CN-121981673-A - Automatic acquisition and intelligent checking system and method for coal laboratory data

CN121981673ACN 121981673 ACN121981673 ACN 121981673ACN-121981673-A

Abstract

The invention provides an automatic acquisition and intelligent checking system and method for coal laboratory data, which relate to the technical field of coal assay, and comprise a data layer, a processing layer, a checking layer and an application layer, wherein the data layer is used for automatically acquiring experimental data in real time and pushing the experimental data to a server through a data acquisition program deployed on a computer connected with heterogeneous experimental equipment; the invention realizes the full-flow automation from data acquisition, synthesis, calculation to checking and distribution, shortens the data processing period to the minute level in the conventional mode by hours or even days, greatly shortens the report presentation time, cancels the manual handover and input of a plurality of links, and remarkably improves the overall working efficiency and service response speed of a laboratory.

Inventors

LI WENXUE
ZHANG KAI
YU HUIMIN
ZHANG GUANGLI

Assignees

山东宏桥新型材料有限公司

Dates

Publication Date: 20260505
Application Date: 20260109

Claims (10)

1. The system is characterized in that the data layer is used for automatically acquiring experimental data in real time and pushing the experimental data to a server through a data acquisition program deployed on a computer connected with heterogeneous experimental equipment; the processing layer is used for monitoring the integrity of the sample data and automatically triggering the synthesis and calculation of the data when the data are complete; The checking layer is used for carrying out rationality checking on the calculated data based on an intelligent algorithm and carrying out real-time early warning on unreasonable data, and the application layer is used for displaying operation information through a visual billboard and realizing data integration with an external service platform.
2. The automatic acquisition and intelligent checking system for the coal laboratory data, as set forth in claim 1, is characterized in that the data acquisition program in the data layer is a lightweight data crawler program matched with specific experimental equipment, unified access and management of heterogeneous experimental equipment are achieved by configuring different data crawlers, the lightweight data crawler program supports a custom data field mapping rule, and native fields and system standard fields of data output by the experimental equipment are associated through JSON configuration files.
3. The automatic acquisition and intelligent checking system for the coal laboratory data, which is disclosed by claim 1, is characterized in that an intelligent algorithm in the checking layer comprises a history comparison method and a regression analysis method, wherein the history comparison method is used for comparing current data with a history data interval of the same type of sample, the regression analysis method is used for deducing an expected value of an index by using a regression equation established based on the history data to check the rationality of the data, the history data interval of the history comparison method is an effective data set of the same coal type under the same experimental condition for about 36 months, the data confidence is more than or equal to 95%, and the fitting goodness R2 of the regression analysis method is more than or equal to 0.92.
4. The automatic acquisition and intelligent checking system for coal laboratory data according to claim 3, wherein a regression equation of the regression analysis method is established based on more than 12 ten thousand historical data samples, and a proprietary equation is set for different coal types, wherein the regression equation adopts a multiple linear regression model, and the specific expression is as follows: Y=a 0 +Σ（i=1ton）a i X i +ε, Wherein Y is an expected value of an index to be checked, a 0 is a regression constant term obtained by fitting historical data, i is a source index number, a positive integer, a value range of 1≤i≤n, n is a source index number, a positive integer is set according to coal characteristics, a value range of 3≤n≤ 8;a i is a regression coefficient of an ith source index, obtained by fitting historical data and reflecting the influence weight of the source index on Y, X i is an actual measurement value of the ith source index, epsilon is a random error term and is caused by experimental environment fluctuation and measurement precision factors, and the value of epsilon is less than or equal to 0.05 XY is satisfied.
5. The coal laboratory data automatic acquisition and intelligent checking system according to claim 1 is characterized in that the real-time early warning mode of the checking layer comprises on-site voice broadcasting and abnormal information pushing to a terminal of an laboratory staff, the on-site voice broadcasting adopts a TTS voice synthesis technology, broadcasting frequency is 1 time every 30 seconds until abnormal data processing is completed, and the terminal pushing information comprises abnormal data numbers, corresponding equipment numbers, checking algorithm types and deviation amplitude, and one-key jumping to a data detail page is supported.
6. The automatic acquisition and intelligent checking system for the coal laboratory data, which is disclosed in claim 1, is characterized in that the information displayed by the visual display board of the application layer comprises sample quantity, completion quantity, re-detection quantity, experimental quantity of each device, unreasonable data device distribution and unreasonable data numbering, the visual display board supports data statistics dimension switching according to day/week/month, unreasonable data device distribution is displayed by adopting a thermodynamic diagram, the data updating frequency is less than or equal to 5 seconds, and the export of abnormal data is supported.
7. The automatic acquisition and intelligent checking system for the coal laboratory data, which is disclosed in claim 1, is characterized in that the external service platform integrated by the application layer comprises a coal yard command vehicle platform and a primary numbering workbench, and is used for realizing automatic receiving of sample numbers and automatic pushing of test results, wherein the data integration is realized by adopting a RESTfulAPI interface, the response time of the interface is less than or equal to 300ms, the data transmission is encrypted by adopting AES-256, and the bidirectional data synchronization with the external service platform is supported, and the synchronization delay is less than or equal to 1 minute.
8. The automatic acquisition and intelligent checking method for the coal laboratory data is applied to the automatic acquisition and intelligent checking system for the coal laboratory data, which is characterized by comprising the following steps of: s1, automatically acquiring experimental data in real time through a data acquisition program deployed on a computer connected with heterogeneous experimental equipment and pushing the experimental data to a server; s2, monitoring the integrity of sample data, and automatically triggering the synthesis and calculation of the data when the data are complete; s3, performing rationality check on the calculated data by adopting an intelligent algorithm, and performing real-time early warning on unreasonable data; and S4, displaying the operation information through a visual billboard, and integrating and cooperating the processing result with an external service platform.
9. The method for automatically acquiring and intelligently checking the data of the coal laboratory according to claim 8, wherein the step of performing rationality checking by adopting an intelligent algorithm comprises the following steps: Checking based on a history comparison method, namely comparing index data of a current sample with a history data interval of the same type of sample, and identifying an abnormal value which is obviously deviated; And/or checking based on a regression analysis method, namely calculating expected values of the indexes to be checked according to a plurality of source indexes by using a regression equation established based on historical data, and checking the rationality of the data by comparing the actual values with the expected values; when the data is checked based on a regression analysis method, when the relative error delta= | (actual value-Y)/actual value|multipliedby 100% >5 between an actual value and an expected value, the data is judged to be unreasonable data.
10. The method for automatically collecting and intelligently checking the data in the coal laboratory according to claim 8, further comprising the step of realizing unified data collection access to heterogeneous experimental equipment of different manufacturers and models by configuring a lightweight data crawler program matched with specific experimental equipment, wherein the lightweight data crawler program supports two deployment modes, namely a local process mode and a cloud agent mode, supports offline data caching of equipment, and automatically synchronizes data caching after the equipment is re-networked.

Description

Automatic acquisition and intelligent checking system and method for coal laboratory data Technical Field The invention relates to the technical field of coal assay, in particular to a system and a method for automatically acquiring and intelligently checking data in a coal laboratory. Background Currently, coal laboratories are commonly operated in a "data integration" mode. In the mode, experimental data are stored in scattered and heterogeneous experimental equipment or computers connected with the experimental equipment, and the data acquisition is highly dependent on manual operation, such as manual setting of shared folders, copying and pasting of data one by one or complex software docking; The prior art scheme has the obvious defects that firstly, all links from data acquisition, synthesis calculation to checking distribution in the whole process need manual intervention, and the steps are complex, low in efficiency and extremely prone to error. Secondly, the verification of the rationality of the data is seriously lagged and highly depends on the personal experience of an inspector, objective and intelligent automatic verification means are lacked, and the data quality is difficult to guarantee. Moreover, because laboratory equipment is usually purchased from different manufacturers, models and batches, the existing solution often needs to be deeply customized and developed for each equipment, so that the system has high coupling degree, poor expansibility and high maintenance cost, and is difficult to adapt to the updating and expansion of the equipment, therefore, the invention provides an automatic acquisition and intelligent checking system and method for coal laboratory data, and solves the problems in the prior art. Disclosure of Invention Aiming at the problems, the invention provides the system and the method for automatically acquiring and intelligently checking the data in the coal laboratory, which realize the full-flow automation from data acquisition, synthesis, calculation to checking and distribution, shorten the data processing period in the conventional mode from hour to day to minute, greatly shorten the report issuing time, cancel the manual handover and input of a plurality of links and obviously improve the overall working efficiency and service response speed of the laboratory. The system comprises a data layer, a processing layer, a checking layer and an application layer, wherein the data layer is used for automatically acquiring experimental data in real time and pushing the experimental data to a server through a data acquisition program deployed on a computer connected with heterogeneous experimental equipment; The checking layer is used for carrying out rationality checking on the calculated data based on an intelligent algorithm and carrying out real-time early warning on unreasonable data, and the application layer is used for displaying operation information through a visual billboard and realizing data integration with an external service platform. The method is further improved in that the data acquisition program in the data layer is a lightweight data crawler program matched with specific experimental equipment, unified access and management of heterogeneous experimental equipment are achieved through configuration of different data crawlers, the lightweight data crawler program supports a custom data field mapping rule, and the native fields and the system standard fields of the output data of the experimental equipment are associated through a JSON configuration file. The intelligent algorithm in the checking layer comprises a history comparison method and a regression analysis method, wherein the history comparison method is used for comparing current data with a history data interval of a sample of the same type, the regression analysis method is used for deducing an expected value of an index by using a regression equation established based on the history data so as to check data rationality, the history data interval of the history comparison method is an effective data set of the same coal type under the same experimental condition for about 36 months, the data confidence is more than or equal to 95%, and the goodness of fit R2 of the regression analysis method is more than or equal to 0.92. The regression equation of the regression analysis method is established based on more than 12 ten thousand historical data samples, and a proprietary equation is set for different coal types, wherein the regression equation adopts a multiple linear regression model, and the specific expression is as follows: Y=a0+Σ（i=1ton）aiXi+ε, Wherein Y is an expected value of an index to be checked, a 0 is a regression constant term obtained by fitting historical data, i is a source index number, a positive integer, a value range of 1≤i≤n, n is a source index number, a positive integer is set according to coal characteristics, a value range of 3≤n≤ 8;a i is a regression coefficient