Abstract:It is a challenging to ensure the quality of microholes during the laser drilling of bulk metallic glasses (BMGs), where the defects such as heat-affected zones, burrs, and low smoothness are usually obtained. Based on machine vision, back-propagation neural network (BPNN), and genetic algorithm (GA) in artificial intelligence technology, a relationship model between the microhole features and laser processing parameters was proposed, so as to predict and optimize the processing parameters during the laser drilling of BMGs. Firstly, the geometric parameters and texture information of microholes were obtained using fractal theory and machine vision techniques, including the gray-level co-occurrence matrix (GLCM) algorithm, image binarization, and edge extraction. A comprehensive numerical characterization of various microhole features was conducted, providing the basis for subsequent prediction and optimization. Secondly, the previously-extracted microhole features were set as input layer, and the processing parameters were set as output layer. After proper training, the BP neural network optimization model (GABPNN), based on the genetic algorithm, was established. As compared to the original BPNN model, the GABPNN model exhibited improved performance, reducing the maximum prediction error from 4.860% to 2.285%. Thirdly, with a set of input layer parameters, the required processing parameters were predicted by the model. The predicted results were verified in experiments, where good surface quality of the laser-drilled BMG microholes with good roundness and small ablation was obtained. This shows that it is feasible to predict and optimize processing parameters by GABPNN based on feature parameters extracted by machine vision.

Research of PCB Micro-Drilling Quality Modeling Based on Bayesian-Networks

Research on the Fault Diagnosis of the CNC Machine Tool Based on Bayesian Network

Thermal Error Modeling of NC Machine Tools Based on Bayesian Networks

A Machine Learning-Based Framework for Automatic Visual Inspection of Microdrill Bits in PCB Production

Intelligent Prediction for Surface Roughness of Cnc Surface Grinding Machine Tool Based on Bayesian Network

Research on Intelligent Monitoring Technique of Machining State for Surface Grinder Based on Bayesian Network

Research on remote intelligent fault-diagnosis of CNC lathe based on Bayesian Networks

A Naive Bayesian Belief Network Model for Predicting Effort Deviation Rate in Software Testing

Accurate temperature characterization and experimental verification in the PCB drilling process

Geometry Parameters Estimation of Defects in Multi-layered Structures Based on Eddy Current Nondestructive Testing Technique with Bayesian Networks

Robust Bayesian Networks for Low-Quality Data Modeling and Process Monitoring Applications

Dynamic Bayesian Network for Robust Latent Variable Modeling and Fault Classification

Bayesian networks modeling for thermal error of numerical control machine tools

Micro drill defect detection with hybrid BP networks, clusters selection and crossover

Data Rectification Based on Bayesian Network

Research on Modeling Thermal Dynamic Errors of Precision Machine Based on Fuzzy Logic and Artifical Neural Network

Process modeling and optimization in laser drilling of bulk metallic glasses based on GABPNN and machine vision

Uncertainty quantification in machining deformation based on Bayesian network

The micro-drilling process fine modeling with the strain characteristics of undeformed chip considered

Predictive Model for the Interlayer Burr Height During Drilling of Stacked Aluminum Plates

Is the drill pipe safe during drilling process?—A new method for evaluating drill pipe failure risk based on the Noisy-OR gate and bayesian network