Fault prediction and simulation method based on digital twinning in smart factory

Abstract

The invention discloses a fault prediction and simulation method based on digital twinning in a smart factory. The fault prediction and simulation method comprises the following steps: step S1, planning a comprehensive work flow of the smart factory; s2, building a digital twinborn simulation model of the smart factory; s3, training a fault prediction model; and S4, fault prediction analysis and positioning. According to the method, the digital twinborn technology and fault prediction are combined, so that the smart factory can establish the digital twinborn simulation model reflecting the real state of the workshop on the virtual digital level based on the digital twinborn technology; the fault prediction model is used for performing virtual testing on the digital twinborn simulation modeling to predict faults possibly occurring in the working process of the smart factory, so that the working reliability and safety of the smart factory are improved, powerful support is provided for subsequent rapid maintenance of the smart factory, and the training sample of the fault prediction model is increased, so that the fault prediction efficiency is improved. Therefore, the fault prediction capability of the fault prediction model is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis methods, and specifically to a fault prediction and simulation method based on digital twins in a smart factory. Background Art

[0002] With the continuous development of science and technology, smart factories have become a new trend in the development of modern factories. Based on digital factories, smart factories use the technology of the Internet of Things and equipment monitoring technology to strengthen information management and services, clearly grasp the production and sales process, and improve the controllability of the production process; During the operation of smart factories, in order to ensure the reliability and safety of smart factories, preventive maintenance is usually required for smart factories. The general preventive maintenance method is to conduct regular inspections and fault repairs on smart factories, but this preventive maintenance method is inefficient and rapid fault detection is difficult.

[0003] The defects of existing fault prediction and simulation methods are: 1. Application document CN116719658A discloses a method for predicting faults in data center operation, but the method does not consider using digital twin technology to predict faults at the virtual digital information level; 2. The fault prediction methods in the prior art basically only use historical fault data as sample alarm data for the fault prediction method, which is not comprehensive enough, resulting in poor training effect of the fault alarm model; 3. Application document CN114330010A proposes a prediction method for digital nuclear power plant simulation operation data, but this prediction method does not help the staff to accurately locate the fault after predicting the fault; 4. The fault prediction and simulation methods in the prior art do not consider real-time updating of the fault data training samples of the fault data to expand the number of training samples of the fault prediction model. Summary of the invention

[0004] The purpose of the present invention is to provide a fault prediction and simulation method based on digital twins in a smart factory to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a fault prediction and simulation method based on digital twins in a smart factory, the fault prediction and simulation method comprising the following steps: Step S1, planning the overall workflow of the smart factory: planning the process steps of the smart factory, and determining the production equipment and equipment operating parameters at each working stage of the smart factory; Step S2, building a digital twin simulation model of the smart factory: building a three-dimensional simulation model of the smart factory based on the actual site information of the smart factory, and associating the real-time operation status of the smart factory with the three-dimensional simulation model, monitoring the operation data of the smart factory in real time, and generating a digital twin simulation model of the smart factory based on the digital twin technology; Step S3, training of the fault prediction model: inputting different categories of alarm fault data information stored in the sample database of the smart factory system into the digital twin simulation model of the smart factory respectively, and obtaining corresponding fault simulation data through batch simulation calculation, inputting different categories of alarm fault data and fault simulation data into the fault prediction model, and training the fault prediction model; Step S4, fault prediction analysis and location: input the real-time monitored smart factory operation data into the fault prediction model, and perform fault prediction on the monitored smart factory operation data. When fault simulation data is detected, the fault location is determined according to the fault data corresponding to the fault simulation data, the root cause of the fault is located, and a fault work order is issued based on the prediction results.

[0006] Preferably, in step S2, the digital twin simulation model is refined by full-stack three-dimensional visualization technology, and the factory area, workshop, production line and operating equipment of the smart factory are visualized step by step.

[0007] Preferably, in step S2, the method for obtaining the actual operating parameters of the smart factory based on the digital twin technology is: obtaining production and manufacturing information through the manufacturing execution system, obtaining equipment operating status information through sensors, obtaining equipment operating current and voltage conditions through electrical detectors, collecting workshop environment information through temperature and humidity sensors, and capturing whether the equipment is in an operating state through smart cameras.

[0008] Preferably, in step S3, the alarm fault data information stored in the sample database of the smart factory system includes historical operation abnormal data of the smart factory and manually entered abnormal data, and the alarm fault data information stored in the sample database of the smart factory system is classified according to different working stages in the working process of the smart factory.

[0009] Preferably, in step S3, data cleaning and preprocessing are performed on the fault simulation data obtained after batch simulation calculations, and the processed fault simulation data and corresponding fault alarm information are entered into the fault prediction model to train the fault prediction model.

[0010] Preferably, in step S3, the digital twin simulation model configures model parameters through a CPU simulation machine to achieve association and interaction between the test parameters entered into the model and the running simulation parameters.

[0011] Preferably, in step S4, when the fault prediction model detects fault simulation data, it determines the corresponding fault data, grades the fault according to its impact on the production process, and determines the type of fault according to the location where the fault occurs. The fault prediction model searches the database of the smart factory through data retrieval technology based on the type and level of the fault and the current operating conditions to obtain recommended solutions to the fault.

[0012] Preferably, in step S4, the content of the fault work order issued by the fault prediction model includes fault data, fault level, fault type, fault location and recommended fault solution.

[0013] Preferably, in step S4, when the fault prediction model detects a fault, the corresponding fault data is saved to the database sample of the smart factory, the database sample of the smart factory is updated, and the fault work order is fed back to the physical space to execute the corresponding solution.

[0014] Preferably, the digital twin simulation model and fault training model of the smart factory respond to the update of the data set samples of the smart factory. The digital twin simulation model of the smart factory inputs the updated fault data in the database samples of the smart factory for simulation to obtain updated fault simulation data. The fault training model enters the updated fault data and fault simulation data into the fault prediction model to update and train the fault prediction model.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention combines digital twin technology with fault prediction, so that the smart factory can establish a digital twin simulation model that can reflect the real state of the workshop at the virtual digital level based on the digital twin technology, and predict the possible faults that may occur in the working process of the smart factory through virtual testing of the digital twin simulation modeling by the fault prediction model, thereby improving the working reliability and safety of the smart factory and providing strong support for the subsequent rapid maintenance of the smart factory.

[0016] 2. The present invention not only uses the historical operation abnormal data of the smart factory as samples of fault alarm data, but also adds manually entered abnormal data as fault alarm samples in the sample database, and adds abnormal data manually entered by staff based on past experience to the sample database, making the fault sample data more comprehensive. At the same time, by inputting the fault alarm samples into the digital twin simulation model for batch simulation to obtain fault simulation data, the sample data of the fault prediction model is more complete, thereby improving the fault prediction ability of the fault prediction model.

[0017] 3. The present invention predicts faults in smart factories through fault prediction models, and after detecting fault simulation data, classifies the faults according to their impact on the production process, determines the fault type according to the location where the fault occurs, and uses data retrieval technology to search the database of the smart factory to obtain recommended solutions to the faults. Finally, the fault work order is fed back to the physical space to execute the corresponding solution, thereby helping the staff to more accurately locate the fault location more quickly, providing the staff with fault solutions, and improving the fault-solving capabilities of the smart factory.

[0018] 4. The fault prediction model of the present invention is trained with different categories of alarm fault data information and fault simulation data in the sample database, so that the fault prediction model can identify fault conditions during the operation of the smart factory, and update the fault data and fault simulation data in time after a new fault is detected, and the fault data is re-invested in the digital twin simulation model for batch simulation, thereby expanding the training data of the fault prediction model and improving the accuracy of the fault prediction model in predicting faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0020] 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.

[0021] The following combination Figure 1 , the technical solution of the present invention is further explained: The embodiment provided by the present invention is a fault prediction and simulation method based on digital twins in a smart factory, comprising the following steps: Step S1, planning the overall workflow of the smart factory: planning the process steps of the smart factory, and determining the production equipment and equipment operating parameters at each working stage of the smart factory; Step S2, building a digital twin simulation model of the smart factory: building a three-dimensional simulation model of the smart factory based on the actual site information of the smart factory, and associating the real-time operation status of the smart factory with the three-dimensional simulation model, monitoring the operation data of the smart factory in real time, and generating a digital twin simulation model of the smart factory based on the digital twin technology; Step S3, training of the fault prediction model: inputting different categories of alarm fault data information stored in the sample database of the smart factory system into the digital twin simulation model of the smart factory respectively, and obtaining corresponding fault simulation data through batch simulation calculation, inputting different categories of alarm fault data and fault simulation data into the fault prediction model, and training the fault prediction model; Step S4, fault prediction analysis and location: input the real-time monitored smart factory operation data into the fault prediction model, and perform fault prediction on the monitored smart factory operation data. When fault simulation data is detected, the fault location is determined according to the fault data corresponding to the fault simulation data, the root cause of the fault is located, and a fault work order is issued based on the prediction results.

[0022] In step S2, the digital twin simulation model is refined through full-stack three-dimensional visualization technology, and the factory area, workshop, production line and operating equipment of the smart factory are visualized step by step.

[0023] In step S2, the method for obtaining the actual operating parameters of the smart factory based on the digital twin technology is: obtaining production and manufacturing information through the manufacturing execution system, obtaining equipment operating status information through sensors, obtaining equipment operating current and voltage conditions through electrical detectors, collecting workshop environment information through temperature and humidity sensors, and capturing whether the equipment is in operation through smart cameras.

[0024] Furthermore, the digital twin simulation model combines the physical system of the actual factory with the digital model to achieve accurate modeling and simulation of the production process, which allows factory managers to optimize and improve in a digital environment to improve production efficiency and quality; Through digital twin technology, fault prediction is carried out in smart factories, enabling them to monitor equipment operating status and data in real time, quickly detect faults and diagnose them. At the same time, the fault prediction model is trained, combining machine learning and artificial intelligence algorithm technology to achieve fault prediction, take measures in advance for smart factory faults, and avoid production interruptions and losses.

[0025] In step S3, the alarm fault data information stored in the sample database of the smart factory system includes the historical operation abnormal data of the smart factory and the manually entered abnormal data. The alarm fault data information stored in the sample database of the smart factory system is classified according to different working stages in the working process of the smart factory.

[0026] In step S3, the fault simulation data obtained after batch simulation calculation is cleaned and preprocessed, the processed fault simulation data and the corresponding fault alarm information are entered into the fault prediction model, and the fault prediction model is trained.

[0027] In step S3, the digital twin simulation model configures the model parameters through the CPU simulator to achieve the association and interaction between the test parameters entered into the model and the running simulation parameters.

[0028] Furthermore, digital twin technology is used to simulate the actual operating conditions of smart factories, and the actual operating parameters of smart factories are monitored through various devices. The real-time monitored smart factory operating data is input into the fault prediction model to realize fault prediction and simulation of smart factories. The smart factory uses digital twin technology to simulate and display the overall process assembly line of the smart factory, showing the key actions of the equipment in the smart factory, the production process and the degree of production fluency. It also simulates the status of the smart factory assembly line based on 3D scenes, and simulates and displays the production working status of the smart factory.

[0029] Furthermore, the fault data is put into the digital twin simulation model for batch simulation processing, and abnormal data manually input by staff based on their work experience is added to the fault data, which helps expand the sample range of fault data and improves the ability to predict faults. By simulating fault data in batches, sufficient fault simulation data can be obtained, which makes the training effect of the fault prediction model of the smart factory excellent. There is no need to conduct a large number of fault experiments in actual working scenarios, which reduces the cost of fault prediction.

[0030] In step S4, when the fault prediction model detects the fault simulation data, it determines the corresponding fault data, classifies the fault according to its impact on the production process, and determines the fault type according to the location of the fault. The fault prediction model searches the database of the smart factory through data retrieval technology based on the type, level and current operating conditions of the fault to obtain recommended solutions to the fault.

[0031] In step S4, the content of the fault work order issued by the fault prediction model includes fault data, fault level, fault type, fault location and recommended fault solution.

[0032] In step S4, when the fault prediction model detects a fault, the corresponding fault data is saved to the database sample of the smart factory, the database sample of the smart factory is updated, and the fault work order is fed back to the physical space to execute the corresponding solution.

[0033] The digital twin simulation model and fault training model of the smart factory respond to the update of the data set samples of the smart factory. The digital twin simulation model of the smart factory inputs the updated fault data in the database samples of the smart factory for simulation and obtains updated fault simulation data. The fault training model enters the updated fault data and fault simulation data into the fault prediction model to update and train the fault prediction model.

[0034] Furthermore, the fault work order issued by the fault prediction model includes fault data, fault level, fault type, fault location and recommended fault solution. These accurate fault information enable the staff to intuitively understand the upcoming faults in the smart factory. At the same time, the recommended fault solution provided after the fault is predicted is obtained through higher education data analysis and intelligent decision-making mechanism, which can help the staff to solve the upcoming fault problems in the production site of the smart factory in a timely manner; During the operation of the smart factory, the digital twin simulation model is synchronized with the various operating parameters of the smart factory, realizing the digitization and virtualization of all elements of the factory, real-time and visualization of production management, and collaborative and intelligent equipment operation and maintenance. The continuous updating of fault data and fault simulation data can improve the accuracy of subsequent fault prediction models in predicting faults in smart factories.

[0035] Furthermore, the fault prediction and simulation method can help smart factories effectively prevent faults from occurring, and there is no need for manual regular maintenance and overhaul of the smart factories. The fault prediction and simulation method synchronizes the digital twin simulation model of the smart factory with the actual operating status of the smart factory to achieve real-time monitoring of the operating status of the smart factory, and can combine the digital twin simulation model with data visualization technology to display the complex work production processes in the smart factory in a graphical and visual manner, so that managers and staff can understand and grasp the production situation of the smart factory more intuitively, and can also analyze and simulate data to continuously improve the production process of the smart factory and improve production efficiency and quality.

[0036] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A fault prediction and simulation method based on digital twins in a smart factory, characterized in that: The fault prediction and simulation method comprises the following steps: Step S1, planning the overall workflow of the smart factory: planning the process steps of the smart factory, and determining the production equipment and equipment operating parameters at each working stage of the smart factory; Step S2, building a digital twin simulation model of the smart factory: building a three-dimensional simulation model of the smart factory based on the actual site information of the smart factory, and associating the real-time operation status of the smart factory with the three-dimensional simulation model, monitoring the operation data of the smart factory in real time, and generating a digital twin simulation model of the smart factory based on the digital twin technology; Step S3, training of the fault prediction model: inputting different categories of alarm fault data information stored in the sample database of the smart factory system into the digital twin simulation model of the smart factory respectively, and obtaining corresponding fault simulation data through batch simulation calculation, inputting different categories of alarm fault data and fault simulation data into the fault prediction model, and training the fault prediction model; Step S4, fault prediction analysis and location: input the real-time monitored smart factory operation data into the fault prediction model, and perform fault prediction on the monitored smart factory operation data. When fault simulation data is detected, the fault location is determined according to the fault data corresponding to the fault simulation data, the root cause of the fault is located, and a fault work order is issued based on the prediction results.

2. According to claim 1, the fault prediction and simulation method based on digital twin in a smart factory is characterized by: In step S2, the digital twin simulation model is refined through full-stack three-dimensional visualization technology, and the factory area, workshop, production line and operating equipment of the smart factory are visualized step by step.

3. The fault prediction and simulation method based on digital twin in a smart factory according to claim 1, characterized in that: In step S2, the method for obtaining the actual operating parameters of the smart factory based on the digital twin technology is: obtaining production and manufacturing information through the manufacturing execution system, obtaining equipment operating status information through sensors, obtaining equipment operating current and voltage conditions through electrical detectors, collecting workshop environment information through temperature and humidity sensors, and capturing whether the equipment is in operation through smart cameras.

4. The fault prediction and simulation method based on digital twins in a smart factory according to claim 1, characterized in that: In step S3, the alarm fault data information stored in the sample database of the smart factory system includes the historical operation abnormal data of the smart factory and the manually entered abnormal data. The alarm fault data information stored in the sample database of the smart factory system is classified according to different working stages in the working process of the smart factory.

5. The fault prediction and simulation method based on digital twin in a smart factory according to claim 1, characterized in that: In step S3, the fault simulation data obtained after batch simulation calculation is cleaned and preprocessed, the processed fault simulation data and the corresponding fault alarm information are entered into the fault prediction model, and the fault prediction model is trained.

6. The fault prediction and simulation method based on digital twins in a smart factory according to claim 1, characterized in that: In step S3, the digital twin simulation model configures the model parameters through the CPU simulator to achieve the association and interaction between the test parameters entered into the model and the running simulation parameters.

7. The fault prediction and simulation method based on digital twin in a smart factory according to claim 1, characterized in that: In step S4, when the fault prediction model detects the fault simulation data, it determines the corresponding fault data, classifies the fault according to its impact on the production process, and determines the fault type according to the location of the fault. The fault prediction model searches the database of the smart factory through data retrieval technology based on the type, level and current operating conditions of the fault to obtain recommended solutions to the fault.

8. The fault prediction and simulation method based on digital twins in a smart factory according to claim 7, characterized in that: In step S4, the content of the fault work order issued by the fault prediction model includes fault data, fault level, fault type, fault location and recommended fault solution.

9. The fault prediction and simulation method based on digital twin in a smart factory according to claim 1, characterized in that: In step S4, when the fault prediction model detects a fault, the corresponding fault data is saved to the database sample of the smart factory, the database sample of the smart factory is updated, and the fault work order is fed back to the physical space to execute the corresponding solution.

10. The fault prediction and simulation method based on digital twins in a smart factory according to claim 9, characterized in that: The digital twin simulation model and fault training model of the smart factory respond to the update of the data set samples of the smart factory. The digital twin simulation model of the smart factory inputs the updated fault data in the database samples of the smart factory for simulation and obtains updated fault simulation data. The fault training model enters the updated fault data and fault simulation data into the fault prediction model to update and train the fault prediction model.

Images