基于BP神经网络的液压缸市场价格估算模型.pdfVIP

中国科技论文在线 基于BP 神经网络的液压缸市场价格估算模 型 周敏，高挺，陈艳霞* 5 （武汉科技大学机械自动化学院，武汉430081 ） 摘要：液压缸品种多、使用量大、价格变化范围大，其价格估算对采购计划制定、维修及再 制造经济性评价等具有重要意义。针对液压缸价格估算的影响因素多、难以建立直接的数学 模型的问题，本文采用德尔菲法，通过总结本领域多位专家的经验知识，找出了内径、行程、 10 压力等级、是否伺服缸、安装形式、厂家资质和生产者物价指数等7 个影响液压缸价格的关 键因素，建立了基于BP 神经网络的液压缸市场价格估算模型，最后用实例验证了该模型的 有效性。模型中所引入的生产者物价指数，可以提高该模型对市场环境变化的敏感度，使基 于该模型的市场价格估算准确度更高。 关键词：工业工程；液压缸；BP 神经网络；市场价格；估算模型 15 中图分类号：F272.3 A Market Prices Forecast Model of Hydraulic Cylinders via BP Neural Networks ZHOU Min, GAO Ting, CHEN Yanxia 20 (College of Machinery and Automation,Wuhan University of Science and Technology,Wuhan 430081) Abstract: Hydraulic cylinders are widely used with various kinds and a large variation range of price. It has great significance to establish hydraulic cylinder market price forecast model for the issues of making procurement plan and economical efficiency evaluation of maintenance and 25 remanufacturing. However, there are many influencing factors of hydraulic cylinder market price and it is hard to establish a direct mathematical model. The objective of this paper is to develop a forecast model of market price for the hydraulic cylinder via the application of back-propagation (BP) neural networks. Seven critical factors related to market price of hydraulic cylinder have been found out by the means of Delphi, which are cylinder bore diameter, stroke, pressure rating, 30 servo cylinder or not, installation type, manufacturer qualification and Producer Price Index (PPI). Case study is used to prove that neural