Fuzzy Coding of Genetic Algorithms

A well-cited paper suggesting fuzzy coding as an alternative to the conventional binary, grey and floating-point representations used in genetic algorithms.

Detection of Anaerobic Bacteria in High Numbers in Sputum from Patients with Cystic Fibrosis

Rationale: Pulmonary infection in cystic ?brosis (CF) is polymicrobial and it is possible that anaerobic bacteria, not detected by routine aerobic culture methods, reside within infected anaerobic airway
mucus.
Objectives: To determine whether anaerobic bacteria are present in the sputum of patients with CF.
Methods: Sputum samples were collected from clinically stable adults with CF and bronchoalveolar lavage ?uid (BALF) samples from children with CF. Induced sputum samples were collected from healthy volunteers who did not have CF. All samples were processed using anaerobic bacteriologic techniques and bacteria within the samples were quanti?ed and identi?ed.
Measurements and Main Results: Anaerobic species primarily within the genera Prevotella,Veillonella, Propionibacterium, andActinomyces were isolated in high numbers from 42 of 66 (64%) sputum samples from adult patients with CF. Colonization with Pseudomonas aeruginosa signi?cantly increased the likelihood that anaerobic bacteria would be present in the sputum. Similar anaerobic species were identi?ed in BALF from pediatric patients with CF. Although anaerobes were detected in induced sputum samples from 16 of 20 volunteers, they were present in much lower numbers and were
generally different species compared with those detected in CF sputum. Species-dependent differences in the susceptibility of the anaerobes to antibiotics with known activity against anaerobes were apparent with all isolates susceptible to meropenem.
Conclusions: A range of anaerobic species are present in large numbers in the lungs of patients with CF. If these anaerobic bacteria are contributing signi?cantly to infection and in?ammation in the CF
lung, informed alterations to antibiotic treatment to target anaerobes, in addition to the primary infecting pathogens, may improve management.

A recursive rule base adjustment algorithm for a fuzzy logic controller

This paper introduces a recursive rule base adjustment to enhance the performance of fuzzy logic controllers. Here the fuzzy controller is constructed on the basis of a decision table (DT), relying on membership functions and fuzzy rules that incorporate heuristic knowledge and operator experience. If the controller performance is not satisfactory, it has previously been suggested that the rule base be altered by combined tuning of membership functions and controller scaling factors. The alternative approach proposed here entails alteration of the fuzzy rule base. The recursive rule base adjustment algorithm proposed in this paper has the benefit that it is computationally more efficient for the generation of a DT, and advantage for online realization. Simulation results are presented to support this thesis. (c) 2005 Elsevier B.V. All rights reserved.

Robust fuzzy Gustafson-Kessel clustering for nonlinear system identification

This paper deals with Takagi-Sugeno (TS) fuzzy model identification of nonlinear systems using fuzzy clustering. In particular, an extended fuzzy Gustafson-Kessel (EGK) clustering algorithm, using robust competitive agglomeration (RCA), is developed for automatically constructing a TS fuzzy model from system input-output data. The EGK algorithm can automatically determine the 'optimal' number of clusters from the training data set. It is shown that the EGK approach is relatively insensitive to initialization and is less susceptible to local minima, a benefit derived from its agglomerate property. This issue is often overlooked in the current literature on nonlinear identification using conventional fuzzy clustering. Furthermore, the robust statistical concepts underlying the EGK algorithm help to alleviate the difficulty of cluster identification in the construction of a TS fuzzy model from noisy training data. A new hybrid identification strategy is then formulated, which combines the EGK algorithm with a locally weighted, least-squares method for the estimation of local sub-model parameters. The efficacy of this new approach is demonstrated through function approximation examples and also by application to the identification of an automatic voltage regulation (AVR) loop for a simulated 3 kVA laboratory micro-machine system.