基于遗传算法的关联规则在AGV系统中的研究与应用

发布时间：2019-06-09 12:11

【摘要】：自动导引车(Automated Guided Vehicle，AGV)是现代物流系统备受关注的关键设备。越来越多的大中型企业开始应用AGV系统，尝试以自动货运机器人来替代人工作业，节省人力资源成本，逐步实现工业自动化。AGV系统运行过程中积累了大量的无规则数据。如何利用数据挖掘技术对AGV系统的数据进行有效的分析，从中提取出有用的信息，并利用这些信息提高AGV系统的运行效率是一个值得研究的问题。本文提出了一种基于遗传算法的关联规则方法，并对AGV系统中的数据进行关联分析。主要内容如下：阐述了数据挖掘、遗传算法和关联规则的相关知识。针对“支持度-置信度”关联规则模型的不足，引入理解度和兴趣度这两个评价标准,根据支持度、置信度、理解度和兴趣度综合评价一条关联规则。针对进行关联规则挖掘时，计算各个评价标准要重复扫描数据库的问题，提出了一种属性目录结构，根据该结构能够有效地减少扫描数据库的次数，从而减少关联规则挖掘的时间。根据遗传算法的全局寻优的特点，提出了一种基于遗传算法的关联规则算法，详细介绍了该算法的染色体编码方式，使用支持度、置信度、理解度和兴趣度构造适应度函数并结合属性目录计算适应值，产生初始群体，设计遗传算子等方面。最终将该算法应用于AGV系统中，得到一些有价值的关联规则，并将结果与其他算法进行比较，，证明该算法的高效性。通过分析解释这些规则，对AGV系统优化，仓库货物安排，货物备货量，工作人员分配等方面提供有价值的信息。
[Abstract]:Automatic guided vehicle (Automated Guided Vehicle,AGV) is the key equipment of modern logistics system. More and more large and medium-sized enterprises begin to apply AGV system, try to replace manual operation with automatic freight robot, save human resource cost and realize industrial automation step by step. AGV system accumulates a lot of irregular data in the process of operation. How to use data mining technology to effectively analyze the data of AGV system, extract useful information from it, and use this information to improve the operation efficiency of AGV system is a problem worthy of study. In this paper, an association rule method based on genetic algorithm is proposed, and the data in AGV system are analyzed. The main contents are as follows: the related knowledge of data mining, genetic algorithm and association rules is described. In view of the shortcomings of the "support-confidence" association rule model, two evaluation criteria, understanding degree and interest degree, are introduced to evaluate an association rule according to the degree of support, confidence, understanding and interest. In order to solve the problem that each evaluation standard should scan the database repeatedly when mining association rules, an attribute directory structure is proposed, according to which the number of scanning databases can be effectively reduced. In order to reduce the mining time of association rules. According to the characteristics of global optimization of genetic algorithm, an association rule algorithm based on genetic algorithm is proposed. The chromosome coding method, support degree and confidence level of the algorithm are introduced in detail. The fitness function is constructed by understanding degree and interest degree, and the fitness value is calculated by combining the attribute directory to generate the initial population, design genetic operator and so on. Finally, the algorithm is applied to AGV system, and some valuable association rules are obtained, and the results are compared with other algorithms to prove the efficiency of the algorithm. Through the analysis and interpretation of these rules, the optimization of AGV system, warehouse cargo arrangement, cargo reserve, staff distribution and other aspects of valuable information.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP311.13

【参考文献】