Record Details

Automatic generation control of interconnected power systems using artificial neural network techniques

Shodhganga@INFLIBNET

View Archive Info
 
 
Field Value
 
Title Automatic generation control of interconnected power systems using artificial neural network techniques
 
Contributor Ghodekar, J G
 
Description In an interconnected power system, any sudden small load perturbation in any of the interconnected areas causes the deviation of the area frequencies and also of the tie line powers. The main objectives of Automatic Generation Control (AGC) are: 1. To maintain the desired megawatt output and the nominal frequency in an interconnected power system. 2. To maintain the net interchange of power between control areas at predetermined values. Thus, an AGC scheme for an interconnected power system basically incorporates suitable control system, which can bring the area frequencies and tie line powers back to nominal or very close to nominal values effectively after the load perturbations. Conventionally this is achieved with the help of Integral Controllers. However, they have certain disadvantages like; they are slow in action, they do not take into account the inherent nonlinearities of various power system components, it is quite difficult to decide the integrator gain settings as per the changes in operating point. Intelligent control systems have many advantages over integral controllers. They provide a high adaption to changing conditions and have ability to make decisions quickly by processing imprecise information. Also, they can perform effectively even with nonlinearities. In present research work, the artificial neural network (ANN) technique has been used for AGC of interconnected power systems. The feedforward neural network controllers capable of working in real time have been developed. They have been trained using Lavenberg-Marquardt (LM) back propagation algorithm under supervised training method with adequate amount of data which has been generated by using optimal and suboptimal control strategies. Once trained properly, these controllers can be interfaced with the power system as real time controllers. The trained ANN controllers give the outputs which act as control inputs of the power system.
List of figures, symbols, references and appendix included.
 
Date 2010-05-31T11:41:52Z
2010-05-31T11:41:52Z
2010-05-31
May 2009
September 30, 2009
 
Type Ph.D.
 
Identifier http://hdl.handle.net/10603/48
 
Language English
 
Rights Bharat University
 
Format xv, 142p.
CD
 
Publisher Chennai
Bharath University
Bharat University. Faculty of Engineering and Technology. Department of Electrical Engineering
 
Source INFLIBNET