K. Sujatha

Intelligent Parallel Networks for Combustion Quality Monitoring in Power Station Boilers

This research work includes a combination of Fisher’s Linear Discriminant (FLD) analysis by combining Radial Basis Function Network (RBF) and Back Propagation Algorithm (BPA) for monitoring the combustion conditions of a coal fired boiler so as to control the air/fuel ratio. For this two dimensional flame images are required which was captured with CCD camera whose features of the images, average intensity, area, brightness and orientation etc., of the flame are extracted after pre-processing the images. The FLD is applied to reduce the n-dimensional feature size to 2 dimensional feature size for faster learning of the RBF. Also three classes of images corresponding to different burning conditions of the flames have been extracted from a continuous video processing. In this the corresponding temperatures, the Carbon monoxide (CO) emissions and other flue gases have been obtained through measurement. Further the training and testing of Parallel architecture of Radial Basis Function and Back Propagation Algorithm (PRBFBPA) with the data collected have been done and the performance of the algorithms is presented.

Monitoring Power Station Boilers Using ANN and Image Processing

This project deals with the monitoring the combustion quality of the power station boilers using Artificial Intelligence for improvement in the combustion quality in the power station boiler. The colour of the flame indicates whether the combustion taking place is complete, partial or incomplete. When complete combustion takes place the flue gases released are within the permissible limits otherwise its level is high which is out of limit. By analyzing the flame color which is captured using infrared camera and displayed on CCTV the quality of combustion is estimated. If combustion is partial or incomplete the flue gases released will create air pollution. So this work includes enhancement in the quality of combustion, saving of energy as well as check on the pollution level. The features are extracted from the flame images such as average intensity, area, brightness and orientation are obtained after preprocessing. Three classes of images corresponding to different burning conditions are taken from continuous video. Further training, testing and validation with the data collected have been carried out and performance of the various intelligent algorithms is presented.

Foetal Heartbeat and Volume Detection Using Spherical Harmonics Shape Model for Level Set Segmentation

The paper is aimed to estimate the heartbeat and volume of a foetus based on three-dimensional images from ultrasonic using image processing algorithms. Since human soft tissues show a minimal intensity difference, the ultrasonic produces poor quality image. Enhancement of the boundary is done to recover the deprived image which is the input image for segmentation. Simultaneously, the superfluous edges in the foetus image are removed to enhance the quality. Nonlinear dispersal techniques were incorporated for foetal image enhancement method. Foetus images are processed using segmentations based on iso-intensity and edge focusing to achieve apt and comprehensible edge image. The iso-intensity shape method is applied to sort the pixel values with equal intensity. Another shape-based method, which has been used, is edge focusing, and this method depends on the disparities of significance between dissimilar boundaries in the foetus picture. When outline is traced out as a series of the points in the three-dimensional volume, then the shape of the foetus can be fitted. The normal profile of anterior or posterior region resembles a sphere or an ellipse which needs to be fitted. As soon as an analytical figure is fit, the volume of the foetus and therefore the heartbeat and volume can be premeditated. These image processing algorithms are developed using image acquisition tool from MATLAB 2013a to acquire and process the foetal images. The results are finally validated using the error value.

An Approach to Wireless Sensor Networks for Healthcare Scrutiny in Hospitals

In a hospital, a system by which a patient’s physical condition and physiological parameters can be continuously monitored is essential. For example, parameters such as blood pressure (BP), cardiac rate, and fetal movements need to be taken in order to properly manage their condition. This chapter focuses on a scheme that has the ability to monitor physiological parameters from various patients. In the planned scheme, a director node is attached to the surface of the patient’s skin to gather information from the unwired sensors and transmit it to the ground station. This scheme can sense anomalous conditions and can produce a corresponding alert signal for the patient. At the same time it can pass on a message to the mobile service or send an e-mail to the patient’s general practitioner. In addition, the planned scheme contains a number of reliable, unwired relay nodes used to transmit the information passed on by the director node to the ground station. The key benefit achieved during assessment is a decrease in the energy used to extend the system’s existence, speed up and the extent to which the communication exposure to boost the liberty for increasing the patient’s life. Thus a multi-user design for infirmary healthcare has been developed and its performance weighed against existing accessible networks. It is supported by a multi-hop relay node in line with exposure to energy usage and speed.

IoT-Based Multimodal Biometric Identification for Automation Railway Engine Pilot Security System

Railways are the most convenient mode of transport, but safety precaution is lagging. Train accidents, due to an unknown person operating the engine, will lead to the end of many lives and also loss of railway property. The optimal solution to meet this problem here proposes the effective system of “Automation of Railway Engine Pilot Security System using Multimodal Biometrics Identification” (AREPSS using MBI). Iris and Fingerprint inputs are given by engine pilot from cabin to control room using Internet of things (IoT). In control room, identifications take place by fusing the inputs and then pass the decision signal to automatically start the engine. The common unimodal biometric system can be seen in most of the places due to its popularity. Its reliability has decreased because it requires larger memory footprint, higher operational cost, and it has slower processing speed. So, we are introducing multimodal biometric identification system which uses iris and fingerprint for security reason. The major advantage of this several modality method is that as both modalities utilized the same matcher component, the reminiscence footprint of the system is reduced. High performance is achieved by integrating multiple modalities in user verification and identification causing high dependability and elevated precision. So this procedure improves the safety in engine and thus helps in saving lives and property.

Automation of Railway Engine Pilot Security System Using Multimodal Biometric Identification

Railways are the most convenient mode of transport, but safety precaution is lagging. Train accidents due to an unknown person operating the engine will lead to the end of many lives and also loss of railway property. The golden solution to meet this problem here the proposed effective system is ‘Automation of Railway Engine Pilot Security System using Multimodal Biometrics Identification’ (AREPSS using MBI). Iris and fingerprint inputs are given by engine pilot from cabin to control room. In control room, identification takes place by fusing inputs, then passing the decision signal to automatically start the engine. It is the most commonly used unimodal biometric system, which can be seen in most of the places due to its popularity. Its reliability has decreased because it requires larger memory footprint and higher operational cost and it has slower processing speed. So, we are introducing Multimodal Biometric Identification System which uses iris and fingerprint for security purpose. The major advantage of this multimodal analysis is based on the template-matching phenomenon which utilizes less memory for storage as compared with footprint. User corroboration by multiple modality methods yields high output, high reliability, and high accuracy. So this technique enhances security in engine and thus saves lives and property.

Lighting Electrical Energy Audit and Management in a Commercial Building

In the present scenario, the world is dependent upon the ways to conserve electrical energy in an effective manner with less cost investment in India. The demand that is lagging in the year 2016–2017 is 300 GW. It is seen that day by day the demand is increasing the government is behind the generation part but conservation is very much essential to reduce the demand in an effective manner, looking over this scenario an initiative has been taken in our University to conduct electrical energy audit and management in an effective manner to reduce the demand and save 10 MW generation in 10 years. The initial work was started under the vision MGR-VISION 10 MW which was inaugurated in our University. The team has completed audit in 25 residential flats, 2 commercial building and 2 industries so far. This paper delivers a lighting layout for a commercial building in which it consist of six floor. Lighting layout of one particular floor is done with electrical energy audit and energy management. The recommendations for the benefits of implementation with breakeven chart are given to reduce the consumption in an effective manner. Recommendation for usage of renewable energy is given so as to reduce the consumption to reduce demand and save electrical utilization bills.

Internet of things for flame monitoring power station boilers

Analysis of combustion quality of flame images of thermal and gas turbine power plants is of great importance. In the domain of image processing the detection, recognition and understanding is the foundation for identifying the combustion condition. Soft sensors are the state of the art. So the flame temperature based on subsequent combustion quality estimation is done using Back Propagation Algorithm (BPA) and Ant Colony Optimization (ACO). The basic idea utilizes the colour information from the flame images. The first step is to define a feature vector. The 9 feature elements, from the samples of 51 flame images are used to train and test the model. Experiments prove this method to be effective. The classification of flame images based on combustion quality is dependent on the flame temperature and colour. The solution includes the Internet of Things (IoT). The intelligent sensors are embedded in the computing system to monitor the combustion quality and flame temperature. This flexible and dispensable form of environment needs continuous monitoring, controlling and behavior analysis in power plants. The prototype implementation consists of Arduino UNO board, intelligent sensors with Arduino hardware support package. The implementation is tested for monitoring the combustion quality and its subsequent flame temperature to provide a feed control for combustion quality monitoring and to make the environment smart.

Condition monitoring of gas turbine power plant using image processing (CMGTPPIP)

In the gas turbine power plants, the images of the flame are obtained from the video image captured by the infrared camera. This flame image is analyzed for image processing, detection, recognition and to understand the combustion condition. The flame image obtained is classified by soft sensors using feed forward neural network trained with Back propagation algorithm (BPA). Classification and other process are done based on the colour of the flame images which are determined by the combustion quality. First for each flame image a feature vector is determined. The feature vector describes 7 features of the flame like brightness of flame, the area of high temperature flame, brightness of high temperature flame, the rate of area of high temperature flame, the flame centroid. Image classification and feature vector is done to measure the temperature and flue gas emission from the colour of the flame obtained from the image. Here 51 sample images are taken for training, testing, and with proposed algorithm which are then recognized and classified.

Automation of solar system for Maximum Power Point tracking using Artificial Neural Networks and IoT

The importance of this project focuses on the efficiency of the solar panel which can be increased by incorporating indigenous solar tracking systems in order to increase solar panel efficiency. However, implementation of this technology requires accurate control which is crucial to develop a refined tracking system. In this work, a solar tracking system using Artificial Neural Network (ANN) based Image Processing (IPT) Techniques to estimate the azimuth angle of the sun from Global Positioning System (GPS) and image sensor is proposed here. The features extracted using IP algorithms with a decision making AI process is adopted to differentiate whether the present weather condition is sunny or cloudy. With reference to the results obtained, the solar tracking system establishes the usage of astronomical calculations approximately. The proposed hi-tech arrangement is evaluated and validated through experimentation results which are made available on the cloud service for coordination.