Loading...
7 results
Search Results
Now showing 1 - 7 of 7
- Fault isolation based on HSFNN applied to DAMADICS benchmark problemPublication . Calado, João Manuel Ferreira; Louro, R.; Mendes, Mário J. G. C.; Costa, J. M. G. Sá da; Kowal, M.The present paper is concemed with the application of a hierarchical structure of fuzzy newal networks (HSFNN) to fault isolation on a pneumatic servo-motor actuated valve that is the benchmark considered for all the DAMADICS (Development and Application of Methods for Actuator Diagnosis in IndusIrial Control Systems) project partners. The adoption of a hierarchical structure of fuzzy newal netwoIks for fault isolation pwposes aims the development of an architecture that can localise abrupt and incipient single and multiple faults correctly or at least with a minimum misclassification rate and be easily trained, ftom only single abrupt fault symptoms.
- Control of the archimedes wave swing using neural networksPublication . Beirão, Pedro; Mendes, Mário J. G. C.; Valério, Duarte; Costa, José Sá daThis paper addresses the control of the Archimedes Wave Swing, a fully-submerged Wave Energy Converter (WEC), of which a prototype has already been built and tested. Simulation results are presented in which Internal Model Control (IMC) is used, both with linear models and with non-linear neural network (NN) models. To the best of our knowledge this is the first time NN-based control is being applied to design a controller for a WEC. NNs are a mathematical tool suitable to model the behaviour of dynamic systems, both linear and non-linear (as in our case). Significant absorbed wave energy increases were found, both using linear models and NNs. Results were better when IMC with NNs was employed (with a nearly sixfold increase against a fivefold increase), except for the May—September period, when IMC with linear models performs better.
- Fault detection system for the Évora irrigation canalPublication . Louro, Diogo; Mendes, Mário J. G. C.; Valério, Duarte; Costa, José Sá daA model-based fault detection (FD) system was developed for a Simulink simulation of a four pool irrigation canal in ´Evora, Portugal. Incipient and abrupt faults in the gates, the water off-take valves and the water level sensors were considered. Neural Networks were used to model the canal and find the residue. The training algorithm employed for the NNs was found to be an important factor determining the success of the FD system.
- Fault diagnosis system based in agentsPublication . Mendes, Mário J. G. C.; Calado, João Manuel Ferreira; Costa, J. M. G Sá daIn this work is proposed a new agent based fault diagnosis system for complex and dynamic processes. The fault diagnosis systems of the future should have present the distribution and complexity of the processes and they must be able to cooperate and communicate with other systems to achieved a satisfactory performance. The fault detection and isolation (FDI) agents proposed here have hybrid architectures based in a horizontal layered architecture. The reactive layer of the FDI agents are based in decomposition wavelet methods for the fault detection and in neural networks for the fault isolation task. The new agent based FDI system is applied to fault diagnosis in a three tank process.
- Neuro and neuro-fuzzy hierarchical structures comparison in FDI: a case studyPublication . Calado, João Manuel Ferreira; Mendes, Mário J. G. C.; Costa, J. M. G. Sá da; Korbicz, JózefIn this paper a hierarchical structure of several artificial neural networks has been developed for fault isolation purposes. Two different approaches have been considered. The hierarchical structure is the same for both approaches, but one uses multi-layer feedforward artificial neural networks and the other uses fuzzy neural networks. A result comparison between the two architectures will be presented. It is aimed to isolate multiple simultaneous abrupt and incipient faults from only single abrupt fault symptoms. A continuous binary distillation column has been used as test bed of the current approaches.
- Identification and control of the AWS using neural network modelsPublication . Valério, Duarte; Mendes, Mário J. G. C.; Beirão, Pedro; Costa, José Sá daThe Archimedes Wave Swing (AWS) is a a fully-submerged Wave Energy Converter (WEC), that is to say, a device that converts the energy of sea waves into electricity. A first prototype of the AWS has already been built and tested. In this paper, neural network (NN) models for this AWS prototype are developed. NNs are then used together with proven control strategies (phase and amplitude control, internal model control and switching control) to maximise energy production. Simulations show an yearly average electricity production increase of 160% over the performance of the original AWS controller.
- Switching control of the archimedes wave swingPublication . Valério, Duarte; Costa, José Sá da; Mendes, Mário J. G. C.; Beirão, PedroControl switching is applied to the Archimedes Wave Swing (AWS), a device designed to convert the energy of sea waves into electricity. Previous simulations showed that energy production can be significantly increased using Internal Model Control, together with direct and inverse Elman Neural Network (NN) models of the AWS, and a reference based upon the phase and amplitude control strategy. Since the best performance was achieved by different NN models depending on the month of the year, further simulations were carried out showing that switching between diferente controllers, corresponding to diferente models, according to the spectrum of the incoming wave,further increases energy production.