Mamdani-type Fuzzy Inference System Research Articles

A comparative study of fuzzy logic systems approach for river discharge prediction

Summary In recent years, flood disasters resulting from extreme rainfall have been on the increase in many regions of the world. In developed countries, the usual practice of mitigating flood disasters is by structural means which can reduce infrastructural damages as well as casualties but are unaffordable in most developing countries. The alternative then is to look for non-structural means that involve, among other things, early warning systems which can reduce casualties. The basic technical components of an early warning system involves a measurable input data set that trigger floods, a measurable output data set that quantify the extent of flood and an appropriate mathematical model that transforms the input data set into a corresponding output data set. There are many types of mathematical models that can be used to transform the input data into corresponding output data. The crux of this paper is on one type of data driven mathematical models, namely the use of fuzzy logic approach. The reliability and robustness of the approach are demonstrated with daily and 6-hourly discharge predictions in 4 rivers in 3 countries having contrasting climatological, geographical and land use characteristics. The first application is for two tropical rivers in Sri Lanka using daily upstream rainfall and discharge data to predict downstream discharge with the minimum implication function type Mamdani fuzzy inference system. The second application is for another tropical river in Fiji using similar type of data with daily and 6-h time scales. Both Mamdani type fuzzy inference system with minimum and product implication functions as well as Larsen type inference systems were used. In the third application, daily upstream and tributary discharges were used to predict downstream discharges in a temperate-climate river in China using the TSK type fuzzy inference system with clustering. The methods are robust and the results obtained are within reasonable agreement with observations.

Grey Fuzzy Multiobjective Optimization of Process Parameters for CNC Turning of GFRP/Epoxy Composites

For obtaining close fits and tolerances, certain amount of machining has to be carried out on GFRP (Glass Fibre Reinforced Plastic) composites, produced by primary manufacturing processes. A number of cylindrical GFRP composite parts are finish machined by turning. These include axles, bearings, spindles, rolls and steering columns. There is always a tradeoff between quality and productivity during machining operations. Hence it becomes essential to evaluate the optimal cutting parameters setting in order to satisfy these opposing requirements. In this study, a hybrid multiobjective optimization algorithm involving grey and fuzzy coupled with Taguchi methodology is used. Four process parameters, each at three levels are selected for the study viz. cutting tool nose radius, cutting speed, feed rate and depth of cut. Surface roughness parameter Ra, tangential cutting force Fz and material removal rate MRR are the chosen output performance measures. The experimental plan is laid according to Taguchi's orthogonal array L27. Woven fabric based GFRP/ Epoxy tubes produced using hand layup process are finish turned using PCD cutting tool. Grey relational coefficients of the three performance measures are converted into a single multi performance characteristics index (MPCI) using Mamdani type fuzzy inference system. This MPCI is then optimized using Taguchi analysis. The parameter combination of A2B1C1D3, i.e. tool nose radius of 0.8mm, cutting speed of 120 m/min, feed rate of 0.05mm/rev and depth of cut of 1.6mm, is evaluated as the optimum combination. The confirmatory experiment at these settings gave maximum value of MPCI, validating the results.

Study of fuzzy systems with Sugeno and Mamdani-type fuzzy inference systems for determination of heartbeat cases on Electrocardiogram (ECG) signals

This paper discusses the suitability of implementing Sugeno- and Mamdani-type FISs for heartbeat case determination based on the Electrocardiogram (ECG) signals. The heartbeat cases are Normal Sinus Rhythm (NORM), Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB), Ventricular Premature Contractions (VPC) and Atrial Premature Contractions (APC). Overall, fuzzy system with Sugeno-type FIS was developed based on the FL method used in a paper. System modifications were carried out to create an alternative system for the application, implemented using Mamdani-type FIS. Both systems were verified with 3000 sets of random data for systems’ performance comparison. Sugeno’s system sensitivities in determining each heartbeat case are 100%, which leads to a TCA value of 100%, whereas in Mamdani’s system, the sensitivities are all 100%, except for NORM heartbeat case which is 99.8% and thus TCA value is 99.9667%. It is also found that the Sugeno’s system processing time is always less compared to Mamd...

Multi-response optimisation of electrical discharge machining process using combined approach of RSM and FIS

In this work, an integrated approach for optimising multiple responses using response surface methodology (RSM) combined with fuzzy inference system (FIS) and particle swarm optimisation (PSO) algorithm is proposed for obtaining best parametric combination in electrical discharge machining (EDM) process. Four process parameters (factors) such as discharge current (Ip), pulse-on-time (Ton), duty factor (τ) and flushing pressure (Fp) and four important responses like material removal rate (MRR), tool wear rate (TWR), surface roughness (Ra) and circularity (r1/r2) of machined component are considered in this study. As the responses are conflicting in nature, a single combination of parameters cannot produce best performance satisfying all the responses. A Box-Behnken design is used to generate data from experimental set up and Mamdani-type fuzzy inference system is used to convert multiple responses into a single equivalent characteristic index known as multi-response performance characteristic index (MPCI). The relationship between MPCI and process parameters is expressed mathematically using multiple regression analysis. Finally, particle swarm optimisation technique is employed to obtain best parametric combination that maximises MPCI. The machined surfaces are analysed through scanning electron microscope (SEM) to study the effect of statistically significant parameters on pores and micro crack found in the of wok piece material.

Fuzzy-logic modeling of Fenton’s strong chemical oxidation process treating three types of landfill leachates

Three multiple input and multiple output-type fuzzy-logic-based models were developed as an artificial intelligence-based approach to model a novel integrated process (UF-IER-EDBM-FO) consisted of ultrafiltration (UF), ion exchange resins (IER), electrodialysis with bipolar membrane (EDBM), and Fenton's oxidation (FO) units treating young, middle-aged, and stabilized landfill leachates. The FO unit was considered as the key process for implementation of the proposed modeling scheme. Four input components such as H(2)O(2)/chemical oxygen demand ratio, H(2)O(2)/Fe(2+) ratio, reaction pH, and reaction time were fuzzified in a Mamdani-type fuzzy inference system to predict the removal efficiencies of chemical oxygen demand, total organic carbon, color, and ammonia nitrogen. A total of 200 rules in the IF-THEN format were established within the framework of a graphical user interface for each fuzzy-logic model. The product (prod) and the center of gravity (centroid) methods were performed as the inference operator and defuzzification methods, respectively, for the proposed prognostic models. Fuzzy-logic predicted results were compared to the outputs of multiple regression models by means of various descriptive statistical indicators, and the proposed methodology was tested against the experimental data. The testing results clearly revealed that the proposed prognostic models showed a superior predictive performance with very high determination coefficients (R (2)) between 0.930 and 0.991. This study indicated a simple means of modeling and potential of a knowledge-based approach for capturing complicated inter-relationships in a highly non-linear problem. Clearly, it was shown that the proposed prognostic models provided a well-suited and cost-effective method to predict removal efficiencies of wastewater parameters prior to discharge to receiving streams.

Estimation of fracture energy of high-strength steel fibre-reinforced concrete using rule-based Mamdani-type fuzzy inference system

AbstractIn this study, we worked to estimate the fracture energy of steel fibre-reinforced concrete (SFRC) according to the water/cement ratio (w/c), tensile strength of steel fibre, steel fibre volume fraction and flexural strength of concrete sample as inputs using the Mamdani-type fuzzy inference system (FIS). In the study, the values obtained from the model and experimental divided three groups (each group has six experimental results) according to the w/c ratios to evaluate the fuzzy logic (FL) model approximate reasoning ability. As a result, the Mamdani-type FIS has shown a satisfying relation with the experimental results and suggests an alternative approach to evaluate the fracture energy estimation using related inputs.

Performance Evaluation of Mamdani-type and Sugeno-type Fuzzy Inference System Based Controllers for Computer Fan

Nowadays, there are several models of computer systems finding their ways into various offices, houses, organizations as well as remote locations. Any slight malfunction of the computer system's components could lead to loss of vital data and information. One of the sources of computer system malfunction is overheating of the electronic components. A common method of cooling a computer system is the use of cooling fan(s). Therefore, it is essential to have an appropriate control mechanism for the operation of computer system's cooling fan in order to save energy, and prevent overheating. Failure to adopt a well designed and efficient performance controller could lead to the malfunction of a computer system. Presently, most controllers in computer systems are pulse width modulation based. That is, they make use of pulses in form of digits, 0 and 1. It was observed that inherent noise is still prevalent in the operation of computer system. Also, eventual breakdown of components is common. A new approach is therefore investigated through the use of fuzzy logic to serve as a base or platform to build an intelligent controller using a set of well defined rules to guide its operational performance. Mamdani-type fuzzy inference system and Sugeno-type fuzzy inference system were used with two input sets each and a single output function each. Simulation was carried out in MATLAB R2007a platform and operational performances of the two approaches were compared. Simulated results of the performances of the Mamdani-type fuzzy inference system based controller and the Sugeno-type fuzzy inference system based controller are presented accordingly.

Modelling Risk Perception in ATIS Context: A Comparison of Different Fuzzy Logic-Based Approaches

In this paper travellers’ reactions to Advanced Traveller Information Systems (ATIS) are analysed. In particular two kinds of information (descriptive and prescriptive) and four levels of reliability have been tested. A web-based tool has been adopted in order to carry out a stated preference experiment for data collection. The presented research continues previous studies of the authors in the field of travellers’ compliance with information and travellers’ route choices under ATIS. In previous studies both a discrete choice theory approach and a Mamdani-type Fuzzy Inference System (FIS) were tested. Here several FIS approaches are analysed more in detail. Some preliminary analyses, are recalled from previous research work, furthermore collected data have been deeply analysed through the Sugeno FIS-type approach and by Adaptive-Network-Based FIS. The methods are applied to reproduce travellers’ behaviour and are compared with each other to find the best approach.

Fuzzy-logic modeling of Fenton's oxidation of anaerobically pretreated poultry manure wastewater

A multiple inputs and multiple outputs (MIMO) fuzzy-logic-based model was proposed to estimate color and chemical oxygen demand (COD) removal efficiencies in the post-treatment of anaerobically pretreated poultry manure wastewater effluent using Fenton's oxidation process. Three main input variables including initial pH, Fe+2, and H2O2 dosages were fuzzified in a new numerical modeling scheme by the use of an artificial intelligence-based approach. Trapezoidal membership functions with eight levels were conducted for the fuzzy subsets, and a Mamdani-type fuzzy inference system was used to implement a total of 70 rules in the IF-THEN format. The product (prod) and the center of gravity (centroid) methods were applied as the inference operator and defuzzification methods, respectively. Fuzzy-logic predicted results were compared with the outputs of two first-order polynomial regression models derived in the scope of this study. Estimated results were also compared to the multiple regression approach by means of various descriptive statistical indicators, such as root mean-squared error, index of agreement, fractional variance, proportion of systematic error, etc. Results of the statistical analysis clearly revealed that, compared to conventional regression models, the proposed MIMO fuzzy-logic model produced very smaller deviations and demonstrated a superior predictive performance on forecasting of color and COD removal efficiencies with satisfactory determination coefficients over 0.98. Due to high capability of the fuzzy-logic methodology in capturing the non-linear interactions, it was demonstrated that a complex dynamic system, such as Fenton's oxidation, could be easily modeled.

Pitch Control of an Aircraft Using Artificial Intelligence

This paper presents the investigation into the design, simulation and analysis of two autopilots: a fuzzy Proportional-Integral-Derivative (PID) controller and, its hybrid with a PID controller for the control of pitch plane dynamics of an aircraft. The Mamdani-type fuzzy inference system is employed for the Fuzzy Inference System (FIS) in the fuzzy logic controller design. The dynamic modeling of system begins with a derivation of suitable mathematical model to describe the longitudinal motion of an aircraft. This research set the platform for thorough investigation into the various structures available for PID-FLC and its hybrids. Considering hardware implementation challenges and limitations, not all PID-FLC and its hybrid structures are viable. The PID-FLC is constructed as a parallel structure of a PD-FLC and a PI-FLC, with the output signal of the control loop, y serving as the input for the derivative parameter of the PD-FLC. The output of the PID-FLC is formed by algebraically adding the outputs of the two fuzzy control blocks as suggested in Guanrong et al., 2000. Also, the proposed hybrid fuzzy PID autopilot consists of the PID-FLC with a traditional PID controller structured by algebraically adding the outputs of the two control blocks. Result of simulation in MATLAB ® /Simulink ® shows that the proposed PID-FLC autopilot gave an unacceptable

A Prognostic Approach Based on Fuzzy-Logic Methodology to Forecast PM10 Levels in Khaldiya Residential Area, Kuwait

A prognostic approach is proposed based on a fuzzy-logic model to estimate suspended dust concentrations, related to PM10, in a specific residential area in Kuwait with high traffic and industrial influences. Seven input variables, including four important meteorological parameters (wind speed, wind direction, relative humidity and solar radiation) and the ambient concentrations of three gaseous pollutants (methane, carbon monoxide and ozone) were fuzzified using a sytem with a graphical user interface (GUI) and an artificial intelligence-based approach. Trapezoidal membership functions with ten and fifteen levels were employed for the fuzzy subsets of each model variable. A Mamdani-type fuzzy inference system (FIS) was developed to introduce a total of 146 rules in the IF-THEN format. The product (prod) and the centre of gravity (centroid) methods were performed as the inference operator and defuzzification methods, respectively, for the proposed FIS. The results obtained using uzzy-logic were compared with the outputs of an exponential regression model. The predictive performances of the models were compared based on various descriptive statistical indicators, and the proposed method was tested against additional observed data. The prognostic model presented in this work produced very small deviations from the actual results, and showed better predictive performance than the other model with regard to forecasting PM10 levels, with a very high determination coefficient of over 0.99.

Hybrid intelligent system for cardiac arrhythmia classification with Fuzzy K-Nearest Neighbors and neural networks combined with a fuzzy system

In this paper we describe a hybrid intelligent system for classification of cardiac arrhythmias. The hybrid approach was tested with the ECG records of the MIT-BIH Arrhythmia Database. The samples considered for classification contained arrhythmias of the following types: LBBB, RBBB, PVC and Fusion Paced and Normal, as well as the normal heartbeats. The signals of the arrhythmias were segmented and transformed for improving the classification results. Three methods of classification were used: Fuzzy K-Nearest Neighbors, Multi Layer Perceptron with Gradient Descent and momentum Backpropagation, and Multi Layer Perceptron with Scaled Conjugate Gradient Backpropagation. Finally, a Mamdani type fuzzy inference system was used to combine the outputs of the individual classifiers, and a very high classification rate of 98% was achieved.

An easy-to-use MATLAB program (MamLand) for the assessment of landslide susceptibility using a Mamdani fuzzy algorithm

In this study, landslide susceptibility mapping using a completely expert opinion-based approach was applied for the Sinop (northern Turkey) region and its close vicinity. For this purpose, an easy-to-use program, “MamLand,” was developed for the construction of a Mamdani fuzzy inference system and employed in MATLAB. Using this newly developed program, it is possible to construct a landslide susceptibility map based on expert opinion. In this study, seven conditioning parameters characterising topographical, geological, and environmental conditions were included in the FIS. A landslide inventory dataset including 351 landslide locations was obtained for the study area. After completing the data production stage of the study, the data were processed using a soft computing approach, i.e., a Mamdani-type fuzzy inference system. In this system, only landslide conditioning data were assessed, and landslide inventory data were not included in the assessment approach. Thus, a file depicting the landslide susceptibility degrees for the study area was produced using the Mamdani FIS. These degrees were then exported into a GIS environment, and a landslide susceptibility map was produced and assessed in point of statistical interpretation. For this purpose, the obtained landslide susceptibility map and the landslide inventory data were compared, and an area under curve (AUC) obtained from receiver operating characteristics (ROC) assessment was carried out. From this assessment, the AUC value was found to be 0.855, indicating that this landslide susceptibility map, which was produced in a data-independent manner, was successful.

Development of In-process Surface Roughness Evaluation System for Cast Nylon 6 Turning Operation

This paper aims to develop in-process surface roughness evaluation system for cast nylon 6 turning operation. Construction of the developed system composes of data acquisition system and fuzzy logic system. The data acquisition system includes a miniature load cell inserted in a tool holder, signal conditions and a signal interface card in order to detect cutting force signal and transmit signal to the data analysis module constructed by LabVIEW program. The mamdani-type fuzzy inference system was utilized and twenty fuzzy rules were determined based on relationship of cutting speed, feed rate and cutting force for prediction both of surface roughness value (Rz) and symbolic representation. The results indicated that prediction by the developed system had 87% of accuracy with an average absolute error of 4.8μm. The method validation results exhibited that precision and surface roughness value obtained from conventional measuring method and the developed prediction system was not significant different.

A fuzzy-logic-based model to predict biogas and methane production rates in a pilot-scale mesophilic UASB reactor treating molasses wastewater

A MIMO (multiple inputs and multiple outputs) fuzzy-logic-based model was developed to predict biogas and methane production rates in a pilot-scale 90-L mesophilic up-flow anaerobic sludge blanket (UASB) reactor treating molasses wastewater. Five input variables such as volumetric organic loading rate (OLR), volumetric total chemical oxygen demand (TCOD) removal rate ( R V), influent alkalinity, influent pH and effluent pH were fuzzified by the use of an artificial intelligence-based approach. Trapezoidal membership functions with eight levels were conducted for the fuzzy subsets, and a Mamdani-type fuzzy inference system was used to implement a total of 134 rules in the IF-THEN format. The product (prod) and the centre of gravity (COG, centroid) methods were employed as the inference operator and defuzzification methods, respectively. Fuzzy-logic predicted results were compared with the outputs of two exponential non-linear regression models derived in this study. The UASB reactor showed a remarkable performance on the treatment of molasses wastewater, with an average TCOD removal efficiency of 93 (±3)% and an average volumetric TCOD removal rate of 6.87 (±3.93) kg TCOD removed/m 3-day, respectively. Findings of this study clearly indicated that, compared to non-linear regression models, the proposed MIMO fuzzy-logic-based model produced smaller deviations and exhibited a superior predictive performance on forecasting of both biogas and methane production rates with satisfactory determination coefficients over 0.98.

Modeling changes in biomass composition during microwave‐based alkali pretreatment of switchgrass

This study used two different approaches to model changes in biomass composition during microwave-based pretreatment of switchgrass: kinetic modeling using a time-dependent rate coefficient, and a Mamdani-type fuzzy inference system. In both modeling approaches, the dielectric loss tangent of the alkali reagent and pretreatment time were used as predictors for changes in amounts of lignin, cellulose, and xylan during the pretreatment. Training and testing data sets for development and validation of the models were obtained from pretreatment experiments conducted using 1-3% w/v NaOH (sodium hydroxide) and pretreatment times ranging from 5 to 20 min. The kinetic modeling approach for lignin and xylan gave comparable results for training and testing data sets, and the differences between the predictions and experimental values were within 2%. The kinetic modeling approach for cellulose was not as effective, and the differences were within 5-7%. The time-dependent rate coefficients of the kinetic models estimated from experimental data were consistent with the heterogeneity of individual biomass components. The Mamdani-type fuzzy inference was shown to be an effective approach to model the pretreatment process and yielded predictions with less than 2% deviation from the experimental values for lignin and with less than 3% deviation from the experimental values for cellulose and xylan. The entropies of the fuzzy outputs from the Mamdani-type fuzzy inference system were calculated to quantify the uncertainty associated with the predictions. Results indicate that there is no significant difference between the entropies associated with the predictions for lignin, cellulose, and xylan. It is anticipated that these models could be used in process simulations of bioethanol production from lignocellulosic materials.

SU-FF-T-377: Self-Learning AI Technique for Parameter Optimization of IMRT Treatment Planning

Purpose: The conventional application of Fuzzy Inference System (FIS) on parameter optimization of IMRT treatment planning requires that the FIS model (membership function, inference rules) to be predetermined in an intuitive way based on prior planner's knowledge. This study is aimed to develop an automated update scheme using an Adaptive Neuro-Fuzzy Inference System (ANFIS). Method and Materials: An ANFIS is a technique developed for modelling a special type of FIS originally introduced by Sugeno [1]. The Sugeno-type FIS has similar structure as the conventional Mamdani-type FIS, and can update itself based on the observation data. Without the prior planner's knowledge, a Sugeno-type FIS can adjust itself to adapt supervising signal, i.e. training data. It is a great leap over conventional Mamdani-type FIS. For experimental purpose, one set of training and testing data were obtained from the observations of the input-output responses of a Mamdani-type FIS previously constructed by human knowledge. Then, the model of Sugeno-type FIS was iteratively adjusted by a gradient-descent algorithm. As learning procedure was completed, it was examined by the testing data and the error between the outputs of Sugeno-type and Mamdani-type FIS was evaluated. Results: With the same training data set and testing data set for both systems, the discrepancy between the outputs of Sugeno-type FIS and Mamdani-type FIS is within 2% and varies by 3%. Conclusion: With adequate training data and training epoch, the Sugeno-type FIS can effectively approximate the function of Mamdami-type FIS. The result indicates that it is possible to build a self-learning AI system for parameter optimization of IMRT treatment planning. Acknowledgement: this project is partially supported by Varian Research Grant.