| S.No | Title | Author | Description | Download | |
|---|---|---|---|---|---|
| 1 | Design Of Functional Observer Based Load Frequency Controller For Inter-Connected Power System | Anand Gondesi,R Vijaya Santhi ,K R Sudha |
This paper presents a novel peculiar functional observer based on quasi-decentralized load frequency control scheme for power systems. The quasi-decentralized functional observers are designed to implement any given state feedback controller based on functional observer theory. The designed functional observers are decoupled from each other to have a simpler structure when compared to the state observer based schemes. The proposed design method is based on the network topology and the proposed functional observer scheme is further applied to a complex nonlinear power system. |
||
| 2 | Reactive Power Flow Control Using Voltage Dependent Reactive Power Control Strategy In PV Based Distributed Grid System | Shravankumar Venumula |
Utilization of grid connected photovoltaic systems is facing rise in voltage and reverse power flow problems which are indirectly complicating the reactive power flow control in power system network. It is highly needed to develop an appropriate solution which can control reactive power and for better voltage regulation. In this paper, the reactive power flow control is obtained with the help of reactive power centralized management system in a high voltage distribution along with characteristics of voltage dependent reactive power with photovoltaic system. The control of rise in voltage and reverse power flow is obtained at the medium voltage level which results in reactive power control interchange between in medium and high levels of voltages. The proposed system consists of two radial feeders at the grid connected PV system. The proposed system is modelled using Simulink and the results are provided to explain the reactive power control by controlling rise in voltage and reverse power flow.
|
||
| 3 | EMAIL Spam Detection Using NLP | 1.I JOSEPH KISHORE , 2.SIVA RAMA KRISHNA T |
Email has become the most widely used and economic form of communication in this digital era. Email users generally get bombarded with unsolicited messages regarding direct marketing often sent to multiple users using bots. Users have to spend a considerable amount of time on clearing such messages. Study shows that there is a sharp increase in spam emails , it is estimated that they are almost 89% of the total email traffic. Spam emails can create havoc by causing financial loss or identity theft of users.Spammers use many techniques to bypass manual filters such as misspelled words by adding extra letters to words (eg: amazing, amaze-on etc..), synonyms of generally used words etc.. Use of Machine learning models can handle such data. Creating text classifiers that precisely filter such emails from the user's mailbox to spam folder is more efficient than manual filters. |
||
| 4 | Image Based Plant Disease Detection Using Deep Learning | 1.B.Sathish Kumar Reddy,2.B.Laxmi Vara Prasad,3.K.Susanna,4.P.Harsha |
With increasing population the crisis of food is getting bigger day by day. During crisis crop diseases are a major threat to food industry, but their rapid identification remains difficult in many parts of the world due to the unavailability of the experts and necessary infrastructure. The recent expansion of deep learning methods has found solution for detection of various plant diseases.Using the dataset which contains several leaf images which were taken in various weather conditions, at different angles. Entire dataset is split into train and test sets. We do image processing using deep learning on the train set and we evaluate our trained model on the test set. During the training we want to get rid of many problems like overfitting and underfitting by make use of regularization or dropout or by increasing number of layers. This model may recognize only some diseases because of the limited dataset available, but we can use this model for detection of any type of disease by providing the large dataset which contains almost all diseases. |
||
| 5 | Early Detection And Prevention Of Anomalies Using Mcnn | T. Suvarna Kumari |
Crowd detection and density estimation from crowded images have a wide range of applications such as crime detection, congestion, public safety, crowd abnormalities, visual surveillance and urban planning. CNN based techniques helps to estimate the crowd detection count and density. The job of detecting a face in the crowd is complicated due to its variability present in human faces including color, pose, expression, position, orientation, and illumination. The proposed approach is simple but effective Multi-column Convolutional Neural Network (MCNN) architecture to map the image to its crowd density map. The proposed MCNN allows the input image to be of arbitrary size or resolution. By utilizing filters with receptive fields of different sizes, the features learned by each column CNN are adaptive to variations in people/head size due to perspective effect or image resolution. |
||
| 6 | Earthquake-Related Behavior Of Piles In Liquid Deposits | 1.G. Shirisha, 2.B. Sudhakar , 3.B. Harish |
This article discusses the behaviour of piles in liquefied soils based on the results of recent investigations carried out in Japan. These studies were carried out in recent years. These studies include (a) field performance and damage features of piles observed in the 1995 Kobe earthquake; (b) experimental findings from benchmark tests on fullsize piles; and (c) simplified design methodology for piles undergoing lateral spreading. (a) Field performance and damage features of piles were observed in the 1995 Kobe earthquake. An investigation of piles that are simplified gives particular emphasis to the consequences of enormous lateral displacements of liquefied soils and their modelling in order to get a deeper understanding of these effects. |