In this study, we evaluated the predictive performance of an adaptiveneuro-fuzzyinferencesystem (ANFIS) with six di ﬀerent membership functions (MFs). Using a geographic information system (GIS), we applied ANFIS to landsubsidencesusceptibilitymapping (LSSM) in the study area of Amol County, northern Iran. As a novelty, we derived a landsubsidence inventory from the di ﬀerential synthetic aperture radar interferometry (DInSAR) of two Sentinel-1 images. We used 70% of surface subsidence deformation areas for training, while 30% were reserved for testing and validation. We then investigated regions that are susceptible to subsidence via the ANFIS method and evaluated the resulting prediction maps using receiver operating characteristics (ROC) curves. Out of the six di ﬀerent versions, the most accurate map was generated with a Gaussian membership function, yielding an accuracy of 84%.
tried to develop more and more accurate methods to map susceptible areas, either through combining different methods such as combining AHP and fuzzy logic (Feizizadeh et al. 2013), using factor score matrix with AHP that called modified AHP (M-AHP) (Nefeslioglu et al. 2013), integrating interval pairwise comparison matrices with the GIS-MCDAs (Feizizadeh and Ghorbanzadeh, 2017) and combining ensemble frequency ratio with logistic regression models (Umar et al. 2014) and analytic network process (ANP) (Pirnazar et al. 2017). There are also several attempts to develop new methods such as the methodology of using the global sensitivity analysis for GIS-MCDAs developed by Ligmann-Zielinska and Jankowski (2012) (Ghorbanzadeh et al. 2017). Our research utilizes the ANFIS by applying the Takagi–Sugeno rule format. This format is a combination of optimized premise MFs with an optimized consequent equation (Bui et al. 2012). ANFIS is an inferencesystem with a high capacity (Sezer et al. 2011). It has also some advantages compared to other suscepti- bility mapping methods such as expert-knowledge-basedGIS-MCDA. GIS-MCDA is sometimes criticized for the expert knowledge to be a major source of uncertainty among the results (S¸alap-Ayc¸a and Jankowski 2016; Feizizadeh and Kienberger 2017; Erlacher et al. 2017; Feizizadeh and Ghorbanzadeh 2017; Cabrera-Barona and Ghorbanzadeh 2018) or ordinary neural networks because they use if–then rules (Bardestani et al. 2017). The pro- posed method does not apply expert opinions at any stage. In addition, it provides the possibility of using a variety of fuzzy MFs. This method is also known for fast convergence times (Pandey and Sinha 2015). By using the input and target data, ANFIS can provide a fuzzyinference structure (Cakıt and Karwowski 2017) in which the MF parameters use hybrid learning algorithms to adjust themselves (Bui et al. 2012). The main differences between the literature discussed above and our study are that we used
Abstract. Groundwater is one of the most valuable natu- ral resources in the world (Jha et al., 2007). However, it is not an unlimited resource; therefore understanding ground- water potential is crucial to ensure its sustainable use. The aim of the current study is to propose and verify new artifi- cial intelligence methods for the spatial prediction of ground- water spring potential mapping at the Koohdasht–Nourabad plain, Lorestan province, Iran. These methods are new hy- brids of an adaptiveneuro-fuzzyinferencesystem (ANFIS) and five metaheuristic algorithms, namely invasive weed op- timization (IWO), differential evolution (DE), firefly algo- rithm (FA), particle swarm optimization (PSO), and the bees algorithm (BA). A total of 2463 spring locations were iden- tified and collected, and then divided randomly into two sub- sets: 70 % (1725 locations) were used for training models and the remaining 30 % (738 spring locations) were utilized for evaluating the models. A total of 13 groundwater con- ditioning factors were prepared for modeling, namely the slope degree, slope aspect, altitude, plan curvature, stream power index (SPI), topographic wetness index (TWI), ter- rain roughness index (TRI), distance from fault, distance from river, land use/land cover, rainfall, soil order, and lithol- ogy. In the next step, the step-wise assessment ratio analysis (SWARA) method was applied to quantify the degree of rele- vance of these groundwater conditioning factors. The global performance of these derived models was assessed using the area under the curve (AUC). In addition, the Friedman and Wilcoxon signed-rank tests were carried out to check and
to generate GIS-based flood susceptibility maps. Today, powerful machine learning methods, such as adaptiveneuro- fuzzy (Mukerji et al., 2009), genetic algorithm (Chau et al., 2005), decision tree (DT), and support vector machine (SVM) (Adeli & Hung, 1994) have replaced traditional methods. Many of these methods have been rarely used in flood modeling, while they are highly able to cope with other natural disasters such as landslides (Pradhan, 2013 ; Yilmaz, 2010 ; Tien Bui et al., 2019). The mentioned models, when applied alone, have weaknesses and limitations in modeling. For instance, ANFIS (or other similar data mining and machine learning methods) has to deal with inconsistent input values. It also should cope with input error values due to the type of inputs in which the weight of each class of criteria is estimated through this method. As the case study, the Jahrom town is suffering from extreme seasonal floods that have always damaged the city. The purpose of this study is to prepare a flood susceptibility map for the Jahrom town by an ensemble of frequency ratio and adaptiveneuro-fuzzyinferencesystem.
plied a two-dimensional (2D) look-up table of power coefficient and power mapping method to estimate the wind speed, but the technique needed huge memory space and suboptimum solution was often caused by the inherent slow searching mechanism. H. Li et al.  used the Artificial Neuro Network (ANN) to establish a sen- sorless wind speed estimator but the neuro-network is easy to be over-fitting or fall into the local minima. V. Calderaro et al. [7,8] combined the advantages of T-S fuzzysystem , Genetic Algorithms (GA) and Fuzzy C-Means clustering (FCM). Then, an adaptive optimum setting strategy was realized. However, the GA was un- stableness and also time- consuming to train the parame- ters of T-S fuzzysystem.
AdaptiveNeuroFuzzyInferenceSystem (ANFIS) which is an integration of neural networks' features and fuzzy logic has the potential to capture the benefits of both fields in a single framework. The ANFIS utilizes linguistic information from the fuzzy logic as well as the learning capabili- ties of an artificial neural network(Czogala and Leski, 2000; Ehret et al., 2011; Snehal and Sandeep, 2014). An ANFIS is a kind of artificial neural network that is based on Takagi–Sugeno fuzzyinferencesystem. It is considered generally as a multilayer feed forward adaptive network, where each node performs a particular function with its cor- responding input parameter set.
II. ADAPTIVE NEUROFUZZY INFERENCESYSTEM The adaptiveneurofuzzyinferencesystem (ANFIS), first introduced by Jang (1993), is a universal approximator and as such is capable of approximating any real continuous function on a compact set to any degree of accuracy (Jang et al. 1997). ANFIS is functionally equivalent to fuzzyinference systems (Jang et al. 1997). Specifically the ANFIS system of interest here is functionally equivalent to the Sugeno first-order fuzzy model (Jang et al. 1997; Drake 2000). Below, the hybrid learning algorithm, which combines gradient descent and the least-squares method, is introduced. As a simple example we assume a fuzzyinferencesystem with two inputs x and y and one output z. The first-order Sugeno fuzzy model, a typical rule set with two fuzzy If–Then rules can be expressed as
The hydrologic behavior of rainfall-runoff process is very complicated phenomenon which is controlled by large number of climatic and physiographic factors that vary with both the time and space. The relationship between rainfall and resulting runoff is quite complex and is influenced by factors relating the topography and climate. In recent years, artificial neural network (ANN), fuzzy logic, genetic algorithm and chaos theory have been widely applied in the sphere of hydrology and water resource. ANN have been recently accepted as an efficient alternative tool for modeling of complex hydrologic systems and widely used for prediction. Some specific applications of ANN to hydrology include modeling rainfall-runoff process. Fuzzy logic method was first developed to explain the human thinking and decision system by . Several studies have been carried out usingfuzzy logic in hydrology and water resources planning . Adaptiveneuro- fuzzyinferencesystem (ANFIS) which is integration of neural networks and fuzzy logic has the potential to capture the benefits of both these fields in a single framework. ANFIS utilizes linguistic information from the fuzzy logic as well learning capability of an ANN. Adaptiveneurofuzzyinferencesystem (ANFIS) is a fuzzymapping algorithm that is based on Tagaki-Sugeno-Kang (TSK) fuzzyinferencesystem  and . ANFIS used for many applications such as, database management, system
SRM working is based on switched reluctance principle. The reluctance value in SRM is varied from unaligned position to aligned position. In aligned position the air gap as well as reluctance values are low. Since the reluctance is directly proportional to length of air gap. In unaligned position the air gap as well as reluctance is more. Reluctance is nothing but opposition to creation of flux. Hence the rotor poles are attracted by stator poles thereby movement of the poles taking place from unaligned position to aligned position. This is done when excitation of rotor poles occurred. For proper working of SRM rotor position details are essential. Rotor position information is sensed by sensors. But the usage of sensors has certain limitations such as occupies more space, electrical and mechanical loose contacts, deposit of dust particles and costly. So the sensors are replaced by sensorless methods such as fuzzy logic, artificial neural networks and ANFIS. Moreover SRM has salient poles in the stator and rotor. This introduces the nonlinear nature of SRM. The nonlinear characteristics can be easily analyzed by soft computing techniques. The inputs of the soft computing techniques are current and flux linkage, which are derived from the terminals of SRM stator. Using these values rotor position has been estimated by soft computing techniques. The rotor position information is fed to a processor. Based on the rotor position information firing pulses are produced for triggering of power electronic devices of a converter. The converter output excites the phase winding of the SRM. The same process repeated for other phases. The voltage equation for SRM is given in eq. (1).
As traditional authentication methods fall short of security and usability, implicit authentication (IA) is gaining attention as a complementary authentication method which is capable of authenticating a legitimate user and detecting imposters transparently and continuously without the explicit involve- ment of the user. According to a survey , IA is widely accepted, and 73% of the participants consider IA more secure. Recent researches – demonstrate the use of behavioural- based IA to provide continuous protection on mobile devices. Although the results are promising, maintaining a balance between accuracy, adaptiveness and practical feasibility is still an unresolved challenge. Hence, an ANFIS-based implicit authentication system is presented in this paper. By utilizing its self-learning capabilities, the ANFIS-basedsystem is able to ascertain the unique behavioural pattern for every single user. Based on the perceived pattern, the ANFIS-basedsystem is able to provide continuous and transparent authentication in real-time. Our proposed ANFIS-based IA system incorporates a time-window based profiling approach in order to enable granular and continuous re-authentication which also ensures timely adversary detection. The proposed system also utilizes anomaly based scoring, which together with an adaptively computed reference, provides real-time inputs for the ANFIS system. The anomaly based scoring system applies a ranking algorithm that maintains a sorted list of the most relevant events that are used in the computation of the anomaly-based scores.
The FEM is one of the most popular numerical methods used for computer simulation. The key advantage of the FEM over other numerical methods in engineering applications is the ability to handle nonlinear, time- dependent and circular geometry problems. Therefore, this method is suitable for solving the problem involving magnetic field effects around the transmission line caused by circular cross-section of high voltage conductors. By literature, these research works are conducted based on electromagnetic theory or image theory , . With defining a line of calculation and assuming very thin power lines, two- dimensional problems of magnetic field analysis governed by empirical mathematical expressions can be applied. However, these conventional methods are unable to include effects of bundled conductors that are typical for EHV power transmission systems. To provide a potential tool of simulation, the FEM is flexible and suitable to estimate magnetic field distribution. As mentioned where a normal steady-state operation is assumed, the current does not suddenly change its value.
Load forecasting plays important tasks in power system planning, operation and control. It has received an increasing attention over the years by academic researchers and practitioners. Control, security assessment, optimum planning of power production required a precise medium term load forecasting. Electric load forecasting is a real-life problem in industry. Electricity supplier’s use forecasting models to predict the load demand of their customers to increase/decrease the power generated and to minimize the operating costs of producing electricity. In addition to the conventional classical models, several models based on artificial intelligence have been proposed in the literature, in particular, neural network for their good performance. Other nonparametric approaches of artificial intelligence have also been applied. Nevertheless, all these models are inaccurate when used in real time operation. This paper presents the novel techniques of usingAdaptiveNeuroFuzzy Interference System for prediction of hourly load power system data which combines neural network and fuzzy logic to predict future load. ANFIS model is constructed using one complete year load data from TATA POWER Company, Mumbai, applying Genfis 2 and Genfis 3 to train & test the data. The RMSE, MAPE & SD are used as Performance indices to evaluate the model.
This work attempted to test the hypothesis that AI techniques can significantly help in forecasting future readings by minimizing the maximum prediction error rates and reducing the uncertainties involved by handling input data in a more simple and efficient manner. The main aim was to use ANFIS to predict the groundwater recharge using some easily measured parameters such as temperature, rainfall, relative humidity and sunshine radiation. Other objectives included investigating whether or not the number of input parameters have a direct effect on the results. Additionally, to investigate the effect of different membership function types on the results. The benefits and merits of using ANFIS modelling for predicting groundwater recharge and the obtained results corresponding to our objectives are discussed in the following paragraphs.
The dynamics of robot manipulators are highly nonlinear with strong couplings existing between joints and are frequently subjected to structured and unstructured uncertainties. Fuzzy Logic Controller can very well describe the desired system behavior with simple “if-then” relations owing the designer to derive “if-then” rules manually by trial and error. On the other hand, Neural Networks perform function approximation of a system but cannot interpret the solution obtained neither check if its solution is plausible. The two approaches are complementary. Combining them, Neural Networks will allow learning capability while Fuzzy-Logic will bring knowledge representation (Neuro-Fuzzy). This paper presents the control of six degrees of freedom robot arm (PUMA Robot) usingAdaptiveNeuroFuzzyInferenceSystem (ANFIS) based PD plus I controller. Numerical simulation using the dynamic model of six DOF robot arm shows the effectiveness of the approach in trajectory tracking problems. Comparative evaluation with respect to PID, Fuzzy PD+I controls are presented to validate the controller design. The results presented emphasize that a satisfactory tracking precision could be achieved using ANFIS controller than PID and Fuzzy PD+I controllers
There are various linearization techniques utilized as part of an attempt to lessen the non-linearity of the thermistor signal conditioning circuits (SCC). They can be hardware or software based methods of linearization. A variety of analog signal conditioning methods [3-16] have been implemented for the negative temperature coefficient (NTC) thermistors to produce linear or quasi-linear temperature-resistance relation. These methods comprise of different resistor combinations attached to the thermistor sensor which is an integral part of a SCC [3-4]. They produce linear or quasi-linear temperature-resistance change over a narrow range of temperature but with reduced sensitivity. Some of the methods used for linearization connect the thermistor into SCCs having timing circuitry [5-6], log and antilog circuits [7-8], dividers , analog multipliers [10-11], analog to digital
Neural networks are a form of multiprocessor computer system, with simple processing elements, a high de- gree of interconnection, adaptive interaction between elements; it is also referred as an “artificial” neural net- work (ANN). According to Dr. Robert Hecht-Nielsen, a neural network is “...a computing system made up of a number of simple, highly interconnected processing elements, which process information by their dynamic state response to external inputs”. There are many different kinds of learning rules used by neural networks. ANNs can learn from data and feedback and have learning capabilities. On the other hand, fuzzy logic models are rule-based models and do not have learning capabilities, therefore so for learning, fuzzyinferencesystem per- forms the following operations:
Applied the combination of ANFIS to classify the lymphoma microarray data set. The first choose some important genes using a feature importance ranking scheme. This ranking Scheme are based on the t- score (TS) is a t-statistic between the centroid of a specific class and the overall centroid of all the classes. Another possible model for TS could be a t-statistic between the centroid of a specific class and the centroid of all the other classes. First added the selected 50 genes one by one to the network according to their TS ranks starting with the gene ranked. At first used only a single gene that is ranked 1 as the input to the network. Then trained the network with the training data set, and subsequently tested the network with the test data set then repeat the process so on.
Reservoir water release decision is one of the critical actions in determining the quantity of water to be retained or released from the reservoir. Typically, the decision is influenced by the reservoir inflow that can be estimated based on the rainfall recorded at the reservoir’s upstream areas. Since the rainfall is recorded at several different locations, the use of temporal pattern alone may not be appropriate. Hence, in this study a spatial temporal pattern was used to retain the spatial information of the rainfall’s location. In addition, rainfall recorded at different locations may cause fuzziness in the data representation. Therefore, a hybrid computational intelligence approach, namely the AdaptiveNeuroFuzzyInferenceSystem (ANFIS), was used to develop a reservoir water release decision model. ANFIS integrates both the neural network and fuzzy logic principles in order to deal with the fuzziness and complexity of the spatial temporal pattern of rainfall. In this study, the Timah Tasoh reservoir and rainfall from five upstream gauging stations were used as a case study. Two ANFIS models were developed and their performances were compared based on the lowest square error achieved from the simulation conducted. Both models utilized the spatial temporal pattern of the rainfall as input. The first model considered the current reservoir water level as an additional input, while the second model retained the existing input. The result indicated that the application of ANFIS could be used successfully for modeling reservoir water release decision. The first model with the additional input showed better performance with the lowest square error compared to the second model.
– Today most organizations have discovered that advanced trainings can be considered as the key asset for modern organizations. This study presents a forecasting model that predicts intangible assets on the basis of innovation performance in organizational training using widely applied innovation criteria. The research focused on criteria, such as organization culture, ability to respond to organizational technology changes, relationship with other organizations in the training process and the use of high technology in education. The adaptiveneuro-fuzzyinference systems (ANFIS) approach has been used to verify the proposed model. It is possible to predict innovation performance and it can also adjust allocated resources to organizational training system for its innovation objectives with this method. Originality/value – A great deal of work has been published over the past decade on the application of neural networks in diverse fields. This paper presents a model that measure and forecasts the intangible assets by the development of an Adaptive Neural Network with FuzzyInferencesystem (ANFIS), using data that concern human capital, organizational support and innovativeness. The results indicate that fuzzy neural networks could be an efficient system that is easy to apply in order to accurately measure and forecast the intangible assets by measuring innovation capabilities of an organization or firm.
Large consumers of electric power need more than one source additionally to the utility grid for supplying their load demands. In many cases, the large consumers of electrical power combined between several types of power sources, which called hybrid power system. Hybrid power systems achieve the integrity of power generation sources. The main reasons of using HPSs are frequently interruptions of the utility grid and random load demands, in addition to the high generation budget of the traditional power sources. The high generation budget of the traditional power sources are high fuel cost and the periodicity maintenance requirements cost.