Adaptive Neuron-fuzzy Inference System Research Articles

Adaptive Equalization for UWB communication System based on ANFIS

Ultra-wideband (UWB) communication systems cover enormous bandwidths that have strongly low-power spectral densities. At UWB communication system with high data rate, owing to multipath propagation, the spread delay in inter symbol interference (ISI) will raise the bit error rate (BER) considerably. ISI which is formed via the UWB channels can be removed by equalization, which is one of the most significant signal processing techniques. Furthermore, LMS algorithm represents a very efficient tool for determining adaptive equalizer coefficients values in communication systems, in spite of that, the LMS adaptive equalizer encounters response diminishing besides slow convergence rate. The current paper adopts an adaptive equalizer based adaptive neuron-fuzzy inference system (ANFIS). The simulation outcomes reveal that the convergence rates as well as accuracy of identification of ANFIS based algorithm are surpass the traditional LMS algorithm, moreover, simulation outcomes prove that ISI is effectively limited and the performance of the system is clearly improved.

Solar energy density estimation using ANFIS based on daily maximum and minimum temperature

The data of solar energy density in one area is very important when the area will constructed photovoltaic (PV) system. The data is as preliminary study to decide what the area is suitable or not to be constructed the PV application system. But, sometime the available data is missing because the limitation of weather equipment. An alternative technique for the available data of solar energy density should be done for the continuity of PV application system decision. An estimation technique of solar energy density is one part of good alternative to solve this problem. This paper presents the estimation of solar energy density using Adaptive Neuron Fuzzy Inference System (ANFIS). The ANFIS system has two input data of the measured daily minimum, maximum temperature and difference between maximum and minimum temperature. The measured solar energy density is as target data of ANFIS system. The data is recorded from Medan meteorological station through the web site of world weather online for the year of 2018. The result shows that the average estimated solar energy density is classified in the very high solar energy density and based on the percentage error shows that the estimated solar energy density is acceptable.

Multiperiod-Ahead Wind Speed Forecasting Using Deep Neural Architecture and Ensemble Learning

Accurate forecasting of wind speed plays a fundamental role in enabling reliable operation and planning for large-scale integration of wind turbines. It is difficult to obtain the accurate wind speed forecasting (WSF) due to the intermittent and random nature of wind energy. In this paper, a multiperiod-ahead WSF model based on the analysis of variance, stacked denoising autoencoder (SDAE), and ensemble learning is proposed. The analysis of variance classifies the training samples into different categories. The stacked denoising autoencoder as a deep learning architecture is later built for unsupervised feature learning in each category. The ensemble of extreme learning machine (ELM) is applied to fine-tune the SDAE for multiperiod-ahead wind speed forecasting. Experimental results are made to demonstrate that the proposed model has the best performance compared with the classic WSF methods including the single SDAE-ELM, ELMAN, and adaptive neuron-fuzzy inference system (ANFIS).

Experimental and artificial neural network investigation on the effect of inclination angle on the interface temperature of CPU/metal foam heat sink

Purpose This study aims to clarify the relationship between inclination angle of hot surface of CPU and its temperature in absence and presence of aluminum foam as a cooling system. It proposes application of the artificial neural [multi-layer perceptron (MLP) and radial basis function] networks and adaptive neuron-fuzzy inference system (ANFIS) to predict interface temperature of central processing unit (CPU)/metal foam heat sink. Design/methodology/approach To provide a consistent set of data, the surface of an aluminum cone with and without installing Duocel aluminum foam was heated in a natural convection using an electrical resistor. The hot surface temperature was measured using five K-type thermocouples (±0.1°C). To develop the predictive models, ambient temperature, input power and inclination angle are taken as input which varied from 23°C to 32°C, 4 to 20 W and 0° to 90°, respectively. The hot surface temperature is taken as the output. Findings The results show that in the presence of foam, the hot surface temperature was less sensitive to the variations of angle, and the maximum enhancement of the heat transfer coefficient was 23 per cent at the vertical position. Both MLP network and ANFIS are comparable, but the values predicted by MLP network are in more conformity with the measured values. Originality/value The effect of metal foam on the inclination angle/hot surface temperature dependence is identified. The optimum angle is clarified. The applicability of the MLP networks to predict interface temperature of CPU/heat sink is approved.

Battery Aging Prediction Using Input-Time-Delayed Based on an Adaptive Neuro-Fuzzy Inference System and a Group Method of Data Handling Techniques

In this article, two techniques that are congruous with the principle of control theory are utilized to estimate the state of health (SOH) of real-life plug-in hybrid electric vehicles (PHEVs) accurately, which is of vital importance to battery management systems. The relation between the battery terminal voltage curve properties and the battery state of health is modelled via an adaptive neuron-fuzzy inference system and a group method of data handling. The comparison of the results demonstrates the capability of the proposed techniques for accurate SOH estimation. Moreover, the estimated results are compared with the direct actual measured SOH indicators using standard tests. The results indicate that the adaptive neuron-fuzzy inference system with fifteen rules based on a SOH estimator has better performances over the other technique, with a 1.5% maximum error in comparison to the experimental data.

Further model development for prediction of viscosity of mixed oils

ABSTRACTIn the present study a new model named adaptive neuron fuzzy inference system optimized by hybrid method (Hybrid-ANFIS) is developed for estimation of viscosity of mixed oils. The experimental data for development of the new model were gathered from literature. Various methods were taken into consideration to examine the accuracy and precision of the model. The outcomes of the developed model were also put into comparison with four well-known literature models. Results show that the Hybrid-ANFIS model provides acceptable and precise estimations and more satisfactory predictions in comparison with literature models. The model exhibits an overall AARD% value of 2.19%.

A Novel Online Data-Driven Algorithm for Detecting UAV Navigation Sensor Faults.

The use of Unmanned Aerial Vehicles (UAVs) has increased significantly in recent years. On-board integrated navigation sensors are a key component of UAVs’ flight control systems and are essential for flight safety. In order to ensure flight safety, timely and effective navigation sensor fault detection capability is required. In this paper, a novel data-driven Adaptive Neuron Fuzzy Inference System (ANFIS)-based approach is presented for the detection of on-board navigation sensor faults in UAVs. Contrary to the classic UAV sensor fault detection algorithms, based on predefined or modelled faults, the proposed algorithm combines an online data training mechanism with the ANFIS-based decision system. The main advantages of this algorithm are that it allows real-time model-free residual analysis from Kalman Filter (KF) estimates and the ANFIS to build a reliable fault detection system. In addition, it allows fast and accurate detection of faults, which makes it suitable for real-time applications. Experimental results have demonstrated the effectiveness of the proposed fault detection method in terms of accuracy and misdetection rate.

Fault detection using thermal image based on soft computing methods: Comparative study

This paper presents Integrated Circuit (IC) fault detection of a Printed Circuit Board (PCB) model using thermal image processing. The thermal image is captured and processed from the PCB model by the finite element method (FEM). The histogram features are extracted from the ICs hotspots which are used as inputs in a classifier model. The effective features are minimized by the principal component analysis method. In this work, a comparative study for image classification and detection is performed based on three soft computing techniques: multilayer perceptron, support vector machine, and adaptive neuron-fuzzy inference system. The effectiveness of the models is evaluated by comparing the performance and accuracy of the classification. To validate the model, the experimental evaluation is performed on Arduino UNO in order to detect the fault condition on the real time operating PCB.

Intelligent testing for Arduino UNO based on thermal image

The goal of this paper is to develop a tool that will aid manufacturers of Printed Circuit Boards (PCBs) in testing their production lines with an acceptable rate of fault detection and minimize the test time. The PCB unit, namely the Arduino UNO board, is used to generate the thermal profile for the unit under test. This work is based on using a classification approach that classifies the PCB defects into the Integrated Circuit (IC) level. In the proposed technique, histogram features are extracted from the ICs hotspots which are used as inputs into a classifier model. The number of effective features are minimized by the principal component analysis. The image classification and detection are performed based on three soft computing techniques; multilayer perceptron, support vector machine, and adaptive neuron-fuzzy inference system. The effectiveness of the models is evaluated by comparing their performance and accuracy of classification.

A new congruency-based hysteresis modeling and compensating of a piezoactuator incorporating an adaptive neuron fuzzy inference system

This paper proposes a new approach to modeling and compensating for a rate-independent hysteresis of a piezoactuator. The model—namely, congruency-based hysteresis—is developed based on two very important characteristics of the hysteresis. These are congruency and wipe-out. The proposed approach consists of two branches for cases of monotonic increase and monotonic decrease of input excitation. In order to realize this model, datasets of first-order minor-loop values should be determined in advance. This can be done using the adaptive neuron fuzzy system (ANFIS) technique and experimental data. With this technique, an input-output relationship of first-order minor-loop values is estimated effectively. In addition, the ANFIS technique is also used in constructing datasets of inverse first-order minor-loop values, which are essential parts of a congruency-based hysteresis compensator. Several experiments in modeling and open-loop control are conducted to show the effectiveness of the proposed approach. In addition, a comparative work between the proposed approach and one of previous works is undertaken to demonstrate the benefit of the proposed method.

Damping of Power System Oscillations by using ANFIS based UPFC supplementary controller

The main purpose of this paper is to investigate an Adaptive Neuron- Fuzzy Inference System (ANFIS) based supplementary Unified Power Flow Controller (UPFC) in superimposing a damping function on the control signal of UPFC who’s incorporated in multi-machines power system. By using a hybrid learning procedure, the proposed ANFIS construct an input –output mapping based on stipulated input-output data pairs.This control scheme designed to produce supplementary signal for damping oscillations in the interconnected power systems. Oscillations in power systems have to be taken a serious note when the fault takes place in any part of the system; else this might lead to the instability mode and shutting down of the power system. The UPFC device is modelled using dynamic phasor method and interfaced with an interconnected power system. The improved damping performance was studied and it is validated on a benchmark power system example taken from literature. The systems model is evaluated via simulations during unbalanced faults. The control actions are formulated as a set of local controllers, namely the generator excitation controller, power system stabilizer (PSS) and unified power flow controller, which are coordinated to maintain transient angle stability, transient voltage stability and damping of inter-area and local oscillations following a disturbance.

Parameter optimization for intelligent phishing detection using Adaptive Neuro-Fuzzy

Phishing attacks has been growing rapidly in the past few years. As a result, a number of approaches have been proposed to address the problem. Despite various approaches proposed such as feature-based and blacklist-based via machine learning techniques, there is still a lack of accuracy and real-time solution. Most approaches applying machine learning techniques requires that parameters are tuned to solve a problem, but parameters are difficult to tune to a desirable output. This study presents a parameter tuning framework, using adaptive Neuron-fuzzy inference system with comprehensive data to maximize systems performance. Extensive experiment was conducted. During ten-fold cross-validation, the data is split into training and testing pairs and parameters are set according to desirable output and have achieved 98.74% accuracy. Our results demonstrated higher performance compared to other results in the field. This paper contributes new comprehensive data, novel parameter tuning method and applied a new algorithm in a new field. The implication is that adaptive neuron-fuzzy system with effective data and proper parameter tuning can enhance system performance. The outcome will provide a new knowledge in the field.

INCOMING AND OUTGOING CT READINGS BASED AI TECHNIQUES FOR BUS BAR DIFFERENTIAL PROTECTION FAULTS ANALYSIS

The application of artificial intelligent approaches was introduced recently in protection of distribution networks. In this paper, the application of Adaptive Neuron Fuzzy Inference System (ANFIS) as well as artificial Neural Network (ANN) for protection of bus bars will be illustrated. The objective of this paper is firstly, to detect the fault occurrence on bus bar, secondly location of the fault. The fault detection and location are firstly trained using ANN and ANFIS techniques and then they are tested in diversity of system conditions with respect to the fault types, inception instant, resistance and pre fault conditions. The fault detection and location units used current magnitude difference change at fault and Pre-fault conditions for incoming and outgoing CTs. The fault location unit in this paper is to detect if the fault occurs internally on the bus bar or externally. The protection system selection depends on the time as the system should take the correct action at fault condition, sending correct tripping signal for internal fault conditions and prevent the mal-operation due to external fault conditions. On other hand, the fault clearing time is very important issue to avoid more damage to power system equipments. From the simulation, it is found that the proposed system success to detect the fault and clarify the fault location at different fault cases within a short detection time than obtained in the previous work.

Research on Condition Monitoring of Bearing Health Using Vibration Data

A reliable condition monitoring system is very useful in a wide range of industries to detect the occurrence of incipient defects so as to prevent machinery performance degradation, malfunction and sudden failure. Among the rotating machinery, many mechanical problems are attributed due to bearing failures. So implementing condition monitoring for bearing is critically needed. Considering that most research for condition monitoring only focus on detecting the existing fault, this paper add degradation tendency prognostics into the condition monitoring process. The kernel of bearing condition monitoring method presented in this paper is related to condition features extraction and remaining useful life prediction. The former is realized by the comprehensive vibration analysis for specific fault frequencies. The latter is achieved by adaptive neuron-fuzzy inference system based on extracted degradation signal. For illustration purpose, a bearing case from NASA data repository is used to validate the feasibility of the proposed method. The result indicates that the performance degradation of bearing can be effectively monitored and the predicted remaining useful life with 5.6% relative error can be the important reference for maintenance decision making.

Identification and Equalization of Nonlinear Channel Based on ANFIS

The demand of high-speed communication leads that the application for resource of channel exceeds the range of linear model. The channel should be described as a nonlinear model. So, the paper uses the adaptive neuron-fuzzy inference system (ANFIS) to identify and equalize the nonlinear channel of the high-speed communication system. Meanwhile, the subtract cluster is applied to identify the construction of the ANFIS and the hybrid learning algorithm based on the least square and back-propagation is used to train network. The simulation results show that the convergence rate and identification accuracy of the ANFIS are better than BP network and the efficiency of the ANFIS based on subtract cluster partition algorithm is higher than that of the ANFIS based on the grid partition algorithm.

Adaptive neuron-fuzzy inference system combined with principal components analysis for determination of compound thiamphenicol powder on near-infrared spectroscopy

The adaptive neuron-fuzzy inference system (ANFIS) is an effective modeling tool developed recently. It has gained much interest in solving classification and function approximation. In this paper, a new application based on ANFIS was presented for nondestructive determination of thiamphenicol powder drug with near-infrared (NIR) spectroscopy. The principal component analysis (PCA) technique was applied to extract relevant features from a number of spectral data in order to reduce the input variables of the ANFIS. The generated scores of the principal components (PCs) subsequently were used as the input variables of the ANFIS instead of the spectra data and constituted the principal component analysis-adaptive neuron-fuzzy inference system (PCA-ANFIS) model. A hybrid-learning algorithm which combined the least squares method and the gradient descent method was applied to optimize the parameters of PCA-ANFIS. Various optimum PCA-ANFIS models based on the conventional spectra and pretreated spectra (standard normal variate (SNV), multiplicative scatter correction (MSC) and the first-derivative) were established and compared. Experiment results indicated that the PCA-ANFIS model obtained from data sets achieved satisfactory accuracy, and the PCA-ANFIS approach with MSC pretreated spectra was found that it provided the best results. In order to present the advantages of PCA-ANFIS, the principal component regression (PCR) was also used, which was compared with PCA-ANFIS. Experiment results showed that the proposed PCA-ANFIS was more efficient than PCR.

Fuzzy Controller Design of TCSC with ANFIS to Improve the Dynamic Stability of Power System

In this paper, a new variable fuzzy control algorithm is proposed for thyristor-controlled series capacitor (TCSC) to improve power system dynamic performance. Rules of adaptive neuron fuzzy inference system (ANFIS) devise this fuzzy controller for TCSC in power system. ANFIS uses a hybrid learning algorithm to identify the membership function parameters of Sugeno-type fuzzy inference system. Consequences of this control trend are compared with PI controller and it is shown by using fuzzy controller the system has more stable power flow and less power oscillations. Simulation results signify power stability improvement of two-area power system by using fuzzy controller in TCSC than traditional PI controller.

Adaptive neuro-fuzzy inference approach for prediction the stiffness modulus on asphalt concrete

In this study, stiffness modulus parameters of asphalt concrete were determined experimentally for different temperature and exposure times. The stiffness modules were calculated according to Nijboer stiffness module. Basic physical properties and the quantity of bitumen of asphalt core samples were designated for determining the stiffness modules. The samples were exposed to 17^oC (reference temperature), 30, 40 and 50^oC temperatures for 1.5, 3, 4.5 and 6h respectively and then Marhall Stability tests were done for each samples. By using the test results a prediction model with Sugeno type based on the adaptive neuron-fuzzy inference system (ANFIS) was alternatively developed to predict the stiffness modules of asphalt core samples. As a result, it was seen that the developed prediction model could be used as a prediction model for unperformed situations which are not suitable for experiments.

Timing of resources exploration in the behavior of firm—Empirical simulation by intelligent hybrid model

We have insight into the importance of resource exploration derived from the quest for sustaining competitive advantage as well as the growth of the firm, which are well-explicated in the resources point of view. However, we really do not know when the firm will seriously commit to this kind of activities. Therefore, this study proposes an intelligent hybrid model using quantum minimization (QM) to tune a composite model adaptive support vector regression (ASVR) and nonlinear generalized autoregressive conditional heteroscedasticity (NGARCH) such that it constitutes the relationship among five indicators, the growth rate of long-term investment, the firm size, the return on total asset, the return on common equity, and the return on sales. In particular, this proposed approach outperforms several typical methods such as auto-regressive moving-average regression (ARMAX), back-propagation neural network (BPNN), or adaptive neuron-fuzzy inference system (ANFIS) for this timing problem in term of comparing their achievement and the goodness of fit. Consequently, the preceding methods involved in this problem truly explain the timing of resources exploration in the behavior of firm. Meanwhile, the performance summary among methods is compared quantitatively.

Can quetiapine induce delirium in bipolar disorder?

The goal of this paper is to develop a tool that will aid manufacturers of Printed Circuit Boards (PCBs) in testing their production lines with an acceptable rate of fault detection and minimize the test time. The PCB unit, namely the Arduino UNO board, is used to generate the thermal profile for the unit under test. This work is based on using a classification approach that classifies the PCB defects into the Integrated Circuit (IC) level. In the proposed technique, histogram features are extracted from the ICs hotspots which are used as inputs into a classifier model. The number of effective features are minimized by the principal component analysis. The image classification and detection are performed based on three soft computing techniques; multilayer perceptron, support vector machine, and adaptive neuron-fuzzy inference system. The effectiveness of the models is evaluated by comparing their performance and accuracy of classification.