Sugeno Fuzzy Inference Research Articles

Competency mapping with Sugeno fuzzy inference system for variable pay determination: A case study

Abstract In this work we propose a novel approach for computing the variable pay/pay for performance component of employees in a technical institution based on Takagi Sugeno Kang (TSK)/Sugeno fuzzy inference system. Sugeno fuzzy inference system is an inference model based on fuzzy set theory and is known to handle MCDM (Multiple criteria decision making) problems quite efficiently. In the proposed model the Sugeno fuzzy inference system has been used to compute the variable pays of the faculties based on their individual API (Academic Performance Indicator) scores in a particular appraisal year. The model aims to overcome the pitfalls of the current forced distribution technique being used for appraisal and subsequent variable pay decisions. The performance of the proposed model has been tested with the API scores of all the faculties of the said institute. Along with that a function was used to generate random inputs in the range of API scores to extensively test the model. All the methods were implemented in MATLAB and the results were analyzed in EXCEL using the CORREL and PEARSON’s coefficient. Experimental results show that the proposed model is reliable and demonstrates significant improvement over the traditional forced distribution approach currently adopted.

Generating realistic intrusion detection system dataset based on fuzzy qualitative modeling

Prior to deploying any intrusion detection system, it is essential to obtain a realistic evaluation of its performance. However, the major problems currently faced by the research community is the lack of availability of any realistic evaluation dataset and systematic metric for assessing the quantified quality of realism of any intrusion detection system dataset. It is difficult to access and collect data from real-world enterprise networks due to business continuity and integrity issues. In response to this, in this paper, firstly, a metric using a fuzzy logic system based on the Sugeno fuzzy inference model for evaluating the quality of the realism of existing intrusion detection system datasets is proposed. Secondly, based on the proposed metric results, a synthetically realistic next generation intrusion detection systems dataset is designed and generated, and a preliminary analysis conducted to assist in the design of future intrusion detection systems. This generated dataset consists of both normal and abnormal reflections of current network activities occurring at critical cyber infrastructure levels in various enterprises. Finally, using the proposed metric, the generated dataset is analyzed to assess the quality of its realism, with its comparison with publicly available intrusion detection system datasets for verifying its superiority.

Optimized sugeno fuzzy clustering algorithm for wireless sensor networks

Clustering is the most common approach to achieve energy efficiency in wireless sensor networks. The existing clustering techniques exhibit some drawbacks which limit their usage for practical networks. First, cluster heads are typically selected among all sensor nodes within the network, and consequently, unbalanced clusters may be generated. Second, the controllable parameters are defined manually. Third, the protocol is not adjusted and tuned based on application specifications. In this paper, we propose an adaptive fuzzy clustering protocol (named LEACH-SF), in order to overcome the mentioned drawbacks. In LEACH-SF, fuzzy c-means algorithm is used to cluster all sensor nodes into balanced clusters, and then appropriate cluster heads are selected via Sugeno fuzzy inference system. The fuzzy inputs of the Sugeno fuzzy inference system include the residual energy, the distance from sink, and the distance from cluster centroid. Unlike the existing fuzzy-based routing protocols in which the fuzzy rule base table is defined manually, we utilize artificial bee colony algorithm to adjust the fuzzy rules of LEACH-SF. The fitness function of the algorithm is defined to prolong the network lifetime, based on the application specifications. In other words, LEACH-SF not only prolongs the lifetime, but also is applicable to any kind of application. Simulations over 10 heterogeneous wireless sensor networks show that LEACH-SF outperforms the existing cluster-based routing protocols.

A Hybrid Fuzzy-Based Multi-Objective PSO Algorithm for Conjunctive Water Use and Optimal Multi-Crop Pattern Planning

This paper focuses on extracting an optimal multi-crop pattern plan through multi-objective conjunctive surface-ground water use management. Minimizing shortages in meeting irrigation demands, maximizing groundwater resources sustainability and maximizing agricultural net benefits are the three main goals of the multi-objective optimization problem solved in this paper. A new robust fuzzy-based multi-objective PSO algorithm called f-MOPSO is adopted and modified to solve a three-objective real-world conjunctive use management problem presented in this paper after testing on standard test problems revealed f-MOPSO superiority as compared to the well-known multi-swarm vector evaluated PSO (VEPSO) algorithm. The f-MOPSO benefits from a well-organized Sugeno fuzzy inference system (SFIS) designed for handling multi-objective nature of the optimization problems. The unique performance of f-MOPSO is not only presenting the better final solutions, but also aggregating the capabilities for measurement of dominance and diversity of the solutions in one stage by one index named comprehensive dominance index, in contrast to a wide range of multi-objective algorithms that evaluate dominance and diversity in two separate stages resulting in excessive computational burden. The optimization model is carried out on a 10-year long-term simulation period, resulting in increasing irrigation efficiency i.e. decreasing water losses, decreasing water consumption per unit cultivated area and increasing water productivity compared to those similar criteria observed in actual operation in the study area. The wheat and rice crops were identified as the dominant crops, while the optimization model was the least interested to onion cultivation, assigning the least average cultivation area to this crop over the whole planning period.

Shear wave prediction using committee fuzzy model constrained by lithofacies, Zagros basin, SW Iran

The main purpose of this study is to introduce the geological controlling factors in improving an intelligence-based model to estimate shear wave velocity from seismic attributes. The proposed method includes three main steps in the framework of geological events in a complex sedimentary succession located in the Persian Gulf. First, the best attributes were selected from extracted seismic data. Second, these attributes were transformed into shear wave velocity using fuzzy inference systems (FIS) such as Sugeno's fuzzy inference (SFIS), adaptive neuro-fuzzy inference (ANFIS) and optimized fuzzy inference (OFIS). Finally, a committee fuzzy machine (CFM) based on bat-inspired algorithm (BA) optimization was applied to combine previous predictions into an enhanced solution. In order to show the geological effect on improving the prediction, the main classes of predominate lithofacies in the reservoir of interest including shale, sand, and carbonate were selected and then the proposed algorithm was performed with and without lithofacies constraint. The results showed a good agreement between real and predicted shear wave velocity in the lithofacies-based model compared to the model without lithofacies especially in sand and carbonate.

Fuzzy-Based Flat Anomaly Diagnosis and Relief Measures in Distributed Wireless Sensor Network

This paper bestows a distributed adaptive scheme for diagnosing inaccurate data (anomaly) in wireless sensor networks. Faults occurring in sensor nodes are routine owing to the sensor device itself and the harsh environment in which the sensor nodes are deployed. It is mandatory for the WSNs to discover the anomaly and take actions to avoid further seediness of the network lifetime for confirming data accuracy. In this standpoint, we propose two perspectives for diagnosing and alleviating anomalies. The first view depicts input space partitioning by subtractive clustering method with robust density measure. Later, Takagi–Sugeno fuzzy inference model is applied for selection of several parameters and its membership functions, and rule-based construction is practiced to spot anomalies in distributed clustering wireless sensor network. By exploring combined correlation analysis with second perspective, the eliminated anomalous data are replaced by imputed data. Experimental results infer accuracy and reliability with a reduced amount of energy consumption than the state-of-the-art techniques.

Convergence analysis of the batch gradient-based neuro-fuzzy learning algorithm with smoothing L1/2 regularization for the first-order Takagi–Sugeno system

It has been proven that Takagi–Sugeno systems are universal approximators, and they are applied widely to classification and regression problems. The main challenges of these models are convergence analysis and their computational complexity due to the large number of connections and the pruning of unnecessary parameters. The neuro-fuzzy learning algorithm involves two tasks: generating comparable sparse networks and training the parameters. In addition, regularization methods have attracted increasing attention for network pruning, particularly the Lq(0<q<1) regularizer after L1 regularization, which can obtain better solutions to sparsity problems. The L1/2 regularizer has a specific sparsity capacity and it is representative of Lq(0<q<1) regularizations. However, the nonsmoothness of the L1/2 regularizer may lead to oscillations in the learning process. In this study, we propose a gradient-based neuro-fuzzy learning algorithm with a smoothing L1/2 regularization for the first-order Takagi–Sugeno fuzzy inference system. The proposed approach has the following three advantages: (i) it enhances the original L1/2 regularizer by eliminating the oscillation of the gradient in the cost function during the training; (ii) it performs better by pruning inactive connections, where the number of the redundant connections for removal is higher than that generated by the original L1/2 regularizer, while it is also implemented by simultaneous structure and parameter learning processes; and (iii) it is possible to demonstrate the theoretical convergence analysis of this learning method, which we focus on explicitly. We also provide a series of simulations to demonstrate that the smoothing L1/2 regularization can often obtain more compressive representations than the current L1/2 regularization.

Optimization of Sinter Plant Operating Conditions Using Advanced Multivariate Statistics: Intelligent Data Processing

Blast furnace operators expect to get sinter with homogenous and regular properties (chemical and mechanical), necessary to ensure regular blast furnace operation. Blends for sintering also include several iron by-products and other wastes that are obtained in different processes inside the steelworks. Due to their source, the availability of such materials is not always consistent, but their total production should be consumed in the sintering process, to both save money and recycle wastes. The main scope of this paper is to obtain the least expensive iron ore blend for the sintering process, which will provide suitable chemical and mechanical features for the homogeneous and regular operation of the blast furnace. The systematic use of statistical tools was employed to analyze historical data, including linear and partial correlations applied to the data and fuzzy clustering based on the Sugeno Fuzzy Inference System to establish relationships among the available variables.

Induction Motor Relaying Scheme for External Faults Detection and Classification Using Subtractive Clustering Based Sugeno Fuzzy Inference System

Conventional thermal relays generally suffer from inaccurate induction motor external faults identification. Recent developments in diagnosis systems have led to the scope of radically different diagnosis strategies based on artificial intelligence techniques. This study use fuzzy logic as a software tool and relay logic for external faults identification of an induction motor. In this study, a subtractive clustering based Sugeno-type fuzzy inference system is utilized to classify five external faults and the normal condition of a low-voltage three-phase induction motor using simulation and practical data sets. Five external faults- overload, over-voltage, under-voltage, single-phasing, and voltage unbalance-along with normal conditions are simulated on an induction motor over a wide range of operating voltage and load torque; sampled three-phase RMS voltages and currents are used as input feature vectors to train the fuzzy inference system. Average total classification accuracy and average root means square error obtained with ten-times random sub-sampling cross-validation of test data sets are used to find the best accurate and generalized fuzzy inference system configuration among different cluster radius fuzzy inference systems for both simulation and real-time data sets. This method accurately detects external faults with 94.37% average total classification accuracy for practical test data sets.

Method for Higher Order polynomial Sugeno Fuzzy Inference Systems

A method for Higher Order polynomial Sugeno Fuzzy Inference Systems formation is presented. Compared to other existing Higher Order Sugeno implementations, it uses fewer parameters; and compared to Zero and 1st Order Sugeno Fuzzy Systems it has overall improved model performance. While best models are not always obtained via a Higher Order representation, in our proposed method it is possible to choose the polynomial Order which best fits the desired data. Its input is a previously established model found by a clustering algorithm (subtractive algorithm in this case). Afterward, parameters of all Higher Order polynomials are adjusted using Recursive Least Square algorithm. For experimental validation, multiple benchmark datasets are tested using Hold-Out and K-fold validation as well as data forecasting. Various performance measures are used, although Akaike Information Criterion is used as a primary measure to demonstrate that our proposed Higher Order polynomials have overall better model performance over Zero and 1st Order polynomials.

Fuzzy collision avoidance system for ships

The paper discusses the problem of maritime traffic control. A model of the relative motion of two vessels is considered. An algorithm for the generation of alarms of various types in accordance with the verbal ship---ship danger level is considered. Navigation situations are separated into levels based on the ship's maneuvering intensity and time to collision. A fuzzy decision-making system about the motion's danger level that combines Mamdani and Sugeno fuzzy inference systems is proposed. The results of the numerical experiment that demonstrates the system's operation under standard conditions and the results of the system's field tests based on real ship traffic data in the waters adjacent to the port of Vladivostok are given.

Sugeno fuzzy inference algorithm and its application in epicentral intensity prediction

In this paper, the epicentral intensity prediction model based on Sugeno fuzzy inference system has been proposed where the earthquake magnitude and the depth of hypocenter are input and the epicentral intensity is output. The parameters of the proposed model are estimated by statistical analysis based on the historical data for Sichuan earthquake in the last decades. Finally, by comparison with the existing fitting model, the proposed model has better efficiency and higher accuracy.

Predicting building's corners hygrothermal behavior by using a Fuzzy inference system combined with clustering and Kalman filter

The hygroscopic characteristics of building materials can affect thermal gain or losses that are directly associated to energy consumption due to the latent heat transport. Moreover, some specific regions can accumulate humidity on building structures, and some of this regions, known as building corners, are still barely explored due to modelling complexity, high computer run time, numerical divergence, and highly moisture-dependent properties. This article presents an alternative to predict temperature, vapor pressure, and moisture content profiles in specific points where moisture can be easily accumulated, increasing mould growth risks and/or causing structural damage to the building. In order to avoid time-consuming numerical models, this article uses a Takagi–Sugeno fuzzy inference system with a multiple-input, single-output (MISO) structure to predict building corners hygrothermal behavior. Due to the ability of nonlinearity detection, associated with a small number of “if–then” rules with fuzzy antecedents and crisp mathematical functions or linear functions in the resultant part, the fuzzy system was combined with subtractive clustering method and Kalman filter to enhance its performance. The results suggested that the developed Takagi–Sugeno fuzzy model has achieved good accuracy in terms of precision when the results were compared to the analytical model. Moreover, in terms of simulation time, after the tuning and optimization procedures, the prediction of temperature, relative humidity, and vapor pressure on specific nodes are faster than the numerical model.

Emotional inference by means of Choquet integral and λ-fuzzy measurement in consideration of ambiguity of human mentality

Research on human emotions has started to address psychological aspects of human nature and has advanced to the point of designing various models that represent them quantitatively and systematically. Based on the findings, a method is suggested for emotional space formation and emotional inference that enhance the quality and maximize the reality of emotion-based personalized services. In consideration of the subjective tendencies of individuals, AHP was adopted for the quantitative evaluation of human emotions, based on which an emotional space remodeling method is suggested in reference to the emotional model of Thayer and Plutchik, which takes into account personal emotions. In addition, Sugeno fuzzy inference, fuzzy measures, and Choquet integral were adopted for emotional inference in the remodeled personalized emotional space model. Its performance was evaluated through an experiment. Fourteen cases were analyzed with 4.0 and higher evaluation value of emotions inferred, for the evaluation of emotional similarity, through the case studies of 17 kinds of emotional inference methods. Matching results per inference method in ten cases accounting for 71% are confirmed. It is also found that the remaining two cases are inferred as adjoining emotion in the same section. In this manner, the similarity of inference results is verified.

Fuzzy - Based Method of Detecting the Enviroment Character for UAV Optical Stabilization

An optical stabilization of UAV (UAS) is a very important part of a structure in their control systems. Not only as a backup stabilization system in a case of IMU failure, but also as a main system, used for stabilization or navigation. In this paper the concept of a system for environment character detection is presented. The system can classify a surrounding environment depending on chosen characteristics. Such system can be used for a better horizon detection due to switching to a correct horizon detection algorithm, which can be used for determining the position of UAV. The system is based on Takagi - Sugeno fuzzy inference system and fuzzy artificial neural networks. An earlier work on this subject was presented last year, but concept of the system was redesigned with a usage of fuzzy artificial neural network for a more precisive outputs and automatic determination of characteristics of fuzzy sets on input.

Monitoring the osmotic dehydration process of quince by the novel fusion modular neural networks - fuzzy logic (Fmnn-Fl)

This paper presents a novel approach to monitor food process based on Modular Neural Networks (MNNs) and fuzzy inference system. The proposed MNN consists of three separate modules, each using different image features as input including: edge detection, wavelet transform, and Hough transform. The sugeno fuzzy inference system was used to combine the outputs from each of these modules to classify the images of quince during osmotic dehydration process. To test the method, for classification, database was made of 108 quince samples’ images (12 classes). In experiments, the developed architecture achieved 91.6% recognition accuracy. Next step, solid gain, water loss and moisture content of quince samples were considered as MNNs outputs, whereas osmotic dehydration time and classified images were MNNs inputs. The minimum %MRE (18.153) with 89% prediction ability for water loss (WL) was obtained when applying two hidden layers with 6 neurons per each two layers. The lowest %MRE (35.5335) with 93% prediction ability for solid gain (SG) was obtained when using 6 and 8 neurons per first and second layer, respectively. And finally %MRE was at least (7.4759) with 96% prediction ability for moisture content (MC) by 6 and 5 neurons per first and second layer, respectively. The results show that this model could be commendably implemented for quantitative modeling and monitoring of food quality changes during osmotic dehydration process.

OPTIMASI FUZZY INFERENCE SYSTEM SUGENO DENGAN ALGORITMA HILL CLIMBING UNTUK PENENTUAN HARGA JUAL RUMAH

A house selling price can be determined by two methods, financially and technically. Hovewer, the selling price that determined by the methods are often different. It makes the manager having a problem when determining the house final selling price accurately. This paper proposes Sugeno Fuzzy Inference System (FIS) to calculate an accurate price. To get better result, Hill Climbing Algorithms is used to optimize the membership function of Sugeno FIS. A series of computational experimens prove that the approach is effective. Hill Climbing Algorithms can improve the accuracy of results.

Fuzzy Docking Guidance Using Augmented Navigation System on an AUV

A fundamental successful docking operation requires the autonomous underwater vehicle (AUV) to be able to guide, navigate, and control itself into the docking station in a strategic manner and even possibly execute different maneuvers at different mission phases, depending on docking scenario, requirements, and homing sensor type. A docking station, due to environmental or mission requirements, is possibly oriented at a specific direction instead of allowing omnidirectional homing, and necessitates vehicle docking in only this direction. Depending on the operating environment, either wave or current presence or both can result in a dominating disturbance to the vehicle docking operation. In this work, an inverted ultrashort baseline (USBL) system is used as the main homing sensor to complement the existing navigation suite on the DSO-developed AUV. A docking guidance system was designed and implemented using the Sugeno fuzzy inference system (FIS). A desired heading vector field and the fuzzy rules were developed to perform the fuzzy docking maneuver. An error-state Kalman filter (KF) was designed, formulated, and implemented successfully on the AUV and has proven to perform excellent relative positioning estimation in sea trials. A software architecture was designed for the docking algorithms, and implemented onto a single board computer in the AUV. A sensor fusion approach to the software programming was adopted to ensure that navigation data from all navigation sensors are properly acquired and synchronized. A docking station was designed and eventually deployed at sea for docking trials. Successful AUV docking attempts at sea trials were demonstrated, thus showing the effectiveness of the implemented docking algorithms.

HARmonic–LINear (HarLin) model for solar irradiation estimation

The solar energy has potential future effectiveness in a variety of areas and it helps to decrease the dominance of fossil fuels, which cause atmospheric pollution, global warming and climate change impacts. In the literature, there are different methodologies for its modeling, but this study suggests the harmonic analysis application to solar irradiation and sunshine duration data for more refined relevant prediction of solar irradiation. The basis of the methodology is combined application of the HARmonic and the classical LINear regression analyses, and therefore, it is referred to as the HarLin model. It isolates first the periodicity from the daily averages of records and then the linear regression analysis is applied elegantly to first order stationary data. The results are tested and compared with the Adaptive-Neuro Fuzzy Inference System (ANFIS) model based on the Sugeno fuzzy logic inference system and Angström–Prescott model in the form of a linear regression analysis. In the application, three solar irradiation sites are considered from different solar energy potential locations in Turkey, namely, at Adana, Gaziantep and Silifke cities. The predictions by the HarLin model appear more successful than ANFIS and the classical Angström–Prescott approaches.

A comparison of Mamdani and Sugeno fuzzy based packet scheduler for MANET with a realistic wireless propagation model

The mobile nature of the nodes in a wireless mobile ad-hoc network (MANET) and the error prone link connectivity between nodes pose many challenges. These include frequent route changes, high packet loss, etc. Such problems increase the end-to-end delay and decrease the throughput. This paper proposes two adaptive priority packet scheduling algorithms for MANET based on Mamdani and Sugeno fuzzy inference system. The fuzzy systems consist of three input variables: data rate, signal-to-noise ratio (SNR) and queue size. The fuzzy decision system has been optimised to improve its efficiency. Both fuzzy systems were verified using the Matlab fuzzy toolbox and the performance of both algorithms were evaluated using the riverbed modeler (formally known as OPNET modeler). The results were compared to an existing fuzzy scheduler under various network loads, for constant-bit-rate (CBR) and variable-bit-rate (VBR) traffic. The measuring metrics which form the basis for performance evaluation are end-to-end delay, throughput and packet delivery ratio. The proposed Mamdani and Sugeno scheduler perform better than the existing scheduler for CBR traffic. The end-to-end delay for Mamdani and Sugeno scheduler was reduced by an average of 52% and 54%, respectively. The performance of the throughput and packet delivery ratio for CBR traffic are very similar to the existing scheduler because of the characteristic of the traffic. The network was also at full capacity. The proposed schedulers also showed a better performance for VBR traffic. The end-to-end delay was reduced by an average of 38% and 52%, respectively. Both the throughput and packet delivery ratio (PDR) increased by an average of 53% and 47%, respectively. The Mamdani scheduler is more computationally complex than the Sugeno scheduler, even though they both showed similar network performance. Thus, the Sugeno scheduler is more suitable for real-time applications.