國立虎尾科技大學 |

Global and Local Sensitivity-Based Design of Robust Precision Motion Controllers.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Global and Local Sensitivity-Based Design of Robust Precision Motion Controllers./
Author:	Stein, Adrian.
Description:	1 online resource (254 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
Contained By:	Dissertations Abstracts International85-12B.
Subject:	Mechanical engineering. -
Online resource:	click for full text (PQDT)
ISBN:	9798382831275

Global and Local Sensitivity-Based Design of Robust Precision Motion Controllers.
Stein, Adrian.

Global and Local Sensitivity-Based Design of Robust Precision Motion Controllers. - 1 online resource (254 pages)

Source: Dissertations Abstracts International, Volume: 85-12, Section: B.

Thesis (Ph.D.)--State University of New York at Buffalo, 2024.

Includes bibliographical references

This dissertation focuses on precision motion control of vibratory systems in the presence of model parameter uncertainties. The use of local sensitivity and global sensitivity measure in the design of robust control profiles for point-to-point motion are explored. Extensive experimental validation is used to validate the simulation results and generate statistical measures of the output of the tests. The dissertation starts with the exploration of minimum-time point-to-point motion control of a gantry crane system with velocity bounds. Optimal control theory is used to guide the synthesis of optimal bang-off-bang profiles control profiles whose parameters are solved by formulating a nonlinear programming problem. Sensitivity states are used to augment the system model for the synthesis of robust motion control profiles. Experimental validation is performed on a two-degree-of-freedom gantry crane system with a suspended payload which includes a mechanism to consistently vary the length of the pendulum to emulate uncertainties in the natural frequency of vibration of the pendulum. Box and whisker charts are used to illustrate and compare the performance of the time-optimal and robust time-optimal controllers. The second approach proposed for precision motion control profiles are FIR filters which target an uncertain low frequency mode of vibration and can deal with higher modes of excitation as unmodeled dynamics. Magnitude constraints in the frequency domain are imposed in a convex programming framework to ensure attenuation of dominant resonant modes and minimal excitation of high-frequency unmodeled modes. Different cost functions for filter optimization, filter lengths, transition widths, and robustness in the notch area are studied too. Experimental validation is conducted using a clamped beam with a tip mass, comparing time-delay filters with the proposed Finite Impulse Response filters. The third contribution related to precision motion control focuses on real-time estimation of modal parameters to permit an adaptive input-shaping design to regulate payloads of gantry cranes undergoing point-to-point maneuvers. The modal parameter estimation is enabled by the use of computer vision and ArUco markers and the experiments illustrate the feasibility of real-time estimation and control of uncertain vibratory systems. Another core contribution of this dissertation is the synthesis of a new global sensitivity measure based on Shapley effects. This is a variance-based metric which involves evaluation of numerous marginal contributions of uncertain variables to characterize the uncertainties in the output of interest. Benchmark problems are studied to gauge the performance of the Shapley-based measure and Shapley effects are shown to routinely outperform Sobol indices in correctly rank ordering the importance of uncertain variables. Since the Monte Carlo cost of evaluating Shapley effects is high, polynomial chaos based surrogate models are developed and Shapley effects are shown to be algebraic maps of the polynomial chaos coefficients. The Shapley effects are used to illustrate a robust design using a Euler-Bernoulli beam with two lumped masses, where the locations of the masses are uncertain. The aim is to minimize or maximize the impact uncertain second mass location on the second mode of the beam structure. Finally, a controller design approach is proposed based on Shapley effects where it is experimentally confirmed that Shapley-based controller designs outperform the traditional time-delay filter structure when uncertainty is applied to both lumped masses.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798382831275Subjects--Topical Terms:

557493
Mechanical engineering.
Subjects--Index Terms:

Global sensitivity analysisIndex Terms--Genre/Form:

554714
Electronic books.

Global and Local Sensitivity-Based Design of Robust Precision Motion Controllers.
LDR:05103ntm a22004217 4500 001 1149738
005 20241022112627.5
006 m o d
007 cr bn ---uuuuu
008 250605s2024 xx obm 000 0 eng d
020 $a 9798382831275
035 $a (MiAaPQ)AAI31293452
035 $a AAI31293452
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Stein, Adrian. $3 1476068
245 1 0 $a Global and Local Sensitivity-Based Design of Robust Precision Motion Controllers.
264 0 $c 2024
300 $a 1 online resource (254 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
500 $a Advisor: Singh, Tarunraj.
502 $a Thesis (Ph.D.)--State University of New York at Buffalo, 2024.
504 $a Includes bibliographical references
520 $a This dissertation focuses on precision motion control of vibratory systems in the presence of model parameter uncertainties. The use of local sensitivity and global sensitivity measure in the design of robust control profiles for point-to-point motion are explored. Extensive experimental validation is used to validate the simulation results and generate statistical measures of the output of the tests. The dissertation starts with the exploration of minimum-time point-to-point motion control of a gantry crane system with velocity bounds. Optimal control theory is used to guide the synthesis of optimal bang-off-bang profiles control profiles whose parameters are solved by formulating a nonlinear programming problem. Sensitivity states are used to augment the system model for the synthesis of robust motion control profiles. Experimental validation is performed on a two-degree-of-freedom gantry crane system with a suspended payload which includes a mechanism to consistently vary the length of the pendulum to emulate uncertainties in the natural frequency of vibration of the pendulum. Box and whisker charts are used to illustrate and compare the performance of the time-optimal and robust time-optimal controllers. The second approach proposed for precision motion control profiles are FIR filters which target an uncertain low frequency mode of vibration and can deal with higher modes of excitation as unmodeled dynamics. Magnitude constraints in the frequency domain are imposed in a convex programming framework to ensure attenuation of dominant resonant modes and minimal excitation of high-frequency unmodeled modes. Different cost functions for filter optimization, filter lengths, transition widths, and robustness in the notch area are studied too. Experimental validation is conducted using a clamped beam with a tip mass, comparing time-delay filters with the proposed Finite Impulse Response filters. The third contribution related to precision motion control focuses on real-time estimation of modal parameters to permit an adaptive input-shaping design to regulate payloads of gantry cranes undergoing point-to-point maneuvers. The modal parameter estimation is enabled by the use of computer vision and ArUco markers and the experiments illustrate the feasibility of real-time estimation and control of uncertain vibratory systems. Another core contribution of this dissertation is the synthesis of a new global sensitivity measure based on Shapley effects. This is a variance-based metric which involves evaluation of numerous marginal contributions of uncertain variables to characterize the uncertainties in the output of interest. Benchmark problems are studied to gauge the performance of the Shapley-based measure and Shapley effects are shown to routinely outperform Sobol indices in correctly rank ordering the importance of uncertain variables. Since the Monte Carlo cost of evaluating Shapley effects is high, polynomial chaos based surrogate models are developed and Shapley effects are shown to be algebraic maps of the polynomial chaos coefficients. The Shapley effects are used to illustrate a robust design using a Euler-Bernoulli beam with two lumped masses, where the locations of the masses are uncertain. The aim is to minimize or maximize the impact uncertain second mass location on the second mode of the beam structure. Finally, a controller design approach is proposed based on Shapley effects where it is experimentally confirmed that Shapley-based controller designs outperform the traditional time-delay filter structure when uncertainty is applied to both lumped masses.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Mechanical engineering. $3 557493
650 4 $a Applied mathematics. $3 1069907
650 4 $a Energy. $3 784773
650 4 $a Mechanics. $3 527684
653 $a Global sensitivity analysis
653 $a Input shaping
653 $a Optimal control
653 $a Precision motion control
653 $a Uncertainty quantification
653 $a Vibration control
655 7 $a Electronic books. $2 local $3 554714
690 $a 0548
690 $a 0346
690 $a 0364
690 $a 0791
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a State University of New York at Buffalo. $b Mechanical and Aerospace Engineering. $3 845600
773 0 $t Dissertations Abstracts International $g 85-12B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31293452 $z click for full text (PQDT)