Defect prediction method based on multi-feature parallel multi-stage neural network (MF-PMSNN)
A defect prediction method based on a multi-feature parallel multi-stage neural network (MF-PMSNN), includes: obtaining a trajectory dataset, and preprocessing data of a defect of a workpiece in additive manufacturing (AM); building an MF-PMSNN, and evaluating an output classification result based on evaluation indicators; and performing real-time defect prediction, and deploying a trained MF-PMSNN model to a production environment. The present disclosure combines and effectively matches thermal imaging-based in-situ monitoring data and X-ray computed tomography (XCT)-based in-situ monitoring data to ensure temporal and spatial consistency between the thermal imaging-based in-situ monitoring data and the XCT-based in-situ monitoring data. In this way, a molten pool status and a pore of the workpiece can be captured more comprehensively. The MF-PMSNN is proposed to obtain a molten pool status and the porosity distribution in the data and perform defect prediction.
1 . A defect prediction method based on a multi-feature parallel multi-stage neural network (MF-PMSNN), comprising the following steps:
step 1 , obtaining a trajectory dataset, and preprocessing data of a defect of a workpiece in additive manufacturing (AM);
step 2 , building an MF-PMSNN, wherein the MF-PMSNN has N branch networks, and each of the branch networks comprises a Conv-Stem layer, a Stage X module, an average pooling layer, a fully connected layer, and a classifier that are sequentially connected, capturing space and timing sequence information in multimodal data, and obtaining a molten pool status and a porosity distribution in the data, wherein specific steps are as follows:
step 2 . 1 , inputting preprocessed data into the Conv-Stem layer for feature extraction to obtain an eigenvalue;
step 2 . 2 , inputting the eigenvalue into the Stage X module for data processing, wherein the Stage X module comprises a pooling layer and a local-to-global feature extraction module that are sequentially connected;
step 2 . 3 , inputting the eigenvalue processed by the Stage X module into the average pooling layer for average pooling, and extracting a feature of the processed eigenvalue; and
step 2 . 4 , inputting the feature of the processed eigenvalue into the fully connected layer, constructing an output layer through the fully connected layer, using an activation function to map the feature inputted into the fully connected layer, and inputting mapped feature into the classifier for defect category recognition to obtain a defect category label;
step 3 , evaluating an output classification result based on evaluation indicators; and
step 4 , performing real-time defect prediction, and deploying a trained MF-PMSNN model to actual production of the workpiece.
2 . The defect prediction method according to claim 1 , wherein the obtaining a trajectory dataset in the step 1 comprises:
collecting trajectory data by using thermal imaging-based in-situ monitoring and X-ray computed tomography (XCT)-based in-situ monitoring devices, wherein the trajectory data comprises thermal imaging-based in-situ monitoring data and XCT-based in-situ monitoring data;
defining a label related to the defect of the workpiece through supervised learning, wherein the label comprises the molten pool status and the porosity distribution; and
effectively matching the thermal imaging-based in-situ monitoring data and the XCT-based in-situ monitoring data to ensure temporal and spatial consistency between the thermal imaging-based in-situ monitoring data and the XCT-based in-situ monitoring data.
3 . The defect prediction method according to claim 2 , wherein the preprocessing the data of the defect of the workpiece in AM in the step 1 comprises:
step 1 . 1 , obtaining the trajectory dataset;
step 1 . 2 , performing data cleaning to clean a missing value, an outlier, or an incorrect label in the trajectory dataset;
step 1 . 3 , standardizing and normalizing the data;
step 1 . 4 , labeling a defect in the trajectory dataset and a category to which each sample in the trajectory dataset belongs;
step 1 . 5 , performing data enhancement on the trajectory dataset; and
step 1 . 6 , dividing the trajectory dataset into a training set, a validation set, and a test set.
4 . The defect prediction method according to claim 3 , wherein the trajectory dataset in the step 1 . 1 comprises:
a model of a processing engine bed, ambient humidity, an ambient temperature, a molten pool length, a molten pool width, a molten pool eccentricity ratio, an average molten pool temperature, and a maximum molten pool temperature in the AM.
5 . The defect prediction method according to claim 1 , wherein the inputting the preprocessed data into the Conv-Stem layer for feature extraction in the step 2 . 1 comprises:
performing 3×3 convolution with a stride of 2 on the preprocessed data to halve a number of an output eigenvalue, performing batch normalization, and using an activation function Gaussian Error Linear Unit (GELU) in network training:
GELU
(
x
)
=
0.5
x
(
1
+
tanh
(
2
x
(
x
+
0
.
0
4
4
7
1
5
x
3
)
)
)
.
6 . The defect prediction method according to claim 5 , wherein the data processing in the step 2 . 2 comprises pooling the eigenvalue by the Stage X module which specifically comprises:
pooling the eigenvalue outputted from the Conv-Stem layer, and capturing local information and global information of the eigenvalue, wherein the local-to-global feature extraction module comprises a contextual perception block, a first batch normalization layer, a lightweight multi-head self-attention mechanism, a second batch normalization layer, and a multi-layer perceptron that are sequentially connected;
enhancing the eigenvalue by using the contextual perception block to obtain enhanced data, performing 3×3 depth-separable convolution on the enhanced data, and inputting obtained data into the first batch normalization layer, performing layer normalization and calculating a statistical magnitude for each input data sample by the first batch normalization layer; and
executing the lightweight multi-head self-attention mechanism to capture a long distance dependency in data inputted into the lightweight multi-head self-attention mechanism; and processing the long distance dependency by using the second batch normalization layer and the multi-layer perceptron, to obtain an output vector.
7 . The defect prediction method according to claim 1 , wherein the constructing the output layer through the fully connected layer in the step 2 . 4 comprises:
connecting each input neuron of the fully connected layer to each output neuron to form a fully connected network, and converting the mapped feature into a final output, wherein an expression of the fully connected layer is as follows:
output
=
activation
(
input
×
weights
+
biases
)
wherein input represents data inputted into the fully connected layer, weights represents a weight matrix, biases represents a bias, and activation represents the activation function.
8 . The defect prediction method according to claim 1 , after the step 3 , further comprising: setting a loss function, using an Adam optimizer to optimize the MF-PMSNN, and obtaining an optimal MF-PMSNN when the loss function has a minimum value.
9 . The defect prediction method according to claim 8 , further comprising:
feedback and continuous improvement: collecting feedback information from a user and a professional on an evaluation result of the model after testing, adjusting the evaluation indicator, a model parameter, and a training method based on the feedback information, and regularly updating the trajectory dataset.