Traction network simplified modeling method and power flow and fault analysis method

Abstract

The invention discloses a traction network simplified modeling method and a power flow and fault analysis method. In order to reduce calculation errors caused thereby, additional self-impedance and additional mutual impedance are introduced, and an additional impedance matrix is formed by the additional self-impedance and the additional mutual impedance. An additional admittance matrix may be generated from the additional impedance matrix. And establishing a node admittance matrix of the traction power supply system, fusing the node admittance matrix with the additional admittance matrix to obtain an extended admittance matrix and a node voltage equation of the traction power supply system, solving the equation by using a continuous linear power flow method, and performing proper processing to obtain steel rail potential and traction network voltage and current distribution. The method is further applied to traction power supply system load flow calculation and short circuit analysis. According to the method, computer memory occupancy and modeling time are reduced, and traction power supply system load flow calculation time is effectively shortened. Meanwhile, fault analysis based on the simplified traction network has the advantages of simple calculation, clear concept, high accuracy and the like.

Description

Technical Field

[0001] This invention belongs to the field of rail transit traction power supply technology, and particularly relates to a simplified modeling method for traction networks and a power flow and fault analysis method. Background Technology

[0002] In the traction power supply system, the traction substation supplies power to the train through the overhead contact line, and forms a circuit through the rails or a dedicated return line.

[0003] Modeling and simulating the traction power supply system is a fundamental basis for the planning and design of electrified lines. The main functions of traction power supply system simulation are as follows:

[0004] (1) Check the voltage level: During train operation, the voltage level of the traction network should meet the needs of the train to prevent the train from failing to operate normally due to the traction network voltage exceeding the limit.

[0005] (2) Verify whether the rail potential meets the requirements: The rail potential determines the safety of passengers and maintenance personnel or the risk of equipment insulation breakdown.

[0006] (3) Design reasonable relay protection settings: The short-circuit current of the traction power supply system is affected by many factors such as power supply capacity, line impedance, fault location and the lateral connection of the contact network and return path. When a fault occurs in the traction network, the relay protection device should be able to operate reliably and avoid false operation as much as possible.

[0007] Currently, the analysis of traction power supply systems mostly uses a unified chain circuit model for modeling and simulation. Unlike DC systems, the return path of a traction power supply system not only includes the rails but also involves the ground, return lines, or through ground wires, and there are lateral connections between the rails, return lines, or through ground wires, making the traction network modeling quite complex. With the development of bilateral power supply and through power supply, the problems of long simulation time and large computational resource consumption in power flow calculation have become increasingly prominent. Therefore, while ensuring the reliability of simulation results, simplifying the model complexity, shortening the simulation time, and saving computational resources are of great significance for improving the computational efficiency of traction power supply systems. The traction network structure is complex, and the existing traction network short-circuit calculation models have relatively large errors, making modeling using a chain circuit model quite complex. Summary of the Invention

[0008] To address the problems of complex traction network modeling, long dynamic simulation time, and high computational resource consumption associated with the unified chain circuit model, this invention provides a simplified traction network modeling method as well as power flow and fault analysis methods.

[0009] The present invention provides a simplified modeling method for traction networks, including unified chain circuit modeling of the traction power supply system, specifically comprising the following steps:

[0010] Step 1: Obtain the unit impedance matrix Z and unit admittance matrix Y of the traction network.

[0011] For a traction network consisting of n conductors, the unit impedance matrix and unit admittance matrix of the traction network are obtained using the Carson method or the mirror method, and the conductivity of the rails and the ground wire to the ground is included in the unit admittance matrix. The conductor merging algorithm is used to merge the contact wires and catenary wires in the same line and the rails in the same line. The number of conductors after merging is n′, and the unit impedance matrix Z and unit admittance matrix Y of the traction network are obtained.

[0012] Step 2: Obtain the additional mutual impedance matrix of the return network.

[0013] If a certain number of p conductors in the traction network are connected laterally at intervals, then the p conductors are divided into two parts; one part consists of conductors into which current is injected or out, denoted as conductor C1, and the remaining conductors form conductor set C2; the conductors in C2 are merged and equalized, and the merged set is denoted as conductor C2; then conductors C1 and C2 are merged and equalized, and denoted as conductor C.

[0014] Since conductors C1 and C2 are not cross-connected at every point, calculating power flow or short circuits based on conductor C will cause significant calculation errors. To reduce these errors, an additional self-impedance is introduced onto conductor C.

[0015]

[0016] Among them, Z C1 and Z C2 Z represents the self-impedance of conductors C1 and C2. C12 Let be the mutual impedance between the two, s be the spacing between the cross-connections of a certain p conductor, and m be the distance from the current injection point to the cross-connection on one side; it can be seen that when m = 0 or m = s, the additional self-impedance Z aC =0.

[0017] To accurately describe the mutual influence between the two current injection points, an additional mutual impedance is introduced; the cross-connector divides the p conductors into different segments; if the two current injection points belong to different segments, the additional mutual impedance Z... aC = 0; otherwise, use equation (2) to calculate the additional mutual impedance between the two current injection points:

[0018]

[0019] Among them, s ij The coupling distance between the two traction loads is called the coupling distance between the two traction loads and is calculated according to equation (3):

[0020]

[0021] Using equations (2) and (3), the additional self-impedance and additional mutual impedance of the return network are obtained, and the additional impedance matrix Z is formed. aC Its order is equal to the number of current injection points.

[0022] The traction network return current network is divided into the upward rail R1 and the remaining conductors Q, including the upward return current line and the through ground wire, and the downward rail, return current line and through ground wire. Using the above method, the additional impedance matrix Z of the return current network is obtained. aR .

[0023] Step 3: Obtain the additional impedance matrix of the contact wire.

[0024] Similar to step 2, use equations (1) to (3) to determine the additional matrix Z caused by the overhead contact line. aT .

[0025] Step 4: Use the impedance and admittance merging method to merge and equalize the conductors of the traction network; whereby the upstream and downstream contact networks are merged and referred to as conductor T, and the upstream and downstream return networks are merged and referred to as conductor H, to obtain the unit impedance matrix Z′ and unit admittance matrix Y′ of the traction network composed of conductor T and conductor H.

[0026] Step 5: Obtain the total additional impedance matrix.

[0027] In the power flow calculation of the traction network, the additional impedance matrix Z is used. aR Z aT By combining the results, we obtain the total additional impedance matrix:

[0028] Z a =Z aR +Z aT (4)

[0029] Total additional impedance matrix Z a It connects the overhead contact line and the traction load.

[0030] The present invention provides a method for calculating the power flow of a traction power supply system based on a simplified model of a traction network, comprising the following steps:

[0031] Step 1: Establish a simplified correlation matrix between the traction network section and the traction load.

[0032] A simplified correlation matrix M is used to describe the relationship between traction load and catenary nodes; the rows of the simplified correlation matrix represent sections, and the columns represent branches, i.e., traction loads; the elements of the simplified correlation matrix are defined as follows:

[0033]

[0034] Step 2: Obtain the additional admittance matrix.

[0035] The correlation matrix representing the contact wire nodes and the newly added nodes caused by the additional impedance is as follows:

[0036]

[0037] Where k is the number of traction loads, E k×k It is a k-order identity matrix.

[0038] From the additional impedance matrix Z a Determine the additional impedance admittance matrix:

[0039]

[0040] Step 3: Use the chain circuit model of the traction network to establish the node admittance matrix Y of the traction power supply system. S0 .

[0041] Based on the unit impedance matrix Z′ and unit admittance matrix Y′ of the traction network composed of conductors T and H, a chain circuit model is used to model the traction power supply system. Ignoring additional impedance, the nodal admittance matrix Y′ of the traction power supply system is formed. S0 .

[0042] Step 4: Incorporate the additional admittance matrix into the nodal admittance matrix Y S0 The nodal admittance matrix Y of the entire system is obtained. S .

[0043] Modify the system's node admittance matrix, and add the admittance matrix Y. a Include it; the method is: include the additional admittance matrix Y a Divide the network into appropriate blocks according to existing overhead contact line nodes and newly added nodes, and obtain...

[0044] If the traction substation is located without traction load, the node admittance matrix Y of the traction power supply system will be... s0 Also, divide it into appropriate blocks, and get Where Y0 is a complex number, Y 0t It is a row vector.

[0045] matrix Y a And matrix Y s0 By combining them appropriately, we obtain the following system node admittance matrix:

[0046]

[0047] Where H3 = [1, 0] t , It is the Kronecker product of the matrix.

[0048] If the traction substation is located under a traction load, then Y s0 =Y t,but:

[0049]

[0050] Step 5: Obtain the node voltage equations of the traction power supply system.

[0051] Define H1 = [0, 1] t , If the traction substation is located without traction load, the node voltage equation of the traction power supply system is:

[0052]

[0053] If the traction substation is located under a traction load, then:

[0054]

[0055] Among them, I s To be with M2·I L Vectors of the same order, and the vector consisting of their first two elements is:

[0056] Step 6: Solve the node voltage equations of the traction power supply system.

[0057] The continuous linear power flow method is used to solve the nodal voltage equations (11) or (12) of the traction power supply system and achieve the convergence condition to obtain the traction network tangential voltage vector U. t and the newly added node voltage vector U v .

[0058] Step 7: Solve for the rail potential and traction network current distribution.

[0059] Use equation (13) to calculate the rail potential at the traction load:

[0060]

[0061] Suppose there are p conductors connected in parallel. We need to determine the current distribution across these p conductors; and write the voltage drop equations between two adjacent cross sections of the p conductors:

[0062]

[0063] Among them, E p×1 Let q be a p-dimensional column vector of all 1s, and let q be the combined equivalent conductor of the traction network, excluding the p parallel conductors.

[0064] We obtain the following from equation (14):

[0065]

[0066] The current distribution of p parallel conductors can be obtained from equation (15).

[0067] The present invention provides a short-circuit analysis method for a traction power supply system based on a simplified model of a direct-supply traction network, specifically as follows:

[0068] When a short circuit occurs in the traction power supply system, the traction grid voltage drops significantly. The control system of the electric locomotive / motor unit will block the converter trigger signal, causing the power of the electric locomotive / motor unit to be close to zero. Therefore, the impact of traction load is generally not considered when the traction power supply system is short-circuited.

[0069] When a short-circuit fault occurs in the traction network, if the rail potential is not a concern, the traction network can be considered as a single conductor, and its unit impedance can be calculated according to formula (16):

[0070]

[0071] At the short circuit point, the additional impedances caused by the contact wire and return network are Z, respectively. aT and Z aR Solve for the Thevenin equivalent circuit as seen from the traction bus to the system, where, and Z s These represent the electromotive force and impedance of the Thevenin equivalent circuit as viewed from the traction bus toward the power system.

[0072] Find the short-circuit impedance Z as seen from the traction bus to the short-circuit point. bus,m :

[0073] Z bus,m =Z e x+Z aT +Z aR (17)

[0074] The short-circuit current at the short-circuit point can be obtained using equation (18):

[0075]

[0076] The relay protection device is set based on the measured impedance of the traction bus and the calculated short-circuit current.

[0077] The present invention provides a short-circuit analysis method for a traction power supply system based on a simplified model of a fully parallel traction network, specifically as follows:

[0078] Draw the equivalent circuit when a short circuit fault occurs in the fully parallel AT traction network.

[0079] For the return current network in the equivalent circuit, the node injected with short-circuit current is denoted as node b, and the node connected to the negative terminal of the power supply is denoted as node c; the remaining nodes are arbitrarily numbered; node c is grounded; and the node admittance matrix Y of the return current network is written according to the connection relationship between the nodes. RCAnd node b is the first node; divide the nodes of the return network into two subsets, node b is subset 1, and the remaining nodes are subset 2; divide the admittance matrix into subset blocks, and we get:

[0080]

[0081] The superscript t indicates transpose.

[0082] The nodal voltage equations for the return network are:

[0083]

[0084] Since no current is injected into the nodes in subset 2, I2 is a zero vector; by using Kron's elimination method to eliminate the nodes in subset 2 of equation (20), we get:

[0085]

[0086] The impedance between nodes b and c is:

[0087]

[0088] Calculate the short-circuit impedance Z at 25kV as seen from the traction bus to the short-circuit point. bus,m :

[0089] Z bus,m =z′1L+Z f +Z bc (twenty three) The short-circuit current is:

[0090]

[0091] The relay protection device is set based on the measured impedance of the traction bus and the calculated short-circuit current.

[0092] The beneficial technical effects of this invention are as follows:

[0093] The simplified traction network modeling method proposed in this invention can significantly reduce the number of cross sections and the dimension of the node admittance matrix in the traction power supply system, thereby reducing computer memory usage and effectively shortening the computation time for power flow calculation in the traction power supply system. The fault analysis based on the simplified traction network presented in this invention has the advantages of simple calculation, clear concepts, and high accuracy. Attached Figure Description

[0094] Figure 1 A simplified schematic diagram of the return network conductor.

[0095] Figure 2 A simplified model circuit diagram for direct supply to the traction network.

[0096] Figure 3 It is a simplified model circuit diagram of a fully parallel AT traction network.

[0097] Figure 4 It is a flow chart of power flow calculation based on a simplified traction network.

[0098] Figure 5 It is that the positions of traction loads V1 and V2 coincide.

[0099] Figure 6 It is a schematic diagram of a short - circuit fault occurring in a direct - supply traction network.

[0100] Figure 7 It is an analysis of short - circuit faults based on a simplified model of a direct - supply traction network.

[0101] Figure 8 It is an analysis of short - circuit faults based on a simplified model of a fully parallel AT traction network.

[0102] Figure 9 It is Figure 8 an analysis of the return network. Specific implementation manners

[0103] The following further elaborates on the present invention in detail in conjunction with the attached drawings and specific implementation methods.

[0104] A method for simplifying the modeling of a traction network according to the present invention is specifically as follows:

[0105] Step 1: Obtain the unit impedance matrix Z and unit admittance matrix Y of the traction network.

[0106] For a traction network composed of n conductors, use Carson's formula or the mirror method to obtain the unit impedance matrix and unit admittance matrix of the traction network, and account for the distributed conductance of the rail and the through - ground wire to the ground in the unit admittance matrix; use the conductor merging algorithm to merge the contact wire and the messenger wire in the same row, and merge the rails in the same row. After merging, the number of conductors is n′, and obtain the unit impedance matrix Z and unit admittance matrix Y of the traction network.

[0107] Step 2: Obtain the additional mutual impedance matrix of the return network.

[0108] If p (p < n′) conductors in the traction network are transversely connected at intervals, divide the p conductors into two parts; one part is the conductors with current injection or outflow, denoted as conductor C1, and the remaining conductors form conductor set C2; merge and equivalent the conductors in C2, and denote it as conductor C2 after merging; then merge and equivalent conductor C1 and C2, and denote it as conductor C.

[0109] Since conductor C1 and conductor C2 are not transversely connected everywhere, performing power flow or short - circuit calculations according to conductor C will cause relatively large calculation errors; to reduce the calculation errors caused thereby, introduce an additional self - impedance on conductor C:

[0110]

[0111] Among them, Z C1 and Z C2 Z represents the self-impedance of conductors C1 and C2. C12 Let be the mutual impedance between the two, s be the spacing between the cross-connections of a certain p conductor, and m be the distance from the current injection point to the cross-connection on one side; it can be seen that when m = 0 or m = s, the additional self-impedance Z aC =0.

[0112] To accurately describe the mutual influence between the two current injection points, an additional mutual impedance is introduced; the cross-connector divides the p conductors into different segments; if the two current injection points belong to different segments, the additional mutual impedance Z... aC = 0; otherwise, use equation (2) to calculate the additional mutual impedance between the two current injection points:

[0113]

[0114] Among them, s ij The coupling distance between the two traction loads is called the coupling distance between the two traction loads and is calculated according to equation (3):

[0115]

[0116] Using equations (2) and (3), the additional self-impedance and additional mutual impedance of the return network are obtained, and the additional impedance matrix Z is formed. aC Its order is equal to the number of current injection points.

[0117] The traction network return path is divided into upward rails R1 and other conductors Q (upward return line, through ground wire; downward rails, return line, and through ground wire). A conductor merging algorithm is used to merge the conductors in Q; the merged equivalent conductor is called conductor Q. The conductor merging algorithm is also used to merge conductors R1 and Q; the merged conductor is called conductor H. Figure 1 As shown.

[0118] Assume that the mutual impedances of other conductors to conductor R1 and conductor Q are equal.

[0119] When the return network is described by conductor H, the calculated rail potential is lower than its true value. Therefore, an impedance, called additional self-impedance, is introduced between the traction load and conductor H. Furthermore, the rail potentials of the uplink and downlink traction loads influence each other, leading to the introduction of additional mutual impedance.

[0120] For a system where conductors R1 and Q are connected transversely at regular intervals, assuming the load current is injected into conductor R1, the additional self-impedance of the return network can be obtained using equation (1). Then, the additional mutual impedance between the two traction loads can be calculated using equations (2) and (3), forming the additional impedance matrix Z caused by the return network. aR Its order is equal to the traction load number. For example, Figure 2 (a) The power supply arm of the direct-supply traction network has two traction loads, V1 and V2. Section 3 is at the end of the power supply arm. To accurately account for the ground admittance of the traction network, it is assumed that section 3 also has a load with zero current, which can be obtained as follows: Figure 2 The equivalent circuit is shown in (b). For simplicity, Figure 2 (b) The additional mutual impedances are not shown, and the traction loads are all represented using current sources. The equivalent circuit viewed from the traction bus to the system uses the Norton equivalent circuit. Assuming that V1 and V2 are very close, their additional mutual impedances are not zero. Therefore, the additional impedance matrix at this time is:

[0121]

[0122] To make matrix Z aR Reversible, let Z be... aR3 =1Ω.

[0123] Step 3: Obtain the additional impedance matrix of the contact wire.

[0124] Since the overhead contact line is not horizontally connected everywhere, the actual voltage drop caused by the traction load on the contact line is greater than the voltage drop calculated after combining the up and down contact lines into equal values. The additional matrix Z caused by the contact line is determined using equations (1) to (3) for the up and down traction loads of the power supply arm. aT Its order is equal to the number of traction loads (including the imaginary load at the end of the power supply arm). Figure 2 In (a), the additional impedance caused by the contact wire is shown below. Figure 2 (b)

[0125] Step 4: Use the impedance and admittance merging method to merge and equalize the conductors of the traction network; whereby the upstream and downstream contact networks are merged and referred to as conductor T, and the upstream and downstream return networks are merged and referred to as conductor H, to obtain the unit impedance matrix Z′ and unit admittance matrix Y′ of the traction network composed of conductor T and conductor H.

[0126] Step 5: Obtain the total additional impedance matrix.

[0127] In the power flow calculation of the traction network, the additional impedance matrix Z is used. aR Z aT By combining the results, we obtain the total additional impedance matrix:

[0128] Z a =ZaR +Z aT (4)

[0129] Considering the effect of additional impedance, the total additional impedance matrix Z can be considered as follows: a Connected between the overhead contact line and the traction load, by Figure 2 (b) can be obtained as follows Figure 2 (c) shows the equivalent circuit of the traction network.

[0130] For a fully parallel AT traction network, first, the additional impedance caused by the contact wire and return network is calculated. Then, using the single-row AT equivalent circuit method, its equivalent circuit at 25kV based on a simplified model can be obtained, such as... Figure 3 As shown.

[0131] The flowchart of the power flow calculation method for a traction power supply system based on a simplified traction network model of the present invention is as follows: Figure 4 As shown, it includes the following steps:

[0132] Step 1: Establish a simplified correlation matrix between the traction network section and the traction load.

[0133] The relationship between traction load and catenary nodes is described using a reduced incidence matrix (RIM). Rows in the RIM represent cross-sections, and columns represent branches (traction loads). The elements of the RIM are defined as follows:

[0134]

[0135] Figure 2 In (c), the RIM matrix M is:

[0136]

[0137] and Figure 5 In the middle, the positions of traction loads V1 and V2 coincide, therefore the RIM matrix M is:

[0138]

[0139] Step 2: Obtain the additional admittance matrix.

[0140] The correlation matrix representing the contact wire nodes and the newly added nodes caused by the additional impedance is as follows:

[0141]

[0142] Where k is the number of traction loads, E k×k It is a k-order identity matrix.

[0143] From the additional impedance matrix Z a Determine the additional impedance admittance matrix:

[0144]

[0145] Step 3: Use the chain circuit model of the traction network to establish the node admittance matrix Y of the traction power supply system. S0 .

[0146] Based on the unit impedance matrix Z′ and unit admittance matrix Y′ of the traction network composed of conductors T and H, a chain circuit model is used to model the traction power supply system. Ignoring additional impedance, the nodal admittance matrix Y′ of the traction power supply system is formed. S0 .

[0147] The traction network is modeled using a chain circuit model to obtain the system admittance matrix Y. S0 This method has two advantages: 1) there are only two conductors in each cross section; 2) the number of cross sections is also very small.

[0148] For example, such as Figure 2 In the circuit shown in (c), a traction power supply system supplies power to traction loads at three different locations on the same power supply arm from a traction substation. Using the above method to model the system, the system has four cross-sections: the traction substation, each traction load, and the end of the power supply arm.

[0149] Step 4: Incorporate the additional admittance matrix into the nodal admittance matrix Y S0 The nodal admittance matrix Y of the entire system is obtained. S .

[0150] Modify the system's node admittance matrix, and add the admittance matrix Y. a Include it; the method is: include the additional admittance matrix Y a Divide the network into appropriate blocks according to existing overhead contact line nodes and newly added nodes, and obtain...

[0151] If the traction substation is located without traction load, the node admittance matrix Y of the traction power supply system will be... s0 Also, divide it into appropriate blocks, and get Where Y0 is a complex number, Y 0t It is a row vector.

[0152] matrix Y a And matrix Y s0 By combining them appropriately, we obtain the following system node admittance matrix:

[0153]

[0154] Where H3 = [1, 0] t , It is the Kronecker product of the matrix.

[0155] If the traction substation is located under a traction load, then Y s0 =Y t ,but:

[0156]

[0157] Step 5: Obtain the node voltage equations of the traction power supply system.

[0158] Define H1 = [0, 1] t , If the traction substation is located without traction load, the node voltage equation of the traction power supply system is:

[0159]

[0160] If the traction substation is located under a traction load, then:

[0161]

[0162] Among them, I s To be with M2·I L Vectors of the same order, and the vector consisting of their first two elements is:

[0163] Step 6: Solve the node voltage equations of the traction power supply system.

[0164] The continuous linear power flow method is used to solve the nodal voltage equations (11) or (12) of the traction power supply system and achieve the convergence condition to obtain the traction network tangential voltage vector U. t and the newly added node voltage vector U v .

[0165] Step 7: Solve for the rail potential and traction network current distribution.

[0166] Use equation (13) to calculate the rail potential at the traction load:

[0167]

[0168] Suppose there are p conductors connected in parallel. We need to determine the current distribution across these p conductors; and write the voltage drop equations between two adjacent cross sections of the p conductors:

[0169]

[0170] Among them, E p×1 Let q be a p-dimensional column vector of all 1s, and let q be the combined equivalent conductor of the traction network, excluding the p parallel conductors.

[0171] We obtain the following from equation (14):

[0172]

[0173] The current distribution of p parallel conductors can be obtained from equation (15).

[0174] Similarly, it can be applied to Figure 3 The simplified equivalent circuit shown is used for power flow calculation based on the fully parallel AT traction network.

[0175] The present invention provides a short-circuit analysis method for a traction power supply system based on a simplified model of a direct-supply traction network, specifically as follows:

[0176] When a short circuit occurs in the traction power supply system, the traction grid voltage drops significantly. The control system of the electric locomotive / motor unit will block the converter trigger signal, causing the power of the electric locomotive / motor unit to be close to zero. Therefore, the impact of traction load is generally not considered when the traction power supply system is short-circuited.

[0177] When a short circuit occurs in the direct traction network, the circuit is as follows: Figure 6 As shown. When the rail potential is not a concern, the traction network is considered as a single conductor, and its unit impedance can be calculated using equation (16):

[0178]

[0179] Considering the effect of the additional impedance between the conductors, we can obtain the following: Figure 7 The equivalent circuit shown is shown.

[0180] At the short circuit point, the additional impedances caused by the contact wire and return network are Z, respectively. aT and Z aR Solving for the Thevenin equivalent circuit as seen from the traction bus to the system yields the following results: Figure 7 The equivalent circuit is shown. Wherein, and Z s These represent the electromotive force and impedance of the Thevenin equivalent circuit as viewed from the traction bus toward the power system.

[0181] Depend on Figure 7 The short-circuit impedance Z, viewed from the traction bus towards the short-circuit point, can be calculated. bus,m :

[0182] Z bus,m =Z e x+Z aT +Z aR (17)

[0183] The short-circuit current at the short-circuit point can be obtained using equation (18):

[0184]

[0185] The relay protection device is set based on the measured impedance of the traction bus and the calculated short-circuit current.

[0186] The present invention provides a short-circuit analysis method for a traction power supply system based on a simplified model of a fully parallel traction network, specifically as follows:

[0187] The equivalent circuit for a short circuit fault in a fully parallel AT traction network is as follows: Figure 8 As shown. To determine the short-circuit current, it is necessary to determine... Figure 8 The equivalent impedance between node b and node c in the return network. Figure 8 The backflow network between node b and node c is redrawn in Figure 9 .

[0188] Will Figure 9 Node c in the middle is grounded and connected to Figure 9 Number the nodes. In the backflow network, except for node b, the other nodes can be numbered arbitrarily.

[0189] Write the node admittance matrix of the system based on the connection relationships between the nodes. Figure 9 The nodal admittance matrix of the system shown is denoted as Y. RC .Will Figure 9 The nodes in the matrix are divided into two subsets: node b is subset 1, and the remaining nodes are subset 2. Dividing the admittance matrix into subset blocks yields:

[0190]

[0191] The superscript t indicates transpose.

[0192] Figure 9 The nodal voltage equations of the system shown are:

[0193]

[0194] Since no current is injected into the nodes in subset 2, I2 is a zero vector; by using Kron's elimination method to eliminate the nodes in subset 2 of equation (20), we get:

[0195]

[0196] The impedance between nodes b and c is:

[0197]

[0198] Calculate the short-circuit impedance Z at 25kV as seen from the traction bus to the short-circuit point. bus,m :

[0199] Z bus,m =z1′L+Z f +Z bc (23) The short-circuit current is:

[0200] The relay protection device is set based on the measured impedance of the traction bus and the calculated short-circuit current.

Claims

1. A simplified modeling method for traction networks, characterized in that, This includes unified chain circuit modeling of the traction power supply system, specifically comprising the following steps: Step 1: Obtain the unit impedance matrix Z and unit admittance matrix Y of the traction network; For a traction network consisting of n conductors, the unit impedance matrix and unit admittance matrix of the traction network are obtained using the Carson method or the mirror method, and the conductivity of the rails and the ground wire to the ground is included in the unit admittance matrix; the conductor merging algorithm is used to merge the contact wires and catenary wires in the same line, and the rails in the same line are merged, and the number of conductors after merging is n′, thus obtaining the unit impedance matrix Z and unit admittance matrix Y of the traction network; Step 2: Obtain the additional mutual impedance matrix of the return network; If p conductors in a traction network are laterally connected at regular intervals, then the p conductors are divided into two parts; one part consists of conductors into which current is injected or out, denoted as conductor C1, and the remaining conductors form conductor set C2; the conductors in C2 are merged and equalized, and the merged set is denoted as conductor C2; then conductors C1 and C2 are merged and equalized, and denoted as conductor C. Since conductors C1 and C2 are not cross-connected at every point, calculating power flow or short circuits based on conductor C will cause significant calculation errors. To reduce these errors, an additional self-impedance is introduced onto conductor C. where Z C1 and Z C2 are the self-impedances of the conductors C1 and C2, Z C12 is the mutual impedance between them, s is the transposition pitch of the p conductors, and m is the distance of the current injection point from the transposition pitch on one side. It can be seen that when m = 0 or m = s, the additional self-impedance Z aC = 0. To accurately describe the interaction between two current injection points, additional mutual impedances are introduced; a transposition line divides the p conductors into different sections; if the two current injection points belong to different sections, the additional mutual impedance Z aC = 0; otherwise, the additional mutual impedance between the two current injection points is calculated using equation (2): where s ij The coupling distance, referred to as the two-pull load, is calculated according to equation (3): The additional self-impedance, additional mutual-impedance of the reflux network are obtained by using formula (2), formula (3), and the additional impedance matrix Z is composed aC The order is equal to the number of current injection points. The traction net return network is divided into an uplink rail R1, the rest conductors Q including an uplink return line, a through ground line, a downlink rail, a return line and a through ground line, using the above method, an additional impedance matrix Z of the return network is obtained aR ; Step 3: Obtain the additional impedance matrix of the overhead contact line; As in step 2, the additional matrix Z caused by the catenary is determined using the formulae (1) to (3) aT ; Step 4: Use the impedance and admittance merging method to merge and equalize the conductors of the traction network; whereby the upstream and downstream contact networks are merged and referred to as conductor T, and the upstream and downstream return networks are merged and referred to as conductor H, to obtain the unit impedance matrix Z′ and unit admittance matrix Y′ of the traction network composed of conductor T and conductor H; Step 5: Obtain the total additional impedance matrix; In the current calculation of the traction network, the additional impedance matrix Z aR , Z aT is combined to obtain the total additional impedance matrix: Z a = Z aR + Z aT (4) total additional impedance matrix Z a coupled between the catenary and the traction load.

2. A method for calculating the power flow of a traction power supply system based on a simplified traction network model, based on the simplified traction network modeling method of claim 1, characterized in that, Includes the following steps: Step 1: Establish a simplified correlation matrix between the traction network section and the traction load; A simplified correlation matrix M is used to describe the relationship between traction load and catenary nodes; the rows of the simplified correlation matrix represent sections, and the columns represent branches, i.e., traction loads; the elements of the simplified correlation matrix are defined as follows: Step 2: Obtain the additional admittance matrix; The correlation matrix representing the contact wire nodes and the newly added nodes caused by the additional impedance is as follows: where k is the number of traction loads, E k×k is a k-order identity matrix; from the additional impedance matrix Z a determining an additional impedance admittance matrix: Step 3: Establishing the nodal admittance matrix Y of the traction power supply system using chain-link circuit model of the traction network S0 ; According to the unit impedance matrix Z' and the unit admittance matrix Y' of the traction network composed of the conductor T and the conductor H, the traction power supply system is modeled by using a chain circuit model, additional impedance is ignored, and a node admittance matrix Y of the traction power supply system is formed S0 ; Step 4: Incorporate the additional admittance matrix into the node admittance matrix Y S0 , to obtain the node admittance matrix Y S of the entire system. Modify the system's node admittance matrix, and add the admittance matrix Y. a Include it; the method is: include the additional admittance matrix Y a Divide the network into appropriate blocks according to existing overhead contact line nodes and newly added nodes, and obtain... If the traction substation is located without traction load, the node admittance matrix Y of the traction power supply system will be... s0 Also, divide it into appropriate blocks, and get Where Y0 is a complex number, Y 0t It is a row vector; The matrix Y a and the matrix Y s0 Suitably combined, the following system node admittance matrix is obtained: Where H3 = [1, 0] t , The Kronecker product of matrices; If the traction substation is located in a traction load, then Y s0 = Y t then: Step 5: Obtain the node voltage equations of the traction power supply system; Define H1 = [0, 1] t , If the traction substation is located without traction load, the node voltage equation of the traction power supply system is: If the traction substation is located under a traction load, then: Among them, I s To be with M2·I L Vectors of the same order, and the vector consisting of their first two elements is: Step 6: Solve the nodal voltage equations of the traction power supply system; The continuous linear flow method is used to solve the node voltage equation (11) or (12) of the traction power supply system, and the convergence condition is reached, to obtain the traction network cut surface voltage vector U t and the new node voltage vector U v ; Step 7: Solve for the rail potential and traction network current distribution; Use equation (13) to calculate the rail potential at the traction load: Suppose there are p conductors connected in parallel. We need to determine the current distribution across these p conductors; and write the voltage drop equations between two adjacent cross sections of the p conductors: wherein E p×1 is a p-dimensional all-ones column vector, and q is the combined equivalent conductor of the other conductors of the traction network, excluding the p conductors in parallel. We obtain the following from equation (14): The current distribution of p parallel conductors can be obtained from equation (15).

3. A short-circuit analysis method for a traction power supply system based on a simplified model of a direct-supply traction network, based on the simplified modeling method of the traction network described in claim 1, characterized in that, Specifically: When a short circuit fault occurs in the traction power supply system, the voltage of the traction network drops significantly. The control system of the electric locomotive / motor unit will block the converter trigger signal, resulting in the power of the electric locomotive / motor unit being close to zero. Therefore, the impact of traction load is generally not considered when the traction power supply system is short-circuited. When a short-circuit fault occurs in the traction network, if the rail potential is not a concern, the traction network can be considered as a single conductor, and its unit impedance can be calculated according to formula (16): At the short circuit point, the additional impedances caused by the contact wire and return network are Z, respectively. aT and Z aR Solve for the Thevenin equivalent circuit as seen from the traction bus to the system, where, and Z s These are the electromotive force and impedance of the Thevenin equivalent circuit as viewed from the traction bus toward the power system; The short-circuit impedance Z is determined from the traction bus to the short-circuit point bus,m : Z bus,m = Z e x + Z aT + Z aR (17) The short-circuit current at the short-circuit point can be obtained using equation (18): The relay protection device is set based on the measured impedance of the traction bus and the calculated short-circuit current.

4. A short-circuit analysis method for a traction power supply system based on a simplified model of a fully parallel traction network, based on the simplified modeling method of the traction network described in claim 1, characterized in that, Specifically: Draw the equivalent circuit when a short circuit fault occurs in the fully parallel AT traction network; For the return network in the equivalent circuit, the short-circuit current injection node is numbered as node b, and the node connected with the negative pole of the power supply is numbered as node c; the rest of the nodes are numbered arbitrarily; ground node c; write the node admittance matrix Y of the return network according to the connection relationship between the nodes RC , and node b is the first node; divide the nodes of the return network into two subsets, node b is subset 1, and the rest of the nodes are subset 2; block the admittance matrix according to the subsets, and the following is obtained: Wherein, the superscript t indicates transpose; The nodal voltage equations for the return network are: Since no current is injected into the nodes in subset 2, I2 is a zero vector; by using Kron's elimination method to eliminate the nodes in subset 2 of equation (20), we get: The impedance between nodes b and c is: Calculate the short-circuit impedance Z at 25kV as seen from the traction bus to the short-circuit point. bus,m : WITH bus,m =z1′L+Z f +Z bc (23) The short-circuit current is: The relay protection device is set based on the measured impedance of the traction bus and the calculated short-circuit current.

Images