Computational Intelligence–Driven Multi-Objective Optimization of Rib-Stiffened Composite Panels Under Combined Loading Conditions

Abstract

This study focuses on identifying an optimal trade-off between mass and cost characteristics of rib-stiffened composite panels operating under combined loading conditions. The optimization framework is developed with consideration of real operating environments, where even minor variations in individual parameters may significantly affect overall structural performance. The research integrates principles of engineering mechanics and mathematical modeling within a vector-based formulation of the optimization problem. To solve this problem, an integral quality criterion is constructed, incorporating expert judgment and the guaranteed result principle. The proposed approach enables the determination of optimal design parameters for panels with different stiffening configurations while ensuring adaptability to varying load conditions and design constraints. The effectiveness of the model is validated through six loading scenarios applied to panels with rectangular and angular stiffeners. The findings contribute to the advancement of multi-objective optimization of composite structures by refining modeling approaches and extending methods for evaluating trade-offs between competing criteria. The practical relevance of the study lies in the development of a software tool for selecting rational stiffening configurations based on given constraints and requirements. The implemented human-machine procedure allows the incorporation of expert knowledge into the optimization process without compromising the accuracy of mathematical modeling, which is particularly important in applications where both weight and cost are critical factors. A limitation of the study is the consideration of only two optimization criteria - mass and cost. Future research should focus on expanding the set of performance indicators and enhancing the complexity of evaluation models. From the standpoint of computational intelligence, the proposed human-machine procedure may be interpreted as an intelligent decision-support mechanism for multicriteria structural design under incomplete certainty, where expert preferences are iteratively formalized within the optimization loop.