M.A. Ansari

A Survey: Soft Computing in Intelligent Information Retrieval Systems

This paper provides an in-depth survey of challenges in the design of intelligent information retrieval systems, pointing out some similarities and differences in the core data mining and web based search operations. The procedures for evaluation of search engine performance and implicit feedback of the user with respect to a search result are studied with references to the different algorithms. We have proposed a novel neural satisfaction based feedback vector in contrast to the existing activity pattern based feedback as a future research direction in Intelligent IR. We addressed the select research work in the area of soft information retrieval using fuzzy sets, artificial neural networks, genetic algorithms and probabilistic information retrieval. As an instance of information retrieval, web mining is a set of operations to retrieve relevant documents from preprocessed, crawled and indexed web, and it can be categorized into more specialized tasks of web content mining, web structure mining and web usage mining. We have given a survey of the important reviews on topic of web mining and its associated tasks. On the basis of Identified challenges in information retrieval in general, and web mining in particular, we have concentrated on applicability of soft computing techniques and their hybrids in web mining, the performance related issues of select solutions, future of web mining and the next generation users expectations from a search engine.

Wind speed and power prediction of prominent wind power potential states in India using GRNN

This paper introduces Generalized Regression Neural Network (GRNN) for long term wind speed prediction of major wind power potential states in India. The performance of proposed GRNN model is evaluated using the publicly available online dataset of National Aeronautics and Space Administration (NASA). Data samples of 26 cities are used for training the generalized regression neural network and remaining 5 cities data samples are used for testing purpose. Air temperature, earth temperature, relative humidity, daily solar radiation, elevation, latitude, heating degree days, cooling degree days, frost days, longitude and atmospheric pressure are used as input variables. Mean square error between measured and forecasted wind speed using training data samples and testing data samples are found to be 0.000042279 and 0.1543. Here it is important to impart that the proposed GRNN model is trained and tested with data samples of different geographical locations in order to make it feasible for wind speed prediction of any other location. Wind power of prominent wind power potential states in India are predicted by a variable pitch and speed control wind turbine G80-2MW.

Communication and load balancing using SCADA model based integrated substation

With the increase in automation, the industrial control systems are automated with computers. Computer control and monitor the industrial operations that exist in the physical world with virtual simulation tools. In this paper, we present a SCADA (supervisory control and data acquisition) based fully automated master station in the integrated substations environment. The idea is, to enable all the substations to communicate with master station and to balance the load through SCADA model. The scenarios we are taking into the consideration include automatic load balancing. Load is balanced in two modes namely, manual mode and automatic mode. This model helps the engineers and research scholar to test their system before the real implementation. This model for integrated substations drastically reduces the time latency if compared with conventional model. It reduces the time as well as human intervention in the critical tasks such as responding to faults, load balancing actions, etc. The proposed model is fast, scalable, flawless and robust.

Implementation of particle swarm optimization for dynamic economic load dispatch problem

The economic operation of the generating systems has always occupied an important position in the electric power industry. It is one of the complex problems of the power system. The aim of the dynamic economic load dispatch problem is to find the optimal combination of generators in order to minimize the operating costs of the system. The load demand must be appropriately shared among the various generating units of the system. This work is done by using the particle swarm optimization (PSO) algorithm. PSO is applied to search for the optimal schedule of all the generator units that can supply the required load demand at minimum fuel cost while satisfying all system constraints such as Generator constraints, Ramp rate limits, Transmission losses and valve point effect. The PSO method was developed through the simulation of a simplified social system. The simulations were performed over various test systems with 5 generation units. So by using PSO we have done the dynamic economic load dispatch thereby reducing the operating costs of the system.

A Flexible Scheme to Fault Detection for Electrical Assets Using Infrared Thermography

This paper approaches an infrared thermography methodology that can help to find fault and diagnosis for the electrical equipment. It uses noncontact and nondestructive technology. This technique is fast and reliable for inspecting the system without any interruption. In the field of electrical area maintenance, reliability of transmission and distribution system is one of the most critical issues and it suffers from some problems like loose connection, corrosion, and unbalanced loads. In this paper noninvasive methods are used to monitor the temperature of electrical assets and analysis of the hot region, and it used the watershed transform for the image segmentation and color-based segmentation which separates the red, green, blue area of the image. Dark red hot region areas are detected and the maximum temperature is 194 °C and the reference temperature is 40.8 °C of fuse cabinet. The proposed method is to detect the hotness and hot region of the electrical assets for fault detection.

Mitigation of voltage sag/swell and harmonics using self-supported DVR

Unbalanced voltage supply is an important issue in distributed network system. In this paper, we proposed a new scheme to control the self-supported dynamic voltage restorers (DVRs). In this scheme, three phase harmonic filter (double tuned) is used to moderate harmonics, generated by voltage source converter (VSC). Total harmonic distortion (THD) is minimized at load bus. By using a reduced-rating DVR, the reimbursement of the voltage dip, voltage swell, and harmonics is demonstrated. Parks Transformation is used to convert the voltages from rotating vectors to the stationary frame. The load voltage is sustained sinusoidal and in phase by inserting properly required compensation voltage through DVR. Extensive experiments are performed on MATLAB platform to analyze the performance of the proposed scheme.

Design and development of hybrid wind-hydro power generation system

In this paper, we are trying to find the conditions in India in order to develop a combined wind/hydro power plant in the areas where they are easily available. India has a great wind potential but the existing installed wind power cannot be fully absorbed. Moreover, the cost of electricity is very high and the operation of conventional plants is polluting the environment and it is expensive too. The large wind farms help to increase the stability problems when they are connected to the grid. In this research work, a combined wind-hydro power system has been developed aiming to produce low cost electricity and in increasing the penetration of renewable energy source in India. Additionally, the design and performance analysis is performed and the results have found of this hybrid system using MATLAB/Simulink.

An Efficient Technique for Power Management in Hybrid Solar PV and Fuel Cell System

ABSTRACT In this paper, a hybrid system with solar PV, fuel cell (FC), and battery energy storage system (BESS) for the efficient power-management scheme have been proposed. The combined system is designed for the load of 1.2 kW consist of 0.9 kW AC and 0.3 kW of DC load. A power-management scheme is implemented and the modeling of the PV array with maximum power point tracking technique is used here. This hybrid system is providing power to a mutual DC bus from where the AC and DC type of loads are taking the supply. The power-management scheme used in this paper depends on power sharing and regulation of voltage at DC bus. The hybrid system provides the advantage of the reduction in capacity of battery banks to be used as it is integrated with a fuel cell system. Graphical Abstract

A Novel Scheme of Fault Detection in Transmission Line Using Image Processing

This paper gives a scheme to find the fault or monitor the condition of transmission line using image processing. The technique of Digital Image Processing (DIP) and Wavelet Shrinkage Function (WSF) is used for fault detection and diagnosis. In other context, image of a transmission line is taken by the thermovision camera with the coordinates of transmission line, which are current, voltage, and temperature. The algorithm for image segmentation is used which divides the image into the set of part and objects. Application of WSF is done to read the image characteristics and standard deviation which gives the image quality where the fault occurs. This proposed method gives results in terms of visual quality and peak signal-to-noise ratio (PSNR).

Efficient detection of brain tumor from MRIs using K-means segmentation and normalized histogram

Magnetic resonance imaging (MRI) is a technique which is used for the evaluation of the brain tumor in medical science. In this paper, a methodology to study and classify the image de-noising filters such as Median filter, Adaptive filter, Averaging filter, Un-sharp masking filter and Gaussian filter is used to remove the additive noises present in the MRI images i.e. Gaussian, Salt & pepper noise and speckle noise. The de-noising performance of all the considered strategies is compared using PSNR and MSE. A novel idea is proposed for successful identification of the brain tumor using normalized histogram and segmentation using K-means clustering algorithm. Efficient classification of the MRIs is done using Naïve Bayes Classifier and Support Vector Machine (SVM) so as to provide accurate prediction and classification.