Intelligent application system of fluid geochemical data

Abstract

The application relates to the technical field of data management, and discloses a fluid geochemical data intelligent application system; the system comprises a data acquisition module, a database construction module, an abnormality judgment module, an index calculation module and a report analysis module; the data acquisition module is used for combining original data into a data set; the database construction module is used for constructing a traceability database; the abnormality judgment module is used for judging whether abnormal data exist; the index calculation module is used for calculating an application abnormality index; and the report analysis module is used for determining an application level and generating an application authentication report; the application can associate and integrate original data in a discrete state into a structured data set, avoids the phenomenon of repeated interference between different types of original data, can ensure that a traceability link can be quickly and accurately traced back to data with abnormal phenomena, fundamentally solves the data management and application defects of data isolation between different dimensions and difficult data tracing, and provides an accurate and reliable implementation path for complete traceability required by a laboratory CNAS system authentication.

Description

Technical Field

[0001] This invention relates to the field of data management technology, and more specifically, to an intelligent application system for fluid geochemical data. Background Technology

[0002] Fluid geochemistry laboratories continuously generate massive amounts of multi-source and heterogeneous data during operation. These data collectively constitute the core data chain of the laboratory's quality management system. The effectiveness of data management and application directly affects data quality, report credibility, and the certification results of laboratory compliance. In order to improve the reliability of laboratory data management and application and accelerate the certification of the laboratory's quality management system, it is necessary to implement intelligent management and application of the laboratory's diverse data.

[0003] Reference patent application CN112200547A discloses a laboratory scientific data management system, comprising: a document acquisition module for acquiring documents that need to be processed and stored; a document parsing and recognition module for extracting information from the documents acquired by the document acquisition module; a data storage module for storing the document data parsed and recognized by the document parsing and recognition module; a document storage module for saving the original files of the document data parsed and recognized by the document parsing and recognition module; and an information query and usage module for searching, querying, and importing the document data stored in the document storage module and the original files of the documents. In existing laboratory data application systems, data from different dimensions are often stored in isolation or discretely, lacking an effective mechanism for correlation and integration between different dimensions. This leads to data redundancy and duplicate interference. When it is necessary to assess data quality or conduct specific problem analysis, it is difficult to quickly and dynamically correlate data from different sources, and it is also impossible to accurately trace and locate abnormal data. Consequently, the laboratory data management and application process suffers from the defects of isolated data from different dimensions and difficulty in tracing the source.

[0004] In view of this, the present invention proposes an intelligent application system for fluid geochemical data to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a fluid geochemical data intelligent application system, comprising: The data acquisition module is used to extract the data features of the raw data, including the recording time, content attributes and target objects, and to combine the raw data into a data set based on the data features. The database construction module is used to identify baseline data from the dataset, use the baseline data as the central node, and combine the node association criteria to construct a traceability database with traceability links. The anomaly detection module is used to construct an evaluation scenario with an interactive channel, dynamically integrate the traceability database with the evaluation scenario, and set the evaluation mechanism of the interactive channel to determine whether there is abnormal data in the traceability database. The index calculation module is used to trace abnormal data as the starting point and the tracing link as the path trajectory, to spread the tracing starting point into an abnormal path, and to calculate the application abnormal index through the path impact parameters of the abnormal path, including the proportion of abnormal points and the proportion of hierarchical association. The report analysis module is used to determine the application level of the laboratory based on the application anomaly index, and to summarize the data of the laboratory to generate an application certification report.

[0006] Furthermore, the methods for combining data sets are as follows: Query out the target objects of all the original data one by one, and summarize the original data with the same target objects to generate a basic set; The overlap of the content attributes of the original data in the base set is compared sequentially, and the original data with the same overlap level are aggregated to generate a secondary set. Mark the recording time of the original data in the secondary set one by one. According to the chronological order, summarize the original data in the same time period to generate the top-level set. Then, arrange the original data in the base set, secondary set and top-level set respectively to generate A data sets.

[0007] Furthermore, the method for identifying the benchmark data is as follows: A1: Import A sets of data into a three-dimensional space, and mark the spatial coordinates of the spatial points where the original data is located, based on the standard that one original data corresponds to one spatial point. A2: Randomly select a spatial point as the standard point, calculate the Euclidean distance between the standard point and the remaining spatial points one by one using the Euclidean distance calculation formula, and then sum all the Euclidean distance values ​​and calculate the average to obtain the effective distance value; A3: Traverse all spatial points, repeat A2, and calculate B effective distance values; A4: Calculate the overlap difference by taking the absolute value of the difference between the overlap of the original data at the standard point and the overlap of the remaining original data. Then, calculate the standard deviation of the overlap difference using the standard deviation calculation formula. A5: Iterate through all the original data and repeat A4 to obtain B standard deviations; A6: After assigning corresponding scaling factors to the effective distance value and standard deviation respectively, perform a weighted comparison to calculate the benchmark index, and record the original data corresponding to the maximum value of the benchmark index as the benchmark data.

[0008] Furthermore, the node association criteria are as follows: nodes within the same data set are associated point by point, while nodes in different data sets are associated centrally. The method for constructing the traceability database is as follows: Establish a blank database with A library units, import A data sets one by one into the A library units to generate A related libraries, and calculate the length of C nodes by multiplying the baseline index of C original data in the data sets by the calibrated unit length. Using the baseline data as the central node, C non-adjacent end nodes are marked in a ring shape. The tracing distance from the end nodes to the central node is continuously adjusted until the C tracing distances correspond one-to-one with the C node lengths. The original data is then bound to the central node and the end nodes one-to-one, causing the associated database to be converted into the base database. Within the same basic database, a two-way data channel is established between the central node as the starting point of the traceability and the end node as the ending point of the traceability. A bidirectional data transmission channel is established between two central nodes within two different base libraries; Using the central node as a reference, pointing arrows are configured on the data channel to point to the end nodes, simulating a tree-like traceability link. All traceability links are then numbered in ascending order to construct a traceability database.

[0009] Furthermore, the method for constructing the evaluation scenario is as follows: A scene outline with closed boundaries is constructed using virtual simulation software, and A non-adjacent evaluation units are established within the scene outline. Using the preset boundary length as a standard, C protocol points are marked at equal intervals on the closed boundary, and evaluation permissions and evaluation parameters are configured on each of the C protocol points to generate a protocol fence. Using the same data capacity as a standard, dividers are set within the evaluation unit to divide the evaluation unit into evaluation positions and standard positions. D bidirectional interactive channels are established on the dividers to upgrade the scene outline into an evaluation scene.

[0010] Furthermore, the evaluation mechanism is as follows: all interaction channels during a single evaluation maintain time synchronization. The method for determining whether there is abnormal data is as follows: The data fusion algorithm dynamically integrates the traceability database with the evaluation scenario, and imports A basic databases one by one into the evaluation positions of A evaluation units; Based on the original data of the evaluation position, the corresponding certification data is indexed from the database, the certification data is imported into the standard position of the evaluation unit one by one, and the original data is bound to the corresponding certification data one by one to obtain E evaluation samples. Based on the standard of one-to-one correspondence between interactive channels and evaluation samples, the evaluation samples are embedded in the interactive channels, and anomaly analysis instructions are sent synchronously to all interactive channels to analyze the authentication status of the original data. When the authentication status is non-compliant, it is determined that there is abnormal data. When the authentication status is compliant, it is determined that there is no abnormal data.

[0011] Furthermore, the method for tracing and propagating abnormal paths is as follows: B1: Record the node where the abnormal data is located in the traceability database as the starting point of the path, and compare the data characteristics of the remaining nodes with the original data at the starting point of the path for consistency. B2: When the remaining node has the same recording time, content attributes and target object as the path start point, the node is recorded as a waypoint, and the tracing link is used as the path trajectory. The tracing start point is traced and spread to the waypoint according to the pointing arrow. B3: Repeat steps B1-B2 to obtain F waypoints, and then connect all the waypoints in sequence to generate an abnormal path.

[0012] Furthermore, the calculation method for the anomaly index is as follows: In the traceability database, the number of transit points and the number of nodes are counted separately, and the percentage of abnormal points is calculated by dividing the number of transit points by the number of nodes. Record the path points that have two or three consistent records in the time, content attributes, and target objects as first-level points and second-level points, respectively. Divide the number of first-level points and the number of second-level points by the number of path points, add them together, and calculate the average to obtain the hierarchical association ratio. The application anomaly index is calculated by weighting and summing the corresponding proportional coefficients assigned to the proportion of abnormal points and the proportion of hierarchical associations.

[0013] Furthermore, the application levels include low-utility, medium-utility, and high-utility levels: The method for determining the application level is as follows: The laboratory application abnormality index Each is compared with the preset first-level index Second-level index Comparison, Greater than ; when Greater than or equal to At that time, the application level was determined to be inefficient. when Less than ,and Greater than or equal to At that time, the application level was determined to be the medium utility level; when Less than At that time, the application level was determined to be the high-efficiency application level.

[0014] Furthermore, when generating the application certification report, a blank basic electronic report is created. Seven link units are set up in the basic report. The laboratory's original data, data set, traceability database, abnormal data, abnormal paths, abnormal index, and application level are compressed and linked to obtain seven data links. The seven data links are then imported into the seven link units one by one, and the unit explanation is noted on the link unit to generate the application certification report.

[0015] The technical advantages of the intelligent application system for fluid geochemical data of this invention are as follows: (1): This invention can integrate discrete raw data into a structured data set by combining the three-level combination method from the basic set to the secondary set and then to the top set, avoiding the phenomenon of repeated interference between different types of raw data. By taking the benchmark data as the central node and combining the principle of linking the same set point by point and linking different sets centrally, a traceability database with traceability function can be constructed to ensure that the traceability link can quickly and accurately trace back to the data with abnormal phenomena. This fundamentally solves the defects of laboratory data management and application such as data isolation and difficulty in tracing between different dimensions, and provides an accurate and reliable implementation path for the complete traceability required by the laboratory CNAS system certification.

[0016] (2): By constructing an evaluation scenario that matches the traceability database and setting a time-synchronized evaluation mechanism, this invention can provide a quick channel for analyzing and evaluating different types of raw data, achieve accurate location of abnormal data, and, combined with the method of generating abnormal paths through traceability diffusion, dynamically correlate the negative impact of abnormal data in the traceability database. This allows for correlation analysis of potential hidden dangers in data management and application in the laboratory, avoiding the limitations of single data analysis and evaluation, and thus meeting the comprehensive evaluation requirements of the laboratory's CNAS system certification. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of a fluid geochemical data intelligent application system provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart illustrating a method for intelligent application of fluid geochemical data provided in Embodiment 2 of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Please refer to Figure 1 As shown in this embodiment, a fluid geochemical data intelligent application system includes: The data acquisition module collects multi-dimensional raw data from the laboratory, extracts the data features of the raw data, and combines the raw data into a dataset. Raw data refers to the unprocessed initial data collected by the laboratory during the CNAS system certification review. Raw data can provide data support for the quantitative evaluation of fluid geochemistry laboratories in CNAS system certification from multiple different dimensions and levels, and lay the foundation for laboratory data management and application in the earthquake system. In this embodiment, the raw data needs to include and cover the entire process data during normal laboratory operation to meet the comprehensive requirements of CNAS system certification and review.

[0020] Specifically, the raw data includes fluid geochemical data, equipment operation and maintenance calibration data, material scheduling and management data, and experimental operation record data.

[0021] Fluid geochemical data refers to the relevant geological data generated during the monitoring and analysis of fluid geochemistry in the laboratory, and serves as the professional data for the laboratory to obtain CNAS system certification. Specifically, fluid geochemical data includes, but is not limited to, major ion concentration units, trace element concentration units, isotope ratios, isotope radioactivity, dissolved gas ratios, etc.; in this embodiment, the fluid geochemical data is collected using equipment such as ion chromatographs, inductively coupled plasma mass spectrometers, and isotope ratio mass spectrometers.

[0022] Equipment operation and maintenance calibration data refers to the relevant maintenance data of related equipment and instruments in the laboratory during the testing, operation and maintenance process, and is used as the equipment data for the laboratory to conduct CNAS system certification; Specifically, equipment operation and maintenance calibration data includes, but is not limited to, instrument calibration certificates, historical verification records, maintenance work logs, core operating parameters, etc.; in this embodiment, the equipment operation and maintenance calibration data is collected through the system database and various sensors.

[0023] Material scheduling and management data refers to the relevant data on the daily scheduling, use, and operation of materials and reagents in the laboratory, and is used as material data for CNAS system certification; Specifically, material scheduling management data includes, but is not limited to, standard material balance, reagent warehousing time, historical requisition time, and current inventory information; in this embodiment, the material scheduling management data is obtained by querying the system database.

[0024] Experimental operation record data refers to the relevant data generated during the experimental operation and testing process in the laboratory, and is used as experimental data for CNAS system certification; Specifically, the experimental operation record data includes, but is not limited to, the qualifications of the experimental personnel, the experimental environment, and the work record manual; in this embodiment, the experimental operation record data is obtained by querying the system database.

[0025] Once the raw data is obtained, it can be identified and its data features extracted to provide reliable data support for subsequent classification of the raw data. Data characteristics are used to represent the type and meaning of raw data in the laboratory, and serve as a direct basis for subsequent differentiation of raw data; Specifically, data characteristics include recording time, content attributes, and target object; recording time refers to the point in time when the original data was collected and recorded, content attributes refer to the specific type of meaning of the original data content, and target object refers to the direct object that the original data is directed to and associated with.

[0026] In this embodiment, data features are obtained after parsing using natural language processing technology.

[0027] After obtaining the data characteristics of the raw data, the raw data can be distinguished and combined based on the data characteristics, thereby forming a data set that can be directly used for subsequent analysis and evaluation, and ensuring that the raw data in each data set can maintain a direct correlation. The methods for combining data sets are as follows: Query out the target objects of all the original data one by one, and summarize the original data with the same target objects to generate a basic set; The overlap of the content attributes of the original data in the base set is compared sequentially, and the original data with the same overlap level are aggregated to generate the secondary set. The overlap level is a numerical representation used to keep the repetition of the content attributes in the original data within a certain range, so that the overlap of the content attributes of the original data in the secondary set can be the same or close. The recording times of the original data within each secondary set are marked one by one. The original data within the same time period are then aggregated according to chronological order to generate a top-level set. The original data within the base set, secondary set, and top-level set are then arranged to generate A data sets. A time period is used to represent the recording times of the original data within a certain range, ensuring that the recording times of the original data within the top-level set remain within a similar time span.

[0028] It should be noted that the original data within each dataset is unique, ensuring that the same original data will not appear in different datasets. This avoids the impact of duplicate data and provides a unique traceability target for subsequent tracing of the original data.

[0029] The database construction module identifies baseline data from the dataset, uses the baseline data as the central node, and constructs a traceability database with traceability links through node association criteria. Benchmark data refers to the original data in a dataset that maintains an optimal spatial distance and overlap with any original data. Benchmark data can serve as the basis for subsequent dataset management. The method for identifying benchmark data is as follows: A1: Import A sets of data into a three-dimensional space, and mark the spatial coordinates of the spatial points where the original data is located, based on the standard that one original data corresponds to one spatial point. A2: Randomly select a spatial point as the standard point, calculate the Euclidean distance between the standard point and the remaining spatial points one by one using the Euclidean distance calculation formula, and then sum all the Euclidean distance values ​​and calculate the average to obtain the effective distance value; the Euclidean distance calculation formula is existing technology in this field and will not be described in detail here; A3: Traverse all spatial points, repeat A2, and calculate B effective distance values; A4: Calculate the overlap difference by taking the absolute value of the difference between the overlap of the original data at the standard point and the overlap of the remaining original data. Then, calculate the standard deviation of the overlap difference using the standard deviation calculation formula. The standard deviation calculation formula is existing technology in this field and will not be described in detail here. A5: Iterate through all the original data and repeat A4 to obtain B standard deviations; A6: After assigning corresponding proportional coefficients to the effective distance value and standard deviation respectively, perform weighted comparisons to calculate the benchmark index, and record the original data corresponding to the maximum value of the benchmark index as the benchmark data; The formula for calculating the benchmark index is: ; In the formula, As a benchmark index, For valid distance values, Standard deviation This is the proportionality coefficient of the effective distance value. is the proportionality coefficient of the standard deviation, and , All are greater than 0.

[0030] It should be noted that each dataset has a unique baseline, which enables a quantitative matching effect between the baseline and the dataset.

[0031] After identifying the baseline data, the original data in the same and different data sets can be dynamically linked based on the baseline data, thereby constructing the data set into a traceability database that can store and query the original data; In this embodiment, when constructing the traceability database, it is necessary to integrate different data sets. However, the original data in each data set is independent and discrete, making it impossible for the directly constructed traceability database to dynamically associate the original data in different data sets. Therefore, it is necessary to use node association criteria to dynamically associate different original data and ultimately ensure the interactive effect of the traceability database.

[0032] Specifically, the node association criteria are as follows: the same data set is associated point by point, and different data sets are associated centrally; this can achieve the effect of as-needed association between the original data and the benchmark data within the same data set and different data sets, ensuring the dynamic association of the subsequent traceability database.

[0033] The method for constructing the traceability database is as follows: Create a blank database with A library units, and import A data sets one by one into the A library units, thereby converting the A library units into A related databases; The lengths of the C nodes are calculated by multiplying the baseline indices of the C original data in the dataset by the calibrated unit length. The calibrated unit length is the pre-defined interval between two adjacent nodes and provides a numerical basis for the specific location of the subsequent end nodes. Using the baseline data as the central node, C non-adjacent end nodes are marked in a ring shape. The tracing distance from the end nodes to the central node is continuously adjusted until the C tracing distances correspond one-to-one with the C node lengths. The original data is then bound to the central node and the end nodes one-to-one, causing the associated database to be converted into the base database. Within the same basic database, a two-way data channel is established between the central node as the starting point of the traceability and the end node as the ending point of the traceability. A bidirectional data transmission channel is established between two central nodes within two different base libraries; Using the central node as a reference, directional arrows pointing to the end nodes are configured on the data channel to simulate a tree-like traceability link. All traceability links are then sequentially numbered in ascending order, transforming a blank database into a traceability database. This tree-like traceability link ensures the multidirectional and orderly nature of data during the traceability process, thus providing rapid forward and reverse traceability for subsequent data tracing.

[0034] It should be noted that the traceability link is a channel used to dynamically associate raw data from different locations and levels in the traceability database, avoiding the isolation of different raw data. The tree-like distribution method can ensure the multi-directionality and comprehensiveness of the traceability link, and improve the reliability of subsequent raw data traceability results.

[0035] The anomaly detection module constructs an evaluation scenario that matches the traceability database, dynamically integrates the traceability database with the evaluation scenario, and sets up an evaluation mechanism to determine whether there is abnormal data in the traceability database. The assessment scenario is a virtual scenario used to comprehensively analyze and evaluate the authenticity, rationality, and validity of the original data stored in the traceability database. It provides an assessment environment for whether the original data in the traceability database meets the quantitative requirements under the CNAS system certification. Since the traceability database consists of multiple base databases and traceability links, and the original data stored in different base databases are not consistent, while the traceability links can provide an interactive transmission channel for the original data in different base databases, the constructed evaluation scenario needs to be adapted to the form and structure of the traceability database to improve the accuracy of subsequent evaluation of the original data in the traceability database.

[0036] Specifically, the method for constructing the evaluation scenario is as follows: A scene outline with closed boundaries is constructed using virtual simulation software, and A non-adjacent evaluation units are established within the scene outline. Using a preset boundary length as a standard, C protocol points are marked at equal intervals on the closed boundary, and evaluation permissions and evaluation parameters are configured on each of the C protocol points to generate a protocol fence. The preset boundary length refers to the distance between two adjacent protocol points to avoid two adjacent protocol points being too close or too far apart, thus laying the distance foundation for the components of the protocol fence. Using the same data capacity as a standard, dividers are set within the evaluation unit to divide the evaluation unit into evaluation positions and standard positions. D bidirectional interactive channels are established on the dividers to upgrade the scene outline into an evaluation scene.

[0037] It should be noted that evaluation permissions are used to limit the scope of the evaluation simulation operation of the raw data to be evaluated, ensuring that raw data at the same permission level can be accurately evaluated and avoiding the phenomenon of evaluation beyond the permission level; evaluation parameters are used to impose various restrictions on the evaluation process of raw data. Specifically, evaluation parameters include, but are not limited to, evaluation duration, evaluation order, and unit evaluation quantity.

[0038] After constructing the assessment scenario, the traceability database can be dynamically integrated with the assessment scenario, enabling the traceability database to perform CNAS system certification assessment within the assessment scenario, and ultimately determine whether there is any abnormal data in the traceability database that does not meet the quantitative requirements under the CNAS system certification.

[0039] When evaluating the raw data in the traceability database, it is necessary to conduct the evaluation process within the constraints of the evaluation mechanism in order to ensure that the evaluation process is orderly and reasonable. The evaluation mechanism is as follows: the interaction channels during each evaluation maintain time synchronization, which ensures the time consistency of each evaluation of the raw data.

[0040] The method for determining whether there is abnormal data is as follows: The data fusion algorithm dynamically integrates the traceability database with the evaluation scenario, and imports A basic databases one by one into the evaluation positions of A evaluation units; Based on the original data of the evaluation position, the corresponding certification data is indexed from the database, and the certification data is imported into the standard position of the evaluation unit one by one. The original data and the corresponding certification data are bound one by one to obtain E evaluation samples. The certification data is the original data that meets the CNAS system certification and is pre-stored in the database. Based on the standard of one-to-one correspondence between interactive channels and evaluation samples, the evaluation samples are embedded in the interactive channels, and anomaly analysis instructions are sent synchronously to all interactive channels to analyze the authentication status of the original data. When the authentication status is non-compliant, there is an anomaly between the original data and the authentication data, and it is determined that there is abnormal data. When the authentication status is compliant, there are no anomalies between the original data and the authentication data, so it is determined that there is no abnormal data.

[0041] In this embodiment, the anomaly analysis instruction refers to the instruction to analyze whether the original data and the corresponding certification data comply with CNAS system certification, thereby determining whether the original data in the laboratory meets the quantitative requirements of CNAS system certification. The certification status is used to represent the results of the anomaly analysis and serves as a direct basis for judging abnormal data. For example, when the original data is the equipment maintenance cycle, the certification data is the equipment maintenance certification duration. When the equipment maintenance cycle is greater than the equipment maintenance certification duration, the certification status of the original data is non-compliant; conversely, the opposite is true.

[0042] The index calculation module, if there is abnormal data, takes the abnormal data as the starting point of the path and the tracing link as the path trajectory, performs tracing diffusion on the tracing starting point, simulates the abnormal path, and calculates the application abnormal index through the path influence parameters of the abnormal path. When abnormal data exists, the traceability database contains original data that does not meet the quantitative requirements of CNAS system certification. This abnormal data will negatively affect the laboratory's CNAS system certification. In order to clarify the degree of negative impact caused by abnormal data and facilitate subsequent intelligent management and application optimization of abnormal data, it is necessary to determine the impact trajectory and path of abnormal data in the traceability database, and then simulate the abnormal path. In this embodiment, the abnormal path is a data trajectory that uses abnormal data as the starting point and the traceability link as the path trajectory to represent the abnormal impact range of the abnormal data within the traceability database.

[0043] The method for tracing and propagating abnormal paths is as follows: B1: Record the node where the abnormal data is located in the traceability database as the starting point of the path, and compare the data characteristics of the remaining nodes with the original data at the starting point of the path for consistency. B2: When the remaining node has the same recording time, content attributes and target object as the starting point of the path, then there is a dynamic relationship between the node and the abnormal data. The node is recorded as a waypoint, and the tracing link is used as the path trajectory. The tracing starting point is traced and spread to the waypoint according to the pointing arrow. B3: Repeat steps B1-B2 to obtain F waypoints, and then connect all the waypoints in sequence to generate an abnormal path.

[0044] Once the abnormal path is obtained, the path influence parameters in the abnormal data can be collected, so that the path influence parameters can serve as the data basis for judging the degree of abnormality of the abnormal path, thereby providing data support for the subsequent calculation of the abnormality index. In this embodiment, the anomaly application index is a numerical representation of the anomaly path corresponding to the abnormal data that has deviated from CNAS system certification in the traceability database, which can provide a reference for the subsequent correction and optimization of the original data in the laboratory.

[0045] Specifically, the path impact parameters include the proportion of abnormal points and the proportion of hierarchical associations.

[0046] The calculation method for the application anomaly index is as follows: In the traceability database, the number of transit points and the number of nodes are counted separately, and the percentage of abnormal points is calculated by dividing the number of transit points by the number of nodes. The formula for calculating the percentage of abnormal locations is: ; In the formula, This represents the percentage of abnormal locations. The number of waypoints, The number of nodes; Record the path points that have two or three consistent records in the time, content attributes, and target objects as first-level points and second-level points, respectively. Divide the number of first-level points and the number of second-level points by the number of path points, add them together, and calculate the average to obtain the hierarchical association ratio. The formula for calculating the percentage of hierarchical association is: ; In the formula, For the proportion of hierarchical association, The number of points in the first level. This represents the number of points in the second level. After assigning corresponding proportional coefficients to the proportion of abnormal points and the proportion of hierarchical associations respectively, the weighted sum is calculated to obtain the application anomaly index; The formula for calculating the anomaly index is as follows: ; In the formula, To apply the anomaly index, This is the percentage coefficient of abnormal locations. This is the proportional coefficient for the hierarchical association ratio, and , The setting logic is the same as the above. , The setting logic is consistent.

[0047] The report analysis module determines the application level of the laboratory by applying anomaly indices, and summarizes the data of the laboratory to generate an application certification report; After obtaining the application anomaly index, the anomaly index can be used as the basis for judging the compliance of the laboratory's management and application under the CNAS system certification, and thus determine the laboratory's application level, so that the application level serves as the output result of the laboratory's performance in actual data management and application. Application levels include low-utility, medium-utility, and high-utility levels; and the data management and application-level performance corresponding to the low-utility, medium-utility, and high-utility levels increases from low to high.

[0048] The method for determining the application level is as follows: The laboratory application abnormality index Each is compared with the preset first-level index Second-level index Comparison, Greater than The preset first-level index and second-level index refer to the critical values ​​of the application anomaly index corresponding to the low-utility level and the medium-utility level, and the medium-utility level and the high-utility level, respectively, which can provide a reasonable numerical basis for determining the application level. when Greater than or equal to This indicates that the laboratory's performance under the CNAS system certification is poor, and the application level is determined to be a low-efficiency level. when Less than ,and Greater than or equal to This indicates that the laboratory's performance under the CNAS system certification is average, and the application level is determined to be medium utility level; when Less than This indicates that the laboratory has performed well under the CNAS system certification, and the application level is determined to be the high-efficiency application level.

[0049] After obtaining the application level, the laboratory's data management and application performance under the CNAS system certification are analyzed and quantified, and corresponding output results are given for each performance and dimension. In order to better demonstrate the laboratory's comprehensive performance in fluid geochemical data and provide accurate and reliable improvement suggestions for subsequent CNAS system certification, it is necessary to summarize the laboratory's multi-dimensional data to generate an application certification report for external display. In this embodiment, the application certification report is used to summarize and display the laboratory's raw data, datasets, traceability database, abnormal data, abnormal paths, abnormal indices, and application levels, and serves as the final output result.

[0050] Specifically, when generating the application certification report, a blank basic electronic report is first created. Seven linking units are set within the basic report. Then, the laboratory's raw data, dataset, traceability database, abnormal data, abnormal paths, abnormal indices, and application levels are compressed and linked to obtain seven data links. Finally, the seven data links are imported into the seven linking units one by one, and the unit explanation is noted on the linking units, so that the basic electronic report can generate the application certification report.

[0051] It should be noted that the unit definition is a concise representation of the data information contained in each linked unit, so that the data type and meaning of each linked unit can be accurately understood without clicking the data link; for example, the unit definition of the linked unit that imports raw data is "raw", and the unit definition of the linked unit that imports abnormal paths is "path".

[0052] Example 2: Please refer to Figure 2 As shown, the parts not described in detail in this embodiment are described in Embodiment 1. A method for intelligent application of fluid geochemical data is provided, implemented through an intelligent application system for fluid geochemical data, including: S01: Extract the data features of the original data and combine the original data into a dataset based on the data features; S02: Identify baseline data from the dataset, use the baseline data as the central node, and construct a traceability database with traceability links by combining the node association criteria; S03: Construct an evaluation scenario with an interactive channel, dynamically integrate the traceability database with the evaluation scenario, and set the evaluation mechanism of the interactive channel to determine whether there is abnormal data in the traceability database; if abnormal data exists, execute S04; if no abnormal data exists, end. S04: Using abnormal data as the starting point of the path and the tracing link as the path trajectory, the tracing starting point is traced and spread into an abnormal path, and the application anomaly index is calculated through the path influence parameters of the abnormal path. S05: Determine the application level of the laboratory based on the application anomaly index, summarize the data of the laboratory, and generate an application certification report.

[0053] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A fluid geochemical data intelligent application system, characterized in that, include: The data acquisition module is used to extract the data features of the raw data, including the recording time, content attributes and target objects, and to combine the raw data into a data set based on the data features. The database construction module is used to identify baseline data from the dataset, use the baseline data as the central node, and combine the node association criteria to construct a traceability database with traceability links. The anomaly detection module is used to construct an evaluation scenario with an interactive channel, dynamically integrate the traceability database with the evaluation scenario, and set the evaluation mechanism of the interactive channel to determine whether there is abnormal data in the traceability database. The index calculation module is used to trace abnormal data as the starting point and the tracing link as the path trajectory, to spread the tracing starting point into an abnormal path, and to calculate the application abnormal index through the path impact parameters of the abnormal path, including the proportion of abnormal points and the proportion of hierarchical association. The report analysis module is used to determine the application level of the laboratory based on the application anomaly index, and to summarize the data of the laboratory to generate an application certification report.

2. The intelligent application system for fluid geochemical data according to claim 1, characterized in that, The methods for combining data sets are as follows: Query out the target objects of all the original data one by one, and summarize the original data with the same target objects to generate a basic set; The overlap of the content attributes of the original data in the base set is compared sequentially, and the original data with the same overlap level are aggregated to generate a secondary set. Mark the recording time of the original data in the secondary set one by one. According to the chronological order, summarize the original data in the same time period to generate the top-level set. Then, arrange the original data in the base set, secondary set and top-level set respectively to generate A data sets.

3. The intelligent application system for fluid geochemical data according to claim 2, characterized in that, The method for identifying benchmark data is as follows: A1: Import A sets of data into a three-dimensional space, and mark the spatial coordinates of the spatial points where the original data is located, based on the standard that one original data corresponds to one spatial point. A2: Randomly select a spatial point as the standard point, calculate the Euclidean distance between the standard point and the remaining spatial points one by one using the Euclidean distance calculation formula, and then sum all the Euclidean distance values ​​and calculate the average to obtain the effective distance value; A3: Traverse all spatial points, repeat A2, and calculate B effective distance values; A4: Calculate the overlap difference by taking the absolute value of the difference between the overlap of the original data at the standard point and the overlap of the remaining original data. Then, calculate the standard deviation of the overlap difference using the standard deviation calculation formula. A5: Iterate through all the original data and repeat A4 to obtain B standard deviations; A6: After assigning corresponding scaling factors to the effective distance value and standard deviation respectively, perform a weighted comparison to calculate the benchmark index, and record the original data corresponding to the maximum value of the benchmark index as the benchmark data.

4. The intelligent application system for fluid geochemical data according to claim 3, characterized in that, The node association criteria are: nodes within the same data set are associated point by point, while nodes in different data sets are associated centrally. The method for constructing the traceability database is as follows: Establish a blank database with A library units, import A data sets one by one into the A library units to generate A related libraries, and calculate the length of C nodes by multiplying the baseline index of C original data in the data sets by the calibrated unit length. Using the baseline data as the central node, C non-adjacent end nodes are marked in a ring shape. The tracing distance from the end nodes to the central node is continuously adjusted until the C tracing distances correspond one-to-one with the C node lengths. The original data is then bound to the central node and the end nodes one-to-one, causing the associated database to be converted into the base database. Within the same basic database, a two-way data channel is established between the central node as the starting point of the traceability and the end node as the ending point of the traceability. A bidirectional data transmission channel is established between two central nodes within two different base libraries; Using the central node as a reference, pointing arrows are configured on the data channel to point to the end nodes, simulating a tree-like traceability link. All traceability links are then numbered in ascending order to construct a traceability database.

5. The intelligent application system for fluid geochemical data according to claim 4, characterized in that, The method for constructing the evaluation scenario is as follows: A scene outline with closed boundaries is constructed using virtual simulation software, and A non-adjacent evaluation units are established within the scene outline. Using the preset boundary length as a standard, C protocol points are marked at equal intervals on the closed boundary, and evaluation permissions and evaluation parameters are configured on each of the C protocol points to generate a protocol fence. Using the same data capacity as a standard, dividers are set within the evaluation unit to divide the evaluation unit into evaluation positions and standard positions. D bidirectional interactive channels are established on the dividers to upgrade the scene outline into an evaluation scene.

6. The intelligent application system for fluid geochemical data according to claim 5, characterized in that, The evaluation mechanism is as follows: all interaction channels during a single evaluation maintain time synchronization. The method for determining whether there is abnormal data is as follows: The data fusion algorithm dynamically integrates the traceability database with the evaluation scenario, and imports A basic databases one by one into the evaluation positions of A evaluation units; Based on the original data of the evaluation position, the corresponding certification data is indexed from the database, the certification data is imported into the standard position of the evaluation unit one by one, and the original data is bound to the corresponding certification data one by one to obtain E evaluation samples. Based on the standard of one-to-one correspondence between interactive channels and evaluation samples, the evaluation samples are embedded in the interactive channels, and anomaly analysis instructions are sent synchronously to all interactive channels to analyze the authentication status of the original data. When the authentication status is non-compliant, it is determined that there is abnormal data. When the authentication status is compliant, it is determined that there is no abnormal data.

7. The intelligent application system for fluid geochemical data according to claim 6, characterized in that, The method for tracing and propagating abnormal paths is as follows: B1: Record the node where the abnormal data is located in the traceability database as the starting point of the path, and compare the data characteristics of the remaining nodes with the original data at the starting point of the path for consistency. B2: When the remaining node has the same recording time, content attributes and target object as the path start point, the node is recorded as a waypoint, and the tracing link is used as the path trajectory. The tracing start point is traced and spread to the waypoint according to the pointing arrow. B3: Repeat steps B1-B2 to obtain F waypoints, and then connect all the waypoints in sequence to generate an abnormal path.

8. The intelligent application system for fluid geochemical data according to claim 7, characterized in that, The calculation method for the application anomaly index is as follows: In the traceability database, the number of transit points and the number of nodes are counted separately, and the percentage of abnormal points is calculated by dividing the number of transit points by the number of nodes. Record the path points that have two or three consistent records in the time, content attributes, and target objects as first-level points and second-level points, respectively. Divide the number of first-level points and the number of second-level points by the number of path points, add them together, and calculate the average to obtain the hierarchical association ratio. The application anomaly index is calculated by weighting and summing the corresponding proportional coefficients assigned to the proportion of abnormal points and the proportion of hierarchical associations.

9. The intelligent application system for fluid geochemical data according to claim 8, characterized in that, Application levels are categorized into low-utility, medium-utility, and high-utility levels: The method for determining the application level is as follows: The laboratory application abnormality index Each is compared with the preset first-level index Second-level index Comparison, Greater than ; when Greater than or equal to At that time, the application level was determined to be inefficient. when Less than ,and Greater than or equal to At that time, the application level was determined to be the medium utility level; when Less than At that time, the application level was determined to be the high-efficiency application level.

10. The intelligent application system for fluid geochemical data according to claim 9, characterized in that, When generating the application certification report, a blank basic electronic report is created. Seven link units are set up in the basic report. The laboratory's original data, dataset, traceability database, abnormal data, abnormal paths, abnormal indices, and application level are compressed and then linked to obtain seven data links. The seven data links are then imported into the seven link units one by one, and the unit explanation is noted on the link unit to generate the application certification report.

Images