Study On Some Problems Of Traction Power Supply Systems In High-Speed Railway

Along with the social development, the existed ordinary-speed railway does not meet the transportation requirements of Chinese people. Due to the large population and vast land, high-speed railway (HSR) is most suitable to China. AT-fed traction power supply system is usually adopted in HSR. Lots of problems have not been totally illustrated in AT-fed traction power network (TPN) yet, some of which will be discussed in this paper.The methods to derive per-unit-length parameters of AT-fed TPN are discussed. As there are transverse connections in AT-fed TPN, it would be very complicated for the accurate simulation of the TPN, and reduction methods always are adopted in practice. Reduction methods are compared, and the deviation is distinguished. Chain circuit model and generalized method of sequence components are briefly described, and the method to cancel the effect of mutual impedance between rail and other conductors on sequence parameters is illustrated. Composite sequence network with AT leakage reactance taken into account is firstly presented, equivalent circuit of the traction power network thus is derived.In the simulation of AT-fed TPN, the existed method to derive characteristic impedance matrix (CIM) of the current-return circuit in AT-fed TPN is incorrect in principle as it does not take into account the transverse connections between conductors. Based on the chain parameter matrix of the current-return circuit, it is deduced in this paper that CIM should meet complex algebraic Reccati equation, to which generalized eigen-problem algorithm can be used. A case is presented, and sensitivity analysis is conducted. It is shown that only small deviation exits between the results of the two methods, and rail potential is not sensitive to the variation of CIM. Thus the existed method or the simplified version proposed in this paper can also be used for the derivation of CIM.Three modes of AT-fed TPNs are briefly described in this paper, and are compared on traction transformer capacity, TPN’s transmission capacity and rail potential, etc. The results show that rail potential, TPN’s transmission capacity, voltage level and circuit breaker’s disconnection voltage of the three modes are nearly identical, while the novel mode doesn’t necessitate substation AT and traction transformer tapping. However the ground electrode current is the highest and the existed fault location method, i.e., AT neutral current ratios, does not work. French mode AT-fed TPN requires a higher transformer capacity in the secondary windings, and tapping is also needed, however the AT in the traction substation is omitted. Tapping is not required in the Japanese mode, however substation AT is needed to be installed instead. As the novel mode TPN has drawback of fault location, French model and Japanese mode TPNs have the same equivalent circuit, the reseach is focused on Japanese mode TPN hereinafter.Decoupled method is extended to arbitrary multi-conductor system. Based on the extended decoupled method, rail potential distribution in double-line AT-fed TPN is analyzed. Storage capacity and calculation burden is reduced compared to chain circuit model which is general used in traction power system analysis. And the accuracy depends greatly on the current distribution calculation. This method is more suitable for theorital analysis than chain circuit model, thus deserve the study devoted to.The normalization of the AT leakage reactance is discussed, and typical range of the normalized value is presented. The influences of the AT leakage reactance is analyzed theoretically in several aspects. The results show that the influences on rail potential and voltage level are trivial, such as rail potential, voltage level, fault location and ground electrode current. Based on the TPN’s chain circuit model, simulation is conducted and the theoretical analysis is verified. It is found that the influences on fault location, ground electrode current are obvious. Thus leakage reactance should be selected as small as possible if permitted.In track circuit of high-speed railway station, the middle points of impedance bonds at several IRJs are isolated to avoid returning interference through other paths. However, IRJ and rail injuries often occur as multiple unit train passing the isolated IRJ. Equivalent circuit of insulated rail joint (IRJ) is derived first, and the mechanism of IRJ and rail injuries is investigated. A scheme is presented to decrease the injuries. Based on PSCAD and Mayr arc model, simulation is conducted to study the effect of the proposed scheme. The results show that the power and energy of the arc are decreased to such an extend as the normal IRJ.