材料试验机施力系统的优化设计

给出了采用X-Q自适应控制器构成的一类非线性控制器以及实现多目标优化控制系统的计算机辅助设计参数寻优方法，讨论怎样应用X—Q自适应控制器实现多目标优化控制，并给出实现多目标优化材料试验机施力系统的实例，计算机仿真校验其性能指标的结果表明，多目标优化系统的性能指标好，设计方法简单、通用、适用于工程设计。     

轮胎非稳态侧偏特性非线性仿真模型

在稳态指数统一模型的基础上,根据动态过程有效滑移率的微分表达,建立轮胎非稳态侧偏特性非线性仿真模型。在实验中发现当用有效滑移率和准稳态概念来计算回正力臂 Ｄｘ 时,有滑移率超前而回正力臂滞后现象,该现象可以通过一个一阶滤波器来表达。模型的计算结果与实验结果对比,表明该模型可以反映轮胎非线性侧偏特性,可以用于汽车操纵动力学研究。     

Mechanical and electromechanical behaviour of fibre brushes

Investigations of the mechanics of metal fibres capable of carrying very high currents are presented. The role played by the Laplace force is emphasized. This force appears whenever a fibre carrying a current is subjected to a magnetic field whose direction is different from that of the fibre. The role played by the coefficient of friction is also taken into account. An exact non-linear differential equation is used to describe the problem, i.e. no assumptions are made about the magnitudes of the deflections. When the normal force exerted on the track by the fibre is an increasing function of the Laplace force, it is found that the maximum stress within the fibre has a non-linear dependence on the Laplace force. When the same normal force is a decreasing function of the Laplace force, it is shown that the normal force vanishes for a critical value of the Laplace force. A perturbation method is used to explain the behaviour of the fibre.     

基于粒子滤波跟踪方法研究

文章针对道路上的车辆跟踪问题,提出了粒子滤波跟踪算法。粒子滤波通过非参数化的蒙特卡罗模拟方法来实现递推贝叶斯滤波,适用于任何能用状态空间模型表示的非线性系统,以及传统卡尔曼滤波无法表示的非线性系统,精度可以逼进最优估计。粒子滤波方法的使用非常灵活,容易实现,具有并行结构,实用性强。文章的主要研究内容包括粒子滤波理论及其实现方法;利用粒子滤波理论来解决目标跟踪问题,构建基于粒子滤波的跟踪框架。     

Application and conceptual explanation of an energy-based approach for the modelling and prediction of sliding wear

Establishing an accurate predictive model for wear will result in major improvements in the efficiency, lifetime, cost and performance of many engineering systems. The authors have previously proposed a practical energy-based model to describe the tribological damage using energy dissipation. In this work, the capacity of this method to simplify the complex wear phenomenon was determined, and the applicability of this method to forecast the wear behaviour of materials was investigated. The wear results for a W–Cu electrical contact composite were utilised to determine the practical value of the model. Numerical integration of friction force–distance diagrams was used to evaluate the energy dissipation. Linear and non-linear least squares methods were applied to find the optimal curve fitting. It is shown that this model can be applied in two general forms: (1) a graphical method and (2) an explicit formulation. Both of these forms are shown to be capable of providing the necessary information, based on a limited number of initial tests, to predict unknown wear data (volume loss, lifetime, energy) without the need for morphological observations.     

有限样本下基于迁移学习的铣削稳定性预测方法

传统铣削稳定性分析因采用静态刀尖点频响函数和平均切削力系数而使其在真实工况下的预测精度降低。为此,引入迁移学习提出一种基于少量实验样本的铣削稳定性预测方法。首先,生成静态刀尖点频响函数和平均切削力系数在全转速范围内多个系列的随机值,并在各系列下进行铣削稳定性分析,通过计算少量极限切削深度实验值与对应的预测值之间的误差,确定最优系列并以其构造源域稳定域数据;然后,利用大量源域数据建立极限切削深度的预训练模型,通过少量实验样本全局微调此模型使其适应真实加工场景。以40组颤振实验样本展开实例验证,所提方法比采用少样本建模的预测精度提升32%,并对比不同数据规模下各类模型预测精度,共同验证所提方法的有效性。     

用线性化方法研究直齿圆柱齿轮的动态性能

首先建立了一对啮合齿轮的有限元模型,然后将用非线性方程表示的模型线性化,通过结合应用有限元理论和接触力学理论,得到齿轮的动态传输误差(eDTE)和接触力以及动态响应的关系。此方法在齿轮动态研究领域显示出了较强的优越性。由于将非线性问题转化成线性问题,避开了冗长的迭代过程,加速了求解过程。再者,在接触及其附近区域使用接触力学理论,仅用数目较少的单元就可得到精度足够的解。     

Identification of multi-degree of freedom non-linear systems using an extended modal space model

The identification of non-linear dynamic systems is an increasingly important area of research, with potential application in many industries. Current non-linear identification methodologies are, in general, mostly suited to small systems with few degrees of freedom (DOF) and few non-linearities. In order to develop a practical identification approach for real engineering structures, the capability of such methods must be significantly extended. In this paper, it is shown that such an extension can be achieved using multi-exciter techniques in order to excite specific modes or DOF of the system under investigation.A novel identification method for large non-linear systems is presented, based on the use of a multi-exciter arrangement using appropriated excitation applied in bursts. This proposed non-linear resonant decay method is applied to a simulated system with 5 DOF and an experimental clamped panel structure. The technique is essentially a derivative of the restoring force surface method and involves a non-linear curve fit performed in modal space. The effectiveness of the resulting reduced order model in representing the non-linear characteristics of the system is demonstrated. The potential of the approach for the identification of large continuous non-linear systems is also discussed.     

An analytical model to investigate skidding in rolling element bearings during acceleration

Skidding, which occurs in rolling element bearings during shaft rotational acceleration, causes wear and incipient failure. This paper presents an analytical model to investigate skidding during rolling element bearing acceleration, taking account of the contact force and friction force between the rolling elements and the races and the cage, gravity, and the centrifugal force of the rolling elements. The Hertzian contact theory is applied to calculate the non-linear contact force. The Coulomb friction law is used to calculate the friction force. All forces above are included in force equilibrium equations to derive the non-linear governing equations of the bearing during acceleration, and are solved using a fourth-order Runge-Kutta algorithm with fixed time step. The proposed model is verified by comparison to other published results and with experimental results. The proposed model can be used to investigate skidding in rolling element bearings during acceleration and the transient motion behavior of rolling elements, and it will lay the theoretical foundations for eliminating skidding in rolling element bearings.     

Modelling of ionic polymer–metal composites by a multi-field finite element method

Ionic polymer–metal composites (IPMCs), a new type of smart material, offer the attractive features of high sensitivity and light weight for developing novel designs in the fields of dynamic sensors, robotic actuators and artificial muscles. The non-linear water molecule diffusion resistance force exists in the process of sodium ion migration, which may cause a significant effect on the dynamic electro-mechanical properties of this kind of material. In this paper, Tadokoro's model constructed by taking the electrostatic force, the viscous resistance force, the sodium ion diffusion force, and the water molecule diffusion resistance force into account was applied for simulation of electro-mechanical properties of IMPC. By coupling these forces and based on the principle of internal force equilibrium and the laws of motion, a multi-field finite element method (MFFEM) was derived using the dynamic electro-mechanical model, and a numerical solving scheme was proposed. Using the MFFEM, the electro-active behaviours of an IPMC beam have been simulated and discussed. The simulation results show that the water molecule diffusion resistance force has various effects on an IPMC beam, such as the water concentration, charge density of sodium ions and internal equilibrium forces. Furthermore, MFFEM has also been applied to study the migration speed of the hydrated sodium ions, electric-field intensity and bending displacement distribution across the thickness direction of the IPMC beam. Compared with the experimental findings reported by other investigators, it has been demonstrated that the proposed method provided a more realistic and scientific way to predict the electro-mechanical behaviours of IPMC by introduction of the non-linear force component in the model. Hence, the accuracy of prediction may also be improved and more information may be provided for developing new IPMC devices.     

基于AHP-NSGA-Ⅲ的车辆悬架预瞄最优控制研究

针对车辆主动悬架线性二次最优控制存在的权值系数确定问题,提出了一种结合主客观性能评价的权值系数选择方法---层次分析-三代遗传算法。利用层次分析法确定变量范围,以6个主动控制悬架性能指标为目标函数,利用三代遗传算法进行多目标优化并以控制力作为限制条件筛选出最优解。以半车模型为例,结合前轮预瞄信息与卡尔曼估计器估计的状态变量,设计预瞄最优控制器,并采用虚拟激励法直接在频域范围内进行求解。仿真结果表明,所提出的控制策略可改善车辆的乘坐舒适性。     

考虑铰间隙的折叠翼展开机构展开过程碰撞动力学仿真分析

基于“二状态模型”的非线性等效弹簧阻尼模型,建立折叠翼展开机构的铰间隙碰撞模型,利用ADAMS软件对折叠翼展开机构的展开过程动力学进行仿真分析．包括间隙大小和间隙摩擦在折叠翼展开机构展开过程中引起的碰撞力变化规律。结果表明,在折叠翼展开机构整个展开过程中,由于间隙的影响,产生随机持续的碰撞力;随着间隙增大,碰撞次数减少,碰撞力幅值有增大的趋势;摩擦加快系统能量的损耗,一定程度上抑制了间隙碰撞次数和展开机构动力学扰动效应。     

Semi-active vibration control of an eleven degrees of freedom suspension system using neuro inverse model of magnetorheological dampers

A semi-active controller-based neural network for a suspension system with magnetorheological (MR) dampers is presented and evaluated. An inverse neural network model (NIMR) is constructed to replicate the inverse dynamics of the MR damper. The typical control strategies are linear quadratic regulator (LQR) and linear quadratic gaussian (LQG) controllers with a clipped optimal control algorithm, while inherent time-delay and non-linear properties of MR damper lie in these strategies. LQR part of LQG controller is also designed to produce the optimal control force. The LQG controller and the NIMR models are linked to control the system. The effectiveness of the NIMR is illustrated and verified using simulated responses of a full-car model. The results demonstrate that by using the NIMR model, the MR damper force can be commanded to follow closely the desirable optimal control force. The results also show that the control system is effective and achieves better performance and less control effort than the optimal in improving the service life of the suspension system and the ride comfort of a car.     

Design strategy for optimal iterative learning control applied on a deep drawing process

Metal forming processes in general can be characterised as repetitive processes; this work will take advantage of this characteristic by developing an algorithm or control system which transfers process information from part to part, reducing the impact of repetitive uncertainties, e.g. a gradual changes in the material properties. The process is highly non-linear and the system plant is modelled using a non-linear finite element and the gain factors for the iterative learning controller is identified solving a non-linear optimal control problem. The optimal control problem is formulated as a non-linear least square problem where the system response is evaluated using a non-linear finite element model of the process.     

Intelligent semi-active vibration control of eleven degrees of freedom suspension system using magnetorheological dampers

A novel intelligent semi-active control system for an eleven degrees of freedom passenger car’s suspension system using magnetorheological (MR) damper with neuro-fuzzy (NF) control strategy to enhance desired suspension performance is proposed. In comparison with earlier studies, an improvement in problem modeling is made. The proposed method consists of two parts: a fuzzy control strategy to establish an efficient controller to improve ride comfort and road handling (RCH) and an inverse mapping model to estimate the force needed for a semi-active damper. The fuzzy logic rules are extracted based on Sugeno inference engine. The inverse mapping model is based on an artificial neural network and incorporated into the fuzzy controller to enhance RCH. To verify the performance of the NF controller (NFC), comparisons with existing semi-active techniques are made. The typical control strategy are linear quadratic regulator (LQR) and linear quadratic Gaussian (LQG) controllers with clipped optimal control algorithm, while inherent time-delay and non-linear properties of MR damper lie in these strategies. Simulation results demonstrated that the NFC has better control performance and less control effort than the optimal in improving the service life of the suspension system and the ride comfort of a car.     

全局力平衡PDC钻头布齿优化设计

针对聚晶金刚石复合片(PDC)钻头在钻井破岩过程中因受力不平衡，导致钻井倾斜、井径扩大、钻头产生横向振动和涡动，造成钻头早期失效的问题，基于已有PDC钻头切削力学知识，构建了PDC钻头弯曲力矩及全局力平衡布齿优化设计模型，并提出模型求解方法。布齿设计实例结果表明，采用全局力平衡布齿设计方法得到的布齿结构可使PDC钻头在不同进尺条件下均能保持全局力平衡状态，尤其低进尺（小于1 mm）时的力平衡状态得到极大改善，提高了PDC钻头的钻井稳定性，对于改善钻头受力、提升钻头使用寿命、提高钻井质量与效率具有重要意义。     

微滑移离散模型及在干摩擦阻尼叶片振动分析中的应用

建立了阻尼器干摩擦力微动滑移离散模型,该模型可以较好地描述摩擦界面粘滞-滑移共存状态和阻尼器因局部拉、压应变不同而产生的局部摩擦力方向的不同,适用于各种激振作用下的阻尼器动力学分析,克服了微动滑移解析模型仅适用于单谐波作用力的局限。采用时域法对简化的凸肩干摩擦阻尼结构叶片系统进行了强迫振动响应分析,验证了微动滑移离散模型在阻尼叶片动力学分析中的可用性。     

基于Z-MAP方法的球头铣刀铣削力的建模

瞬时刚性切削力的建模是铣削加工物理仿真的基础,然而,球头铣刀的刀齿形状复杂,加工过程中姿态多变,瞬时刚性铣削力的建模难度较大。在考虑刀具姿态调整的情况下,通过齐次坐标变换建立了刀齿的运动轨迹,提出了一种识别刀具和工件瞬时接触区的改进Z-MAP算法,通过计算当前刀齿的参考线与工件的边界面或刀齿扫掠面的交点求出瞬时未变形切屑厚度,并采用非线性回归的方法辨识了切削力系数,在此基础上使用微元积分法建立了瞬时切削力的计算模型。为了验证仿真模型的可靠性,分别进行了垂直加工和倾斜加工试验,试验和仿真结果具有较高的一致性,表明该建模仿真方法是有效的,可以为实际加工中参数的选择和优化提供理论依据。     

NON-LINEAR CONTROL OF BUCKLED BEAMS UNDER STEP LOADING

The present paper studies a strategy for the active non-linear control of the oscillations of simply supported buckled beams, in order to mitigate the effects of dynamic loading on the vibration amplitudes and prevent dangerous instability phenomena. First, an analysis of the symmetric non-linear behaviour of the structure without any control system is carried out. In order to control the non-linear vibrations of the beam, an active tendon control system is adopted. A control method based on non-linear optimal control using state feedback is developed and the solution of the non-linear optimal control problem is obtained by representing system non-linearities and performance indices by power series with the help of algebraic tensor theory. In this work, general polynomial representations of the non-linear control law are obtained up to the fifth order. This solution procedure is employed to analyse the influence of the resulting non-linear control laws on the dynamic behaviour of a buckled beam under a lateral step load. This arch-like structural system is highly non-linear and under compressive lateral loading may suffer snap-through buckling. This may cause undesirable stresses and/or displacements, leading as a rule to a failure of the structural system. So, special attention is given to the determination of the potential of the present control methodology for efficiently limiting extreme state responses and preventing the snap-through buckling. Numerical results indicate that the control algorithm can effectively increase the load-carrying capacity of the buckled beam without demanding large control forces. Also, this study can be used as a basis for the non-linear control of more complex structures and for the design of control systems.