Identification and decentralized control of a dual-actuator hard disk drive system

The dual-actuator hard disk drive with two voice coil motor actuators mounted in the diagonal corners of the baseplate allows faster data access with relatively slow spindle speed. However, the mechanical interaction between the actuators tends to affect the head positioning accuracy. In this paper, we use the circle-fit modal testing method to identify the dual-input dual-output frequency response model of the dual-actuator plant. We discuss the decentralized control scheme for the dual-actuator tracking servo. The servo loop decomposition reveals that the interaction induces narrow-band vibrations near the crossover frequency. We propose a phase-lead interaction filter with a slight change to the loop stability to suppress the vibrations. Experimental results on a dual-actuator prototype show that the position error signal 6/spl sigma/ value due to interaction vibrations is reduced by 61% with the interaction filter in combination with the decentralized control scheme.     

Longitudinal recording in a stray field

We have studied the sensitivity of soft error rate (SER) to an external magnetic field in longitudinal hard disk drives. We found that during writing SER is generally quadratic with field amplitude and symmetric with field direction. The application of an external field during writing also shifts the center of the track. In contrast, during reading SER is predominantly linear with field amplitude and antisymmetric with field direction. There is no shift in the track center when the field is applied during reading up to 100 Oe. We explain the difference in the track center shift between writing and reading in terms of the effect of the stray field on the servo positioning system.     

Synchronous servo scheme using maximum-likelihood detectors

For magnetic disk drives, we have developed a scheme for synchronous servo with maximum-likelihood (SSML) detection that is synchronous around the entire circumference of the track and utilizes ML estimation to detect head position, clock mark phase, and track addresses. Compared to the conventional asynchronous sector servo scheme, the SSML scheme has many, much shorter, servo areas-typically several hundred-distributed on the disk. This results in higher servo information rate, which will be needed for future high-track-density disk drives with a wide servo bandwidth. We analyze the performance of the SSML scheme, including the format efficiency. We discuss a limitation of the synchronous servo scheme with respect to the number of servo areas, and we show an appropriate region for its application. Feasibility of the SSML scheme has been demonstrated in prototype 2.5-in hard disk drives.     

Robust Compensation of Periodic Disturbances by Multirate Control

We propose an adaptive feedforward cancellation (AFC) scheme based on multirate discretization for compensation of periodic disturbance in hard disk drive servo systems. By discretizing the plant by a multirate scheme, we obtain the inverse of the closed-loop transfer function. Our scheme cancels selected repeatable runout (RRO) components more rapidly even when the disturbance signal is time-varying in both frequency and amplitude. We also show by simulation and experiment that the scheme can attenuate selected RRO harmonics effectively in a wide frequency range without amplifying other frequency harmonics, which cannot be achieved by conventional AFC. Our scheme is robust against time-varying periodic disturbances and actuator resonance frequency shifting.     

Use of an upwind finite volume method to solve the air bearing problem of hard disk drives

In this paper we present an upwind higher order finite volume numerical scheme over unstructured triangular mesh to solve the slider air bearing problem of hard disk drives. The scheme is nodal based, which uses the median dual as the control volume. The convection part of the generalized Reynolds equation is modeled by the flux difference splitting technique. Higher order accuracy in space is achieved by a linear reconstruction technique with a flux limiting technique incorporated to prevent oscillation in the high-pressure gradient regions. A linear Galerkin method is used to discretize the diffusion terms. In addition, a non-nested multi-grid iteration technique is used to increase the convergence rate. Finally, the steady state flying attitude of a slider subject to pre-applied suspension force and torques is obtained by a Quasi–Newton iteration method, and the results of this scheme are compared with two other schemes. Received 16 July 2000     

基于自适应RBF模糊神经网络的旋转柔性铰接梁的振动控制

由柔性关节连接中心刚体和挠性附件的刚柔耦合系统广泛应用于卫星太阳能帆板、空间机器人等领域中,在调姿或者外部扰动带来振动时,将影响系统的稳定性和指向精度,对带有铰接结构的柔性梁的影响更甚。设计并建立了带有柔性关节（谐波齿轮）的旋转柔性铰接梁实验平台,进行了基于压电传感器测量信号的振动频响特性分析,分别采用PD控制和自适应RBF模糊神经网络控制算法,进行了基于电机驱动的位置设定点弯曲振动的主动控制研究。实验比较结果验证设计的自适应RBF模糊神经网络控制算法能够快速抑制振动。

      

Modified dual-stage controller for dealing with secondary-stage actuator saturation

This paper proposes a modified decoupled master-slave (DMS) dual-stage control scheme for hard disk drives that offers enhanced stability when the secondary-stage actuator is saturated. A nonlinear secondary-stage actuator model is used instead of a linear model in the controller. This modification allows microactuators with relatively smaller displacement range to be used in the servo system without causing stability problems in the DMS controller. Simulation and experimental results confirm the effectiveness and robustness of the scheme.     

Neural Networks in Applied Statistics

Artificial neural networks are computer algorithms or computer programs derived in part from attempts to model the activity of nerve cells. They have been applied to pattern recognition, classification, and optimization problems in the physical and chemical sciences, as well as in other fields. We introduce the principles of the multilayer feedforward network that is among the most commonly used neural networks in practical problems. The relevance of neural network models for the applied statistician is considered using a time series prediction problem as an example. The multilayer feedforward neural network uses a nonlinear function of the predictors to obtain predictions for future time series values. We illustrate the considerations involved in specifying a neural network model and evaluate the accuracy of neural network models relative to the accuracy obtained using other computer-intensive, nonmodel-based techniques.     

A novel damping scheduling scheme for proximate time optimal servomechanisms in hard disk drives

The proximate time-optimal servomechanism (PTOS) is widely used in hard disk drives because of its simplicity and good performance. There have been many studies to improve PTOSs, though they are complex or need a time-consuming tuning process. Here, we propose a novel damping scheduling PTOS (DSPTOS) for hard disk drives. The proposed algorithm keeps the closed-loop damping coefficient at a predetermined value. Compared to the original PTOS, smaller damping in deceleration will boost the rising speed, and larger damping in the settling phase reduces the overshoot and residual vibrations. Moreover, the DSPTOS maintains the simplicity of the PTOS by introducing just a predetermined damping coefficient. We applied the proposed DSPTOS algorithm to a voice coil motor in a hard disk drive. Through simulations and real-time experiments, we verified the effectiveness of the DSPTOS. In particular, with respect to long stroke seeks, the DSPTOS performs better than the PTOS.     

基于软件误差修正的通用直线工作台的纳米级定位控制

提出了一种用于通用压电陶瓷线性马达(PLCM)驱动平台的纳米级定位数控方案,该平台上配有栅距20μm的线性光栅.传统的信号细分多数基于硬件电路,无法对纳米级误差进行灵活的调整.本项研究开发了基于Labview的数字控制系统,该系统具有失真波形的调整、1/4栅距脉冲计数、基于Lissajous圆环的波形细分、后传神经网络和PID控制以及平台的位置反馈控制等特征.介绍了定位控制的3个步骤,即用于长行程连续移动控制的AC模式,用于低速短行程的开关模式控制马达驱动,以及以压电激励方式使马达工作的DC模式,这些步骤能够实现几纳米的精确定位.在每种运动模式下采用了专门的信号处理技术.实验结果表明,此方法可以很容易地应用于线性平台,取得小于31 nm的定位精度,9 mm行程的标准误差小于30 nm.相比于原来移动台±1μm的重复性和±1μm/25 mm的精度,该数字控制系统能够将定位精度提高一个数量级.     

Nonrepeatable Run-out Rejection Using Online Iterative Control for High-Density Data Storage

The spectra of disturbances and noises affecting precise servo positioning for ultrahigh-density storage in future hard disk drives are time-varying and remain unknown. In this paper, we propose an online iterative control algorithm that sets the measured position error signal (PES) into the servo system to achieve high track densities by minimizing the square of the H₂-norm of the transfer function from nonrepeatable run-out (NRRO) disturbances to the true PES. It is not necessary to solve any algebraic Riccati equations and linear matrix inequalities. The algorithm constructs an online repeatable run-out estimator to extract NRRO components for gradient estimates, thereby preventing the controller parameters from being trapped in a local minima. Experimental results on a PC-based servo system for a spinstand show an improvement of 22% in 3sigma NRRO and suppression of baseline NRRO spectrum     

Noninvasive Vibration Diagnostics to Detect Head–Disk Contact in Hard Disk Drives

Reliability of modern hard disk drives (HDDs) is a major concern especially with decreasing head-to-disk spacing in order to attain higher recording densities. One of the reasons for HDD failure is the occurrence of intermittent or sustained head–disk contacts. It is therefore important to be able to detect such head–disk contacts early on so that backup safety schemes can be implemented before catastrophic HDD failure and permanent data loss. In the study reported here, we used noninvasive vibration diagnostics and multiaxial vibration measurements to assess the condition and functionality of HDDs. We chose a miniature triaxial accelerometer for the diagnostic study, given that it is relatively inexpensive and easy to use. We measured contact vibration, induced by a scratch on the top surface of the rotating disk, on the outside of the HDD by attaching an accelerometer on the actuator shaft. We found that contact vibrations between the head and disk inside the HDD can reliably be measured on the outside of the HDD with an accelerometer. We also found that a miniature accelerometer does not cause significant mass loading, as verified by noncontacting laser vibrometry.     

基于余弦基神经网络的FIR滤波器设计

提出了一种基于余弦基神经网络FIR滤波器的设计方法。根据线性相位FIR滤波器的幅频特性是有限项的傅里叶级数,构造了一个三层余弦基神经网络模型,并给出了最优隐层神经元的个数。经模拟仿真,滤波器性能非常理想,具有同时输出低通、高通、带通、带阻各种功能,通带、阻带无过冲无波动,边界频率可以精确控制。     

面向微装配的压电陶瓷微夹钳建模与控制

建立了悬臂梁位移与外电场和外力之间的动静态模型以及压电陶瓷迟滞环的Backlash算子模型.以所建立的迟滞模型为前馈环节与PID反馈构成复合控制对悬臂梁进行位置控制.现场测试数据验证了该模型的正确性和复合控制方法的可行性,并分别用逆控制、PID控制以及复合控制方法进行了正弦曲线跟踪实验,这些控制方法的比较结果表明,复合控制策略具有最高的控制精度.     

模糊自适应滤波的主动控制方法研究

由于主动控制中广泛使用的自适应滤波 - x L MS算法只适用于线性控制问题 ,针对一些非线性问题 ,本文提出利用一种非线性自适应滤波方法——基于模糊逻辑系统的自适应滤波方法来解决一类参考信号与外扰呈非线性函数关系的前馈主动控制问题。仿真结果表明 ,该模糊自适应滤波器优于线性滤波器的控制效果     

Novel method for minimizing track seeking residual vibrations ofhard disk drives

Track access time is an important parameter for high-performance hard disk drives (HDDs). Presently, the resonances of the head actuator caused by the pivot-bearing stiffness are an even more urgent problem, since these resonances have become the main design factor that limits further reduction of track access time and improvement of the servo control bandwidth in HDDs. Our objective in this paper is to investigate, from a mechanics and dynamics viewpoint, the influence of the driving force waveforms on residual vibrations and to identify the characteristics of the waveform, which will lead to zero or very low residual vibrations. Using a simplified model of a head actuator assembly and finite-element simulation, we show that the residual vibration of the head actuator assembly depends strongly on the waveform and the time duration of the driving impulse force. If suitable waveform and time duration of the seeking impulse force are appropriately chosen, very low track seeking residual vibrations and quick track access can be achieved     

基于模糊神经网络的包装机械臂定位方法研究

目的 为提高包装机械臂的抓取精度,文中基于模糊神经网络设计一种包装机械臂定位方法。方法 将激光测距仪与工业相机融合,可实现目标点的初步定位并得到位姿偏差。以机械臂末端位置误差补偿为例,设计一种模糊神经网络控制器,可实现PID控制关键参数的在线调整以提高误差补偿精度。进一步地,采用果蝇优化算法实现神经网络控制器初始值的优化,可提高控制系统性能。最后,进行实验研究。结果 实验结果表明,机械臂定位算法可使最大绝对误差从7.704 9 mm下降到1.424 2 mm;平均绝对定位误差降低约82.5%;机械臂执行效率与对照组相当。结论 该定位方法可以大幅度提高包装机械臂定位精度,可满足包装、化工、食品等相关行业要求。     

漂浮基闭链空间机械臂抓持载荷基于神经网络的反演自适应控制

讨论漂浮基闭链空间机械臂抓持系统目标载荷的位置、力混合协调控制问题。结合系统动量、动量矩守恒关系及闭合运动链几何约束关系,利用多刚体动力学建模方法建立了漂浮基闭链双臂空间机器人抓持系统的合成动力学方程。针对参数不确定和外部干扰的情况,提出一种基于反演神经网络的位置、力自适应协调控制方案,从而达到对抓持负载位置与所受内力的双重控制效果。此控制方案无需要求系统动力学方程关于惯性参数呈线性函数关系;同时,它可保证系统渐近稳定性。数值仿真结果证实了该控制方案可有效地消除参数不确定对系统的影响。     

非线性质量补偿方法在微量包装系统中的应用

目的 为解决微量包装系统中称量传感器输出电压与质量之间的非线性关系、提高称量精度,基于改进BP神经网络设计一种非线性补偿方法.方法 阐述电阻应变式称量传感器的非线性补偿原理,根据称量传感器输入和输出之间的关系,设计一种神经网络补偿器.为提高神经网络控制性能,引入一种惩罚因子,可解决因训练不足导致的误差偏大等问题.结果 经对比发现,改进型BP神经网络具有较快的收敛速度、较高的精度,可提高微量称量包装系统的控制性能.高速模式下,称量误差可以控制在0.5％以内,实际称量结果较理想.结论 该方法能够改善系统动态性能,提高测量精度,可满足称量、包装行业等精度要求.     

Fly Height Measurement Based on Phase Comparison Michelson Interferometry Using Low-Coherence Light Source

We present a method for measuring head fly height in hard disk drives using phase comparison Michelson interferometry (PCMI), which compares the phases of two interference fringe patterns formed respectively on the inner surface of a glass disk and the air-bearing surface of a fly head slider through the glass disk. To suppress interference noise and further enhance measurement accuracy, we adopted a low-coherence light source as an illumination source to replace a high-coherence laser. The captured fringe images enabled us to extract ridge lines much more accurately than with a laser. We compared our measurements in the sub-10 nm spacing range with calculations and found excellent agreement.