Integration of Machine Learning Techniques in Topology Optimization for Enhancing Parallel Kinematics Mechanisms Performance

EasyChair Preprint no. 13198

Abstract

This research paper investigates the integration of machine learning techniques into the topology optimization process to enhance the performance of Parallel Kinematics Mechanisms (PKMs). PKMs offer advantages in precision, stiffness, and dynamic performance but face challenges in structural integrity and weight reduction. Topology optimization, a computational design approach, systematically redistributes material within the design space to achieve optimal performance criteria. However, addressing computational complexity and scalability issues remains a challenge. This paper explores the integration of machine learning techniques to optimize the PKM design process. By leveraging machine learning algorithms, such as neural networks and reinforcement learning, engineers can develop more efficient optimization strategies, overcome computational challenges, and achieve superior PKM designs. Case studies, performance evaluation metrics, and future directions are discussed to illustrate the potential and implications of integrating machine learning with topology optimization for PKM design.

Keyphrases: computational complexity, machine learning, optimization strategies, Parallel Kinematics Mechanisms, structural integrity, topology optimization, weight reduction

@Booklet{EasyChair:13198,
  author = {Wahaj Ahmed and Lanying Fang},
  title = {Integration of Machine Learning Techniques in Topology Optimization for Enhancing Parallel Kinematics Mechanisms Performance},
  howpublished = {EasyChair Preprint no. 13198},

  year = {EasyChair, 2024}}