Intelligent manufacturing system based on digital twinning

Abstract

The invention belongs to the technical field of intelligent manufacturing, and provides an intelligent manufacturing system based on digital twinning, and the system comprises the steps: obtaining intelligent manufacturing production line data in a historical period, carrying out the operation feature analysis of a production line, and evaluating whether the intelligent manufacturing production line is in a healthy state or not; if the production line is in the non-healthy state, carrying out volatility analysis on the production efficiency of the production line in a plurality of historical periods, identifying the change rule of the production efficiency of the non-healthy production line in the plurality of historical periods, and judging whether a production efficiency fluctuation signal is generated or not; if the production efficiency fluctuation signal is generated, analyzing the dominant frequency when the production efficiency fluctuation signal is generated and the vibration frequency when the fault part vibration signal is generated through Fourier transform, and judging the coupling degree; according to the method, the fault source causing the fluctuation of the production efficiency is positioned, and passive maintenance is converted into active intervention, so that sudden shutdown is effectively avoided, and the risk of non-planned shutdown is remarkably reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent manufacturing, and specifically relates to an intelligent manufacturing system based on digital twin. Background Art

[0002] With the rapid development of industry and intelligent manufacturing technology, the complexity and intelligent level of production line equipment have been significantly improved. In the field of intelligent manufacturing, production line health management and fault prediction are the core links to ensure production continuity and efficiency.

[0003] However, the existing technologies have the following defects. On the one hand, for intelligent manufacturing production lines, the assessment of the production line status mainly relies on manual inspection, lacking the fusion analysis of multi-dimensional data such as the distribution of fault parts and production efficiency, resulting in strong subjectivity and insufficient accuracy of the health assessment results. On the other hand, the maintenance mode of the production line is mainly based on after-sales maintenance or regular maintenance, with a high risk of sudden shutdown. The fault location depends on "trial-and-error troubleshooting", and it is difficult to quickly locate the specific fault source. In addition, the existing technologies do not integrate digital twin technology, and cannot dynamically simulate the wear process of key components and predict the remaining life, resulting in the lag of maintenance decisions behind the actual deterioration state of the equipment, and the risk of sudden shutdown is difficult to effectively control.

[0004] Therefore, the present invention provides an intelligent manufacturing system based on digital twin. Summary of the Invention

[0005] In order to make up for the deficiencies of the existing technologies and solve at least one technical problem proposed in the background art.

[0006] The technical solution adopted by the present invention to solve its technical problems is: an intelligent manufacturing system based on digital twin, specifically including the following modules: Health assessment module: Obtain the data of the intelligent manufacturing production line within the historical cycle, and conduct an analysis of the production line operation characteristics to evaluate whether the intelligent manufacturing production line is in a healthy state. Among them, the production line data includes the number of production line failures, the production line fault locations, and the production line output. Efficiency fluctuation analysis module: If it is in an unhealthy state, determine the production line efficiency within the historical cycle according to the output corresponding to the unhealthy production line within the historical cycle, combined with the production duration corresponding to the production line output, and conduct a volatility analysis of the production line efficiency in multiple historical cycles to determine whether a production efficiency fluctuation signal is generated. Coupling correlation module: If a production efficiency fluctuation signal is generated, extract the production line fault locations in the corresponding unhealthy production line, analyze the dominant frequency when the production efficiency fluctuation signal is generated and the vibration frequency when the fault location vibration signal is generated through Fourier transform, and determine the level of coupling. Wear analysis module: If the coupling degree is high, through the digital twin platform, simulate the wear process of the fault part with a high coupling degree, construct a wear degradation model using a long short-term memory network, input the wear volume threshold of the key part into the wear degradation model, and determine the remaining wear period.

[0007] As a further aspect of the present invention: The specific process of evaluating whether the intelligent manufacturing production line is in a healthy state is as follows: Based on any production line, analyze the data of the intelligent manufacturing production line within the historical cycle to obtain the production line fault frequency ratio and the production line fault location ratio; Perform weighted summation processing on the production line fault frequency ratio and the production line fault location ratio to obtain a comprehensive health score; If the comprehensive health score is greater than or equal to the comprehensive health score threshold, it indicates that the intelligent manufacturing production line is in an unhealthy state.

[0008] As a further aspect of the present invention: The process of obtaining the production line fault frequency ratio is as follows: Count the number of production line faults within the historical cycle, perform difference processing with the standard value of the number of faults, and then perform ratio processing with the standard value of the number of faults to obtain the production line fault frequency ratio; As a further aspect of the present invention: The process of obtaining the production line fault location ratio is as follows: Count the number of production line fault locations within the historical cycle, perform difference processing with the standard value of the number of production line fault locations, and then perform ratio processing with the standard value of the number of production line fault locations to obtain the production line fault location ratio.

[0009] As a further aspect of the present invention: The specific process of determining whether to generate a production efficiency fluctuation signal is as follows: Record the intelligent manufacturing production line in an unhealthy state as an unhealthy production line, and calculate the production efficiency mean and production efficiency standard deviation of the current unhealthy production line through the mean and standard deviation formulas respectively; Perform ratio processing on the production efficiency standard deviation of the current unhealthy production line and the production efficiency mean to obtain the efficiency fluctuation value of the current unhealthy production line; If the efficiency fluctuation value of the current unhealthy production line is greater than or equal to the efficiency fluctuation threshold, generate a production efficiency fluctuation signal.

[0010] As a further aspect of the present invention: The specific process of determining the high or low coupling degree is as follows: Extract the production line fault locations in the production line when the production efficiency fluctuation signal is generated, collect the vibration signals of the production line fault locations, and perform Fourier transform on the production efficiency fluctuation signal and the vibration signals of the production line fault locations and perform analysis to obtain the amplitude correlation value and the frequency difference ratio; Weighted sum of the amplitude-related value and the frequency difference ratio is performed to obtain the coupling-related value; If the coupling-related value is less than the coupling-related threshold, it indicates that the coupling degree between the dominant frequency when the production efficiency fluctuation signal is generated and the vibration frequency when the vibration signal of the current production line fault location is generated is high.

[0011] As a further solution of the present invention: the process of obtaining the amplitude-related value is as follows: For the production efficiency fluctuation signal and the vibration signal of the production line fault location perform Fourier transform to obtain the frequency-domain signals and , conjugate multiply the frequency-domain signals and to calculate the cross-spectral density of the production efficiency fluctuation signal and the vibration signal; According to the cross-spectral density of the production efficiency fluctuation signal and the vibration signal, calculate the amplitude of the cross-spectral density corresponding to the current frequency; Take the absolute value after subtracting the amplitude of the cross-spectral density at the current frequency from the amplitude standard value, and then perform ratio processing with the amplitude standard value to obtain the amplitude-related value.

[0012] As a further solution of the present invention: the process of obtaining the frequency difference ratio is as follows: Obtain the frequency-domain amplitude spectrum after Fourier transform of the production efficiency fluctuation signal, and extract the frequency with the largest amplitude from it, which is the dominant frequency of the production efficiency fluctuation signal; Obtain the frequency-domain amplitude spectrum after Fourier transform of the vibration signal of the fault location, and determine the vibration frequency of the fault location; Take the absolute value after subtracting the vibration frequency of the fault location from the dominant frequency of the production efficiency, obtain the frequency difference, and perform ratio processing with the frequency difference standard value to obtain the frequency difference ratio.

[0013] As a further solution of the present invention: the process of obtaining the vibration frequency of the fault location is as follows: According to the frequency-domain amplitude spectrum after Fourier transform of the vibration signal of the fault location, extract the fault characteristic frequency through time-frequency analysis of the vibration signal, take the absolute value after subtracting it from the frequency standard value, and then perform ratio processing with the frequency standard value to obtain the frequency error ratio; If the frequency error ratio is within the frequency error ratio range, the frequency standard value corresponding to the frequency signal is the vibration frequency of the fault location; otherwise, extract the fault characteristic frequency through time-frequency analysis of the vibration signal, which is the vibration frequency of the fault location.

[0014] As a further solution of the present invention: the process of obtaining the remaining wear period is as follows:

[0015] According to Archard wear theory, the fault parts with high coupling degree are recorded as key parts, and the wear volume of the key parts is calculated. Taking the wear volume of the key parts in multiple historical periods as the training set, the long short-term memory network is trained so that the long short-term memory network model can learn the change law of the wear volume of the key parts when the production efficiency fluctuation signal is generated. The wear volume threshold of the key parts is input into the trained long short-term memory network, and the time point when the wear volume of the key parts reaches the wear volume threshold is output, and the difference is calculated with the time point when the production efficiency fluctuation signal is generated to obtain the remaining wear period.

[0016] The beneficial effects of the present invention are as follows: The present invention obtains and preprocesses the intelligent manufacturing production line data, analyzes the production line operation characteristics of the preprocessed intelligent manufacturing production line data, and evaluates whether the intelligent manufacturing production line is in a healthy state. Among them, the production line data includes the number of production line failures, the production line failure parts, and the production line output. If the intelligent manufacturing production line is in an unhealthy state, the intelligent manufacturing production line in the unhealthy state is recorded as an unhealthy production line. According to the output of the unhealthy production line in the historical period and combined with the production duration corresponding to the production line output, the production efficiency of the production line in the historical period is determined, and the volatility analysis of the production efficiency of the production line in multiple historical periods is carried out to identify the change law of the production efficiency of the unhealthy production line in multiple historical periods, and determine whether a production efficiency fluctuation signal is generated. Through multi-dimensional data fusion and intelligent analysis, the present invention realizes the refined health management of the intelligent manufacturing production line, quantitatively evaluates the health state of the production line, improves the accuracy of fault warning, quantitatively determines the production stability through the coefficient of variation, determines the hidden efficiency decline trend of the production line, traces the root cause of efficiency fluctuation, and provides data support for preventive maintenance.

[0017] The present invention generates the faulty parts in the production line when generating the production efficiency fluctuation signal. By performing Fourier transform analysis on the dominant frequency when generating the production efficiency fluctuation signal and the vibration frequency when generating the vibration signal of the faulty parts, and judging the coupling degree between the two. If the coupling degree is high, the faulty parts with high coupling degree are extracted and recorded as key parts. Through the digital twin platform, the wear process of the key parts is simulated, and a wear degradation model is constructed using a long short-term memory network. The wear volume threshold of the key parts is input into the wear degradation model, and the time point when the wear volume of the key parts reaches the wear volume threshold is output to determine the remaining wear period. Before the end of the remaining wear period, the key parts are replaced and repaired. The present invention constructs an intelligent operation and maintenance system for an intelligent manufacturing production line, conducts spectral correlation analysis on the production efficiency fluctuation and the equipment vibration characteristics, and quantitatively evaluates through cross-spectral density and frequency difference ratio, which can accurately locate the fault source causing the efficiency fluctuation, improve the fault tracing efficiency, construct a wear degradation model on the digital twin platform, can predict the remaining life of key components in advance, convert passive maintenance into active intervention, effectively avoid sudden shutdowns, significantly reduce the risk of unplanned shutdowns, improve the operation stability of the production line, and provide a full-life cycle health management solution for the intelligent manufacturing system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The present invention will be further described below with reference to the accompanying drawings.

[0019] Figure 1 is a system block diagram of an intelligent manufacturing system based on digital twin according to an embodiment of the present invention; Figure 2 is a step flow chart of an intelligent manufacturing method based on digital twin according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] In order to make the technical means, creative features, achieved purposes and effects of the present invention easy to understand, the present invention will be further described below in conjunction with specific embodiments.

[0021] Example 1, please refer to Figure 1 As shown, an intelligent manufacturing system based on digital twin according to an embodiment of the present invention includes the following modules: Health assessment module: Obtain the data of the intelligent manufacturing production line within the historical period and perform preprocessing. By analyzing the operation characteristics of the intelligent manufacturing production line on the preprocessed data of the intelligent manufacturing production line, evaluate whether the intelligent manufacturing production line is in a healthy state. Among them, the production line data includes the number of production line failures, the faulty parts of the production line, and the production line output; During the historical period, the number of failures and the failure locations of the intelligent manufacturing production line are extracted from the system logs, and the output of the intelligent manufacturing production line is obtained using fiber optic sensors. All the collected production line data is cleaned by removing outliers and filling in missing values. To eliminate the influence of different data dimensions and value ranges, the Min-Max normalization method is used to normalize the production line data, and the processed data is mapped within the range of 0 to 1; Based on any production line, the number of production line failures during the historical period is counted. After subtracting the standard value of the number of production line failures from the number of production line failures and then performing a ratio process with the standard value of the number of production line failures, the production line failure ratio is obtained; Count the number of production line failure locations during the historical period. After subtracting the standard value of the number of production line failure locations from the number of production line failure locations and then performing a ratio process with the standard value of the number of production line failure locations, the production line failure location ratio is obtained; It should be noted that the standard value of the number of production line failures and the standard value of the number of production line failure locations are set by those skilled in the art according to historical experience; Using the weighted comprehensive scoring method, the production line failure ratio and the production line failure location ratio are weighted and summed to obtain the comprehensive health score of the intelligent manufacturing production line. The weight allocation is set by those skilled in the art in combination with industry experience; In some embodiments, the comprehensive health score is compared with the comprehensive health score threshold. The specific comparison process is as follows: If the comprehensive health score is greater than or equal to the comprehensive health score threshold, it indicates that the intelligent manufacturing production line is in an unhealthy state; If the comprehensive health score is less than the comprehensive health score threshold, it indicates that the intelligent manufacturing production line is in a healthy state; Efficiency fluctuation analysis module: If the intelligent manufacturing production line is in an unhealthy state, the intelligent manufacturing production line in the unhealthy state is recorded as an unhealthy production line. Based on the output of the unhealthy production line during the historical period and combined with the production duration corresponding to the production line output, the production line production efficiency during the historical period is determined, and the volatility analysis of the production line production efficiency in multiple historical periods is performed to identify the change law of the production line production efficiency of the unhealthy production line in multiple historical periods and determine whether to generate a production efficiency fluctuation signal; Extract the production line output of all intelligent manufacturing production lines in a healthy state in multiple historical periods; Based on any unhealthy production line, according to the output of the unhealthy production line during the historical period, the production duration corresponding to the output of the unhealthy production line during the historical period is obtained through a time timer, and the production efficiency SX of the unhealthy production line during the historical period is calculated. The specific calculation formula is: ; In the formula, CL represents the output of the unhealthy production line during the historical period, Indicates the production time corresponding to the output of unhealthy production lines in the historical period; The production line production efficiency of the unhealthy production line in multiple historical cycles is integrated into a production line production efficiency data group, and the mean value of all data in the production line production efficiency data group is calculated by the mean formula to obtain the mean production efficiency of the current unhealthy production line. The standard deviation of all data in the production line production efficiency data group is calculated by the standard deviation formula to obtain the standard deviation of the production efficiency of the current unhealthy production line. The standard deviation of the production efficiency of the current unhealthy production line is compared with the mean of the production efficiency to obtain the efficiency fluctuation value of the current unhealthy production line; It should be noted that the efficiency fluctuation value is essentially the coefficient of variation in statistics, which is used to measure the degree of dispersion of data relative to the mean. The smaller the efficiency fluctuation value, the smaller the fluctuation of the production efficiency of the production line in multiple historical cycles, and the higher the stability. The larger the efficiency fluctuation value, the more drastic the fluctuation of the production efficiency of the production line, and the worse the stability. Even if the production line passes the health assessment (no significant faults), if the efficiency fluctuation value is too high, it means that the production efficiency of the production line is on a downward trend. In some embodiments, the efficiency fluctuation value of the current unhealthy production line is compared with the efficiency fluctuation threshold, and the specific comparison process is: If the efficiency fluctuation value of the current unhealthy production line is greater than or equal to the efficiency fluctuation threshold, it means that the production efficiency change of the corresponding unhealthy production line in multiple historical cycles fluctuates, and a production efficiency fluctuation signal is generated; If the efficiency fluctuation value of the current unhealthy production line is less than the efficiency fluctuation threshold, it means that the production efficiency change of the corresponding unhealthy production line in multiple historical cycles is stable, and a production efficiency stability signal is generated; The technical solution of this embodiment is: obtaining the data of the intelligent manufacturing production line in the historical period and preprocessing it, and evaluating whether the intelligent manufacturing production line is in a healthy state by analyzing the production line operation characteristics of the preprocessed intelligent manufacturing production line data, wherein the production line data includes the number of production line failures, the location of the production line failures, and the production line output. If the intelligent manufacturing production line is in an unhealthy state, the intelligent manufacturing production line in an unhealthy state is recorded as an unhealthy production line. According to the output of the unhealthy production line in the historical period, combined with the production time corresponding to the production line output, the production efficiency of the production line in the historical period is determined, and the production efficiency of the production line in multiple historical periods is analyzed for fluctuations, the production efficiency change law of the unhealthy production line in multiple historical periods is identified, and it is determined whether a production efficiency fluctuation signal is generated. The present invention realizes the refined health management of the intelligent manufacturing production line through multi-dimensional data fusion and intelligent analysis, quantitatively evaluates the health status of the production line, improves the accuracy of fault warning, quantifies production stability through the coefficient of variation, determines the implicit efficiency decline trend of the production line, traces the root cause of efficiency fluctuations, and provides data support for preventive maintenance; Example 2, please refer to Figure 1As shown in the figure, an intelligent manufacturing system based on digital twin according to an embodiment of the present invention includes the following modules: Coupling correlation module: If a production efficiency fluctuation signal is generated, the production line fault location in the production line when the production efficiency fluctuation signal is generated is determined. The dominant frequency when the production efficiency fluctuation signal is generated and the vibration frequency when the vibration signal of the fault location is generated are analyzed through Fourier transform, and the coupling degree between the two is judged; Use an acceleration sensor to collect the vibration signal of the fault location at the production line fault location; It can be understood that the sampling frequencies of the production efficiency fluctuation signal and the fault location fluctuation signal are the same, and the sampling frequency needs to meet the Nyquist criterion; Based on any fault location; For the production efficiency fluctuation signal and the vibration signal of the fault location perform Fourier transform to obtain the frequency domain signal and , specifically: , ; In the formula, f is the frequency point, j is the imaginary unit, N represents the total number of sampling points, and t = 0, 1,..., N - 1; For the frequency domain signal and perform conjugate multiplication to calculate the cross-spectral density of the production efficiency fluctuation signal and the vibration signal , and the specific calculation formula is: ; In the formula, is the conjugate of the frequency domain signal , is the phase difference between the production efficiency fluctuation signal and the vibration signal at the frequency f; Obtain the frequency domain amplitude spectrum corresponding to the production efficiency fluctuation signal after Fourier transform, and extract the frequency with the largest amplitude from it, which is the dominant frequency of the production efficiency fluctuation signal ; Obtain the frequency domain amplitude spectrum corresponding to the vibration signal of the fault location after Fourier transform, and extract the fault characteristic frequency through time-frequency analysis of the vibration signal. Then, take the absolute value of the difference from the frequency standard value and perform ratio processing with the frequency standard value to obtain the frequency error ratio; It should be noted that the frequency standard value is set by those skilled in the art through physical model calculation of the theoretical characteristic frequency; If the frequency error ratio is within the frequency error ratio range, the frequency standard value corresponding to the frequency signal is the vibration frequency of the fault location ; If the frequency error ratio is not within the frequency error ratio range, the fault characteristic frequency is extracted through the time-frequency analysis of the vibration signal, which is the vibration frequency of the fault location. ; Take the absolute value after subtracting the dominant frequency of production efficiency from the vibration frequency of the fault location to obtain the frequency difference, and then perform a ratio process on the frequency difference and the frequency difference standard value to obtain the frequency difference ratio. It should be noted that the frequency difference standard value is set by those skilled in the art through calculating the theoretical characteristic frequency with a physical model. According to the cross-spectral density of the production efficiency fluctuation signal and the vibration signal , calculate the amplitude of the cross-spectral density corresponding to the current frequency. The specific calculation formula is: ; In the formula, and respectively represent the real part and the imaginary part; Take the absolute value after subtracting the amplitude of the cross-spectral density of the current frequency from the amplitude standard value, and then perform a ratio process with the amplitude standard value to obtain the amplitude correlation value. It should be noted that the amplitude standard value is set by those skilled in the art according to historical experience. Perform a weighted sum on the amplitude correlation value and the frequency difference ratio to obtain the coupling correlation value. It can be understood that the amplitude correlation value reflects the relative magnitude of the cross-spectral density amplitude. The smaller the amplitude correlation value, the stronger the amplitude correlation between the production efficiency fluctuation signal and the vibration signal at the current frequency. The frequency difference ratio reflects the frequency consistency. The smaller the frequency difference ratio, the smaller the difference between the dominant frequency of production efficiency and the vibration frequency of the fault location, indicating that the production efficiency fluctuation is directly caused by the vibration of the fault location. In some embodiments, compare the coupling correlation value with the coupling correlation threshold. The specific comparison process is as follows: If the coupling correlation value is less than the coupling correlation threshold, it indicates a high coupling degree between the dominant frequency of production efficiency fluctuation and the current vibration frequency of the fault location; If the coupling correlation value is greater than or equal to the coupling correlation threshold, it indicates a low coupling degree between the dominant frequency of production efficiency fluctuation and the current vibration frequency of the fault location; The purpose of analyzing the coupling degree between the dominant frequency of production efficiency fluctuation and the vibration frequency of the fault location is as follows: Function 1: Establish a dynamic mapping relationship between production efficiency and the equipment operation state through frequency correlation analysis. When a fault occurs at a certain location, its vibration frequency will affect the overall operation state through mechanical conduction, resulting in fluctuations in production efficiency. By comparing whether the frequencies of the two match, it is beneficial to determine the specific fault location that causes the production efficiency fluctuation and avoid the inefficiency of the traditional "trial-and-error troubleshooting". Function 2: By analyzing the coupling degree between the dominant frequency of production efficiency fluctuations and the vibration frequency of the fault location, it is beneficial to identify early equipment faults in advance when the production efficiency has not significantly decreased, convert passive maintenance to proactive maintenance, and reduce the risk of sudden downtime; Wear analysis module: If the coupling degree is high, extract the fault locations with a high coupling degree, record them as key locations, simulate the wear process of the key locations through the digital twin platform, construct a wear degradation model using a long short-term memory network, input the wear volume threshold of the key locations into the wear degradation model, output the time point when the wear volume of the key locations reaches the wear volume threshold, determine the remaining wear period, and replace and repair the key locations before the end of the remaining wear period; According to Archard's wear theory, the specific formula for calculating the wear volume of the key location is: ; where V is the wear volume of the key location, K is the wear coefficient set according to the material properties of the key location, is the sliding friction force calculated through finite element analysis, H is the material hardness, and L is the sliding distance; On the digital twin platform, simulate the operating conditions of the key locations in the intelligent manufacturing production line when the production efficiency fluctuation signal is generated, preset the simulation operation period, divide the simulation operation period into several simulation time points with equal time intervals, run the simulation model, and output the wear volume of the key locations at the simulation time points; Compare the wear amount of the key location output at the simulation time point with the actual wear amount of the key location in the historical period, and make the output of the simulation model as close to the true value as possible by minimizing the mean square error; Construct a wear degradation model using a long short-term memory network (LSTM). Use the wear volume of the key location in multiple historical periods as the training set to train the long short-term memory network. The long short-term memory network contains 2 hidden layers, with 128 neurons in each layer. Optimize the network parameters through training to enable the long short-term memory network model to learn the change law of the wear volume of the key location when the production efficiency fluctuation signal is generated. Input the wear volume threshold of the key location into the trained long short-term memory network, output the time point when the wear volume of the key location reaches the wear volume threshold, and perform a difference operation with the time point when the production efficiency fluctuation signal is generated to obtain the remaining wear period. Replace and repair the key location before the end of the remaining wear period to avoid the production line shutdown caused by sudden failures; The technical solution of this embodiment is as follows: If a production efficiency fluctuation signal is generated, the faulty part of the production line when the production efficiency fluctuation signal is generated is determined. The dominant frequency when the production efficiency fluctuation signal is generated and the vibration frequency when the vibration signal of the faulty part is generated are analyzed through Fourier transform, and the coupling degree between the two is judged. If the coupling degree is high, the faulty part with a high coupling degree is extracted and recorded as the key part. Through the digital twin platform, the wear process of the key part is simulated, a wear degradation model is constructed using a long short-term memory network, the wear volume threshold of the key part is input into the wear degradation model, the time point when the wear volume of the key part reaches the wear volume threshold is output, the remaining wear period is determined, and the key part is replaced and repaired before the end of the remaining wear period. The present invention constructs an intelligent operation and maintenance system for an intelligent manufacturing production line, conducts spectral correlation analysis on the production efficiency fluctuation and the equipment vibration characteristics, and through cross-spectral density and frequency difference ratio quantification evaluation, can accurately locate the fault source causing the efficiency fluctuation, improve the fault tracing efficiency, construct a wear degradation model on the digital twin platform, can predict the remaining life of key components in advance, convert passive maintenance into active intervention, effectively avoid sudden shutdowns, significantly reduce the risk of unplanned shutdowns, improve the operation stability of the production line, and provide a full-life-cycle health management solution for the intelligent manufacturing system; Example 3, please refer to Figure 2 As shown in the figure, a digital-twin-based intelligent manufacturing system according to an embodiment of the present invention includes the following steps: Step 1: Obtain the intelligent manufacturing production line data within a historical period and perform preprocessing. By analyzing the production line operation characteristics of the preprocessed intelligent manufacturing production line data, evaluate whether the intelligent manufacturing production line is in a healthy state. Among them, the production line data includes the number of production line failures, the faulty parts of the production line, and the production line output; Step 2: If the intelligent manufacturing production line is in an unhealthy state, record the intelligent manufacturing production line in an unhealthy state as an unhealthy production line. According to the output of the unhealthy production line within the historical period, combined with the production duration corresponding to the production line output, determine the production efficiency of the production line within the historical period, and conduct volatility analysis on the production efficiency of the production line in multiple historical periods, identify the change law of the production efficiency of the unhealthy production line in multiple historical periods, and judge whether a production efficiency fluctuation signal is generated; Step 3: If a production efficiency fluctuation signal is generated, determine the faulty part of the production line when the production efficiency fluctuation signal is generated. Analyze the dominant frequency when the production efficiency fluctuation signal is generated and the vibration frequency when the vibration signal of the faulty part is generated through Fourier transform, and judge the coupling degree between the two; Step 4: If the coupling degree is high, extract the fault parts with a high coupling degree, denoted as key parts. Through the digital twin platform, simulate the wear process of the key parts, construct a wear degradation model using a long short-term memory network, input the wear volume threshold of the key parts into the wear degradation model, output the time point when the wear volume of the key parts reaches the wear volume threshold, determine the remaining wear period, and replace and repair the key parts before the end of the remaining wear period.

[0022] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification only illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent manufacturing system based on digital twin, characterized in that: Specifically, it includes the following modules: Health assessment module: Obtain the data of the intelligent manufacturing production line within the historical period, analyze the operation characteristics of the production line, and evaluate whether the intelligent manufacturing production line is in a healthy state. Among them, the production line data includes the number of production line failures, the failure locations of the production line, and the production output of the production line; Efficiency fluctuation analysis module: If it is in an unhealthy state, determine the production efficiency of the production line within the historical period according to the production output corresponding to the unhealthy production line within the historical period, combined with the production duration corresponding to the production output of the production line, and conduct a volatility analysis of the production efficiency of the production line in multiple historical periods to determine whether a production efficiency fluctuation signal is generated; Coupling-related module: If a production efficiency fluctuation signal is generated, extract the failure locations of the production line in the corresponding unhealthy production line, analyze the dominant frequency when the production efficiency fluctuation signal is generated and the vibration frequency when the vibration signal of the failure location is generated through Fourier transform, and determine the level of coupling; Wear analysis module: If the coupling level is high, through the digital twin platform, simulate the wear process of the failure location with a high coupling level, construct a wear degradation model using a long short-term memory network, and input the wear volume threshold of the key location into the wear degradation model to determine the remaining wear period.

2. The intelligent manufacturing system based on digital twin according to claim 1, wherein: The specific process of evaluating whether the intelligent manufacturing production line is in a healthy state is as follows: Based on any production line, analyze the data of the intelligent manufacturing production line within the historical period to obtain the production line failure frequency ratio and the production line failure location ratio; Perform a weighted summation process on the production line failure frequency ratio and the production line failure location ratio to obtain a comprehensive health score; If the comprehensive health score is greater than or equal to the comprehensive health score threshold, it means that the intelligent manufacturing production line is in an unhealthy state.

3. The intelligent manufacturing system based on digital twin according to claim 2, characterized in that: The acquisition process of the production line failure frequency ratio is as follows: Count the number of production line failures within the historical period, perform a difference process with the standard value of the failure frequency, and then perform a ratio process with the standard value of the failure frequency to obtain the production line failure frequency ratio.

4. An intelligent manufacturing system based on digital twin according to claim 2, wherein: The acquisition process of the production line failure location ratio is as follows: Count the number of production line failure locations within the historical period, perform a difference process with the standard value of the number of production line failure locations, and then perform a ratio process with the standard value of the number of production line failure locations to obtain the production line failure location ratio.

5. The intelligent manufacturing system based on digital twin according to claim 2, wherein: The specific process of determining whether a production efficiency fluctuation signal is generated is as follows: Record the intelligent manufacturing production line in an unhealthy state as an unhealthy production line, and calculate the production efficiency mean and production efficiency standard deviation of the current unhealthy production line through the mean and standard deviation formulas respectively; Perform a ratio process on the production efficiency standard deviation of the current unhealthy production line and the production efficiency mean to obtain the efficiency fluctuation value of the current unhealthy production line; If the efficiency fluctuation value of the current unhealthy production line is greater than or equal to the efficiency fluctuation threshold, a production efficiency fluctuation signal is generated.

6. The intelligent manufacturing system based on digital twin according to claim 5, characterized in that: The specific process of determining the level of coupling is as follows: Extract the production line fault location in the production line when generating the production efficiency fluctuation signal, collect the vibration signal of the production line fault location, and for the production efficiency fluctuation signal and the vibration signal of the production line fault location perform Fourier transform and analyze to obtain the amplitude correlation value and the frequency difference ratio; Perform a weighted summation on the amplitude correlation value and the frequency difference ratio to obtain the coupling correlation value; If the coupling correlation value is less than the coupling correlation threshold, it means that the dominant frequency when the production efficiency fluctuation signal is generated is highly coupled with the vibration frequency when the vibration signal of the current production line failure location is generated.

7. An intelligent manufacturing system based on digital twin according to claim 6, characterized in that: The acquisition process of the amplitude correlation value is as follows: For the production efficiency fluctuation signal and the vibration signal of the production line failure location perform Fourier transform to obtain the frequency domain signal and . For the frequency domain signal and take the conjugate multiplication to calculate the cross spectral density of the production efficiency fluctuation signal and the vibration signal; Calculate the amplitude of the cross-spectral density corresponding to the current frequency according to the cross-spectral density of the production efficiency fluctuation signal and the vibration signal; Take the absolute value after subtracting the amplitude of the current frequency cross-spectral density from the amplitude standard value, and then perform a ratio process with the amplitude standard value to obtain the amplitude correlation value.

8. An intelligent manufacturing system based on digital twin according to claim 6, characterized in that: The process of obtaining the frequency difference ratio is as follows: Obtain the frequency-domain amplitude spectrum after Fourier transform of the production efficiency fluctuation signal, and extract the frequency with the largest amplitude from it, which is the dominant frequency of the production efficiency fluctuation signal; Obtain the frequency-domain amplitude spectrum after Fourier transform of the vibration signal of the fault location, and determine the vibration frequency of the fault location; Take the absolute value after subtracting the dominant frequency of the production efficiency from the vibration frequency of the fault location to obtain the frequency difference, and perform a ratio process with the frequency difference standard value to obtain the frequency difference ratio.

9. The intelligent manufacturing system based on digital twin according to claim 8, characterized in that: The process of obtaining the vibration frequency of the fault location is as follows: According to the frequency-domain amplitude spectrum after Fourier transform of the vibration signal of the fault location, extract the fault characteristic frequency through time-frequency analysis of the vibration signal, take the absolute value after subtracting it from the frequency standard value, and then perform a ratio process with the frequency standard value to obtain the frequency error ratio; If the frequency error ratio is within the frequency error ratio range, the frequency standard value corresponding to the frequency signal is the vibration frequency of the fault location; otherwise, extract the fault characteristic frequency through time-frequency analysis of the vibration signal, which is the vibration frequency of the fault location.

10. The intelligent manufacturing system based on digital twin according to claim 6, wherein: The process of obtaining the remaining wear period is as follows: According to Archard wear theory, mark the fault location with high coupling degree as the key location, and calculate the wear volume of the key location; Take the wear volume of the key location in multiple historical cycles as the training set to train the long short-term memory network, so that the long short-term memory network model can learn the change rule of the wear volume of the key location when the production efficiency fluctuation signal is generated. Input the wear volume threshold of the key location into the trained long short-term memory network, output the time point when the wear volume of the key location reaches the wear volume threshold, and perform a difference process with the time point when the production efficiency fluctuation signal is generated to obtain the remaining wear period.

Images