Performance enhancement of hard disk drives through data-driven control design and population clustering

A data driven control design approach in the frequency domain is used to design track following feedback controllers for dual-stage hard disk drives using multiple data measurements. The advantage of the data driven approach over model based approach is that, in the former approach the controllers are directly designed from frequency responses of the plant, hence avoiding any model mismatch. The feedback controller is considered to have a Sensitivity Decoupling Structure. The data driven approach utilizes H∞ and H2 norms as the control objectives. The H∞ norm is used to shape the closed loop transfer functions and ensure closed loop stability. The H2 norm is used to constrain and/or minimize the variance of the relevant signals in time domain. The control objectives are posed as a locally convex optimization problem. Two design strategies for the dual-stage hard disk drive are presented.

Download Full-text

Application of MIMO Data Driven Feedback Control Design to Dual Stage Hard Disk Drives

Volume 1: Adaptive/Intelligent Sys. Control; Driver Assistance/Autonomous Tech.; Control Design Methods; Nonlinear Control; Robotics; Assistive/Rehabilitation Devices; Biomedical/Neural Systems; Building Energy Systems; Connected Vehicle Systems; Control/Estimation of Energy Systems; Control Apps.; Smart Buildings/Microgrids; Education; Human-Robot Systems; Soft Mechatronics/Robotic Components/Systems; Energy/Power Systems; Energy Storage; Estimation/Identification; Vehicle Efficiency/Emissions ◽

10.1115/dscc2020-3260 ◽

2020 ◽

Author(s):

Prateek Shah ◽

Zhi Chen ◽

Roberto Horowitz

Keyword(s):

Feedback Control ◽

Frequency Response ◽

Design Methodology ◽

Hard Disk Drives ◽

Control Design ◽

Hard Disk ◽

Data Driven ◽

Disk Drives ◽

Feedback Controllers ◽

Dual Stage

Abstract With increasing data density on hard disk drives, there is need to develop more robust and better performing track following control systems. We present a multi-input multi-output (MIMO) data driven feedback control design methodology. The design considers multiple frequency response measurements of all actuators, simultaneously, ensuring robustness of the control design system. A mixed H2 – H∞ norm locally convex optimization algorithm is used to synthesize the feedback controllers for MIMO systems. Feedback controllers are developed for dual stage hard disk drives using the MIMO data driven control design technique. A dual stage hard disk drive comprises of two actuators in series, controlling a read/write head onto a rotating disk. Our objective is to stabilize the closed loop of the actuators and minimize the error position signal of the read/write head. H2 norm and H∞ norm control objectives are used to formulate the MIMO data driven control problem. The design is based on a set of five frequency response measurements of the two actuators. We also compare the MIMO design methodology to a single-input multi-output (SIMO) design methodology presented earlier [1].

Download Full-text

Mixed H2/H∞ Data-Driven Control Design for Hard Disk Drives

2018 Asia-Pacific Magnetic Recording Conference (APMRC) ◽

10.1109/apmrc.2018.8601117 ◽

2018 ◽

Cited By ~ 1

Author(s):

Omid Bagherieh ◽

Roberto Horowitz

Keyword(s):

Hard Disk Drives ◽

Control Design ◽

Hard Disk ◽

Data Driven ◽

Disk Drives

Download Full-text

Moving towards Real-time Data-driven Quality Monitoring: A Case Study of Hard Disk Drives

Procedia Manufacturing ◽

10.1016/j.promfg.2018.07.147 ◽

2018 ◽

Vol 26 ◽

pp. 1107-1115 ◽

Cited By ~ 5

Author(s):

Ardeshir Raihanian Mashhadi ◽

Willie Cade ◽

Sara Behdad

Keyword(s):

Real Time ◽

Hard Disk Drives ◽

Hard Disk ◽

Quality Monitoring ◽

Data Driven ◽

Disk Drives ◽

Time Data ◽

Real Time Data

Download Full-text

HEAD-POSITIONING CONTROL IN TRIPLE-STAGE-ACTUATOR ENTERPRISE HARD DISK DRIVES USING MIXED H2/H∞ SYNTHESIS METHODOLOGIES

ASME Letters in Dynamic Systems and Control ◽

10.1115/1.4051251 ◽

2021 ◽

pp. 1-8

Author(s):

Zhi Chen ◽

Prateek Shah ◽

Roberto Horowitz

Keyword(s):

Hard Disk Drives ◽

Voice Coil Motor ◽

Hard Disk ◽

Data Driven ◽

Disk Drives ◽

Positioning Accuracy ◽

Positioning Control ◽

Model Based ◽

Head Positioning ◽

Resonance Frequencies

Abstract The recent rapid growth in the cloud storage industry has strongly increased the demand for high-capacity hard disk drives (HDDs). Increasing the areal density brings new challenges to the high-accuracy head-positioning control in the next generation HDD development. Triple-stage-actuator (TSA) system is one of the emerging technologies that can achieve higher bandwidth than that of a dual-stage-actuator (DSA) system and improve the track-following performance. In this paper, we focus on the track-following controller design for TSA system with one voice coil motor (VCM) and two piezoelectric (PZT) actuators. Two types of mixed H2/Hinf synthesis methodologies based on model-based optimization and data-driven optimization are proposed to design the track-following controller for the TSA system. The TSA system can increase the bandwidth of the servo system and decrease the sensitivity to disturbances at the low-frequency range. While increasing the stroke limitation and the resonance frequency of the micro-actuator, the 3sigma the position error signals (PES) is reduced. The data-driven controller can achieve comparable head-positioning accuracy to the model-based controller when it converges to a local optimal solution. The simulation results show the feasibility and effectiveness of the TSA systems with a tertiary PZT actuator. We also analyze the effects of stroke limitations and resonance frequencies of the second/third-stage PZT actuators on the head-positioning accuracy. The results might provide a guideline for the TSA mechanical design.

Download Full-text