共查询到18条相似文献,搜索用时 62 毫秒
1.
2.
基于BP神经网络的钢轨磨损量预测 总被引:1,自引:1,他引:0
随着列车运行速度和轴重的提高,轮轨系统的磨损越来越严重,其中曲线半径、轴重和运行速度是影响轮轨磨损的重要因素。建立了钢轨磨损量影响规律的径向BP基函数神经网络模型,该网络具有3路输入,3个神经层;在JD-1大型轮轨模拟试验机上通过改变试验参数进行钢轨磨损试验,获得不同试验参数下的钢轨磨损量;以钢轨磨损数据作为BP神经网络的目标样本,对不同试验参数下的磨损量进行了预测。结果表明,模型可较准确地计算轮轨冲角和速度对钢轨磨损量的影响规律,利用BP神经网络对钢轨磨损量预测具有较高的精度,可在一定程度上验证试验结果。 相似文献
3.
4.
5.
6.
7.
8.
针对42CrMo钢精密切削刀具磨损量预测研究小样本、非线性的特点,将量子粒子群算法(QPSO)、卷积神经网络(CNN)及长短期神经网络(LSTM)相结合,构建了QPSO-CNN-LSTM组合预测模型。采用QPSO算法对CNN-LSTM模型的隐藏层单元数、学习率、卷积核等进行优化,结合CNN网络特征提取能力强、LSTM网络具备记忆能力的特点,对实际加工实验的刀具磨损量进行预测,并通过误差评价指标分析,与CNN、LSTM、BP等单一模型以及PSO-GRNN组合模型进行预测效果对比研究。研究结果表明,本文构建的组合预测模型相对于单一预测模型,其预测值与真实值吻合程度更高;相对于PSO-GRNN组合模型,三种误差评价指标的误差值至少降低了27%,其泛化性和稳定性较好,预测精度与非线性拟合能力更强。 相似文献
9.
库存控制是电信供应链管理过程中重要技术之一,针对BP神经网络库存预测方法中存在的局部极小问题,引入GA-BP网络算法,将影响库存控制的因素抽象出来,并且把电信某物资的库存历史数据作为训练样本,对库存进行有效预测。经实验证明,GA-BP算法不仅避开BP网络的弊端,继承神经网络很强的学习、训练能力,同时也提高了库存的预测精度。 相似文献
10.
库存控制是电信供应链管理过程中重要技术之一,针对BP神经网络库存预测方法中存在的局部极小问题,引入GA-BP网络算法,将影响库存控制的因素抽象出来,并且把电信某物资的库存历史数据作为训练样本,对库存进行有效预测.经实验证明,GA-BP算法不仅避开BP网络的弊端,继承神经网络很强的学习、训练能力,同时也提高了库存的预测精度. 相似文献
11.
12.
13.
针对BP算法在系统辨识应用中的不足,提出了一种基于遗传算法(GA)的BP神经网络建模方法.该算法充分利用了遗传算法和BP算法各自的优点.采用该算法完成了具有复杂非线性的某伺服系统的建模工作.实验结果表明了该算法的有效性. 相似文献
14.
针对目前钢轨强化采用的欠速淬火方法存在强化层硬度偏低、耐磨性难以满足重载线路使用要求的问题,使用3种不同光斑宽度的激光,研究了U71Mn材质钢轨的激光淬火强化工艺,获得了不同扫描速度下的临界功率和淬火层深,并测试了淬火层在滚动接触条件下的磨损与接触疲劳性能。结果表明:在临界熔化的激光能量密度下,光斑宽度由6 mm增加到20 mm时,淬火层深度提高了38%,或在获得相同的淬火层深度情况下,处理效率提高6.8倍;淬火层组织为针状马氏体,硬度从原来的300HV提高到800HV以上;20万周次的磨损试验后,激光淬火试样的磨损量只有未处理试样磨损量的25%,未处理试样以表面接触疲劳剥落和塑性变形为主,激光淬火试样仅有轻微的疲劳磨损,耐磨性和抗接触疲劳性能优异。 相似文献
15.
16.
P. Palanisamy I. Rajendran S. Shanmugasundaram 《The International Journal of Advanced Manufacturing Technology》2008,37(1-2):29-41
Tool wear prediction plays an important role in industry for higher productivity and product quality. Flank wear of cutting
tools is often selected as the tool life criterion as it determines the diametric accuracy of machining, its stability and
reliability. This paper focuses on two different models, namely, regression mathematical and artificial neural network (ANN)
models for predicting tool wear. In the present work, flank wear is taken as the response (output) variable measured during
milling, while cutting speed, feed and depth of cut are taken as input parameters. The Design of Experiments (DOE) technique
is developed for three factors at five levels to conduct experiments. Experiments have been conducted for measuring tool wear
based on the DOE technique in a universal milling machine on AISI 1020 steel using a carbide cutter. The experimental values
are used in Six Sigma software for finding the coefficients to develop the regression model. The experimentally measured values
are also used to train the feed forward back propagation artificial neural network (ANN) for prediction of tool wear. Predicted
values of response by both models, i.e. regression and ANN are compared with the experimental values. The predictive neural
network model was found to be capable of better predictions of tool flank wear within the trained range. 相似文献
17.
18.
W. R. Tyfour 《Tribology Letters》2008,29(3):229-234
Rail grinding has become an increasingly permanent way maintenance practice to tackle rail corrugation, as well as extending
the rolling contact fatigue life of rails. However, and as far as material loss is concerned, such a grinding is considered
as an artificial wear process added to the natural wear. The work presented in this article investigates the quantitative
effect of grinding the whole deformed rail surface layer on the overall wear process of the running surface of pearlitic rail
steel. Results show that if wear behavior is known as an empirical model which can predict the effect of grinding that could
be obtained. 相似文献