Solving time-varying inverse kinematics problem of wheeled mobile manipulators using Zhang neural network with exponential convergence

A mobile manipulator is a robotic device composed of a mobile platform and a stationary manipulator fixed to the platform. The forward kinematics problem for such mobile manipulators has a mathematical analytic solution; however, the inverse kinematics problem is mathematically intractable (especially for satisfying real-time requirements). To obtain the accurate solution of the time-varying inverse kinematics for mobile manipulators, a special class of recurrent neural network, named Zhang neural network (ZNN), is exploited and investigated in this article. It is theoretically proven that such a ZNN model globally and exponentially converges to the solution of the time-varying inverse kinematics for mobile manipulators. In addition, the kinematics equations of the mobile platform and the manipulator are integrated into one system, and thus the resultant solution can co-ordinate simultaneously the wheels and the manipulator to fulfill the end-effector task. For comparison purposes, a gradient neural network (GNN) is developed for solving time-varying inverse kinematics problem of wheeled mobile manipulators. Finally, we conduct extensive tracking-path simulations performed on a wheeled mobile manipulator using such a ZNN model. The results substantiate the efficacy and high accuracy of the ZNN model for solving time-varying inverse kinematics problem of mobile manipulators. Besides, by comparing the simulation results of the GNN and ZNN models, the superiority of the ZNN model is demonstrated clearly.     

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...     

面内变刚度薄板弯曲问题的挠度?弯矩耦合神经网络方法

发展了一种求解面内变刚度功能梯度薄板弯曲问题的神经网络方法. 面内变刚度薄板弯曲问题的偏微分控制方程为一复杂的4阶偏微分方程, 传统的基于强形式的神经网络解法在求解该偏微分方程时可能会遇到难以收敛、边界条件难以处理的情况. 本文基于Kirchhoff薄板弯曲理论, 提出了一种直角坐标系下任意面内变刚度薄板弯曲问题的神经网络解法. 神经网络模型包含挠度网络与弯矩网络, 分别用于预测薄板的挠度与弯矩, 从而将求解4阶偏微分方程转换为求解一系列二阶偏微分方程组, 通过对挠度、弯矩试函数的构造可使得神经网络计算结果严格满足边界条件. 在误差的反向传播中, 根据本文提出的误差函数公式计算训练误差并结合Adam优化算法更新模型的内部参数. 求解了不同边界条件、形状的面内变刚度薄板弯曲问题, 并将所得计算结果与理论解、有限元解进行对比. 研究表明, 本文模型对于求解面内变刚度薄板弯曲问题具备适应性, 虽然模型中的弯矩网络收敛较挠度网络要慢, 但本文方法在试函数的构造上更为简单、适应性更强.      

一种结构动力响应分析的快速高阶算法

为了提高基于高阶格式的结构动力响应微分求积分析方法的计算效率,发展了一种求解动力方程的快速算法.利用微分求积原理将结构动力方程转化为标准Sylvester方程的形式,通过对系数矩阵进行矩阵分解,进而将动力响应Sylvester方程化为一系列标准线性方程组,采用相关成熟算法求解这些线性方程组后即可获得结构动力时程响应的全...     

Solving Huxley equation using an improved PINN method

Although many effective methods for solving partial differential equations (PDEs) have been proposed, there is no universal method that can solve all PDEs. Therefore, solving partial differential equations has always been a difficult problem in mathematics, such as deep neural network (DNN). In recent years, a method of embedding some basic physical laws into traditional neural networks has been proposed to reveal the dynamic behavior of equations directly from space-time data [i.e., physics-informed neural network (PINN)]. Based on the above, an improved deep learning method to recover the new soliton solution of Huxley equation has been proposed in this paper. As far as we know, this is the first time that we have used an improved method to study the numerical solution of the Huxley equation. In order to illustrate the advantages of the improved method, we use the same network depth, the same hidden layer and neurons contained in the hidden layer, and the same training sample points. We analyze the dynamic behavior and error of Huxley’s exact solution and the new soliton solution and give vivid graphs and detailed analysis. Numerical results show that the improved algorithm can use fewer sample points to reconstruct the exact solution of the Huxley equation with faster convergence speed and better simulation effect.

     

泊松白噪声激励下强非线性系统的半解析瞬态解

自然界与工程中都普遍存在着随机扰动,且大多数呈现出固有的非高斯性质,若采用高斯激励建模可能会导致巨大的误差.泊松白噪声作为一种典型且重要的非高斯激励模型,已引起了广泛的关注.目前,泊松白噪声激励下系统的动态特性分析主要集中于稳态响应的研究,而针对瞬态响应的求解难度仍较大,需进一步发展.本文引入径向基神经网络,提出了一种泊松白噪声激励下单自由度强非线性系统瞬态响应预测的高效半解析方法.首先将广义Fokker-Plank-Kolmogorov (FPK)方程的瞬态解表示为一组含时变待定权值系数的高斯径向基神经网络;然后采用有限差分法离散时间导数项,并结合随机取样技术构造含时间递推式的损失函数;最后通过拉格朗日乘子法使得损失函数最小化获得时变最优权值系数.作为算例,探究了两个经典强非线性系统,并采用蒙特卡罗模拟方法对解析结果加以验证.结果表明:本文方法所获得的瞬时概率密度函数与蒙特卡罗模拟数据吻合地较好,并且算法具备较高的计算效率.在系统响应的整个演化过程中,本文所提方法能够非常有效地捕捉到系统响应在各个时刻下的复杂非线性特征.此外,本文方法所获得的高精度半解析瞬态解,不仅可作为基准解检验其...     

Time-dependent solutions of viscous incompressible flows in moving co-ordinates

A time-accurate solution method for the incompressible Navier-Stokes equations in generalized moving coordinates is presented. A finite volume discretization method that satisfies the geometric conservation laws for time-varying computational cells is used. The discrete equations are solved by a fractional step solution procedure. The solution is second-order-accurate in space and first-order-accurate in time. The pressure and the volume fluxes are chosen as the unknowns to facilitate the formulation of a consistent Poisson equation and thus to obtain a robust Poisson solver with favourable convergence properties. The method is validated by comparing the solutions with other numerical and experimental results. Good agreement is obtained in all cases.     

FD-PINN:频域物理信息神经网络

物理信息神经网络(physics-informed neural network, PINN)是将模型方程编码到神经网络中,使网络在逼近定解条件或观测数据的同时最小化方程残差,实现偏微分方程求解.该方法虽然具有无需网格划分、易于融合观测数据等优势,但目前仍存在训练成本高、求解精度低等局限性.文章提出频域物理信息神经网络(frequency domain physics-informed neural network, FD-PINN),通过从周期性空间维度对偏微分方程进行离散傅里叶变换,偏微分方程被退化为用于约束FD-PINN的频域中维度更低的微分方程组,该方程组内各方程不仅具有更少的自变量,并且求解难度更低.因此,与使用原始偏微分方程作为约束的经典PINN相比, FD-PINN实现了输入样本数目和优化难度的降低,能够在降低训练成本的同时提升求解精度.热传导方程、速度势方程和Burgers方程的求解结果表明, FD-PINN普遍将求解误差降低1～2个数量级,同时也将训练效率提升6～20倍.     

基于神经网络的差分方程快速求解方法

近年来, 人工神经网络(artificial?neural?networks, ANN), 尤其是深度神经网络(deep?neural?networks, DNN)由于其在异构平台上的高计算效率与对高维复杂系统的拟合能力而成为一种在数值计算领域具有广阔前景的新方法. 在偏微分方程数值求解中, 大规模线性方程组的求解通常是耗时最长的步骤之一, 因此, 采用神经网络方法求解线性方程组成为了一种值得期待的新思路. 但是, 深度神经网络的直接预测仍在数值精度方面仍有明显的不足, 成为其在数值计算领域广泛应用的瓶颈之一. 为打破这一限制, 本文提出了一种结合残差网络结构与校正迭代方法的求解算法. 其中, 残差网络结构解决了深度网络模型的网络退化与梯度消失等问题, 将网络的损失降低至经典网络模型的1/5000; 修正迭代的方法采用同一网络模型对预测解的反复校正, 将预测解的残差下降至迭代前的10?5倍. 为验证该方法的有效性与通用性, 本文将该方法与有限差分法结合, 对热传导方程与伯格方程进行了求解. 数值结果表明, 本文所提出的算法对于规模大于1000的方程组具有10倍以上的加速效果, 且数值误差低于二阶差分格式的离散误差.      

Analytical Solution for Multi-Species Contaminant Transport in Finite Media with Time-Varying Boundary Conditions

Most analytical solutions available for the equations governing the advective–dispersive transport of multiple solutes undergoing sequential first-order decay reactions have been developed for infinite or semi-infinite spatial domains and steady-state boundary conditions. In this study, we present an analytical solution for a finite domain and a time-varying boundary condition. The solution was found using the Classic Integral Transform Technique (CITT) in combination with a filter function having separable space and time dependencies, implementation of the superposition principle, and using an algebraic transformation that changes the advection–dispersion equation for each species into a diffusion equation. The analytical solution was evaluated using a test case from the literature involving a four radionuclide decay chain. Results show that convergence is slower for advection-dominated transport problems. In all cases, the converged results were identical to those obtained with the previous solution for a semi-infinite domain, except near the exit boundary where differences were expected. Among other applications, the new solution should be useful for benchmarking numerical solutions because of the adoption of a finite spatial domain.     

Adaptive sliding mode control of dynamic system using RBF neural network

This paper presents a robust adaptive sliding mode control strategy using radial basis function (RBF) neural network (NN) for a class of time varying system in the presence of model uncertainties and external disturbance. Adaptive RBF neural network controller that can learn the unknown upper bound of model uncertainties and external disturbances is incorporated into the adaptive sliding mode control system in the same Lyapunov framework. The proposed adaptive sliding mode controller can on line update the estimates of system dynamics. The asymptotical stability of the closed-loop system, the convergence of the neural network weight-updating process, and the boundedness of the neural network weight estimation errors can be strictly guaranteed. Numerical simulation for a MEMS triaxial angular velocity sensor is investigated to verify the effectiveness of the proposed adaptive RBF sliding mode control scheme.     

基于改进位移模式的一维有限元超收敛算法

将多尺度方法的思想与超收敛计算的解析公式结合起来,提出了改进有限元位移模式的算法。利用超收敛计算的解析公式,将高阶有限元解的位移模式用常规有限元解的位移模式表示。用常规有限元解的位移模式与高阶有限元解的位移模式之和构造新的位移模式,采用积分形式推导了单元刚度矩阵。该算法在前处理和后处理两个阶段都使用超收敛计算公式,在常规试函数的基础上,增加了高阶试函数,使得单元内平衡方程的残差减少,从而达到提高精度的目标。对于线性单元,本文结点和单元的位移、导数都达到了h4阶的超收敛精度。     

Prescribed-time adaptive neural tracking control for a class of uncertain robotic manipulators with dead zone input

This paper solves the prescribed-time control problem for a class of robotic manipulators with system uncertainty and dead zone input. To make the system stable within a given convergence time T, a novel prescribed-time adaptive neural tracking controller is proposed by using the temporal scale transformation method and Lyapunov stability theory. Unlike the finite-time and the fixed-time stability where the convergence time depends on the controller parameters, the convergence time constant T is introduced into the proposed controller so that the closed-loop system will be stable within T. To cope with the system uncertainty, radial basis function neural networks (RBFNNs) are used and only need to update one parameter online. In addition, by choosing the same structure and parameters of RBFNNs, the proposed method can shorten the convergence time of the neural networks. Finally, simulation results are presented to demonstrate the effectiveness of the prescribed-time controller.

     

CONSTITUTIVE MODEL OF ARTIFICIAL FROZEN SOIL BASED ON CASCADE-CORRELATION NEURAL NETWORK

为解决BP (back propagation) 神经网络收敛速度慢,网络结构需事先定义等缺点,采用了级连相关神经网络模型来建立人工冻土应力和应变之间的关系. 基于该模型推导了冻土的一致刚度矩阵形式,利用人工冻土三轴试验数据对神经网络模型进行训练,并用其替换有限元计算中的传统本构模型,将计算结果与性质及含水率相同的冻土的试验结果进行了对比,发现该神经网络本构模型很好地反应了材料的非线性,能够改善数值计算结果,与实测结果吻合地很好,比具有相同隐含层神经元个数的BP 模型更接近实测结果.     

High order symplectic conservative perturbation method for time-varying Hamiltonian system

This paper presents a high order symplectic conservative perturbation method for linear time-varying Hamiltonian system.Firstly,the dynamic equation of Hamiltonian system is gradually changed into a high order perturbation equation,which is solved approximately by resolving the Hamiltonian coefficient matrix into a "major component" and a "high order small quantity" and using perturbation transformation technique,then the solution to the original equation of Hamiltonian system is determined through a series of inverse transform.Because the transfer matrix determined by the method in this paper is the product of a series of exponential matrixes,the transfer matrix is a symplectic matrix;furthermore,the exponential matrices can be calculated accurately by the precise time integration method,so the method presented in this paper has fine accuracy,efficiency and stability.The examples show that the proposed method can also give good results even though a large time step is selected,and with the increase of the perturbation order,the perturbation solutions tend to exact solutions rapidly.     

一种有限元模型动力缩聚移频迭代法

提出了一种基于矩阵广义逆的有限元模型动力缩聚移频迭代方法,该方法首先直接从原系统特征方程出发,导出反映系统主,副自由度之间位移关系的动力缩聚矩阵的控制方程,然后给出了相应的迭代求解方法和收敛准则。为了减少求矩阵广义逆的计算工作量,本文给出了一种替代方法,把对一个高阶满阵求逆转化为对一个同阶高度稀疏矩阵求逆。与已有的动力缩聚迭代法相比,本文提出的方法具有两个显著的优点：其一是迭代收敛速度高,其二是通     

二维对流扩散方程的高精度全隐式多重网格方法

提出了数值求解二维非定常变系数对流扩散方程的一种时间二阶、空间四阶精度的三层全隐紧致差分格式。为了加快迭代求解隐格式时在每一个时间步上的收敛速度，采用多重网格加速技术，建立了适用于本文高精度金隐紧致格式的多重网格算法。数值实验结果验证了本文方法的精确性、稳定性和对高网格雷诺数问题的强适应性。     

Exponential state estimation for delayed recurrent neural networks with sampled-data

In this paper, the sampled-data state estimation problem is investigated for a class of recurrent neural networks with time-varying delay. Instead of the continuous measurement, the sampled measurement is used to estimate the neuron states, and a sampled-data estimator is constructed. By converting the sampling period into a bounded time-varying delay, the error dynamics of the considered neural network is derived in terms of a dynamic system with two different time-delays. Subsequently, by choosing an appropriate Lyapunov functional and using the Jensen??s inequality, a sufficient condition depending on the sampling period is obtained under which the resulting error system is exponentially stable. Then a sampled-data estimator is designed in terms of the solution to a set of linear matrix inequalities (LMIs) which can be solved by using available software. Finally, a numerical example is employed to demonstrate the effectiveness of the proposed sampled-data estimation approach.     

Stabilization of a nonlinear rotating disk-beam system with localized thermal effect

This paper presents two methods for finite-time topology identification for the complex spatio- temporal networks with coupling time delay. By introducing the auxiliary systems, a relationship between the unknown topology and two measurable matrix signals is developed. Based on the relationship, one method of identifying the topology is to compute the invertibility of the matrix. Besides this method, an adaptive law is developed to infer the topology online in finite time. The proposed methods do not require the differentiability of the time-varying delay. Furthermore, the methods can also be applied to complex spatio-temporal networks with unknown system parameters. Finally, an illustrative example is provided to show the effectiveness of the proposed methods.     

Mixed time discontinuous space-time finite element method for convection diffusion equations

A mixed time discontinuous space-time finite element scheme for secondorder convection diffusion problems is constructed and analyzed. Order of the equation is lowered by the mixed finite element method. The low order equation is discretized with a space-time finite element method, continuous in space but discontinuous in time. Stability, existence, uniqueness and convergence of the approximate solutions are proved. Numerical results are presented to illustrate efficiency of the proposed method.