Browsing by Author "Fonseca, C. M."
Now showing 1 - 10 of 12
Results Per Page
Sort Options
- Accelerating multi-objective control system design using a neuro-genetic approachPublication . Duarte, N. M.; Ruano, Antonio; Fonseca, C. M.; Fleming, P. J.Designing control systems using multiobjective genetic algorithms can lead to a substantial computational load as a result of the repeated evaluation of the multiple objectives and the population-based nature of the search. Here, a neural network approach, based on radial basis functions, is introduced to alleviate this problem by providing computationally inexpensive estimates of objective values during the search. A straightforward example demonstrates the utility of the approach.
- An overview of nonlinear identification and control with neural networks.Publication . Ruano, Antonio; Ferreira, P. M.; Fonseca, C. M.The aim of this chapter is to introduce background concepts in nonlinear systems identification and control with artificial neural networks. As this chapter is just an overview, with a limited page space, only the basic ideas will be explained here. The reader is encouraged, for a more detailed explanation of a specific topic of interest, to consult the references given throughout the text. Additionally, as general books in the field of neural networks, the books by Haykin [1] and Principe et al. [2] are suggested. Regarding nonlinear systems identification, covering both classical and neural and neuro-fuzzy methodologies, Reference 3 is recommended. References 4 and 5 should be used in the context of B-spline networks.
- Bacterial memetic algorithm for fuzzy rule base optimizationPublication . Cabrita, Cristiano Lourenço; Botzheim, J.; Gedeon, Tamas D.; Ruano, Antonio; Kóczy, László T.; Fonseca, C. M.In our previous works model identification methods were discussed. The bacterial evolutionary algorithm for extracting a fuzzy rule base from a training set was presented. The Levenberg-Marquardt method was also proposed for determining membership functions in fuzzy systems. The combination of evolutionary and gradient-based learning techniques – the bacterial memetic algorithm – was also introduced. In this paper an improvement of the bacterial memetic algorithm is shown for fuzzy rule extraction. The new method can optimize not only the rules, but can also find the optimal size of the rule base.
- Developing redundant binary representations for genetic searchPublication . Fonseca, C. M.; Correia, Marisol B.This paper considers the development of redundant representations for evolutionary computation. Two new families of redundant binary representations are proposed in the context of a simple mutationselection evolutionary model. The first is a family of linear encodings in which the connectivity of the search space may be designed directly via a decoding matrix. The second is a family of representations exhibiting various degrees of neutrality, and is constructed using mathematical tools from error-control coding theory. The study of these representations provides additional insight into the properties of redundant encodings, such as synonymity, locality, and connectivity, and into their interrelationships.
- Evolutionary multiobjective design of radial basis function networks for greenhouse environmental controlPublication . Ferreira, P. M.; Ruano, Antonio; Fonseca, C. M.In this work a multiobjective genetic algorithm is applied to the identi cation of radial basis function neural network coupled models of humidity and temperature in a greenhouse. Models are built as one-step-ahead predictors and then used iteratively to produce long term predictions. The number of neurons and input terms used in both models de ne the search space. Two combinations of performance and complexity criteria are used to steer the selection of model structures, resulting in distinct sets of solutions. It is shown that minimisation of one-step-ahead prediction errors negatively in uences long term prediction performance. Long term prediction results are presented for a pair of models selected from sets of models obtained in the experiments.
- Genetic assisted selection of RBF model structures for greenhouse inside air temperature predictionPublication . Ferreira, P. M.; Ruano, Antonio; Fonseca, C. M.This paper presents results on the application of Multi-Objective Genetic Algorithms to the selection of Radial Basis Function Neural Networks structures. The neural networks are to be incorporated in a real-time predictive greenhouse environmental control strategy, as' predictors of the inside air temperature. Previous research conducted by the authors modelled the inside air temperature, as a function of the inside relative humidity and of the outside temperature and solar radiation. A second-order model structure previously selected in the context of dynamic temperature models identification was used. Several training and learning methods were compared, and the application of the Levenberg-Frquardt optimisation method was found to be the best way to determine the neural network parameters. The application of correlation-based model-validity tests revealed that the validity of such a second-order model structure could be manually improved after inspection of the tests results. Both network performance and validity are certainly affected by the number of neurons, the input variables considered and the time delays used. As the number of alternatives is huge, Multi-Objective Genetic Algorithms are applied here to the selection of network inputs and number of neurons.
- How redundancy and neutrality may affect evolution on NK fitnessPublication . Correia, Marisol B.; Fonseca, C. M.An experimental study was performed to determine whether it is neutrality itself or the larger neighborhoods associated with neutral representations that allow good results to be achieved on NK fitness landscape problems. Markov chains were used to model a stochastic hill climber on NK fitness landscapes, using three types of representation: a neutral network representation, a redundant representation without neutrality which exhibits the same neighborhood of the neutral representation and a non-redundant representation.
- Neuro-genetic Pid autotuningPublication . Lima, João; Azevedo, Ana Beatriz da Piedade de; Duarte, N. M.; Fonseca, C. M.; Ruano, Antonio; Fleming, P. J.A new PID autotuning technique, involving neural networks and genetic algorithms is proposed. The validity of this approach is shown, through the results of several experiments. Special attention is given to the off-line training of one of the auto-tuner models, the criterion networks. Procedures used to obtain good training data are described.
- Nonlinear identification of aircraft gas-turbine dynamicsPublication . Ruano, Antonio; Fleming, P. J.; Teixeira, C. A.; Rodriguez-Vázquez, K.; Fonseca, C. M.Identi cation results for the shaft-speed dynamics of an aircraft gas turbine, under normal operation, are presented. As it has been found that the dynamics vary with the operating point, nonlinear models are employed. Two di7erent approaches are considered: NARX models, and neural network models, namely multilayer perceptrons, radial basis function networks and B-spline networks. A special attention is given to genetic programming, in a multiobjective fashion, to determine the structure ofNARMAX and B-spline models.
- On the roles of redundancy and neutrality in evolutionary optimizationPublication . Correia, Marisol B.; Fonseca, C. M.An experimental study was performed to explore whether it is neutrality itself or simply the larger neighborhoods associ- ated with neutral representations that influence the results achieved by evolutionary algorithms on NK fitness landscape problems. Markov chains were used to model the behaviour of a stochastic hill-climber on NK fitness landscapes, using two different types of representation: a neutral network rep- resentation which exhibits neutrality and a redundant rep- resentation without neutrality which implements the same neighborhood induced by the corresponding neutral repre- sentation.