Neuro-fuzzy Classification Model Research Articles

Evolving neuro-fuzzy network for real-time high impedance fault detection and classification

This paper concerns the application of a neuro-fuzzy learning method based on data streams for high impedance fault (HIF) detection in medium-voltage power lines. A wavelet-packet-transform-based feature extraction method combined with a variation of evolving neuro-fuzzy network with fluctuating thresholds is considered for recognition of spatial–temporal patterns in the data. Wavelet families such as Haar, Symlet, Daubechie, Coiflet and Biorthogonal were investigated as a way to provide the most discriminative features for fault detection. The proposed evolving neuro-fuzzy classification model has shown to be particularly suitable for the problem because the HIF environment is subject to concept changes. Different from other statistical and intelligent approaches to the problem, the developed neuro-fuzzy model for HIF classification is not only parametrically, but also structurally adaptive to cope with nonstationarities and novelties. New neurons and connections are incrementally added to the neuro-fuzzy network when necessary for the identification of new patterns, such as faults and usual transients including sag, swell and spikes due to the switching of 3-phase capacitors and energization of transformers. Experimental evaluations compare the proposed classifier with other well-established computational intelligence methods, viz. multilayer perceptron neural network, learning vector quantization neural network and a support vector machine model. Results have shown that the evolving neuro-fuzzy system is effective and robust to changes. The system is able to maintain its detection and classification accuracy even in situations in which other classifiers exhibit a significant drop in accuracy due to gradual and abrupt changes of the fault patterns. Fuzzy rules are useful for interpretability purposes and help to enhance model credibility for decision making.

A Neuro-Fuzzy Rule-Based Classifier Using Important Features and Top Linguistic Features

The efficient feature selection for predictive and accurate classification is highly desirable in many application domains. Most of the attempts to neuro-fuzzy classifier lose information to build interpretable neuro-fuzzy classification model. This paper proposes an interpretable neuro-fuzzy classification model with significant features without loss of knowledge, which is an extension of an existing interpretable neuro-fuzzy classification model. The proposed model is designed based on the consideration of feature importance that is determined by frequency of linguistic features. The rules are then made based on important features. Therefore, the knowledge acquired in network can be comprehended to logical rules using only important features. The proposed model finally performs classification task by rule-based approach. The average accuracy calculated by 10-fold cross validation finds that the proposed model can increase performance of the already proven neuro-fuzzy system for classification tasks.

Separability indexes and accuracy of neuro-fuzzy classification in Geographic Information Systems for assessment of coastal environmental vulnerability

The aim of this study was the development, evaluation and analysis of a neuro-fuzzy classifier for a supervised and hard classification of coastal environmental vulnerability due to marine aquaculture using minimal training sets within a Geographic Information System (GIS). The neuro-fuzzy classification model NEFCLASS‐J, was used to develop learning algorithms to create the structure (rule base) and the parameters (fuzzy sets) of a fuzzy classifier from a set of labeled data. The training sites were manually classified based on four categories of coastal environmental vulnerability through meetings and interviews with experts having field experience and specific knowledge of the environmental problems investigated. The inter-class separability estimations were performed on the training data set to assess the difficulty of the class separation problem under investigation. The two training data sets did not follow the assumptions of multivariate normality. For this reason Bhattacharyy and Jeffries–Matusita distances were used to estimate the probability of correct classification. Further evaluation and analysis of the quality of the classification achieved low values of quantity and allocation disagreement and a good overall accuracy. For each of the four classes the user and producer values for accuracy were between 77% and 100%.In conclusion, the use of a neuro-fuzzy classifier for a supervised and hard classification of coastal environmental vulnerability demonstrated an ability to derive an accurate and reliable classification using a minimal number of training sets.

A neuro-fuzzy evaluation of steel beams patch load behaviour

This work presents a neuro-fuzzy system developed to predict and classify the behaviour of steel beam web panels subjected to concentrated loads. A good performance was obtained with a previously developed neural network system [Fonseca ET, Vellasco MMBR, Vellasco PCGdaS, de Andrade SAL, Pacheco MAC. A neural network system for patch load prediction. J Intell Robot Syst 2001;31(1/3):185–200; Fonseca ET, Vellasco PCGdaS, de Andrade SAL, Vellasco MMBR. A patch load parametric analysis using neural networks. J Constr Steel Res 2003;59(2):251–67; Fonseca ET, Vellasco PCGdaS, de Andrade SAL, Vellasco MMBR. Neural network evaluation of steel beam patch load capacity. Adv Eng Software 2003;34(11–12):763–72] when compared to available experimental data. The neural network accuracy was also significantly better than existing patch load prediction formulae [Lyse I, Godfrey HJ. Investigation of web buckling in steel beams. ASCE Trans 1935;100:675–95, paper 1907; Bergfelt A. Patch loading on slender web. Influence of horizontal and vertical web stiffeners on the load carrying capacity, S79:1. Goteborg: Chalmers University of Technology, Publication; 1979, p. 1–143; Skaloud M, Drdacky M. Ultimate load design of webs of steel plated structures – Part 3 webs under concentrated loads. Staveb Cas 1975;23(C3):140–60; Roberts TM, Newark ACB. Strength of webs subjected to compressive edge loading. J Struct Eng Am Soc Civil Eng 1997;123(2):176–83]. Despite this fact, the system architecture did not explicitly considered the fundamental different structural behaviour related to the beam collapse (web and flange yielding, web buckling and web crippling). Therefore this paper presents a neuro-fuzzy system that takes into account the patch load ultimate limit state. The neuro-fuzzy system architecture is composed of one neuro-fuzzy classification model and one patch load prediction neural network. The neuro-fuzzy model is used to classify the beams according to its pertinence to a specific structural response. Then, a neural network uses the pertinence established by the neuro-fuzzy classification model, to finally determine the beam patch load resistance.

Assessing daily tropical rainfall variations using a neuro-fuzzy classification model

Rainfall variations in tropical areas like Indonesia are dependent upon the tropical climate variability that has two seasons, the dry and wet seasons. However, the significant variations inherent in tropical climate are frequently affected by the combination of the atmosphere phenomena such as the El Niño-southern oscillation (ENSO) influence and tropical cyclone. This in turn lead to uncertainty of rainfall, making it difficult to develop an analysis technique that adequately assesses and interprets variations in rainfall periodicity. Many previous studies of rainfall variation have used techniques such as artificial neural networks and fuzzy methods. Each method uses a different rationale for the way in which the analysis purposes are preserved during analysis. The current study presents the use of a supervised learning of the neuro-fuzzy classification model in order to assess the rainfall variations in a tropical area. This method is a special example of a model within the field of neuro-fuzzy systems that enables the construction of the model output that could be represented by fuzzy classification rules. The classification procedures were started to derive the cluster information for the datasets by using the fuzzy C-means (FCM) clustering. Here, the process of clustering was arranged to provide two clusters of datasets, by adapting to the rainfall of dry (small rainfall) and wet (large rainfall) seasons over the study areas. Based on the prior clusters from the FCM, a neuro-fuzzy algorithm was trained to develop a set of the rule base of the classification models. The pruning strategies for the given rule base in the trained classifier were then exploited to improve the accuracy of the resulting model. The results of analysis gave strong performance by yielding a simpler rule base with a high accuracy. This enabled improved interpretability of variation in rainfall.

Neuro-Fuzzy Approaches for Modeling the Wet Season Tropical Rainfall

Information on rainfall variations is a matter of great importance in agricultural countries. Climate and rainfall are non-linear natural phenomena whose measurement leads to complex data, primarily due to noise patterns and distribution heterogeneity. Therefore, it is difficult to develop an appropriate model in practice by using conventional modeling techniques. This study presents the use of a neuro-fuzzy system for modeling wet season tropical rainfall. The advantage of this technique was the possibility of a modified environment of input parameters for improving the data representation. Two approaches used in the neuro-fuzzy models were classification and prediction. The neuro-fuzzy classification model firstly produced a simple rule base that enables improved interpretability of variation in rainfall rate. The given fuzzy classification rules were then utilized to generate a neuro-fuzzy inference system in order to predict rainfall variation. This approach measured the accuracy of the prediction model according to the root mean square error (RMSE) estimation. The models resulted low values of the RMSE indicated that the prediction models are reliable in representing the recent inter-annual variation of the wet season tropical rainfall.

Hybrid modeling approach to on-line estimation of yeast biomass concentration in industrial bioreactor

The industrial fed-batch yeast cultivation process has been divided into four different metabolic phases (adaptation, carbon limited, oxygen limited and maturation) by a neuro-fuzzy classification model that consists of 4 applied linguistic rules on 2 state variables: oxygen uptake rate and liquid volume. The membership functions have been automatically adapted by this fuzzy perceptron, i.e., by a supervised learning algorithm initialized by prior operator's knowledge. Process compartmentalization has made easier and more realistic a subsequent state estimation of the biomass concentration with separate artificial neural networks combined with balance equations. Static networks with local recurrent memory structures were used, and the inputs were standard cultivation state variables: respiratory quotient, molasses feed rate, ethanol concentration, etc. This hybrid approach is generally applicable to state estimation or prediction when different sources of process information and knowledge have to be integrated.