Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.306765
Title: Multi-objective scheduling and control of a nonlinear automotive powertrain
Author: Garbett, Kay Susan
ISNI:       0000 0001 3488 2660
Awarding Body: University of Warwick
Current Institution: University of Warwick
Date of Award: 1991
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
The automotive industry is faced with the challenge of ever-increasing emission legislation. This study demonstrates the effective use of nonlin­ear techniques in automotive control for the problem of fuel and emission minimisation. A review of previous work highlights the inadequacy of traditional optimisation formulations. The conflicting requirements of both low fuel and emissions is a design problem for which compromise and trade-offs are unavoidable. This study attacks the problem through powertrain scheduling, an approach ideally suited to both S.I. and diesel engines, and demonstrates how the novel application of multi-objective optimisation methods provides a solution more akin to the real physical problem. The modern control theory approach presented is a three stage pro­cess : formulation of the mathematical model, including the essential dy­namics, constraints, and objectives of the physical problem; optimisation of the control strategy with respect to the relevant performance criteria; and synthesis of the optimal control design. The optimisation model is finite-dimensional and nonlinear, the use of which demands a knowledge of nonlinear systems and available methods. These are classified. Re­sults for single and multi-objective optimisations are compared and fully demonstrate the advantages of the latter for the scheduling problem. Op­timal schedules are generated and from them, implementable rule-based control laws are derived. Performance, in terms of the ability to track a legislative test cycle and to retain the optimal design specification, is demonstrated through dynamic simulation, as is their driveability and robustness. This study specifically considers a diesel-engined vehicle incorporating a CVT. The methods tire widely applicable however, to other engine and transmissions types, and to other automotive control problems.
Supervisor: Not available Sponsor: Science and Engineering Research Council ; Lucas Automotive
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.306765  DOI: Not available
Keywords: TA Engineering (General). Civil engineering (General)
Share: