ICTACT Journal on Soft Computing

A. Soundarrajan ¹, S. Sumathi ², G. Sivamurugan ²

Affiliations
1 Department of Information Technology, PSG College of Technology, Tamil Nadu, IN
2 Department of Electrical and Electronics Engineering, PSG College of Technology, Tamil Nadu, IN

Abstract

Power generating system has the responsibility to ensure that adequate power is delivered to the load, both reliably and economically. Any electrical system must be maintained at the desired operating level characterized by nominal frequency and voltage profile. But the ability of the power system to track the load is limited due to physical and technical consideration. Hence, a Power System Control is required to maintain a continuous balance between power generation and load demand. The quality of power supply is affected due to continuous and random changes in load during the operation of the power system. Load Frequency Controller (LFC) and Automatic Voltage Regulator (AVR) play an important role in maintaining constant frequency and voltage in order to ensure the reliability of electric power. The fixed gain PID controllers used for this application fail to perform under varying load conditions and hence provide poor dynamic characteristics with large settling time, overshoot and oscillations. In this paper, Evolutionary Algorithms (EA) like, Enhanced Particle Swarm Optimization (EPSO), Multi Objective Particle Swarm Optimization (MOPSO), and Stochastic Particle Swarm Optimization (SPSO) are proposed to overcome the premature convergence problem in a standard PSO. These algorithms reduce transient oscillations and also increase the computational efficiency. Simulation results demonstrate that the proposed controller adapt themselves appropriate to varying loads and hence provide better performance characteristics with respect to settling time, oscillations and overshoot.

Keywords

Load Frequency Control (LFC), Automatic Voltage Regulator (AVR), Evolutionary Algorithm (EA), Enhanced Particle Swarm Optimization (EPSO), Multi Objective Particle Swarm Optimization (MOPSO), and Stochastic Particle Swarm Optimization (SPSO).

Full Text

     

P. Surekha ¹, S. Sumathi ¹

Affiliations
1 Department of Electrical and Electronics Engineering, PSG College of Technology, Tamil Nadu, IN

Abstract

This paper proposes a new optimization method for digital images in the Discrete Wavelet Transform (DWT) domain. Digital image watermarking has proved its efficiency in protecting illegal authentication of data. The amplification factor of the watermark is the significant parameter that helps in improving the perceptual transparency and robustness against attacks. The tradeoff between the transparency and robustness is considered as an optimization problem and is solved by applying Genetic Algorithm. The experimental results of this approach prove to be secure and robust to filtering attacks, additive noise, rotation, scaling, cropping and JPEG compression. The Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE), and computational time are evaluated for a set of images obtained from the Tampere University of Technology, Finland using the MATLAB R2008b software.

Keywords

DWT, Genetic Algorithm, Robustness, Transparency, PSNR, MSE, Computational Time.

Full Text