Industrial metering instrument data self-correction calculation method

By collecting and analyzing the data of industrial metrology instruments, and using historical data and regression models to automatically correct abnormal data, the problems of low instrument fault detection efficiency and inability to automatically correct, and the stability and processing efficiency of data are improved.

CN119022980APending Publication Date: 2024-11-26JIANGSU JINHENG INFORMATION TECH CO LTD
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411196994.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Industrial metrology instruments are prone to failure during use, resulting in unstable data and cannot be corrected quickly and automatically. The manual inspection method has problems such as large workload, long fault detection cycle and inconsistent correction standards.

Method used

Provide a method for self-correcting calculation of industrial metrology data. By collecting data from target metrology, acquisition equipment and network equipment, we can judge whether the acquisition link, acquisition equipment and instruments are abnormal, and use historical data to perform arithmetic average calculation and regression model analysis. Automatically supplement and correct the cumulative data of abnormal periods to generate data self-correction results.

Benefits of technology

It realizes rapid identification and automatic correction of instrument data abnormalities, improves fault detection efficiency, reduces the workload and correction time of manual inspection, and ensures the stability and accuracy of data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119022980A_ABST
    Figure CN119022980A_ABST
Patent Text Reader

Abstract

The invention provides an industrial metering device data self-correction calculation method, and relates to the technical field of industrial metering devices, and the method comprises the steps: collecting the data of a target metering device, a collection device, and a network device; judging whether the acquisition link, the acquisition equipment, the instrument and the instrument data are abnormal or not according to the data of the target metering instrument, the acquisition equipment and the network equipment; if the acquisition link, the acquisition device and the instrument are normal and the instrument data are abnormal, performing arithmetic average calculation according to the historical data of the target metering instrument to supplement the cumulant data in the abnormal time period; calculating the daily consumption condition of each instrument; if the daily consumption of the instrument is counted to be zero, the consumption of the instrument is judged according to the state of production operation equipment, and the cumulative value of the instrument is deduced reversely to supplement the hourly value of the metering instrument; and generating a data self-correction result according to the cumulant data in the abnormal time period and the hourly value of the metering instrument. The problems that the industrial metering instrument is low in fault detection efficiency and cannot be automatically corrected are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of industrial metering instruments, and in particular to a method for calculating industrial metering instrument data self-correction. Background Art

[0002] With the continuous development of industrial metering instrument technology, instruments have acquired communication functions, which can realize real-time collection and remote analysis of instrument data. However, industrial metering instruments with communication functions are prone to malfunctions during use, resulting in unstable instrument data and prone to errors.

[0003] In order to ensure the stability and accuracy of metering data, some users use manual inspection to realize abnormal perception, fault handling and data correction of industrial metering instruments. The manual inspection method is to conduct daily inspections manually, handle faults after abnormal problems are found, and then adjust the settlement results through manual correction. However, this method has the following disadvantages: First, a large number of instruments cannot be inspected once a day, and the workload of inspectors is huge; second, it is impossible to quickly identify whether it is a network transmission problem or a problem with the instrument itself, resulting in a long fault judgment cycle; third, after the data is abnormal, manual correction is required, and there is a problem of inconsistent correction standards. Summary of the invention

[0004] The present application provides a method for calculating the self-correction of industrial metering instrument data, so as to solve the problem that the fault detection efficiency of industrial metering instruments is low and cannot be automatically corrected.

[0005] The present application provides a method for self-correction calculation of industrial metering instrument data, the method comprising:

[0006] Collect data from target measuring instruments, collection equipment, and network equipment;

[0007] Judging whether the collection link, collection equipment, instrument and instrument data are abnormal based on the data of the target meter, collection equipment and network equipment;

[0008] If the collection link, collection equipment and instrument are normal, but the instrument data is abnormal, the arithmetic average calculation is performed based on the historical data of the target meter to supplement the cumulative data during the abnormal period;

[0009] Calculate the daily consumption of each meter;

[0010] If the daily consumption of the meter is zero, the consumption of the meter is determined by the status of the production equipment, and the cumulative value of the meter is reversely deduced to supplement the hourly value of the meter;

[0011] The data self-correction result is generated based on the accumulated data during the abnormal period and the hourly value of the measuring instrument.

[0012] The above method quickly determines the abnormality of instrument data based on the data of the target metering instrument, collection equipment, and network equipment, and uses historical data to calculate the metering consumption during the abnormal period and supplement the accounting data to automatically correct the instrument data during the abnormal period, so as to solve the problem of low efficiency in fault detection and inability to automatically correct industrial metering instruments.

[0013] Optionally, the data of the target meter, collection device, and network device include:

[0014] The real-time data of the temperature, pressure, instantaneous quantity and cumulative quantity of the target meter and the data refresh time; the collection operation status and forwarding operation status of the collection equipment.

[0015] By capturing the above data from the data of the target metering instrument, collection equipment, and network equipment, it is possible to quickly determine whether the collection link, collection equipment, instrument, and instrument data are abnormal, thereby improving processing efficiency.

[0016] Optionally, after collecting data from the target meter, collection device, and network device, the method further includes:

[0017] Grouping each meter in the target meter to form a network topology;

[0018] extracting historical data in the network topology;

[0019] Analyze the refresh time of the historical data and the state of the collection link to obtain a complete and accurate data set;

[0020] The refresh time of the data set and the state of the acquisition link are trained by a regression model to obtain historical data of the target meter.

[0021] The above method analyzes the refresh time of historical data and the state of the collection link in the network topology to check the integrity and accuracy of the historical data so as to identify missing or erroneous data in the historical data, and then trains the refresh time of the data set and the state of the collection link through a regression model to make the historical data of the target meter more accurate.

[0022] Optionally, the method for determining whether the collection link is abnormal includes:

[0023] Determine the network status of the acquisition device;

[0024] If the network status of the acquisition device is abnormal, the acquisition link is abnormal;

[0025] Generate a collection link abnormality instruction and send it to the network device maintenance terminal.

[0026] Whether the acquisition link is abnormal can be quickly determined by collecting whether the network status of the device is abnormal, and by sending the abnormal information of the acquisition link to the network device maintenance end in the form of instructions, the corresponding personnel can make corrections as soon as possible, thereby improving processing efficiency.

[0027] Optionally, the method for determining whether the acquisition device is abnormal includes:

[0028] If the acquisition link is normal, sending the acquired communication message to the acquisition device;

[0029] If the communication message is sent multiple times but no reply message is received within the preset time, the collection device is abnormal;

[0030] Generate an abnormal instruction for the acquisition device and send it to the acquisition device maintenance terminal.

[0031] Whether the collection device is abnormal can be quickly determined by whether a reply message is received within a preset time, and by sending the abnormal information of the collection device to the collection device maintenance terminal in the form of instructions, the corresponding personnel can make corrections as soon as possible, thereby improving processing efficiency.

[0032] Optionally, methods for determining whether the instrument is abnormal include:

[0033] If the acquisition link and the acquisition device are normal, determine in sequence whether the data refresh time of each meter in the target meter is the current latest time;

[0034] If the data refresh time of the first and last meters is the latest time, and the data refresh time of the middle meter is not the latest time, set the meter unchanged monitoring time according to user needs;

[0035] If the temperature, pressure, instantaneous quantity and cumulative quantity of each instrument do not change for a set period of time, the instrument is abnormal;

[0036] Generate instrument abnormality instructions and send them to the meter maintenance terminal.

[0037] By checking whether the temperature, pressure, instantaneous quantity and cumulative quantity of the instrument change simultaneously, it is possible to quickly determine whether the instrument is abnormal. By sending the abnormal information of the instrument to the maintenance end of the meter in the form of instructions, the corresponding personnel can make corrections as soon as possible to improve processing efficiency.

[0038] Optionally, methods for determining whether instrument data is abnormal include:

[0039] If the data refresh time of each meter in the target meter is the current latest time, and the accumulated value changes instantaneously by multiples, the meter data is abnormal.

[0040] The data refresh time and the accumulated value can be used to quickly determine whether the instrument data is abnormal, so as to facilitate subsequent maintenance and automatic correction.

[0041] Optionally, an arithmetic average calculation is performed based on the historical data of the target meter to supplement the cumulative data during the abnormal period, including:

[0042] Perform arithmetic average calculation based on historical data of the target meter to obtain arithmetic average calculation data;

[0043] Calculate the numerical range of the arithmetic mean calculation data;

[0044] Perform multi-layer nested correction on the arithmetic average calculation data through the Drools rule engine to obtain corrected instrument data;

[0045] Store the corrected instrument data in a separate data list and record the corrected instrument data information.

[0046] The above method can quickly and accurately correct and supplement the accumulated data during the abnormal period, and store the corrected instrument data in a separate data list to facilitate subsequent regression analysis.

[0047] Optionally, the corrected instrument data information includes the instrument name, abnormal data type, abnormal value, corrected value, abnormal time and corrected time;

[0048] While storing the corrected instrument data in a separate data list, the method further includes:

[0049] The corrected value is stored in a real-time database.

[0050] The correction value is stored in a real-time database so as to subsequently calculate the daily consumption of each meter.

[0051] It can be seen from the above technical scheme that the present application provides a method for self-correction calculation of industrial metering instrument data, the method comprising: collecting data of target metering instruments, collection equipment, and network equipment; judging whether the collection link, collection equipment, instrument, and instrument data are abnormal based on the data of the target metering instruments, collection equipment, and network equipment; if the collection link, collection equipment, and instrument are normal, and the instrument data is abnormal, performing arithmetic average calculation based on the historical data of the target metering instrument to supplement the cumulative data during the abnormal period; calculating the daily consumption of each instrument; if the daily consumption of the instrument is zero, judging the consumption of the instrument through the status of the production and operation equipment, and reversely deriving the cumulative value of the instrument to supplement the hourly value of the metering instrument; generating a data self-correction result based on the cumulative data during the abnormal period and the hourly value of the metering instrument to solve the problem that the fault detection efficiency of industrial metering instruments is low and cannot be automatically corrected. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0053] Figure 1 This is a flow chart of the industrial metering instrument data self-correction calculation method described in an embodiment of the present application;

[0054] Figure 2 This is a flow chart of a method for obtaining historical data of a target meter according to an embodiment of the present application;

[0055] Figure 3 A schematic flow chart of a method for supplementing accumulated data within an abnormal period of time according to an embodiment of the present application;

[0056] Figure 4 It is a collection link diagram in the industrial metering instrument data self-correction calculation method described in the embodiment of the present application;

[0057] Figure 5 This is a schematic diagram of instrument data regression analysis in the industrial metering instrument data self-correction calculation method described in an embodiment of the present application;

[0058] Figure 6 It is a curve diagram of instrument data before the abnormal data processing of the pelletizing three lines described in the embodiment of the present application;

[0059] Figure 7 This is a graph of instrument data after processing the three-line abnormal data of pellets described in the embodiment of the present application. DETAILED DESCRIPTION

[0060] The following embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following embodiments do not represent all implementations consistent with the present application. They are only examples of systems and methods consistent with some aspects of the present application as detailed in the claims.

[0061] With the continuous development of industrial metering instrument technology, instruments have acquired communication functions, which can realize real-time collection and remote analysis of instrument data. However, industrial metering instruments with communication functions are prone to malfunctions during use, resulting in unstable instrument data and prone to errors.

[0062] Taking steel enterprises as an example, as the awareness and requirements of energy control in enterprises continue to strengthen, the dimensions of enterprise energy management have gradually sunk to the process level and equipment level. In order to ensure accurate measurement of energy consumption and carbon emissions, enterprises have installed a large number of process-level and equipment-level metering instruments, and equipped with multi-state collection methods. The stability and accuracy of metering data directly affect the calculation of energy costs, carbon emission tax rates, etc. In order to ensure the stability and accuracy of metering data, manual inspection is used to realize abnormal perception, fault handling and data correction of industrial metering instruments. The manual inspection method is to conduct manual daily inspections, handle faults after abnormal problems are found, and then adjust the settlement results through manual corrections. However, this method will have the following disadvantages: First, a large number of instruments cannot be inspected once a day, and the workload of the inspectors is huge; second, it is impossible to quickly identify whether it is a network transmission problem or a problem with the instrument itself, resulting in a long fault judgment cycle; third, after the data is abnormal, manual correction is required, and there is a problem of inconsistent correction standards.

[0063] To solve the problem that the fault detection efficiency of industrial measuring instruments is low and cannot be automatically corrected. Figure 1 , Figure 1 The present invention provides a method for self-correction calculation of industrial metering instrument data, the method comprising:

[0064] S100: Collect data from target measuring instruments, collection equipment, and network equipment.

[0065] S200: judging whether the collection link, the collection device, the meter and the meter data are abnormal according to the data of the target meter, the collection device and the network device.

[0066] S300: If the collection link, collection equipment and instrument are normal, but the instrument data is abnormal, an arithmetic average calculation is performed based on the historical data of the target meter to supplement the accumulated data during the abnormal period.

[0067] S400: Calculate the daily consumption of each meter.

[0068] S500: If the daily consumption of the meter is zero, the consumption of the meter is determined by the status of the production equipment, and the cumulative value of the meter is reversely derived to supplement the hourly value of the meter.

[0069] By reversely deriving the meter's cumulative value to supplement the meter's hourly value, hourly, daily, and monthly consumption accounting can be achieved.

[0070] S600: Generate a data self-correction result according to the accumulated data in the abnormal period and the hourly value of the meter.

[0071] It should be understood that the daily consumption of the meter can be calculated and counted through the energy performance module. The energy performance module is an important part of the enterprise energy management system. It is used to monitor, analyze and optimize the energy efficiency of the enterprise. The energy performance module can not only collect the energy consumption data of the enterprise in real time through data acquisition equipment to ensure the real-time and accuracy of the data. It can also help enterprises understand the patterns and trends of energy consumption, identify high-energy-consuming links and energy-saving potential, and improve the efficiency and sustainability of energy use by analyzing the collected data.

[0072] The above method quickly determines the abnormality of instrument data based on the data of target metering instruments, collection equipment, and network equipment, and uses historical data to calculate the metering consumption during the abnormal period and supplement the accounting data to automatically correct the instrument data during the abnormal period, so as to solve the problem of low efficiency in fault detection and inability to automatically correct industrial metering instruments.

[0073] In some embodiments, a data self-correction result is generated according to the accumulated data in the abnormal period and the hourly value of the meter, including:

[0074] Based on the accumulated data during the abnormal period and the hourly value of the meter, the corrected meter data is obtained.

[0075] Generate data self-correction results based on the corrected instrument data.

[0076] The data self-correction results include the self-correction result report and detailed correction process for subsequent storage and analysis. The self-correction result report contains interactive visual charts, and the detailed correction process includes the method used, formula, reference historical data, instrument readings before and after correction, daily consumption, equipment operating status, etc., which is convenient for subsequent analysis and adjustment.

[0077] In some embodiments, the data of the target meter, collection device, and network device include:

[0078] Real-time data and data refresh time of temperature, pressure, instantaneous quantity and cumulative quantity of target measuring instruments; collection operation status and forwarding operation status of collection equipment.

[0079] By capturing the above data from the target metering instruments, collection equipment, and network equipment, it is possible to quickly determine whether the collection link, collection equipment, instrument, and instrument data are abnormal, thereby improving processing efficiency.

[0080] In some embodiments, see Figure 2 , Figure 2 The following is a flow chart of a method for obtaining historical data of a target meter. After collecting data of the target meter, collection equipment, and network equipment, the method further includes:

[0081] S110: Group each meter in the target meter to form a network topology.

[0082] S120: extracting historical data from the network topology.

[0083] S130: Analyze the refresh time of historical data and the state of the collection link to obtain a complete and accurate data set.

[0084] S140: training the refresh time of the data set and the state of the collection link through a regression model to obtain historical data of the target meter.

[0085] It should be understood that the grouping can be based on the function, location or type of the instrument. For example, the instruments on the same production line can be grouped into one group, or all the instruments measuring temperature can be grouped into another group. Network topology refers to the physical layout of various devices interconnected by transmission media, that is, the specific physical or logical arrangement between the members constituting the network. Network topology can be divided into many types. The network topology of the present application can be selected as a tree topology. The tree topology combines the advantages of star and bus types, is suitable for hierarchical network design, and can expand the network by adding branches. Regression model is a statistical tool for predictive analysis, time series model establishment, and finding causal relationships between different variables. Regression model can be divided into many types. The regression model of the present application can be selected as a logistic regression model. The logistic regression model has the advantages of fast training speed, good model interpretability, suitable for binary classification problems, no need to scale input features, good mathematical properties, low resource requirements, and good robustness, and is suitable for the present application.

[0086] The above method analyzes the refresh time of historical data and the status of the collection link in the network topology to check the integrity and accuracy of the historical data, so as to find out the missing or erroneous data in the historical data, and then trains the refresh time of the data set and the status of the collection link through the regression model to make the historical data of the target meter more accurate.

[0087] In some embodiments, the acquisition link of the present application is a link of a hand-in-hand network topology structure in an RS485 bus network. Figure 4 , Figure 4 This is a collection link diagram in a self-correction calculation method for industrial metering instrument data. The collection link includes an instrument collection server, an instrument collector, and an instrument; the instrument collection server is connected to the instrument collector, and the instrument collector then connects instrument 1, instrument 2, instrument 3, and other instruments in series in sequence to form a hand-in-hand network topology. The hand-in-hand network topology has the advantages of high flexibility, high speed, and strong anti-interference.

[0088] In some embodiments, the method for determining whether the acquisition link is abnormal includes:

[0089] Determine the network status of the acquisition device;

[0090] If the network status of the acquisition device is abnormal, the acquisition link is abnormal;

[0091] Generate a collection link abnormality instruction and send it to the network device maintenance terminal.

[0092] It should be understood that the judgment of abnormal network status can be judged and checked through a series of network diagnostic commands and technologies. For example, use the ping command to test whether the network connection is normal. If error messages such as "request timed out" or "host unreachable" are returned, it usually means that there is a problem with the network connection; use the traceroute command to help view the routing path of the data packet from the local machine to the remote host, so as to identify possible problems in the transmission process; use the ifconfig command to view the status and configuration information of the network interface, including IP address, subnet mask, broadcast address, etc. In addition, if the network status of the acquisition device is normal, the acquisition link is normal, and you can judge whether the acquisition device is abnormal in turn.

[0093] By checking whether the network status of the acquisition device is abnormal, it is possible to quickly determine whether the acquisition link is abnormal, and by sending the acquisition link abnormal information to the network equipment maintenance end in the form of instructions, the corresponding personnel can make corrections as soon as possible to improve processing efficiency.

[0094] In some embodiments, the method for determining whether the acquisition device is abnormal includes:

[0095] If the acquisition link is normal, the acquired communication message is sent to the acquisition device;

[0096] If the communication message is sent multiple times but no reply message is received within the preset time, the collection device is abnormal;

[0097] Generate an abnormal instruction for the acquisition device and send it to the acquisition device maintenance terminal.

[0098] It should be understood that the preset time can be set to 1 minute, and the number of times the communication message is sent can be set to 5. If the reply message information is received within the preset time, the acquisition device is normal, and it can be judged whether the instrument is abnormal in turn.

[0099] Whether the collection device is abnormal can be quickly determined by whether a reply message is received within the preset time. By sending the abnormal information of the collection device to the collection device maintenance end in the form of instructions, the corresponding personnel can make corrections as soon as possible to improve processing efficiency.

[0100] In some embodiments, the method for determining whether the meter is abnormal includes:

[0101] If the acquisition link and acquisition equipment are normal, determine in sequence whether the data refresh time of each meter in the target meter is the current latest time;

[0102] If the data refresh time of the first and last meters is the latest time, and the data refresh time of the middle meter is not the latest time, set the meter unchanged monitoring time according to user needs;

[0103] If the temperature, pressure, instantaneous quantity and cumulative quantity of each instrument do not change at the same time for a set time, the instrument is abnormal;

[0104] Generate instrument abnormality instructions and send them to the meter maintenance terminal.

[0105] Whether the instrument is abnormal can be quickly determined by checking whether the temperature, pressure, instantaneous quantity and cumulative quantity of the instrument change at the same time. By sending the abnormal information of the instrument to the maintenance end of the meter in the form of instructions, the corresponding personnel can make corrections as soon as possible to improve processing efficiency.

[0106] In some embodiments, the method for determining whether the instrument data is abnormal includes:

[0107] If the data refresh time of each instrument in the target meter is the current latest time, and the accumulated value changes instantaneously by multiples, the instrument data is abnormal.

[0108] It should be understood that the instantaneous change in the cumulative value by multiples may specifically be that the cumulative value suddenly increases or decreases by several times. This is a problem of data collection quality, and the acquired data will affect the calculation results. In some embodiments, the multiple of increase or decrease can be set to 10 times.

[0109] The data refresh time and accumulated value can be used to quickly determine whether the instrument data is abnormal, so as to facilitate subsequent maintenance and automatic correction.

[0110] In some embodiments, see Figure 3 , Figure 3 A flow chart of a method for supplementing the accumulated data during an abnormal period. Performing arithmetic average calculation based on the historical data of the target meter to supplement the accumulated data during the abnormal period includes:

[0111] S310: Performing arithmetic mean calculation according to historical data of the target meter to obtain arithmetic mean calculation data.

[0112] S320: Calculate the numerical range of the arithmetic mean calculation data.

[0113] S330: Perform multi-layer nested correction on the arithmetic mean calculation data through the Drools rule engine to obtain corrected instrument data.

[0114] S340: storing the corrected instrument data in a separate data list, and recording the corrected instrument data information.

[0115] It should be understood that Drools is a Java-based rule engine that defines rules in a declarative way and automatically executes these rules at runtime. This approach makes it easier to manage and execute rules, improving flexibility and responsiveness. The specific methods of multi-layer nested correction are: comparison of arithmetic mean values, analysis of the corrected cumulative value storage, and real-time storage and accounting of the corrected value, and finally obtaining the corrected instrument data.

[0116] The above method can quickly and accurately correct and supplement the accumulated data during the abnormal period, and store the corrected instrument data in a separate data list to facilitate subsequent regression analysis.

[0117] It should be understood that the corrected instrument data stored in a separate data table can be re-analyzed using the regression model. Figure 5 , Figure 5 This is a schematic diagram of instrument data regression analysis in a self-correction calculation method for industrial metering instrument data. The abnormality of 0 value can be collected first: when the start value is not equal to 0 and the end value is equal to 0, re-take the data, remove the 0 value, and take the value normally in this time period; when the start value is equal to 0 and the end value is equal to 0, the value in this time period is the last value of the previous time period; when the start value is equal to 0 and the end value is not equal to 0, the last value in this time period is taken normally, but the last value calculated by the first value; when the start value is not equal to 0 and the end value is not equal to 0, no abnormal processing is performed. Then handle the abnormality of jump in the collected value. If the accumulated value in the current time is more than multiple times of the previous time period, the end time is pushed forward 1 minute, and after re-taking the value, it is judged again according to the conditions until no jump occurs; if the accumulated value in the current time is not more than multiple times of the previous time period, no abnormal processing is performed.

[0118] See also Figure 6-Figure 7 , Figure 6 It is a curve diagram of instrument data before the abnormal data of the three lines of pelletizing is processed; Figure 7 This is an instrument data curve after the abnormal data of the pelletizing line 3 is processed. The data correction result after regression analysis can be taken as an example of the pelletizing line 3 5517, and the changes before and after the abnormal data processing are clearly seen. It can be seen that the data correction result has a high accuracy.

[0119] In some embodiments, the corrected meter data information includes meter name, abnormal data type, abnormal value, corrected value, abnormal time, and corrected time;

[0120] While storing the corrected instrument data in a separate data list, the method further includes: storing the corrected values ​​in a real-time database.

[0121] The corrected values ​​are stored in a real-time database to facilitate subsequent calculation of the daily consumption of each meter.

[0122] In some embodiments, an encryption engine and a key management system are provided in the real-time database to encrypt the data in the real-time database, thereby encrypting the fields, tables or the entire database in the real-time database to ensure the confidentiality of the data.

[0123] In some embodiments, the real-time database can be communicatively connected to a cloud platform, and the cloud platform can be used to store and analyze correction values, thereby providing large-scale data processing capabilities and redundant backup to ensure the security of the correction values ​​and the reliability of calculations.

[0124] In some embodiments, after performing multi-layer nested correction on the arithmetic mean calculation data through the Drools rule engine to obtain the corrected instrument data, the method further includes:

[0125] The corrected instrument data information is compared with the actual measured value to obtain the deviation value between the corrected instrument data and the actual measured value.

[0126] If the deviation value exceeds the preset deviation value, the fused historical data and the corrected instrument data information are obtained.

[0127] It should be understood that the preset deviation value can be set to 0.01; the fused historical data is obtained by fusing data from different sources and the same type of collection target metering instruments, collection equipment, and network equipment on the cloud platform to provide more accurate correction data.

[0128] Build a machine learning model with the deviation value as the learning target.

[0129] Train the machine learning model to obtain a trained machine learning model.

[0130] It should be understood that the fused historical data and the corrected instrument data information can be divided into a training set and a test set. The training set is used to train the selected machine learning model. During the training process, the performance of the model is optimized by adjusting the parameters of the model. Grid search, random search or Bayesian optimization strategies can be used to find the optimal parameter combination. The test set is used to evaluate the performance of the model. If the model performance is poor, the model parameters are readjusted.

[0131] Use the trained machine learning model to predict the new corrected meter data to obtain the prediction results.

[0132] Adjust Drools rules based on the prediction results.

[0133] The above method combines the Drools rule engine and machine learning technology to realize an efficient, intelligent and adaptive instrument data correction system to continuously optimize the correction process and improve data quality.

[0134] It can be seen from the above technical scheme that the embodiment of the present application provides a method for self-correction calculation of industrial metering instrument data, the method comprising: collecting data of target metering instruments, collection equipment, and network equipment; judging whether the collection link, collection equipment, instrument, and instrument data are abnormal based on the data of the target metering instruments, collection equipment, and network equipment; if the collection link, collection equipment, and instrument are normal, and the instrument data is abnormal, performing arithmetic average calculation based on the historical data of the target metering instrument to supplement the cumulative data during the abnormal period; calculating the daily consumption of each instrument; if the daily consumption statistics of the instrument are zero, judging the consumption of the instrument through the status of the production and operation equipment, and reversely deriving the cumulative value of the instrument to supplement the hourly value of the metering instrument; generating a data self-correction result based on the cumulative data during the abnormal period and the hourly value of the metering instrument to solve the problem that the fault detection efficiency of industrial metering instruments is low and cannot be automatically corrected.

[0135] Similar parts between the embodiments provided in this application can be referenced to each other. The specific implementation methods provided above are only a few examples under the general concept of this application and do not constitute a limitation on the protection scope of this application. For those skilled in the art, any other implementation methods expanded based on the scheme of this application without creative work belong to the protection scope of this application.

Claims

1. A method for self-correction calculation of industrial metering instrument data, characterized in that: The method comprises: Collect data from target measuring instruments, collection equipment, and network equipment; Judging whether the collection link, collection equipment, instrument and instrument data are abnormal based on the data of the target meter, collection equipment and network equipment; If the collection link, collection equipment and instrument are normal, but the instrument data is abnormal, the arithmetic average calculation is performed based on the historical data of the target meter to supplement the cumulative data during the abnormal period; Calculate the daily consumption of each meter; If the daily consumption of the meter is zero, the consumption of the meter is determined by the status of the production equipment, and the cumulative value of the meter is reversely deduced to supplement the hourly value of the meter; The data self-correction result is generated based on the accumulated data during the abnormal period and the hourly value of the measuring instrument.

2. The industrial metering instrument data self-correction calculation method according to claim 1 is characterized in that: The data of the target measuring instruments, collection equipment, and network equipment include: The real-time data of the temperature, pressure, instantaneous quantity and cumulative quantity of the target meter and the data refresh time; the collection operation status and forwarding operation status of the collection equipment.

3. The industrial metering instrument data self-correction calculation method according to claim 2 is characterized in that: After collecting data from the target meter, collection equipment, and network equipment, the method further includes: Grouping each meter in the target meter to form a network topology; extracting historical data in the network topology; Analyze the refresh time of the historical data and the state of the collection link to obtain a complete and accurate data set; The refresh time of the data set and the state of the acquisition link are trained by a regression model to obtain historical data of the target meter.

4. The industrial metering instrument data self-correction calculation method according to claim 3 is characterized in that: Methods for determining whether the collection link is abnormal include: Determine the network status of the acquisition device; If the network status of the acquisition device is abnormal, the acquisition link is abnormal; Generate a collection link abnormality instruction and send it to the network device maintenance terminal.

5. The industrial metering instrument data self-correction calculation method according to claim 4 is characterized in that: Methods for determining whether the acquisition equipment is abnormal include: If the acquisition link is normal, sending the acquired communication message to the acquisition device; If the communication message is sent multiple times but no reply message is received within the preset time, the collection device is abnormal; Generate an abnormal instruction for the acquisition device and send it to the acquisition device maintenance terminal.

6. The industrial metering instrument data self-correction calculation method according to claim 5 is characterized in that: Methods for determining whether the instrument is abnormal include: If the acquisition link and the acquisition device are normal, determine in sequence whether the data refresh time of each meter in the target meter is the current latest time; If the data refresh time of the first and last meters is the latest time, and the data refresh time of the middle meter is not the latest time, set the meter unchanged monitoring time according to user needs; If the temperature, pressure, instantaneous quantity and cumulative quantity of each instrument do not change for a set period of time, the instrument is abnormal; Generate instrument abnormality instructions and send them to the meter maintenance terminal.

7. The industrial metering instrument data self-correction calculation method according to claim 6 is characterized in that: Methods for determining whether instrument data is abnormal include: If the data refresh time of each meter in the target meter is the current latest time, and the accumulated value changes instantaneously by multiples, the meter data is abnormal.

8. The industrial metering instrument data self-correction calculation method according to claim 3 is characterized in that: Perform arithmetic average calculation based on the historical data of the target meter to supplement the cumulative data during the abnormal period, including: Perform arithmetic average calculation based on historical data of the target meter to obtain arithmetic average calculation data; Calculate the numerical range of the arithmetic mean calculation data; Perform multi-layer nested correction on the arithmetic average calculation data through the Drools rule engine to obtain corrected instrument data; Store the corrected instrument data in a separate data list and record the corrected instrument data information.

9. The industrial metering instrument data self-correction calculation method according to claim 8 is characterized in that: The corrected instrument data information includes the instrument name, abnormal data type, abnormal value, corrected value, abnormal time and corrected time; While storing the corrected instrument data in a separate data list, the method further includes: The corrected value is stored in a real-time database.

10. The industrial metering instrument data self-correction calculation method according to claim 9, characterized in that: An encryption engine and a key management system are arranged in the real-time database, and encryption processing is performed on the real-time database through the encryption engine and the key management system.

Citation Information

Cited By

  • Metering device online detection method, system and equipment based on Internet of Things

    CN120547201A

  • Ammonia spraying method and device based on denitration system

    CN120679341A

  • Ammonia Injection Method and Apparatus Based on Denitrification System

    CN120679341B