Modeling and Optimization for The Tensile Properties of 3D-Printed FRP using Artificial Neural Network and Artificial Bee Colony Algorithm
IABSE Congress, Nanjing 2022: Bridges and Structures: Connection, Integration and Harmonisation IABSE Congress Reports（2022）
Fiber-reinforced polymer (FRP) has multiple applications as a primary material or reinforcing material for the structural elements. Controlling the quality of the 3D printed FRP is critical to guarantee a FRP material of high performance. In this research, machine learning (ML) model based on data collected from experimental studies was developed by artificial neural network (ANN) to control the quality of 3D printed FRP. ANN model predicts the ultimate tensile strength (UTS) of the FRP as function of 7 material and printing parameters. The UTS of the FRP was maximized via optimizing the printing and material parameters by using artificial bee colony (ABC) algorithm. ANN and ABC algorithms were coded by MATLAB. The results showed that the developed ANN model can predict with good accuracy the UTS of FRP. Moreover, it was found that the ABC optimization algorithm can design the input parameters such that a FRP with maximum UTS can be obtained.
