DEVELOPMENT AND VIBRATION ANALYSIS OF CERAMIC ROCKER ARM

ＤＥＶＥＬＯＰＭＥＮＴＡＮＤＶＩＢＲＡＴＩＯＮＡＮＡＬＹＳＩＳＯＦＣＥＲＡＭＩＣＲＯＣＫＥＲＡＲＭＤＥＶＥＬＯＰＭＥＮＴＡＮＤＶＩＢＲＡＴＩＯＮＡＮＡＬＹＳＩＳＯＦＣＥＲＡＭＩＣＲＯＣＫＥＲＡＲＭＹａｎｇＨｅｌａｉ；ＣｈｅｎｇＧｅｎｇ；ＳｕＦｅｎｇ（...     

STUDY OF THE 3-DIMENSIONAL STRESSES OF COLD STRIP ROLLING UNDER CENTRE WAVE

ＳＴＵＤＹＯＦＴＨＥ３－ＤＩＭＥＮＳＩＯＮＡＬＳＴＲＥＳＳＥＳＯＦＣＯＬＤＳＴＲＩＰＲＯＬＬＩＮＧＵＮＤＥＲＣＥＮＴＲＥＷＡＶＥ￥ＬｉｕＨｏｎｇｍｉｎ；ＢｉａｎＹｕｈｏｎｇ；ＬｉａｎＪｉａｃｈｕａｎｇ（ＹａｎｓｈａｎＵｎｉｖｅｒｓｉｔｙ）Ａｂｓｔｒａ...     

ANALYSIS OF TOOTH FORM ERROR OF EQUAL BASE CIRCLE BEVEL GEAR

ＡＮＡＬＹＳＩＳＯＦＴＯＯＴＨＦＯＲＭＥＲＲＯＲＯＦＥＱＵＡＬＢＡＳＥＣＩＲＣＬＥＢＥＶＥＬＧＥＡＲＡＮＡＬＹＳＩＳＯＦＴＯＯＴＨＦＯＲＭＥＲＲＯＲＯＦＥＱＵＡＬＢＡＳＥＣＩＲＣＬＥＢＥＶＥＬＧＥＡＲＧｏｎｇＹｕｎｐｅｎｇ；ＤｉｎｇＳｈｉｃｈａｏ；Ｃ...     

RESEARCH ON CAD/CAM FOR HY-DRAULIC MANIFOLD BLOCKS

ＲＥＳＥＡＲＣＨＯＮＣＡＤ／ＣＡＭＦＯＲＨＹ－ＤＲＡＵＬＩＣＭＡＮＩＦＯＬＤＢＬＯＣＫＳＲＥＳＥＡＲＣＨＯＮＣＡＤ／ＣＡＭＦＯＲＨＹ－ＤＲＡＵＬＩＣＭＡＮＩＦＯＬＤＢＬＯＣＫＳＸｉｅＬｉｎ；ＤｏｎｇＨｕａ；ＬｉｕＮｅｎｇｈｏｎｇ；ＳｕｎＱｉｎｇｈｕａ...     

POWER FLOW ANALYSIS FOR A NEW ISOLATION SYSTEM-FLEXIBLE FLOATING RAFT

ＰＯＷＥＲＦＬＯＷＡＮＡＬＹＳＩＳＦＯＲＡＮＥＷＩＳＯＬＡＴＩＯＮＳＹＳＴＥＭ－ＦＬＥＸＩＢＬＥＦＬＯＡＴＩＮＧＲＡＦＴＰＯＷＥＲＦＬＯＷＡＮＡＬＹＳＩＳＦＯＲＡＮＥＷＩＳＯＬＡＴＩＯＮＳＹＳＴＥＭ－ＦＬＥＸＩＢＬＥＦＬＯＡＴＩＮＧＲＡＦＴＸｉｏｎｇ...     

STUDY ON DOUBLE CIRCULAR ARC PROFILE SPIRAL BEVEL GEAR TRANSMISSION

ＳＴＵＤＹＯＮＤＯＵＢＬＥＣＩＲＣＵＬＡＲＡＲＣＰＲＯＦＩＬＥＳＰＩＲＡＬＢＥＶＥＬＧＥＡＲＴＲＡＮＳＭＩＳＳＩＯＮＳＴＵＤＹＯＮＤＯＵＢＬＥＣＩＲＣＵＬＡＲＡＲＣＰＲＯＦＩＬＥＳＰＩＲＡＬＢＥＶＥＬＧＥＡＲＴＲＡＮＳＭＩＳＳＩＯＮＬｉＪｉｎｂａｏ；...     

OPTIMAL DYNAMIC LOAD DISTRIBUTION OF MULTIPLE COOPERATING ROBOT MANIPULATORS

ＯＰＴＩＭＡＬＤＹＮＡＭＩＣＬＯＡＤＤＩＳＴＲＩＢＵＴＩＯＮＯＦＭＵＬＴＩＰＬＥＣＯＯＰＥＲＡＴＩＮＧＲＯＢＯＴＭＡＮＩＰＵＬＡＴＯＲＳ￥ＺｈａｏＹｏｎｇｓｈｅｎｇ；ＨｕａｎｇＺｈｅｎ；ＧａｏＦｅｎｇ（ＹａｎｓｈａｎＵｎｉｖｅｒｓｉｔｙ）Ａｂｓｔｒａ...     

DISTRIBUTION OF REACHABLE WORKSPACE AREAS IN PHYSICAL MODEL OF THE SOLUTIONS PACE OF 2－DOFPARALLEL PLANAR MANIPULAT

ＤＩＳＴＲＩＢＵＴＩＯＮＯＦＲＥＡＣＨＡＢＬＥＷＯＲＫＳＰＡＣＥＡＲＥＡＳＩＮＰＨＹＳＩＣＡＬＭＯＤＥＬＯＦＴＨＥＳＯＬＵＴＩＯＮＳＰＡＣＥＯＦ２－ＤＯＦＰＡＲＡＬＬＥＬＰＬＡＮＡＲＭＡＮＩＰＵＬＡＴＯＲＳＤＩＳＴＲＩＢＵＴＩＯＮＯＦＲＥＡＣＨＡＢＬ...     

FAST GENERATION OF OPTIMAL ASSEMBLY PLANS

ＦＡＳＴＧＥＮＥＲＡＴＩＯＮＯＦＯＰＴＩＭＡＬＡＳＳＥＭＢＬＹＰＬＡＮＳＦＡＳＴＧＥＮＥＲＡＴＩＯＮＯＦＯＰＴＩＭＡＬＡＳＳＥＭＢＬＹＰＬＡＮＳＱｉａｎＷｅｎｈａｎ（ＳｈａｎｇｈａｉＪｉａｏｔｏｎｇＵｎｉｖｅｒｓｉｔｙ）ＡｂｓｔｒａｃｔＥｘｉｓｔｉｎ...     

NEW METHOD FOR VIBRATION MEASUREMENT AND CONTROL OF MOVING WIRE

ＮＥＷＭＥＴＨＯＤＦＯＲＶＩＢＲＡＴＩＯＮＭＥＡＳＵＲＥＭＥＮＴＡＮＤＣＯＮＴＲＯＬＯＦＭＯＶＩＮＧＷＩＲＥＨｕＸｕａｎｌｉ；ＣｈｅｎＴｉａｎｎｉｎｇ；ＴａｎｇＹｏｎｇｊｉｅ；ＺｈａｎｇＳｈｅｎｇｂｉ（Ｘｉ＇ａｎＪｉａｏｔｏｎｇＵｎｉｖｅｒｓｉｔｙ）...     

基于自适应模糊聚类的多模型染纱能耗软测量

针对染纱生产的工艺能耗测量问题,提出一种基于自适应模糊聚类的多神经网络的染纱能耗软测量方法.该方法采用自适应模糊C均值聚类算法,基于实时采集的样本数据,将训练集划分成不同聚类中心的子集,并自适应修正.每个子集用径向基函数网络训练得到子模型,然后根据聚类后的隶属度,将各子模型的输出加权求和获得最后结果.通过对染缸能耗软测量建模,并对其进行仿真和典型实例研究,表明该方法具有良好的预测精度和鲁棒性,且与制造执行系统结合具有良好的在线测量能力.     

g<1永磁直线同步电动机齿槽力波动的径向基神经网络预估器

永磁直线同步电动机(Permanent magnet linear synchronous motor,PMLSM)齿槽力是影响电动机性能的主要因素之一,特别是在高精度,低速情况下,问题尤为突出。根据q<1分数槽绕组PMLSM结构,采用有限元法计算齿槽力的影响,建立以径向基神经网络为基础的PMLSM齿槽力预估器,其学习算法首先采用快速模糊C均值算法(Accelerated fuzzy C-means,AFCM)对数据进行聚类,选取基函数传播因子,再由最小正交平方算法(Orthogonal least squares learning algorithm,OLSA)选取中心矢量,该预估器与带动量的BP网络(Back propagation neural net- work,BPNN)预估器相比较表明,能够在加快网络学习速度的前提下,保证精度,缩小网络规模,提高网络分类能力。试验结果表明,采用q<1分数槽绕组PMLSM能够有效地减小齿槽力的影响。预估器的建立,能够在设计阶段对PMLSM齿槽结构参数进行有效地预估,使得电动机在满足推力波动指标条件下,实现快速敏捷设计,提高PMLSM的整体设计水平。     

RBFNDOB-based neural network inverse control for non-minimum phase MIMO system with disturbances

An adaptive control strategy combining neural network inverse controller (NNIC) with RBFN disturbance observer (RBFNDOB) is developed for a multi-input-multi-output (MIMO) system with non-minimum phase, internal and external disturbances in this paper. Since the inverse model of system is unstable due to the non-minimum phase, a pseudo-plant is constructed, then the RBFN is used to identify the inverse model of pseudo-plant, which can track the parameter variations of system. By copying the structure and parameters of the identifier, the NNIC is obtained. Cascading the NNIC with the original plant, the MIMO system can be decoupled and linearized into independent SISO systems. For the independent decoupled system, the RBFNDOB employs a RBFN to observe the external disturbances and this estimate value is used as a feed-forward compensation term in controller. The case study on ball mill grinding circuit is presented. The effectiveness of the proposed method is demonstrated by simulation results and comparisons.     

一种通用的RBF网络非线性动态系统逆控制

针对任意复杂非线性系统,提出一种通用的RBF网络动态系统逆控制方案,并提出了一种通过对聚类半径进行“粗调”和“细调”的最近邻聚类学习算法,利用构造伪系统的方法构成一种对非最小相位同样有效的神经网络动态逆控制与PID控制相结合的在线自学习控制策略。仿真研究证明,该控制策略不仅能使多种非线性对象跟踪多种参考信号,而且抗干扰能力和鲁棒性也很好。     

Comparison between response surface methodology and radial basis function network for core-center drill in drilling composite materials

Drilling using twist drill is the most frequently used secondary machining for fiber-reinforced composite laminates and delamination is the most important concern during drilling. The drill design and drilling parameters associated with thrust distribution on the drilling-induced delamination are presented. The core-center drill has been found to be more advantageous than the core drill in reference and practice experiences. Response surface methodology (RSM) is a very practical, economical, and useful tool for the modeling and analysis of experimental results using polynomials as local approximations to the true input/output relationship. Due to the radial basis function network’s (RBFN) fast learning speed, simple structure, local tuning, and global generalization power, researchers in the field of manufacturing engineering have been using RBFN in nonlinear manufacturing studies. The present paper compares these two techniques using various drilling parameters (diameter ratio, feed rate, and spindle speed) to predict the thrust force for a core-center drill in drilling composite materials. The obtained results indicated that RBFN is a practical and an effective way for the evaluation of drilling-induced thrust force.     

Modeling of TIG welding and abrasive flow machining processes using radial basis function networks

Input-output relationships of tungsten inert gas (TIG) welding and abrasive flow machining (AFM) processes were determined using radial basis function networks (RBFNs). A batch mode of training was adopted to implement the principle of back-propagation (BP) algorithm (which works based on a steepest descent algorithm) and a genetic algorithm (GA), separately. The performances of RBFN tuned by a BP algorithm and that trained by a GA were compared, on some test cases related to the above two manufacturing processes. The GA-optimized RBFN was found to perform slightly better than the BP-tuned RBFN. The back-propagation algorithm works based on the principle of a steepest descent method, whose solutions have the chance of getting stuck at the local minima, whereas the probability of the GA-solutions for being trapped at the local minima is less. However, their performances may depend on the nature of the deviation (error) function.     

双相干谱和RBF网络在旋转机械故障诊断中的应用

双相干谱保留了信号的相位信息，可以用来描述非线性相位的耦合，径向基函数网络具有良好的推广能力和分类能力。文中将双相干谱和径向基函数网络结合，提出一种基于双相干谱与径向基函数网络相结合的旋转机械故障诊断方法，即以双相干谱为故障特征向量，以径向基函数网络作为分类器，对旋转机械的故障进行分类，并以转子不平衡、转轴碰摩、油膜涡动为例进行实验研究。实验结果表明，结合双相干谱和径向基函数网络的旋转机械故障诊断方法是有效的。     

径向基函数网络在虚拟加工对象重构中的应用

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.     

径向基粒子滤波状态跟踪测量算法

状态跟踪测量的过程噪声降低了目标信噪比,增加了自适应滤波跟踪的难度。当误差较大时,基本粒子滤波算法的预测累积误差效应将导致系统发散。无迹粒子滤波算法利用无迹卡尔曼滤波提高重要性函数估计精度,减少后验概率密度分布误差,但同时也将大幅增加运算时间。提出一种基于径向基函数网络(RBFN)的改进型粒子滤波算法PF-RBF,利用RBFN通过目标状态观测值和全局预测值拟合状态变化趋势,更新各粒子状态,提高先验概率密度分布估计精度,消除过程噪声引起的估计误差。与无迹粒子滤波(UPF)算法相比,该算法无需构造无迹卡尔曼滤波(UKF)重要性函数,提高了运算速度。机动目标跟踪试验表明,径向基粒子滤波算法在线性和非线性观测方程下的状态跟踪测量精度和算法稳定性均优于UKF、PF和UPF算法,可有效实现对状态变化的实时鲁棒跟踪。当参与运算的粒子数增加时,PF-RBF算法执行时间的增长速率较UPF算法更低,可满足高精度状态跟踪应用。     

压电陶瓷驱动器的滑模神经网络控制

由于压电陶瓷驱动器的迟滞非线性严重影响其定位精度,本文提出了一种滑模神经网络控制方法来改善它的性能.用径向基函数神经网络的输出作滑模控制的等价控制量,由迟滞补偿器估计控射器参数误差、外部扰动和近似计算所造成的不确定量对神经网络的输出控制量进行补偿,从而使驱动器系统状态保持在滑模平面上.基于Lyapunov稳定性理论推导了控制器和补偿器的自适应调节律,分析了控制系统的收敛性和稳定性,以可变幅值的低频三角波为参考位移量对控制系统进行了实验测试与分析,结果表明,只采用神经网络控制时的平均定位误差为0.43 μm,最大误差为0.77 μm,而采用滑模控制方法对神经网络控制量进行补偿后,平均定位误差减小为0.27 μm,最大误差减小为0.49μm,定位精度有了显著的提高.