Optimisation of Flexible Forming Processes Using Multilayer Perceptron Artificial Neural Networks and Genetic...

Area of Science:

• Materials Science
• Manufacturing Engineering
• Computational Intelligence

Background:

• Flexible forming processes enable diverse product shapes without retooling.
• Single-point incremental forming (SPIF) offers flexibility but struggles with excessive sheet metal thinning.
• Pre-forming phases can enhance thickness uniformity in SPIF.

Purpose of the Study:

• To develop a generalized approach for minimizing sheet metal thinning in SPIF.
• To optimize SPIF process parameters for improved formability, particularly for advanced high-strength steels (AHSSs).
• To integrate artificial intelligence for predicting and controlling thinning.

Main Methods:

• Utilized a multilayer perceptron artificial neural network (MLP ANN) to predict thinning based on input parameters.
• Employed a genetic algorithm (GA) to optimize these input parameters for minimal thinning.
• Focused on the application and formability of advanced high-strength steels (AHSSs).

Main Results:

• Successfully demonstrated a generalized method to predict and minimize sheet metal thinning.
• Optimized SPIF parameters effectively reduced material thinning.
• Provided valuable insights into the formability of AHSSs during incremental forming.

Conclusions:

• The proposed MLP ANN and GA approach is effective in minimizing sheet metal thinning during SPIF.
• This method offers significant advancements in flexible forming technologies.
• The approach is applicable to a wide range of materials and industrial applications, especially AHSSs.