Study On Structural Optimization Of Steel Truss Based On Improved Atomic Search Algorithm

Structural optimization is through the adjustment of the geometric shape,material,and size parameters of a structure to achieve the best performance meeting design requirements and constraints.With the complexity of the structures,the traditional optimization methods are more and more limited to handle complex structural optimization problems due to computational accuracy and convergence speed issues.The computational intelligence methods are becoming increasingly popular in structural optimization due to their faster computation speed and higher accuracy.Atom Search Optimization(ASO)is a powerful computational intelligence method that uses atomic motion rules and the interaction forces between atoms in a population to guide intelligent optimization searches.Compared to traditional methods,ASO can adaptively handle nonlinear,non-convex,multi-modal,and constrained problems,and is insensitive to initial values,able to handle large-scale problems,and easy to implement in parallel computing.However,ASO also has the shortcoming of premature which means to get stuck at local optimal solutions.To overcome these limitations and find a more effective method for structure optimization,this dissertation proposes an improved Atom Search Optimization method called Immune Clone Atom Search Optimization(ICASO).This method uses a reverse learning strategy to enhance population diversity and improve the balance between global and local searches.It also incorporates cloning,immune,and selection operations into the atomic position update process to expand the search range and improve search efficiency and accuracy.The effectiveness of this strategy is verified through simulation experiments of benchmark test functions,0-1knapsack problem,and TSP problem.Furthermore,the improved ICASO algorithm is applied to the size optimization,shape optimization,and topology optimization of steel truss structures.The candidate solutions are represented as structures composed of atoms,and the positions and types of atoms are changed in each iteration to realize the optimization.Result of simulation experiments and comparison with other optimization algorithms shows the efficiency of the improved Atom Search Optimization algorithm.This research provides a new and effective method for structural optimization,which can be used to promote the development of related fields.