An evolutionary neural network approach to machining process planning: A proof of concept

Chen, Niechen

doi:10.1016/j.promfg.2021.06.083

Cited by 3 publications

(1 citation statement)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CAPP also saw many attempts to use ML for better results. Methods have been proposed based on the Genetic Algorithm [11], ANN [12], a combination of GA, ANN, and analytical hierarchical process (AHP) [13]. Besides process planning, ML techniques have been widely used in optimizing the toolpath parameters such as feed rate [14], depth of cut [15], machining direction [1], and stepovers [16].…”

Section: Related Literaturementioning

confidence: 99%

Optimum Toolpath Planning Strategy Prediction using Machine Learning Technique

Kukreja¹,

Pande

2022

Preprint

View full text Add to dashboard Cite

Finding the best toolpath planning strategy for Computer Numerical Control (CNC) machining requires a trial-and-error approach. One needs to follow a number of steps, viz., generate the toolpath, perform simulations/machining, measure various parameters to quantify the quality of the toolpath, and then select the best toolpath. This conventional iterative approach is time-consuming and often error-prone. This paper presents a novel Machine Learning based system for choosing the best toolpath planning strategy for CNC machining (finishing) of complex freeform surfaces directly from the CAD model. Three tool path planning strategies are considered, viz. Adaptive planar, Iso-scallop, and Hybrid. At first, a novel toolpath analysis module is presented to evaluate the quality of the toolpath considering performance parameters such as the surface finish, toolpath length, and smoothness. The toolpath analysis module was extensively tested for robustness and accuracy. This quality measurement technique is then used to analyze and label a large number of CAD models to create a dataset for supervised learning for the three toolpath strategies. A Convolutional Neural Network (CNN) model is designed to predict the best toolpath planning strategy. Results show that the proposed data-driven model achieved 96.8% test accuracy.

show abstract