Application of adaptive harmony search and machine learning on optimization problems about strength of materials

Abstract

In structural design, the strength of materials is the key factor in design and applications including optimization covers the strength of material theory in design. The constraints are generally related to the strength of the material for different types of stress that occur under various external loading. The number of design problems using metaheuristics in the subject is high and these problems are generally used as benchmark examples in structural optimization. In this chapter, adaptive harmony search is presented for these problems and multiple cases of these problems are solved to obtain machine learning data. Then, artificial intelligence models that predict optimum results without a rerun of the iterative optimization process are generated. The models used are compared with performance metrics. These performance metrics are Coefficient of determination (R2), Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Mean Square Error (MSE). When the success of all regression models was analysed, it was seen that the model with the highest R2 (0.9994) and the low error values was Random Forest. © 2025 Elsevier Ltd. All rights reserved.