result, and the deviation and limitation possibly caused by single-node analysis are effectively avoided. The method fully plays the advantages of distributed calculation, can more reliably identify factors which have obvious influence on the detection result of the iodine content of the salt, and provides a basis for the confidence sorting of the follow-up detection data.

Further, the salt iodine content detection process type and the preset influence index set are sent to a first distributed node, correlation analysis is performed, and a first correlation influence index set is obtained, including:

S111, sending the salt iodine content detection process type and the preset influence index set to a first distributed node, and collecting salt iodine content detection historical data, wherein the salt iodine content detection historical data comprise salt production record parameters, preset influence index record characteristic values and salt iodine content detection record values;

s112, performing cluster analysis on the salt iodine content detection historical data according to the salt production recording parameters to obtain multi-cluster salt iodine content detection historical data;

s113, traversing the multi-cluster salt iodine content detection historical data to perform gray correlation degree analysis based on the preset influence index record characteristic value and the salt iodine content detection record value, executing correlation analysis to obtain a plurality of initial correlation influence index sets, taking a union set, and adding the union set into the first correlation influence index set.

In a preferred embodiment, firstly, the salt iodine content detection process type and the preset influence index set are sent to a first distributed node, and salt iodine content detection historical data are collected, wherein the salt iodine content detection historical data comprise salt production record parameters, preset influence index record characteristic values and salt iodine content detection record values. The salt production recording parameters comprise information such as production batch, production date, raw material source, process parameters and the like, are used for identifying sample characteristics under different production conditions, the preset influence index recording characteristic values refer to actual monitoring values of preset influence indexes such as specific values of factors such as light source stability, temperature and humidity in the history detection process, and the salt iodine content detection recording values are salt iodine content measurement results obtained in the history detection. The first distributed node acquires relevant historical detection data from a local database or a central database through a data interface according to the received salt iodine content detection process type and a preset influence index set, and provides data support for subsequent analysis.

And then, carrying out cluster analysis on the salt iodine content detection historical data according to salt production record parameters to obtain multi-cluster salt iodine content detection historical data. Cluster analysis is an unsupervised learning method aimed at classifying data samples with similar features. Based on key characteristics in salt production record parameters, such as raw material sources, production processes, production equipment and the like, adopting algorithms such as K-means, hierarchical clustering or density clustering and the like to divide salt iodine content detection historical data into a plurality of clusters. The samples in each cluster have higher similarity in the table salt production condition, and the classification mode is helpful for identifying key factors influencing the detection result under specific production conditions, so that the pertinence and the accuracy of correlation analysis are improved.

And then, traversing multi-cluster salt iodine content detection historical data to perform gray correlation degree analysis based on the preset influence index record characteristic value and the salt iodine content detection record value, performing correlation analysis to obtain a plurality of initial correlation influence index sets, taking a union set, and adding the union set into a first correlation influence index set. Gray correlation analysis is an effective method for processing uncertainty problems, and is particularly suitable for the conditions of small sample size and incomplete information. Specifically, for each salt iodine content detection historical data cluster, extracting a preset influence index record characteristic value as a comparison sequence, taking the salt iodine content detection record value as a reference sequence, calculating gray association degree between each preset influence index and a detection result, and screening out indexes with association degree higher than a threshold value to form an initial association influence index set. After traversing all the data clusters, a plurality of initial association influence index sets are obtained, and then a union set of the sets is taken to form a first association influence index set of a first distributed node.

By the method based on data clustering and gray correlation analysis, rule information contained in historical detection data can be fully mined, and factors which have obvious influence on salt iodine content detection results under different production conditions can be identified. The method comprehensively considers the diversity of production conditions and the complexity of influence factors, can more comprehensively and accurately determine the associated influence indexes, and provides a basis for the confidence sorting of the follow-up salt iodine content detection data. By adopting the historical data analysis method, the accurate identification of the associated influence indexes is realized on a distributed computing architecture, the confidence sorting accuracy and adaptability of the salt iodine content detection data are improved, and technical guarantee is provided for obtaining high-quality detection results.

Further, based on the preset influence index record characteristic value and the salt iodine content detection record value, traversing the multi-cluster salt iodine content detection historical data to perform gray correlation analysis, and performing correlation analysis to obtain a plurality of initial correlation influence index sets, including:

s1131, extracting first cluster salt iodine content detection historical data from the multi-cluster salt iodine content detection historical data;

S1132, extracting a first cluster of preset influence index record characteristic value set from the first cluster of salt iodine content detection historical data, performing dimension removal treatment, and setting the first cluster of preset influence index record characteristic value set as a comparison sequence;

S1133, extracting a first cluster salt iodine content detection record value set from the first cluster salt iodine content detection historical data, performing dimension removal treatment, and setting the first cluster salt iodine content detection record value set as a reference sequence;

s1134, according to the comparison sequence and the reference sequence, gray correlation analysis is carried out to obtain a first cluster correlation set;

S1135, extracting preset influence indexes which are larger than or equal to a relevance threshold value in the first cluster relevance set, and adding the preset influence indexes into a first initial relevance influence index set;

S1136, adding the first initial association-influencing index set to the plurality of initial association-influencing index sets.

In a preferred embodiment, first, the first cluster of salt iodine content detection history data is extracted from a plurality of clusters of salt iodine content detection history data. The first cluster refers to one of a plurality of data clusters formed after the cluster analysis in step S112. Each cluster data represents a collection of test records under a particular salt production condition, with similar production background and process characteristics. Firstly, processing the first cluster of salt iodine content detection historical data, and subsequently sequentially processing other clusters of data to ensure that influence factors under different production conditions are comprehensively analyzed.

Then, a first cluster of preset influence index record characteristic value sets are extracted from the first cluster of salt iodine content detection historical data, and the first cluster of preset influence index record characteristic value sets are set as comparison sequences after dimension removal processing. The first cluster of preset influence index record characteristic value sets comprises actual values of preset influence indexes recorded in the history detection process, such as specific measured values of factors including light source stability, temperature, humidity and the like. Because the dimension and the magnitude of different indexes possibly have larger difference, the dimension removing treatment is carried out on the characteristic values, such as the methods of initial value, average or interval, and the like, so that the influence of the dimension difference on the analysis result is eliminated. The data after the dimensionality removal is used as a comparison sequence for gray correlation analysis and is used for comparing with a reference sequence to evaluate the correlation of the data. Meanwhile, a first cluster salt iodine content detection record value set is extracted from the first cluster salt iodine content detection historical data, and the first cluster salt iodine content detection record value set is set as a reference sequence after dimension removal treatment. The first cluster of salt iodine content detection record value sets comprises salt iodine content measurement results obtained in the history detection process. The detected values are also subjected to dimensionality treatment to ensure comparability with the alignment sequence. The processed data is used as a reference sequence for gray correlation analysis and is a reference standard for evaluating the correlation of each preset influence index. And then, according to the comparison sequence and the reference sequence, gray correlation analysis is performed to obtain a first cluster correlation set. Specifically, firstly, calculating a difference sequence between a reference sequence and each comparison sequence, then, calculating a correlation coefficient at each moment based on a minimum difference, a maximum difference and a resolution coefficient, and finally, calculating an average value to obtain gray correlation degree. By this calculation, the degree of similarity and the degree of association between each preset influence index and the detection result of the iodine content of table salt are quantified, and a first cluster association degree set containing the association degrees of all the preset influence indexes is formed.

And then, extracting a preset influence index which is larger than or equal to the association degree threshold value in the first cluster association degree set, and adding the preset influence index into a first initial association influence index set. The association degree threshold is a preset judgment standard, and represents the minimum association degree required for accepting an index as an association influence index. Screening out preset influence indexes with the association degree higher than an association degree threshold, wherein the indexes have obvious correlation with the detection result of the iodine content of the salt, and the indexes are included in a first initial association influence index set. And then, adding the first initial association influence index set into a plurality of initial association influence index sets, so that the result of analysis on the first cluster of salt iodine content detection historical data is saved into an overall result set, and preparation is made for the subsequent integration of analysis results of each cluster. And in a similar manner, sequentially processing the second cluster until the last cluster of data to obtain a plurality of initial association influence index sets, and finally obtaining the association influence index sets of which the union sets form the distributed nodes.

Through the gray correlation analysis-based method, the correlation between each preset influence index and the detection result of the iodine content of salt can be accurately estimated under the conditions of limited data and incomplete information, thereby providing a basis for determining the correlation influence index and providing technical support for improving the reliability and the accuracy of the detection data. By adopting the gray correlation degree analysis method, the accurate identification of the correlation influence indexes is realized, the influence degree of each preset influence index is comprehensively evaluated on the basis of fully considering different production conditions, and a reliable technical means is provided for confidence sorting of the salt iodine content detection data.

Further, counting the iodine content detection sample set meeting the association influence index deviation vector time sequence information, including:

s41, constructing a deviation state similarity evaluation function based on the association influence index deviation vector time sequence information:

,

Wherein, the Characterizing the similarity of the bias vectors of the ith attribute association influence indicators,Characterizing the total number of i-th attribute association influence index monitoring moments, t-th moment of monitoring moments,Characterizing the ith attribute association affects indicator departure vector timing information,Characterizing sample ith attribute association affects indicator departure vector timing information,Characterizing the ith attribute association effect index t moment offset vector,Characterizing a sample ith attribute association impact index tth moment offset vector,Characterizing all attribute association impact indicator departure vector timing information,The correlation of all attributes of the characterization samples affects the index deviation vector time sequence information, M characterizes the total number of the attributes,,Characterizing a small constant;

S42, evaluating the similarity of the deviation state of the correlation influence index deviation vector time sequence information of the salt sample to be selected and the correlation influence index deviation vector time sequence information according to the similarity evaluation function of the deviation state;

S43, when the similarity of the deviation states is larger than or equal to a similarity threshold, adding the salt sample to be selected into the iodine content detection sample set.

Specifically, a sample screening method based on deviation state similarity evaluation is provided, and an iodine content detection sample set with similar detection conditions is screened out by quantitatively evaluating the deviation state similarity among samples, so that a reliable data basis is provided for subsequent error analysis.

First, based on the associated influence index deviation vector timing information, a deviation state similarity evaluation function is constructed as follows:

,。

the evaluation function adopts the design idea of the structural similarity index, and can effectively quantify the similarity degree of different samples in the association influence index deviation state. The function is divided into two layers, namely, the similarity of the offset vectors of the attribute association influence indexes is calculated firstly, and then the overall similarity is determined based on the similarity of the attributes.

At the attribute level, calculating the similarity of the offset vectors of the ith attribute association influence indexes by adopting a modified cosine similarity formula. The formula takes the dynamic characteristics of the time sequence information into consideration, and the direction consistency and the amplitude similarity of the two time sequence information are quantized through the ratio of the sum of the product and the square sum of the deviation vectors at each moment.The offset vector time sequence information representing the ith attribute association influence index of the current detection sample is a vector sequence which changes along with time; the i-th attribute associated influence index of the sample to be selected in the characterization history database deviates from the vector time sequence information and is also a time sequence vector sequence. The two sequences record the change of the deviation state of different samples on the same associated influence index with time. The method comprises the steps of representing the total number of monitoring moments of an ith attribute associated influence index, namely the number of time points for monitoring the index in the whole detection process, wherein T represents the specific moment in the monitoring process, the value range is1 to T, and the specific moment represents different points in a time sequence. For each time instant t of the day,The deviation vector of the ith attribute association influence index of the current detection sample at the t moment is represented, namely the deviation degree of the monitoring value and the constraint interval at the moment, and the same is true,And (3) representing the deviation vector of the ith attribute association influence index of the sample to be selected at the t moment. These deviation vectors are basic units for calculating the similarity, and reflect the deviation states of the index at each time.Is a small constant for avoiding the situation that the denominator is zero and improving the stability of the algorithm. And obtaining the comprehensive similarity of the whole time sequence by averaging the similarity of all the moments.

At the overall level, taking the minimum value of each attribute similarity as the overall similarity. Wherein, the The time sequence information of the deviation vector representing all attribute association influence indexes of the current detection sample is a set containing all attribute deviation information, and similarly,And (5) representing the deviation vector time sequence information of all attribute association influence indexes of the sample to be selected. M characterizes the total number of attributes, i.e., the number of associated impact indicators. And searching the value with the lowest similarity among all M attributes as an overall evaluation result. Relation typeThe deviation vector indicating each moment is a component of the corresponding time sequence information, and the time sequence information of each attribute is a component of the whole deviation information, so that the hierarchical relationship between the data is clearly described.

And then, evaluating the similarity of the deviation state of the correlation influence index deviation vector time sequence information and the correlation influence index deviation vector time sequence information of the salt sample to be selected according to the similarity evaluation function of the deviation state. Specifically, the time sequence information of the correlation influence index deviation vector of the current detection sample is used as a reference, the time sequence information of the correlation influence index deviation vector of the salt sample to be selected is compared with the time sequence information of the correlation influence index deviation vector of the salt sample to be selected, and the similarity of the deviation states between the two is calculated. The comparison not only considers the direction and the amplitude of the deviation, but also considers the dynamic change characteristic on the time sequence, and can comprehensively evaluate the comparability of the detection conditions among samples. And when the similarity of the deviation states is greater than or equal to a similarity threshold value, adding the salt sample to be selected into the iodine content detection sample set. The similarity threshold is a preset criterion that indicates the minimum similarity required to accept a sample as a reference sample. By means of the screening, a sample set with high similarity in the correlation influence index deviation state is formed, and the samples are comparable with the current detection samples in the aspect of being interfered by influence factors and are suitable to be used as a reference basis for evaluating system errors.

Through the sample screening method based on the similarity evaluation of the deviation states, samples similar to the current detection conditions can be accurately screened out from the historical data, and effective data support is provided for subsequent error analysis. The method fully considers the time sequence characteristic and the multi-attribute characteristic of the correlation influence index deviation, can evaluate the similarity between samples more comprehensively and accurately, and improves the accuracy and reliability of confidence sorting of the salt iodine content detection data. The similarity evaluation method for the deviation state realizes the accurate screening of the detection samples, ensures that the data base of error analysis has high comparability and representativeness, provides a guarantee for obtaining the accurate iodine content detection error mode value, and further improves the accuracy and reliability of the salt iodine content detection.

Further, counting the salt sample iodine content detection value set with the trusted mark to obtain a salt iodine content confidence value, and sending the salt iodine content confidence value to the user side, including:

s61, performing a box graph analysis on the salt sample iodine content detection value set to obtain a box salt sample iodine content detection value set;

And S62, carrying out concentrated trend evaluation on the iodine content detection value set of the box body salt sample to obtain a salt iodine content distribution interval, and setting the salt iodine content distribution interval as the salt iodine content confidence value and sending the salt iodine content confidence value to a user side.

In a preferred embodiment, first, a box chart analysis is performed on a salt sample iodine content detection value set to obtain a box salt sample iodine content detection value set. Specifically, first, the quartiles of the salt sample iodine content detection value set are calculated, including a first quartile Q1 (25% quantile), a second quartile Q2 (median, 50% quantile), and a third quartile Q3 (75% quantile), and the quartile IQR (i.e., Q3-Q1) is determined. Based on the quartile and the quartile range, an upper boundary and a lower boundary are defined, wherein the upper boundary is Q3+1.5IQR, and the lower boundary is Q1-1.5IQR. Samples with iodine content detection values outside these boundaries of the table salt sample are considered as potential outliers or outliers, which are excluded from subsequent analysis. The detection values between the upper boundary and the lower boundary form a detection value set of the iodine content of the salt sample of the box body, and the detection value set is suitable for subsequent concentrated trend analysis. Through the analysis of the box type diagram, extreme values possibly influenced by unknown factors or caused by measurement errors can be effectively identified and eliminated, and the accuracy and reliability of subsequent statistical analysis are improved.

And then, carrying out concentrated trend evaluation on the iodine content detection value set of the salt sample of the box body to obtain a salt iodine content distribution interval, setting the salt iodine content distribution interval as a salt iodine content confidence value, and sending the salt iodine content confidence value to a user side. The concentrated trend evaluation aims at extracting statistics which can represent integral characteristics from the screened box body salt sample iodine content detection value set, and provides basis for determining a final detection result. Firstly, basic statistics, such as mean value, median value, standard deviation or variance, of a box body salt sample iodine content detection value set are calculated, and distribution characteristics of data are comprehensively grasped. On the basis, a statistical method is selected based on the distribution type of the data to determine the distribution interval of the iodine content of the salt. If the data distribution does not accord with the normal assumption, a method based on quantiles, such as taking the median as a center point, extending to two sides to specific quantiles to form a non-parameterized distribution interval. And the determined salt iodine content distribution interval is used as a salt iodine content confidence value and is sent to the user side through a communication network.

Through the steps, abnormal values are effectively removed from the detection value set with the trusted identification, representative centralized trend characteristics are extracted, and the table salt iodine content confidence value with high reliability is formed. The method based on the analysis of the box diagram and the evaluation of the concentrated trend fully considers the distribution characteristic and possible uncertainty of the data, provides accurate and reliable detection results for users, and effectively supports the quality control and supervision management of the salt iodine content. By adopting the method, the scientific conversion from the trusted detection result to the final confidence value is realized, the accuracy and the reliability of the detection of the iodine content of the salt are improved, and the high-quality detection result is provided for the user.

In a second embodiment, as shown in fig. 2, based on the same inventive concept as the confidence sorting method of the salt iodine content detection data provided in the first embodiment, the embodiment of the present invention further provides a confidence sorting system of the salt iodine content detection data, including:

The correlation analysis module 11 is configured to perform correlation analysis on a preset influence index set according to the type of the salt iodine content detection process, so as to obtain a correlation influence index set;

A constraint interval receiving module 12, configured to receive an associated impact indicator constraint interval of the associated impact indicator set;

The deviation vector analysis module 13 is used for collecting the time sequence information of the monitoring value of the associated influence index when the salt sample is detected, and comparing the time sequence information with the constraint interval of the associated influence index to obtain the time sequence information of the deviation vector of the associated influence index;

An error mode statistics module 14, configured to count an iodine content detection error mode value of the iodine content detection sample set that satisfies the correlation impact indicator deviation vector timing information;

The detection value identification module 15 is configured to perform trusted identification on the iodine content detection value of the salt sample when the iodine content detection error count value is less than an error threshold value;

the confidence value statistics module 16 is configured to count the salt sample iodine content detection value set with the trusted identifier, obtain a confidence value of the salt iodine content, and send the confidence value to the user side.

Further, the correlation analysis module 11 includes the following steps:

Transmitting the salt iodine content detection process type and the preset influence index set to a first distributed node, and performing correlation analysis to obtain a first correlation influence index set;

Until the salt iodine content detection process type and the preset influence index set are sent to an N-th distributed node, performing correlation analysis to obtain an N-th associated influence index set;

And sorting the trigger frequency based on the first association influence index set to the Nth association influence index set to obtain the association influence index set.

Further, the correlation analysis module 11 further includes the following steps:

Carrying out statistics on the same attribute triggering frequencies from the first association influence index set to the Nth association influence index set to obtain a first index triggering frequency to a Q index triggering frequency;

Calculating the ratio of the first index trigger frequency to N, and setting the ratio as a first index association degree;

Setting the ratio of the Q index triggering frequency to N as the Q index association degree until the ratio of the Q index triggering frequency to N is calculated;

Extracting the first index association degree until indexes which are larger than or equal to an association degree threshold value in the Q index association degree are added into the association influence index set.

Transmitting the salt iodine content detection process type and the preset influence index set to a first distributed node, and collecting salt iodine content detection historical data, wherein the salt iodine content detection historical data comprise salt production record parameters, preset influence index record characteristic values and salt iodine content detection record values;

Performing cluster analysis on the salt iodine content detection historical data according to the salt production recording parameters to obtain multi-cluster salt iodine content detection historical data;

And traversing the multi-cluster salt iodine content detection historical data to perform gray correlation degree analysis based on the preset influence index record characteristic value and the salt iodine content detection record value, performing correlation analysis to obtain a plurality of initial correlation influence index sets, taking a union set, and adding the union set into the first correlation influence index set.

extracting a first cluster of salt iodine content detection historical data from the multi-cluster salt iodine content detection historical data;

extracting a first cluster of preset influence index record characteristic value sets from the first cluster of salt iodine content detection historical data, performing dimension removal treatment, and setting the first cluster of preset influence index record characteristic value sets as a comparison sequence;

Extracting a first cluster salt iodine content detection record value set from the first cluster salt iodine content detection historical data, performing dimension removal treatment, and setting the first cluster salt iodine content detection record value set as a reference sequence;

According to the comparison sequence and the reference sequence, gray correlation analysis is carried out, and a first cluster correlation set is obtained;

Extracting a preset influence index which is larger than or equal to a relevance threshold value in the first cluster relevance set, and adding the preset influence index into a first initial relevance influence index set;

the first initial set of associated influence indicators is added to the plurality of initial sets of associated influence indicators.

Further, the error mode statistics module 14 includes the following steps:

Based on the associated influence index deviation vector time sequence information, a deviation state similarity evaluation function is constructed:

,

according to the deviation state similarity evaluation function, evaluating the deviation state similarity of the deviation vector time sequence information of the correlation influence indexes of the salt sample to be selected and the deviation vector time sequence information of the correlation influence indexes;

And when the similarity of the deviation state is greater than or equal to a similarity threshold, adding the salt sample to be selected into the iodine content detection sample set.

Further, the confidence value statistics module 16 includes the following steps:

Performing a box graph analysis on the salt sample iodine content detection value set to obtain a box salt sample iodine content detection value set;

and carrying out concentrated trend evaluation on the iodine content detection value set of the box body salt sample to obtain a salt iodine content distribution interval, and setting the salt iodine content distribution interval as the salt iodine content confidence value and sending the salt iodine content confidence value to a user side.

In the foregoing embodiments, the descriptions of the embodiments are focused on, and for those portions of one embodiment that are not described in detail, reference may be made to the related descriptions of other embodiments.

It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts.

It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the present invention and the equivalent techniques thereof, the present invention is also intended to include such modifications and variations.

Claims

1. A confidence sorting method for salt iodine content detection data, comprising:

According to the type of salt iodine content detection process, correlation analysis is performed on the preset influencing indicator set to obtain the associated influencing indicator set;

receiving an associated impact indicator constraint interval of the associated impact indicator set;

When the salt sample is tested, the time series information of the monitoring value of the correlation influence indicator is collected and compared with the constraint interval of the correlation influence indicator to obtain the time series information of the deviation vector of the correlation influence indicator;

Counting the iodine content detection error mode value of the iodine content detection sample set that satisfies the correlation influence indicator deviation vector time series information;

When the iodine content detection error mode value is less than the error threshold, the iodine content detection value of the salt sample is marked as credible;

Collect statistics on the iodine content test values of salt samples with trustworthy identification, obtain the confidence value of the iodine content of salt and send it to the user end;

Among them, according to the type of salt iodine content detection process, a correlation analysis is performed on the preset influencing indicator set to obtain a set of associated influencing indicators, including:

Sending the salt iodine content detection process type and the preset impact indicator set to a first distributed node, performing correlation analysis, and obtaining a first correlation impact indicator set;

Until the salt iodine content detection process type and the preset impact indicator set are sent to the Nth distributed node, correlation analysis is performed, and the Nth correlation impact indicator set is obtained;

Trigger frequency sorting is performed based on the first association impact indicator set to the Nth association impact indicator set to obtain the association impact indicator set.

2. The confidence sorting method for salt iodine content detection data according to claim 1, characterized in that trigger frequency sorting is performed based on the first correlation influence indicator set to the Nth correlation influence indicator set to obtain the correlation influence indicator set, comprising:

Performing same-attribute trigger frequency statistics on the first associated impact indicator set to the Nth associated impact indicator set to obtain first indicator trigger frequencies to Qth indicator trigger frequencies;

Calculate the ratio of the first indicator trigger frequency to N, and set it as the first indicator correlation degree;

Until the ratio of the trigger frequency of the Qth indicator to N is calculated and set as the Qth indicator correlation degree;

Indicators from the first indicator correlation degree to the Qth indicator correlation degree that are greater than or equal to a correlation degree threshold are extracted and added into the correlation impact indicator set.

3. The confidence sorting method for salt iodine content detection data according to claim 1, characterized in that the salt iodine content detection process type and the preset impact indicator set are sent to a first distributed node, and correlation analysis is performed to obtain a first correlation impact indicator set, comprising:

Sending the salt iodine content detection process type and the preset influencing indicator set to a first distributed node, and collecting salt iodine content detection history data, wherein the salt iodine content detection history data includes salt production record parameters, preset influencing indicator record characteristic values, and salt iodine content detection record values;

performing cluster analysis on the historical data of iodine content detection of salt according to the salt production record parameters to obtain multiple clusters of historical data of iodine content detection of salt;

Based on the preset influencing indicator record characteristic values and the salt iodine content detection record values, the multiple clusters of salt iodine content detection historical data are traversed to perform grey correlation analysis, and correlation analysis is performed to obtain multiple initial correlation influencing indicator sets, take the union set, and add it to the first correlation influencing indicator set.

4. The confidence sorting method for salt iodine content detection data according to claim 3 is characterized in that, based on the preset influence index record characteristic value and the salt iodine content detection record value, traversing the multiple clusters of salt iodine content detection historical data to perform grey correlation analysis, performing correlation analysis, and obtaining multiple initial correlation influence index sets, including:

Extracting a first cluster of table salt iodine content detection historical data from the multiple clusters of table salt iodine content detection historical data;

Extracting a first cluster of preset influencing indicator record feature value sets from the first cluster of table salt iodine content detection historical data, de-dimensionalizing the sets and setting them as a comparison sequence;

Extracting a first cluster of table salt iodine content detection record value sets from the first cluster of table salt iodine content detection historical data, de-dimensionalizing the sets, and setting the sets as a reference sequence;

Performing grey relational analysis on the comparison sequence and the reference sequence to obtain a first cluster relational degree set;

Extracting preset impact indicators greater than or equal to a correlation threshold from the first cluster correlation set, and adding them to a first initial correlation impact indicator set;

The first initial correlation impact indicator set is added to the multiple initial correlation impact indicator sets.

5. The confidence sorting method for salt iodine content detection data according to claim 1 is characterized in that the iodine content detection sample set that satisfies the correlation influence index deviation vector time series information is statistically analyzed, comprising:

Based on the time series information of the deviation vector of the correlation impact indicator, a deviation state similarity evaluation function is constructed:

,

in, Characterizes the deviation vector similarity of the i-th attribute association impact index, represents the total number of monitoring moments of the i-th attribute-related impact indicator, t represents the t-th moment of the monitoring time, Characterizes the time series information of the deviation vector of the impact index of the i-th attribute association, Characterizes the time series information of the deviation vector of the impact index of the i-th attribute of the sample, Characterizes the deviation vector of the impact index of the i-th attribute association at the t-th moment, Characterizes the deviation vector of the sample's attribute-i correlation impact index at time t, Characterizes the time series information of the deviation vector of all attribute association impact indicators, Represents the deviation vector time series information of all attribute correlation impact indicators of the sample, M represents the total number of attributes, , Characterize small constants;

According to the deviation state similarity evaluation function, the deviation state similarity between the time series information of the deviation vector of the correlation influence indicator of the selected salt sample and the time series information of the deviation vector of the correlation influence indicator is evaluated;

When the deviation state similarity is greater than or equal to the similarity threshold, the salt sample to be selected is added to the iodine content detection sample set.

6. The confidence sorting method for salt iodine content detection data according to claim 1 is characterized in that the iodine content detection value set of salt samples with a trustworthy identifier is counted to obtain a salt iodine content confidence value and send it to a user terminal, comprising:

Performing a box plot analysis on the set of iodine content detection values of the salt samples to obtain a set of iodine content detection values of the box salt samples;

A central tendency assessment is performed on the set of iodine content detection values of the box salt samples to obtain a distribution interval of the iodine content of the salt, which is set as the confidence value of the iodine content of the salt and sent to the user end.

7. A confidence sorting system for salt iodine content test data, characterized in that it is used to implement the method according to any one of claims 1 to 6, and the system comprises:

A correlation analysis module is used to perform correlation analysis on a preset influencing indicator set according to the type of salt iodine content detection process to obtain a related influencing indicator set;

A constraint interval receiving module, configured to receive the associated impact indicator constraint interval of the associated impact indicator set;

The deviation vector analysis module is used to collect the time series information of the monitoring value of the correlation influence indicator when the salt sample is tested, compare it with the constraint interval of the correlation influence indicator, and obtain the time series information of the deviation vector of the correlation influence indicator;

An error mode statistics module, configured to count the iodine content detection error mode values of the iodine content detection sample set that satisfies the correlation influence index deviation vector time series information;

A detection value identification module, configured to identify the iodine content detection value of the salt sample as credible when the iodine content detection error mode value is less than an error threshold;

The confidence value statistics module is used to collect statistics on the iodine content detection value set of salt samples with a trustworthy identifier, obtain the confidence value of the iodine content of the salt and send it to the user end.