The design of a neuro-fuzzy system based on a radial basis function (RBF) network architecture and using support vector learning is considered. Typically, a neuro-fuzzy model structure is created from numerical data, however the common modeling techniques may introduce unnecessary redundancy into the rule base. It is of great interest to reduce the number of fuzzy rules. The proposed method pro...

An irregular activity on electric power distribution feeder, which does not draw adequate fault current to be detected by general protective devices, is called as High impedance fault (HIF). This paper presents the algorithm for HIF detection based on the amplitude of third and fifth harmonics of current, voltage and power. This paper proposes an intelligent algorithm using the Takagi SugenoKan...

In this paper the superheating system of a 325MW steam power generating plant is modeled by usage of recurrent neuro-fuzzy networks and subtractive clustering. The experimental data are obtained from a complete set of field experiments under various operating conditions. Neuro-fuzzy models are constructed for each subsystem of the superheating unit. The nine fuzzy models are then constructed in...

An adaptive neuro-fuzzy inference system (ANFIS) was developed using the subtractive clustering technique to study the air demand in low-level outlet works. The ANFIS model was employed to calculate vent air discharge in different gate openings for an embankment dam. A hybrid learning algorithm obtained from combining back-propagation and least square estimate was adopted to identify linear and...

In this paper, the hydro power plant model (with penstock-wall elasticity and compressible water column effect) is simulated at random load disturbance variation with output as turbine speed for random gate position as input. The multilayer perceptron neural network (i.e. NNARX) and fused neural network and fuzzy inference system (i.e. ANFIS) for identification of turbine speed as output variab...

The present work proposes a new approach in modelling the biological wastewater treatment plant: The anaerobic digestion by multiple models. Although a mathematical model already exists, but the interaction between the various variables is strongly not linear and is modelled in the form of rather complex algébro-differential equations. So, by using the method of subtractive clustering, the mult...

In this study, we apply a non-negative matrix factorization approach for the extraction and detection of concepts or topics from electronic mail messages. For the publicly released Enron electronic mail collection, we encode sparse term-by-message matrices and use a low rank non-negative matrix factorization algorithm to preserve natural data non-negativity and avoid subtractive basis vector an...

Composite Strain Encoding (C-SENC) is an Magnetic Resonance Imaging (MRI) technique for acquiring simultaneous viability and functional and images of the heart. It combines two imaging techniques, Delayed Enhancement (DE) and Strain Encoding (SENC). In this work, a novel multi-stage method is proposed to identify ventricular infarction in the functional and viability images provided by C-SENC M...

In this research, input/output data of a MIMO nonlinear system are used to create intelligent models. Multi layer perceprtrons and neuro-fuzzy networks are utilized for this purpose. For the purpose that these models suit predictive control in their best, a variety of subtle points should be considered. Recurrent models and subtractive clustering are used in this research, and a pre-processing ...

In this paper, a fuzzy model predictive control (FMPC) strategy is proposed to regulate the output variables of a coagulation hemical dosing unit. A multiple-input, multiple-output (MIMO) process model in form of a linearised Takagi–Sugeno (T–S) fuzzy odel is derived. The process model is obtained through subtractive clustering from the plant’s data set. The MIMO model is escribed by a set of c...