| Permanent magnetic actuator (PMA) is very suitable for driving vacuum circuit breaker due to good performances including simple structure, few components, high reliability and long life as well as small actuated error, and is the only operating mechanism which can realize synchronous control. Although the PMA that fits high voltage vacuum circuit breaker has been expected by the users, the news concerned with this aspect hasn't been reported. In order to development high voltage vacuum circuit breaker (HVVCB) 126kV PMA for HVVCB with synchronous control is researched in this dissertation.By means of Ansoft Maxwell 3D/2D software, the dynamitic and static characteristics of PMA are simulated. As results, a novel excitation method with main coil and auxiliary coil in series is presented so as to make the armature obtain enough energy to move all through. By which, the dynamitic characteristics of PMA is evidently improved such as the opening time and closing time is reduce, the energy consumption of the energy storage capacitor is decreased, the peak current of exciting coil is minished, and the collision impact at the close moment of the armature is lightened. The results show that it may satisfy the demands of 126kV high voltage vacuum interrupter.Support vector machine (SVM) is the best general machine learning theory developed from statistical learning theory, and suit to do prediction from small samples by learning. SVM has strict theoretical basis and good predictive ability as well as simple computing process, also may automatically fix the structure of the model. PMA is supplied by the energy storage capacitor which is controlled by the power electronic device. The opening time and closing time of PMA is related to the charging voltage of the energy storage capacitor and the ambient temperature as well as the contact's electrodynamics force caused by the short circuit current, in order to improve the synchronous control precision, the predictive models of the opening time and closing time of PMA are built by minimum mean-square value SVM so as to obtain the predictive function by means of lattice search method. The predictive function related to few support vectors is esay to be programmed. Synchronous control of circuit breaker is trend of development and one of the important research topics. Now synchronous closing switch has come into practical application, but synchronous opening switch has been made theoretic ready. Because the short circuit current related to the primary phase of the power supply and the line impedance is asymmetric, thus it is hard to obtain accurately the synchronous opening control signal. According to the general expression of short circuit current, the short circuit current is analyzed and simulated based on self-adapting parameter identification and arithmetic recursion, and then the predictive model of the short circuit current zero is built. By Matlab simulations the precision of the predictive model under differ sampling rate and effective number of bits (ENoB) of analog to digital converter (ADC) as well as signal to noise ratio (SNR) of the measurement circuit is analyzed, as a result, 7/8 wave arithmetic recursion formula presented in this dissertation is adopted because it is easy to come true on the condition that ENoB of ADC is equal to 12, sampling period is equal to 32 times in every fundamental wave and SNR of measuring circuit is equal of 60dB.Field programmable gate array (FPGA) is a general programmable electronic component, and is provided with goods of high integrated level, versatility, flexibility, fast operational speed and low power waste so as to be suit to the complex system. In order to control PMA the signle-chip and FPGA is united to make the parallel process, thus the real time the reliability is improved. |
PDF Full Text Request