Nonlinear control system design using variable complexity modelling and multiobjective optimization
AUTOR(ES)
Silva, Valceres V. R. E, Khatib, Wael, Fleming, Peter J.
FONTE
Sba: Controle & Automação Sociedade Brasileira de Automatica
DATA DE PUBLICAÇÃO
2006-03
RESUMO
To design controllers for complex non-linear systems usually involves the use of expensive computational models. A non-linear thermodynamic model of a gas turbine engine is used to evaluate a selection of designs for a multivariable PI controller configuration. An approach using variable complexity modelling (VCM) is introduced to allow more designs to be evaluated and also to speed up the design process. Response surface methodology (RSM) is a statistical technique in which smooth functions are used to model an objective function. RSM employs statistical methods to create functions, typically polynomials, to model the response or outcome of a numerical experiment in terms of several independent variables. Regression analysis is applied to fit polynomial models to this data for various control responses. These control responses models are evaluated by a multiobjective genetic algorithm to design the controller parameters. The final designs are checked using the original non-linear model.
Documentos Relacionados
- A fly-by-wire rudder control system design using multiobjective optimization.
- Design of distributed optical-fiber raman amplifiers using multi-objective particle swarm optimization
- Control system design for a gas turbine engine using evolutionary computing for multidisciplinary optimization
- Saddle point and second order optimality in nondifferentiable nonlinear abstract multiobjective optimization
- Multiobjective engineering design optimization problems: a sensitivity analysis approach