Adaptive Nero-Fuzzy Inference System Research Articles

Machine learning models for prediction the Higher Heating Value (HHV) of Municipal Solid Waste (MSW) for waste-to-energy evaluation

Time-resolved knowledge of physico-chemical properties of Municipal Solid Waste (MSW) materials and their thermal energy content is one of the important subjects needed to build waste incineration power plants around the world. For this purpose, machine learning models were used to predict the Higher Heating Value (HHV) of MSW based on the initial materials. Four types of machine learning methods: Radial Bias Function Artificial Neural Network (RBF-ANN), Multilayer Perceptron Artificial Neural Network (MLP-ANN), Support Vector Machine (SVM) and Adaptive Nero-Fuzzy Inference System (ANFIS) were used for modeling the HHV with six different inputs (carbon, water, hydrogen, oxygen, nitrogen, sulfur, and ash). The results showed that RBF-ANN can predict the HHV of MSW with higher accuracy than other models. The overall Mean Absolute Percentage Error (MAPE) for MLP-ANN, SVM and ANFIS models were 7.3, 11.77 and 23.76%, respectively. The MAPE of the best topology for RBF model (6-17-1) with spread factor of 0.8 reach to 0.45%. Finally, the results of this study proved that ANN's can be used as a practical tool with high accuracy and reliability for design and management of waste incineration plants.

ANFIS Based Parameter Estimating of a Two-Phase Interleaved Dual Cascaded DC-DC Boost Converter for Fuel Cell Supplied Electric Vehicles

Fuel cells (FCs), one of the renewable energy sources, have started to be preferred as a power source in electric vehicles in recent years and research studies are continuing on designs in this direction. Although their efficiency is low <50%, they produce Direct Current (DC) electrical energy by electrochemical conversion without requiring battery systems, which can be used in electric vehicle drive systems. There are zero-emission effects such as water and temperature rise with waste and environmental aspects. One of the major disadvantages is the DC voltage amplitude they produce is inversely proportional to the temperature increase. In this context, parameter estimation is required to adapt the fluctuating FC voltage to a certain value adaptively with the DC-DC boost converter circuit. In this study, parametric simulation studies were carried out with Ansys-Electronics 2019-R3 software to determine the DC voltage level of a certain number of series and parallel connected FC cells depending on different temperature values. Duty ratio values of two-phase interleaved dual cascaded DC-DC boost converter circuit for desired output voltage were determined by using Adaptive Nero-Fuzzy Inference System (ANFIS) modeling of 1300 data determined by simulation studies. Thus, the output voltage of the converter is adaptively fixed at a certain value.

Neuro-fuzzy system based model for prediction of project performance in downstream sector of petroleum industry in Iran

Purpose Planning phase of a project results to series of crucial decisions which determine the path to objectives achievement. At the same time, in this phase, project encounters the highest level of uncertainty in comparison of all phases of project lifecycle. This paper aims to support early decisions of project based on the progress forecasting. Design/methodology/approach The scope of study is limited to downstream projects of petroleum industry in Iran, and the proposed model is trained and tested based on 75 Iranian completed petroleum projects. First, types of progress curve functions are investigated, and various types are studied and the most appropriate ones are selected through curve fitting. In the next step, using questionnaire, dependent and independent variables are recognized. Finally, using historical data and s-curve generator functions, a fuzzy inference system (ANFIS) based model have been developed to support early phases decision-making processes. Findings Based on the analysis of received questionnaires, six functional criteria in two groups as dependent variables and 25 independent variables, in two groups and four clusters are determined and categorized. Eventually, performance prediction model of a project has been developed by using Adaptive Nero Fuzzy Inference System. Originality/value The main contribution of this study to construction management knowledge is categorizing two groups of variables, which first one defines the project dynamic and the other calculates the key effects on previous one. Also, this investigation improves the current knowledge by analyzing the project system from the dynamic behavior perspective and modeling the defined variables using ANFIS tools.

Developing Convective Mass Transfer of Nanofluids in Fully Developed Flow Regimes in a Circular Tube: Modeling using Fuzzy Inference System and ANFIS

In this study, actual nanofluids Sherwood numbers were compared by the Look-up Table-based Fuzzy Inference System (LUT-based FIS) and the Fuzzy C-means Adaptive Nero Fuzzy Inference System (FCM-ANFIS) predicted data in a circular tube under both laminar and turbulent flow regimes. Three influential parameters including material types, Reynolds numbers, and nanoparticles volume concentrations were used as the input attributes. Intelligent models had also one output presented by Sherwood numbers. The results revealed that nanoparticles size and type could play an important role in determining the mass transfer in nanofluids and the optimum nanoparticle concentration. The maximum increase in mass transfer for all nanofluids occurred in laminar flow, 36% in a 0.03% TiO2, 21% in a 0.0057% SiO2, and 16.8% in 0.01% Al2O3 nanofluids. Comparison of the experimental data with the predicted ones by FIS and ANFIS models showed that these models could predict convective mass transfer in nanofluids, while ANFIS models with only 5 rules and the shorter running time (19 s) had a simpler structure and precisely imitated the mass transfer. The best RMSE and R2 for the test data under laminar and turbulent flow were 0.044, 0.98, and 0.022, 0.99, respectively.

Risk Management Prediction for Overcrowding in Railway Stations Utilising Adaptive Nero Fuzzy Inference System (ANFIS)

In this research, an intelligent system for managing risks is developed with a framework to aid in managing the risks in the railway stations. A method to advance risk management in the railway stations is needed in order to minimize risk through an automated process taking into consideration all the factors in the system and how they work together to provide an acceptable level of safety and security. Thus, the Adaptive Nero Fuzzy Inference System (ANFIS) is proposed to improve risk management as an intelligently selected model which is powerful in dealing with uncertainties in risk variables. The methods of artificial neural network (ANN) and Fuzzy interface system (FIS) have been proven as tools for measuring risks in many fields. In this case study, the railway is selected as a place for managing the risks of overcrowding in the railway stations taking two parameters as input for risk value output using a hybrid model, which has the potency to deal with risk uncertainties and to learn by ANN training processes. The results show that the ANFIS method is more promising in the management of station risks. The framework can be applied for other risks in the station and more for a wide range of other systems. Also, ANFIS has the ability to learn from past risk records for future prediction. Clearly, the risk indexes are essential to reflect the actual condition of the station and they can indicate a high level of risks at the early stage, such as with overcrowding. The dynamic model of risk management can define risk levels and aid the decision makers by convenient and reliable results based on recorded data. Finally, the model can be generalised for other risks.

A fuzzy Anti-lock braking system (ABS) controller using CMOS circuits

In this paper, an Anti-lock braking system (ABS) is designed using a fuzzy logic controller (FLC). The FLC is designed in 0.35 μm standard CMOS process and this design of the circuit makes it having high speed and low power consumption. Also, all the controlling signals and input signals are in voltage form and there is no need for digital programming. Voltage mode design of the circuits leads to the convenient use of FLC with sensors. The objective function is designed to keep the wheel slip to a desired level so that maximum wheel tractive force and maximum vehicle deceleration are achieved. By training this function to the Adaptive Nero-Fuzzy Inference System (ANFIS) of MATLAB software the main parameters of fuzzy controller singletons and shape of Membership Function Generators (MFGs) are measured, then all the circuits of the proposed fuzzy logic controller is designed using Complementary Metal-Oxide-Semiconductor (CMOS) circuits, at the end simulation results of the fuzzy logic controller using Hspice simulator and BSIM3v3 (Berkeley Short-Channel Insulated gate field effect transistor Model) parameters verifies the performance and functionality of the design.

Experimental Measurement of Erosive Wear and Development of Prediction Model Using Adaptive Nero Fuzzy Inference System

An investigation has been made to predict the erosion wear of slurry. Experimental setup was developed and experiments were conducted with slurry erosion tester. Based on results of experimentation regression model was developed results of this and experimental results were then compared with prediction model developed with adaptive Nero Fuzzy Inference System. It is observed that the regression model has an error of 25%. However with ANFIS prediction model wear prediction error was reduced to about 6%.

Editorial

This issue of the Journal of Flood Risk Management (JFRM) commences the eighth year of our publication, having become an established and recognised platform for the dissemination of knowledge and practice in our area. It has long been my view that flood risk is essentially a human problem since risk is a measure of the potential for harm or damage that flooding can cause, and, harm and damage are human constructs rather than being intrinsic to the natural world. However the management of the flood risk involves a combination of activities or actions which may have physical effects on the propagation of the flood waters or have effects on the social and environmental ‘receptors’ of the impact of inundation. The papers in this issue illustrate the breadth of flood risk management as an area of interdisciplinary professional activity. To manage flood risk it is essential to understand the characteristics of the flood source through to the likely impacts of inundation. Two of the papers in this Issue of JFRM cover computational modelling of flood propagation. Huthoff et al. demonstrate the value of 2D modelling using standard commercial software for understanding the evolution of the nature of risk in the large flood plains of the Middle Mississippi River following the rupture of a levee in a major flood, introducing a categorisation of three phases inundation with differing potential impacts. On the other hand the contribution by Ransom and Younis presents a development in the numerical techniques for accurate 2D simulation of dambreak flows. These extreme flows are a more severe test of the numerical algorithms than slowly rising and falling floods; such advances will also be of value in similar contexts for example flash floods and the initial phase of levee failure as noted by Huthoff et al. The paper by Almoradie et al. illustrates another use of information technology in flood risk management, that of enabling and facilitating stakeholder dialogue and participation in flood risk management plans and strategies, a topic I raised in an earlier Editorial (Samuels 2012). This paper illustrates how this can be achieved at a local level through two real case studies from the UK and Germany. One means of mitigating flood risk – particularly residual risk – is by flood insurance, although the availability and uptake of flood insurance varies markedly between different countries as is illustrated in the introduction of the paper by Aliagha et al. This paper then discusses the factors that influence the decisions for residential property owners in Malaysia on whether or not to purchase flood insurance. The starting point for assessing interventions in the flood risk system is data on the factors that determine the existing nature of the risk; this issue of the JFRM contains two papers relating to the analysis of data. The paper by Paulikas and Rahman provides a temporal assessment of flooding fatalities in Pakistan covering a period of over 60 years using data from the EM-DAT International Disaster Database. The analysis of this data leads to the identification that whilst summer monsoonal flooding is the most frequent cause of deaths across the country, the greatest number of fatalities are caused by the landfall of tropical cyclones. Hasanpour Kashani et al. discuss a new approach to a common problem in analysing river flood data, the construction of a rating curve; this paper compares the performance of three artificial intelligence algorithms with more conventional approaches. The methods are assessed against data obtained on the Kizilirmak River in Turkey with the conclusion on this case that the Adaptive Nero-Fuzzy Inference System provided the best performance. Flood management can bring benefits to society beyond the avoidance of damage; this is central to the WMO Associated Programme on Flood Management (see: http://www.apfm.info/ ) that views Integrated Flood Management as integrating land and water resources development in a river basin, within the context of Integrated Water Resources Management. The remaining paper in this issue of JFRM by Saher et al. considers ‘spate irrigation’ that is used for flood water harvesting, in hill torrent areas of Pakistan. The paper describes application of GIS to facilitate sustainable water planning through the conservation of excess water resources thus reducing the flash flood hazard of hill torrents and making flood water available for irrigation. I hope that my brief sketch of the content of this issue, drawing on applications in America, Europe and Asia, will encourage you to read the papers and also consider offering a paper about innovation or insights from your own work in flood risk management whether as a researcher or as a practitioner.

Comparison of different methods for developing a stage–discharge curve of the Kizilirmak River

AbstractPrediction of a stage–discharge relationship is of immense importance for reliable planning, design and management of most water resources projects. The aim of this study is to compare the performance of artificial intelligence methods, namely, Artificial Neural Network, Adaptive Nero‐Fuzzy Inference System (ANFIS), and Gene‐Expression Programming, with those of two conventional methods, i.e., stage‐rating curve and Regression techniques in deriving a stage–discharge curve in the Kizilirmak River, Turkey. The daily minimum, mean and maximum river discharge, and stage data for period of 2005–2007 were used for training and testing the models. The comparison includes visual and parametric approaches, namely coefficient of correlation, mean absolute error and root mean square error. The results proved the high ability of the artificial intelligence methods in developing stage–discharge relationship. Furthermore, the performance of the ANFIS model was found to be superior to all the models.

Comparison of Soft Computing Techniques for Modelling of the EDM Performance Parameters

Selection of appropriate operating conditions is an important attribute to pay attention for in electrical discharge machining (EDM) of steel parts. The achievement of EDM process is affected by many input parameters; therefore, the computational relations between the output responses and controllable input parameters must be known. However, the proper selection of these parameters is a complex task and it is generally made with the help of sophisticated numerical models. This study investigates the capacity of Adaptive Nero-Fuzzy Inference System (ANFIS), genetic expression programming (GEP) and artificial neural networks (ANN) in the prediction of EDM performance parameters. The datasets used in modelling study were taken from experimental study. According to the results of estimating the parameters of all models in the comparison in terms of statistical performance is sufficient, but observed that ANFIS model is slightly better than the other models.

Comparison of three artificial intelligence techniques for discharge routing

The inter-comparison of three artificial intelligence (AI) techniques are presented using the results of river flow/stage timeseries, that are otherwise handled by traditional discharge routing techniques. These models comprise Artificial Neural Network (ANN), Adaptive Nero-Fuzzy Inference System (ANFIS) and Genetic Programming (GP), which are for discharge routing of Kizilirmak River, Turkey. The daily mean river discharge data with a period between 1999 and 2003 were used for training and testing the models. The comparison includes both visual and parametric approaches using such statistic as Coefficient of Correlation (CC), Mean Absolute Error (MAE) and Mean Square Relative Error (MSRE), as well as a basic scoring system. Overall, the results indicate that ANN and ANFIS have mixed fortunes in discharge routing, and both have different abilities in capturing and reproducing some of the observed information. However, the performance of GP displays a better edge over the other two modelling approaches in most of the respects. Attention is given to the information contents of recorded timeseries in terms of their peak values and timings, where one performance measure may capture some of the information contents but be ineffective in others. Thus, this makes a case for compiling knowledge base for various modelling techniques.

Adapted Neuro-Fuzzy Inference System on indirect approach TSK fuzzy rule base for stock market analysis

Nowadays because of the complicated nature of making decision in stock market and making real-time strategy for buying and selling stock via portfolio selection and maintenance, many research papers has involved stock price prediction issue. Low accuracy resulted by models may increase trade cost such as commission cost in more sequenced buy and sell signals because of insignificant alarms and otherwise bad diagnosis in price trend do not satisfy trader’s expectation and may involved him/her in irrecoverable cost. Therefore, in this paper, Neuro-Fuzzy Inference System adopted on a Takagi–Sugeno–Kang (TSK) type Fuzzy Rule Based System is developed for stock price prediction. The TSK fuzzy model applies the technical index as the input variables and the consequent part is a linear combination of the input variables. Fuzzy C-Mean clustering implemented for identifying number of rules. Initial membership function of the premise part approximately defined as Gaussian function. TSK parameters tuned by Adaptive Nero-Fuzzy Inference System (ANFIS). Proposed model is tested on the Tehran Stock Exchange Indexes (TEPIX). This index with high accuracy near by 97.8% has successfully forecasted with several experimental tests from different sectors.

Extracting compact fuzzy rules for nonlinear system modeling using subtractive clustering, GA and unscented filter

This paper presents a two stage procedure for building optimal fuzzy model from data for nonlinear dynamical systems. Both stages are embedded into Genetic Algorithm (GA) and in the first stage emphasis is placed on structural optimization by assigning a suitable fitness to each individual member of population in a canonical GA. These individuals represent coded information about the structure of the model (number of antecedents and rules). This information is consequently utilized by subtractive clustering to partition the input space and construct a compact fuzzy rule base. In the second stage, Unscented Filter (UF) is employed for optimization of model parameters, that is, parameters of the input–output Membership Functions (MFs). The proposed hybrid approach exploits the advantages and utilizes the desirable characteristics of all three algorithms for extracting accurate and compact fuzzy models. Case studies are given to illustrate the efficiency of the modeling procedure. Benchmark examples are analyzed and the results are compared with those obtained by Adaptive Nero-fuzzy Inference System (ANFIS). In all cases enhanced performance and superior results are obtained from the proposed procedure.