Improvement of Transient Stabilty Using Neuro Fuzzy System

Computational Intelligence combines neural network, fuzzy systems and evolutional computing. Neurofuzzy integrated system utilizes features of both Neural and Fuzzy networks together for better results by which generalization of the unseen data from seen data by forming the fuzzy rules and training. In this project training of the system with training data which usually is 70% of the whole available data, rest 30% data is used for testing. The algorithm used in project is hybrid algorithm. A characteristic inherent to Electric power system is that they operate under the influence of disturbance The analysis required to know whether, the power system will survive the transients and move into stable operation or lead to loss of synchronous operation is subject of primary concern and is referred to as transient stability assessment. Methods of improvement stability try to achieve – Reduction in disturbing influence by minimizing the fault severity and duration. Increase of the restoring synchronizing forces. In Fast Valving it is a technique applicable to thermal units to assist in maintaining power system transient stability. The Application of Artificial neural network to fast valving for transient stability improvement has the input Accelerating power and rotor speed. The output is Fault clearing time. The Neural fuzzy system are stable, efficient, reliable. Here an Adaptive Network based fuzzy system is used. The data with power, speed and fault clearing time is used in order to train network with rule based structure. Then this training data is used to test the output with unseen data and error is obtained then minimization of this error is done in order to obtain an accurate neuro-fuzzy system which can correctly predict the opening and closing of the thermal valve.