Platt and Platt , and Cheng et al.  used a Logitmodel to analyze pre-warning model and to a build financialdistressmodel, while Zhang et al. , and O’leary  used artificial neural networks. One problem of the logistic regression is that serial correlation might exist in the explanatory variables. Another problem is the inconsistency generated from the errors on construction of the dummy variable indicating the financial stage, distress or stability periods, crisis or no crisis periods, leading to misclassification of time points. With ANFIS approach we do not face these problems, which are very usual in conventional econometric modelling. A significant study was made by Cheng et al. . The authors study a pre-warning financialdistressmodel for the TSE listed companies and they apply a binary logit and a fuzzy regression model with triangular membership function. Their results support fuzzy regression, where the correctly predicted percentage of fuzzy regression is 90.98 percent versus logit regression which predicts correctly the 90.30 percent. In this paper we expand this approach by taking panel data as we have a group of companies. Because we have various companies among time periods we need to examine logistic regressions through panel data analysis and to investigate if random or fixed effects are more appropriate. With this approach we show that the overall percentage, and especially the correct percentage of financialdistressperiods, of panelLogitmodel is significant higher to simple binary Logitmodel without panel data analysis examined by Cheng et al. . Additionally, we propose ANFIS because the overall correct classified percentage of financialdistress and stability periods is significant superior to Logit and fuzzy regressions.
One of the advantages and features of Logit and Probit models is the non-linearity of the effect of the right-hand side variables on the left-hand side variable. This is an attractive property of the model because the literature on currency crises has demonstrated that such effects are at work (Eichengreen and Rose 1998; Demirguc-Kunt and Detragrache 1998). Furthermore, there is complete freedom to choose which variables enter into the vector X t , with the selection being based on various theories in the financial and economic crisis literature. Another positive point about the Probit and Logit approach is that it summarizes all the information in one easily interpretable number – the probability of a crisis occurring. In addition, Probit and Logit models consider all variables together and look only at the marginal contributions of each indicator; the discrete choice models disregard variables that do not contribute information that is not already captured in the other variables. Lastly, Probit and Logit lend themselves readily to statistical testing.
The purpose of this paper is to present a neuro-fuzzy approach of financialdistress pre- warning model appropriate for risk supervisors, investors and policy makers. We examine a sample of the financial institutions and electronic companies of Taiwan Security Exchange (TSE) from 2002 through 2008. We present an adaptiveneuro-fuzzysystem with triangle and Gaussian membership functions. We conclude that neuro-fuzzymodel presents almost perfect forecasts for financialdistressperiods as also very high forecasting performance for financial stability periods, indicating that ANFIS technology is more appropriate for financial credit risk control and management and for the forecasting of bankruptcy and distressperiods. On the other hand we propose the use of both models, because with Logit and generally with discrete choice models we can examine and investigate the effects of the inputs or the independent variables, while we can simultaneously use ANFIS for forecasting purposes. The wise and the most scientific option are to combine both models and not taking only one of them.
Hepatitis B is one of the liver diseases that is difficult to discover at an early stage of its attack and prominent public health problem. As at 2017, medical statistic recorded that over 23 million of Nigerians were living with Hepatitis B. Several decision support systems used in diagnosing liver diseases derived their efficiencies from artificial intelligence techniques in tackling the challenges facing physician in respect to complexity of the numerous variables involved in liver diseases diagnosis. In this paper, AdaptiveNeuro-FuzzyInferenceSystem (ANFIS) was employed to invoke neural network that provided structures for fuzzyinference engine (FIE) in order to learn information about the normalized dataset on hepatitis B. The neural network (NN) triggers backpropagation and least square methods for tuning the membership functions at the fuzzification stage while the center of area (COA) was used as defuzzification method to compute the weighted average of the fuzzy set and intensity level of the disease for each record. The system was implemented with technical computing language, MATHLAB, on a dataset that consists of 155 instances and 20 attributes of which only the most five liver function tests (LFTs) attributes were selected as input parameters and the corresponding linguistic values and intensity levels were generated as output in order to identify the severity level of the infection. After the system was evaluated, the performance metric gave accuracy of 90.2%.
Here are some more reasons you might want a solar power system, despite the inefficiencies and costs involved. Two or more reasons might be related, such as green energy and independent living. No problems choose the relevant ones and decide if they give you the value you want. Perhaps you live in an area where commercial electrical power is frequently interrupted by storms or other events A solar power system can act like a large uninterruptable power supply (UPS) for your house. You might not even know when the grid power fails unless you're using a high-power appliance at the time. Your computer and TV will stay on, and the lights might not even flicker. Any solar power system will lower your electric bill, but a grid-tied system will lower it more than a non-grid-tied system.
The main motivation for applying a Neuro- fuzzy computing approach is that it combines the generalization capabilities of Neural Networks with the ease of interpretability and high expressive power of fuzzy rules in an effective way. Vibration signals were obtained from the Feed Pump using Tri-axial Accelerometer and FFT. These signals were processed in MATLAB using ANFIS tool for training, testing and checking to simulate the Feed Pump. The performance criterion of the ANFIS classifier was evaluated using confusion matrix. The total classification accuracy of 95% obtained, proves the validation of the Feed Pump model. Neuro-Fuzzy Systems have high potential in diagnosis of machinery. The proposed ANFIS model has been found to be an effective tool for diagnosing faults.
two voltage-sourced converters (VSCs) using thyristors which operate from a common dc-circuit consisting of a dc-storage capacitor. The UPFC could be described as consisting of a parallel and a series branch. Each converter can independently generate or absorb reactive power. This arrangement enables free flow of active power in either direction between the ac-terminals of the two converters. The function of the parallel converter is to supply or absorb the active power demanded by the series branch. This converter is connected to the ac- terminal through a parallel-connected transformer. If required, it may also inject leading or lagging reactive power directly into the connection busbar. The second (series connected) converter provides the main function of the UPFC by injecting an ac-voltage with controllable magnitude and phase angle. The transmission line current flows through this voltage source, resulting in an active and reactive power exchange with the ac-system.
cordings have been obtained from PhysioNet apnea-ECG database. Wavelet transforms have been applied on the 1-minute and 3-minute length recordings. According to the preliminary tests, the variances of 10 th and 11 th detail components can be used as discriminative features for apneas. The features obtained from total 8 recordings have been used for training and testing of an adaptiveneurofuzzyinferencesystem (ANFIS). Training and testing process have been repeated by using the randomly obtained five different sequences of whole data for ge- neralization of the ANFIS. According to results, ANFIS based classification has sufficient accuracy for apnea detection considering of each type of respiratory. How- ever the best result has been obtained by analyzing the 3-minute length nasal based respiratory signal. In this study, classification accuracies have been obtained greater than 95.2% for each of the five sequences of en- tire data. Due to the results of the 1-minute based analy- sis, the classification accuracies of ANFIS have obtained between 80.6% - 81.5%, 89.2% - 90.9%, 90.8% - 92.9% and 88.6% - 90.4% respectively for the chest, nasal, ab- dominal respiratory and EDR signals. For the analysis of 3-minute length data, the classification accuracies have obtained between 84.8% - 86.5%, 95.2% - 96.5%, 93.4% - 95.4% and 92.0% - 94.0%, respectively. According to these results, both of the 1-minute and 3-minute length of chest, nasal, abdominal based respiratory and EDR sig- nals can be used sufficiently for proposed method. How- ever the best result can be obtained by analyzing the sec- tion of the 3-minute length nasal based respiratory Table 1. ANFIS based classification accuracies for the analysis of 1-minute length signal.
Furthermore, neural networks possess a variety of alternative features such as massive parallelism, distributed representation and computation, generalization ability, adaptability and inherent contextual information processing. On the other hand, fuzzy sets constitute the oldest and most reported soft computing paradigm. They are well suited to modelling different forms of uncertainties and ambiguities, often encountered in real life. The objective of the hybridization through ANFIS has been to overcome the weaknesses in one technology during its application, with the strengths of the other by appropriately integrating them.
Almost all mechanical properties of concrete could be estimated by the most important structural property of concrete, CS. Selected mechanical properties of HSC in this study include Splitting Tensile Strength (STS), Modulus of Elasticity (MOE) and CS that are essential in all type of design and evaluation of HSC structures. This study considers MOE and STS as input and CS as output in ANFIS model. More than 100 sets of experimental studies in the last 15 years has been collected from Giaccio and Zerbino(1998) , Jin-Kuen and Sang-Hun (1999) , Shannag (2000) , Ajdukiewicz and Kliszczewicz (2002) , Jin-Keun et al. (2004) , Bissonnette et al. (2007) , Almeida et al. (2008) , Pablo (2008) , Yin, J. et al. (2010) , K. M. Ng et al. (2010) , Ozbay et al. (2011) , Parra et al. (2011) , Das and Chatterjee (2012) , Ranaivomanana et al. (2013) . To include wide range of experimental data in the model, lower and upper limit of CS for HSC is selected 400 MPa and 1000 MPa respectively. Table 1, represents the range of collected experimental mechanical properties for HSC in this study.
The estimation of traffic noise under heterogeneous traffic is usually complex due to varying traffic flow, driving behavior which leads to irregular pattern of honking and other factors. At present traffic noise models are not available which considers honking as one of the parameter for traffic noise prediction. Literature reveals that honking has significant impact on traffic noise. In present study, honking has been considered for traffic noise prediction. The comparative study illustrate that model performs better than some of the popular noise models and its performance could further enhanced by incorporating honk equivalent of different vehicles for better noise pollution assessment and control.
analyses are much more difficult. This paper presents the use of neuro-fuzzy networks as means of implementing algorithms suitable for nonlinear black-box prediction and control. In engineering applications, two attractive tools have emerged recently. These two attractive tools are: the artificial neural networks and the fuzzy logic system. One area of particular importance is the design of networks capable of modeling and predicting the behavior of systems that involve complex, multi-variable processes. To illustrate the applicability of the neuro-fuzzy networks, a case study involving air-fuel ratio is presented here. Air-fuel ratio represents complex, nonlinear and stochastic behavior. To monitor the engine conditions, an adaptiveneuro-fuzzyinferencesystem (ANFIS) is used to capture the nonlinear connections between the air-fuel ratio and control parameters such manifold air pressure, throttle position, manifold air temperature, engine temperature, engine speed, and injection opening time. This paper describes a fuzzy clustering method to initialize the ANFIS.
The adaptive network-based fuzzyinference systems (ANFIS) is used to solve problems related to parameter identification. This parameter identification is done through a hybrid learning rule combining the back- propagation gradient descent and a least-squares method. ANFIS is basically a graphical network representation of Sugeno-type fuzzy systems endowed with the neural learning capabilities. The network is comprised of nodes with specific functions collected in layers. ANFIS is able to construct a network realization of IF / THEN rules. Consider a Sugeno type of fuzzysystem having the rule base
A suspension system is a mechanism which consist of spring and damping element connected between wheel and car body. The suspension plays an important role to control the vertical dynamics of car body. The performance and characteristics of suspension system mainly depends on ride comfort and stability control of vehicle . A better ride comfort can be achieved by using soft suspension, whereas better stability can be achieved with the help of hard suspension. The design of suspension involves optimization process where the elements are selected between soft and hard suspension. A suspension is normally classified into passive suspension, semi-active suspension and active suspension. Nowadays, lot of research works are going on [2-5] active suspension system because of its ability to operate wide range of frequency and forces. The performance of active suspension system obtained by measuring suspension travel and acceleration of vehicle body. Due to the development of microcontroller and computers [6-8], the real time implementation of active suspension can be done more effectively. The effect of ride comfort on suspension can be measured with the help of body acceleration of vehicle. Similarly, the performance of stability can be measured with the help of suspension travel.
Prediction will provide a powerful tool for managers to be more successful in the long and short term planning for their organization. Prediction can be done in two ways: it can either be the result of deduction and analysis of an expert in a given field of knowledge, or the analysis and evaluation of raw data and statistics. In this study we consider prediction using time series data. Time series show different trends in different cases. If we want to divide this behavior into two general categories, we can say that data have either linear or non-linear trend. The purpose of this study is to analyze Auto-Regressive Integrated Moving Average model and artificial neural network model in fuzzy systems. Then with comparing these models we can conclude whether the classical Auto-Regressive Integrated Moving Average Model has the same prediction power as the neuro-fuzzymodel or not. Classical model of moving average or Box-Jenkins model, have conventionally been used with data having linear trend. But in the real world there are fewer cases where data having linear trend or static state in average and variance, so recently more accurate methods of modeling non-linear systems are invented.
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
Abstract The main aim of this paper is to model the industrial power consumption in Nigeria with the AdaptiveNeuro-FuzzyInferenceSystem (ANFIS) model and then forecast the industrial power consumed for the next five years beyond the available data. About 45 years (1970 to 2015) dataset was obtained from the Central Bank of Nigeria (CBN), the National Bureau of Statistics (NBS) and other relevant organizations. The data includes population, rainfall, electricity connectivity and temperature which are the explanatory variables. Matlab was used along with the dataset to train and evaluate the ANFIS model which was then used to forecast the industrial power consumption in Nigeria for the years 2016 to 2020.The prediction performance of the ANFIS model was compared to those of Autoregressive Moving Average model and Moving Average model. From the result obtained, ANFIS gave R-square value of 0.9977 (99.77%), SSE value of 395.3674 and RMSE value of 2.9641. The regression coefficient of 99.77% shows that about 99.77% of the variations in the industrial power consumption in Nigeria for the years 1970 to 2015 are explained by the selected explanatory variables. The forecast result showed that the Nigerian industrial power consumption would be about 374.7 MW at the end of 2020 which is about 73.1% increase from the industrial power consumption in 2015. As such, based on the industrial power consumption in 2015, over 73% increment in power supply to the industrial sector will be required to satisfy the industrial sector’s power demand in 2020.
However, the pattern can be shifted from the pattern predefined by previous studies. A study shows Indonesia is facing the challenge of climate change that is significant enough to alter the intensity of the rain and have a negative impact on the economy, health and environment. So, we need an effective and appropriate strategies to address climate change . Climate change is a dynamic situation which several studies give different result eventhough the time between studies is short .Thus further studies are required in developing forecasting model that can adapt with the rapid climate change.