Digital Twin Manufacturing and Testing System for Key Components of Aerospace Equipment

Abstract

The present invention discloses a digital twin intelligent manufacturing and testing system for key components of aerospace equipment, belonging to the field of intelligent manufacturing, including: an intelligent manufacturing module, a surface detection module, a non-destructive flaw detection module, and a mechanical property detection module. The intelligent manufacturing module digitizes the manufacturing process by means of digital twin technology, facilitating the mapping of the process state during machining to subsequent detection results; the surface detection module mainly performs visual detection, uses two-dimensional or three-dimensional visual information, and combines artificial intelligence algorithms to achieve the detection of surface machining quality; the non-destructive flaw detection module detects internal defects of components with the help of internal defect flaw detection equipment; the mechanical property detection module mainly simulates the detection of the mechanical properties and life cycle of components under dynamic working conditions such as stress and temperature. The present invention realizes the digital mapping of the intelligent manufacturing process - surface characteristics - internal characteristics - mechanical properties by constructing a neural network architecture, thereby achieving the purpose of gradually replacing physical tests with digital tests.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent manufacturing, and particularly relates to a digital twin intelligent manufacturing and testing system for key components of aerospace equipment. Background Art

[0002] With the rapid development of the aerospace industry, countries have increasingly attached importance to the improvement of the digital level of aerospace equipment manufacturing and testing processes. Aerospace equipment is a complex system integrating multiple subsystems and mechanical components, with a wide variety of supporting components, complex and variable manufacturing processes, many testing links, strong process correlation, and mutual restriction. In the manufacturing and testing processes of key components of aerospace equipment, due to the large variety of components, unclear testing schemes, unclear boundary conditions, and the need for a large number of physical tests and repeated iterations in the design and manufacturing processes, the physical test cost is high and the cycle is long, which greatly restricts the improvement of the design efficiency of aerospace equipment. Therefore, the digital twin intelligent manufacturing and testing system may be one of the effective ways to solve the above problems.

[0003] Existing intelligent manufacturing and testing systems mostly rely on single testing equipment to detect the characteristics of key components, unable to cover all-round testing functions such as surface defects - internal defects - mechanical properties, and unable to apply to the testing tasks of multi-variety components. Therefore, it is difficult to meet the digital twin intelligent manufacturing and testing requirements of key components of aerospace equipment.

[0004] For example, Chinese Patent Application CN116579660A discloses a method for detecting the processing quality of aero-engine components and an interactive detection system. However, this system mainly relies on two-dimensional and three-dimensional drawings to construct a management system for digital detection, with poor coverage of the internal and mechanical properties of components; Chinese Patent CN115034147B discloses an intelligent manufacturing system based on digital twin. However, this system mainly relies on physical sensors to collect data, calculate processing errors, and then generate correction signals, with poor coverage of the types of key component detections; Chinese Patent Application CN117825521A mainly aims at problems such as low probe coverage of ultrasonic testing equipment, and designs a method of rotating workpieces on an automated production line to achieve detection.

[0005] In the design and intelligent manufacturing processes of aerospace equipment, the high cost, long cycle, and limited mode of physical tests of key components are the key problems restricting the improvement of design efficiency. However, existing manufacturing and testing systems generally have a single detection type and have not achieved the integration of multiple types and multi-platform detection modes at the equipment and digital information levels. Therefore, there is an urgent need to develop a digital twin intelligent manufacturing and testing system for key components of aerospace equipment. Summary of the Invention

[0006] To solve the problems of high test costs and long cycles caused by numerous physical tests during the design process of key components of aerospace equipment, the present invention provides a digital twin intelligent manufacturing and testing system for key components of aerospace equipment. By collecting data during the detection process through four modules: an intelligent manufacturing module, a surface detection module, a non-destructive flaw detection module, and a mechanical property detection module, a small sample data set is established. Combining with the actual data under actual working conditions, by constructing a neural network architecture, the fusion of small sample data and actual data and the weight sharing during the mapping process of each module are realized. With the help of a reinforcement learning mechanism, continuous training is carried out for weight iteration to achieve the digital mapping of the manufacturing process - surface characteristics - internal characteristics - mechanical properties, complete the overall digital twin intelligent manufacturing and testing system, and achieve the purpose of gradually replacing physical tests with digital tests. With the help of digital twin technology, the present invention achieves the purpose of gradually replacing physical tests with digital tests, reducing both the test cost and cycle in the design process of key components of aerospace equipment such as engine blisks and blades.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] A digital twin intelligent manufacturing and testing system for key components of aerospace equipment, comprising an intelligent manufacturing module, a surface detection module, a non-destructive flaw detection module, and a mechanical property detection module;

[0009] The intelligent manufacturing module digitizes the manufacturing process with the help of digital twin technology, facilitating the mapping of processing technology parameters with the detection results of the surface detection module, the non-destructive flaw detection module, and the mechanical property detection module, converging towards the real working condition direction to achieve digital prediction of the processing process;

[0010] The surface detection module mainly focuses on visual detection. Based on two-dimensional or three-dimensional visual information and combined with artificial intelligence algorithms, it realizes the detection of surface processing quality;

[0011] The non-destructive flaw detection module detects internal defects of key components with the help of internal defect flaw detection equipment;

[0012] The mechanical property detection module simulates the mechanical properties of key components under dynamic working conditions of stress and temperature and conducts life cycle detection;

[0013] Collect the processing technology parameters and geometric data of key components using the intelligent manufacturing module, collect the surface feature data of key components using the surface detection module, collect the internal damage and porosity data of key components using the non-destructive testing module, collect the mechanical property data of key components using the mechanical testing module, establish a small sample data set through the geometric data, surface feature data, internal damage, porosity data, and mechanical property data, combine with the actual data under actual working conditions, and through constructing a neural network, realize the fusion of small sample data and actual data and the weight sharing in the mapping process of the intelligent manufacturing module, surface detection module, non-destructive testing module, and mechanical testing module. With the help of the reinforcement learning mechanism, continuously train and perform weight iteration to realize the digital mapping of the manufacturing process - surface characteristics - internal characteristics - mechanical properties.

[0014] Furthermore, the key components of the aerospace equipment include blisks and blades.

[0015] Furthermore, the intelligent manufacturing module processes the key components of the aerospace equipment, collects the processing technology parameters of the components relying on the digital twin technology, completes the digital modeling of the processing process, serves as the mapping basis for digital twin manufacturing and testing, and realizes the digital prediction of the processing process.

[0016] Furthermore, the surface detection module completes the key point analysis required for modeling key components through two-dimensional or three-dimensional visual information, digitally enhances the surface features near the key points of key components based on generative AI and deep learning algorithms, and completes the detection of the processing quality of key components according to the enhanced surface features; the key points include holes and curved surfaces.

[0017] Furthermore, the non-destructive testing module detects the internal damage and porosity of key components through the digital information of eddy current or ultrasonic waves, and according to the intensity of the detection signal and the anomalies in the time domain and frequency domain of the oscillation, and at the same time scans the key areas including holes and curved surfaces according to the processing characteristics of different key components, and improves the detection efficiency in the way of adaptively adjusting the scanning density of the preset key areas and non-key areas; the non-key areas refer to regular planes.

[0018] Furthermore, the mechanical testing module simulates the dynamic working conditions of key components under real working conditions by setting dynamic working condition information, combines the small sample data set of mechanical properties collected by the mechanical testing module with the actual data under actual working conditions, fits the mechanical property degradation law of key components under specified stress and temperature working conditions, and provides a reference for adjusting the processing parameters in the processing process and inferring the results of surface and internal defect detection.

[0019] Furthermore, the digital mapping of the intelligent manufacturing process - surface characteristics - internal characteristics - mechanical properties constructs a shared neural network relying on the shared mode of multi-task learning, designs the input layer and hidden layer of the shared neural network, uses the data of the intelligent manufacturing module and surface detection module in the constructed small sample dataset as input, uses the data of the intelligent manufacturing module and surface detection module under actual working conditions as comparative input, and completes the shared learning of the input layer and hidden layer with the help of the kernel of shared learning; uses internal damage, pore data and mechanical properties as the output of the neural network, designs a loss function, continuously trains and performs weight iteration with the help of the reinforcement learning mechanism, converges the model of the digital mapping of the intelligent manufacturing process - surface characteristics - internal characteristics - mechanical properties towards the real working condition direction, and constructs a digital mapping model.

[0020] Furthermore, the non-destructive testing module includes an X-ray flaw detector, an eddy current flaw detector or an ultrasonic flaw detector.

[0021] Furthermore, the machining quality includes the quality of tool marks, surface cracks, roughness, and machining deformation.

[0022] Beneficial effects:

[0023] 1. By setting up the intelligent manufacturing module, the present invention digitizes the manufacturing process with the help of digital twin technology, meets the processing procedure requirements of multi-variety key components, constructs a digital model, replaces the traditional method of machining first and then detecting, and combines with the subsequent digital twin test model to achieve the purpose of predicting the machining quality result and adjusting the process parameters during the machining process.

[0024] 2. By setting up the surface detection module, non-destructive testing module and mechanical testing module, relying on the real data provided by specific testing equipment and the virtual data generated based on generative AI, the present invention uses artificial intelligence algorithms to realize the fusion of virtual and real data, and solves the problems of incomplete acquisition of physical test data and high cost during the testing process.

[0025] 3. By constructing a neural network architecture, relying on the small sample datasets established by the four modules during the intelligent manufacturing and testing processes, and combining with actual data, the present invention realizes the fusion of small sample data and actual data and the weight sharing during the mapping process of each module. With the help of the reinforcement learning mechanism, continuous training is carried out for weight iteration to realize the digital mapping of the intelligent manufacturing process - surface characteristics - internal characteristics - mechanical properties, which not only realizes the digital prediction or testing of mechanical properties based on intelligent manufacturing parameters and surface characteristics, but also realizes the reverse deduction of processing conditions based on the expected mechanical properties, thereby reducing the number of physical experiments to a certain extent and improving the test efficiency. Description of the Drawings

[0026] Figure 1 is the principle block diagram of the digital twin intelligent manufacturing and testing system for key components of aerospace equipment of the present invention;

[0027] Figure 2 This is the working principle diagram of the surface detection module of the present invention;

[0028] Figure 3 This is the pre-experiment data diagram of the surface detection module of the present invention. Specific implementation manners

[0029] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0030] The present invention provides a digital twin intelligent manufacturing and testing system for key components of aerospace equipment, including an intelligent manufacturing module, a surface detection module, a non-destructive flaw detection module, and a mechanical property detection module. Among them, the intelligent manufacturing module, the surface detection module, the non-destructive flaw detection module, and the mechanical property detection module have communication and data collection functions.

[0031] As Figure 1 shown, the working process of the digital twin intelligent manufacturing and testing system for key components of aerospace equipment of the present invention is as follows: First, the intelligent manufacturing module digitizes the processing process of key components of aerospace equipment by means of digital twin technology, and then uses the surface detection module, the non-destructive flaw detection module, and the mechanical property detection module, through artificial intelligence algorithms, to complete the testing process of key components of aerospace equipment. By constructing a neural network, a small sample data set established based on the data collected by the four modules is fused with the actual working condition data, and converges towards the real working condition direction, realizing the digital mapping of the manufacturing process - surface characteristics - internal characteristics - mechanical properties, and realizing the gradual replacement of physical testing with digital testing, so as to solve the problems of high cost and long cycle caused by numerous physical tests in component design. Further, the intelligent manufacturing module digitizes the processing process by means of digital twin technology, which is convenient for mapping the state of the processing process to the subsequent detection results.

[0032] Specifically, the intelligent manufacturing module processes key components of aerospace equipment, digitally processes the parameters of the components by relying on digital twin technology, and completes the digital modeling of the overall processing process through the collection of intelligent manufacturing processing data, providing a digital model basis.

[0033] Further, the surface detection module mainly performs visual detection, and realizes the detection of surface processing quality based on two-dimensional or three-dimensional visual information and in combination with artificial intelligence algorithms.

[0034] Specifically, the surface detection module digitizes and enhances the surface features near the key points of the components through two-dimensional or three-dimensional visual information, combined with the model features of the key components of aerospace equipment, based on the algorithms of generative AI and deep learning, and judges the surface defect conditions through the sparsity information of the point cloud model, so as to complete the detection of the processing quality of the components.

[0035] Further, the non-destructive testing module detects the internal defects of the components with the help of internal defect detection equipment.

[0036] Specifically, the non-destructive testing module uses internal defect detection equipment to detect the internal structure of the components through electromagnetic or acoustic signals such as eddy current or ultrasonic waves, and judges whether there is any abnormality inside according to the different response signals of the signals in the time domain or frequency domain, that is, according to the detection data analysis, so as to digitally detect the internal defects of the components such as pores and cracks. At the same time, according to the processing characteristics of different components, key areas are scanned intensively, and the detection efficiency is improved by adaptively adjusting the scanning density of the preset key area - non-key area.

[0037] Further, the mechanical property testing module mainly simulates the detection of the mechanical properties and life cycle of key components under dynamic working conditions such as stress and temperature.

[0038] Specifically, the mechanical property testing module combines conditions such as the material properties of existing key components, sets stress and temperature command information according to the mechanical properties to be tested, and applies different loads (i.e., Figure 1 the applied load in

[0039] ), simulates the dynamic mechanical, temperature and other environments of the components under real working conditions, fits the degradation law of the mechanical properties of the components under specific working conditions through big data, and then predicts the mechanical properties and life cycle of the components. Specifically, a neural network structure is constructed relying on the shared mode of multi-task learning, the input layer and hidden layer of the shared neural network are designed, the data of the manufacturing module and surface characteristic module in the constructed small sample data set are used as inputs, the data of the two modules under actual working conditions are used as comparison inputs, and with the help of the shared learning kernel, the shared learning of the input layer and hidden layer is completed, realizing the fusion of small sample data and measured data and the weight sharing in the mapping process of each module. Taking the internal characteristics and mechanical properties as the neural network output, a loss function is designed, and the weight iteration is continuously trained with the help of the reinforcement learning mechanism, so that the digital mapping model of the manufacturing process - surface characteristics - internal characteristics - mechanical properties converges towards the real working conditions. The digital mapping model constructed in this way can not only realize the digital prediction or test of mechanical properties based on manufacturing parameters and surface characteristics, but also inversely deduce the processing conditions based on the expected mechanical properties, thus achieving the purpose of replacing physical tests with digital twin tests. Figure 2As shown, the surface detection module first conducts key point analysis for modeling of key components. Based on the deep reinforcement learning robot path planning algorithm, it plans the paths when two-dimensional or three-dimensional scanning robots scan components of different types and working conditions, realizes key point analysis for modeling of key components, realizes the deep reinforcement learning robot path planning algorithm, completes the comprehensive scanning of key components, and constructs the basic digital model of the surface features of key components. Then, it uses the point cloud upsampling algorithm based on generative AI to locate the local sparse areas of the digital model, and relies on the surface defect detection algorithm for point cloud deep learning to realize the defect probability inference for key components, determine the types and location information of surface defects, and finally complete the point cloud update to confirm the defect parts. The specific algorithm is as follows: First, extract the high-frequency point cloud through the graph filter:

[0040] ;

[0041] Among them, is a high-pass filter, is the adjacency matrix of the point cloud, is the identity matrix.

[0042] ;

[0043] Among them, is the current point, is the current point 's adjacent point, is the element of the adjacency matrix A, , P is an arbitrary matrix, R represents rational numbers, is the dimension of the matrix, , i, j represent the serial numbers of points, is the extraction filtering result of the current point . The current point is a high-frequency point.

[0044] Secondly, constrain the geometric consistency between the upsampled point cloud and the real point cloud through the Earth Mover's Distance (EMD) loss function, and ensure the uniform distribution of the upsampled point cloud on the surface through the uniformity loss function.

[0045] The EMD loss function is as follows:

[0046] ;

[0047] Among them, is the upsampled point cloud, is the target point cloud, is the mapping function of the point pair, is the EMD loss function, is The point cloud in represents the Euclidean distance between the sampled points and the mapped points. represents the operation of the Euclidean distance.

[0048] The uniformity loss function is as follows:

[0049] ;

[0050] where, is a subset of points, is the global uniformity function, is the local uniformity function, is the uniformity loss function, , is the number of upsampled point clouds of . Finally, through the analysis of the EMD loss function and the uniform distribution of the point cloud, the analysis of the point cloud upsampling algorithm based on generative AI is completed relying on the Chamfer evaluation index.

[0051] The Chamfer evaluation index is as follows:

[0052] ;

[0053] where, is the evaluation index of the point cloud, is the sampled points in , is the sampled points in .

[0054] Furthermore, it is stated that the surface detection module can perform digital detection on the processing quality of parts under various different working conditions, including but not limited to quality characteristics such as tool marks, surface cracks, roughness, and processing deformation.

[0055] As Figure 3 shown are the pre-experiment data of the surface detection module. First, the depth reinforcement learning path planning algorithm is used to generate a scanning path according to the sample part features, and the robot executes this path to complete the scanning of the sample part to obtain the original point cloud data of the part surface. Then, the point cloud upsampling algorithm based on generative AI is used to complete the upsampling processing of the local sparse point cloud, and the surface flatness is calculated on the upsampled point cloud data. The pre-experiment is carried out on the same sample part five times successively (corresponding to Figure 3Flatness detection experiments were carried out on the first to fifth groups), and the percentage deviations were 0.026%, 0.208%, 0.098%, 0.094%, and 0.042% respectively. The percentage deviation here refers to the ratio of the deviation between the flatness tested in the experiment and the true flatness to the true flatness. The average percentage deviation of the five groups of experiments was 0.0936%. This result indicates that the test data of the surface detection module is highly consistent with the true experimental data, proving that the surface detection module of the present invention can be used for the detection of key components of aerospace equipment.

[0056] Specifically, the above two-dimensional or three-dimensional scanning robot completes the comprehensive scanning of the blade part by planning the scanning path of the blade part; then uses the point cloud upsampling algorithm based on generative AI to complete the positioning of the local sparse area of the digital model, and completes the detection of the hole features and surface flatness features of the blade part. Using digital detection reduces the process of scanning and detecting multiple surfaces, establishes a data sample of hole features and surface flatness, and quickly conducts comparison detection through the hole and surface feature data of different blade parts, improving the test efficiency.

[0057] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that for those skilled in the art, any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included within the protection scope of the present invention.

Claims

1. A digital twin manufacturing and testing system for key components of aerospace equipment, characterized in that, It includes an intelligent manufacturing module, a surface detection module, a non-destructive flaw detection module, and a mechanical property detection module; The intelligent manufacturing module digitizes the manufacturing process by means of digital twin technology, facilitating the mapping of processing technology parameters to the detection results of the surface detection module, the non-destructive flaw detection module, and the mechanical property detection module, converging towards the real working conditions, and realizing the digital prediction of the processing process; The surface detection module mainly uses visual detection. Based on two-dimensional or three-dimensional visual information and combined with artificial intelligence algorithms, it realizes the detection of surface processing quality; The non-destructive flaw detection module uses internal defect flaw detection equipment to detect internal defects of key components; The mechanical property detection module simulates the mechanical properties of key components under dynamic working conditions of stress and temperature, and conducts life cycle detection; Use the intelligent manufacturing module to collect processing technology parameters and geometric data of key components, use the surface detection module to collect surface feature data of key components, use the non-destructive flaw detection module to collect internal damage and pore data of key components, use the mechanical property detection module to collect mechanical property data of key components, establish a small sample data set through geometric data, surface feature data, internal damage, pore data, and mechanical property data, combine with actual data under actual working conditions, realize the fusion of small sample data and actual data and weight sharing in the mapping process of the intelligent manufacturing module, the surface detection module, the non-destructive flaw detection module, and the mechanical property detection module through constructing a neural network, and continuously train and perform weight iteration with the help of the reinforcement learning mechanism to realize the digital mapping of the manufacturing process - surface characteristics - internal characteristics - mechanical properties; The digital mapping of the manufacturing process - surface characteristics - internal characteristics - mechanical properties relies on the shared mode of multi-task learning to construct a shared neural network, design the input layer and hidden layer of the shared neural network, use the data of the intelligent manufacturing module and the surface detection module in the constructed small sample data set as input, use the data of the intelligent manufacturing module and the surface detection module under actual working conditions as comparative input, and complete the shared learning of the input layer and the hidden layer with the help of the core of shared learning; use internal damage, pore data, and mechanical properties as the output of the neural network, design a loss function, continuously train and perform weight iteration with the help of the reinforcement learning mechanism, converge the model of the digital mapping of the manufacturing process - surface characteristics - internal characteristics - mechanical properties towards the real working conditions, and construct a digital mapping model.

2. The digital twin intelligent manufacturing and testing system for key components of aerospace equipment according to claim 1, characterized in that The key components of the aerospace equipment include blisks or blades.

3. The digital twin intelligent manufacturing and testing system for key components of aerospace equipment according to claim 1, characterized in that, The intelligent manufacturing module processes the key components of the aerospace equipment, collects the processing technology parameters of the key components relying on digital twin technology, completes the digital modeling of the processing process, serves as the mapping basis for digital twin manufacturing and testing, and realizes the digital prediction of the processing process.

4. The digital twin intelligent manufacturing and testing system for key components of aerospace equipment according to claim 3, characterized in that, The surface detection module completes the key point analysis required for the modeling of key components through two-dimensional or three-dimensional visual information. Based on generative AI and deep learning algorithms, it digitally enhances the surface features near the key points of the key components, and completes the detection of the processing quality of the key components according to the digitally enhanced surface features; the key points include holes or curved surfaces.

5. The digital twin intelligent manufacturing and testing system for key components of aerospace equipment according to claim 4, characterized in that, The non-destructive testing module detects internal damages and pores of key components through eddy current or ultrasonic digital information, based on the intensity of the detection signal and anomalies in the time domain and frequency domain of the oscillation. At the same time, it scans key areas including holes and curved surfaces according to the processing characteristics of different key components, and improves the detection efficiency by presetting and adaptively adjusting the scanning density of key areas and non-key areas; the non-key areas refer to regular planes.

6. The digital twin intelligent manufacturing and testing system for key components of aerospace equipment according to claim 5, characterized in that, The mechanical testing module simulates the dynamic working conditions of key components under real working conditions by setting dynamic working condition information, combines the small sample data set of mechanical properties collected by the mechanical testing module with the actual data under actual working conditions, and fits the mechanical property degradation law of key components under specified stress and temperature conditions, providing a reference for the adjustment of processing parameters during the processing process and the inference of the results of surface and internal defect detection.

7. The digital twin intelligent manufacturing and testing system for key components of aerospace equipment according to claim 1, characterized in that The non-destructive testing module includes an X-ray flaw detector, an eddy current flaw detector or an ultrasonic flaw detector.

8. The digital twin intelligent manufacturing and testing system for key components of aerospace equipment according to claim 4, characterized in that, The machining quality includes the quality of tool marks, surface cracks, roughness and machining deformation.

Images