摘要随着高新技术和现代生产的发展,如飞机等复杂产品的性能以及其结构和功能上的复杂性不断增加,所以在产品投入研发或者使用之前,采用科学的费用估算方法提前对复杂产品的费用进行估算有必要的。但这类复杂产品的费用数据有限,因此本论文对小样本条件下基于支持向量回归机的费用估算展开了研究,并集成于一个通用的面向复杂产品的费用估算系统。主要工作如下:
(1)在介绍课题背景及相关方法的国内外研究现状的基础上,给出了传统费用估算的方法及支持向量回归机的原理等;27038
(2)对小样本条件下的基于SVR的费用估算算法进行了分析,详细阐述了算法的流程及数据处理方法,并给出实例对算法的可行性进行了验证;
(3)设计了基于SVR的费用估算子系统的架构、数据库、类与函数;
(4)实现了基于SVR的费用估算算法,并集成于总系统,通过界面展现了实现效果。
关键词 费用预测,复杂产品,小样本,支持向量回归机 毕业论文设计说明书外文摘要
Title Research and Implementation of Cost Estimation Method of Complex Products under the Condition of Small Sample
Abstract
With the development of high and new technology and modern production, the performance of aircraft or other complex products and the complexity of their structure and function increase continually. So it is necessary to use scientific cost estimation method to estimate the costs of complex products in advance before the products are put into research or use. But this kind of complex product cost data is limited, so the cost estimation based on support vector regression under the condition of small sample is studied, and has been integrated with a general cost estimation system for complex products in this paper. The main work is as follows:
(1)On the basis of introducing the background of project and the overseas and domestic research status of related methods, the method of the traditional cost estimation and the principle of support vector regression are given.
(2)The cost estimation algorithms based on SVR under the condition of small sample are analyzed, the process of the algorithms and the method of data processing are described in detail, and the feasibility of the algorithm is verified by the examples.
(3)The architecture, database, class and function of the cost estimation subsystem based on SVR are designed.
(4)The cost estimation algorithms based on SVR ,which are integrated into the total system, are realized, and the effect is realized through the interface.
Keywords Cost estimation,Complex products,Small sample,Support vector regression
目 次
1 绪论 1
1.1 问题研究背景 1
1.2 国内外研究现状 1
1.2.1 费用估算算法研究现状 1
1.2.2 支持向量回归机(SVR)的国内外发展现状 2
1.3 本文主要内容与组织结构 3
2 相关技术和概念 4
2.1 传统费用估算算法 4
2.2 支持向量回归机原理 5
2.2.1 支持向量回归机算法 5
2.2.2 SVR在小样本条件下费用估算的优势 6
2.3 开发所要使用的技术及工具 6
2.3.1 Ext JS前端框架简介 6
2.3.2 ORACLE数据库简介 6
2.3.3 C#语言简介 7
3 小样本条件下基于SVR的费用估算算法设计 8
3.1 需求分析 8
3.2 算法描述 10
3.2.1 算法流程图 10
3.2.2 基于支持向量机的费用估算 10