论文部分内容阅读
对减速机故障形式进行分析,采用BP神经网络建立故障诊断模型,利用遗传算法优化神经网络权值、阈值、网络结构.将遗传神经网络模型应用到远程减速机的故障诊断的设计中.比较单一的神经网络和遗传神经网络的训练误差曲线,得出遗传神经网络在训练速度和准确性上远远高于神经网络训练模型.
基于遗传算法和神经网络的远程减速机故障诊断系统的设计
【摘 要】
：
对减速机故障形式进行分析,采用BP神经网络建立故障诊断模型,利用遗传算法优化神经网络权值、阈值、网络结构.将遗传神经网络模型应用到远程减速机的故障诊断的设计中.比较单一的神经网络和遗传神经网络的训练误差曲线,得出遗传神经网络在训练速度和准确性上远远高于神经网络训练模型.
【机 构】
：
南通职业大学 机械系 江苏南通 226007
【出 处】
：
第三届中国智能计算大会
【发表日期】
：
2009年6期
其他文献
In order to establish direction fuzzy clustering,direction fuzzy relations and matrices are introduced in this paper.Furthermore,some basic properties of them are studied.Finally,two direction fuzzy c
The Expected Value Programming Model with Net Present Value for Multiproduct Aggregate Production Pl
In this paper,net present value is taken into account while discussing multiproduct aggregate production planning decision making problems in bifuzzy environments.Then,an expected value programming mo
In order to solve the portfolio problem when security returns are birandom variables,firstly we propose a definition of risk on the basis of the average chance of birandom variable,then one type of po
A Hybrid Self Organizing Migrating Algorithm for Solving Fourth-Party Logistics Routing Problem with
From the beginning of the 21st century,fourth party logistics(4PL)has been attracting more and more attention in the field of academic and industry.In this paper,a 4PL routing problem(4PLPR)with fuzzy
Convergence is a very important concept in credibility theory.There are four main types of fuzzy variables in credibility theory:equipossible fuzzy variable,triangular fuzzy variable,trapezoidal fuzzy
This paper proposes the new concept of fuzzy maximum likelihood estimator based on credibility distribution of fuzzy variables.The invariance properties of fuzzy maximum likelihood estimator are inves
Given a fuzzy variable,what is the degree of difficulty of predicting the specified value that fuzzy variable will take? Entropy provides a measure to characterize the fuzziness.As we know,image segme
In this paper,we present a multidimensional model describing the hybrid financial market.The definition of no-arbitrage,No-arbitrage theorem and determinant theorem of no-arbitrage are given based on
设A是Banach空间X中的闭的稠定线性算子,且A生成一个指数有界的(C)-正则预解算子族[R(t)]t≥0.本文主要研究了(C)-正则预解算子族的加法扰动,并给出了扰动后的算子族的公式表示.
Bonferroni curve and Bonferroni index are used to provide a sensitive measure of income inequalities.In this paper,we proposes the new concept of credibilistic Bonferroni curve and Bonferroni index ba