Use this URL to cite or link to this record in EThOS:
Title: New optimization methods in predictive control
Author: Shahzad, Amir
ISNI:       0000 0004 2693 8339
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2010
Availability of Full Text:
Access from EThOS:
Access from Institution:
This thesis is mainly concerned with the efficient solution of a linear discrete-time finite horizon optimal control problem (FHOCP) with quadratic cost and linear constraints on the states and inputs. In predictive control, such a FHOCP needs to be solved online at each sampling instant. In order to solve such a FHOCP, it is necessary to solve a quadratic programming (QP) problem. Interior point methods (IPMs) have proven to be an efficient way of solving quadratic programming problems. A linear system of equations needs to be solved in each iteration of an IPM. The ill-conditioning of this linear system in the later iterations of the IPM prevents the use of an iterative method in solving the linear system due to a very slow rate of convergence; in some cases the solution never reaches the desired accuracy. A new well-conditioned IPM, which increases the rate of convergence of the iterative method is proposed. The computational advantage is obtained by the use of an inexact Newton method along with the use of novel preconditioners. A new warm-start strategy is also presented to solve a QP with an interior-point method whose data is slightly perturbed from the previous QP. The effectiveness of this warm-start strategy is demonstrated on a number of available online benchmark problems. Numerical results indicate that the proposed technique depends upon the size of perturbation and it leads to a reduction of 30-74% in floating point operations compared to a cold-start interior point method. Following the main theme of this thesis, which is to improve the computational efficiency of an algorithm, an efficient algorithm for solving the coupled Sylvester equation that arises in converting a system of linear differential-algebraic equations (DAEs) to ordinary differential equations is also presented. A significant computational advantage is obtained by exploiting the structure of the involved matrices. The proposed algorithm removes the need to solve a standard Sylvester equation or to invert a matrix. The improved performance of this new method over existing techniques is demonstrated by comparing the number of floating-point operations and via numerical examples.
Supervisor: Kerrigan, Eric ; Constantinides, George A. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral