A fuzzy sensitivity analysis approach to estimate brain effective connectivity and its application to epileptic seizure detection

The brain is considered to be the most complicated organ in human body. Inferring and quantification of effective (causal) connectivity among regions of the brain is an important step in characterization of its complicated functions. The proposed method is comprised of modeling multivariate time series with Adaptive Neurofuzzy Inference System (ANFIS) and carrying out a sensitivity analysis using Fuzzy network parameters as a new approach to introduce a connectivity measure for detecting causal interactions between interactive input time series. The results of simulations indicate that this method is successful in detecting causal connectivity. After validating the performance of the proposed method on synthetic linear and nonlinear interconnected time series, it is applied to epileptic intracranial Electroencephalography (EEG) signals. The result of applying the proposed method on Freiburg epileptic intracranial EEG data recorded during seizure shows that the proposed method is capable of discriminating between the seizure and non-seizure states of the brain.

Improved Estimation of Bio-Oil Yield Based on Pyrolysis Conditions and Biomass Compositions Using GA- and PSO-ANFIS Models

This paper incorporates the adaptive neurofuzzy inference system (ANFIS) technique to model the yield of bio-oil. The estimation of this parameter was performed according to pyrolysis conditions and biomass compositions of feedstock. For this purpose, this paper innovates two optimization methods including a genetic algorithm (GA) and particle swarm optimization (PSO). Primary data were gathered from previous studies and included 244 data of biodiesel oils. The findings showed a coefficient determination ( R 2 ) of 0.937 and RMSE of 2.1053 for the GA-ANFIS model, and a coefficient determination ( R 2 ) of 0.968 and RMSE of 1.4443 for PSO-ANFIS. This study indicates the capability of the PSO-ANFIS algorithm in the estimation of the bio-oil yield. According to the performed analysis, this model shows a higher ability than the previously presented models in predicting the target values and can be a suitable alternative to time-consuming and difficult experimental tests.

Prediction of the Loss of Feed Water Fault Signatures Using Machine Learning Techniques

Fault diagnosis occurrence and its precise prediction in nuclear power plants are extremely important in avoiding disastrous consequences. The inherent limitations of the current fault diagnosis methods make machine learning techniques and their hybrid methodologies possible solutions to remedy this challenge. This study sought to develop, examine, compare, and contrast three robust machine learning methodologies of adaptive neurofuzzy inference system, long short-term memory, and radial basis function network by modeling the loss of feed water event using RELAP5. The performance indices of residual plots, mean absolute percentage error, root mean squared error, and coefficient of determination were used to determine the most suitable algorithms for accurately diagnosing the loss of feed water transient signatures. The study found out that the adaptive neurofuzzy inference system model outperformed the other schemes when predicting the temperature of the steam generator tubes, the radial basis function network scheme was best suited in forecasting the mass flow rate at the core inlet, while the long short-term memory algorithm was best suited for the estimation of the severities of the loss of the feed water fault.

Vibration signal diagnosis and analysis of rotating machine by utilizing cloud computing

Vibration signal diagnosis and analysis plays an important role in the industrial machinery since it enhances the machinery performance under supervision. The information regarding the future condition is given by vibration diagnosis techniques which is growing interest for the scientific and industrial communities. Information for failure diagnostic and prediction are provided by the motor vibration through signal processing. The development of mechanical systems fault prognosis and in the last decades, research is done at a very rapid rate. The examination of vibration signal monitoring is done in this paper with the aid of Cyber-Physical Systems (CPS) and Cloud Technology (CT). The machines maintenance strategies are implemented by using the data collected from machines which are based on the fault prognosis. The cloud computing platform is presented in this paper which is having three layers and the unlabelled data is received to generate an interpreted online decision. Feature extraction of the vibration signal is obtained in terms of range, mean value, root mean square value, and standard deviation and crest values. The performance of the model is evaluated by utilizing the classical statistical metrics such as RMSE Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) of the vibration signal. It is obtained that the proposed technique is 25% and 90% better than the Adaptive Neurofuzzy Inference System and the Single Modeling System respectively in terms of RMSE. The performance in terms of MAPE, then the proposed technique outperforms the existing Adaptive Neurofuzzy Inference System and the Single Modeling System by 8 % and 60% respectively. The presented technique is better than the existing Adaptive Neurofuzzy Inference System and the Single Modeling techniques by average of 15% and 30 % respectively.

Application of Adaptive Neurofuzzy Control in the Field of Credit Insurance

Credit insurance is of vital importance for the trade sector and almost every related business. Moreover, every policy in credit insurance is tailor-made in order to suit in the best available way the unique needs and demands of the insured business. Thus, pricing of such service can be tricky for an insurance company. In the present chapter, this pricing problem in the field of credit insurance will be addressed through the use of intelligent control mechanisms. More specifically, a way of calculating the price of insurance policies that has to be paid by a prospective client of an insurance company will be suggested. The model will be created and implemented with the use of fuzzy logic, and more specifically, through the implementation of an adaptive neurofuzzy inference system. The training data that will be used for the tuning of the system will be derived from real anonymous insurance policies of the Greek insurance market.

Prediction of Oilfield-Increased Production Using Adaptive Neurofuzzy Inference System with Smoothing Treatment

A novel modified adaptive neurofuzzy inference system with smoothing treatment (MANFIS) is proposed. The MANFIS model considered the smoothing treatment of initial data basing on the adaptive neurofuzzy inference system, and we used it to predict oilfield-increased production under the well stimulation. Numerical experiments show the prediction result of the novel considering smoothing treatment is better than that without smoothing treatment. This study provides a novel and feasible method for prediction of oilfield-increased production under well stimulation, and it can be helpful in the further study of oilfield development measure planning.

A leakage current estimation based on thermal image of polymer insulator

Polymer insulators tend to fail because of the climatic and environmental conditions. The failure occurs when the surface of insulator is contaminated by sea salt or cement dust which lead to partial discharge (PD). Leakage currents will increase by PD that causes deterioration of insulation. To predict the insulation failures, an adaptive neurofuzzy inference system (ANFIS) method using initial color detection processes are proposed to estimate the leakage currents based on the polymer insulator thermal images (infrared signature). In this study, the sodium chloride and kaolin are used as pollutants of the polymer insulator according to IEC 60507 standards. Then, the insulator is tested in the laboratory using AC high voltage applied at 18 kV where the temperature detection is controlled at 26° C and 70% RH (relative humidity). The percentage of colors (Red, Yellow, and Blue) from the thermal image is measured using the color detection method. Correspond to the color percentage, the ANFIS method predicts leakage currents from polymer insulators. Furthermore, this system interprets measured data from insulators that need to be categorized as Safe, Need Maintenance or Harmful. The final application of the system can be a non-contact tool to predict the polymer insulators used by technicians in the field.

Prediction of Oscillatory Heat Transfer Coefficient in Heat Exchangers of Thermo-Acoustic Systems

The characterisation of heat transfer in oscillatory flow of thermo-acoustic based heat exchangers is a cumbersome issue. This is due to the nature of the heat transfer between the gas particles moving along the device at high amplitude and the solid surface of the heat exchangers. In addition, the change in velocity, pressure and temperature induces nonlinear effect. As a result, the performance of heat exchangers negatively affects the efficiency of thermo-acoustic systems. Hence, it is necessary to determine to oscillatory heat transfer coefficient in order to measure the performance of heat exchangers in thermo-acoustic systems. Although it is possible to conduct experimental investigation or perform numerical analysis in order to determine oscillatory heat transfer coefficient, the former requires costly time consuming experiment while the latter involves the resolution of complex mathematical models. In this paper, an improved adaptive neurofuzzy inference system and artificial neural network trained by particle swarm optimization are proposed to predict oscillatory heat transfer coefficient. This paper is intending to provide clarity on the benefits of these new approaches on the computation of geometrical configuration and the working parameters of heat exchangers in thermo-acoustic systems.