| Genetic Algorithm (GA) is a method for searching for the optimum solution to a complex problem, based on the mechanics of natural selection, the process of evolution. It is flexible and robust. Its goal is to achieve similar breadth of performance by abstracting nature抯 adaptation algorithm of choice in artificial form. GA is basically an automated, artificial intelligent approach to trial and error. In the aspect of solving large, non-linear and poorly understood problems where expert knowledge is scarce or difficult to encode and traditional methods fail, GA has great advantages. It is one of kernel techniques related with intelligent computing in 21 century.A multi-object combinatorial optimize problem glass-block typesetting optimize problem solved by genetic algorithm is studied. The data related with study on optimization of glass-block typesetting based on GA has not been found at home and abroad. Some new methods are provided. The glass-block typesetting optimization system and its optimization algorithm are designed and realized.The primary work of this paper as follows: In order to lessen the difficulty of solving, the multi-object problem is disassembled into several single object problems, and these problems are solved one by one by applying Genetic Algorithm. Some tryouts on treatment for the basic realization technique of GA are made. Different decoding methods is used to calculate the fitness. Running frame of different GA is constructed. Different coding method is used to code chromosome, Different genetic operators in allusion to different coding are designed for the sake of guaranteeing the validity of individuals. Niche technique and hybrid GA is applied to boost up local search ability of GA. Summed up some treat skills about disassembling a large, complex and multi-object problem and solving it one by one. And the ways of improvement on every GA is also provided. Finally, the solution relationship among the three problems is analyzed, and some conclusion are obtained, which as follows: CD When the number of glass-block that will be incised is few and the total area of the glassblock exceed the area of glass board, different algorithms used to solve the knapsack problem, the differences of the solutions are little, and the influence on the system抯 optimized result is little, and the result is foreign to the specifications of glass-block.?When the number of glass-block that will be incised is more, different algorithms used to solve the knapsack problem, the differences of the solutions are great, and which have great influence on the solution of layout design problem and the optimized result of the system.ç”'hen the specifications of glass-block that will be incised differ, for a same passel of glass-block, if GA used to solve the knapsack problem use penalize method to decode, in its solution the glass-block whose area is big is prone to be selected into glass board, however, if the GA use greedy method to decode, in its solution the glass-block selected into glass board isIIIcomparatively random.@I~Ythe rational typesetting scheme found out by solving layout design problem is not always a optimized scheme, however, a more optimized typesetting scheme is more in favor of solution of TSP, that is it is more easy to find out a more optimized incise path.Chapter 1 introduced the work of the paper and its background. Compared traditional methods used to solve combinatorial optimization problem and GA. Chapter 2 introduced genetic algorithm systematically, which include the relationship between GA and natural evolution and the characteristics of genetic algorithm, basic implement techniques of (IA and so on. Chapters 3 presented glass-block typeset optimizing system and it optimize algorithm. Chapter 4 discussed in detail the design and implement of (IA used to solve the.9knapsack problem in the system. Chapter 5 discussed in detail the design and implement of GA used to solve the layout problem in the system. And the relationship between knapsack proble... |
PDF Full Text Request