Recognition And Suppression Methods For Ferroresonance And Grounding Arc Over-voltage In Distribution Power System

In the power system, over-voltage is an important factor affecting safe operation of power grid, ferroresonance and grounding arc belongs to the internal over-voltage phenomena in the power system, have a longer duration and tremendous threat to the security and stability of power system. The establishment of over-voltage monitoring system for real-time tracking; profiling and flexible defense system stll is an outstanding problem. The development mechanism of ferroresonance over-voltage and grounding arc over-voltage are studied and analyzed, and the characteristic parameters are provided to distinguish these two kinds of overvoltage from the rest ones. Based on the characteristic parameters, the pattern recognition algorithm is studied to rapidly extract ferroresonance over-voltage and grounding arc over-voltage from the overvoltage wave forms. The suppression methods for both over-voltage have been studied, chaos control method based on T-S fuzzy model for ferroresonance over-voltage suppression, and Two-Tuning methods for grounding arc over-voltage suppression have been separately proposed, and the over-voltage suppression effect was evaluated. Methods proposed in this paper can effectively inhibit grounding arc over-voltage and ferroresonance over-voltage through simulation analysis.In this paper, the research of the ferroresonance and the grounding arc over-voltage simulation model, and its influencing factors, waveform characteristics have been analyzed based on mathematical models. Wavelet transform is introduced into the over-voltage characteristic quantities extraction. Taking into account the over-voltage signals non-stationary characteristics, the db4 wavelet is chosen in over-voltage 10-layer wavelet decomposition, which has orthogonality, time-domain compact support and regular character. The detail coefficients and the energy amplitude at all levels, the length of over-voltage continuous, three-phase voltage RMS, three-phase waveform cross-correlation coefficient is selected as characteristic quantities, which make up the lack of time-domain analysis or frequency domain analysis as a separate analysis, and can achieve the purpose of extracting useful information on the effective.By constructing a multi-layered over-voltage classification tree, the hierarchical structure identification easy to analyze and synthesize the various features by hierarchical structure, fractionize the over-voltage layer by layer, so the relationship between different over-voltage is clear, it's easy to understand, update and adjust the features, it would be more effective and more pertinence to choice the features. The eigenvector set is selected through different modules and different levels, forming module-level structure, and constitute the best features of the layer. According to the classification tree, each node combines the best features, the thesis adopts the genetic algorithms to class fuzzy cluster center of over-voltage, improve the module node features and genetic algorithm crossover and mutation operators, at last obtain the optimal clustering centers and fuzzy membership of over-voltage at different layers.Chaos control method for the ferroresonance's chaotic characteristics in neutral grounded system has been studied in this paper. A chaos control method based on T-S fuzzy model was proposed, which based on nonlinear dynamics theory, this model can approximate any nonlinear systems of arbitrary-precision, in this theory, a nonlinear system considered as the sum of linear subsystems with their weight multiplied, in this way, the linear control theory can be applied to the nonlinear system. The model's consequent of a rule consists of linear state space subsystem, the fuzzy controller obtained by fuzzy systems theory, and than the controller can be used to suppress the ferroresonance over-voltage. This method can cut into the control at different times, ferromagnetic resonance system can be stability to different control objectives, and the method has the advantages of rapid control process, simple calculation and good effect.The calculation of the power grid-to-ground capacitive current and arc-suppression coil tuning parameters is derived, based on Petersen-coil self-loss conductance and line-to-ground conductance. By sampling the voltage and current on the neutral point in the system during normal operation, calculating the system-to-ground capacitance in real-time and finding the value of Petersen-coil inductance, the basis for real-time Petersen-coil tuning is provided. This method can calculate the capacitive current with a relative error of less than 1%, and the compensation precision is increased compared with the conventional method, which is able to carry out a more accurate compensation for the arc grounding. Petersen-coil input strategy based on this method can avoid the difficult balance between the difficulties of Petersen-coil parameter adjustment in grid normal operation and fault operation, this strategy can suppress the arc ground over-voltage and has little effect during normal operation of the power grid.