Excitation Control of Self Excited Induction Generator using Genetic Algorithm and Artificial Neural Network

Induction generators which may be operated in grid or self-excited mode, are found to be successful machines for wind energy conversion. Out of these two selfexcited mode is gaining importance due to its ability to convert the wind energy into electrical energy for large variations in operating speed. However it has been found that these machine exhibits a poor voltage regulation. Steady-state analysis of self excited induction generator reveals that such generators are not capable to maintain the terminal voltage and frequency in the absence of expensive controllers. In turn addition of such controllers may result into a fall in popularity of this machine due to its simplicity. Another simple way to control the terminal voltage is through excitation control using series compensation. In this paper artificial intelligent techniques are used to model the control strategy for proper reactive compensation under different operating conditions. Genetic algorithm along with artificial neural network has been proposed to estimate the values of shunt and series excitation capacitance to maintain the terminal and load voltage. Simulated results as found using proposed control technique are verified using experimental results on a test machine. Simulated results are found to be in close agreement with experimental results. Keywords— Artificial Neural Network, Genetic Algorithm, Optimization, Self Excited Induction Generator, Wind Energy Generation.