Improved ACO Algorithm And Its Application In Truss Structural Optimization Design Based On Reliability

At present, the economy and the practicality of structural optimization design have caused widespread concern. The truss structure is one of the most common structures in engineering structures. Its optimization design is to chose structural bars with reasonable cross-sectional area, in the condition of the structure constraints, to reduce the structural weight mostly. Traditional design methods based on statistical knowledge regard the uncertainty existing in structural design process as safety factor, and it cannot reflect the degree of reliability of the structure effectively. Therefore, studying the truss structure optimization design based on the reliability has important theory and practical significance.In order to find an effective method for the reliability based optimization of the truss structure, an improved ant colony optimization based on chaos theory was proposed. The chaos theory was introduced to the local pheromone update, and was used to control the selection of the main parameters of the ant colony optimization algorithm. The strong optimization capability and the high optimization efficiency are indicated during the process of solving a case with classical function optimization problem. Then the improved algorithm was applied to discrete variables and conditions variables respectively optimization design of truss structure based on reliability. Convergence criteria of the algorithm were designed according to the specific problem. The effectiveness and superiority of the improved algorithm were obviously verified by several examples of the structural optimization and performance comparison with other algorithms. This study provides a new and effective method for the reliability-based structural optimization design of complex problems.