CN118034136A - Digital twin method and system applied to industrial research and development design - Google Patents
Digital twin method and system applied to industrial research and development design Download PDFInfo
- Publication number
- CN118034136A CN118034136A CN202410231111.7A CN202410231111A CN118034136A CN 118034136 A CN118034136 A CN 118034136A CN 202410231111 A CN202410231111 A CN 202410231111A CN 118034136 A CN118034136 A CN 118034136A
- Authority
- CN
- China
- Prior art keywords
- industrial equipment
- digital twin
- industrial
- twin model
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000013461 design Methods 0.000 title claims abstract description 28
- 238000012827 research and development Methods 0.000 title claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 41
- 238000011161 development Methods 0.000 claims description 30
- 239000011159 matrix material Substances 0.000 claims description 30
- 230000008859 change Effects 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 230000010354 integration Effects 0.000 claims description 4
- 238000001228 spectrum Methods 0.000 claims description 4
- 238000013519 translation Methods 0.000 claims description 4
- 238000012797 qualification Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 44
- 238000010586 diagram Methods 0.000 description 13
- 238000004590 computer program Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 9
- 238000003860 storage Methods 0.000 description 9
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to a digital twin method and a digital twin system applied to industrial research and development design, wherein the method comprises the following steps: acquiring operation data of industrial equipment to be detected in a research and development stage through a sensor; the operational data includes temperature, pressure and vibration data of the industrial equipment; extracting the characteristics of the operation data to obtain corresponding frequency domain characteristics and time domain characteristics; processing the frequency domain features and the time domain features to obtain a state vector of the industrial equipment; constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment; the external control signal is a signal for controlling industrial equipment to adjust the working state; calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model; and carrying out fault prediction on the industrial equipment based on the optimized digital twin model to obtain a fault prediction rate corresponding to the industrial equipment so as to evaluate the industrial equipment. The invention can improve the accuracy of industrial equipment fault prediction, and the research and development efficiency and effect.
Description
Technical Field
The present invention relates to the field of digital twinning technology, and in particular, to a digital twinning method, system, electronic device and non-transitory computer readable storage medium applied to industrial development design.
Background
In the current industrial technical background, digital twin is a technology for sensing, diagnosing and predicting physical entity objects in real time through a digital model, and combines actual measurement, simulation and data analysis to optimize and guide the full life cycle management of products in an industrial research and development design scene.
However, due to the complexity of industrial equipment and environmental diversity, digital twinning has some errors in predicting industrial equipment failure during the development phase, which may lead to a hysteresis effect on the development of industrial equipment.
Disclosure of Invention
Aiming at the technical problems in the prior art, the invention provides a digital twin method, a system, electronic equipment and a non-transitory computer readable storage medium applied to industrial research and development design, which can improve the accuracy of industrial equipment fault prediction and the research and development efficiency and effect of industrial equipment.
The technical scheme for solving the technical problems is as follows:
the invention provides a digital twin method applied to industrial development design, which comprises the following steps:
acquiring operation data of industrial equipment to be detected in a research and development stage through a sensor; the operational data includes temperature, pressure, and vibration data of the industrial equipment;
Extracting the characteristics of the operation data to obtain corresponding frequency domain characteristics and time domain characteristics;
Processing the frequency domain features and the time domain features to obtain a state vector of the industrial equipment;
Constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment; the external control signal is a signal for controlling the industrial equipment to adjust the working state;
calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model;
And carrying out fault prediction on the industrial equipment based on the optimized digital twin model to obtain a fault prediction rate corresponding to the industrial equipment so as to evaluate the industrial equipment.
Optionally, the frequency domain features are expressed as:
Wherein F (t) is the frequency domain feature; This part is used to calculate the discrete fourier transform of the first input signal x (n) on the different frequency components, representing the contribution of each frequency to the overall spectrum; This part introduces an additional term, modulated by a first parameter f k, comprising cosine components at a plurality of frequencies f k, term/> A weight representing each frequency component; |2 denotes performing an absolute value squaring operation on the result of the whole sum; /(I)The representation normalizes the total number of samples N that pass.
Alternatively, the time domain features are expressed as:
Wherein H (t) is the time domain feature for capturing changes in the original signal x (t) over different time periods; x (t) is the original signal, i.e. the second input signal which varies with time; Is a window function; /(I) For time normalization of the integration result.
Optionally, the processing the frequency domain feature and the time domain feature to obtain a state vector of the industrial device includes:
Processing the frequency domain features and the time domain features according to a mapping function to obtain an estimated value of the motion state of the industrial equipment; the estimated value is used for representing an approximate value for determining the state of the industrial equipment based on actual observation data;
Acquiring a calibration value of the motion state of the industrial equipment;
The state vector is constructed based on the estimated value and the calibration value.
Alternatively, the estimated value is expressed as:
Xest(t)＝M(F(t),H(t))；
Wherein X est (t) is the estimated value, M is the mapping function;
The state vector is expressed as:
X(t)＝Xest(t)+ΔX(t)；
wherein X (t) is the state vector and DeltaX (t) is the calibration value.
Optionally, the constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment includes:
Constructing a first matrix, a second matrix and a third matrix related to the state of the industrial equipment according to the operation data;
constructing an initial digital twin model for generating a state change rate of the industrial equipment according to the first matrix, the second matrix, the third matrix, the state vector and the external control signal;
The expression of the initial digital twin model is:
Wherein, The state change rate is the state change rate, U (t) is the external control signal, A, B and G are the first matrix, the second matrix and the third matrix, respectively, α·sin (ωt) is a sine function, α is an amplitude control parameter, ω is a frequency control parameter.
Optionally, the failure prediction rate is expressed as:
Wherein R (t) is the failure prediction rate, γ is a translation parameter, δ is a curvature control parameter, and α is the amplitude control parameter, also referred to as a scaling parameter;
after obtaining the failure prediction rate, the method further includes:
obtaining a plurality of fault prediction rates of the digital twin model on a plurality of industrial devices, and obtaining the accuracy, recall rate and precision rate of the digital twin model based on a plurality of fault prediction rates;
Constructing an objective function based on the accuracy, the recall and the precision, and determining the qualification of the digital twin model according to the objective function; the objective function is expressed as:
Wherein Objective (A, R, P) is the Objective function, A is the accuracy, R is the recall, P is the accuracy, ω1, ω2 and ω3 are the first weight, the second weight and the third weight, respectively.
Optionally, the calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model includes:
acquiring a loss function of the initial digital twin model;
according to the loss function, determining an optimal parameter set for minimizing the loss function, and optimizing and calibrating the initial digital twin model based on the optimal parameter set;
Wherein the optimal parameter set is expressed as:
wherein, theta * is the optimal parameter set, As a function of the loss in question,Is a regularization term for controlling the complexity of the initial digital twin model, λ is a first scaling parameter, argmin θ represents optimizing the parameter vector θ.
Optionally, the acquiring, by the sensor, the operation data of the industrial equipment to be detected in the development stage includes:
Deploying a temperature sensor, a pressure sensor, and a vibration sensor on the industrial device;
Acquiring raw data of the operation of the industrial equipment according to the temperature sensor, the pressure sensor and the vibration sensor;
preprocessing the original data to obtain the operation data; the preprocessing includes denoising processing, missing value processing, normalization processing, outlier processing, and time alignment processing.
The invention also provides a digital twin system for use in an industrial development design, the system comprising:
The data acquisition module is used for acquiring the operation data of the industrial equipment to be detected in the research and development stage through the sensor; the operational data includes temperature, pressure, and vibration data of the industrial equipment;
The feature extraction module is used for carrying out feature extraction on the operation data to obtain corresponding frequency domain features and time domain features;
The characteristic processing module is used for processing the frequency domain characteristics and the time domain characteristics to obtain a state vector of the industrial equipment;
The model construction module is used for constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment; the external control signal is a signal for controlling the industrial equipment to adjust the working state;
the model optimization module is used for calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model;
And the fault prediction module is used for predicting the faults of the industrial equipment based on the optimized digital twin model to obtain the corresponding fault prediction rate of the industrial equipment so as to evaluate the industrial equipment.
In addition, to achieve the above object, the present invention also proposes an electronic device including: a memory for storing a computer software program; and the processor is used for reading and executing the computer software program so as to realize the digital twin method applied to the industrial development design.
In addition, to achieve the above object, the present invention also proposes a non-transitory computer readable storage medium having stored therein a computer software program which, when executed by a processor, implements a digital twin method applied to an industrial development design as described above.
The beneficial effects of the invention are as follows:
(1) According to the invention, by adopting a complex nonlinear model and comprehensive multidimensional characteristics, efficient digital twin modeling is realized, the state evolution factors of industrial equipment are more comprehensively considered, and the simulation precision of the digital twin model is improved;
(2) According to the invention, the digital twin model is subjected to parameter calibration by introducing a complex mathematical optimization algorithm, so that the performance of the digital twin model is further improved, and the digital twin model can be better fitted with the running condition of real industrial equipment;
(3) According to the invention, a plurality of performance indexes such as accuracy, recall rate and precision rate can be comprehensively considered through the objective function, comprehensive performance evaluation can be carried out on the digital twin model by balancing the indexes, and meanwhile, the S-shaped function is introduced to adjust the precision rate, so that the model has more practical significance for predicting the faults of industrial equipment;
(4) According to the invention, by establishing a complex digital twin model, accurate prediction of the faults of the industrial equipment can be realized, including the occurrence time, the type, the severity and the like of the faults, potential problems can be found in advance, preventive maintenance measures are taken, and the influence of the faults of the equipment on production is reduced;
(5) The super parameters, the weights, the threshold values and the like of the invention are adjustable, so that the invention has stronger flexibility under different scenes and requirements, and the user can adjust according to specific conditions so as to realize better performance and adaptability.
In conclusion, the modeling precision and the prediction performance of the digital twin model of the industrial equipment are improved by comprehensively utilizing the advanced modeling technology, the mathematical optimization method and the multi-index comprehensive evaluation, and the method is beneficial to realizing more reliable operation and management of the industrial equipment.
Drawings
FIG. 1 is a schematic diagram of a digital twinning method for use in an industrial development design in accordance with the present invention;
FIG. 2 is a flow chart of a digital twinning method applied to an industrial development design provided by the invention;
FIG. 3 is a schematic diagram of a digital twin system applied to industrial development design according to the present invention;
fig. 4 is a schematic hardware structure of one possible electronic device according to the present invention;
Fig. 5 is a schematic hardware structure of a possible computer readable storage medium according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
In the description of the present invention, the term "for example" is used to mean "serving as an example, instance, or illustration. Any embodiment described as "for example" in this disclosure is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the invention. In the following description, details are set forth for purposes of explanation. It will be apparent to one of ordinary skill in the art that the present invention may be practiced without these specific details. In other instances, well-known structures and processes have not been described in detail so as not to obscure the description of the invention with unnecessary detail. Thus, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.
Referring to fig. 1, fig. 1 is a schematic diagram of a digital twin method applied to industrial development design according to the present invention. As shown in fig. 1, the terminal and the server are connected through a network, for example, a wired or wireless network connection. The terminal may include, but is not limited to, mobile terminals such as mobile phones and tablets, and fixed terminals such as computers, inquiry machines and advertising machines, where applications of various network platforms are installed. The server provides various business services for the user, including a service push server, a user recommendation server and the like.
It should be noted that, the scene diagram of the digital twin method applied to the industrial development design shown in fig. 1 is only an example, and the terminal, the server and the application scenario described in the embodiment of the present invention are for more clearly describing the technical solution of the embodiment of the present invention, and do not generate the limitation of the technical solution provided by the embodiment of the present invention, and as a person of ordinary skill in the art can know that, with the evolution of the system and the appearance of a new service scenario, the technical solution provided by the embodiment of the present invention is also applicable to similar technical problems.
Wherein the terminal may be configured to:
acquiring operation data of industrial equipment to be detected in a research and development stage through a sensor; the operational data includes temperature, pressure, and vibration data of the industrial equipment;
Extracting the characteristics of the operation data to obtain corresponding frequency domain characteristics and time domain characteristics;
Processing the frequency domain features and the time domain features to obtain a state vector of the industrial equipment;
Constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment; the external control signal is a signal for controlling the industrial equipment to adjust the working state;
calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model;
And carrying out fault prediction on the industrial equipment based on the optimized digital twin model to obtain a fault prediction rate corresponding to the industrial equipment so as to evaluate the industrial equipment.
Referring to fig. 2, a flowchart of a digital twin method applied to industrial development design of the present invention is provided, comprising the steps of:
step 201, acquiring operation data of industrial equipment to be detected in a development stage through a sensor.
The operational data may include, among other things, temperature, pressure, and vibration data of the industrial equipment. It will be appreciated that in the development stage, in order to digitally twin model an industrial plant, it is necessary to obtain real-time data concerning the operating state of the plant, which can be achieved by deploying sensors on the industrial plant, which can measure various physical quantities. Through these sensors, operational data of the industrial equipment can be obtained. Here, the operation data, which are important to pay attention to, may include temperature, pressure, and vibration of the industrial equipment, which are important indicators of the operation state of the industrial equipment, having a key influence on the performance and health state of the equipment.
Optionally, step 201 may include:
Deploying a temperature sensor, a pressure sensor, and a vibration sensor on the industrial device;
Acquiring raw data of the operation of the industrial equipment according to the temperature sensor, the pressure sensor and the vibration sensor;
preprocessing the original data to obtain the operation data; the preprocessing includes denoising processing, missing value processing, normalization processing, outlier processing, and time alignment processing.
In some embodiments, temperature, pressure, and vibration sensors may be installed on an industrial device, with the above sensors deployed to monitor and acquire parameters such as temperature, pressure, and vibration during operation of the industrial device in real time.
In some embodiments, deployed sensors may be used to collect raw data generated at the time of operation of an industrial device. This includes real-time data such as temperature, pressure and vibration measured by the sensor.
In some embodiments, the raw data collected may be subjected to a series of preprocessing steps to obtain operational data for subsequent modeling. In a specific implementation, the denoising process can filter noise in the original data so as to ensure the data quality; missing value processing may detect and process missing data that may be present, filling in or removing missing values to preserve the integrity of the data; the data acquired by different sensors can be subjected to standardization processing, so that the data have similar dimensions and distribution, and the influence of different orders of magnitude on the model is avoided; abnormal value processing can detect and process possible abnormal values, and robustness of the model is ensured; the time alignment process may ensure that the data from the different sensors remain consistent in time for efficient data analysis and modeling.
And 202, extracting the characteristics of the operation data to obtain corresponding frequency domain characteristics and time domain characteristics.
In some embodiments, key information about the operating characteristics of the system may be extracted from the operating data that has been preprocessed for use in building a digital twin model. Through the feature extraction process, a frequency domain feature F (t) and a time domain feature H (t) about the operational data are obtained, which play a key role in the digital twin model for more fully describing the dynamic behavior of the system.
In some embodiments, the frequency domain features may be expressed as:
Wherein F (t) is the frequency domain feature; This part is used to calculate the Discrete Fourier Transform (DFT) of the first input signal x (n) on the different frequency components, representing the contribution of each frequency to the overall spectrum; This part introduces an additional term, modulated by a first parameter f k, comprising cosine components at a plurality of frequencies f k, term/> A weight representing each frequency component; |2 denotes performing an absolute value squaring operation on the result of the whole sum; /(I)The representation normalizes the total number of samples N that pass.
In particular, the method comprises the steps of,Representing the contribution of each frequency to the overall spectrum. /(I)Ensuring that the result is independent of the length of the signal.
In some embodiments, the time domain features may be expressed as:
Wherein H (t) is the time domain feature for capturing changes in the original signal x (t) over different time periods; x (t) is the original signal, i.e. the second input signal which varies with time; Is a window function; /(I) For time normalization of the integration result.
In particular, the method comprises the steps of,Is a window function that adjusts the original signal at different locations in time t, where α, and σ are parameters of the window function.
The time domain feature H (t) reflects the dynamic characteristics of the signal in the time domain by weighted integration of the original signal in different time windows, provides information about the response modes of the system at different time points, and provides a more comprehensive time domain description for the digital twin model.
Before feature extraction, the physical quantity of continuous change in time of operation data such as temperature, pressure, vibration, etc. is measured. In digital signal processing, these continuously varying physical quantities can be regarded as original signals. For example, the change in temperature over time may be represented as x (t), where t is time. Also, the discretely sampled data may be represented as x (n), where n is a discrete point in time.
And 203, processing the frequency domain features and the time domain features to obtain a state vector of the industrial equipment.
Optionally, step 203 may include:
Processing the frequency domain features and the time domain features according to a mapping function to obtain an estimated value of the motion state of the industrial equipment; the estimated value is used for representing an approximate value for determining the state of the industrial equipment based on actual observation data;
Acquiring a calibration value of the motion state of the industrial equipment;
The state vector is constructed based on the estimated value and the calibration value.
In some embodiments, the frequency domain feature F (t) and the time domain feature H (t) may be processed by a mapping function M to obtain an estimated value X est (t) of the motion state of the industrial device. This step is based on the results of frequency and time domain feature extraction, which are converted by a mapping function into an estimate of the state of motion of the industrial equipment.
In some embodiments, the estimate may be expressed as:
Xest(t)＝M(F(t),H(t))；
Wherein X est (t) is the estimated value, M is the mapping function;
in some embodiments, the state vector may be represented as:
X(t)＝Xest(t)+ΔX(t)；
wherein X (t) is the state vector and DeltaX (t) is the calibration value.
In a specific implementation, the estimated value X est (t) can process the frequency domain and time domain features extracted from the actual observed data, so that the motion state of the industrial equipment is approximately represented, the estimated value is an approximate description of the state of the industrial equipment, and the system is modeled and analyzed through a digital twin model.
In some embodiments, the estimated value may be corrected by obtaining a calibration value Δx (t) of the state of motion of the industrial equipment, which may be understood as correcting the difference between the estimated value and the actual state of motion.
By the method, comprehensive estimation of the motion state of the industrial equipment can be obtained, the information of the frequency domain and the time domain features is considered, and correction is carried out through the calibration value, so that the state vector reflects the actual motion state more accurately.
Step 204, constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment;
in some embodiments, an initial digital twin model may be constructed based on the resulting state vector X (t) and external control signals of the industrial equipment, the model being a mathematical description of the behavior and performance of the industrial equipment, taking into account the information of the estimated and calibrated values in the state vector, and the effect of the external control signals on the system.
Wherein the external control signal is a signal for controlling the industrial equipment to adjust the working state. The external control signal may be a signal to control the industrial equipment to adjust the operating state, such as may be from an operator command, feedback from an automation system, or other control source. The external control signals can influence the input of the digital twin model, and by adjusting the signals, the behavior of the industrial equipment in different working states can be simulated and predicted.
Optionally, step 204 may include:
Constructing a first matrix, a second matrix and a third matrix related to the state of the industrial equipment according to the operation data;
constructing an initial digital twin model for generating a state change rate of the industrial equipment according to the first matrix, the second matrix, the third matrix, the state vector and the external control signal;
The expression of the initial digital twin model is:
Wherein, The state change rate is the state change rate, U (t) is the external control signal, A, B and G are the first matrix, the second matrix and the third matrix, respectively, α·sin (ωt) is a sine function, α is an amplitude control parameter, ω is a frequency control parameter.
It will be appreciated that, based on the operational data, a first matrix, a second matrix, and a third matrix may be constructed that are related to the status of the industrial equipment, the matrices reflecting status information contained in the operational data, the relationship between the status and the operational data being expressed by a mathematical model.
In particular, the method comprises the steps of,Representing the derivative of the state vector X (t) over time, representing the evolution speed of the system over time. A. B, G are state-dependent matrices, respectively, by which interactions between different states are represented. U (t) is an external control signal representing a control input to the industrial equipment, and may be a control command from an operator or system feedback. Alpha.sin (ωt) is a sine function part, and a vibration component is introduced, wherein alpha is an amplitude control parameter, and ω is a frequency control parameter.
It can be understood that the invention expresses the evolution process of the state of the industrial equipment along with time through the initial digital twin model, and can simulate the running state of the equipment under different working conditions by adjusting external control signals. This provides the basis for subsequent parameter calibration, optimization and fault prediction.
And 205, calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model.
Optionally, step 205 may further include:
acquiring a loss function of the initial digital twin model;
according to the loss function, determining an optimal parameter set for minimizing the loss function, and optimizing and calibrating the initial digital twin model based on the optimal parameter set;
Wherein the optimal parameter set is expressed as:
wherein, theta * is the optimal parameter set, As a function of the loss in question,Is a regularization term for controlling the complexity of the initial digital twin model, λ is a first scaling parameter, argmin θ represents optimizing the parameter vector θ.
Specifically, Y k is the actual observation data, H (X k, θ) is the predicted value of the digital twin model, θ is the parameter vector of the model, andThen, as a regularization term, used to control the complexity of the model, argmin θ represents optimizing the parameter vector θ such that the loss function reaches a minimum. By optimizing the loss function, an optimal parameter set θ * capable of minimizing the prediction error is found, and the optimization aims at making the actual observed data and the model prediction as close as possible while considering the influence of the regularization term.
In some embodiments, the initial digital twin model may be optimized and calibrated using the resulting optimal parameter set θ *. By adjusting parameters of the model, the model is better adapted to actual observation data, and the prediction performance of the model is improved. Through the mode of carrying out mathematical optimization on the initial digital twin model, the digital twin model can more accurately reflect the state change of industrial equipment, and the generalization capability of the model is improved so as to adapt to different working conditions.
And 206, carrying out fault prediction on the industrial equipment based on the optimized digital twin model to obtain a fault prediction rate corresponding to the industrial equipment so as to evaluate the industrial equipment.
In some embodiments, the failure prediction rate may be expressed as:
Where R (t) is the failure prediction rate, γ is a translation parameter, δ is a curvature control parameter, and α is the amplitude control parameter, also referred to as a scaling parameter.
Specifically, in the digital twin model, the failure prediction rate R (t) is based on the state change rateIs a probabilistic model of (a). The model expresses the possibility of industrial equipment failure at time t by processing the state change rate through an S-shaped function (Sigmoid function), and parameters in the formula are explained as follows: alpha (amplitude control parameter) is used to adjust the amplitude of the prediction rate. Beta (slope parameter) is used to control the slope of the sigmoid function, affecting the speed of change. Gamma (translation parameter) is used to translate the position of the sigmoid function on the time axis, affecting the reference level of the prediction rate. Delta (curvature control parameter) is used to adjust the curvature of the sigmoid function, affecting the curve shape of the prediction rate.
After obtaining the failure prediction rate, the method of the invention further comprises the following steps:
obtaining a plurality of fault prediction rates of the digital twin model on a plurality of industrial devices, and obtaining the accuracy, recall rate and precision rate of the digital twin model based on a plurality of fault prediction rates;
Constructing an objective function based on the accuracy, the recall and the precision, and determining the qualification of the digital twin model according to the objective function; the objective function is expressed as:
Wherein Objective (A, R, P) is the Objective function, A is the accuracy, R is the recall, P is the accuracy, ω1, ω2 and ω3 are the first weight, the second weight and the third weight, respectively.
Specifically, the accuracy is that the ratio of correct predictions is calculated by comparing the predicted value of the model with the actual observed value, specifically the number of correct predictions of the digital twin model divided by the total number of samples. The recall is used to measure the ability of the model to correctly identify faults, i.e., the proportion of faults that are successfully predicted in the actual occurrence, specifically the number of faulty industrial devices successfully predicted by the digital twinning model divided by the total number of industrial devices that are actually faulty. The accuracy rate indicates how many of the positive categories of model predictions are truly positive, specifically the number of successfully predicted failed industrial devices divided by the total number of model predicted failed industrial devices.
The objective function comprehensively considers accuracy, recall and precision and adjusts the balance between them through weights. And the optimal parameter set theta * is determined by optimizing an objective function, wherein parameters to be adjusted in the model are included. Therefore, the digital twin model can be ensured to have more comprehensive performance in the aspect of fault prediction of a plurality of industrial equipment, and the reliability of the digital twin model in practical application is improved.
Referring to fig. 3, fig. 3 is a schematic structural diagram of a digital twin system applied to industrial development and design according to the present invention.
As shown in fig. 3, a digital twin system applied to industrial development design according to an embodiment of the present invention includes:
the data acquisition module 301 is configured to acquire, by using a sensor, operation data of an industrial device to be detected in the development stage; the operational data includes temperature, pressure, and vibration data of the industrial equipment;
the feature extraction module 302 is configured to perform feature extraction on the operation data to obtain corresponding frequency domain features and time domain features;
A feature processing module 303, configured to process the frequency domain feature and the time domain feature to obtain a state vector of the industrial device;
A model building module 304 for building an initial digital twin model from the state vector and an external control signal of the industrial equipment; the external control signal is a signal for controlling the industrial equipment to adjust the working state;
The model optimization module 305 is configured to calibrate and optimize the initial digital twin model to obtain an optimized digital twin model;
And the fault prediction module 306 is configured to predict a fault of the industrial equipment based on the optimized digital twin model, and obtain a fault prediction rate corresponding to the industrial equipment, so as to evaluate the industrial equipment.
Referring to fig. 4, fig. 4 is a schematic diagram of an embodiment of an electronic device according to an embodiment of the invention. As shown in fig. 4, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420, wherein the processor 420 executes the computer program 411 to implement the following steps:
acquiring operation data of industrial equipment to be detected in a research and development stage through a sensor; the operational data includes temperature, pressure, and vibration data of the industrial equipment;
Extracting the characteristics of the operation data to obtain corresponding frequency domain characteristics and time domain characteristics;
Processing the frequency domain features and the time domain features to obtain a state vector of the industrial equipment;
Constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment; the external control signal is a signal for controlling the industrial equipment to adjust the working state;
calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model;
And carrying out fault prediction on the industrial equipment based on the optimized digital twin model to obtain a fault prediction rate corresponding to the industrial equipment so as to evaluate the industrial equipment.
Referring to fig. 5, fig. 5 is a schematic diagram of an embodiment of a computer readable storage medium according to an embodiment of the invention. As shown in fig. 5, the present embodiment provides a computer-readable storage medium 500 having stored thereon a computer program 411, which computer program 411, when executed by a processor, performs the steps of:
acquiring operation data of industrial equipment to be detected in a research and development stage through a sensor; the operational data includes temperature, pressure, and vibration data of the industrial equipment;
Extracting the characteristics of the operation data to obtain corresponding frequency domain characteristics and time domain characteristics;
Processing the frequency domain features and the time domain features to obtain a state vector of the industrial equipment;
Constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment; the external control signal is a signal for controlling the industrial equipment to adjust the working state;
calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model;
And carrying out fault prediction on the industrial equipment based on the optimized digital twin model to obtain a fault prediction rate corresponding to the industrial equipment so as to evaluate the industrial equipment.
In the foregoing embodiments, the descriptions of the embodiments are focused on, and for those portions of one embodiment that are not described in detail, reference may be made to the related descriptions of other embodiments.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a system for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (10)
1. A digital twinning method for use in an industrial development design, the method comprising:
acquiring operation data of industrial equipment to be detected in a research and development stage through a sensor; the operational data includes temperature, pressure, and vibration data of the industrial equipment;
Extracting the characteristics of the operation data to obtain corresponding frequency domain characteristics and time domain characteristics;
Processing the frequency domain features and the time domain features to obtain a state vector of the industrial equipment;
Constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment; the external control signal is a signal for controlling the industrial equipment to adjust the working state;
calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model;
And carrying out fault prediction on the industrial equipment based on the optimized digital twin model to obtain a fault prediction rate corresponding to the industrial equipment so as to evaluate the industrial equipment.
2. The digital twinning method applied to an industrial development design of claim 1, wherein the frequency domain features are represented as:
Wherein F (t) is the frequency domain feature; This part is used to calculate the discrete fourier transform of the first input signal x (n) on the different frequency components, representing the contribution of each frequency to the overall spectrum; This part introduces an additional term, modulated by a first parameter f k, comprising cosine components at a plurality of frequencies f k, term/> A weight representing each frequency component; |2 denotes performing an absolute value squaring operation on the result of the whole sum; /(I)The representation normalizes the total number of samples N that pass.
3. The digital twinning method applied to an industrial development design of claim 2, wherein the time domain features are expressed as:
Wherein H (t) is the time domain feature for capturing changes in the original signal x (t) over different time periods; x (t) is the original signal, i.e. the second input signal which varies with time; Is a window function; /(I) For time normalization of the integration result.
4. The digital twinning method for use in an industrial development design of claim 3, wherein said processing said frequency domain features and said time domain features to obtain a state vector for said industrial equipment comprises:
Processing the frequency domain features and the time domain features according to a mapping function to obtain an estimated value of the motion state of the industrial equipment; the estimated value is used for representing an approximate value for determining the state of the industrial equipment based on actual observation data;
Acquiring a calibration value of the motion state of the industrial equipment;
The state vector is constructed based on the estimated value and the calibration value.
5. The digital twinning method applied to an industrial development design of claim 4, wherein the estimated value is expressed as:
Xest(t)＝M(F(t),H(t))；
Wherein X est (t) is the estimated value, M is the mapping function;
The state vector is expressed as:
X(t)＝Xest(t)+ΔX(t)；
wherein X (t) is the state vector and DeltaX (t) is the calibration value.
6. The digital twinning method applied to an industrial development design of claim 5, wherein the constructing an initial digital twinning model from the state vector and an external control signal of the industrial equipment comprises:
Constructing a first matrix, a second matrix and a third matrix related to the state of the industrial equipment according to the operation data;
constructing an initial digital twin model for generating a state change rate of the industrial equipment according to the first matrix, the second matrix, the third matrix, the state vector and the external control signal;
The expression of the initial digital twin model is:
Wherein, The state change rate is the state change rate, U (t) is the external control signal, A, B and G are the first matrix, the second matrix and the third matrix, respectively, α·sin (ωt) is a sine function, α is an amplitude control parameter, ω is a frequency control parameter.
7. The digital twinning method applied to an industrial development design of claim 6, wherein the failure prediction rate is expressed as:
Wherein R (t) is the failure prediction rate, γ is a translation parameter, δ is a curvature control parameter, and α is the amplitude control parameter, also referred to as a scaling parameter;
after obtaining the failure prediction rate, the method further includes:
obtaining a plurality of fault prediction rates of the digital twin model on a plurality of industrial devices, and obtaining the accuracy, recall rate and precision rate of the digital twin model based on a plurality of fault prediction rates;
Constructing an objective function based on the accuracy, the recall and the precision, and determining the qualification of the digital twin model according to the objective function; the objective function is expressed as:
Wherein Objective (A, R, P) is the Objective function, A is the accuracy, R is the recall, P is the accuracy, ω1, ω2 and ω3 are the first weight, the second weight and the third weight, respectively.
8. The digital twin method applied to industrial development design according to claim 1, wherein the calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model comprises:
acquiring a loss function of the initial digital twin model;
according to the loss function, determining an optimal parameter set for minimizing the loss function, and optimizing and calibrating the initial digital twin model based on the optimal parameter set;
Wherein the optimal parameter set is expressed as:
wherein, theta * is the optimal parameter set, For the loss function,/>Is a regularization term for controlling the complexity of the initial digital twin model, λ is a first scaling parameter, argmin θ represents optimizing the parameter vector θ.
9. The digital twin method applied to industrial development design according to any one of claims 1-8, wherein the acquiring, by sensors, operational data of the industrial equipment to be detected in the development phase comprises:
Deploying a temperature sensor, a pressure sensor, and a vibration sensor on the industrial device;
Acquiring raw data of the operation of the industrial equipment according to the temperature sensor, the pressure sensor and the vibration sensor;
preprocessing the original data to obtain the operation data; the preprocessing includes denoising processing, missing value processing, normalization processing, outlier processing, and time alignment processing.
10. A digital twin system for use in an industrial development design, the system comprising:
The data acquisition module is used for acquiring the operation data of the industrial equipment to be detected in the research and development stage through the sensor; the operational data includes temperature, pressure, and vibration data of the industrial equipment;
The feature extraction module is used for carrying out feature extraction on the operation data to obtain corresponding frequency domain features and time domain features;
The characteristic processing module is used for processing the frequency domain characteristics and the time domain characteristics to obtain a state vector of the industrial equipment;
The model construction module is used for constructing an initial digital twin model according to the state vector and an external control signal of the industrial equipment; the external control signal is a signal for controlling the industrial equipment to adjust the working state;
the model optimization module is used for calibrating and optimizing the initial digital twin model to obtain an optimized digital twin model;
And the fault prediction module is used for predicting the faults of the industrial equipment based on the optimized digital twin model to obtain the corresponding fault prediction rate of the industrial equipment so as to evaluate the industrial equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410231111.7A CN118034136A (en) | 2024-02-29 | 2024-02-29 | Digital twin method and system applied to industrial research and development design |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410231111.7A CN118034136A (en) | 2024-02-29 | 2024-02-29 | Digital twin method and system applied to industrial research and development design |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118034136A true CN118034136A (en) | 2024-05-14 |
Family
ID=91002265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410231111.7A Pending CN118034136A (en) | 2024-02-29 | 2024-02-29 | Digital twin method and system applied to industrial research and development design |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118034136A (en) |
-
2024
- 2024-02-29 CN CN202410231111.7A patent/CN118034136A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110442936B (en) | Equipment fault diagnosis method, device and system based on digital twin model | |
| US7275018B2 (en) | Method of condition monitoring | |
| CN108549908B (en) | Chemical process fault detection method based on multi-sampling probability kernel principal component model | |
| CN109598052B (en) | Intelligent ammeter life cycle prediction method and device based on correlation coefficient analysis | |
| JP2010277577A (en) | Real-time scheduling of linear model for control and estimation | |
| CN112100574A (en) | Resampling-based AAKR model uncertainty calculation method and system | |
| KR101953558B1 (en) | Apparatus and Method for Fault Management of Smart Devices | |
| CN111690938A (en) | Cathode protection intelligent control method and system | |
| CN116678489B (en) | Quality control method, system, equipment and medium for force sensor | |
| CN106052847B (en) | Monitoring of a machine having a rotating machine component | |
| CN114329347B (en) | Method and device for predicting metering error of electric energy meter and storage medium | |
| CN117029968A (en) | Traffic data diagnosis method, system, storage medium and electronic equipment | |
| CN117369349B (en) | Management system of remote monitoring intelligent robot | |
| CN117332205B (en) | High-precision automatic optimization method and device for temperature compensation resistance of piezoresistor | |
| CN117216844B (en) | Bridge structure damage detection method, system and storage medium | |
| CN111259338B (en) | Component failure rate correction method and device, computer equipment and storage medium | |
| CN111079591B (en) | Bad data restoration method and system based on improved multi-scale principal component analysis | |
| CN118034136A (en) | Digital twin method and system applied to industrial research and development design | |
| EP3895014A1 (en) | Identifying faults in system data | |
| US20230204549A1 (en) | Apparatus and automated method for evaluating sensor measured values, and use of the apparatus | |
| CN114624791A (en) | Rainfall measurement method and device, computer equipment and storage medium | |
| CN117131469B (en) | Error checking method of intelligent differential pressure transmitter | |
| CN114543646B (en) | Rotor displacement signal angle estimation method, device, medium and bearing controller | |
| Chen et al. | Estimation of fractional SOC for lithium batteries based on OCV hysteretic characteristics | |
| CN117723782A (en) | Sensor fault identification positioning method and system for bridge structure health monitoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |