论文部分内容阅读
描述了以小子样机械系统寿命序列为代表的时间序列系统的浑沌特性,阐述了灰色关联分析和灰色模型在机械产品可靠性分析与预测中可开拓贫信息的工程背景,论述了GM 和BP预测方法在机械系统寿命预测时各自的特长和缺陷,以及作者创立的GM+ BP预测方法具有扬长避短、优势互补的优良特性,同时提出BP神经网络的分形问题和短序列BP网络的构造过程,并以3 个工程实例进一步验证了灰色关联分析和GM+ BP方法的优良特性。本文旨在应用浑沌、分形理论对重大机械产品可靠性分析与预测的小样本开发有所突破。
The chaotic characteristics of the time series system represented by the life span sequence of small submachine mechanical systems are described. The engineering background of gray relational analysis and gray model which can exploit the poor information in the reliability analysis and prediction of mechanical products is expounded. The GM and BP The prediction methods have their own advantages and disadvantages in the prediction of mechanical system life and the good features of GM + BP prediction method founded by the author. They also provide the fractal of BP neural network and the construction process of BP network with Three engineering examples further verify the excellent characteristics of gray relational analysis and GM + BP methods. This paper aims to make breakthroughs in the development of small samples for the reliability analysis and prediction of major mechanical products by using chaos and fractal theory.