Research On Characteristics And Control Of Chaos And Detection Method Of Ferroresonance Overvoltage In Neutral Grounded System

Ferroresonance overvoltage is one of the inner overvoltages in power system. It does great harm to stable and safe running of power system. Especially transmission net developing rapidly at now, it has great relationship with safety and reliability of power grid. So, it has great significance to research ferroresonance overvoltage. Compared with neutral ungrounded power system, ferroresonance overvoltage occurred in neutral grounded power system has more new features, and the research about it is much less. For keeping the transmission and distribution equipments out of damage and malfunction, it is very important to research the ferroresonance overvoltage in neutral grounded power system. The paper uses several kinds of mathematic theorem to research the chaos characteristics, methods of chaos control and extracting features quantities of ferroresonance overvoltage oriented to the neutral grounded system.Based on deeply research on chaos characteristics of neutral grounded power system while the ferroresonance overvoltage happening, through the calculation on heteroclinic orbit, the paper uses Melnikov method to calculate the chaos threshold value, and uses functional analysis to prove the existence of periodic solution to provide the theorem rule for controlling and erasing the chaos in overvoltage.Based on the time-varying mathematic model of three phase neutral grounded system, the paper brings forward the method which uses the stabilization mathematic theorem to research the stability of the system. Especially the paper researches the important influence of zero sequence resistance and interphase capacitance. And the parameters condition which keeps the system running stably is also calculated.The paper proposes the theorem and method of using constant pulse to control the chaos state of ferroresonance in neutral grounded system. It uses the calculation method of constant pulse quantity, and controls the chaos of ferroresonance overvoltage effectively, and realizes the transition from the chaos to sequences. And this method has many merits.Based on the RBF neural network, the paper proposes the method which uses maximum entropy learning algorithm to control the chaos in ferroresonance overvoltage in neutral grounded system. The algorithm optimizes the object function to derive learning rule of central vectors, and use the clustering function of network hidden layers. It improves the regression and learning ability of neural network. And the control result shows that the method could control chaos effectively, and the speed rate is very fast, and could realize several kinds of target.Based on the systematic research on using the wavelet local maxima modulus to calculate the...