Dissertation > Excellent graduate degree dissertation topics show

Solution of the problem of one-dimensional cutting an improved heuristic algorithm

Author: LiuZuo
Tutor: CuiYaoDong
School: Guangxi Normal University
Course: Applied Computer Technology
Keywords: Cutting stock One -dimensional cutting Heuristic Algorithm Multi-objective optimization
CLC: O242.1
Type: Master's thesis
Year: 2009
Downloads: 51
Quote: 1
Read: Download Dissertation

Abstract

Production in the national economy, there is a lot of cutting stock problem. One dimensional cutting stock problem is the raw material and the required dimensions are blank one dimension, in the case of supply conditions are known to consider how to optimize the cutting stock, making the dough needs met and to maximize material utilization, reduce cutting losses. One dimensional cutting stock problem in engineering technology and industrial production has an important and widely used, to discuss the issue is to study two-dimensional, three-dimensional cutting issues such as multi-dimensional basis. With the information technology industry and the rapid development of computer technology, advanced computer-aided design technology is also increasingly being used in the cutting process. Thus, methods of solving the problem in the practical application and theoretical have important significance. This paper discusses a variety of cutting the length of the wire on the issue on heuristic analysis and study, based on the use of an improved heuristic algorithm for sequential one-dimensional cutting problem solving. According to the actual needs of the production environment, through a series of improvement strategies, the algorithm is further optimized to ensure high material utilization while reducing nesting consider ways to increase more than expected last a length of material, the priority use of short materials multiple optimization objectives. Through a large number of instances of the test, confirmed the effectiveness of the algorithm. The main work and innovations are as follows: First, for the research question, given the current optimal nesting methods for solving the mathematical model, the model based on the basic idea of algorithm and implementation process. This paper uses a heuristic based on the order value of the correction algorithm. Each order value correction means generates a new nesting mode, it will use the previous information, a variety of blank correction value and the process is repeated several times, and ultimately make it more reasonable value of coefficient of the state. By adjusting the blank value, their relative popularity reflected on the bad row of blanks given a higher priority, so that priority is selected. Doing so helps build a better way of nesting, and use the information in front of ways to guide subsequent nesting process, thus effectively improve the utilization of raw materials. Based on current optimal nesting mode calculation model, knapsack problem, generating the largest total value of cutting patterns. Consider increasing the number of effective ways of nesting, increase choice. Therefore, in the already acquired nesting mode, select affecting its repetitions of rough to be replaced, get more nesting better way. Judge, followed by selecting the appropriate way to compose the current nesting nesting program. Iterations perform this procedure, the better preserved nesting results. Further refined and improved algorithm. By using a variety of strategies and parameter optimization heuristic methods to improve material utilization and reduce the number of nesting way to simplify the production process. Blank row pre-treat group, each group using the blank candidate generate the current nesting mode, blank subset selection can be an effective way to reduce the number of nesting; design nesting mode selection criteria, priority use a shorter raw materials to reduce the storage capacity; recorded for each of the last nesting nesting program the way I expected length of the algorithm is repeated several times, the better part of the nesting program to save, the preferred solution longer than expected to achieve, I reached to facilitate material recycling and reuse, reduce production costs. Taking into account the diversity of the problem, the algorithm parameters involved in the control calculation using cyclic manner. Corresponding to each set of parameter values, perform multiple iterative algorithm to generate a large number of nesting program. Through multiple calculations, select the best results. Finally, the planning and design of cutting the system's basic function modules, developed a heuristic algorithm based on improved sequence of one-dimensional optimization unloading system. Through a large number of experimental tests, and the results compared with a number of new optimization algorithms to compare the experimental results and analysis, the results show that the proposed algorithm is higher material utilization and cutting patterns to achieve a small number of priority short material, a material more than expected last a long length and many other optimization goal is an effective one-dimensional cutting stock problem solving heuristic algorithms.

Related Dissertations

CLC: > Mathematical sciences and chemical > Mathematics > Computational Mathematics > Mathematical modeling, approximate calculation > Mathematical modeling
© 2012 www.DissertationTopic.Net Mobile