Data medium station based on digital twin production management platform

Abstract

The invention discloses a data medium station based on a digital twin production management platform, and particularly relates to the technical field of digital twin, the recent preset time period of operation of a digital twin model in the data medium station is acquired and divided into a plurality of sub time periods, obtaining a data uploading consistency coefficient, a prediction result deviation coefficient and a data format error coefficient of each sub-time period to obtain a precision influence coefficient; and comparing the precision influence coefficients in all the sub-time periods with a preset precision influence coefficient threshold, and judging whether the precision of the digital twin model in the data center has a problem or not according to a comparison result, so that the precision of the digital twin model is self-detected, and once the precision of the digital twin model has the problem, the digital twin model is not detected. According to the technical scheme, an alarm can be given in time to remind related personnel to take corresponding measures, so that the data medium station can accurately reflect the actual production environment, production management decisions are ensured not to be mistaken or delayed, and the influence on the efficiency of the whole production process and the product quality is reduced.

Description

Technical Field

[0001] The present invention relates to the field of digital twin technology, and in particular to a data center based on a digital twin production management platform. Background Art

[0002] The data middle platform of the production management platform based on digital twin is a solution for integrating and managing production process data. It uses digital twin technology to combine real-time data of the actual production environment with virtual models, thereby optimizing the production process, improving efficiency and reducing costs. Generally, it is achieved through real-time data collection and integration, digital twin model creation, synchronous simulation, data analysis and optimization decision-making, feedback and continuous improvement, and visualization and decision support. The data middle platform integrates and manages multi-source data, synchronizes it with the virtual model, optimizes the production process through analysis and simulation, and continuously provides feedback and improvements, ultimately improving production efficiency, reducing costs, and helping enterprises achieve intelligent manufacturing and digital transformation.

[0003] However, the existing data middle platform based on the digital twin production management platform lacks the self-check function of the digital twin model accuracy. Once there is a problem with the accuracy of the digital twin model, it will cause problems with the data in the data middle platform, making it difficult for the data middle platform to maintain its accurate reflection of the actual production environment, resulting in errors or delays in production management decisions, which in turn affects the efficiency of the entire production process and product quality. Summary of the invention

[0004] The purpose of the present invention is to provide a data center based on a digital twin production management platform to address the deficiencies in the background technology.

[0005] In order to achieve the above-mentioned purpose, the present invention provides the following technical solutions: a data middle platform based on a digital twin production management platform, comprising; Obtain the most recent preset time period of the operation of the Taichung digital twin model in the data, and divide it into several sub-time periods according to the preset time window; Obtain the data upload consistency coefficient, prediction result deviation coefficient, and data format error coefficient of the Taichung digital twin model in the data in each sub-time period; The data upload consistency coefficient, the prediction result deviation coefficient and the data format error coefficient are normalized, and the accuracy influence coefficient is obtained according to the normalized data upload consistency coefficient, the prediction result deviation coefficient and the data format error coefficient; The accuracy impact coefficients in all sub-time periods are compared with the preset accuracy impact coefficient thresholds, and based on the comparison results, it is determined whether there are problems with the accuracy of the Taichung digital twin model in the data.

[0006] Optionally, the data upload consistency coefficient includes: Get the data uploaded by several sensors at different times in each sub-time period and mark them as , Indicates the sequence number of the time. =1, 2, 3, 4, ..., , is the total number of moments, and is a positive integer; Indicates the order number of the sensor. =1, 2, 3, 4, ..., , is the total number of sensors, and is a positive integer; Calculate the data differences between all sensor pairs , and obtain the difference matrix , where =1, 2, 3, 4, ..., ,, =1, 2, 3, 4, ..., ; At each time point The difference matrix is a × The matrix of Indicates sensor and sensors At the point in time The data differences on Each sensor The preset weight factor is expressed as , the weight factor satisfies ; Calculating weighted differences , the calculation formula is: ; Calculate weighted square root , the calculation formula is: .

[0007] Optionally, the data upload consistency coefficient also includes: Calculate adjacent time points and Data change rate , the calculation formula is: ; and when hour, ; Calculate the rate of change difference between sensors , the calculation formula is: ; and using a similar weighted RMS difference calculation method, we get the time series correlation difference ; Combined with the time series correlation and weighted root mean square difference, the data upload consistency coefficient at each moment is calculated , the calculation formula is: ; In the formula, and is the regulating factor, The maximum permissible difference value is preset; Calculate the data upload consistency coefficient , the calculation formula is: .

[0008] Optionally, the prediction result deviation coefficient includes: Get the number of results predicted by the digital twin model in each sub-time period and mark it as , The number indicating the number of results predicted by the digital twin model in each sub-time period. =1, 2, 3, 4, ..., , Represents the total number of results predicted by the digital twin model in each sub-time period, is a positive integer; Compare the predicted results of each digital twin model with the corresponding actual results, and re-mark the results that are inconsistent with the corresponding actual results. , The sequence number indicating the number of inconsistencies between the predicted results and the corresponding actual results. =1, 2, 3, 4, ..., , is a positive integer; Calculate the coefficient of deviation of the forecast results , the calculation formula is: .

[0009] Optionally, the data format error coefficient includes: Data format errors are divided into multiple types, and the preset weights of the impact of different types of format errors on the accuracy of the digital twin model are marked as , A number indicating the type of data format error. =1, 2, 3, 4, ..., , Indicates the total number of data format error types. is a positive integer; and satisfies ; Obtain the total number of data entries uploaded to the digital twin model in each sub-time period and the number of data error entries corresponding to each data format error, and for each error type, calculate the corresponding error ratio , the calculation formula is: , where Indicates The number of data error entries corresponding to the data format error type, Indicates the total number of data entries; Calculate the data format error coefficient , the calculation formula is: .

[0010] Optionally, the accuracy impact coefficient obtained according to the normalized data upload consistency coefficient, the prediction result deviation coefficient and the data format error coefficient includes: In the formula, is the accuracy influence coefficient, , and They are the normalized data upload consistency coefficient, prediction result deviation coefficient, and data format error coefficient. are preset proportional coefficients of the normalized data upload consistency coefficient, the prediction result deviation coefficient, and the data format error coefficient, respectively, and The scores are greater than 0.

[0011] Optionally, the accuracy impact coefficients in all sub-time periods are compared with a preset accuracy impact coefficient threshold, and judging whether there is an accuracy problem of the digital twin model in the data according to the comparison result includes: When the accuracy impact coefficient in any sub-time period is not less than the preset accuracy impact coefficient threshold, it means that the accuracy of the digital twin model in the data center based on the digital twin production management platform has problems and an alarm is issued; When the accuracy impact coefficient in each sub-time period is less than the preset accuracy impact coefficient threshold, it means that there is no problem with the accuracy of the digital twin model in the data center based on the digital twin production management platform, and an alarm is issued.

[0012] Beneficial effects of the present invention: The present invention proposes a data middle platform based on a digital twin production management platform. By obtaining the most recent preset time period of the operation of the digital twin model in the data middle platform, and dividing the most recent preset time period into several sub-time periods, the data upload consistency coefficient, prediction result deviation coefficient and data format error coefficient of the digital twin model in the data middle platform in each sub-time period are obtained to obtain the accuracy influence coefficient; the accuracy influence coefficient in all sub-time periods is compared with the preset accuracy influence coefficient threshold, and according to the comparison result, it is judged whether there is a problem with the accuracy of the digital twin model in the data middle platform. In this way, the accuracy of the digital twin model is self-checked. Once there is a problem with the accuracy of the digital twin model, an alarm can be issued in time to remind relevant personnel to take corresponding measures, so that the data middle platform can maintain its accurate reflection of the actual production environment, ensure that production management decisions will not be wrong or delayed, and reduce the impact on the efficiency of the entire production process and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The present invention will be further described below in conjunction with the accompanying drawings.

[0014] Figure 1 A flowchart of a data middle platform based on a digital twin production management platform is provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0017] The embodiment of the present invention provides a data center based on a digital twin production management platform. Figure 1 , Figure 1 A flowchart of a data center based on a digital twin production management platform provided in an embodiment of the present invention. It includes: Obtain the most recent preset time period of the operation of the Taichung digital twin model in the data, and divide it into several sub-time periods according to the preset time window; Obtain the data upload consistency coefficient, prediction result deviation coefficient, and data format error coefficient of the Taichung digital twin model in the data in each sub-time period; The data upload consistency coefficient, the prediction result deviation coefficient and the data format error coefficient are normalized, and the accuracy influence coefficient is obtained according to the normalized data upload consistency coefficient, the prediction result deviation coefficient and the data format error coefficient; The accuracy impact coefficients in all sub-time periods are compared with the preset accuracy impact coefficient thresholds, and based on the comparison results, it is determined whether there are problems with the accuracy of the Taichung digital twin model in the data.

[0018] It should be noted that the recent preset time period and preset time window are set by professionals according to actual conditions, and are not specifically limited or elaborated.

[0019] Based on a data middle platform based on a digital twin production management platform provided by an embodiment of the present invention, the accuracy of the digital twin model is self-checked in the above manner. Once there is a problem with the accuracy of the digital twin model, an alarm can be issued in time to remind relevant personnel to take corresponding measures, so that the data middle platform can maintain its accurate reflection of the actual production environment, ensure that production management decisions will not be wrong or delayed, and reduce the impact on the efficiency of the entire production process and product quality.

[0020] In one embodiment, the data upload consistency coefficient includes: Get the data uploaded by several sensors at different times in each sub-time period and mark them as , Indicates the sequence number of the time. =1, 2, 3, 4, ..., , is the total number of moments, and is a positive integer; Indicates the order number of the sensor. =1, 2, 3, 4, ..., , is the total number of sensors, and is a positive integer; Calculate the data differences between all sensor pairs , and obtain the difference matrix , where =1, 2, 3, 4, ..., ,, =1, 2, 3, 4, ..., ; At each time point The difference matrix is a × The matrix of Indicates sensor and sensors At the point in time The data differences on Each sensor The preset weight factor is expressed as , the weight factor satisfies ; Calculating weighted differences , the calculation formula is: ; Calculate weighted square root , the calculation formula is: ; Calculate adjacent time points and Data change rate , the calculation formula is: ; and when hour, ; Calculate the rate of change difference between sensors , the calculation formula is: ; and using a similar weighted RMS difference calculation method, we get the time series correlation difference ; Combined with the time series correlation and weighted root mean square difference, the data upload consistency coefficient at each moment is calculated , the calculation formula is: ; In the formula, and is the regulating factor, The maximum permissible difference value is preset; Calculate the data upload consistency coefficient , the calculation formula is: .

[0021] It should be noted that and is the adjustment factor. Generally, Used to adjust the weighted RMS difference Impact on the final consistency coefficient; Used to adjust timing correlation differences Impact on the final consistency coefficient; in general, and The value of is 0.5, and may be other numbers, which are not limited or elaborated in detail; in addition, the preset weight factor and the preset maximum allowable difference value are set by professionals according to actual conditions, which are not limited or elaborated in detail.

[0022] It should be noted that the data upload consistency coefficient refers to the consistency of the data uploaded by the digital twin model in the data center; if the consistency of the uploaded data is small, it means that there is a high possibility that there will be problems with the accuracy of the digital twin model, which will cause problems with the data in the data center, making it difficult for the data center to maintain its accurate reflection of the actual production environment, resulting in errors or delays in production management decisions, and then affecting the efficiency and product quality of the entire production process. The reason is that data upload consistency is a key factor in ensuring the accuracy of the digital twin model. The digital twin model generates an accurate simulation of the production system by receiving and processing data from multiple sensors in real time. If there are inconsistencies in the time or value of these uploaded data, the model will be calculated based on erroneous or incomplete data, resulting in deviations between the simulation results and the actual production situation. This will not only affect the real-time monitoring of the production process, but also mislead management decisions, leading to improper resource scheduling, invalid production plans, and even the risk of equipment failures not being predicted and handled in time. Over time, this deviation will continue to accumulate, weakening the accuracy of the data center's mapping of the actual production environment, ultimately leading to a decrease in the efficiency of the entire production process, unstable product quality, and possibly increased production costs and risks.

[0023] In one embodiment, the prediction result deviation coefficient includes: Get the number of results predicted by the digital twin model in each sub-time period and mark it as , The number indicating the number of results predicted by the digital twin model in each sub-time period. =1, 2, 3, 4, ..., , Represents the total number of results predicted by the digital twin model in each sub-time period, is a positive integer; Compare the predicted results of each digital twin model with the corresponding actual results, and re-mark the results that are inconsistent with the corresponding actual results. , The sequence number indicating the number of inconsistencies between the predicted results and the corresponding actual results. =1, 2, 3, 4, ..., , is a positive integer; Calculate the coefficient of deviation of the forecast results , the calculation formula is: .

[0024] It should be noted that the prediction result deviation coefficient refers to the degree of inconsistency between the predicted results of the digital twin model in the data center and the actual results; if the degree of inconsistency between the predicted results and the actual results is large, it means that there is a high possibility that there is a problem with the accuracy of the digital twin model, which will cause problems with the data in the data center, making it difficult for the data center to maintain its accurate reflection of the actual production environment, thereby causing errors or delays in production management decisions, and then affecting the efficiency and product quality of the entire production process. The reason is that the prediction result deviation coefficient measures the degree of deviation between the predicted results of the digital twin model in the data center and the actual production results, and is an important indicator for evaluating the accuracy of the model. When the deviation coefficient is large, it means that there is a problem with the model's predictive ability, which may be caused by a variety of factors, such as insufficient model training, poor data input quality, or changes in the external environment that are not reflected in a timely manner. Since the digital twin model relies on the fusion of historical data and real-time data for dynamic prediction, a large deviation coefficient indicates that the model's simulation of the production environment is no longer accurate. This problem not only affects the credibility of the model itself, but also directly has a negative impact on the overall data quality of the data center. The core function of the data center is to integrate, process and provide high-quality data to support production management decisions. As a key component, the digital twin model provides management with a basis for decision-making by simulating and predicting the status of the production system. However, when the prediction result deviation coefficient is too large, the prediction result generated by the model will be significantly different from the actual situation. This difference means that the data provided by the data center no longer accurately reflects the real production environment, which may lead to errors or delays in decisions based on these data, resulting in management making wrong resource allocation or production scheduling decisions, resulting in a decrease in production efficiency.

[0025] In one embodiment, the data format error coefficients include: Data format errors are divided into multiple types, and the preset weights of the impact of different types of format errors on the accuracy of the digital twin model are marked as , A number indicating the type of data format error. =1, 2, 3, 4, ..., , Indicates the total number of data format error types. is a positive integer; and satisfies ; Obtain the total number of data entries uploaded to the digital twin model in each sub-time period and the number of data error entries corresponding to each data format error, and for each error type, calculate the corresponding error ratio , the calculation formula is: , where Indicates The number of data error entries corresponding to the data format error type, Indicates the total number of data entries; Calculate the data format error coefficient , the calculation formula is: .

[0026] It should be noted that the preset weights of the impact of different types of format errors on the accuracy of the digital twin model are set by professionals based on actual circumstances, and no specific limitations or elaboration are made.

[0027] It should be noted that the data format error coefficient refers to the severity of errors in the format of the data uploaded to the digital twin model in the data center; if the error severity is high, it means that there is a high possibility that there will be problems with the accuracy of the digital twin model, which will cause problems with the data in the data center, making it difficult for the data center to maintain its accurate reflection of the actual production environment, resulting in errors or delays in production management decisions, and thus affecting the efficiency and product quality of the entire production process. The reason is that if the data format error severity is high, it indicates that there are many problems in the format of the data uploaded to the data center, such as data type mismatch, field order confusion or encoding errors. These format errors will cause difficulties in the digital twin model when parsing and processing data, thereby affecting the validity and accuracy of the model's input data. When the model receives data in an incorrect format, it may not be able to correctly interpret the data or perform effective calculations, resulting in simulation and prediction results deviating from the actual situation. This deviation will weaken the data center's ability to truly reflect the production environment, resulting in production management decisions based on erroneous or incomplete information, which may cause confusion in the production process, improper resource allocation, delays in equipment maintenance, and even a decline in product quality. Over time, these problems will gradually accumulate, leading to reduced production efficiency, increased operating costs, and weakened corporate competitiveness.

[0028] In one embodiment, the accuracy impact coefficient obtained according to the normalized data upload consistency coefficient, the prediction result deviation coefficient and the data format error coefficient includes: In the formula, is the accuracy influence coefficient, , and They are the normalized data upload consistency coefficient, prediction result deviation coefficient, and data format error coefficient. are preset proportional coefficients of the normalized data upload consistency coefficient, the prediction result deviation coefficient, and the data format error coefficient, respectively, and The scores are greater than 0.

[0029] In one embodiment, the accuracy impact coefficients in all sub-time periods are compared with a preset accuracy impact coefficient threshold, and judging whether there is an accuracy problem of the digital twin model in the data according to the comparison result includes: When the accuracy impact coefficient in any sub-time period is not less than the preset accuracy impact coefficient threshold, it means that the accuracy of the digital twin model in the data center based on the digital twin production management platform has problems and an alarm is issued; When the accuracy impact coefficient in each sub-time period is less than the preset accuracy impact coefficient threshold, it means that there is no problem with the accuracy of the digital twin model in the data center based on the digital twin production management platform, and an alarm is issued.

[0030] In one implementation, the above-mentioned preset accuracy influence coefficient reference threshold is set by professionals according to actual conditions, and the details are not repeated here.

[0031] It should be noted that when an alarm signal is issued, it means that there is a problem with the accuracy of the digital twin model in the data center of the production management platform based on the digital twin, which will cause problems with the data in the data center, making it difficult for the data center to maintain its accurate reflection of the actual production environment, resulting in errors or delays in production management decisions, and then affecting the efficiency of the entire production process and product quality. Therefore, an alarm is issued in time to remind staff that there is a problem with the accuracy of the digital twin model in the data center of the production management platform based on the digital twin, and corresponding measures need to be taken in a timely manner to ensure that the data center maintains its accurate reflection of the actual production environment, ensure that production management decisions will not be wrong or delayed, and reduce the impact on the efficiency of the entire production process and product quality.

[0032] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A data center based on a digital twin production management platform, characterized in that: include: Obtain the most recent preset time period of the operation of the Taichung digital twin model in the data, and divide it into several sub-time periods according to the preset time window; Obtain the data upload consistency coefficient, prediction result deviation coefficient, and data format error coefficient of the Taichung digital twin model in the data in each sub-time period; The data upload consistency coefficient, the prediction result deviation coefficient and the data format error coefficient are normalized, and the accuracy influence coefficient is obtained according to the normalized data upload consistency coefficient, the prediction result deviation coefficient and the data format error coefficient; The accuracy impact coefficients in all sub-time periods are compared with the preset accuracy impact coefficient thresholds, and based on the comparison results, it is determined whether there are problems with the accuracy of the Taichung digital twin model in the data.

2. According to a data platform based on a digital twin production management platform according to claim 1, it is characterized in that: The data upload consistency coefficient includes: Get the data uploaded by several sensors at different times in each sub-time period and mark them as , Indicates the sequence number of the time. =1, 2, 3, 4, ..., , is the total number of moments, and is a positive integer; Indicates the order number of the sensor. =1, 2, 3, 4, ..., , is the total number of sensors, and is a positive integer; Calculate the data differences between all sensor pairs and get the difference matrix , where =1, 2, 3, 4, ..., ,, =1, 2, 3, 4, ..., ; At each time point The difference matrix is a × The matrix of Indicates sensor and sensors At the point in time The data differences on Each sensor The preset weight factor is expressed as , the weight factor satisfies ; Calculating weighted differences , the calculation formula is: ; Calculate weighted square root , the calculation formula is: .

3. According to a data platform based on a digital twin production management platform according to claim 2, it is characterized in that: The data upload consistency coefficient also includes: Calculate adjacent time points and Data change rate , the calculation formula is: ; and when hour, ; Calculate the rate of change difference between sensors , the calculation formula is: ; and using a similar weighted RMS difference calculation method, we get the time series correlation difference ; Combined with the time series correlation and weighted root mean square difference, the data upload consistency coefficient at each moment is calculated , the calculation formula is: ; In the formula, and is the regulating factor, The maximum permissible difference value is preset; Calculate the data upload consistency coefficient , the calculation formula is: .

4. According to a data platform based on a digital twin production management platform according to claim 1, it is characterized in that: The prediction result deviation coefficient includes: Get the number of results predicted by the digital twin model in each sub-time period and mark it as , The number indicating the number of results predicted by the digital twin model in each sub-time period. =1, 2, 3, 4, ..., , Represents the total number of results predicted by the digital twin model in each sub-time period, is a positive integer; Compare the predicted results of each digital twin model with the corresponding actual results, and re-mark the results that are inconsistent with the corresponding actual results. , The sequence number indicating the number of inconsistencies between the predicted results and the corresponding actual results. =1, 2, 3, 4, ..., , is a positive integer; Calculate the coefficient of deviation of the forecast results , the calculation formula is: .

5. According to a data platform based on a digital twin production management platform according to claim 1, it is characterized in that: The data format error coefficients include: Data format errors are divided into multiple types, and the preset weights of the impact of different types of format errors on the accuracy of the digital twin model are marked as , A number indicating the type of data format error. =1, 2, 3, 4, ..., , Indicates the total number of data format error types. is a positive integer; and satisfies ; Obtain the total number of data entries uploaded to the digital twin model in each sub-time period and the number of data error entries corresponding to each data format error, and for each error type, calculate the corresponding error ratio , the calculation formula is: , where Indicates The number of data error entries corresponding to the data format error type, Indicates the total number of data entries; Calculate the data format error coefficient , the calculation formula is: .

6. According to a data platform based on a digital twin production management platform according to claim 1, it is characterized in that: According to the normalized data upload consistency coefficient, prediction result deviation coefficient and data format error coefficient, the accuracy impact coefficients include: In the formula, is the accuracy influence coefficient, , and They are the normalized data upload consistency coefficient, the prediction result deviation coefficient, and the data format error coefficient. are preset proportional coefficients of the normalized data upload consistency coefficient, the prediction result deviation coefficient, and the data format error coefficient, respectively, and The scores are greater than 0.

7. According to a data platform based on a digital twin production management platform according to claim 1, it is characterized in that: Compare the accuracy impact coefficients in all sub-time periods with the preset accuracy impact coefficient thresholds, and determine whether there are problems with the accuracy of the digital twin model in the data based on the comparison results, including: When the accuracy impact coefficient in any sub-time period is not less than the preset accuracy impact coefficient threshold, it means that the accuracy of the digital twin model in the data center based on the digital twin production management platform has problems and an alarm is issued; When the accuracy impact coefficient in each sub-time period is less than the preset accuracy impact coefficient threshold, it means that there is no problem with the accuracy of the digital twin model in the data center based on the digital twin production management platform, and an alarm is issued.

Images