Research On Structural Optimization Method Based On Ant Colony Algorithm-FEM

Structural optimization design is a principal aspect of structural design. It has been put a great emphasis in engineering. It is developed from structural optimization to structural reliability optimization. With the further development of structural design, optimization design theory and optimization algorithm are demanded higher and higher. Ant colony algorithm is a novel category of bionic algorithm. Ant colony algorithm which is adaptive has strong robustness and is easy to combine with other methods. So it is very important to research on ant colony algorithm employed in structural optimization. The basic ant colony algorithm and the improved ones are discussed. Then, the theory of ant colony algorithm and its application for structural optimization are researched. The main work can be generalized as follows:1.An improved TACO algorithm is presented. The improved TACO algorithm developed based on TACO. MAX-MIN ant system is introduced. To enlarge search range and simplify search process, a new method of positive-real-number coding is put forward. In order to reduce the influence of the following parameters, including the evaporation coefficientsρandγ, the factorτ0and the value of optimized function f best, a new technology of constant-pheromone update is put forward. As a result, it is propitious to the selection of most pheromone valueτmaxand least pheromone valueτmin. All of them improve the performance and universality of the TACO algorithm. As an example, some mathematic functions are optimized to prove the validity.2.The influences of ant amount m and threshold q 0are studied by the optimization of mathematic functions. The results include the follows:(1)A large amount of ant can enhance search-variety, but an excessive amount makes algorithm fall into local solution due to superfluous pheromone. With more ant amount m , it spends more time on calculation.(2)The influence of threshold q 0is not same to different functions. The value of parameter q 0is limit, which depends on ant amountm . The performance of algorithm is good when q 0is between 0.1 and 0.5.3.The improved TACO algorithm and finite element method(FEM)are combined. A structural optimization method based on improved TACO algorithm—FEM (ITACO-FEM) is presented. The method is applied to the structural optimization of two turbine discs. The results show that the given structural optimization method is valid. 4.The ITACO-FEM and structural reliability theory are combined. A structural reliability optimization method is presented. The method is applied to the structural reliability optimization of the two turbine discs. The results show that the structural reliability optimization method based on ITACO-FEM is valid.5.The basic principle and design flow chart of ITACO-FEM and the structural reliability optimization method are introduced. The improved TACO algorithm code is written with Visual C++6.0. FEM Stress analytic code and FEM reliability analytic code are written with APDL. The program is implemented by combining the above codes with data files in condition of Microsoft Visual C++6.0. The program is convenient and universal.