An Intelligent Multi-objective Design Optimization Method for Nanographite-Based Electrically Conductive Cementitious Composites

Nanographite (NG) is a promising conductive filler for producing effective electrically conductive cementitious composites for use in structural health monitoring methods. Since the acceptable mechanical strength and electrical resistivity are both required, the design of NG-based cementitious composite (NGCC) is a complicated multi-objective optimization problem. This study proposes a data-driven method to address this multi-objective design optimization (MODO) issue for NGCC using machine learning (ML) techniques and non-dominated sorting genetic algorithm (NSGA-II). Prediction models of the uniaxial compressive strength (UCS) and electrical resistivity (ER) of NGCC are established by Bayesian-tuned XGBoost with prepared datasets. Results show that they have excellent performance in predicting both properties with high R² (0.95 and 0.92, 0.99 and 0.98) and low mean absolute error (1.24 and 3.44, 0.15 and 0.22). The influence of critical features on NGCC’s properties are quantified by ML theories, which help determine the variables to be optimized and define their constraints for the MODO. The MODO program is developed on the basis of NSGA-II. It optimizes NGCC’s properties of UCS and ER simultaneously, and successfully achieves a set of Pareto solutions, which can facilitate appropriate parameters selections for the NGCC design.