摘要
基于企业预警指标之上的警界线的划分和警度的评判是企业预警管理系统的研究难点。针对企业预警研究存在的问题,将人工神经网络BP模型应用于企业预警系统中。BP网络模型克服了传统模式难以处理高度非线性模型、容错性差、预警线和预警区域不具备时变性、缺少自学习能力等缺陷。通过模型的网络设计和仿真实验,证实了这种方法的有效性和广泛的应用前景。
Based upon the index of enterprise early-warning system, the partition of watch line and judgment of warning degree, are the research nodi in enterprise early-warning management system. In view of existing problem in this field, the paper applies the BP model of Artificial Neural Networks theory to the early-warning system of enterprise. BP model overcomes many limitations of traditional patterns, such as difficulties in processing highly non-linear model, bad performances of fault tolerance, absence of time-varying volatility in Early-warning line and Early-warning region, lack of self-study and the like. Through the designing of model and simulation, the paper proves the effectiveness and wide prosperity of this method.
出处
《西北农林科技大学学报（社会科学版）》
2004年第3期96-100,共5页
Journal of Northwest A&F University(Social Science Edition)
关键词
企业
预警
人工神经网络
BP模型
enterprise
early-warning system
artificial neural networks
BP model
