A rescheduling method, system and device based on digital twin

Abstract

The present invention provides a rescheduling method, system and device based on digital twins, wherein the method includes obtaining and comparing the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; the deductive model at time T-1 is obtained based on the sampling model at time T; the digital twin model is corrected according to the first deviation value; the rescheduling threshold is corrected based on the corrected digital twin model; the rescheduling threshold is pre-stored; the sampling model at time T+1 and the deductive model at time T+1 are obtained to obtain a second deviation value; if the second deviation value does not meet the corrected rescheduling threshold, rescheduling is performed according to a preset rule. The embodiment of the present invention applies the digital twin model to the rescheduling judgment of the dynamic scheduling of the production line, so that it can better reflect the characteristics of the man-machine-material-method loop and achieve more accurate scheduling.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a rescheduling method, system and device based on digital twins. Background Art

[0002] The manufacturing process is very complex and often changes dynamically. Therefore, production scheduling becomes an important part of workshop manufacturing.

[0003] Production scheduling includes static scheduling and dynamic scheduling. Dynamic scheduling usually refers to scheduling performed when there are unpredictable disturbances in the scheduling environment and tasks. Compared with static scheduling, dynamic scheduling can produce more operational decision-making plans based on the actual situation of the production site. A large number of problems in actual production occur randomly. For example, in the machinery manufacturing industry, due to random events such as the random arrival of workpieces and failures of processing machines, pre-scheduling cannot be performed normally, which requires rescheduling. For example, in the steelmaking-continuous casting-hot rolling integrated production process of steel enterprises, since the logistics connection and rhythm matching between the processes are required to be coordinated and consistent, and high-temperature continuous operation is required, its production scheduling system has the following characteristics: it can generate real-time scheduling online; it can realize online identification of random disturbances, and it can quickly and automatically reschedule.

[0004] At present, in the prior art, the rescheduling judgment of dynamic scheduling is generally based on the existing scheduling model, and the threshold of rescheduling judgment is given. Once the operation deviation of the system exceeds the threshold, rescheduling is triggered. However, in the prior art, the threshold of dynamic rescheduling is set as a fixed value, which cannot fully consider the specificity of the human-machine-material-method loop, is prone to errors, and leads to low efficiency. Summary of the invention

[0005] The present invention provides a rescheduling method, system and device based on digital twins, which are used to solve the problems in the prior art that dynamic rescheduling is prone to errors and low efficiency, and to achieve dynamic, accurate and efficient adjustment of the rescheduling threshold.

[0006] In the first aspect, the present invention provides a rescheduling method based on digital twins, including: obtaining and comparing a sampling model at time T-1 and a deductive model at time T-1 to obtain a first deviation value; the deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1; correcting the digital twin model according to the first deviation value; correcting the rescheduling threshold based on the corrected digital twin model; the rescheduling threshold is pre-stored; obtaining and comparing the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; the second deviation value is the model difference data between the sampling model at time T+1 and the deductive model at time T+1; if the second deviation value does not meet the corrected rescheduling threshold, rescheduling is performed according to a preset rule; wherein T is a positive integer greater than or equal to 2; the sampling model is a digital twin model obtained by sampling a physical entity; the deductive model is a digital twin model obtained by running or reverse running in the digital twin model according to the sampling model at adjacent moments.

[0007] Optionally, the model difference data includes at least one of the following: time deviation data; workpiece precision deviation data; quality information deviation data; sample quantity deviation data.

[0008] Optionally, if the second deviation value does not meet the revised rescheduling threshold, rescheduling is performed according to preset rules, specifically including: if the time deviation data does not meet the revised rescheduling threshold, rescheduling is performed according to the time priority scheduling scheme; if the workpiece precision deviation data does not meet the revised rescheduling threshold, rescheduling is performed according to the precision priority scheduling scheme; if both the time deviation data and the workpiece precision deviation data do not meet the revised rescheduling threshold, rescheduling is performed according to the comprehensive scheduling scheme.

[0009] Optionally, the acquiring and comparing the sampling model at time T-1 and the deductive model at time T-1 to obtain the first deviation value also includes: constructing a digital twin model; the constructing of the digital twin model includes: acquiring process information, and establishing a visual digital model based on the process information; acquiring model features, and performing data projection of the model features and the visual digital model; acquiring operation data, performing data analysis on the operation data to obtain data features, and performing data projection of the data features and the visual digital model.

[0010] Optionally, the model features are modified according to the data features, and data projection is performed between the modified model features and the visualized digital model.

[0011] Optionally, correcting the digital twin model according to the first deviation value includes: correcting the model characteristics according to the first deviation value, and performing data projection on the corrected model characteristics and the visualized digital model.

[0012] In the second aspect, the present invention also provides a rescheduling system based on digital twins, including: a first comparison module, a model correction module, a threshold correction module, a second comparison module and a rescheduling module; the first comparison module is used to obtain and compare the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; the deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1; the model correction module is used to correct the digital twin model according to the first deviation value; the threshold correction module is used to correct the rescheduling threshold based on the corrected digital twin model. ; The rescheduling threshold is pre-stored; a second comparison module is used to obtain and compare the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; the second deviation value is the model difference data between the sampling model at time T+1 and the deductive model at time T+1; a rescheduling module is used to rescheduling according to preset rules if the second deviation value does not meet the revised rescheduling threshold; wherein T is a positive integer greater than or equal to 2; the sampling model is a digital twin model obtained by sampling a physical entity; the deductive model is a digital twin model obtained by running or reverse running the sampling model at adjacent moments in the digital twin model.

[0013] In a third aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the digital twin-based rescheduling method as described above are implemented.

[0014] In a fourth aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the digital twin-based rescheduling methods described above.

[0015] In a fifth aspect, the present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any of the digital twin-based rescheduling methods described above.

[0016] The present invention provides a digital twin-based rescheduling method, system and device, which apply the digital twin model to the rescheduling judgment of dynamic scheduling of production lines, so that it can better reflect the characteristics of the man-machine-material-method loop and achieve more accurate scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flowchart of a digital twin-based rescheduling method provided in an embodiment of the present invention;

[0019] Figure 2 is a sampling schematic diagram of the digital twin model provided by an embodiment of the present invention;

[0020] Figure 3 It is one of the flowcharts of building a digital twin model provided by an embodiment of the present invention;

[0021] Figure 4 This is the second flowchart of building a digital twin model provided by an embodiment of the present invention;

[0022] Figure 5 It is one of the structural schematic diagrams of the digital twin-based rescheduling system provided in an embodiment of the present invention;

[0023] Figure 6 This is the second structural schematic diagram of the digital twin-based rescheduling system provided in an embodiment of the present invention;

[0024] Figure 7 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0026] Combine the following Figure 1-Figure 4 The present invention describes a digital twin-based rescheduling method.

[0027] refer to Figure 1 , Figure 1 : is a flow chart of a rescheduling method based on digital twins provided in an embodiment of the present invention. An embodiment of the present invention provides a rescheduling method based on digital twins, comprising:

[0028] Step 110, obtaining and comparing the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; the deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1;

[0029] Step 120, correcting the digital twin model according to the first deviation value;

[0030] Step 130, modifying the rescheduling threshold based on the modified digital twin model; the rescheduling threshold is pre-stored;

[0031] Step 140, obtaining and comparing the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; the second deviation value is the model difference data between the sampling model at time T+1 and the deductive model at time T+1;

[0032] Step 150, if the second deviation value does not meet the modified rescheduling threshold, rescheduling is performed according to a preset rule;

[0033] Among them, T is a positive integer greater than or equal to 2; the sampling model is a digital twin model obtained by sampling the physical entity; the deductive model is a digital twin model obtained by running or reverse running the sampling model at adjacent moments in the digital twin model.

[0034] The digital twin model in this embodiment is established based on the production workshop and is a digital twin model that is mapped one-to-one with the physical space. Production may include textiles, papermaking, steelmaking, etc., which is not limited in this embodiment.

[0035] The sampling model in this embodiment is a digital twin model obtained by sampling a physical entity; the deductive model is a digital twin model obtained by running or reverse running the sampling model at adjacent moments in the digital twin model system.

[0036] As reference Figure 2 , Figure 2 It is a sampling diagram of the digital twin model provided by an embodiment of the present invention. In terms of rescheduling judgment, the sampling period of system normalization is used as the interval. For example, the physical entity RS1 at the first moment is sampled to obtain the digital twin model corresponding to the physical entity RS1 at the first moment, that is, the sampling model DT1; in the digital twin model, according to the sampling model DT1, the deductive model DT1-T2 at the second moment can be obtained. That is, according to the sampling model DT1 at the first moment, there will be a sampling model DT2 corresponding to the physical entity at the second moment at the second moment, and a deductive model DT&T1 obtained by simulating the operation of the sampling model at the first moment.

[0037] By analogy, the physical entity RS3 at the third moment corresponds to the sampling model DT3, and the physical entity RS4 at the fourth moment corresponds to the sampling model DT4; by running according to the sampling model DT3, the deductive model DT3-T4 at the fourth moment can be obtained.

[0038] At the same time, looking at it the other way around, for the sampling model DT2 at the second moment, according to the reverse operation of the simulation, the deductive model DT2-T1 at the first moment can also be obtained. Similarly, according to the reverse operation of the sampling model DT4, the deductive model DT4-T3 at the third moment can be obtained.

[0039] This embodiment applies the digital twin model to the rescheduling judgment of dynamic scheduling of production lines, so that it can better reflect the characteristics of the man-machine-material-method loop and achieve more accurate scheduling.

[0040] Next, possible implementations of the above steps in specific embodiments are further described.

[0041] Optional, reference Figure 3 , Figure 3 It is a schematic diagram of the process of building a digital twin model provided by an embodiment of the present invention. Before step 110, it also includes: building a digital twin model;

[0042] Building a digital twin model includes:

[0043] Step 310, obtaining process information, and establishing a visual digital model based on the process information;

[0044] Step 320, obtaining model features, and performing data projection on the model features and the visualized digital model;

[0045] Step 330 , obtaining operation data, performing data analysis on the operation data to obtain data features, and performing data projection on the data features and the visualized digital model.

[0046] This embodiment builds a digital twin model based on process information, model characteristics and data characteristics. By establishing a digital twin model with richer data information and more diversified applications, it provides a more accurate basis for rescheduling decisions.

[0047] Step 310: Acquire process information and establish a visual digital model based on the process information.

[0048] In this step, process information refers to all information on how to complete the production process information involved in the production process obtained from the physical workshop system in the physical space, such as workshop resource information, product process information, etc.

[0049] Workshop resource information refers to the material and human resources that need to be invested in the production and processing process, mainly including workshop layout, material resources, equipment resources, human resources, storage resources, logistics resources, etc. Among them, workshop layout can include information such as workshop building facilities and workshop space layout; material resources can include information such as material number, name, category, weight, appearance and supplier; equipment resources can include equipment number, name, spatial location, status, etc.; human resources can include operator information such as name and work number; storage resources can include the location of in-stock resources, in-stock quantity, in-and-out information, etc.; logistics resources can include forklift number, load capacity, etc.

[0050] Product process information refers to the processing information of the product workpiece in the workshop. It mainly includes product information, material composition, process route, process parameters and other information. Among them, product information mainly refers to its basic attributes, material composition mainly includes the types and quantities of parts required and the composition relationship between them; process route includes basic process information, process sequence, processing equipment and workers corresponding to the process, etc.; process parameters refer to the processing details during the process, such as process processing time, processing cost and energy consumption.

[0051] A visual digital model is established based on the process information, and according to the process information, a visual digital model is established which corresponds one to one with the process information in the physical space.

[0052] Step 320 , obtaining model features, and performing data projection on the model features and the visualized digital model.

[0053] In this step, the model feature refers to the production data obtained in the physical workshop system based on the equipment manual, historical experience, big data, etc. This data is a statistic obtained by summarizing a large amount of production data over a long period of time and in a wide range of applications. The statistic can be a statistical average, a normal distribution, etc. For example, the average manufacturing time of workpieces for type A workpiece manufacturing equipment obtained by the manufacturer based on factory tests is: an average of 4.5 minutes to 5.5 minutes to manufacture a workpiece; the logistics company obtains that it takes 36 hours to ship from city B to city C based on big data.

[0054] Model features are generally recorded in the operating instructions. The "man-machine-material-method-environment" of the production site will have an impact on production efficiency. When scheduling, historical statistical values ​​will be used as estimated values ​​for calculation. However, this historical average value is a statistical value obtained from thousands of machines, thousands of people, and thousands of situations.

[0055] In this step, data projection of the model features with the visualized digital model refers to mapping the acquired model features with the visualized digital model one by one, and data projection of the links or modeling points where the model features are located with the corresponding links or modeling points of the visualized digital model, that is, binding the model features with the corresponding links or modeling points. A modeling point is a local model in the visualized digital model, such as a certain device, a certain employee, etc.

[0056] Step 330 , obtaining operation data, performing data analysis on the operation data to obtain data features, and performing data projection on the data features and the visualized digital model.

[0057] In this step, operation data refers to the real-time production status information in the processing and production links, mainly including the operation information of the processing equipment such as status, energy consumption, speed, etc., the basic information of the production tasks such as the number of tasks, delivery time, etc., the real-time processing information of the process such as the selected processing equipment, corresponding workers, processing time, etc., and the real-time information of the operating workers.

[0058] Various sensors can be added to the physical system to collect operation data, such as fiber grating sensors, RFID, sensors, positioning devices, machine scanners and machine tool PLCs; PLCs and sensors can collect data such as equipment power, equipment operating status, equipment energy consumption, processing time, end time, speed signals, pressure signals, vibration signals, displacement signals, etc.; RFID can be used to collect data such as warehouse entry and exit, product location, processing progress, process processing quality, process operation information, workpiece arrival time, workpiece departure time, equipment maintenance, and process operators.

[0059] Data features refer to the key information extracted after completing a large amount of data sampling. In other words, data definition key information can be completed based on data modeling. Data features need to be determined based on specific data characteristics.

[0060] For example, sampling a set of running data, if the set of data is uniformly distributed, then the data characteristics include the central value and range boundaries; if the set of data is normally distributed, then the data characteristics include the standard deviation and mean; if the set of data is linearly variable, then the data characteristics include the change trend.

[0061] Optionally, the data feature type of the collected operating data is predicted based on historical experience, and the data features of the operating data are obtained through data fitting.

[0062] In this step, data features are projected with the visualized digital model, which means mapping the acquired data features with the visualized digital model one by one, and projecting the data features with the corresponding links or modeling points of the visualized digital model, that is, binding the data features with the corresponding links or modeling points. A modeling point is a local model in the visualized digital model, such as a device or an employee.

[0063] Optionally, data projection of the collected operation data with the visualized digital model refers to mapping the acquired operation data with the visualized digital model one by one, and data projection of the link or modeling point where the operation data is located with the corresponding link or modeling point of the visualized digital model, that is, binding the operation data with the corresponding link or modeling point. The modeling point is a local model in the visualized digital model, such as a certain device, a certain employee, etc.

[0064] For example, in the physical system, there is a textile drawing frame, which is 2 meters high and 3 meters wide, with a working speed of 600 to 650 meters per minute, and a weight unevenness rate within 0.4% to 0.5%. The equipment is located 3 meters to the east of the workshop, and the next process after drawing is spinning. In the 1:50 digital twin model, a textile drawing frame is constructed, which is 0.4 meters high and 0.6 meters wide, with a working speed of 600 to 650 meters per minute, and a weight unevenness rate within 0.4% to 0.5%. The equipment is located 0.6 meters to the east of the workshop, and the next process after drawing is spinning.

[0065] Optionally, the model features are modified according to the data features, and the modified model features are projected with the visualized digital model.

[0066] By collecting operation data and analyzing the operation data under a specific operation environment, data features are obtained, and the model features are corrected through the data features. In a human-machine cooperation environment, people's operating experience and proficiency often affect the actual working hours. Therefore, when establishing model features, the estimated working hours are often established as an interval variable. For example, a uniform distribution of 4.5 minutes to 5.5 minutes, or a normal distribution with a median of 10 minutes and a standard deviation. This is the average result after large-scale statistics, which is generally reflected in the operating instructions or the company's work instructions. But at the same time, according to the analysis of operation data, the specific working hours of human-machine cooperation in a specific environment can be located, and the interval of feature statistics can be further subdivided. For example, a uniform distribution of 4.55 minutes to 4.60 minutes. This is conducive to building a more accurate model.

[0067] Another example is logistics time. The estimated logistics time given by the general system is also a statistical average, especially in medium and long-distance logistics, which will be a relatively large range. However, for an accurate model, there will definitely be a statistical gap between the morning and afternoon time for medium and long-distance logistics, and the time required for logistics during peak hours in the morning and evening is also significantly different from the usual logistics time. For example, the actual transportation can obtain data features: it takes 30 hours to ship from City B to City C, and the model feature is corrected to 36 hours based on this data feature.

[0068] This embodiment can assist in establishing more accurate model features by using data features to modify model features, thereby providing a more accurate basis for rescheduling decisions.

[0069] Step 110, obtain and compare the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; the deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1.

[0070] In this step, the physical entity at time T-1 is sampled to obtain the sampling model at time T-1, and the sampling model at time T-1 corresponds to the physical entity at time T-1 one by one; the physical entity at time T is sampled to obtain the sampling model at time T, and the sampling model at time T corresponds to the physical entity at time T one by one; the sampling model at time T is reversed in the digital twin model system to obtain the deductive model at time T-1. By comparing the sampling model at time T-1 and the deductive model at time T-1, the model data of the sampling model at time T-1 and the model data of the deductive model at time T-1 can be compared. The model data may include time data and workpiece precision data. A first deviation value is obtained according to the comparison result of the two model data. The first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1.

[0071] For example, if a workpiece manufacturing workshop is rescheduled, the average manufacturing time of the physical entity workpiece at the first moment is 5 minutes, and the first moment sampling model is obtained based on the physical entity sampling at the first moment; the average manufacturing time of the physical entity workpiece at the second moment is 7 minutes, and the second moment sampling model is obtained based on the physical entity sampling at the second moment; reverse operation is performed in the digital twin model system, and the average manufacturing time of the workpiece in the deductive model at the first moment is 6 minutes. Comparing the deductive model and the sampling model at the first moment, the first deviation is obtained: the average manufacturing time of the workpiece is -1 minute.

[0072] Specifically, the model difference data includes at least one of the following: time deviation data; workpiece accuracy deviation data; quality information deviation data; sample quantity deviation data, etc.

[0073] Time deviation data refers to the difference between the time model data in the deductive model and the sampling model at the same time. Time model data refers to the data involving time in the production process, such as the average production time of workpieces, logistics time, etc.

[0074] Workpiece accuracy deviation data refers to the difference between the workpiece accuracy model data in the deductive model and the sampling model at the same time. Workpiece accuracy model data refers to the data related to accuracy in the production process.

[0075] Quality information deviation data refers to the difference between the quality information data in the deductive model and the sampling model at the same time, that is, the deviation data between the theoretical value and the actual value. Quality information data refers to the index of the one-time inspection pass rate for batch products.

[0076] Sample quantity deviation data refers to the deviation of data such as the batch size of qualified products in a batch of products in the deductive model and the sampling model at the same time.

[0077] Step 120: correct the digital twin model according to the first deviation value.

[0078] In this step, the digital twin model system at time T is corrected according to the first deviation value.

[0079] As in the above example, according to the first deviation value, the average manufacturing time for workpieces is determined to be 6 minutes, while the average manufacturing time for workpieces set in the digital twin model system at time T is 5 minutes. The average manufacturing time for workpieces in the digital twin model system at time T is corrected to 6 minutes.

[0080] Optionally, correcting the digital twin model according to the first deviation value includes: correcting the model characteristics according to the first deviation value, and performing data projection on the corrected model characteristics and the visualized digital model.

[0081] Specifically, the model features of the digital twin model are corrected by the first deviation value. The corresponding data features are obtained by analyzing the first deviation value, the model features are corrected by the data features, and the corrected model features are projected with the visualized digital model.

[0082] Step 130, correcting the rescheduling threshold based on the corrected digital twin model; the rescheduling threshold is pre-stored.

[0083] Specifically, the rescheduling threshold is corrected based on the corrected digital twin model. The rule for correcting the rescheduling threshold can be set according to expert experience, production regulations, equipment manuals, etc., and this embodiment does not limit this.

[0084] If the preset rescheduling threshold is that the average work time deviation of workpiece manufacturing is 1 minute, the average work time threshold correction rule of workpiece manufacturing is set according to expert experience as follows: rescheduling threshold = threshold before correction × average work time of workpiece manufacturing in the digital twin model after correction ÷ average work time of workpiece manufacturing in the digital twin model before correction. It is determined that the average work time of workpiece manufacturing is 6 minutes. Since the average work time of workpiece manufacturing has increased from 5 minutes to 6 minutes, the rescheduling threshold is proportionally adjusted from 1 minute to 1*6 / 5=1.2 minutes.

[0085] The rescheduling threshold is pre-stored, and the rescheduling threshold for the first moment and the second moment can be set manually; the rescheduling threshold for other moments can be set manually or corrected by the digital twin model at the previous moment. The rescheduling threshold can be stored in a historical database, and this embodiment does not limit the storage location and storage method.

[0086] Step 140, obtain and compare the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; the second deviation value is the model difference data between the sampling model at time T+1 and the deductive model at time T+1.

[0087] Specifically, the deductive model at time T+1 is obtained by simulating the operation in the digital twin model system based on the sampling model at time T. Since this step is to determine whether rescheduling is required at time T+1 and to perform rescheduling if necessary, whether or not to schedule at time T+1 is pending, and the operating state of the physical entity at time T+2 is also pending, and the sampling model at time T+2 is also pending, therefore, the deductive model of T+1 here can only be obtained through the sampling model at time T.

[0088] Specifically, the model difference data includes at least one of the following: time deviation data; workpiece accuracy deviation data; quality information deviation data; sample quantity deviation data, etc.

[0089] Time deviation data refers to the difference between the time model data in the deductive model and the sampling model at the same time. Time model data refers to the data involving time in the production process, such as the average production time of workpieces, logistics time, etc.

[0090] Workpiece accuracy deviation data refers to the difference between the workpiece accuracy model data in the deductive model and the sampling model at the same time. Workpiece accuracy model data refers to the data related to accuracy in the production process.

[0091] Quality information deviation data refers to the difference between the quality information data in the deductive model and the sampling model at the same time, that is, the deviation data between the theoretical value and the actual value. Quality information data refers to the index of the one-time inspection pass rate for batch products.

[0092] Sample quantity deviation data refers to the deviation of data such as the batch size of qualified products in a batch of products in the deductive model and the sampling model at the same time.

[0093] For example, the sampling model at time T+1 is obtained, and the average manufacturing time of each workpiece is 8 minutes. The average manufacturing time of each workpiece in the deductive model at time T+1 obtained based on the sampling model at time T is 6 minutes, and the second deviation value is 2 minutes.

[0094] Step 150: If the second deviation value does not meet the modified rescheduling threshold, rescheduling is performed according to a preset rule.

[0095] For example, the second deviation value is 2 minutes, and the revised rescheduling threshold is 1.2 minutes, which does not meet the revised rescheduling threshold, and rescheduling is performed according to the preset rules.

[0096] Optionally, if the second deviation value does not meet the modified rescheduling threshold, rescheduling is performed according to a preset rule, specifically including:

[0097] If the time deviation data does not meet the revised rescheduling threshold, rescheduling is performed according to the time priority scheduling scheme;

[0098] If the workpiece accuracy deviation data does not meet the revised rescheduling threshold, rescheduling is performed according to the accuracy priority scheduling scheme;

[0099] If the quality information deviation data does not meet the revised rescheduling threshold, rescheduling is performed according to the quality priority scheduling scheme;

[0100] If the sample quantity deviation data does not meet the corrected rescheduling threshold, rescheduling is performed according to the accuracy priority scheduling scheme;

[0101] If the comprehensive deviation data does not meet the revised rescheduling threshold, rescheduling is performed according to the comprehensive scheduling plan; the comprehensive deviation data includes at least two of the time deviation data, the workpiece accuracy deviation data, the quality information deviation data and the sample quantity deviation data.

[0102] Specifically, the preset rescheduling scheme can be set according to expert experience and production rules. For single-machine scheduling problems, it includes the WSPT method, WDSPT method, SRPT method, etc. with time priority; the EDD method with maximum completion time priority, etc. For flowshop, there is the SPT-LPT method, and for jobshop, there is the heuristic algorithm.

[0103] For example, the time-priority scheduling scheme can detect available idle devices and put them into production use.

[0104] For example, the accuracy priority solution can be to replace available production lines, detect equipment damage, etc.

[0105] For further explanation of the present invention, refer to Figure 4 , Figure 4 This is the second flow chart of building a digital twin model provided by an embodiment of the present invention:

[0106] Step 1: Understand process-related information through the physical system. At the same time, install various sensors in the physical system to collect operation data.

[0107] Step 2: Build a visual digital model for process information, especially to confirm the order and logical relationship of each process operation. Perform data analysis on the operation data and obtain data features through machine learning and other algorithms. At the same time, build model features from the physical model based on the equipment manual, historical experience, etc. It is particularly important to point out that data features will guide better model features.

[0108] Step 3: Combine data features, model features and digital models to form a complete digital twin model.

[0109] To further supplement the present invention, the following is a second embodiment of the digital twin-based rescheduling method provided by the present invention:

[0110] Step 0: At this point, we have the rescheduling decision threshold at time T, the digital twin model at time T-1, and we can deduce the digital twin model at time T through the digital twin model at time T-1.

[0111] Step 1: Sample the physical model at time T to obtain a digital twin model (digital portrait). At this time, compare the deduced value of the digital twin model at time T-1 with the sampled value at time T, and use the rescheduling value at time T for judgment. If it exceeds the threshold, rescheduling is performed according to the established rules; if it does not exceed the threshold, go to step 2.

[0112] Step 2: Based on the sampling value at time T, first trace back the digital portrait at time T-1, compare it with the digital twin model recorded at time T-1, and confirm the deviation. The model features of the digital twin model are corrected through the deviation. This is what is mentioned above as "using data features to correct model features". Based on the corrected model features, correct the rescheduling threshold at this time (which is also the rescheduling judgment threshold at time T+1).

[0113] Step 3: Sample the physical model at time T+1 to obtain the digital twin model (digital portrait) T+1. Compare the T+1 deduction value of the digital twin model at time T with the sampling value at time T+1, and use the threshold for rescheduling judgment at time T+1 to make a rescheduling decision. If it exceeds the threshold, rescheduling is performed according to the established rules; if it does not exceed the threshold, continue.

[0114] Step 4: Based on the sampling value at time T+1, first trace back the digital portrait at time T, compare it with the digital twin model recorded at time T, and confirm the deviation. The model features of the digital twin model are corrected through the deviation. This is what is mentioned above as "using data features to correct model features". Based on the corrected model features, correct the rescheduling threshold at this time (which is also the rescheduling judgment threshold at time T21).

[0115] To further supplement the present invention, the following is a third embodiment of the digital twin-based rescheduling method provided by the present invention:

[0116] Step 1: Sample the physical model at time T to obtain the digital twin model (digital portrait).

[0117] Step 2: First, trace back the digital portrait at time T-1 and compare it with the digital twin model recorded at time T-1 to confirm the deviation. The model features of the digital twin model are corrected through the deviation. This is what is mentioned above as "using data features to correct model features". Based on the corrected model features, the rescheduling threshold at this time is corrected.

[0118] Step 3: Sample the physical model at time T+1 to obtain the digital twin model (digital portrait) T+1.

[0119] Step 4: Compare the T+1 deduced value of the digital twin model at time T with the sampling value at time T+1 to make a rescheduling decision.

[0120] The digital twin-based rescheduling system provided by the present invention is described below. The digital twin-based rescheduling system described below and the digital twin-based rescheduling method described above can refer to each other.

[0121] refer to Figure 5 , Figure 5 It is one of the structural diagrams of the digital twin-based rescheduling system provided in an embodiment of the present invention. The present invention provides a digital twin-based rescheduling system, including: a first comparison module 510, a model correction module 520, a threshold correction module 530, a second comparison module 540 and a rescheduling module 550.

[0122] The first comparison module 510 is used to obtain and compare the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; the deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1;

[0123] A model correction module 520, configured to correct the digital twin model according to the first deviation value;

[0124] A threshold correction module 530 is used to correct the rescheduling threshold based on the corrected digital twin model; the rescheduling threshold is pre-stored;

[0125] A second comparison module 540 is used to obtain and compare the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; the second deviation value is the model difference data between the sampling model at time T+1 and the deductive model at time T+1;

[0126] A rescheduling module 550, configured to perform rescheduling according to a preset rule if the second deviation value does not meet the modified rescheduling threshold;

[0127] Among them, T is a positive integer greater than or equal to 2; the sampling model is a digital twin model obtained by sampling the physical entity; the deductive model is a digital twin model obtained by running or reverse running the sampling model at adjacent moments in the digital twin model.

[0128] refer to Figure 6 , Figure 6 This is the second structural diagram of the digital twin-based rescheduling system provided in an embodiment of the present invention. A rescheduling judgment is made and a rescheduling decision is made based on the revised rescheduling threshold and the collected data features; and a historical database is updated based on the rescheduling judgment.

[0129] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7As shown, the electronic device may include: a processor (processor) 710, a communication interface (Communications Interface) 720, a memory (memory) 730 and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The processor 710 can call the logic instructions in the memory 730 to execute a rescheduling method based on digital twins, which includes: obtaining and comparing the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; the deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1; correcting the digital twin model according to the first deviation value; correcting the rescheduling threshold based on the corrected digital twin model; the rescheduling threshold is pre-stored; obtaining and comparing the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; the second deviation value is the model difference data between the sampling model at time T+1 and the deductive model at time T+1; if the second deviation value does not meet the corrected rescheduling threshold, rescheduling is performed according to a preset rule; wherein T is a positive integer greater than or equal to 2; the sampling model is a digital twin model obtained by sampling a physical entity; the deductive model is a digital twin model obtained by running or reverse running in the digital twin model according to the sampling model at adjacent times.

[0130] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0131] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a digital twin-based rescheduling method provided by the above methods, the method including: obtaining and comparing the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; the deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1; correcting the digital twin model according to the first deviation value type; correcting the rescheduling threshold based on the corrected digital twin model; the rescheduling threshold is pre-stored; obtaining and comparing the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; the second deviation value is the model difference data between the sampling model at time T+1 and the deductive model at time T+1; if the second deviation value does not meet the corrected rescheduling threshold, rescheduling is performed according to the preset rules; wherein T is a positive integer greater than or equal to 2; the sampling model is a digital twin model obtained by sampling the physical entity; the deductive model is a digital twin model obtained by running or reverse running in the digital twin model according to the sampling model at adjacent times.

[0132] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the processor executes a rescheduling method based on digital twins provided by the above methods, the method comprising: obtaining and comparing the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; the deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1; correcting the digital twin model according to the first deviation value; and based on the corrected digital twin model Modify the rescheduling threshold; the rescheduling threshold is pre-stored; obtain and compare the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; the second deviation value is the model difference data between the sampling model at time T+1 and the deductive model at time T+1; if the second deviation value does not meet the revised rescheduling threshold, rescheduling is performed according to the preset rules; wherein T is a positive integer greater than or equal to 2; the sampling model is a digital twin model obtained by sampling a physical entity; the deductive model is a digital twin model obtained by running or reverse running the sampling model at adjacent moments in the digital twin model.

[0133] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0134] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A rescheduling method based on digital twins, characterized in that: include: Obtain and compare the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; The deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1; Correcting the digital twin model according to the first deviation value; Correct the rescheduling threshold based on the corrected digital twin model; The rescheduling threshold is pre-stored; Acquire and compare the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; the second deviation value is the model difference data between the sampling model at time T+1 and the deductive model at time T+1; If the second deviation value does not meet the revised rescheduling threshold, rescheduling is performed according to a preset rule; Wherein, T is a positive integer greater than or equal to 2; the sampling model is a digital twin model obtained by sampling a physical entity; the deductive model is a digital twin model obtained by running or reverse running the sampling model at adjacent moments in the digital twin model, wherein the model difference data includes at least one of the following: Time deviation data, the time deviation data represents the difference between the time model data in the deductive model and the sampling model at the same time, and the time model data refers to the data related to time in the production process; Workpiece accuracy deviation data; Quality information deviation data; Sample size biased data, The step of obtaining and comparing the sampling model at time T-1 and the deductive model at time T-1 to obtain the first deviation value also includes: constructing a digital twin model; The construction of the digital twin model includes: Acquire process information, and establish a visual digital model based on the process information; Acquiring model features, and performing data projection on the model features and the visualized digital model; The operation data is acquired, data analysis is performed on the operation data to obtain data features, and data projection is performed on the data features and the visualized digital model.

2. The digital twin-based rescheduling method according to claim 1, characterized in that: If the second deviation value does not meet the modified rescheduling threshold, rescheduling is performed according to a preset rule, specifically including: If the time deviation data does not meet the modified rescheduling threshold, rescheduling is performed according to the time priority scheduling scheme; If the workpiece accuracy deviation data does not meet the corrected rescheduling threshold, rescheduling is performed according to the accuracy priority scheduling scheme; If the quality information deviation data does not meet the revised rescheduling threshold, rescheduling is performed according to the quality priority scheduling scheme; If the sample quantity deviation data does not meet the revised rescheduling threshold, rescheduling is performed according to the accuracy priority scheduling scheme; if the comprehensive deviation data does not meet the revised rescheduling threshold, rescheduling is performed according to the comprehensive scheduling scheme; the comprehensive deviation data includes at least two of the time deviation data, the workpiece precision deviation data, the quality information deviation data and the sample quantity deviation data.

3. The digital twin-based rescheduling method according to claim 1, characterized in that: The method further comprises: The model features are modified according to the data features, and data projection is performed between the modified model features and the visualized digital model.

4. The digital twin-based rescheduling method according to claim 1 or 3, characterized in that: The correcting the digital twin model according to the first deviation value includes: The model feature is corrected according to the first deviation value, and data projection is performed between the corrected model feature and the visualized digital model.

5. A rescheduling system based on digital twins, characterized in that: include: A first comparison module is used to obtain and compare the sampling model at time T-1 and the deductive model at time T-1 to obtain a first deviation value; The deductive model at time T-1 is obtained based on the sampling model at time T; the first deviation value is the model difference data between the sampling model at time T-1 and the deductive model at time T-1; A model correction module, used to correct the digital twin model according to the first deviation value; A threshold correction module, used to correct the rescheduling threshold based on the corrected digital twin model; The rescheduling threshold is pre-stored; A second comparison module is used to obtain and compare the sampling model at time T+1 and the deductive model at time T+1 to obtain a second deviation value; The second deviation value is model difference data between the sampling model at time T+1 and the deductive model at time T+1; A rescheduling module, configured to perform rescheduling according to a preset rule if the second deviation value does not meet the revised rescheduling threshold; Wherein, T is a positive integer greater than or equal to 2; the sampling model is a digital twin model obtained by sampling a physical entity; the deductive model is a digital twin model obtained by running or reverse running the sampling model at adjacent moments in the digital twin model. The model difference data includes at least one of the following: Time deviation data, the time deviation data represents the difference between the time model data in the deductive model and the sampling model at the same time, and the time model data refers to the data related to time in the production process; Workpiece accuracy deviation data; Quality information deviation data; Sample size biased data, The step of obtaining and comparing the sampling model at time T-1 and the deductive model at time T-1 to obtain the first deviation value also includes: constructing a digital twin model; The construction of the digital twin model includes: Acquire process information, and establish a visual digital model based on the process information; Acquiring model features, and performing data projection on the model features and the visualized digital model; The operation data is acquired, data analysis is performed on the operation data to obtain data features, and data projection is performed on the data features and the visualized digital model.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the digital twin-based rescheduling method as described in any one of claims 1 to 4 are implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the digital twin-based rescheduling method as described in any one of claims 1 to 4 are implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the digital twin-based rescheduling method as described in any one of claims 1 to 4 are implemented.

Images