Laboratory data quality inspection and traceability analysis system and method based on data fusion

Abstract

This invention discloses a laboratory data quality inspection traceability analysis system and method based on data fusion, belonging to the field of laboratory data quality inspection traceability analysis technology. The invention includes: S10: obtaining abnormal quality inspection traceability data chains matching each historical quality inspection traceability anomaly event, and calculating the trust index between abnormal parameters in the abnormal quality inspection traceability data chains; S20: analyzing the trust pain points in each abnormal quality inspection traceability data chain; S30: finding the quality inspection traceability anomaly interception points in each historical abnormal quality inspection traceability anomaly event; S40: determining the quality inspection traceability path for each historical abnormal quality inspection traceability anomaly event. This invention, by finding the trust pain points and quality inspection traceability anomaly interception points in quality inspection traceability anomaly events, gradually narrows the scope of laboratory data quality inspection traceability anomalies and improves the efficiency of laboratory data quality inspection traceability.

Description

Technical Field

[0001] This invention relates to the field of laboratory data quality inspection traceability analysis technology, specifically to a laboratory data quality inspection traceability analysis system and method based on data fusion. Background Technology

[0002] Laboratory data quality inspection traceability analysis refers to the systematic recording and correlation of data from key stages during laboratory testing or quality inspection to achieve complete tracking and accountability of the entire lifecycle of samples and the history of changes in testing data, thereby ensuring the authenticity, integrity, and compliance of the data.

[0003] In existing laboratory data quality inspection traceability analysis methods, the traceability of abnormal quality inspection data only stays on the surface. For multi-dimensional laboratory data in the same process, it is impossible to quickly identify the laboratory data that caused the abnormal experimental results. At the same time, it is impossible to identify deeper quality problems based on the trust relationship between multi-dimensional laboratory data. Therefore, existing technologies are difficult to meet the requirements of high precision and strong compliance of modern laboratories. Summary of the Invention

[0004] The purpose of this invention is to provide a laboratory data quality inspection traceability analysis system and method based on data fusion, so as to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a laboratory data quality inspection traceability analysis method based on data fusion, the method comprising: S10: Filter out historical quality inspection traceability anomalies in the target laboratory, mark the abnormal laboratory environment parameters and abnormal laboratory equipment operation status parameters in each historical quality inspection traceability anomaly, obtain the abnormal quality inspection traceability data chain that matches each historical quality inspection traceability anomaly, and calculate the trust index between abnormal parameters in the abnormal quality inspection traceability data chain. S20: Using the trust index and positional relationship between abnormal parameters in each abnormal quality inspection traceability data chain as judgment conditions, analyze the trust pain points in each abnormal quality inspection traceability data chain. S30: Based on the positional relationship between the trust pain points in each abnormal quality inspection traceability data chain and the target abnormal parameters in the corresponding abnormal quality inspection traceability data chain, determine the associated abnormal parameters in each abnormal quality inspection traceability data chain, and find the quality inspection traceability interception points in each historical abnormal quality inspection traceability event based on the correlation between the associated abnormal parameters and the trust pain points. S40: Based on the quality inspection traceability interception points in each historical abnormal quality inspection traceability event, determine the quality inspection tracing path for each historical abnormal quality inspection traceability event.

[0006] Furthermore, S10 includes: S101: Filter out historical quality inspection traceability anomalies in the target laboratory. The quality inspection traceability anomalies are used to indicate events in which laboratory data does not meet the standards and the specific steps, laboratory equipment, or laboratory personnel cannot be located through quality inspection traceability. S102: Identify historical quality inspection traceability anomalies Corresponding experimental time period Mark the experimental time period Abnormal laboratory environmental parameters and abnormal laboratory equipment operating status parameters detected internally can be used to trace historical quality inspection anomalies. Corresponding abnormal parameter set The abnormal parameters are used to indicate laboratory environmental parameters and laboratory equipment operating status parameters that do not conform to the standard parameters. The standard parameters refer to the key operating conditions, measurement indicators or performance requirements specified in industry standards during the experiment. These parameters ensure the repeatability, accuracy and compliance of the experiment. S103: Based on the abnormal parameter set The abnormal parameters marked in the middle are sorted according to their acquisition time sequence to obtain the abnormal quality inspection traceability data chain. Calculate the abnormal quality inspection traceability data chain The trust index between other abnormal parameters and the target abnormal parameter is used to represent the ratio between the probability that the target abnormal parameter also occurs given that other abnormal parameters occur, determined based on the historical quality inspection traceability normal event set corresponding to the target abnormal parameter, and the probability that the target abnormal parameter occurs. Other abnormal parameters are used to represent abnormal parameters in the abnormal quality inspection traceability data chain other than the target abnormal parameter.

[0007] By calculating the trust index between abnormal parameters in the abnormal quality inspection traceability data chain, abnormal parameters with low trust indices are screened out. By analyzing the positional relationship of the screened abnormal parameters, it is helpful to quickly pinpoint the scope of abnormalities in laboratory data quality inspection traceability.

[0008] Furthermore, S20 includes: S201: In the abnormal quality inspection traceability data chain, determine the value less than the threshold. The trust index will be less than the threshold. All parameters included in the abnormal parameters corresponding to the trust index are marked as traceability abnormal data in the abnormal quality inspection traceability data chain. Describing a constant and ; S202: In the abnormal quality inspection traceability data chain, obtain the positional relationship of each traceable abnormal data relative to the target abnormal parameter. The positional relationship is used to represent the absolute value of the number difference between the traceable abnormal data and the target abnormal parameter in the abnormal quality inspection traceability data chain. Several traceable abnormal data with the same positional relationship are regarded as trust pain points in the abnormal quality inspection traceability data chain.

[0009] Furthermore, S30 includes: S301: Based on the number value corresponding to the trust pain point in the abnormal quality inspection traceability data chain, determine the associated abnormal parameters in the abnormal quality inspection traceability data chain. The associated abnormal parameters are used to represent the abnormal parameters when the absolute value of the difference between the number value corresponding to the abnormal parameter and the number value corresponding to the trust pain point is 1. S302: Construct a network graph about the traceability data chain of abnormal quality inspection by using the associated abnormal parameters as the end nodes of the network graph and the trust pain points as the middle nodes of the network graph. In particular, if the associated abnormal parameters are also trust pain points, then the associated abnormal parameters are used as the middle nodes of the network graph. The front-end nodes, middle nodes and back-end nodes in the network graph are ordered from left to right, reflecting the time-driven logic of laboratory experimental activities. The trust index between any two adjacent nodes in the network graph is used as the edge weight between the corresponding adjacent nodes. S303: Delete line segments with edge weights equal to 0 in the network graph, obtain all paths in the network graph after deletion, the path is used to represent the route from the end node of the network graph, along the arrow direction, through a series of intermediate nodes and edges, and finally to the end node, and sum the edge weights of each path. Let the intermediate node corresponding to the maximum sum of edge weights be the target intermediate node. The front node corresponding to the maximum sum of edge weights is the target node. Determine the intermediate nodes of the target Are the corresponding parameters the same as those of the target endpoint? The corresponding parameters come from the same monitored object, and if the historical abnormal quality inspection traceability of the abnormal event failed to trace back to the target intermediate node. The corresponding parameters will then target intermediate nodes. The corresponding parameters serve as the interception point for quality inspection tracing in historical abnormal quality inspection events. If the historical abnormal quality inspection tracing event can be traced back to the target intermediate node... The corresponding parameters are analyzed according to the analysis method of the maximum value of the sum of edge weights to determine the interception point of quality inspection traceability in the historical abnormal quality inspection traceability abnormal event. The monitored objects include equipment in the laboratory, experimental steps of various experimental projects, experimental personnel, etc. If not, then trace the historical abnormal quality inspection back to the intermediate node that the abnormal event could not be traced. The corresponding parameters serve as the interception points for quality inspection traceability in historical abnormal quality inspection traceability events.

[0010] By analyzing the associated abnormal parameters of each trust pain point in the abnormal quality inspection traceability data chain, a network graph of the abnormal quality inspection traceability data chain is constructed. The edge weights in the network graph reflect the trust between two nodes, which is helpful to intuitively obtain the source abnormal parameters that appear during the experiment. At the same time, by summing the edge weights corresponding to each path in the network graph, the quality inspection traceability interception points in the abnormal quality inspection traceability events can be accurately analyzed, thereby obtaining an effective quality inspection traceability path.

[0011] Furthermore, S40 includes: combining the complete path containing the quality inspection traceability interception point in the network graph of the abnormal quality inspection traceability data chain with the main data chain in the abnormal quality inspection traceability data chain to obtain the quality inspection traceability path of the historical abnormal quality inspection traceability event, wherein the main data chain is used to represent the link without branch structure.

[0012] A laboratory data quality inspection traceability and analysis system based on data fusion, the system includes an abnormal quality inspection traceability data chain analysis module, a trust pain point analysis module, a quality inspection traceability abnormal interception point finding module, and a quality inspection traceability path determination module; The abnormal quality inspection traceability data chain analysis module is used to analyze the abnormal quality inspection traceability data chain that matches each historical quality inspection traceability abnormal event, and to calculate the trust index between abnormal parameters in the abnormal quality inspection traceability data chain. The trust pain point analysis module is used to analyze the trust pain points in each abnormal quality inspection traceability data chain by using the trust index and positional relationship between abnormal parameters in each abnormal quality inspection traceability data chain as judgment conditions. The quality inspection traceability anomaly interception point finding module is used to find the quality inspection traceability anomaly interception points in each historical abnormal quality inspection traceability anomaly event. The quality inspection traceability path determination module is used to determine the quality inspection traceability path for each historical abnormal quality inspection traceability event.

[0013] Furthermore, the abnormal quality inspection traceability data chain analysis module includes a screening unit, an abnormal parameter set acquisition unit, an abnormal quality inspection traceability data chain analysis unit, and a trust index calculation unit. The screening unit filters out historical quality inspection traceability anomalies existing in the target laboratory; The abnormal parameter set acquisition unit marks the abnormal laboratory environmental parameters and abnormal laboratory equipment operating status parameters monitored during each experimental time period to obtain the abnormal parameter set corresponding to each historical quality inspection traceability abnormal event; The abnormal quality inspection traceability data chain analysis unit sorts the abnormal parameters marked in the abnormal parameter set according to the order of collection time to obtain the abnormal quality inspection traceability data chain. The trust index calculation unit calculates the trust index between other abnormal parameters and the target abnormal parameter in the abnormal quality inspection traceability data chain.

[0014] Furthermore, the trust pain point analysis module includes a traceability abnormal data marking unit and a trust pain point analysis unit; The traceability anomaly marking unit marks traceability anomaly data in the abnormal quality inspection traceability data chain according to the relationship between the trust index and the threshold between the abnormal parameters in the abnormal quality inspection traceability data chain. The trust pain point analysis unit analyzes the trust pain points in the abnormal quality inspection traceability data chain based on the positional relationship between each traceability abnormal data in the abnormal quality inspection traceability data chain and the target abnormal parameter in the abnormal quality inspection traceability data chain.

[0015] Furthermore, the quality inspection traceability anomaly interception point finding module includes an associated anomaly parameter determination unit, a network graph construction unit, a path acquisition unit, and a quality inspection traceability anomaly interception point finding unit; The associated anomaly parameter determination unit determines the associated anomaly parameters in the anomaly quality inspection traceability data chain based on the number value corresponding to the trust pain point in the anomaly quality inspection traceability data chain. The network graph construction unit uses associated abnormal parameters as end nodes of the network graph and trust pain points as intermediate nodes of the network graph to construct a network graph about the abnormal quality inspection traceability data chain. The path acquisition unit performs relevant processing on the network graph based on the edge weight values ​​and acquires all paths in the network graph. The quality inspection traceability anomaly interception point finding unit analyzes the target intermediate node in the network graph based on the sum of the edge weights corresponding to each path, and finds the quality inspection traceability anomaly interception point in each historical abnormal quality inspection traceability event based on the monitored object situation corresponding to the parameters of each target intermediate node.

[0016] Furthermore, the quality inspection traceability path determination module combines the complete path containing the quality inspection traceability anomaly interception point in the network graph of the abnormal quality inspection traceability data chain with the main data chain in the abnormal quality inspection traceability data chain to obtain the quality inspection traceability path of the historical abnormal quality inspection traceability event.

[0017] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention calculates the trust index between abnormal parameters in the abnormal quality inspection traceability data chain, filters out abnormal parameters with low trust indices, and analyzes the positional relationships of the filtered abnormal parameters to lock down multi-dimensional abnormal parameters corresponding to the same link. Through multi-dimensional abnormal parameters and their associated abnormal parameters, a network graph of the abnormal quality inspection traceability data chain is constructed. This can quickly lock down the scope of laboratory data quality inspection traceability anomalies and accurately identify the quality inspection traceability interception points in each historical abnormal quality inspection traceability event based on the trust status between adjacent nodes in the network graph. In other words, it can identify deep-seated quality problems in the experimental process, which is conducive to determining the accurate traceability path of experimental anomalies.

[0018] 2. This invention determines the trust status between multi-dimensional abnormal parameters by using a network diagram of the abnormal quality inspection traceability data chain. The trust status avoids the situation where the traceability of abnormal quality inspection data only stays on the surface, thus improving the traceability accuracy of laboratory quality inspection.

[0019] 3. By identifying trust pain points and interception points in quality inspection traceability anomalies, this invention gradually narrows the scope of laboratory data quality inspection traceability anomalies and improves the efficiency of laboratory data quality inspection traceability. Attached Figure Description

[0020] Figure 1 This is a schematic diagram illustrating the workflow of the laboratory data quality inspection traceability analysis method based on data fusion according to the present invention. Detailed Implementation

[0021] 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.

[0022] like Figure 1 As shown, this invention provides a technical solution for a laboratory data quality inspection traceability analysis method based on data fusion. The method includes: S10: Filter out historical quality inspection traceability anomalies in the target laboratory, mark the abnormal laboratory environment parameters and abnormal laboratory equipment operation status parameters in each historical quality inspection traceability anomaly, obtain the abnormal quality inspection traceability data chain that matches each historical quality inspection traceability anomaly, and calculate the trust index between abnormal parameters in the abnormal quality inspection traceability data chain. S10 includes: S101: Filter out historical quality inspection traceability anomalies in the target laboratory. Quality inspection traceability anomalies are used to indicate events in which laboratory data does not meet the standards and the specific steps, laboratory equipment, or laboratory personnel could not be located through quality inspection traceability. S102: Identify historical quality inspection traceability anomalies Corresponding experimental time period Mark the experimental time period Abnormal laboratory environmental parameters and abnormal laboratory equipment operating status parameters detected internally can be used to trace historical quality inspection anomalies. Corresponding abnormal parameter set Abnormal parameters are used to indicate laboratory environmental parameters and laboratory equipment operating status parameters that do not conform to standard parameters. Standard parameters refer to the key operating conditions, measurement indicators or performance requirements specified in industry standards during the experiment. These parameters ensure the repeatability, accuracy and compliance of the experiment. S103: Based on the abnormal parameter set The abnormal parameters marked in the middle are sorted according to their acquisition time sequence to obtain the abnormal quality inspection traceability data chain. Calculate the abnormal quality inspection traceability data chain The trust index between other abnormal parameters and the target abnormal parameter is used to represent the ratio between the probability that the target abnormal parameter also occurs given that other abnormal parameters occur, based on the historical quality inspection traceability normal event set corresponding to the target abnormal parameter, and the probability that the target abnormal parameter also occurs. Other abnormal parameters are used to represent abnormal parameters in the abnormal quality inspection traceability data chain other than the target abnormal parameter. S20: Using the trust index and positional relationship between abnormal parameters in each abnormal quality inspection traceability data chain as judgment conditions, analyze the trust pain points in each abnormal quality inspection traceability data chain. S20 includes: S201: In the abnormal quality inspection traceability data chain, determine the value less than the threshold. The trust index will be less than the threshold. All parameters included in the abnormal parameters corresponding to the trust index are marked as traceability abnormal data in the abnormal quality inspection traceability data chain. Describing a constant and ; S202: In the abnormal quality inspection traceability data chain, obtain the positional relationship of each traceability abnormal data relative to the target abnormal parameter. The positional relationship is used to represent the absolute value of the difference in the number of the traceability abnormal data relative to the target abnormal parameter in the abnormal quality inspection traceability data chain. Several traceability abnormal data with the same positional relationship are regarded as trust pain points in the abnormal quality inspection traceability data chain. S30: Based on the positional relationship between the trust pain points in each abnormal quality inspection traceability data chain and the target abnormal parameters in the corresponding abnormal quality inspection traceability data chain, determine the associated abnormal parameters in each abnormal quality inspection traceability data chain, and find the quality inspection traceability interception points in each historical abnormal quality inspection traceability event based on the correlation between the associated abnormal parameters and the trust pain points. S30 includes: S301: Based on the number value corresponding to the trust pain point in the abnormal quality inspection traceability data chain, determine the associated abnormal parameters in the abnormal quality inspection traceability data chain. The associated abnormal parameters are used to represent the abnormal parameters when the absolute value of the difference between the number value corresponding to the abnormal parameter and the number value corresponding to the trust pain point is 1. S302: Construct a network graph about the traceability data chain of abnormal quality inspection by using the associated abnormal parameters as the end nodes of the network graph and the trust pain points as the middle nodes of the network graph. In particular, if the associated abnormal parameters are also trust pain points, then the associated abnormal parameters are used as the middle nodes of the network graph. The front-end nodes, middle nodes and back-end nodes in the network graph are ordered from left to right, reflecting the time-driven logic of laboratory experimental activities. The trust index between any two adjacent nodes in the network graph is used as the edge weight between the corresponding adjacent nodes. S303: Delete line segments with edge weights equal to 0 in the network graph, and obtain all paths in the network graph after deletion. A path is used to represent a route starting from an end node in the network graph, passing through a series of intermediate nodes and edges along the arrow direction, and finally reaching the end node. Sum the edge weights of each path. Specifically, if there are two intermediate nodes in the constructed network graph and the acquisition time of the parameters corresponding to the two intermediate nodes is different, then the arrow direction between the two intermediate nodes is: the intermediate node corresponding to the parameter acquired first points to the intermediate node corresponding to the parameter acquired later. If there are two intermediate nodes in the constructed network graph and the acquisition time of the parameters corresponding to the two intermediate nodes is the same, then there is no arrow direction between the two intermediate nodes. Let the intermediate node corresponding to the maximum sum of edge weights be the target intermediate node. The front node corresponding to the maximum sum of edge weights is the target node. Determine the intermediate nodes of the target Are the corresponding parameters the same as those of the target endpoint? The corresponding parameters come from the same monitored object, and if the historical abnormal quality inspection traceability of the abnormal event failed to trace back to the target intermediate node. The corresponding parameters will then target intermediate nodes. The corresponding parameters serve as the interception point for quality inspection tracing in historical abnormal quality inspection events. If the historical abnormal quality inspection tracing event can be traced back to the target intermediate node... The corresponding parameters are analyzed according to the analysis method of the maximum value of the sum of edge weights to determine the interception point of quality inspection traceability in historical abnormal quality inspection traceability events. The monitored objects include equipment in the laboratory, experimental steps of various experimental projects, experimental personnel, etc. If not, then trace the historical abnormal quality inspection back to the intermediate node that the abnormal event could not be traced. The corresponding parameters serve as the interception points for quality inspection traceability in historical abnormal quality inspection traceability events; S40: Based on the quality inspection traceability interception points in each historical abnormal quality inspection traceability event, determine the quality inspection tracing path for each historical abnormal quality inspection traceability event. S40 includes: combining the complete path containing the quality inspection traceability interception point in the network graph of the abnormal quality inspection traceability data chain with the main data chain in the abnormal quality inspection traceability data chain to obtain the quality inspection traceability path of the historical abnormal quality inspection traceability event. The main data chain is used to represent the link without branch structure.

[0023] The laboratory data quality inspection traceability and analysis system based on data fusion includes an abnormal quality inspection traceability data chain analysis module, a trust pain point analysis module, a quality inspection traceability abnormal interception point finding module, and a quality inspection traceability path determination module. The abnormal quality inspection traceability data chain analysis module is used to analyze the abnormal quality inspection traceability data chains that match each historical quality inspection traceability abnormal event, and to calculate the trust index between abnormal parameters in the abnormal quality inspection traceability data chain. The abnormal quality inspection traceability data chain analysis module includes a filtering unit, an abnormal parameter set acquisition unit, an abnormal quality inspection traceability data chain analysis unit, and a trust index calculation unit; The screening unit identifies historical quality inspection traceability anomalies present in the target laboratory. The abnormal parameter set acquisition unit marks the abnormal laboratory environmental parameters and abnormal laboratory equipment operating status parameters monitored during each experimental time period to obtain the abnormal parameter set corresponding to each historical quality inspection traceability abnormal event; The abnormal quality inspection traceability data chain analysis unit sorts the abnormal parameters marked in the abnormal parameter set according to the order of collection time to obtain the abnormal quality inspection traceability data chain. The trust index calculation unit calculates the trust index between other abnormal parameters and the target abnormal parameter in the abnormal quality inspection traceability data chain; The Trust Pain Point Analysis Module is used to analyze the trust pain points in each abnormal quality inspection traceability data chain by using the trust index and positional relationship between abnormal parameters in each abnormal quality inspection traceability data chain as judgment conditions. The Trust Pain Point Analysis module includes a traceability anomaly data marking unit and a trust pain point analysis unit; The traceability anomaly marking unit marks traceability anomaly data in the anomaly quality inspection traceability data chain based on the relationship between the trust index and the threshold between the anomaly parameters in the anomaly quality inspection traceability data chain. The Trust Pain Point Analysis Unit analyzes the trust pain points in the abnormal quality inspection traceability data chain based on the positional relationship between each traceability abnormal data in the abnormal quality inspection traceability data chain and the target abnormal parameter in the abnormal quality inspection traceability data chain. The Quality Inspection Traceability Anomaly Interception Point Finding Module is used to find the quality inspection traceability anomaly interception points in each historical anomaly traceability event. The quality inspection traceability anomaly interception point finding module includes an associated anomaly parameter determination unit, a network graph construction unit, a path acquisition unit, and a quality inspection traceability anomaly interception point finding unit; The associated anomaly parameter determination unit determines the associated anomaly parameters in the anomaly quality inspection traceability data chain based on the corresponding number value of the trust pain point in the anomaly quality inspection traceability data chain. The network graph construction unit uses associated abnormal parameters as end nodes of the network graph and trust pain points as intermediate nodes to construct a network graph about the abnormal quality inspection traceability data chain. The path acquisition unit performs relevant processing on the network graph based on the edge weight values ​​and acquires all paths in the network graph. The quality inspection traceability anomaly interception point finding unit analyzes the target intermediate node in the network graph based on the sum of the edge weights corresponding to each path, and finds the quality inspection traceability anomaly interception point in each historical abnormal quality inspection traceability event based on the parameters of each target intermediate node and the monitored object. The quality inspection traceability path determination module is used to determine the quality inspection traceability path for each historical abnormal quality inspection traceability event; The quality inspection traceability path determination module combines the complete path containing the quality inspection traceability anomaly interception point in the network graph of the abnormal quality inspection traceability data chain with the main data chain in the abnormal quality inspection traceability data chain to obtain the quality inspection traceability path of the historical abnormal quality inspection traceability anomaly event.

[0024] Example 1: Setting the experimental time period The laboratory is The experiment within the time period is called a quantitative analysis experiment, and the monitoring interval for laboratory environmental parameters and laboratory equipment operating status parameters is 3 minutes. Assuming the laboratory to be retrieved is in The temperature at that moment was relative humidity The laboratory environmental parameters used in quantitative analysis experiments include temperature and relative humidity; The retrieved laboratory ultracentrifuge Rotation speed at time Centrifugal force ,temperature The operating parameters of the experimental equipment in the quantitative analysis experiment include rotational speed, centrifugal force, and temperature; Set up the laboratory in The standard temperature at that time is Standard relative humidity Laboratory ultracentrifuges Standard rotational speed at any time Standard centrifugal force , Represents gravitational acceleration, standard temperature ; Then and Stored in historical quality inspection traceability anomaly events Corresponding abnormal parameter set middle.

[0025] Example 2: Establishing an abnormal quality inspection traceability data chain tracing the abnormal quality inspection data chain Terminal abnormal parameters As a target anomaly parameter; When the target anomaly parameter is The historical quality inspection traceability normal event set is used to represent events where laboratory data does not meet standards, and the specific steps, laboratory equipment, or laboratory personnel can be located through quality inspection traceability to determine the abnormal parameters. Caused to abnormal target parameters Total number of occurrences and target anomaly parameters Total number of occurrences ,according to For target abnormal parameters With abnormal parameters Trust index between Perform the calculation.

[0026] Example 3: Establishing an abnormal quality inspection traceability data chain Among them, abnormal parameters The corresponding number is 1, abnormal parameter. The corresponding number is 2, and the target anomaly parameter is... The corresponding number is 3; Set target anomaly parameters With abnormal parameters Trust index between , ; because Then the abnormal parameters and All of these are traceability anomaly data from the abnormal quality inspection traceability data chain; Tracing abnormal data , With target anomaly parameters The positional relationships are all |2-3|=1; Therefore, tracing abnormal data and For abnormal quality inspection traceability data chain The pain point of trust in China.

[0027] Example 4: Abnormal Quality Inspection Traceability Data Chain The main data chain in is .

[0028] Example 5: Establishing an abnormal quality inspection traceability data chain Abnormal parameters The corresponding number is 1, abnormal parameter. The corresponding number is 2, and the target anomaly parameter is... The corresponding number is 3, tracing the abnormal data. and For abnormal quality inspection traceability data chain The trust pain points in the middle are: Trust Pain Points and The associated anomaly parameters are , and Associated abnormal parameters As a network Figure 1 The front-end node will associate abnormal parameters. and As a network Figure 1 The backend nodes, and As a network Figure 1 intermediate nodes, network Figure 1 In this context, the front-end node points to the middle node, and the middle node points to the back-end node; set up and If the corresponding edge weight is 0, then... and The connection between them is from the network Figure 1 Deleted, at this time, network Figure 1 It contains three paths, namely: Path 1: Path 2: Path 3: ; Let the sum of the weights of one side of the path be 4, the sum of the weights of the two sides of the path be 5, and the sum of the weights of the three sides of the path be 3. Then let's denote... intermediate node of the target , For the target node ,because and Since they come from different monitored objects, therefore... As a quality inspection traceability interception point in the historical abnormal quality inspection traceability anomaly event, among which, Source: from the monitored object's ultracentrifuge From the laboratory of the monitored object; At this time, the network Figure 1 The complete path containing the quality inspection traceability anomaly interception point is: ; Establish an abnormal quality inspection traceability data chain The main data chain in is The quality inspection tracing path for historical abnormal quality inspection events is as follows: .

[0029] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A laboratory data quality inspection traceability analysis method based on data fusion, characterized by: The method includes: S10: Filter out historical quality inspection traceability anomalies in the target laboratory, mark the abnormal laboratory environment parameters and abnormal laboratory equipment operation status parameters in each historical quality inspection traceability anomaly, obtain the abnormal quality inspection traceability data chain that matches each historical quality inspection traceability anomaly, and calculate the trust index between abnormal parameters in the abnormal quality inspection traceability data chain. S20: Using the trust index and positional relationship between abnormal parameters in each abnormal quality inspection traceability data chain as judgment conditions, analyze the trust pain points in each abnormal quality inspection traceability data chain. S30: Based on the positional relationship between the trust pain points in each abnormal quality inspection traceability data chain and the target abnormal parameters in the corresponding abnormal quality inspection traceability data chain, determine the associated abnormal parameters in each abnormal quality inspection traceability data chain, and find the quality inspection traceability interception points in each historical abnormal quality inspection traceability event based on the correlation between the associated abnormal parameters and the trust pain points. S40: Based on the quality inspection traceability interception points in each historical abnormal quality inspection traceability event, determine the quality inspection tracing path for each historical abnormal quality inspection traceability event.

2. The laboratory data quality inspection traceability analysis method based on data fusion according to claim 1, characterized in that: S10 includes: S101: Filter out historical quality inspection traceability anomalies in the target laboratory. The quality inspection traceability anomalies are used to indicate events in which laboratory data does not meet the standards and the specific steps, laboratory equipment, or laboratory personnel cannot be located through quality inspection traceability. S102: Determine the experimental time period corresponding to each historical quality inspection traceability anomaly event, mark the abnormal laboratory environment parameters and abnormal laboratory equipment operating status parameters monitored within each experimental time period, and obtain the abnormal parameter set corresponding to each historical quality inspection traceability anomaly event. The abnormal parameters are used to represent laboratory environment parameters and laboratory equipment operating status parameters that do not conform to standard parameters. S103: Sort the abnormal parameters according to the collection time sequence of each abnormal parameter marked in the abnormal parameter set to obtain the abnormal quality inspection traceability data chain. Take the abnormal parameter at the end of the abnormal quality inspection traceability data chain as the target abnormal parameter. Calculate the trust index between the other abnormal parameters in the abnormal quality inspection traceability data chain and the target abnormal parameter. The trust index is used to represent the ratio between the probability that the target abnormal parameter also occurs given that the other abnormal parameters occur, determined based on the historical quality inspection traceability normal event set corresponding to the target abnormal parameter, and the probability that the target abnormal parameter occurs.

3. The laboratory data quality inspection traceability analysis method based on data fusion according to claim 2, characterized in that: S20 includes: S201: Based on the relationship between the trust index and the threshold among the abnormal parameters in the abnormal quality inspection traceability data chain, mark the traceability abnormal data in the abnormal quality inspection traceability data chain; S202: Based on the positional relationship between each traceable abnormal data in the abnormal quality inspection traceability data chain and the target abnormal parameter in the abnormal quality inspection traceability data chain, the trust pain points in the abnormal quality inspection traceability data chain are analyzed.

4. The laboratory data quality inspection traceability analysis method based on data fusion according to claim 3, characterized in that: S30 includes: S301: Based on the number value corresponding to the trust pain point in the abnormal quality inspection traceability data chain, determine the associated abnormal parameters in the abnormal quality inspection traceability data chain. The associated abnormal parameters are used to represent the abnormal parameters when the absolute value of the difference between the number value corresponding to the abnormal parameter and the number value corresponding to the trust pain point is 1. S302: Construct a network graph about the traceability data chain of abnormal quality inspection by using the associated abnormal parameters as the end nodes of the network graph and the trust pain points as the middle nodes of the network graph. In particular, if the associated abnormal parameters are also trust pain points, then use the associated abnormal parameters as the middle nodes of the network graph and use the trust index between any two adjacent nodes in the network graph as the edge weight between the corresponding adjacent nodes. S303: Based on the edge weight values, perform relevant processing on the network graph and obtain all paths in the network graph. Analyze the target intermediate nodes in the network graph based on the sum of the edge weights corresponding to each path. Based on the monitored objects to which the parameters of each target intermediate node belong, find the quality inspection traceability interception points in each historical abnormal quality inspection traceability event.

5. The laboratory data quality inspection traceability analysis method based on data fusion according to claim 4, characterized in that: S40 includes: combining the complete path containing the quality inspection traceability interception point in the network graph of the abnormal quality inspection traceability data chain with the main data chain in the abnormal quality inspection traceability data chain to obtain the quality inspection traceability path of the historical abnormal quality inspection traceability event, wherein the main data chain is used to represent the link without branch structure.

6. A data fusion-based laboratory data quality inspection traceability analysis system applied to the data fusion-based laboratory data quality inspection traceability analysis method according to any one of claims 1-5, characterized in that: The system includes an abnormal quality inspection traceability data chain analysis module, a trust pain point analysis module, a quality inspection traceability abnormal interception point finding module, and a quality inspection traceability path determination module. The abnormal quality inspection traceability data chain analysis module is used to analyze the abnormal quality inspection traceability data chain that matches each historical quality inspection traceability abnormal event, and to calculate the trust index between abnormal parameters in the abnormal quality inspection traceability data chain. The trust pain point analysis module is used to analyze the trust pain points in each abnormal quality inspection traceability data chain by using the trust index and positional relationship between abnormal parameters in each abnormal quality inspection traceability data chain as judgment conditions. The quality inspection traceability anomaly interception point finding module is used to find the quality inspection traceability anomaly interception points in each historical abnormal quality inspection traceability anomaly event. The quality inspection traceability path determination module is used to determine the quality inspection traceability path for each historical abnormal quality inspection traceability event.

7. The laboratory data quality inspection traceability analysis system based on data fusion according to claim 6, characterized in that: The abnormal quality inspection traceability data chain analysis module includes a screening unit, an abnormal parameter set acquisition unit, an abnormal quality inspection traceability data chain analysis unit, and a trust index calculation unit. The screening unit filters out historical quality inspection traceability anomalies existing in the target laboratory; The abnormal parameter set acquisition unit marks the abnormal laboratory environmental parameters and abnormal laboratory equipment operating status parameters monitored during each experimental time period to obtain the abnormal parameter set corresponding to each historical quality inspection traceability abnormal event; The abnormal quality inspection traceability data chain analysis unit sorts the abnormal parameters marked in the abnormal parameter set according to the order of collection time to obtain the abnormal quality inspection traceability data chain. The trust index calculation unit calculates the trust index between other abnormal parameters and the target abnormal parameter in the abnormal quality inspection traceability data chain.

8. The laboratory data quality inspection traceability analysis system based on data fusion according to claim 7, characterized in that: The trust pain point analysis module includes an abnormal data tracking and marking unit and a trust pain point analysis unit. The traceability anomaly marking unit marks traceability anomaly data in the abnormal quality inspection traceability data chain according to the relationship between the trust index and the threshold between the abnormal parameters in the abnormal quality inspection traceability data chain. The trust pain point analysis unit analyzes the trust pain points in the abnormal quality inspection traceability data chain based on the positional relationship between each traceability abnormal data in the abnormal quality inspection traceability data chain and the target abnormal parameter in the abnormal quality inspection traceability data chain.

9. The laboratory data quality inspection traceability analysis system based on data fusion according to claim 8, characterized in that: The quality inspection traceability anomaly interception point finding module includes an associated anomaly parameter determination unit, a network graph construction unit, a path acquisition unit, and a quality inspection traceability anomaly interception point finding unit. The associated anomaly parameter determination unit determines the associated anomaly parameters in the anomaly quality inspection traceability data chain based on the number value corresponding to the trust pain point in the anomaly quality inspection traceability data chain. The network graph construction unit uses associated abnormal parameters as end nodes of the network graph and trust pain points as intermediate nodes of the network graph to construct a network graph about the abnormal quality inspection traceability data chain. The path acquisition unit performs relevant processing on the network graph based on the edge weight values ​​and acquires all paths in the network graph. The quality inspection traceability anomaly interception point finding unit analyzes the target intermediate node in the network graph based on the sum of the edge weights corresponding to each path, and finds the quality inspection traceability anomaly interception point in each historical abnormal quality inspection traceability event based on the monitored object situation corresponding to the parameters of each target intermediate node.

10. The laboratory data quality inspection traceability analysis system based on data fusion according to claim 9, characterized in that: The quality inspection traceability path determination module combines the complete path containing the quality inspection traceability anomaly interception point in the network graph of the abnormal quality inspection traceability data chain with the main data chain in the abnormal quality inspection traceability data chain to obtain the quality inspection traceability path of the historical abnormal quality inspection traceability event.

Images