NEURAL NETWORK MODEL OF MAGNETORHEOLOGICAL DAMPER FOR VIBRATION ISOLATION PLATFORM OF WHOLE-SPACECRAFT

为改善星箭界面振动环境,设计六杆隔振平台,采用磁流变阻尼器作为半主动控制元件,替代原有锥壳过渡支架.对整星隔振平台用磁流变阻尼器进行性能测试,得到反映磁流变阻尼器阻尼特性的实验数据.建立具有两个隐含层的反向传播神经网络对阻尼器进行建模,用于预测磁流变阻尼器阻尼特性以及控制系统设计.提出一种串行算法优化网络结构、权值和阈值,保证网络具有较好的泛化能力和稳定性.仿真结果表明,与参数化模型相比,提出的神经网络模型具有较小的训练误差和较强的泛化能力,能够很好地预测阻尼器的阻尼特性.     

GA based meta-modeling of BPN architecture for constrained approximate optimization

Artificial neural networks (ANN) have been extensively used as global approximation tools in the context of approximate optimization. ANN traditionally minimizes the absolute difference between target outputs and approximate outputs, thereby resulting in approximate optimal solutions being sometimes actually infeasible when it is used as a meta-model for inequality constraint functions. The paper explores the development of the modified back-propagation neural network (BPN) based meta-model that ensures the constraint feasibility of approximate optimal solution. The BPN architecture is optimized via genetic algorithm (GA) to determine integer/continuous decision parameters such as the number of hidden layers, the number of neurons in a hidden layer, and interconnection weights between layers in the network. The verification of the proposed approach is examined by adopting a number of standard structural problems and an optical disk drive (ODD) suspension problem. Finally, GA based approximate optimization of suspension with optical flying head (OFH) is conducted to enhance the shock resistance capability in addition to dynamic characteristics.     

A proposed hybrid neural network for position control of a walking robot

The use of a proposed recurrent neural network control system to control a four-legged walking robot is presented in this paper. The control system consists of a neural controller, a standard PD controller, and the walking robot. The robot is a planar four-legged walking robot. The proposed Neural Network (NN) is employed as an inverse controller of the robot. The NN has three layers, which are input, hybrid hidden and output layers. In addition to feedforward connections from the input layer to the hidden layer and from the hidden layer to the output layer, there is also a feedback connection from the output layer to the hidden layer and from the hidden layer to itself. The reason to use a hybrid layer is that the robot’s dynamics consists of linear and nonlinear parts. The results show that the neural-network controller can efficiently control the prescribed positions of the stance and swing legs during the double stance phase of the gait cycle after sufficient training periods. The goal of the use of this proposed neural network is to increase the robustness of the control of the dynamic walking gait of this robot in the case of external disturbances. Also, the PD controller alone and Computed Torque Method (CTM) control system are used to control the walking robot’s position for comparison.     

中深孔爆破振动参数的BP神经网络预报

以某工程不同爆点不同监测点的爆破振动监测数据为背景 ,在分析爆破振动主要影响因素的基础上 ,建立了能同时对爆破振动速度峰值、振动主频率和振动的持续时间进行预报的BP神经网络模型。该模型的预报结果 (爆破振动的幅值、振动主频率和振动持续时间 )与实际监测结果基本吻合 ,从而得到了该场地不同地质、地形情况下爆破振动预报的BP神经网络模型。     

薄板弯曲问题的神经网络方法

为发展神经网络方法在求解薄板弯曲问题中的应用,基于Kirchhoff板理论,提出一种采用全连接层求解薄板弯曲四阶偏微分控制方程的神经网络方法。首先在求解域、边界中随机生成数据点作为神经网络输入层的参数,由前向传播系统求出预测解;其次计算预测解在域内及边界处的误差,利用反向传播系统优化神经网络系统的计算参数;最后,不断训练神经网络使误差收敛,从而得到薄板弯曲的挠度精确解。以不同边界、荷载条件的三角形、椭圆形、矩形薄板为例,利用所提方法求解其偏微分方程,与理论解或有限元解对比,讨论了影响神经网络方法收敛的因素。研究表明,本文方法对求解薄板弯曲问题的四阶偏微分控制方程具有一定的适应性,其收敛性受多种条件影响。相比有限元,该方法收敛速度较慢,在复杂的边界条件下收敛性不佳,然而其不基于变分原理,无需计算刚度矩阵,便可获得较高的计算精度。     

逐孔起爆震动参数预报的BP神经网络模型

根据神经网络理论,结合逐孔起爆技术的特点,建立了爆破震动参数预报的BP网络模型。以某矿 山深孔台阶爆破为例,利用逐孔起爆过程中收集的原始资料和爆破震动监测数据,对建立的BP网络模型进 行了训练和应用。与实测值比较后发现,BP网络模型的预报结果更接近实测值。     

基于ANN的定向井有杆抽油系统悬点示功图特征参数的计算

有杆抽油系统悬点示功图的特征参数是合理选择地面机电设备的主要依据.由于井下工况的复杂性和部分参数难以确定,使得基于求解高维时变非线性方程的传统方法的计算结果存在偏差.本文将传统方法和神经网络相结合,给出了一种能比较精确地确定定向井有杆抽油系统悬点示功图特征参数的方法,避免了建立和求解复杂的非线性动力学方程.首先依据传统方法,计算出简化悬点示功图的特征参数.然后考虑抽油杆柱弹性振动、抽油杆与油管之问的库仑和粘性摩擦、气体和供液能力等因素的影响,利用BP神经网络和RBF神经网络建立了不同工况下悬点示功图特征参数的计算模型.利用现场实测数据对建立的神经网络进行了训练和测试.测试结果表明了本文方法的正确性和有效性.     

涡激响应尾流双振子模型的参数反演方法

尾流双振子模型是研究圆柱结构涡激振动响应的重要模型,模型参数的准确确定对悬浮隧道设计理论具有重要意义.首先通过降阶法将多变量二阶非线性常微分方程组的尾流双振子模型变换为一阶方程组.然后给出一种新型的圆柱结构水槽试验设计方案,其中试验模型的刚度能够较好反映悬浮隧道等实际工程结构的刚度,基于相机动态捕捉和视频识别计算机程序,获取圆柱结构在水平和竖直方向的位移试验数据.基于试验结果和龙格-库塔方法求解一阶方程组,采用BP神经网络智能算法对模型参数进行反演,同时利用遗传算法对神经元的初始权值和阈值进行优化,所得结果平均误差仅为5.50％,优于遗传算法和未优化的BP神经网络模型.结果表明,基于遗传算法优化的BP神经网络智能算法能够精确实现尾流双振子模型的参数确定,为圆柱结构涡激振动响应分析提供理论基础.     

Chaotic dynamics in a neural network under electromagnetic radiation

In this paper, a small Hopfield neural network with three neurons is studied, in which one of the three neurons is considered to be exposed to electromagnetic radiation. The effect of electromagnetic radiation is modeled and considered as magnetic flux across membrane of the neuron, which contributes to the formation of membrane potential, and a feedback with a memristive type is used to describe coupling between magnetic flux and membrane potential. With the electromagnetic radiation being considered, the previous steady neural network can present abundant chaotic dynamics. It is found that hidden attractors can be observed in the neural network under different conditions. Moreover, periodic motion and chaotic motion appear intermittently with variations in some system parameters. Particularly, coexistence of periodic attractor, quasiperiodic attractor, and chaotic strange attractor, coexistence of bifurcation modes and transient chaos can be observed. In addition, an electric circuit of the neural network is implemented in Pspice, and the experimental results agree well with the numerical ones.     

Empirical correlation study of dryout heat transfer at high pressure using high order neural network and feed forward neural network

A hybrid high order neural network (HHONN) and a feed forward neural network (FNN) are developed and applied to find an optimized empirical correlation for prediction of dryout (DO) heat transfer. The values predicted by the HHONN and FNN models are compared with each other and also with the previous values of empirical correlation. HONN successfully provides an efficient open-box model of nonlinear input–output mapping which provides easier understanding of data mining. By removing the hidden layers, HONN structures become simpler than FNNs and initialization of learning parameters (weights) will not be catastrophic. The RMS results show that the HHONN model has superior fitting specification for prediction of DO heat transfer problem compared to the other prediction methods.     

The Application of Advanced Signal Processing Techniques to Induction Motor Bearing Condition Diagnosis

Four approaches based on bispectral and wavelet analysis of vibration signals are investigated as signal processing techniques for application in the diagnosis of a number of induction motor rolling element bearing faults. The bearing conditions considered are a normal bearing and bearings with cage and inner and outer race faults. The vibration analysis methods investigated are based on the bispectrum, the bispectrum diagonal slice, the summed bispectrum and wavelets. Singular value decomposition (SVD) is used to extract the most significant features from the vibration signatures and the features are used as inputs to an artificial neural network trained to identify the bearing faults. The results obtained show that the diagnostic system using a supervised multi-layer perceptron type neural network is capable of classifying bearing condition with high success rate, particularly when applied to summed bispectrum signatures.     

基于反向传播神经网络的疲劳裂纹扩展分析

材料发生疲劳断裂时往往会引起重大安全事故,而基于传统数值模拟方法求解疲劳裂纹扩展问题时模 型复杂、计算量大．本文基于包含多隐层的反向传播神经网络分析金属材料疲劳裂纹扩展行为,计算了裂纹扩 展过程中的 von Mises应力场和位移场,并与数值解和实验解进行对比,误差分析结果表明其求解精度高．并 基于该神经网络有效预测了裂纹扩展中裂纹长度及裂纹扩展速率的变化过程,预测精度高．该神经网络分析方 法可为材料剩余寿命和疲劳强度预测提供研究基础．     

Condition Monitoring of Ground Anchorages by Dynamic Impulses: GRANIT System

The GRANIT system operates by applying an impulse of known force by means of an impact device that is attached to the tendon of the anchorage. The vibration response signals resulting from this impulse are complex in nature and require analysis to be undertaken in order to extract information from the vibrational response signatures that is relevant to the condition of the anchorage. In the system, the complicated relationship that exists between characteristics of an anchorage and its response to an impulse is identified and learned by a novel artificial intelligence network based on artificial intelligence techniques.The results presented in this paper demonstrate the potential of the GRANIT system to diagnose the integrity of ground anchorages at a site near Stone, England, by using a trained neural network capable of diagnosing the post-tension level of the anchorage. This neural network was used for the diagnosis of load in a second ground anchorage adjacent to the original anchorage used for the training of the neural network. Further tests were taken with a different anchor head configuration of the anchorage and a different relationship between the signature response of the anchorage to an applied impulse and its post-tension level was found.Problems encountered during the diagnosis of this second set of test signatures by the trained neural network are investigated with the use of a lumped parameter dynamic model. This model is able to identify the parameters in the anchorage system that affect this change in response signature. The results from the investigation lead to a new form of classification for the installed anchorages, based on their anchor head configuration.Laboratory strand anchorage tests were undertaken in order to compare with and validate the results obtained from the field tests and the lumped parameter dynamic model.     

VIBRATION SUPPRESSION OF A FLEXIBLE PIEZOELECTRIC BEAM USING BP NEURAL NETWORK CONTROLLER

This paper aims at modeling and developing vibration control methods for a flexible piezoelectric beam. A collocated sensor/actuator placement is used. Finite element analysis (FEA) method is adopted to derive the dynamics model of the system. A back propagation neural network (BPNN) based proportional-derivative (PD) algorithm is applied to suppress the vibration. Simulation and experiments are conducted using the FEA model and BPNN-PD control law. Experimental results show good agreement with the simulation results using finite element modeling and the neural network control algorithm.     

Prediction of void fraction for gas–liquid flow in horizontal,upward and downward inclined pipes using artificial neural network

In this work, the ability of artificial neural networks (ANNs) to predict void fraction of gas–liquid two–phase flow in horizontal and inclined pipes was investigated. For this purpose, an ANN model was designed and trained using a total of 301 experimental data points reported in the literature for inclination angles between –20° and +20°. Pipe inclination angle as well as superficial Reynolds number of gas (Re_sg) and liquid (Re_sl) were chosen as input parameters of different structures of multilayer perceptron (MLP) neural networks, while the corresponding void fraction was selected as their output parameter. A hyperbolic tangent sigmoid and a linear function were employed as transfer functions of hidden and output layers, respectively, and Levenberg–Marquardt back propagation algorithm was used to train the networks. By trial–and–error method, a three–layer network with 10 neurons in the hidden layer was achieved as optimal structure of the ANN which made it possible to predict the void fraction with a high accuracy. Mean absolute percent error (MAPE) of 1.81% and coefficient of determination (R²) of 0.9976 for training data and MAPE of 1.52% and R² value of 0.9948 for testing data were obtained. Also for all data, MAPE of 1.95% and R² value of 0.9972 were calculated, and 96% data were within ±5% error band. In addition, the accuracy of the proposed ANN model was compared with the predictions from 17 void fraction correlations available in the literature for different flow patterns and horizontal and inclined flows. For all cases, the proposed ANN model gave better performance than all of the studied correlations. The results confirm the very good capability of the ANNs to predict the void fractions of gas–liquid flow in inclined pipes, regardless of flow pattern. Finally, by performing interpolation using the trained network, the void fraction values for some other conditions were predicted.     

基于神经网络算法的建筑结构振动分散控制研究

针对地震作用下建筑结构振动分散控制问题,引入神经网络算法,研究结构振动分散神经网络控制策略,来解决分散控制中各子系统的耦合问题和神经网络算法的训练成本问题.利用径向基函数RBF(Radical Basis Function)神经网络模型并基于newrb函数构建了RBF神经网络控制器,对某20层Benchmark结构模型分别进行集中控制和多工况子系统划分分散控制的数值模拟分析,结果表明,提出的各子系统耦合的分散RBF神经网络振动控制策略考虑了子系统间的信息共享,可有效控制结构的振动响应,且子系统达到理想训练结果所需的训练次数与BP网络相比显著降低.     

Free vibration analysis of functionally graded laminated sandwich cylindrical shells integrated with piezoelectric layer

An analytical method for the three-dimensional vibration analysis of a functionally graded cylindrical shell integrated by two thin functionally graded piezoelectric (FGP) layers is presented. The first-order shear deformation theory is used to model the electromechanical system. Nonlinear equations of motion are derived by considering the von Karman nonlinear strain-displacement relations using Hamilton’s principle. The piezoelectric layers on the inner and outer surfaces of the core can be considered as a sensor and an actuator for controlling characteristic vibration of the system. The equations of motion are derived as partial differential equations and then discretized by the Navier method. Numerical simulation is performed to investigate the effect of different parameters of material and geometry on characteristic vibration of the cylinder. The results of this study show that the natural frequency of the system decreases by increasing the non-homogeneous index of FGP layers and decreases by increasing the non-homogeneous index of the functionally graded core. Furthermore, it is concluded that by increasing the ratio of core thickness to cylinder length, the natural frequencies of the cylinder increase considerably.     

考虑摩擦特性的柔性关节空间机器人自适应CMAC神经网络鲁棒控制

讨论了关节摩擦力矩影响下,具有柔性铰关节的漂浮基空间机器人系统的动力学控制问题.设计了基于高斯基函数的小脑神经网络(CMAC)鲁棒控制器和摩擦力矩补偿器.用奇异摄动理论对系统的动力学模型进行快慢变子系统分解,针对快变子系统,设计力矩微分反馈控制器来抑制机械臂关节柔性引起的振动;对于慢变子系统,设计了基于自适应CMAC神...     

地震作用下高层建筑结构的分散神经网络振动控制研究

针对地震作用下高层建筑振动神经网络控制问题,将神经网络理论与分散控制理论相结合,提出分散神经网络振动控制方案,并应用于高层结构地震反应振动控制中。利用多层前馈神经网络建立结构模型,预测结构的振动响应。基于NARMA-L2的神经自校正控制系统设计BP神经网络控制器,研究分散神经网络振动控制效果,并与神经网络集中控制进行比较。对某20层Benchmark结构模型进行数值模拟分析,结果表明,本文提出的分散神经网络振动控制方法简化了神经网络的结构,可有效控制结构振动和消除时滞;同时,相对于集中控制的单一失效,本文方法的可靠性更强且可以保证振动控制系统的实时响应。     

Nonlinear vibration semi-active control of automotive steering using magneto-rheological damper

Traffic accidents are often caused by vibration of automotive steering because the vibration can make a vehicle run like a snake. A?novel semi-active vibration control strategy of automotive steering with magneto-rheological (MR) damper is proposed in this paper. An adaptive RBF neural sliding mode controller is designed for the vibration system. It is showed that an equivalent dynamic model for the vibration system is established by using Lagrange method, and then treats it as actual system partially. A?feedback control law is designed to make this nominal model stable. Uncertain part of system and outside disturbance are estimated using RBF neural network, and their upper boundary is obtained automatically. By constructing reasonable switch function, state variables can arrive at origin asymptotically along the sliding mode. Strong robust character of control system is proved by stability analysis and a numerical simulation example is performed to support this control scheme.