Adaptive Network Based Fuzzy Inference System Network Research Articles

Abnormal discharge detection using adaptive neuro-fuzzy inference method with probability density-based feature and modified subtractive clustering

Abnormal discharge (AD) is a discharge mode in electroencephalogram (EEG) with sharp outlines. Lack of related open-access datasets and insufficient annotation hamper the development of AD detection, while cognitive neuroscience, clinical research and pilots’ neural screening demand stable and reliable AD detection methods. An adaptive neuro-fuzzy inference method for AD detection is proposed in this work. First, a probability-density-based (PD) method is proposed to extract lognormal amplitude features from EEG envelopes. Second, the subtractive clustering method (SCM) is modified so that clustering radii can adapt to cluster shapes for each dimension instead of using identical radii for each cluster. The outputs of modified SCM (mSCM), coordinates of cluster centers and adjusting rates for radius components are used to automatically initialize Gaussian membership functions of adaptive-network-based fuzzy inference system (ANFIS). Finally, we conducted multiple experiments to validate mSCM-based ANFIS, comparing it with traditional machine learning classifiers (support vector machines, multi-layer perceptrons, decision trees, and random forests) and the state-of-the-art deep learning-based time-series classification methods InceptionTime and Minirocket, using synthetic data, small-size datasets and a private EEG dataset. Results show that combining PD features with features proposed in previous studies, such as smoothed nonlinear energy operator features and discrete wavelet transform features, achieved higher accuracy (95.13 ± 0.86%) and recall (89.17 ± 3.06%) than other feature combinations. mSCM created more suitable cluster boundaries than SCM on small-size datasets, and its clustering results demonstrated potential in helping interpretation of classification rules built in the ANFIS network. Results of comparison experiments showed that mSCM-based ANFIS produced competitive results in accuracy and recall for AD detection, with lower computational cost, compared to the top 2 results obtained by InceptionTime and Minrocket.

Fault Location Estimation on Transmission Lines using Neuro-Fuzzy System

Accurate fault location on transmission line is important in ensuring consistent and reliable operation of the power deliver to long distance destination. Conventional methods for locating fault on transmission lines based on travelling wave and impendence-based methods usually suffer from large error due to the complexity of fault modeling on different type of faults. In this paper, an intelligent system for detection of fault location on transmission line using a hybrid model that integrates artificial neural network (ANN) and fuzzy expert system called Adaptive Network-Based Fuzzy Inference System (ANFIS) is proposed. First, a three phase transmission lines is modeled and various types of faults are generated using MATLAB/Simulink. Then, the faulted current signal is segmented from the faulted transmission. Next, feature extraction is performed to obtained information from the faulted current signal. In this study, the extracted features are mean, standard deviation, energy, peak-to-peak and amplitude value. Feature selection is then applied to select important features that correlate with the fault location. For single-phase-to-ground fault, peak-to-peak value and energy is used. Meanwhile, for the line-to-line and double-phase-to-ground faults, only peak-to-peak value is used. Finally, ANFIS network is trained to locate the fault occurrence. Simulation results against two regression models; Linear Regression and Gaussian Process Regression indicated that the ANFIS network is superior in locating the fault. The network achieved the lowest mean squared error (MSE) (0.0012 to 0.0022).

Combination of a gamma radiation-based system and the adaptive network-based fuzzy inference system (ANFIS) for calculating the volume fraction in stratified regime of a three-phase flow

Understanding the volume fraction of water-oil-gas three-phase flow is of significant importance in oil and gas industry. The current research attempts to indicate the ability of adaptive network-based fuzzy inference system (ANFIS) to forecast the volume fractions in a water-oil-gas three-phase flow system. The current investigation devotes to measure the volume fractions in the stratified three-phase flow, on the basis of a dual-energy metering system consisting of the 152Eu and 137Cs and one NaI detector using ANFIS. The summation of volume fractions is equal to 100% and is also a constant, and this is enough for the ANFIS just to forecast two volume fractions. In the paper, three ANFIS models are employed. The first network is applied to forecast the oil and water volume fractions. The next to forecast the water and gas volume fractions, and the last to forecast the gas and oil volume fractions. For the next step, ANFIS networks are trained based on numerical simulation data from MCNP-X code. The accuracy of the nets is evaluated through the calculation of average testing error. The average errors are then compared. The model in which predictions has the most consistency with the numerical simulation results is selected as the most accurate predictor model. Based on the results, the best ANFIS net forecasts the water and gas volume fractions with the mean error of less than 0.8%. The proposed methodology indicates that ANFIS can precisely forecast the volume fractions in a water-oil-gas three-phase flow system.

Short-term wind speed forecasting by an adaptive network-based fuzzy inference system (ANFIS): an attempt towards an ensemble forecasting method

Accurate Wind speed forecasting has a vital role in efficient utilization of wind farms. Wind forecasting could be performed for long or short time horizons. Given the volatile nature of wind and its dependent on many geographical parameters, it is difficult for traditional methods to provide a reliable forecast of wind speed time series. In this study, an attempt is made to establish an efficient adaptive network-based fuzzy interference (ANFIS) for short-term wind speed forecasting. Using the available data sets in the literature, the ANFIS network is constructed, tested and the results are compared with that of a regular neural network, which has been forecasted the same set of dataset in previous studies. To avoid trial-and-error process for selection of the ANFIS input data, the results of autocorrelation factor (ACF) and partial auto correlation factor (PACF) on the historical wind speed data are employed. The available data set is divided into two parts. 50% for training and 50% for testing and validation. The testing part of data set will be merely used for assessing the performance of the neural network which guarantees that only unseen data is used to evaluate the forecasting performance of the network. On the other hand, validation data could be used for parameter-setting of the network if required. The results indicate that ANFIS could not outperform ANN in short-term wind speed forecasting though its results are competitive. The two methods are hybridized, though simply by weightage, and the hybrid methods shows slight improvement comparing to both ANN and ANFIS results. Therefore, the goal of future studies could be implementing ANFIS and ANNs in a more comprehensive ensemble method which could be ultimately more robust and accurate

Fault location in underground cables using ANFIS nets and discrete wavelet transform

This paper presents an accurate algorithm for locating faults in a medium voltage underground power cable using a combination of Adaptive Network-Based Fuzzy Inference System (ANFIS) and discrete wavelet transform (DWT). The proposed method uses five ANFIS networks and consists of 2 stages, including fault type classification and exact fault location. In the first part, an ANFIS is used to determine the fault type, applying four inputs, i.e., the maximum detailed energy of three phase and zero sequence currents. Other four ANFIS networks are utilized to pinpoint the faults (one for each fault type). Four inputs, i.e., the maximum detailed energy of three phase and zero sequence currents, are used to train the neuro-fuzzy inference systems in order to accurately locate the faults on the cable. The proposed method is evaluated under different fault conditions such as different fault locations, different fault inception angles and different fault resistances.

Adaptive neuro-fuzzy approach for reservoir oil bubble point pressure estimation

A new method based on adaptive network-based fuzzy inference system (ANFIS) approach was designed and developed for improved estimation of reservoir oil bubble point pressure using commonly available field data. More than 750 data series from different geographical locations worldwide was gathered for modeling. Two different ANFIS networks (by changing the training optimization algorithms) were compared with evaluation of networks accuracy in bubble point pressure prediction and subsequently the suitable network was determined. The predictions of selected network are in good agreement with the corresponding experimental data with the squared correlation coefficient of 0.97. In addition, a comparative study was carried out between the developed model and other published correlations. In comparison with the published literature correlations, the results showed that proposed ANFIS can be used as a powerful model for improved prediction of reservoir oil bubble point pressure.

Financial forecasting using ANFIS networks with Quantum-behaved Particle Swarm Optimization

To be successful in financial market trading it is necessary to correctly predict future market trends. Most professional traders use technical analysis to forecast future market prices. In this paper, we present a new hybrid intelligent method to forecast financial time series, especially for the Foreign Exchange Market (FX). To emulate the way real traders make predictions, this method uses both historical market data and chart patterns to forecast market trends. First, wavelet full decomposition of time series analysis was used as an Adaptive Network-based Fuzzy Inference System (ANFIS) input data for forecasting future market prices. Also, Quantum-behaved Particle Swarm Optimization (QPSO) for tuning the ANFIS membership functions has been used. The second part of this paper proposes a novel hybrid Dynamic Time Warping (DTW)-Wavelet Transform (WT) method for automatic pattern extraction. The results indicate that the presented hybrid method is a very useful and effective one for financial price forecasting and financial pattern extraction.

Fault Location in Underground Cables using ANFIS Nets and Discrete Wavelet Transform

This paper presents an accurate algorithm for locating faults in a medium voltage underground power cable using Adaptive Network-Based Fuzzy Inference System (ANFIS). The proposed method uses five ANFIS networks and consists of 2 stages, including fault type classification and exact fault location. In the first part, an ANFIS is used to determine the fault type, applying four inputs, i.e., the maximum detailed energy of three phase and zero sequence currents. Other four ANFIS networks are utilized to pinpoint the faults (one for each fault type). Four inputs, i.e., the maximum detailed energy of three phase and zero sequence currents, are used to train the Neuro-fuzzy inference systems in order to accurately locate the faults on the cable. The proposed method is evaluated under different fault conditions such as different fault locations, different fault inception angles and different fault resistances.

A new and accurate fault location algorithm for combined transmission lines using Adaptive Network-Based Fuzzy Inference System

This paper presents a new and accurate algorithm for locating faults in a combined overhead transmission line with underground power cable using Adaptive Network-Based Fuzzy Inference System (ANFIS). The proposed method uses 10 ANFIS networks and consists of 3 stages, including fault type classification, faulty section detection and exact fault location. In the first part, an ANFIS is used to determine the fault type, applying four inputs, i.e., fundamental component of three phase currents and zero sequence current. Another ANFIS network is used to detect the faulty section, whether the fault is on the overhead line or on the underground cable. Other eight ANFIS networks are utilized to pinpoint the faults (two for each fault type). Four inputs, i.e., the dc component of the current, fundamental frequency of the voltage and current and the angle between them, are used to train the neuro-fuzzy inference systems in order to accurately locate the faults on each part of the combined line. The proposed method is evaluated under different fault conditions such as different fault locations, different fault inception angles and different fault resistances. Simulation results confirm that the proposed method can be used as an efficient means for accurate fault location on the combined transmission lines.