Nonlinear model based coordinated adaptive robust control of electro-hydraulic systems
Abstract
The dynamics of hydraulic systems are highly nonlinear. Aside from the nonlinear nature of hydraulic dynamics, hydraulic systems also have large extent of model uncertainties, either due to parametric uncertainties or uncertain nonlinearities. These nasty characteristics make the precision motion control of hydraulic systems rather difficult. As a stepping stone toward the systematic design of high performance control algorithms for hydraulic systems, a nonlinear model based adaptive robust control approach is presented and carefully examined through both rigorous theoretical analysis and experimental verifications. Specifically, nonlinear physical model based analysis and design is used to address the inherent nonlinearities of hydraulic dynamics. Adaptive robust control (ARC) is applied to deal with various model uncertainties effectively. The dissertation starts from the precision motion control of one DOF hydraulic servosystem driven by either a double-rod/double-actuating or a single-rod/double-actuating hydraulic cylinder with constant or time-varying unknown inertia. The stability proof of zero tracking error dynamics associated with single-rod hydraulic cylinders is given. Swing motion control of a three DOF robot arm driven by single-rod hydraulic actuators with the other two joints fixed or actuated simultaneously are used as experimental case studies. For systems that use cheaper proportional directional valves instead of servo valves, addressing strategies are presented to deal with the nonlinear valve characteristics such as deadband and nonlinear flow gain coefficients. Different methods are presented to improve the response of sluggish proportional directional control valves. The coordinated motion control of multi-DOF electro-hydraulic robotic arm is then studied. Two methods are proposed to avoid the need of acceleration feedback for ARC backstepping designs; one uses a nonlinear ARC observer to recover the state needed for the ARC backstepping design, and the other makes full use of the physical property that the adjoint matrix and the determinant of the inertial matrix could be linearly parametrized by certain suitably selected parameters and employs overparametrizing method. An integrated direct/indirect ARC algorithm is also constructed to meet the dual objectives of good tracking performance and parameter estimations. The resulting accurate parameter estimates may be used for other purposes such as machine and component health monitoring and fault detection.
Degree
Ph.D.
Advisors
Yao, Purdue University.
Subject Area
Mechanical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.