共查询到20条相似文献,搜索用时 815 毫秒
1.
为了减小载荷识别过程中的不适定性,稳定地实现动态载荷的时域重构,提出了基于径向基形函数的时域动态载荷识别方法。在离散化系统动力响应卷积关系式的基础上,将载荷在时间域内划分单元,采用径向基函数构造载荷形函数,并在整个时间域内进行组装,得到整体形函数响应矩阵,建立载荷识别的正向模型;在此基础上,利用正则化方法实现动态载荷识别。算例表明,该方法能够有效减小核矩阵的规模,改善核矩阵的病态性,在采样时间步长较大且响应存在测量噪声的情况下,可保证动态载荷反演的稳定性和精确性。 相似文献
2.
提出一种基于径向基函数神经网络的铣刀磨损监控方法,径向基函数神经网络的输出是刀具磨损的具体值,这样有利于对刀具磨损进行各种实时补偿。实验表明,利用径向基函数神经网络进行状态识别可对小型立铣刀的磨损进行监控,能够取得良好的效果,同时证明RBF网络的训练速度优于BP网络。 相似文献
3.
4.
为了解决复杂机械结构输入载荷难以直接测量的问题,基于结构动力学理论,应用Duhamel积分,推导了一种单自由度振动系统动载荷时域识别方法;在此基础上,通过模态坐标变换法,将多自由度振动系统运动方程解耦,以系统加速度响应作为输入,建立了一种多自由度振动系统动载荷识别的时域方法与模型。以一个多自由度的弹簧阻尼系统为对象,研究了载荷识别模型同时在正余弦组合激励、随机激励、三角脉冲激励等三种载荷形式下的稳定性和精确性。研究结果表明:在测量响应无噪声的情况下,载荷识别模型对三种激励的识别均具有很高的精度,识别结果与原始载荷相关系数达到0.99;在测量响应含有30%随机噪声的情况下,三种载荷的识别结果依然有很高的识别精度,相关系数均在0.97以上。 相似文献
5.
6.
针对虚拟制造中无加工对象数据的加工过程仿真,提出适用于离散无规则边界测量数据的径向基函数网络方法,以重构加工对象曲面,阐述了径向基函数网络曲面重构的基本理论,分析了径向基函数网络宽度参数的确定方法;给出了径向基函数网络曲面重构函数的具体实现过程,计算了重构的误差.以实例验证了该方法的可行性,并与传统曲面拟合方法相比较,得出该方法的重构精度高,改进了传统曲面拟合在处理非均匀截面数据点时曲面形状的失真和运算的不稳定现象,从而显示了径向基函数网络重构曲面的优越性. 相似文献
7.
根据线弹性系统的振动特性,采用动载荷识别频域法研究动载荷作用位置识别问题。提出了当量载荷误差判别法,用不同测点响应在相同位置反演出来的当量载荷的误差值建立目标函数,当量载荷误差最小的作用位置就是动载荷的真实作用位置。从单点激励和两点激励方面进行了平面板的载荷位置识别仿真,在加入噪声的情况下效果依然理想,为复杂结构和多个加载位置动载荷识别提供基础。简支梁实验结果表明,该方法能正确识别载荷作用位置,对噪声不敏感,有很好的工程应用前景。 相似文献
8.
9.
《机械科学与技术》2010,(9)
为了解决自适应共振理论(adaptive resonance theory,ART)网络对新的输入样本处理能力差,网络记忆能力差,径向基(radial basis function,RBF)网络选择径向基函数,确定隐节点数目困难的问题,设计了一种基于ART与RBF网络的混合网络模型。将ART网络的特点引入到RBF网络中,通过ART网络的识别与比较功能快速确定RBF网络最少的隐节点数目,同时通过ART理论中警戒门限的检验在线确定是否合并或删除隐节点。并且引入了异常数据修正方法和模糊预处理方法。通过MATLAB神经网络工具箱,对该混合网络进行仿真试验。结果表明:该方法能够有效地减少隐含在数据中的随机性,加快神经网络收敛速度,提高神经网络的建模精度。 相似文献
10.
11.
变速器渗氮齿轮制造技术的研究 总被引:2,自引:0,他引:2
张元增 《传动技术(上海)》1998,12(4):23-30
本文介绍了用于变速器的高强度渗氮齿轮制造技术的开发与研究工作。目前,这项工作已取得可喜进展,这将为变速器齿轮的精密化开辟一条崭新的途径。 相似文献
12.
为分析磁轴承激励下转子系统的振动机理,应用一维有限元方法建立了双盘转子系统动力学特性计算模型,研究了不同类型磁轴承激励下转子系统的动力学行为。研究结果表明:同向旋转的扫频激励力激发了转子系统的正进动模态,而反向旋转的扫频激励力激发了转子系统的反进动模态,两种情况下转子系统均以圆轨迹进动;由于单向简谐激励力可以分解为同向旋转激励力和反向旋转激励力之和,因此在单向简谐扫频激励力作用下,转子系统的反进动和正进动模态均被激发,转子系统以椭圆轨迹进动;在双向简谐扫频激励力作用下,转子系统的进动方向和进动轨迹取决于两个激励力的相位差。研究成果为评估高压压缩机转子系统稳定性时磁轴承激励方案的选取提供了理论依据。 相似文献
13.
Jin-Rae Cho Ki-Yong Jeong Min-Ho Park Dae-Suk Shin O-Kaung Lim No-Gil Park 《Journal of Mechanical Science and Technology》2013,27(7):2053-2059
Gearbox for wind turbine must be designed to have the sufficient structural strength to sustain the extreme torque and forces transferred from rotor blades. Traditionally, the structural analysis of gearbox has been made using the simplified FEM models in which the contacts between gear teeth are replaced with the equivalent forces acting on the gear shafts, because the consideration of the detailed internal gear transmission system requires a huge number of degrees of freedom. But, the traditional method can neither accurately reflect the gear transmission forces, nor is it preferable for the dynamic analysis. In order to solve these problems, a structural analysis method considering the tooth contact of internal gear system is introduced in this paper. The actual tooth contact between a pair of gears is modeled with spring elements and the spring constants are determined through the stiffness analysis of gear teeth. The current analysis technique is justified through the comparison with the simplified gear system model and applied to the structural analysis of a 2-stage differentialtype gearbox for wind turbine. 相似文献
14.
15.
16.
17.
A method of identifying bearing dynamic coefficients by means of step forces is described. Obtaining an accurate fast Fourier transform of the rapidly varying step function is difficult since the limitations of real time analysis by minicomputer make truncation necessary. This paper therefore proposes a feasible approach to direct analysis of the spectrum of the step function and introduces its application to measure the stiffness and damping coefficients of journal bearings. The approach followed is to use a step exciting force on the test bearings. By using the proposed direct spectrum analysis method, the transfer functions of the bearing systems and eventually the dynamic coefficients of the journal bearings can be derived. 相似文献
18.
The paper introduces a new diagnostic feature, which can be used for monitoring the condition of planetary gearboxes in time-variable operating conditions. The novel approach (originally presented in W. Bartelmus, R. Zimroz, Vibration condition monitoring of planetary gearbox under varying external load, Mechanical Systems and Signal Processing 23 (2009) 246–257) exploits the fact that a planetary gearbox in bad condition is more susceptible (yielding) to load than the gearbox in good condition. The diagnostic method based on the new diagnostic feature is very simple: one needs to capture signals for different external load values and calculate a simple spectrum based feature versus operating conditions indicator (current or instantaneous rotation speed). In a certain range of operating conditions the diagnostic relation (i.e. the dependence between the spectral features and the operating conditions indicator) is linear. However, since a gearbox in bad condition is more susceptible to load than the gearbox in good condition the relation will be different for the two cases. Using a simple regression equation one can calculate the slope of the straight line, which expresses the new diagnostic feature. The method is very quick, technically simple, robust and intuitive. This approach has been used for diagnosing the very complex high-power planetary gearbox used in bucket wheel excavators. 相似文献
19.
20.
In this study, we focus on the submerged plate built into the Very Large Floating Structure with the partial openings of 5m
long, which enables the reverse flow of incident wave to generate the wave breaking. The purpose of this study is to investigate
the characteristics of wave exciting forces acting on the submerged plate and the fore part ofVLFS. Firstly, we have carried out the extensive experiments to understand the characteristics of the wave exciting forces. Then
we have performed the numerical simulations by applying the Marker and Cell method (MAC method) and compared with the experimental results. We discuss the validity ofMAC method and the effects of the submerged plate on the motion of VLFS. As a result, we get the conclusion that the submerged
plate is useful for reducing the wave exciting forces acting on the structure behind the submerged plate. 相似文献