Additive Manufacturing Internal Defect Detection Method and Device

Abstract

The present invention provides a method for detecting internal defects in additive manufacturing, comprising the following steps: obtaining three-dimensional point cloud data of an additive manufacturing part; generating a solid model of the additive manufacturing part according to the three-dimensional point cloud data, building a simulation algorithm of the additive manufacturing part through the solid model, and obtaining the detection orientation and detection route variables of the additive manufacturing part through the simulation algorithm of the additive manufacturing part; a laser ultrasonic system outputs laser ultrasonic signals according to the detection orientation and detection route variables of the additive manufacturing part, including a signal map sequence of different orientations of the additive manufacturing part; building a characteristic selection model of the signal map sequence, and selecting the characteristic information of the signal map sequence through the characteristic selection model; building an internal defect detection model based on additive manufacturing, training the internal defect detection model, and introducing the selected characteristic information of the signal map sequence into the trained internal defect detection model to detect internal defects of the additive manufacturing part.

Description

Technical Field

[0001] The present invention relates to the technical field of additive manufacturing, and particularly relates to a method for detecting internal defects in additive manufacturing and a device for detecting internal defects in additive manufacturing. Background Art

[0002] Additive manufacturing technology is a technology for manufacturing entities by accumulating materials. In recent years, additive manufacturing technology has made rapid progress and development, and is also considered to be one of the technologies with the best prospects in the aerospace and defense industries. However, various defects will occur during the processing and manufacturing of additive manufacturing parts, such as caking, spheroidization, internal cracks, etc., which seriously affect the performance of additive manufacturing parts. Therefore, defect detection is particularly important for reducing the hidden dangers of additive manufacturing parts and improving the safety of additive manufacturing parts.

[0003] Most traditional defect detection technologies mostly detect defects on the surface of additive manufacturing parts. For the detection of internal defects in additive manufacturing parts, due to reasons such as the inability of detection tools to accurately locate the defect position, low sensitivity, and slow detection speed, there has been no good solution. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides a method and a device for detecting internal defects in additive manufacturing, which can accurately locate the internal defect position of an additive manufacturing part, and realize non-contact non-destructive detection during the detection process, with high sensitivity and fast detection speed.

[0005] The technical solution adopted by the present invention is as follows:

[0006] A method for detecting internal defects in additive manufacturing includes the following steps: obtaining three-dimensional point cloud data of an additive manufacturing part; generating a three-dimensional model of the additive manufacturing part according to the three-dimensional point cloud data, building a simulation algorithm of the additive manufacturing part through the three-dimensional model, and obtaining the detection orientation and detection route variables of the additive manufacturing part through the simulation algorithm of the additive manufacturing part; a laser ultrasonic system outputs laser ultrasonic signals through the detection orientation and detection route variables of the additive manufacturing part, wherein the laser ultrasonic signals include a signal map sequence of different orientations of the additive manufacturing part; building a characteristic selection model of the signal map sequence, and selecting characteristic information of the signal map sequence through the characteristic selection model; building an internal defect detection model based on additive manufacturing, training the internal defect detection model, and introducing the selected characteristic information of the signal map sequence into the trained internal defect detection model to detect internal defects of the additive manufacturing part.

[0007] In an embodiment of the present invention, the characteristic selection model for building a signal graph sequence specifically includes: collecting a first picture and a second picture of any position picture in the signal graph sequence, wherein both the first picture and the second picture are obtained through more than one processing scheme; respectively performing characteristic selection on the first picture, the second picture, and the remaining pictures to obtain the first picture characteristics of the first picture, the second picture characteristics of the second picture, and the picture characteristics of the remaining pictures, wherein the first picture characteristics are obtained according to a preset first picture characteristic selection algorithm; editing the first picture characteristics through the first picture to obtain an edited picture of the first picture and an editing loss of the first picture characteristics, wherein the editing loss is the characteristic difference value between the edited picture and the first picture; obtaining a comparison loss through the second picture, the second picture characteristics, and the picture characteristics of the remaining pictures, wherein the comparison loss is the difference value between the characteristic distance between the first picture characteristics and the second picture characteristics and the characteristic distance between the first picture characteristics and the picture characteristics of the remaining pictures; updating the first picture characteristic selection algorithm according to the editing loss and the comparison loss to obtain the characteristic selection model.

[0008] In an embodiment of the present invention, the method for selecting the characteristic information of a signal graph sequence through the characteristic selection model specifically includes: obtaining the unreconstructed pictures in the signal graph sequence and transmitting the unreconstructed pictures to the characteristic selection model to obtain the source picture characteristics of the unreconstructed pictures; reconstructing the source picture characteristics through picture transformation to obtain the characteristic information of the unreconstructed pictures.

[0009] In an embodiment of the present invention, the additive manufacturing internal defect detection method further includes: if there are defects, obtaining the orientation and size of the internal defects of the additive manufacturing part in the solid model through the solid model of the additive manufacturing part and the signal graph sequences in different orientations.

[0010] In an embodiment of the present invention, the method for building an internal defect detection model based on additive manufacturing and training the internal defect detection model specifically includes: obtaining an internal defect picture data set of the additive manufacturing part and dividing the internal defect picture data set into a training set and a test set; presetting a randomly generated neural network model and performing simulation training on the randomly generated neural network model according to the random maximum likelihood algorithm; sequentially inputting the internal defect pictures in the training set into the randomly generated neural network model to obtain the high-dimensional characteristics of the internal defect pictures in the training set in multiple dimensions; using the obtained high-dimensional characteristics as the input of the training set and using the internal defect types mapped by the high-dimensional characteristics as the output of the training set.

[0011] An internal defect detection device for additive manufacturing, comprising: a data acquisition unit configured to acquire three-dimensional point cloud data of an additive manufacturing part; a model building unit configured to generate a three-dimensional model of the additive manufacturing part based on the three-dimensional point cloud data, build a simulation algorithm of the additive manufacturing part through the three-dimensional model, and obtain a detection orientation and a detection route variable of the additive manufacturing part through the simulation algorithm of the additive manufacturing part; a signal output unit configured to output a laser ultrasonic signal according to a laser ultrasonic system through the detection orientation and the detection route variable of the additive manufacturing part, wherein the laser ultrasonic signal includes a signal map sequence of different orientations of the additive manufacturing part; a characteristic selection unit configured to build a characteristic selection model of the signal map sequence and select characteristic information of the signal map sequence through the characteristic selection model; and a defect detection unit configured to build an internal defect detection model based on additive manufacturing, train the internal defect detection model, and introduce the selected characteristic information of the signal map sequence into the trained internal defect detection model to perform internal defect detection on the additive manufacturing part.

[0012] In an embodiment of the present invention, the characteristic selection unit is further configured to: collect a first picture and a second picture of any position picture in the signal map sequence, wherein both the first picture and the second picture are obtained through more than one processing scheme; respectively perform characteristic selection on the first picture, the second picture, and the remaining pictures to obtain a first picture characteristic of the first picture, a second picture characteristic of the second picture, and a picture characteristic of the remaining pictures, wherein the first picture characteristic is obtained according to a preset first picture characteristic selection algorithm; edit the first picture characteristic through the first picture to obtain an edited picture of the first picture and an editing loss of the first picture characteristic, wherein the editing loss is a characteristic difference value between the edited picture and the first picture; obtain a comparison loss through the second picture, the second picture characteristic, and the picture characteristics of the remaining pictures, wherein the comparison loss is a difference value between a characteristic distance between the first picture characteristic and the second picture characteristic and a characteristic distance between the first picture characteristic and the picture characteristics of the remaining pictures; and update the first picture characteristic selection algorithm according to the editing loss and the comparison loss to obtain the characteristic selection model.

[0013] In an embodiment of the present invention, the characteristic selection unit is further configured to: obtain an unreconstructed picture in the signal map sequence and transmit the unreconstructed picture to the characteristic selection model to obtain a source picture characteristic of the unreconstructed picture; and reconstruct the source picture characteristic through picture transformation to obtain characteristic information of the unreconstructed picture.

[0014] In an embodiment of the present invention, the additive manufacturing internal defect detection device further includes: a defect acquisition unit, which is used to obtain the orientation and size of the internal defects of the additive manufacturing part in the solid model through the solid model of the additive manufacturing part and a sequence of signal maps in different orientations.

[0015] In an embodiment of the present invention, the defect detection unit is further used to: obtain a dataset of internal defect pictures of the additive manufacturing part, and divide the dataset of internal defect pictures into a training set and a test set; preset a randomly generated neural network model, and perform simulation training on the randomly generated neural network model according to the random maximum likelihood algorithm; by sequentially inputting the internal defect pictures in the training set into the randomly generated neural network model, to obtain the high-dimensional characteristics of the internal defect pictures in the training set in multiple dimensions; use the obtained high-dimensional characteristics as the input of the training set, and use the types of internal defects mapped by the high-dimensional characteristics as the output of the training set.

[0016] Advantages of the present invention:

[0017] The present invention detects the three-dimensional point cloud data of the additive manufacturing part by detecting laser ultrasonic emission, a laser ultrasonic interferometer, and pulsed excitation laser ultrasound, then generates a solid model of the additive manufacturing part according to the three-dimensional point cloud data, constructs a simulation algorithm through the solid model, and obtains the detection orientation and detection route variables of the additive manufacturing part through the simulation algorithm, and outputs a laser ultrasonic signal through the detection orientation and detection route variables of the additive manufacturing part, and selects the characteristic information of the signal map sequence through the laser ultrasonic signal, and finally constructs and trains an internal defect detection model, and introduces the selected characteristic information into the trained internal defect detection model to detect the internal defects of the additive manufacturing part. Thus, it can accurately locate the internal defect position of the additive manufacturing part, and achieve non-contact non-destructive detection during the detection process, with high sensitivity and fast detection speed. Description of the Drawings

[0018] Figure 1 It is a flowchart of the additive manufacturing internal defect detection method according to an embodiment of the present invention;

[0019] Figure 2 It is a block diagram of the additive manufacturing internal defect detection device according to an embodiment of the present invention. Detailed Embodiments

[0020] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] Figure 1 It is a flowchart of the method for detecting internal defects in additive manufacturing according to the embodiments of the present invention.

[0022] As Figure 1 shown, the method for detecting internal defects in additive manufacturing according to the embodiments of the present invention includes the following steps:

[0023] S1. Obtain the three-dimensional point cloud data of the additive manufacturing part.

[0024] In an embodiment of the present invention, the three-dimensional point cloud data of the additive manufacturing part can be obtained through an RGB-D camera, and the three-dimensional coordinates inside the additive manufacturing part can be calculated through the three-dimensional point cloud data.

[0025] S2. Generate a three-dimensional model of the additive manufacturing part according to the three-dimensional point cloud data, build a simulation algorithm for the additive manufacturing part through the three-dimensional model, and obtain the detection orientation and detection route variables of the additive manufacturing part through the simulation algorithm of the additive manufacturing part.

[0026] Specifically, first, a simulation algorithm for additive manufacturing can be built through the three-dimensional model of the additive manufacturing to be detected. By simulating the reaction of the refraction of lasers with various laser intensities, incident directions, and discrete angles on the additive manufacturing part, the attenuation degree of the laser and the corresponding associations in different dimensions of the laser intensity, incident direction, and discrete angle can be obtained. Then, the signal interference-to-noise ratio, gray-scale difference value, and recognition rate of the generated image of the reflected laser ultrasonic signal can be set as the improvement targets for enhancing the image clarity, and the corresponding associations in different dimensions of the laser attenuation degree and the laser intensity, incident direction, and discrete angle can be converted into the boundary factors of several targets to build the improvement algorithms for multiple targets. Secondly, the best combination of the laser attenuation degree and the laser intensity, incident direction, and discrete angle can be extracted to obtain a more comprehensive detection orientation. Finally, due to the possible undetected areas and the possibility of repeated detection, and at the same time, the laser intensity may weaken inside the additive manufacturing part, the detection orientation and detection route variables of the additive manufacturing part can be obtained by improving the emission position and detection route of the laser emission source.

[0027] S3. The laser ultrasonic system outputs laser ultrasonic signals through the detection orientation and detection route variables of the additive manufacturing part, where the laser ultrasonic signals include a sequence of signal maps in different orientations of the additive manufacturing part.

[0028] In one embodiment of the present invention, by obtaining the optimal detection orientation and detection route variables of the additive manufacturing part, a signal map sequence converted from laser ultrasonic signals in different orientations can be obtained. Thus, the internal structure of the additive manufacturing part can be collected in all directions.

[0029] S4. Build a feature selection model for the signal map sequence, and select the feature information of the signal map sequence through the feature selection model.

[0030] In one embodiment of the present invention, building a feature selection model for the signal map sequence may specifically include: collecting a first picture and a second picture of any position picture in the signal map sequence, where both the first picture and the second picture are obtained through more than one processing scheme; respectively performing feature selection on the first picture, the second picture, and the remaining pictures to obtain the first picture feature of the first picture, the second picture feature of the second picture, and the picture features of the remaining pictures, where the first picture feature is obtained according to a preset first picture feature selection algorithm; editing the first picture feature with the first picture to obtain an edited picture of the first picture and an editing loss of the first picture feature, where the editing loss is the feature difference value between the edited picture and the first picture; for the second picture, the second picture feature, and the picture features of the remaining pictures, obtaining a comparison loss, where the comparison loss is the difference value between the feature distance between the first picture feature and the second picture feature and the feature distance between the first picture feature and the picture features of the remaining pictures; updating the first picture feature selection algorithm according to the editing loss and the comparison loss to obtain the feature selection model.

[0031] Specifically, two feature selection algorithms with the same structure can be used to select the first picture feature, the second picture feature, and the picture features of the remaining pictures for the first picture, the second picture, and the remaining pictures. The detailed extraction process can select the first picture feature through the first picture feature selection algorithm, select the second picture feature and the picture features of the remaining pictures through the second picture selection algorithm, where the variable of the second picture selection algorithm can be obtained by performing a moving average operation on the variable of the first picture selection algorithm.

[0032] Further, after obtaining the first picture feature of the first picture, a decoder can be used to edit the first picture feature to obtain an edited picture of the first picture, and a cost function can be used to calculate the loss between the edited picture and the first picture, which is marked as the editing loss.

[0033] Further, the features obtained by extracting the same picture can be arranged into an upsampling group, and the features obtained by extracting different pictures can be arranged into a downsampling group. Then, the difference operation between the predicted probability and the true probability is performed on the upsampling group and the downsampling group, and it is marked as the comparison loss. For example, the first picture feature of the first picture and the second picture feature of the second picture can be arranged into an upsampling group, and the first picture feature of the first picture and each of the remaining pictures can be arranged into a downsampling group, so that several groups of downsampling groups can be obtained. Then, the distance between the upsampling group and each group of downsampling groups can be marked as the comparison loss. Finally, the comparison loss can be calculated through the first picture feature, the second picture feature, and the features of the remaining pictures.

[0034] Further, the editing loss and the comparison loss can be added to obtain the comprehensive loss, and the algorithm for selecting the first picture feature can be updated according to the comprehensive loss. Among them, a constraint standard can be set. When the comprehensive loss does not meet the constraint standard, the variables of the algorithm for selecting the first picture feature can be changed to obtain a better algorithm for selecting the first picture feature relative to the algorithm for selecting the first picture feature, and the better algorithm for selecting the first picture feature can be updated. The update process is as described in the above steps, and it is updated until the comprehensive loss meets the constraint standard, and the updated algorithm for selecting the first picture feature is used as the feature selection model.

[0035] In an embodiment of the present invention, the feature information of the signal graph sequence is selected through the feature selection model, which may specifically include: obtaining the unreconstructed pictures in the signal graph sequence, and transmitting the unreconstructed pictures to the feature selection model to obtain the source picture features of the unreconstructed pictures; reconstructing the source picture features through picture transformation to obtain the feature information of the unreconstructed pictures.

[0036] S5. Build an internal defect detection model based on additive manufacturing, train the internal defect detection model, and introduce the selected feature information of the signal graph sequence into the trained internal defect detection model to detect the internal defects of the additive manufacturing parts.

[0037] In an embodiment of the present invention, building an internal defect detection model based on additive manufacturing and training the internal defect detection model may specifically include: obtaining the internal defect picture data set of the additive manufacturing parts, and dividing the internal defect picture data set into a training set and a test set; presetting a randomly generated neural network model, and performing simulation training on the randomly generated neural network model according to the random maximum likelihood algorithm; inputting the internal defect pictures in the training set into the randomly generated neural network model in sequence to obtain the high-dimensional features of the internal defect pictures in the training set in multiple dimensions; using the obtained high-dimensional features as the input of the training set, and the internal defect types mapped by the high-dimensional features as the output of the training set.

[0038] Specifically, defect pictures of different additive manufacturing parts can be obtained from the network or the additive manufacturing site. The defect pictures should preferably include various types of defects, and labels should be attached to different defect types, thereby forming a dataset that maps pictures to defect types. Based on the labels on the defect pictures, the defect picture dataset can be divided into a standard sampling dataset and a defect sampling dataset, and the standard sampling data and defect sampling data can be further divided into a training set, a test set, and a validation set.

[0039] Furthermore, a randomly generated neural network model can be preset, and the variables in the randomly generated neural network model can be initialized. The randomly generated neural network model can be simulated and trained according to the random maximum likelihood algorithm, where the random maximum likelihood algorithm can perform operations on dimensions and iterate variables by reducing dimensions. After the simulation training of the randomly generated neural network model is completed, the randomly generated neural network model can be verified using the validation set.

[0040] Even further, the internal defect pictures in the training set can be sequentially input into the randomly generated neural network model, and the randomly generated neural network model can be used to select the characteristics of the internal defect pictures, thereby obtaining the high-dimensional characteristics of the defect pictures in multiple dimensions. The high-dimensional characteristics can include partial features and overall structures of the defect pictures. Finally, the obtained high-dimensional characteristics can be used as the input of the training set, and the internal defect types mapped by the high-dimensional characteristics can be used as the output of the training set to train the internal defect detection model.

[0041] According to the additive manufacturing internal defect detection method of the embodiments of the present invention, three-dimensional point cloud data of an additive manufacturing part is collected by detecting laser ultrasonic emission, a laser ultrasonic interferometer, and pulsed excitation laser ultrasound, and then a three-dimensional model of the additive manufacturing part is generated based on the three-dimensional point cloud data. A simulation algorithm is built through the three-dimensional model, and the detection orientation and detection route variables of the additive manufacturing part are obtained through the simulation algorithm. A laser ultrasonic signal is output through the detection orientation and detection route variables of the additive manufacturing part, and the characteristic information of the signal map sequence is selected through the laser ultrasonic signal. Finally, an internal defect detection model is built and trained, and the selected characteristic information is introduced into the trained internal defect detection model to detect the internal defects of the additive manufacturing part. Thus, the internal defect positions of the additive manufacturing part can be accurately located, and non-contact non-destructive detection can be achieved during the detection process, with high sensitivity and fast detection speed.

[0042] To implement the additive manufacturing internal defect detection method of the above embodiments, the present invention also proposes an additive manufacturing internal defect detection device.

[0043] As Figure 2As shown in the figure, the additive manufacturing internal defect detection device according to the embodiment of the present invention includes: a data acquisition unit 100, a model building unit 200, a signal output unit 300, a feature selection unit 400, and a defect detection unit 500. Among them, the data acquisition unit 100 is used to acquire the three-dimensional point cloud data of the additive manufacturing part; the model building unit 200 is used to generate a three-dimensional model of the additive manufacturing part according to the three-dimensional point cloud data, build a simulation algorithm of the additive manufacturing part through the three-dimensional model, and obtain the detection orientation and detection route variables of the additive manufacturing part through the simulation algorithm of the additive manufacturing part; the signal output unit 300 is used to output a laser ultrasonic signal according to the laser ultrasonic system through the detection orientation and detection route variables of the additive manufacturing part. Among them, the laser ultrasonic signal includes a signal map sequence of different orientations of the additive manufacturing part; the feature selection unit 400 is used to build a feature selection model of the signal map sequence and select the feature information of the signal map sequence through the feature selection model; the defect detection unit 500 is used to build an internal defect detection model based on additive manufacturing, train the internal defect detection model, and introduce the selected feature information of the signal map sequence into the trained internal defect detection model to detect the internal defects of the additive manufacturing part.

[0044] In an embodiment of the present invention, the data acquisition unit 100 can acquire the three-dimensional point cloud data of the additive manufacturing part through an RGB-D camera and calculate the three-dimensional coordinates inside the additive manufacturing part through the three-dimensional point cloud data.

[0045] In an embodiment of the present invention, first, the model building unit 200 can build a simulation algorithm of additive manufacturing through the three-dimensional model of the additive manufacturing part to be detected, and can obtain the corresponding relationship between the weakening degree of the laser and different dimensions of the laser intensity, incident orientation, and discrete angle by simulating the reaction of the refraction of the laser with different laser intensities, incident orientations, and discrete angles on the additive manufacturing part. Then, the signal-to-noise ratio, gray-scale difference value, and recognition rate of the generated image of the reflected laser ultrasonic signal can be set as the improvement targets for enhancing the image clarity, and the corresponding relationship between the weakening degree of the laser and different dimensions of the laser intensity, incident orientation, and discrete angle can be converted into the boundary factors of several targets to build an improvement algorithm for multiple targets. Secondly, the best combination of the weakening degree of the laser and the laser intensity, incident orientation, and discrete angle can be extracted to obtain a more comprehensive detection orientation. Finally, due to the possible undetected areas and the possibility of repeated detection, and at the same time, the laser may be weakened inside the additive manufacturing part, the detection orientation and detection route variables of the additive manufacturing part can be obtained by improving the emission position of the laser emission source and the detection route.

[0046] In an embodiment of the present invention, the signal output unit 300 can obtain a signal map sequence of laser ultrasonic signal conversion in different orientations by acquiring the optimal detection orientation and detection route variables of the additive manufacturing part. Thus, the internal structure of the additive manufacturing part can be collected in all directions.

[0047] In an embodiment of the present invention, a characteristic selection model of the signal map sequence is constructed, which specifically includes: collecting a first picture and a second picture of any position in the signal map sequence, where both the first picture and the second picture are obtained through more than one processing scheme; respectively performing characteristic selection on the first picture, the second picture and the remaining pictures to obtain the first picture characteristics of the first picture, the second picture characteristics of the second picture and the picture characteristics of the remaining pictures, where the first picture characteristics are obtained according to a preset first picture characteristic selection algorithm; editing the first picture characteristics through the first picture to obtain an edited picture of the first picture and an editing loss of the first picture characteristics, where the editing loss is the characteristic difference value between the edited picture and the first picture; for the second picture, the second picture characteristics and the picture characteristics of the remaining pictures, obtaining a comparison loss, where the comparison loss is the characteristic distance between the first picture characteristics and the second picture characteristics and the difference value between the first picture characteristics and the picture characteristics of the remaining pictures; updating the first picture characteristic selection algorithm according to the editing loss and the comparison loss to obtain the characteristic selection model.

[0048] Specifically, two characteristic selection algorithms with the same structure can be used to select the first picture characteristics, the second picture characteristics and the picture characteristics of the remaining pictures for the first picture, the second picture and the remaining pictures. The detailed extraction process can select the first picture characteristics through the first picture characteristic selection algorithm, select the second picture characteristics through the second picture selection algorithm and the picture characteristics of the remaining pictures, where the variable of the second picture characteristic selection algorithm can be obtained by performing a moving average operation on the variable of the first picture selection algorithm.

[0049] Further, after obtaining the first picture characteristics of the first picture, a decoder can be used to edit the first picture characteristics to obtain an edited picture of the first picture, and a cost function can be used to calculate the loss between the edited picture and the first picture, which is marked as the editing loss.

[0050] Further, the features obtained by extracting the same picture can be arranged into an upsampling group, and the features obtained by extracting different pictures can be arranged into a downsampling group. Then, the difference operation of the predicted probability and the true probability is performed on the upsampling group and the downsampling group, and it is marked as the comparison loss. For example, the first picture feature of the first picture and the second picture feature of the second picture can be arranged into an upsampling group, and the first picture feature of the first picture and each of the remaining pictures can be arranged into a downsampling group, so that several groups of downsampling groups can be obtained. Then, the distance between the upsampling group and each group of downsampling groups can be marked as the comparison loss. Finally, the comparison loss can be calculated through the first picture feature, the second picture feature, and the features of the remaining pictures.

[0051] Further, the editing loss and the comparison loss can be added to obtain a comprehensive loss, and the first picture feature selection algorithm can be updated according to the comprehensive loss. Among them, a constraint standard can be set. When the comprehensive loss does not meet the constraint standard, the variables of the first picture feature selection algorithm can be changed to obtain a first picture feature selection algorithm that is better than the first picture feature selection algorithm, and the better first picture feature selection algorithm can be updated. The update process is as described in the above steps, and it is updated until the comprehensive loss meets the constraint standard, and the updated first picture feature selection algorithm is used as the feature selection model.

[0052] In an embodiment of the present invention, the feature selection unit 400 can also be used to: obtain the unreconstructed pictures in the signal graph sequence, and transmit the unreconstructed pictures to the feature selection model to obtain the source picture features of the unreconstructed pictures; reconstruct the source picture features through picture transformation to obtain the feature information of the unreconstructed pictures.

[0053] In an embodiment of the present invention, the defect detection unit 500 can also be used to: obtain the internal defect picture dataset of the additive manufacturing part, and divide the internal defect picture dataset into a training set and a test set; preset a randomly generated neural network model, and perform simulation training on the randomly generated neural network model according to the random maximum likelihood algorithm; by sequentially inputting the internal defect pictures in the training set into the randomly generated neural network model, obtain the high-dimensional features of the internal defect pictures in the training set in multiple dimensions; use the obtained high-dimensional features as the input of the training set, and the internal defect types mapped by the high-dimensional features as the output of the training set.

[0054] Specifically, defect pictures of different additive manufacturing parts can be obtained from the network or the additive manufacturing site. The defect pictures should preferably include various types of defects, and labels should be attached to different defect types, thereby forming a dataset that maps pictures to defect types. Based on the labels on the defect pictures, the defect picture dataset can be divided into a standard sampling dataset and a defect sampling dataset, and the standard sampling data and defect sampling data can be further divided into a training set, a test set, and a validation set.

[0055] Furthermore, a randomly generated neural network model can be preset, and the variables in the randomly generated neural network model can be initialized. The randomly generated neural network model can be simulated and trained according to the random maximum likelihood algorithm. Among them, the random maximum likelihood algorithm can perform operations on dimensions and iterate variables by reducing dimensions. After the simulation training of the randomly generated neural network model is completed, the randomly generated neural network model can be verified using the validation set.

[0056] Even further, the internal defect pictures in the training set can be sequentially input into the randomly generated neural network model, and the randomly generated neural network model can be used to select the characteristics of the internal defect pictures, thereby obtaining the high-dimensional characteristics of the defect pictures in multiple dimensions. Among them, the high-dimensional characteristics can include partial features and overall structures of the defect pictures. Finally, the obtained high-dimensional characteristics can be used as the input of the training set, and the internal defect types mapped by the high-dimensional characteristics can be used as the output of the training set to train the internal defect detection model.

[0057] In summary, the present invention acquires three-dimensional point cloud data of an additive manufacturing part through detecting laser ultrasound emission, a laser ultrasound interferometer, and pulsed excitation laser ultrasound, then generates a three-dimensional model of the additive manufacturing part based on the three-dimensional point cloud data, constructs a simulation algorithm through the three-dimensional model, and obtains the detection orientation and detection route variables of the additive manufacturing part through the simulation algorithm. Laser ultrasound signals are output through the detection orientation and detection route variables of the additive manufacturing part, and the characteristic information of the signal map sequence is selected through the laser ultrasound signals. Finally, an internal defect detection model is constructed and trained, and the selected characteristic information is introduced into the trained internal defect detection model to detect internal defects of the additive manufacturing part. Thus, the position of the internal defects of the additive manufacturing part can be accurately located, and non-contact non-destructive detection can be achieved during the detection process, with high sensitivity and fast detection speed.

[0058] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The meaning of "a plurality" is two or more, unless otherwise specifically defined.

[0059] In the present invention, unless otherwise clearly defined and limited, terms such as "installed", "connected", "coupled", "fixed", etc. shall be construed in a broad sense. For example, it may be a fixed connection, a detachable connection, or integrated; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the internal communication of two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0060] In the present invention, unless otherwise clearly defined and limited, the first feature being "on" or "under" the second feature may be that the first and second features are in direct contact, or the first and second features are indirectly in contact through an intermediate medium. Moreover, the first feature being "above", "over" and "on top of" the second feature may be that the first feature is directly above or obliquely above the second feature, or merely indicates that the first feature has a higher horizontal height than the second feature. The first feature being "under", "beneath" and "underneath" the second feature may be that the first feature is directly below or obliquely below the second feature, or merely indicates that the first feature has a lower horizontal height than the second feature.

[0061] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not have to be directed to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0062] Any process or method description shown in the flowchart or described in other ways herein can be understood as representing a module, segment, or part of code including one or more executable instructions for implementing a specific logical function or process. The scope of the preferred embodiments of the present invention includes additional implementations, where the functions may be executed in a substantially simultaneous manner or in a reverse order according to the functions involved, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0063] The logic and / or steps represented in the flowchart or otherwise described herein can, for example, be considered as a definable sequence list of executable instructions for implementing logical functions, which can be embodied specifically in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in conjunction with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection parts with one or more wirings (electronic devices), portable computer disk cartridges (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber devices, and portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other suitable processing as necessary, and then stored in a computer memory.

[0064] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or combinations thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having suitable combinational logic gate circuits, programmable gate arrays (PGA), field-programmable gate arrays (FPGA), etc.

[0065] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried by the method of implementing the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0066] In addition, each functional unit in various embodiments of the present invention may be integrated into one processing module, may exist physically alone for each unit, or two or more units may be integrated into one module. The above integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0067] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for detecting internal defects in additive manufacturing, characterized in that: The following steps are involved: Obtain 3D point cloud data of additively manufactured parts; Generating a three-dimensional model of the additively manufactured part according to the three-dimensional point cloud data, building a simulation algorithm of the additively manufactured part through the three-dimensional model, and obtaining detection position and detection route variables of the additively manufactured part through the simulation algorithm of the additively manufactured part; The laser ultrasonic system outputs a laser ultrasonic signal through the detection position and detection route variables of the additively manufactured part, wherein the laser ultrasonic signal includes a sequence of signal graphs at different positions of the additively manufactured part; Building a feature selection model for a signal graph sequence, and selecting feature information of the signal graph sequence through the feature selection model; An internal defect detection model based on additive manufacturing is built, the internal defect detection model is trained, and characteristic information of the selected signal graph sequence is introduced into the trained internal defect detection model to perform internal defect detection on the additively manufactured part.

2. The additive manufacturing internal defect detection method according to claim 1, characterized in that: The feature selection model for building a signal graph sequence specifically includes: Collecting a first image and a second image of an image at any position in the signal image sequence, wherein the first image and the second image are obtained through more than one processing scheme; Selecting characteristics of the first picture, the second picture, and the remaining pictures respectively to obtain a first picture characteristic of the first picture, a second picture characteristic of the second picture, and picture characteristics of the remaining pictures, wherein the first picture characteristic is obtained according to a preset first picture characteristic selection algorithm; Editing the first picture characteristic through the first picture to obtain an edited picture of the first picture and an editing loss of the first picture characteristic, wherein the editing loss is a characteristic difference value between the edited picture and the first picture; Obtaining a comparative loss through the second picture, the second picture characteristic, and the picture characteristics of the remaining pictures, wherein the comparative loss is a characteristic interval between the first picture characteristic and the second picture characteristic and a difference value between the first picture characteristic and the picture characteristics of the remaining pictures; The first image feature selection algorithm is updated according to the editing loss and the comparison loss to obtain the feature selection model.

3. The additive manufacturing internal defect detection method according to claim 2, characterized in that: The said selecting characteristic information of the signal graph sequence by the said characteristic selection model specifically includes: Acquire an unreconstructed picture in the signal image sequence, and transmit the unreconstructed picture to the feature selection model to obtain source picture features of the unreconstructed picture; The source image characteristics are reconstructed through image transformation to obtain characteristic information of the unreconstructed image.

4. The additive manufacturing internal defect detection method according to claim 3, characterized in that: Also includes: If there is a defect, the position and size of the internal defect of the additively manufactured part in the three-dimensional model are obtained through the three-dimensional model of the additively manufactured part and the signal image sequence in different positions.

5. The additive manufacturing internal defect detection method according to claim 4, characterized in that: The building of the internal defect detection model based on additive manufacturing and training of the internal defect detection model specifically include: Acquire an internal defect image dataset of the additively manufactured part, and divide the internal defect image dataset into a training set and a test set; Presetting a randomly generated neural network model, and performing simulation training on the randomly generated neural network model according to a random maximum likelihood algorithm; The internal defect images in the training set are sequentially input into the randomly generated neural network model to obtain high-dimensional characteristics of the internal defect images in the training set in multiple dimensions; The obtained high-dimensional characteristics are used as the input of the training set, and the internal defect types mapped by the high-dimensional characteristics are used as the output of the training set.

6. An additive manufacturing internal defect detection device, characterized in that: include: A data acquisition unit, the data acquisition unit is used to acquire three-dimensional point cloud data of the additively manufactured part; A model building unit, the model building unit is used to generate a three-dimensional model of the additively manufactured part according to the three-dimensional point cloud data, build a simulation algorithm of the additively manufactured part through the three-dimensional model, and obtain a detection position and a detection route variable of the additively manufactured part through the simulation algorithm of the additively manufactured part; A signal output unit, the signal output unit is used to output a laser ultrasonic signal according to a detection position and a detection route variable of the laser ultrasonic system through the additively manufactured part, wherein the laser ultrasonic signal includes a sequence of signal graphs at different positions of the additively manufactured part; A feature selection unit, the feature selection unit is used to build a feature selection model for the signal graph sequence, and select feature information of the signal graph sequence through the feature selection model; A defect detection unit, wherein the defect detection unit is used to build an internal defect detection model based on additive manufacturing, train the internal defect detection model, and introduce characteristic information of the selected signal image sequence into the trained internal defect detection model to perform internal defect detection on the additively manufactured part.

7. The additive manufacturing internal defect detection device according to claim 6, characterized in that: The feature selection unit is also used for: Collecting a first image and a second image of an image at any position in the signal image sequence, wherein the first image and the second image are obtained through more than one processing scheme; Selecting characteristics of the first picture, the second picture, and the remaining pictures respectively to obtain a first picture characteristic of the first picture, a second picture characteristic of the second picture, and picture characteristics of the remaining pictures, wherein the first picture characteristic is obtained according to a preset first picture characteristic selection algorithm; Editing the first picture characteristic through the first picture to obtain an edited picture of the first picture and an editing loss of the first picture characteristic, wherein the editing loss is a characteristic difference value between the edited picture and the first picture; Obtaining a comparative loss through the second picture, the second picture characteristic, and the picture characteristics of the remaining pictures, wherein the comparative loss is a characteristic interval between the first picture characteristic and the second picture characteristic and a difference value between the first picture characteristic and the picture characteristics of the remaining pictures; The first image feature selection algorithm is updated according to the editing loss and the comparison loss to obtain the feature selection model.

8. The additive manufacturing internal defect detection device according to claim 7, characterized in that: The feature selection unit is also used for: Acquire an unreconstructed picture in the signal image sequence, and transmit the unreconstructed picture to the feature selection model to obtain source picture features of the unreconstructed picture; The source image characteristics are reconstructed through image transformation to obtain characteristic information of the unreconstructed image.

9. The additive manufacturing internal defect detection device according to claim 8, characterized in that: Also includes: A defect acquisition unit is used to acquire the position and size of the internal defects of the additively manufactured part in the three-dimensional model through the three-dimensional model of the additively manufactured part and the signal image sequences in different positions.

10. The additive manufacturing internal defect detection device according to claim 9, characterized in that: The defect detection unit is also used for: Acquire a data set of internal defect images of the additively manufactured part, and divide the data set of internal defect images into a training set and a test set; Presetting a randomly generated neural network model, and performing simulation training on the randomly generated neural network model according to a random maximum likelihood algorithm; The internal defect images in the training set are sequentially input into the randomly generated neural network model to obtain high-dimensional characteristics of the internal defect images in the training set in multiple dimensions; The obtained high-dimensional characteristics are used as the input of the training set, and the internal defect types mapped by the high-dimensional characteristics are used as the output of the training set.

Images