Method for constructing simulation circuit of transient potential rise of GIS considering grounding grid topology and shell broadband transmission characteristics

Abstract

The application discloses a GIS transient potential rise simulation circuit construction method considering grounding net topology and shell broadband transmission characteristics, and belongs to the technical field of power system electromagnetic transient simulation. Firstly, a disconnecting switch multi-arc model containing a fracture dynamic capacitor, a segmented arc resistance and a dynamic arc burning-out criterion is constructed to generate a special fast transient excitation conforming to a physical process; secondly, a unit length GIS bus shell three-dimensional electromagnetic model is established, two-port S parameters are extracted through 0-100MHz broadband frequency sweep, the two-port S parameters are converted into admittance parameters, and a PI type broadband equivalent circuit is obtained through vector matching method. The application simultaneously considers the shell broadband transmission characteristics and the grounding net topology propagation effect, widens the frequency domain application range of simulation, improves the simulation precision and engineering reliability of the GIS shell transient potential rise, and can be applied to substation grounding safety evaluation and electromagnetic compatibility design.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic transient simulation technology for power systems, specifically a method for constructing a GIS transient potential rise simulation circuit that takes into account the grounding grid topology and the broadband transmission characteristics of the casing. Background Technology

[0002] During the opening and closing process of gas-insulated metal-enclosed switchgear (GIS), the contact movement speed is relatively slow, making the break gap prone to multiple breakdowns and reignitions. Each breakdown generates a steep-fronting, wide-spectrum ultra-fast transient overvoltage (VFTO) and ultra-fast transient overcurrent (VFTC). These transient signals propagate along the GIS busbar and form complex high-frequency conduction paths through the core wires, the inner and outer surfaces of the casing, the support structure, and the grounding grid. The resulting transient potential rise in the GIS casing is a core basis for substation grounding safety assessment and electromagnetic compatibility design.

[0003] Currently, simulation models for transient potential rise in GIS enclosures generally employ the three-wave impedance theory or calculate lumped parameter and quasi-static distributed parameter circuits based on geometric dimensions. The GIS busbar is divided into a three-layer structure: core wire, inner shell layer, and outer shell layer. The propagation process of transient signals is characterized by the capacitance of the core wire to the inner shell layer, the capacitance of the outer shell layer to ground, and the inductance and resistance parameters of the grounding lead. This type of model has an intuitive structure, is easy to implement, and has certain engineering applicability in low-frequency transient analysis.

[0004] However, the shell parameters in these models are mostly derived from low-frequency or quasi-static formulas, making it difficult to cover the wide frequency range of tens of kHz to 100 MHz for the main energy distribution of VFTO and transient ground potential rise. This makes it impossible to accurately reflect the axial propagation loss, port reflection characteristics, and ground discharge patterns of the shell under high-frequency operating conditions. Furthermore, existing models often simplify the grounding system to a single centralized grounding resistor, ignoring the fact that the grounding grid is essentially a topological network composed of horizontal grounding electrodes, vertical grounding electrodes, grounding down conductors, soil resistivity, and distributed parameters. This fails to reflect the propagation delay, waveform reflection, and node shunting effects of transient currents in the grounding grid, easily leading to deviations in the calculation results of the transient potential rise amplitude and the potential difference at key insulation breaks in the GIS shell. In summary, existing simulation models cannot simultaneously consider the wideband transmission characteristics of the shell and the propagation effects of the grounding grid topology, resulting in limited simulation accuracy for GIS shell transient potential rise, which is insufficient to meet the needs of high-precision engineering evaluation. Therefore, we propose a method for constructing a GIS transient potential rise simulation circuit that considers both the grounding grid topology and the wideband transmission characteristics of the shell, in order to alleviate or solve the above problems.

[0005] The information disclosed above in this background section is only for enhancing the understanding of the background section of this invention, and therefore may include prior art that is not known to those skilled in the art. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method for constructing a GIS transient potential rise simulation circuit that takes into account the grounding grid topology and the broadband transmission characteristics of the casing. This method solves the problem that existing models cannot simultaneously take into account the broadband transmission characteristics of the casing and the propagation effect of the grounding grid topology, resulting in limited simulation accuracy of GIS casing transient potential rise and difficulty in meeting the requirements of high-precision engineering evaluation.

[0007] To achieve the above objectives, this invention provides a method for constructing a GIS transient potential rise simulation circuit that considers the grounding grid topology and the broadband transmission characteristics of the casing. The method is based on a pre-defined GIS substation basic circuit model and includes the following steps:

[0008] S1. Construct a multiple arcing model for GIS disconnect switches to simulate the generation of ultra-fast transient overvoltage VFTO and ultra-fast transient overcurrent VFTC. The multiple arcing model is constructed by introducing dynamic capacitance of the break, establishing a segmented arc resistance model, and setting arcing and extinction criteria.

[0009] S2. Construct a broadband transmission model of a unit length GIS busbar shell. First, establish an electromagnetic model of a unit length GIS busbar shell, perform a 0-100MHz frequency sweep on the model, extract the S-parameters obtained from the frequency sweep, convert the S-parameters into admittance parameters, and use the vector matching method to obtain the equivalent circuit of the unit length shell.

[0010] S3. In the circuit model of a 330kV voltage level GIS substation, the original inner and outer topologies of the GIS shell are replaced with the obtained equivalent circuit of the shell per unit length, while retaining the topologies of the grounding down conductor, grounding grid and vertical grounding electrode, to obtain the transient ground potential rise and VFTO waveform.

[0011] S4. Compare the replaced circuit waveform with the original circuit waveform in time windows of 2ms, 10ms and 40ms to verify the accuracy and applicability of the broadband transfer model.

[0012] Preferably, in step S1, the input capacitance of the transformer equivalent model is calculated according to the formula... Calculate, where, =3, =540, when adapting to a 330kV voltage level, the inlet capacitor is 2000pF and the inductor is 20mH.

[0013] Preferably, in step S1, the equivalent capacitance of the grounding switch adapted to a 330kV voltage level is 210pF; the wave impedance Z1 of the conductive rod of the GIS busbar is in the range of 60-80Ω, and the wave impedance Z1 of the conductive rod is calculated according to the formula... Calculate, where, The inner radius of the GIS shell; Indicates the outer radius of the busbar conductor rod; the outer casing wave impedance of the GIS busbar. According to the formula Calculate, where, The height of the GIS outer casing conductor from the ground. The outer radius of the GIS shell.

[0014] Preferably, in step S1, the dynamic capacitance model formula for the break dynamic capacitance is: Where C0, C1, and C2 are constants obtained by fitting the measured capacitance data of a 330kV disconnector, and t is time.

[0015] Preferably, in step S1, the segmented arc resistance model includes at least three stages: pre-breakdown, steady-state combustion, and arc extinction. During the pre-breakdown stage, the equivalent resistance of the arc channel decreases over time; during the steady-state combustion stage, a smaller steady-state resistance can be used; and during the arc extinction stage, the resistance rapidly recovers to a high resistance. One possible implementation is as follows:

[0016] Pre-breakdown stage: Arc resistance decreases exponentially, as shown in the formula: Among them, R0 and R a Here are the initial resistance parameters, and T is the time constant;

[0017] Steady-state combustion stage: When time t=t1, the arc resistance takes the steady-state value R. a =2Ω;

[0018] Arc extinction stage: When time t=t2, neglecting the arc gap recovery voltage, the arc resistance formula is as follows: , where R a0 The initial resistance for arc extinguishing is T, and the time constant is T.

[0019] Preferably, in step S1, the specific settings for the arc ignition criterion and the arc extinction criterion are as follows:

[0020] Arc ignition criterion: The dynamic model of breakdown voltage is as follows

[0021] Among them, U K For the dynamic model of the breakdown voltage of the disconnector in the open state, U g Dynamic model of breakdown voltage under closed state, A, B, KV K KD, KV g The fitting coefficients are used; the breakdown field strength of SF6 gas within the GIS is calculated using the formula... The calculation is performed, where Q is the breakdown coefficient related to SF6 gas and electrode structure, and p is the rated pressure of SF6 gas.

[0022] Arc extinction criterion: The arc is extinguished when the product of the current value at the previous moment and the current value at the current moment is less than 0, and the difference between the current values ​​at the two moments is less than a constant K.

[0023] Preferably, in step S2, the specific parameters of the frequency sweep measurement method are: using a sinusoidal voltage signal of 0MHz-100MHz as excitation to measure the broadband transmission characteristics of the GIS busbar shell.

[0024] Preferably, in step S2, the housing broadband transfer model is based on the two-port S-parameters obtained by sweeping the frequency from 0-100MHz. For the reference impedance matrix Z0 or the reference admittance matrix Y0, the transformation from S-parameters to admittance parameter matrix is ​​as follows:

[0025]

[0026] Where I is the identity matrix, s=jω, ω=2πf. When the reference impedance of each port is 50Ω, Y0 can be taken as (1 / 50)S. For the 0MHz endpoint, the low-frequency limit point or the equivalent DC / quasi-static point extrapolated from the adjacent low-frequency point is usually used.

[0027] Preferably, in step S2, for each element Y of the admittance matrix... ij (s) Perform vector matching rational function fitting. Its general expression is:

[0028]

[0029] Among them, P k For common poles, r ijk For the residue, d ij e is a constant term. ij The term is proportional, and N is the total number of poles. During the fitting process, the poles should be located in the left half of the complex plane; when poles appear in the right half of the plane, stability can be ensured by pole flipping or reselecting the fitting order, and passivity checks should be performed to ensure that the integrated circuit does not generate non-physical energy in the ATP-EMTP transient calculation.

[0030] Preferably, the consistency between the fitted curve and the real and imaginary parts of the frequency sweep result is evaluated separately. The correlation coefficient (SCC) can be used to evaluate the real part fitting, and the coefficient of determination (R²) can be used to evaluate the imaginary part fitting. 2 evaluate:

[0031]

[0032] Where, x n For the real or imaginary part samples obtained from the frequency sweep, y n or f nThe values ​​shown are the fitted values, and the values ​​with horizontal lines represent the sample mean. The SCC and R² thresholds can be set in the range of 0.8-0.95; higher thresholds indicate a more consistent fit between the fitted curve and the swept frequency data.

[0033] Preferably, in step S2, when synthesizing the PI-type equivalent network based on the two-port admittance parameters, the derivation process is as follows: A unit-length GIS shell is considered as a linear two-port network between the inlet and outlet annular ports. The voltages at both ports are referenced to the external reference ground of the shell, and the port currents are positive in the direction flowing into the two-port network. After obtaining the admittance parameter matrix from the frequency-sweep S-parameter transformation, the two-port network satisfies:

[0034]

[0035] Where V1 and V2 are the voltages of the inlet and outlet shell nodes relative to the reference ground, respectively, and I1 and I2 are the currents flowing into the network from the two ports. For a passive, reciprocal, and geometrically approximately symmetrical shell per unit length, theoretically, Y... 12 =Y 21 In actual frequency sweeping and numerical fitting, due to grid, port truncation, and fitting errors, it usually manifests as Y 12 ≈Y 21 .

[0036] When the two-port network is equivalent to a PI-type network, the admittance of the parallel branch to ground at the input end is denoted as Y. p1 (s), the admittance of the parallel branch to the ground at the export end is denoted as Y. p2 (s), the series branch impedance between the inlet and outlet is denoted as Z. s (s), the series branch admittance is denoted as Y s (s)=1 / Z s (s). According to the node current relationship, the inlet current and the outlet current are respectively:

[0037]

[0038]

[0039] Comparing the above PI-type network node equations with the two-port admittance matrix term by term, we can obtain:

[0040] ,

[0041] Therefore, when the two ports are strictly reciprocal, we can directly obtain:

[0042]

[0043] When the numerical calculation yields Y 12 With Y 21When they are not completely equal, to avoid directly substituting numerical errors into the equivalent circuit, we can first take the average of the reciprocity terms, letting:

[0044] Therefore, the PI-type equivalent network is a three-branch network uniquely corresponding to the port current equations of the two-port admittance matrix; where the off-diagonal admittance Y 12 Y 21 The series propagation branch between the two ports is determined, and the parallel discharge branch to ground at both ends is obtained after deducting the contribution of the series branch from the diagonal admittance.

[0045] Furthermore, when the vector matching yields Y p1 (s), Y p2 (s) and Z s (s) If the signal can be adequately approximated by low-order passive components within the 0-100MHz range, then the series branch can be written in the form of a series resistor-inductor, and the parallel branch can be written in the form of a resistor and capacitor in parallel:

[0046]

[0047] in, and The equivalent loss and equivalent inductance per unit length of the shell along the axial propagation path are respectively characterized; , Characterizes the broadband discharge and energy storage characteristics of the inlet shell node relative to the reference ground; , Characterizes the broadband discharge and energy storage characteristics of the outlet shell node relative to the reference ground. Since the admittance of the parallel branch is the sum of the conductance and capacitance terms, C... p1 With R p1 They must be connected in parallel, C p2 With R p2 They must also be connected in parallel; if they are mistakenly connected in series, the branch admittance frequency characteristic will change from G... p +sC p It becomes 1 / (R) p +1 / (sC p This is inconsistent with the PI network derived from the admittance matrix.

[0048] The aforementioned low-order parameters can be extracted from the fitted frequency domain function within the target frequency band using the least-squares method. For the series branch, Z can be set... s The real part of (jω) is approximately R s The imaginary part, when divided by ω, is approximately L. s For parallel branches, Y can be set to... p The real part of (jω) is approximately G. p =1 / R p The imaginary part, when divided by ω, is approximately C. pThe set of feasible parameter extraction criteria is as follows:

[0049]

[0050] Where, ω q For the frequency sweep angular sampling point, w q Here, R represents the frequency point weights, and M represents the number of frequency points involved in the fitting. When higher frequency bands have a more significant impact on the transient ground potential rise (TGPR) or VFTO peak value, the weights of the corresponding frequency bands can be appropriately increased; when the goal is to maintain a small average error across the entire 0-100MHz frequency band, uniform weighting can be used. To ensure the stability of the ATP-EMTP transient calculation, the extracted R... s L s R p1 R p2 C p1 C p2 All values ​​should be positive; if a negative value appears, the fitting order should be increased, passive constraints should be reapplied, or a multi-branch Foster / Cauer equivalent network should be used instead of using non-physical parameters directly.

[0051] In one embodiment of the present invention, after the above-mentioned conversion of S-parameters to admittance parameters, vector matching, PI-type network synthesis, and low-order passive parameter extraction, the PI-type broadband equivalent circuit per unit length of the shell is taken as L. s =0.00119mH, C p1 =1.12163×10 -5 μF, C p2 =1.40175×10 -5 μF, R p1 =109.58Ω, R p2 =105.50Ω. This set of values ​​indicates that the ground discharge branches at both the inlet and outlet ends of the casing are resistor-capacitor parallel structures, and the parameters at both ends are allowed to be not completely equal, in order to reflect the slight asymmetry caused by port cutoff, local structure and fitting results.

[0052] The above parameters are equivalent parameters per unit length of the shell. When the GIS busbar shell is represented by multiple cascaded segments, the unit length parameters should be consistently converted according to the actual segment lengths, or the same unit length modules should be repeatedly cascaded. When replacing the busbar core wires in an engineering project, the C0 coupling relationship between the busbar core wires and the inner layer of the shell, as well as the grounding grid topology connection relationship, should be retained to ensure that the broadband shell model and the original ATP-EMTP substation model are consistent in terms of node connection.

[0053] Preferably, in step S3, the grounding grid topology is not simplified to a single grounding resistor, but retains branches such as grounding down conductors, horizontal grounding electrodes, and vertical grounding electrodes. These branches can be represented as follows:

[0054]

[0055] Among them, Z g (s) can represent the series impedance of the grounding lead or horizontal grounding electrode, Y b (s) can represent the discharge admittance of the vertical grounding electrode to the distant ground. By preserving the above topology, it is possible to obtain the potential propagation difference between different grounding points, rather than just obtaining the average potential on an equivalent grounding resistance.

[0056] Preferably, in step S4, the original shell topology model and the replaced shell broadband equivalent model are run respectively to extract TGPR and VFTO waveforms under time windows of 2ms, 10ms, and 40ms. The comparison results are used to verify the consistency of the replaced model in short-term high-frequency oscillation, long-term attenuation trend, and VFTO macroscopic propagation trend. When the replaced model is consistent with the original model in the main trend of VFTO, and can reflect the influence of the shell broadband transmission on the high-frequency oscillation and attenuation of TGPR, it indicates that the broadband equivalent shell model has engineering applicability.

[0057] Compared with the prior art, the beneficial effects of the present invention are:

[0058] This invention constructs a wideband equivalent circuit for the GIS shell based on the 0-100MHz full-band sweep frequency S-parameter and vector matching method. This circuit can reflect the axial propagation loss, port reflection characteristics, and ground discharge law of the GIS shell under high-frequency operating conditions, thus broadening the frequency domain applicability of the simulation model. At the same time, it retains the complete topology of the grounding grid instead of simplifying it to a single resistor, which can realistically reflect the propagation delay, waveform reflection, and node shunting effect of transient current in the grounding grid. Combined with a multiple arcing excitation model of the disconnecting switch with a clear physical mechanism, the three factors work together to improve the calculation accuracy of the transient potential rise amplitude, oscillation characteristics, and attenuation law of the GIS shell, avoiding the underestimation or misjudgment of high-frequency transient response by traditional models.

[0059] The broadband equivalent circuit per unit length shell constructed in this invention adopts a standard PI-type passive network structure with clear parameters and connections. It can be directly embedded into existing mainstream electromagnetic transient simulation software without complex secondary development, making it convenient for engineers to replace and upgrade existing substation simulation models. At the same time, through a waveform comparison and verification system with multiple time windows, the reliability of the model can be verified from multiple dimensions, from short-time high-frequency oscillations and medium-time attenuation trends to long-time stability characteristics. The simulation results can provide more accurate quantitative basis for the design of GIS substation grounding systems, electromagnetic compatibility protection of secondary equipment, and operation and maintenance safety assessment.

[0060] The above overview is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the invention will become readily apparent from the accompanying drawings and the following detailed description. Attached Figure Description

[0061] Figure 1 This is the equivalent schematic diagram of the 330kV GIS system upon which the embodiments of the present invention are based;

[0062] Figure 2 This is a schematic diagram of the equivalent circuit of the dynamic capacitance of the disconnecting switch of the present invention;

[0063] Figure 3 This is a resistance variation curve of the segmented arc model of the present invention;

[0064] Figure 4 This is a schematic diagram illustrating the principle of the arc extinction criterion of the present invention;

[0065] Figure 5 This is a flowchart illustrating the construction of the multiple arcing model of the disconnector switch of the present invention;

[0066] Figure 6 This is a schematic diagram of the multiple arcing model of the disconnector constructed in this invention;

[0067] Figure 7 This is a schematic diagram of a three-dimensional electromagnetic model of the GIS busbar shell of the present invention;

[0068] Figure 8 This is a schematic diagram of the PI-type broadband equivalent circuit structure of the GIS busbar shell per unit length according to the present invention;

[0069] Figure 9 This is a schematic diagram of the equivalent topology of the H-shaped steel bracket of the present invention;

[0070] Figure 10 This is a schematic diagram of the simulation circuit topology for GIS transient potential rise, taking into account the grounding grid topology and the broadband transmission characteristics of the casing, according to the present invention.

[0071] Figure 11 A schematic diagram of the circuit topology for simulating the transient potential of the original GIS busbar casing;

[0072] Figure 12 This is a comparison of the transient ground potential rise waveforms of the original topology and the improved topology under a 2ms time window in this invention;

[0073] Figure 13 This is a comparison diagram of the ultrafast transient overvoltage waveforms of the original topology and the improved topology under the 2ms time window in this invention;

[0074] Figure 14 This is a comparison diagram of the ultrafast transient overvoltage waveforms of the original topology and the improved topology under a 10ms time window in this invention;

[0075] Figure 15 This is a comparison of the transient ground potential rise waveforms of the original topology and the improved topology under a 10ms time window in this invention;

[0076] Figure 16 This is a comparison of the ultrafast transient overvoltage waveforms of the original topology and the improved topology under a 40ms time window in this invention;

[0077] Figure 17 This is a comparison of the transient ground potential rise waveforms of the original topology and the improved topology under a 40ms time window in this invention. Detailed Implementation

[0078] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. It should be noted that the drawings are schematic and not illustrated to scale. For clarity and convenience, the relative sizes and proportions of the parts shown in the drawings have been exaggerated or reduced in size. Any size is only exemplary and not limiting.

[0079] Example 1

[0080] This embodiment uses a 330kV GIS substation as the application scenario, and describes the construction method of a GIS transient potential rise simulation circuit that takes into account the grounding grid topology and the broadband transmission characteristics of the casing. The construction flowchart is attached. Figure 5 As shown, it includes the following steps:

[0081] S1. Construct a multiple arcing model of the GIS disconnector as the excitation source for VFTO and VFTC. The multiple arcing model is constructed by introducing a time-varying dynamic capacitance of the break, a segmented arc resistance model including the pre-breakdown stage, the steady-state combustion stage, and the arc extinction stage, as well as arcing and arc extinction criteria; the equivalent schematic diagram is shown below. Figure 1 As shown. The circuit mainly consists of a transformer, grounding switch, circuit breaker, GIS busbar, and electronic current transformer.

[0082] Specifically, for the transformer equivalent model, in the high-frequency case, the transformer is equivalent to a structure of inductor and capacitor connected in parallel. The method for calculating the transformer's input capacitance is as follows:

[0083]

[0084] For the 330kV voltage level, k=3, n=540. The input capacitor is 2000pF, and the inductance is 20mH.

[0085] For the equivalent model of the grounding switch, it is equivalent to a grounding capacitor. In this embodiment, the equivalent capacitor is taken as 210pF. For the equivalent model of the circuit breaker, a single-port equivalent model in the closed state is selected. For the equivalent model of the GIS busbar, it is simplified to a uniform lossless line based on the Bergeron calculation method. The formulas for calculating the wave impedance of the conductor and the outer shell in the GIS busbar are as follows:

[0086]

[0087]

[0088] in, The wave impedance of the GIS conductive rod; The inner radius of the GIS shell; The outer radius of the busbar conductor rod; The wave impedance of the GIS casing; The height of the conductor above the ground; This is the outer radius of the GIS shell. In this embodiment, we select... The value ranges from 60 to 80 Ω.

[0089] Regarding the dynamic capacitance of the disconnect switch, the disconnect capacitance C1 generated when the moving and stationary contacts of the disconnecting switch are not fully closed is shown in the attached figure. Figure 2 As shown. Its time-varying characteristics are introduced, and the formula used for its dynamic capacitance model is:

[0090]

[0091] Where C0, C1, and C2 are constants obtained by fitting the measured capacitance data of a 330kV disconnector, and t is time.

[0092] The segmented arc resistance model covers three stages: pre-breakdown, steady-state combustion, and arc extinction. Its development curves are shown in the attached figure. Figure 3 As shown.

[0093] During the pre-breakdown stage, the arc resistance decreases exponentially, as shown in the formula:

[0094]

[0095] Among them, R0, R a Here are the initial resistance parameters, and T is the time constant.

[0096] During the steady-state combustion stage, at time t=t1, the disconnecting switch is completely broken down, and the arc resistance reaches its minimum steady-state value. In this embodiment, R is taken as... a =2Ω.

[0097] During the arc extinction phase, when t=t2, the arc current is zero, and the external circuit no longer supplies energy to the arc, thus entering the arc extinction phase. The equations used are:

[0098]

[0099] Where g is the arc conductance, u h P is the arc potential gradient. s Let τ be the arc heat dissipation power, and τ be the arc time constant.

[0100] Let Ra(t) = 1 / g, and we can continue to know:

[0101]

[0102] Based on the above formula, the relationship between arc resistance, time, and recovery voltage is as follows:

[0103]

[0104] When uh=0 and the arc gap recovery voltage is ignored, the arc extinction stage can be represented as:

[0105]

[0106] Among them, R a0 Let T be the initial resistance for arc extinguishing, and T be the arc time constant.

[0107] The arcing criterion is based on a dynamic model of the breakdown voltage U0. The dynamic model formula for the breakdown voltage over time in the open and closed states of the disconnector switch is as follows:

[0108]

[0109] Among them, A, B, Kv K K D Kv G The fitting coefficients are used. The empirical formula for the critical breakdown field strength of SF6 gas in GIS is:

[0110]

[0111] Where Q is a constant and p is the rated pressure of SF6 gas.

[0112] The arc extinction criterion is based on energy balance theory. The arc is considered extinguished when the product of the current value at the previous moment and the current value at the current moment is less than zero, and the difference between the current values ​​at the two moments is less than a constant K. (See attached figure.) Figure 4 As shown.

[0113] The multiple arcing model of the disconnector constructed by combining the above modules is attached. Figure 6As shown, VFTO / VFTC excitation is generated by realistically simulating multiple arcing phenomena during the operation of the disconnector switch.

[0114] S2. Establish a three-dimensional electromagnetic model of the GIS busbar shell of unit length and perform a 0-100MHz frequency sweep to extract the two-port S-parameters; convert the S-parameters into an admittance Y-parameter matrix and perform vector matching rational function fitting; based on the fitting results, construct a PI-type broadband equivalent circuit including axial series branches and parallel branches to ground at both ends using the circuit synthesis method; construct a broadband transfer model of the shell of unit length from the three-dimensional electromagnetic model, as shown in the appendix. Figure 7 As shown.

[0115] Specifically, this 3D model is based on the geometry of a unit-length shell, without altering the shell's physical structure. Two impedance ports are set at the shell's inlet and outlet ends for extracting broadband transmission characteristics. The inlet port injects power into the shell from the outer ring, while the outlet port leads out from the rear of the shell to the outer ring. Both ports are normalized to the same reference impedance, which is uniformly 50Ω, as shown in the attached figure. Figure 7 As shown.

[0116] The model was swept from 0 to 100 MHz to obtain the two-port S-parameters within the frequency band. During frequency sweeping, at each frequency point f q We obtain a two-port S-parameter matrix. When port 1 is the inlet circular port and port 2 is the outlet circular port, then S... 11 Indicates reflection at the entrance end, S 22 Indicates reflection at the outlet end, S 21 S represents the forward flow from the inlet to the outlet. 12 This indicates the reverse flow from the exit to the entrance. Specifically:

[0117]

[0118] Converting S-parameters to admittance parameters: For a reference impedance matrix Z0 or a reference admittance matrix Y0, the transformation from S-parameters to admittance parameter matrices is as follows:

[0119]

[0120] Where I is the identity matrix, s = jω, ω = 2πf. When the reference impedance at each port is 50Ω, Y0 can be taken as (1 / 50)S.

[0121] For each element of the admittance matrix The general expression for vector matching rational function fitting is:

[0122]

[0123] in, For common poles, To leave a residue For constant terms, is the proportional term, and N is the total number of poles.

[0124] For passive electromagnetic structures, the fitted poles should be located in the left half of the complex plane; if poles appear in the right half, the corresponding time-domain response will increase with time and cannot be directly used for ATP-EMTP transient simulations. Unstable poles can be handled by pole flipping or refitting.

[0125]

[0126] Evaluate the consistency between the fitted curve and the real and imaginary parts of the frequency sweep results, respectively. The correlation coefficient (SCC) can be used to evaluate the fit of the real part, while the coefficient of determination (R²) can be used to evaluate the fit of the imaginary part. 2 evaluate:

[0127]

[0128] The difference between the fitted curve and the swept-frequency admittance data is evaluated using the normalized mean square error:

[0129]

[0130] Where M is the number of sweep points and F is the Frobenius norm of the matrix. This index can simultaneously evaluate Y. 11 Y 12 Y 21 Y 22 The overall fit of the four elements.

[0131] After fitting, a passivity check is performed. The shell structure obtained by frequency sweep is itself a passive structure, and the fitted admittance function should also remain passive. For positive real admittance functions, the positive semi-definiteness of the Hermitian part in the frequency domain can be used as one of the checking criteria.

[0132]

[0133] The superscript H indicates conjugate transpose. This represents the smallest eigenvalue.

[0134] The higher-order rational function is further transformed into a lower-order R, L, C equivalent network. When the main behavior of the shell per unit length in the target frequency band can be described by a series propagation branch and two parallel branches to ground, a PI-type network is adopted. When the error is large, a Foster or Cauer branch is added to the PI-type branch.

[0135]

[0136] The GIS shell per unit length is considered as a linear two-port network between the inlet and outlet annular ports. Let the inlet port voltage be V1, the outlet port voltage be V2, and the port currents I1 and I2 be positive in the direction of flowing into the two-port network. Then the two-port admittance equation is:

[0137]

[0138] The two-port network is equivalent to a PI-type network, and the admittance of the parallel branch to ground at the inlet is Y. p1 (s), the admittance of the parallel branch to ground at the export end is Y p2 (s), the admittance of the series branch between the inlet and outlet is Y. s (s), corresponding to a series impedance of Z s (s)=1 / Y s (s). Based on the node current relationship, we can obtain:

[0139]

[0140] Comparing the above PI network node equations with the two-port admittance matrix term by term, we obtain:

[0141]

[0142] Therefore, for strictly reciprocal networks, we have:

[0143]

[0144] When Y 12 When Y21 is not exactly equal to Y21, the average reciprocity term Y can be defined. m :

[0145]

[0146] When a low-order PI network can meet the error requirements within the 0-100MHz frequency band, the series branch can be written as a resistor-inductor series form, and the branch with both ends to ground can be written as a resistor-capacitor parallel form.

[0147]

[0148] in, and The equivalent loss and equivalent inductance per unit length of the shell along the axial propagation path are respectively characterized; , Characterizes the broadband discharge and energy storage characteristics of the inlet shell node relative to the reference ground; , Characterizes the broadband discharge and energy storage characteristics of the outlet shell node relative to the reference ground.

[0149] Will and When connected in series, its admittance becomes:

[0150]

[0151] Low-order parameters can be extracted from the fitted frequency domain function using a least-squares approach. For the series branch, let Zs(jω) approximate R. s +jωL s For parallel branches, let Y p (jω) approaches G p +jωC p The parameter extraction criteria are as follows:

[0152]

[0153] in, For frequency point weights. The extracted parameters should satisfy R... s >0, L s >0、R p1 >0、R p2 >0, C p1 >0, C p2 >0.

[0154] In this embodiment, the equivalent circuit parameters of the PI-type wideband housing per unit length are:

[0155]

[0156] S3. Retain the main propagation branch of the bus core wire and the coupling capacitance from the core wire to the inner layer of the shell in the basic circuit model of the GIS substation, and replace the original lumped parameter topology of the inner and outer layers of the GIS shell with the PI-type broadband equivalent circuit; retain the grounding grid distributed parameter branch composed of the grounding down conductor, horizontal grounding electrode and vertical grounding electrode, and electrically connect the PI-type broadband equivalent circuit with the retained grounding grid topology.

[0157] Specifically, the lumped parameter topology of the inner and outer shell layers in the original 330kV GIS substation model is replaced with the aforementioned PI-type broadband equivalent circuit, while retaining the complete grounding grid topology. The original shell topology is attached. Figure 11 As shown. The shell topology replacement steps are as follows:

[0158] 1. Identify the inlet node, outlet node, and ground reference node for each segment of the GIS busbar shell in the original model;

[0159] 2. Retain L0 between core wire nodes and C0 from the core wire to the inner layer of the housing;

[0160] 3. Delete or bypass the original inner and outer layers of the outer casing to simplify the LC branch;

[0161] 4. Connect between the inlet and outlet nodes of the shell. Side road;

[0162] 5. Connect to ground at the inlet shell node. and Parallel branch lines;

[0163] 6. Connect to ground at the outlet shell node. and Parallel branch lines;

[0164] 7. Continue connecting this casing node to the grounding down conductor, horizontal grounding electrode, and vertical grounding electrode network, as shown in the attached diagram. Figure 10 As shown.

[0165] When the actual GIS bus length is The unit length module length is At that time, adopt Modules are cascaded; during cascading, the node distribution characteristics between modules and the coupling relationship between the core wire and the inner layer of the housing are preserved. When parameter scaling is used, the axial inductance is proportional to the length, the capacitance to ground is proportional to the length, and the bleeder resistor is inversely proportional to the length. Specifically:

[0166]

[0167] The grounding grid topology is not simplified to a single grounding resistor; grounding down conductors, horizontal grounding electrodes, and vertical grounding electrode branches are retained. Specifically:

[0168]

[0169] in, It can represent the series impedance of a grounding down conductor or a horizontal grounding electrode. It can represent the equivalent admittance of a vertical grounding electrode or a soil discharge branch.

[0170] For the horizontal grounding electrode in the grounding grid, this invention discretizes it into several rectangular horizontal grounding electrode units. Each horizontal grounding electrode unit consists of four vertex nodes and four horizontal grounding electrode branches connecting adjacent vertex nodes, wherein the horizontal grounding electrode branches arranged along the unit boundary use series resistors. With series inductor Characterization. This unit is not an isolated circuit unit, but a repeatable unit in the two-dimensional topology of the grounding grid; multiple horizontal grounding electrode units can continue to be cascaded in the up, down, left, and right directions, and adjacent units are interconnected by sharing vertex nodes or sharing side branches, thereby forming a complete grounding grid topology network.

[0171] Let the length of the e-th horizontal grounding electrode branch be l. e The equivalent radius of the conductor is A SGiven that the conductor resistivity is ρs, the permeability is μs, and the angular frequency is ω, the series impedance of this horizontal grounding electrode branch can be expressed as:

[0172] Under low-frequency or DC approximation conditions, the resistance of the horizontal grounding electrode branch is: Under high-frequency transient conditions such as VFTO / TGPR, the conductor skin effect needs to be considered. The skin depth is: When the current is mainly concentrated on the outer surface of the conductor with a thickness of approximately δ s When the conductor is within a thin layer, the effective conductive area can be approximated as the product of the conductor's perimeter and the skin depth, i.e.: Therefore, the resistance of the horizontal grounding electrode branch under high-frequency conditions can be approximated as: To avoid the formula failing at low frequencies, we can take: .

[0173] Horizontal grounding branch inductance L s,e This mainly reflects the magnetic field energy storage formed by the branch current and its soil return path. For a burial depth of h... b A slender, horizontal cylindrical conductor can be approximated in engineering terms based on wire inductance and the method of images: , where h b Let μ0 be the burial depth of the horizontal grounding electrode and μ0 be the free permeability. If the branch lengths of the horizontal grounding electrode unit are different along the x and y directions, they can be denoted as R respectively. s,x L s,x and R s,y L s,y Considering that the horizontal grounding electrode is buried h below the ground surface. b At this location, there are local vertical current diffusion and return paths between the four vertex nodes of the horizontal grounding electrode unit and the reference ground. To characterize the effect of burial depth on the high-frequency transient potential distribution, a burial depth correction grounding branch is set at each vertex node, and its impedance is expressed as: Where k represents the four vertices of the rectangular horizontal grounding electrode unit. L b,k To characterize the magnetic field energy storage effect caused by the local return path along the burial depth near the vertex, it can be calculated based on the equivalent vertical conductive path length h. b Approximation: , where a b The equivalent radius of the local current diffusion region near the vertex can be selected based on the grounding electrode radius, grid size, and soil current diffusion range, and must satisfy the following conditions: R b,k The equivalent damping or bleed resistance used to characterize this local return path can be approximated using a vertical equivalent path: Alternatively, the half-space soil diffusion resistance can be approximated: , where ρ eLet R be the soil resistivity. When the model already includes an independent grounding resistance or soil discharge branch, to avoid repeatedly including the steady-state discharge resistance, R... b,k It can be treated as a high-frequency damping parameter, L b,k It is mainly used to compensate for transient inductive effects in the burial depth direction.

[0174] For the H-beam steel support bracket of GIS equipment, the original vertical grounding electrode branch is replaced with an equivalent topology of the H-beam steel bracket. This topology consists of two vertical steel leg branches on the left and right sides and a horizontal connecting branch. Each conductor segment is theoretically equivalent to an RL series branch and a ground capacitance branch according to the distributed parameters. The H-beam steel bracket can be divided into three main conductive paths: left vertical leg: denoted as Z h1 , corresponding to R h1 L h1 C h1 Right erector limb: denoted as Z h2 , corresponding to R h2 L h2 C h2 Lateral connector: denoted as Z h12 , corresponding to R h12 L h12 Consider C if necessary h12 As attached Figure 9 As shown.

[0175] Convert the H-beam cross-section to an equivalent circular conductor. Let the height of the H-beam cross-section be h, and the flange width be... The flange thickness is The web thickness is Then the cross-sectional area can be approximated as: ;

[0176] When using a circle with the same area radius, the equivalent radius is: For high-frequency surface currents, approximations such as constant perimeter or hydraulic radius can be used: , ,in, This refers to the exposed conductive perimeter of the H-beam.

[0177] Under low-frequency or DC approximation conditions, the resistance per unit length of H-beam is: Under high-frequency transient conditions such as VFTO / TGPR, the skin effect needs to be considered. The skin depth is: If the current is mainly concentrated on the outer surface, the effective conductive area can be approximated as: When modeling an equivalent cylinder, one can take... .

[0178] Therefore, the AC resistance per unit length is: , When the ATP-EMTP model does not use frequency-varying parameters, the target dominant frequency f can be used.e Take the equivalent value: .

[0179] For long, straight conductors close to the ground, their external inductance can be approximated using the image method or an equivalent return path. Let H... s Let represent the distance from the conductor to the equivalent return path, then:

[0180]

[0181] When considering soil recirculation depth, an equivalent recirculation depth D can be introduced. e : , In practical engineering, constant parameters can be used in the main frequency bands: .

[0182] If we consider an H-beam as an equivalent cylindrical conductor with radius req, its capacitance per unit length to ground or a zero-potential surface at infinity can be approximated as: .

[0183] When the support is in significant contact with the foundation or soil, the equivalent dielectric constant of the soil can be used: The electrical conductivity of soil leakage can be written as: When the focus is on high-frequency transient propagation and the impact of soil leakage is relatively weak, the following approach can be taken: .

[0184] In ATP-EMTP, lumped processing is performed first. The parameters for the left erector limb are:

[0185]

[0186] The corresponding impedance and admittance are:

[0187]

[0188] Let the voltage at the midpoint of the left vertical leg be V1, the voltage at the midpoint of the right vertical leg be V2, and the lateral connection admittance be Y. h12 Then the equation for the two nodes can be written as:

[0189]

[0190] in:

[0191]

[0192] When the left and right vertical limbs are perfectly symmetrical and the lateral connection has a negligible impact on the ground, then:

[0193]

[0194] The GIS casing is connected to the grounding grid node via a grounding down conductor. The impedance of the grounding down conductor is:

[0195]

[0196] Example 2

[0197] As attached Figure 11 As shown, this embodiment uses the original shell topology and the replaced broadband shell topology to perform ATP-EMTP transient simulation. The transient ground potential rise (TGPR) and VFTO waveforms are extracted in three time windows of 2ms, 10ms and 40ms. The accuracy and applicability of the broadband equivalent circuit are verified by comparison.

[0198] 2ms short time window verification: verification of corresponding short-time high-frequency oscillation characteristics

[0199] A comparison of transient ground potential rise waveforms under a 2ms time window is attached. Figure 12 As shown, both topologies exhibit transient impacts in the initial stage. The high-frequency oscillation envelope and attenuation pattern of the improved topology differ from those of the original topology, indicating that the wideband PI-type branch of the shell has a direct corrective effect on the short-time response of TGPR, and can more accurately reflect the propagation and discharge patterns of high-frequency transient signals on the shell.

[0200] A comparison of VFTO waveforms under a 2ms time window is attached. Figure 13 As shown, the overall negative VFTO levels of the two topologies are similar, indicating that the shell model replacement did not disrupt the basic propagation relationship of the GIS main loop; the initial high-frequency oscillation details differ, reflecting the fine-tuning effect of the shell broadband parameters on the initial local response of VFTO.

[0201] 10ms time window verification: verification of the corresponding transient decay trend

[0202] A comparison of VFTO waveforms under a 10ms time window is attached. Figure 14 As shown, the VFTO main trends of the original topology and the improved topology are basically consistent, both showing a process of increasing from negative to positive values. The only difference is in the local high-frequency fluctuations and peak positions, indicating that the improved topology and the original topology are consistent in the main propagation mechanism. The broadband shell branch mainly corrects the local oscillation characteristics and does not change the transient propagation law of the main loop of the system.

[0203] A comparison of transient ground potential rise waveforms within a 10ms time window is attached. Figure 15 As shown, TGPR is more sensitive to topology replacement. The modified topology curve shows a different upward trend and oscillation envelope in the middle and later sections compared to the original topology. This indicates that TGPR is mainly controlled by the shell-to-ground branch, grounding down conductor, and grounding grid topology. Preserving the complete grounding grid topology and using a broadband shell model can significantly change the attenuation and oscillation characteristics of TGPR.

[0204] 40ms long-term window verification: corresponding to long-term stability and overall transient characteristics verification

[0205] A comparison of VFTO waveforms under a 40ms time window is attached. Figure 16 As shown, the low-frequency main waveforms of VFTO under the two topologies are generally similar, but there are differences in local spikes and high-frequency oscillations. This indicates that the main propagation path of the bus still controls the overall trend of VFTO over a long time scale, while the wideband shell model mainly affects local high-frequency details. At the same time, the improved topology did not show obvious non-physical divergence, verifying the passivity and numerical stability of the wideband equivalent circuit.

[0206] A comparison of transient ground potential rise waveforms under a 40ms time window is attached. Figure 17 As shown, the amplitude and oscillation pattern of the TGPR under the improved topology differ significantly from those of the original topology, especially with stronger and denser local high-frequency fluctuations appearing in multiple time periods. The evolution of the oscillation envelope also differs significantly from that of the original topology. This phenomenon indicates that the broadband PI-type branch of the shell and the grounding grid topology have a significant impact on the long-term response of the TGPR, and the propagation, reflection, and local enhancement effects of transient potential in the grounding grid cannot be ignored. By retaining the complete grounding grid topology and replacing the broadband equivalent circuit of the shell, this invention can more accurately characterize the evolution of transient potential rise in the GIS shell over long time scales, avoiding the underestimation or misjudgment of the TGPR amplitude and oscillation characteristics by traditional lumped parameter models, and improving the engineering credibility of grounding safety assessment and electromagnetic compatibility analysis.

[0207] Example 3

[0208] A simulation circuit for transient potential rise in GIS, taking into account the grounding grid topology and the broadband transmission characteristics of the casing, is constructed using the method described in Example 1. The circuit comprises a module for multiple arcing excitation of the disconnecting switch, a module for busbar core wire propagation, a broadband equivalent module for the casing, and a grounding grid topology module. These modules form a complete transient conduction link through electrical connections or capacitive coupling, simulating the complete physical process of transient potential rise in the casing under GIS disconnecting switch operation.

[0209] The disconnector switch multiple arcing excitation module serves as the transient excitation source for the entire simulation circuit, generating VFTO and VFTC signals that conform to the physical process of disconnector switch operation. Its output is connected to the input of the bus core wire propagation module. This module includes a time-varying break dynamic capacitance unit, a segmented arc resistance unit, and an arcing and extinction criterion control unit.

[0210] Time-varying break dynamic capacitance unit: corresponding to the capacitive structure between the moving and stationary contacts of the disconnecting switch, the capacitance value changes dynamically with the operation time according to a quadratic polynomial law, and the fitting coefficient is determined by the measured capacitance data of the disconnecting switch of the corresponding voltage level, used to characterize the time-varying characteristics of the break capacitance during the contact opening and closing process.

[0211] Segmented Arc Resistance Unit: Used to simulate the resistance change characteristics of an electric arc from breakdown to extinction, it includes a pre-breakdown sub-unit, a steady-state combustion sub-unit, and an arc-extinguishing sub-unit. The pre-breakdown sub-unit corresponds to an arc resistance that decays exponentially, characterizing the formation process of the arc channel in the initial stage of breakdown; the steady-state combustion sub-unit corresponds to a constant low-value resistance, characterizing a stable arc state; and the arc-extinguishing sub-unit corresponds to an arc resistance that rises exponentially, characterizing the arc gap dielectric recovery process after the current crosses zero, based on the Mayer arc dynamics model.

[0212] Arc ignition and extinction criterion control unit: It incorporates a dynamic breakdown voltage arc ignition criterion and a current zero-crossing arc extinguishing criterion to control the arc extinguishing sequence. The dynamic breakdown voltage arc ignition criterion calculates the dynamic breakdown voltage threshold in real time based on the contact gap; when the voltage between the contacts exceeds the threshold, arc ignition is triggered. The current zero-crossing arc extinguishing criterion determines whether the arc is extinguished based on the zero-crossing characteristics and changes in current at adjacent moments, thereby achieving cyclical triggering of multiple arc ignition processes.

[0213] The bus core wire propagation module is the main transmission channel for ultra-fast transient signals, and includes the main propagation branch of the bus core wire and the coupling capacitor between the core wire and the inner layer of the housing:

[0214] The main propagation branch of the bus core wire is equivalent to a uniform lossless transmission line based on the Bergeron model. Its transmission characteristics are characterized by wave impedance, and it is used to realize the propagation of VFTO / VFTC signals along the axial direction of the GIS bus.

[0215] Core-Shell Inner Layer Coupling Capacitor: Connected between the bus core node and the inner layer node of the shell, it characterizes the capacitive coupling relationship between the core and the metal shell, coupling transient signals on the core to the GIS shell, providing an excitation source for the transient potential rise of the shell. The bus core propagation module forms an electrical connection with the shell broadband equivalent module through this coupling capacitor.

[0216] The shell broadband equivalent module is the core unit that characterizes the broadband transmission characteristics of the GIS shell. It replaces the lumped parameter topology of the inner and outer layers of the shell in the traditional simulation model. It is composed of several unit-length PI-type broadband equivalent circuits cascaded along the axial direction.

[0217] The unit-length PI-type broadband equivalent circuit is a two-port passive network, including an axial series branch, an input-to-ground parallel branch, and an output-to-ground parallel branch:

[0218] Axial series branch: It is composed of equivalent loss resistance and equivalent inductance connected in series, corresponding to the off-diagonal elements of the admittance parameter matrix, and is used to characterize the loss and inductance effect during the propagation of transient signals along the axial direction of the shell.

[0219] The parallel branch to ground at the inlet end is composed of a ground discharge resistor and a ground equivalent capacitance connected in parallel. It corresponds to the inlet-side component of the admittance parameter matrix after deducting the contribution of the series branch from the diagonal elements. It is used to characterize the energy storage and discharge characteristics to ground at the inlet end of the shell.

[0220] The parallel branch to ground at the outlet end is composed of a ground discharge resistor and a ground equivalent capacitance connected in parallel. It corresponds to the outlet-side component of the admittance parameter matrix after deducting the contribution of the series branch from the diagonal elements. It is used to characterize the energy storage and discharge characteristics to ground at the outlet end of the shell.

[0221] When multiple segments are cascaded, the output node of the previous unit length circuit is connected to the input node of the next unit length circuit, and the connection relationship between the core wire and the inner layer coupling capacitor of the housing remains unchanged throughout the process, thus fully restoring the broadband propagation characteristics of the long bus housing.

[0222] The grounding grid topology module is electrically connected to the shell nodes of the shell broadband equivalent module for transient current discharge to ground and transient potential distribution and conduction. It employs a complete distributed parameter topology, not simplified to a single centralized grounding resistor. This module includes grounding down conductor impedance branches, horizontal grounding electrode impedance branches, vertical grounding electrode distributed parameter equivalent circuits, and H-shaped steel support distributed parameter equivalent circuits.

[0223] Grounding down conductor impedance branch: connected between the shell node and the horizontal grounding grid node, its series impedance characteristics are characterized by a resistor-inductor series structure, and it is used to realize the conduction of the shell transient current to the grounding grid.

[0224] The equivalent circuit of the horizontal grounding electrode impedance branch and the vertical grounding electrode distributed parameters together constitute the main topology of the grounding grid, respectively characterizing the transient propagation in the horizontal direction and the ground discharge characteristics in the vertical direction, and fully reflecting the propagation, reflection and spatial distribution differences of transient potential in the grounding grid.

[0225] Equivalent circuit for distributed parameters of H-beam steel support: Connected between the shell node and the grounding grid node, it serves as the equivalent conduction path for the GIS equipment support structure. This equivalent circuit consists of a left vertical branch, a right vertical branch, and a transverse connecting branch. Each branch adopts a lumped parameter structure with a series resistor-inductor connection and a parallel capacitor to ground. This accurately reflects the current shunting, coupling, and grounding characteristics of transient currents in the H-beam steel support, further improving the accuracy of the grounding branch simulation.

[0226] During overall operation, the ultra-fast transient signal generated by the multiple arcing excitation module of the disconnecting switch propagates along the bus core wire, is coupled to the GIS shell via the core wire-shell coupling capacitor, and then the transient signal on the shell is characterized by its broadband transmission characteristics by the broadband equivalent module before flowing into the complete grounding grid topology through the grounding down conductor, ultimately resulting in a transient potential rise in the shell. This circuit simultaneously considers the physical realism of the excitation source, the accuracy of the broadband transmission of the shell, and the integrity of the grounding grid topology, and can achieve high-precision simulation of the transient potential rise of the GIS shell within a wide frequency range of 0-100MHz.

[0227] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0228] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for constructing a simulation circuit for transient potential rise in GIS considering grounding grid topology and broadband transmission characteristics of the casing, characterized in that, The construction is based on a pre-set GIS substation basic circuit model, including the following steps: S1. Construct a GIS disconnect switch multiple arcing model as the excitation source for VFTO and VFTC. The multiple arcing model is constructed by introducing the time-varying dynamic capacitance of the break, a segmented arc resistance model including the pre-breakdown stage, the steady-state combustion stage and the arc extinction stage, as well as the arcing criterion and the arc extinction criterion. S2. Establish a three-dimensional electromagnetic model of the GIS busbar shell of unit length and perform a 0-100MHz frequency sweep to extract the two-port S-parameters; convert the S-parameters into an admittance Y-parameter matrix and perform vector matching rational function fitting; based on the fitting results, construct a PI-type broadband equivalent circuit containing axial series branches and parallel branches to ground at both ends using the circuit synthesis method. S3. Retain the main propagation branch of the bus core wire and the coupling capacitance from the core wire to the inner layer of the shell in the basic circuit model of the GIS substation, and replace the original lumped parameter topology of the inner and outer layers of the GIS shell with the PI-type broadband equivalent circuit; retain the grounding grid distributed parameter branch composed of the grounding down conductor, horizontal grounding electrode and vertical grounding electrode, and electrically connect the PI-type broadband equivalent circuit with the retained grounding grid topology.

2. The method for constructing the GIS transient potential rise simulation circuit according to claim 1, characterized in that, In step S1: The dynamic capacitance of the break changes with time according to a quadratic polynomial law, and the fitting coefficient is obtained from the measured capacitance data of the disconnecting switch. In the segmented arc resistance model, the arc resistance decreases exponentially during the pre-breakdown stage, a fixed resistance is used during the steady-state combustion stage, and the resistance recovery characteristics are calculated based on the Mayer arc dynamic mathematical model during the arc extinction stage. The arc ignition criterion is a dynamic breakdown voltage criterion, which triggers arc ignition when the voltage between the contacts exceeds the dynamic breakdown voltage threshold that changes over time; the arc extinguishing criterion is a current zero-crossing criterion, which determines that the arc is extinguished when the product of the current value at the previous moment and the current value at the current moment is less than zero and the difference is less than a set constant.

3. The method for constructing the GIS transient potential rise simulation circuit according to claim 1, characterized in that, Step S2 specifically includes: Impedance ports with the same reference impedance are set at both ends of the three-dimensional electromagnetic model of the GIS busbar shell of unit length, and the two-port S-parameters are obtained by 0-100MHz frequency sweep. The S-parameters are converted into an admittance Y-parameter matrix according to the port reference impedance; The elements in the Y parameter matrix are fitted with rational functions using the vector matching method. The fitting results are then subjected to pole flipping or the fitting order is reselected to ensure stability, and passivity verification is performed. The axial series branch parameters of the PI-type equivalent circuit are determined based on the off-diagonal elements of the admittance Y-parameter matrix, and the parallel branch parameters to ground at both ends are determined based on the diagonal elements after deducting the contribution of the series branch.

4. The method for constructing the GIS transient potential rise simulation circuit according to claim 3, characterized in that: The axial series branch of the PI-type equivalent circuit is equivalent to a resistor and an inductor connected in series. The parallel branch to ground at the input end and the parallel branch to ground at the output end of the PI-type equivalent circuit are both equivalent to a structure in parallel between a loss resistor and a capacitor.

5. The method for constructing the GIS transient potential rise simulation circuit according to claim 1, characterized in that, In step S3, the grounding grid topology also includes equivalent branches of H-shaped steel supports, and the method for constructing the equivalent branches of H-shaped steel supports is as follows: The H-beam cross-section is equivalent to a cylindrical conductor. The AC resistance per unit length, external inductance, and capacitance to ground are calculated considering the skin effect. The H-shaped steel support is divided into a left vertical branch, a right vertical branch, and a transverse connecting branch. Each conductor segment is equivalent to a lumped parameter circuit consisting of a resistor-inductor series branch and a capacitance-to-ground branch.

6. The method for constructing the GIS transient potential rise simulation circuit according to claim 1, characterized in that, In step S3, when the actual GIS bus length is greater than the unit length module length, multiple PI-type broadband equivalent circuit modules are cascaded in sequence, and the connection relationship between the core wire and the inner layer coupling capacitor of the shell remains unchanged during cascading.

7. The method for constructing the GIS transient potential rise simulation circuit according to claim 1, characterized in that, The method also includes step S4: running the GIS transient simulation circuit, calculating the transient ground potential rise and VFTO waveforms, extracting waveform data under time windows of 2ms, 10ms and 40ms respectively, comparing them with the simulation results of the original shell topology, and verifying the accuracy and applicability of the replaced model.

8. A simulation circuit for transient potential rise in GIS considering grounding grid topology and broadband transmission characteristics of the casing, characterized in that, This includes a multiple arcing excitation module for disconnecting switches, a busbar core wire propagation module, a wideband equivalent module for the casing, and a grounding grid topology module. The disconnector switch multiple arcing excitation module includes a time-varying break dynamic capacitor unit, a segmented arc resistance unit, and an arcing and extinguishing criterion control unit, which is connected to the input terminal of the bus core wire propagation module; The bus core wire propagation module includes the main propagation branch of the bus core wire and the coupling capacitor from the core wire to the inner layer of the housing, and is electrically connected to the housing broadband equivalent module through the coupling capacitor; The housing broadband equivalent module is composed of several unit-length PI-type broadband equivalent circuits cascaded together. The unit-length PI-type broadband equivalent circuit includes an axial series resistor-inductor branch and a parallel resistor-capacitor branch connected to ground at the inlet and outlet ends. The grounding grid topology module includes a grounding down conductor impedance branch, a horizontal grounding electrode impedance branch, and a vertical grounding electrode distributed parameter equivalent circuit, which is electrically connected to the housing broadband equivalent module.

9. The GIS transient potential rise simulation circuit according to claim 8, characterized in that, The grounding grid topology module also includes an equivalent circuit for the distributed parameters of the H-shaped steel support. The equivalent circuit for the distributed parameters of the H-shaped steel support consists of a left vertical branch, a right vertical branch, and a horizontal connecting branch. Each branch is a lumped parameter structure with a series resistor and an inductor connected in series and a parallel capacitor to ground.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-7.

Images