Application Of Genetic Algorithms On Optimal Design Of Supporting Structure In Deep Excavations

The optimal design of the supporting structure in the deep excavations is a complex task that deals with many subjects. It needs various knowledge and professional experience for the structural design. Nowadays, the task done by the scholars inside and outside is not perfect, especially in the selection of the optimal methods. These methods are often based on the traditional operational research which is assumed that the design variables are continuous and the aim function is derivative. But in fact, in the design of the deep excavations, which deals with many factors, the variables are discrete and the aim function has more than one peak value. So the traditional optimal methods often run into the local optimal value. In view of these problems, a method based on the genetic algorithms has been put forward for the optimal design of the supporting structure in the deep excavations. Its takes full advantage of the characteristics of the genetic algorithms such as general use, robust nature, simultaneous disposal and good fitness. It is shown from this paper that the genetic algorithms is an excellent method used in the optimal design of the deep excavations. It is proved by the project that this method possesses a powerful vitality.At first, the category, breaking forms and the function and characteristics of thedeep excavations are introduced, and a series of problems is analyzed. The status quothat the genetic algorithms applies to the structural optimization is analyzed in detail.The methods for the optimal design in the deep excavations are discussed, and the difference is compared between the traditional optimal methods and the modern ones, it safely conclude that the modern optimal methods overmatch the traditional ones. The theory and the implementation of the genetic algorithms are discussed in detail. The question on how to choose the crossover probability, the mutation probability, the scale of population and the numbers of the generation is discussed. Then, the mathematics model of the optimal design is established. Exerting the finite element of elastic base beam and using FORTRAN to write the GAFORTRAN program. The program includes the common genetic algorithms and the modified genetic algorithms. It is shown by the practical engineering that the genetic can be used to analyzed the optimal design in the deep excavations and compared with the traditional method, the results is more optimal. To modify the common genetic algorithms, the niche technique and reserving the best individual are added to it. It is proved by the practical engineering that the modified genetic algorithms has better convergence performance. In order to expedite the speed of convergence, reserving the best individual do not take part in the next generation, thus increasing the speed of convergence and enhancing the efficiency to cut down the running time.