2021 journal article

Design and Development of Longitudinal and Torsional Ultrasonic Vibration-assisted Needle Insertion Device for Medical Applications

Computer-Aided Design and Applications, 19(4), 797–811.

By: Y. Wang, Y. Cai & Y. Lee* 

Source: ORCID
Added: March 15, 2022

Automated feature recognition (AFR) makes it possible to abstract semantic information from neutral CAD models.In an earlier work, we proposed an AFR method for aerospace sheet metal (ASM) parts.In this new work, that method's implementation as an AFR prototype is outlined and the differences between the prototype and the original proposal are pointed out.Then, streamlined data structures are described and explained.They are used to organize the B rep elements extracted from the ASM parts' STEP models, classify and enhance them, and structure the features recognized from the STEP models.Next, a few examples of the algorithms that are implemented in the prototype to manipulate the B rep elements and recognize features are represented and explained.The details of the algorithms are presented in the appendices.To validate the AFR method and verify its correct implementation, a collection of 26 real-world ASM parts was used to create CAD models that were subsequently converted to STEP models.The STEP models were processed to recognize their features, and the results show perfect accuracy.A few of the output feature files are presented in detail.Our results confirm great potential for further AFR method development for rather specialized domains of application.