Author: MOHD FIRASATH ALI, MIRZA HAROON BAIG, MOHAMMED ABDUL MOYEED
Doi: https://jmeb.2024.v9.i02.pp64-78
Abstract:
In the beginning stages of dynamic system development, which involves many disciplines and a hierarchical structure, system requirements need to be pushed down to goal values for each component so that engineers can work together effectively and at the same time. The goal of this work is to suggest a new set-based parallel engineering method for a system that is always changing that uses machine learning. To practise setting goals for concurrent engineering, you need both hierarchical simulations at the system and component levels and a way to use simulations to solve problems that go in the opposite direction. The suggested way is made up of two machine learning techniques that meet these needs. The first one is physics-informed long short-term memory (PI-LSTM), which makes it easier to model how parts behave mechanically. The adaptable range of mechanical modelling can be increased by using the suggested PI-LSTM in places where it is hard to do so. For the dynamic behaviour o
Download PDF