A method for error compensation of a self-coupled induction voltage divider

Abstract

The application discloses a self-coupled induction voltage divider error compensation method and belongs to the field of voltage dividers. The method comprises the following steps: for a structure containing a shielding wire, a core wire-shielding wire double-layer node transfer model is established; the shielding wire node is taken as an internal node to eliminate, and a core wire side equivalent parameter model is obtained; a phased identification strategy is adopted to sequentially determine the core wire side model equivalent parameters; a compensation capacitor branch connected to an end reference tap is set at a first core wire side equivalent node of the target core wire side equivalent parameter model, a compensation calculation model is constructed; according to the current balance relationship of each core wire side equivalent node in the compensation calculation model, a node equation after the compensation capacitor is connected is obtained; and the compensation capacitor set is solved with the minimum square sum of the complex voltage deviation of all the core wire side equivalent nodes after compensation as a target, so that the compensation capacitor values and the access end are obtained. The compensation capacitor and the access position required by each compensation node are accurately calculated, and the error analysis result is converted into an implementable hardware compensation scheme.

Description

Technical Field

[0001] This invention belongs to the field of voltage dividers, and more specifically, relates to an error compensation method for an autotransformer inductive voltage divider. Background Technology

[0002] Inductive voltage dividers are precision electromagnetic proportional devices that achieve AC voltage proportional conversion based on the principle of electromagnetic induction. They are commonly used in AC value traceability, voltage proportionality references, and transformer calibration. With the development of power harmonic measurement, wideband AC metering, and voltage proportionality standards, voltage proportional devices need to maintain a stable and accurate proportional relationship over a wider frequency range. Compared with resistive and capacitive voltage dividers, the proportional relationship of inductive voltage dividers is mainly determined by the number of winding turns and the electromagnetic coupling structure. They offer advantages in proportional stability, low output impedance, and multi-tap output, making them more suitable for wideband, high-accuracy voltage proportionality reference applications.

[0003] In wide-band, high-accuracy proportional applications, inductive voltage dividers typically employ an autotransformer structure to create a multi-tap proportional output. They improve the consistency of parameters across the proportional winding segments through uniform segmentation, symmetrical winding, and multi-stage structures, reducing the impact of uneven leakage impedance on proportional error. Simultaneously, to reduce the influence of distributed capacitance current on the proportional winding tap potentials under wide-band conditions, shielding is usually implemented. These structural measures can reduce the wide-band error of inductive voltage dividers to some extent. However, as the operating frequency range expands and proportional accuracy requirements increase, errors still exist in the tap potentials due to parasitic parameters. Therefore, further error suppression or compensation measures are needed.

[0004] Existing error suppression or compensation measures mainly include the following categories: The first category involves capacitor compensation for the main proportional network or main voltage divider circuit. For example, in inductive voltage dividers, proportional errors caused by capacitive currents can be reduced by connecting external capacitors between taps, balancing capacitance between windings, or optimizing admittance load. In resistive voltage dividers, high-voltage probes, or capacitive voltage dividers, parallel compensation capacitors, compensation electrodes, or structural capacitors are often used to improve high-frequency response. The compensation elements in this type of method typically act on the main proportional winding, tap branches, main voltage divider arm, or output terminal. For autotransformer inductive voltage dividers with shielded wire structures, if compensation capacitors are directly connected between core wire taps or between core wire taps and end taps, although the ratio error at the target frequency may be reduced, the compensation branch will participate in the core wire main proportional network, and the shield potential distribution will not be improved synchronously. When the error source is related to the non-ideal distribution of shield potential, direct compensation of the core wire main network may result in improved ratio error but deteriorated phase error, making it difficult to simultaneously address both the amplitude proportional error and phase error of the tap voltage.

[0005] The second type involves using an auxiliary voltage divider, auxiliary winding, active injection network, or active feedback circuit to drive the shield potential. This method can improve the shield potential distribution, but it typically requires a dedicated auxiliary drive source, feedback network, or complex shield drive structure, making it more suitable for implementation during the instrument design or manufacturing phase. For autotransformer inductive voltage dividers that have already been manufactured or whose structure is already defined, this method increases structural complexity and debugging difficulty.

[0006] Besides the hardware compensation methods mentioned above, existing technologies often employ error calibration and end-user correction methods. These methods typically involve measuring the ratio and angle differences of the inductive voltage divider at different frequencies and tap positions, generating an error table or error curve, and then correcting the measurement results based on this table during subsequent use. While this method can reduce known systematic errors in the final reading, it is essentially a measurement result correction and does not change the potential distribution, capacitor leakage path, or shielding potential state within the inductive voltage divider. Therefore, this type of method cannot reduce errors caused by parasitic capacitance and non-ideal shielding potential distribution at the hardware level, nor can it provide information on the size of the compensation capacitor, the connection node, or the connection terminal. Summary of the Invention

[0007] In view of the shortcomings of related technologies, the purpose of this invention is to provide an error compensation method for an autotransformer inductive voltage divider, which aims to solve the problems that existing passive compensation technology cannot simultaneously take into account both ratio difference and angle difference, while active compensation technology has a complex structure and is difficult to implement.

[0008] To achieve the above objectives, the present invention provides an error compensation method for an autotransformer inductive voltage divider, comprising: S1. For an autotransformer with a shielded wire structure, establish a corresponding transmission model; wherein, the transmission model includes a core wire segmented series impedance branch, a shielded wire segmented series impedance branch, a core wire-shielded wire coupling admittance branch, a shielded wire external reference body admittance branch, and a shielded wire node distributed admittance branch. S2. Eliminate the shielded wire nodes as internal nodes to simplify the transmission model into a core-side equivalent parameter model. The core-side equivalent parameter model includes equivalent series branches, equivalent inter-node admittance branches, and equivalent node external reference body admittance branches. The core-side equivalent parameter model is represented by six equivalent parameters on the core side, including the equivalent leakage inductance on the core side. Equivalent AC resistance on the core wire side Equivalent mutual capacitance on the core wire side Equivalent external reference capacitance on the core wire side Equivalent mutual conductance on the core wire side Equivalent external reference body conductance of the core wire side ; S3. Obtain measured voltage proportional error data at multiple frequency points and multiple tap positions, and use a staged identification strategy to sequentially identify the parameters. , , , , and The equivalent parameter model of the target core wire side with determined parameters is obtained; S4, in the equivalent parameter model on the target core wire side... Equivalent node on each core wire side A compensation capacitor branch connected to the end reference tap is configured to construct a compensation calculation model; wherein, the equivalent node on the core wire side The compensation calculation node is used to calculate the complex potential and tap error before and after compensation; the compensation calculation model includes four types of branches: equivalent series branches between equivalent nodes on adjacent core wire sides, with admittance denoted as... The equivalent mutual admittance branches between equivalent nodes on the core wire side are denoted as follows: The equivalent admittance branch of the core wire side equivalent node to the external reference body, the admittance is denoted as The compensation capacitor branch between the equivalent node on the core side and the end reference tap, and the admittances of the equivalent node on the core side and the two ends are respectively denoted as... and ; S5. Based on the current balance relationship of the equivalent nodes on each core wire side in the compensation calculation model, the node equations after connecting the compensation capacitor are obtained:

[0009] in: It is a complex coefficient matrix containing compensation capacitors. The right-hand vector is formed by the end reference potential, the external reference volume potential, and the compensation branch. This refers to the collection of all compensation capacitors; During the calculation, all compensation capacitors are set to zero to obtain the equivalent node complex potential on the core side in the uncompensated state; the compensation capacitor to be determined is substituted into the above node equation to obtain the equivalent node complex potential on the core side after compensation. S6. Solve for the set of compensation capacitors with the objective of minimizing the sum of squared complex voltage deviations at all equivalent nodes on the compensated core side:

[0010] The compensation capacitance values ​​between each compensation calculation node and the end reference tap are obtained, and the connection point of the compensation capacitor is determined at the same time.

[0011] Optional, also includes: When actually connecting the compensation capacitor, based on the corresponding relationship between the equivalent node on the core wire side and the shield wire node in terms of structural position, the calculated first... The compensation capacitor corresponding to each compensation capacitor value is connected to the shielding wire node. superior.

[0012] Optionally, the establishment of a corresponding transmission model for an autotransformer employing a shielded wire structure includes: The proportional winding in the autotransformer is divided along the tap direction into... Section, obtained Individual core wire nodes The shielded wire is divided into Each shielded wire node ;in, and Given the boundary nodes, to This is the tap node for the core wire to be determined. and For the boundary node of the shielding line, to For internal nodes of the shielding wire; Based on the proportional winding segmented series leakage impedance between adjacent core wire nodes, the segmented series impedance of the shield wire between adjacent shield wire nodes, the core wire-shield coupling admittance between core wire nodes and shield wire nodes at the same location, the external reference volume admittance of the shield wire node, and the distributed admittance between shield wire nodes of different shield wire nodes, corresponding branches are constructed; the transmission model including each branch is expressed as: ;

[0013] ;

[0014]

[0015]

[0016]

[0017] in, The core wire is segmented and connected in series with impedance. AC resistance for segmented core wires. Leakage inductance is segmented for the core wire. The admittance is segmented and connected in series for the core wire. The shielded wire is segmented and connected in series with impedance. Segmented AC resistance for the shielded wire, For segmented inductance of shielded wire, For the segmented series admittance of the shielded wire, For chip-screen local coupling admittance. The coupling conductance between the core wire and the shield wire. This refers to the coupling capacitance between the core wire and the shield wire. For the external reference body admittance of the shielded wire, The equivalent conductivity of the shielded wire node to the external reference body. The equivalent capacitance of the shielded wire node to the external reference body; To shield the distribution of guides between nodes, To shield the equivalent mutual conductance between nodes. This is to shield the equivalent mutual capacitance between nodes.

[0018] Optionally, the step of eliminating elements by treating the shield wire nodes as internal nodes and simplifying the transmission model into an equivalent parameter model on the core wire side includes: The shielded wire node is used as an internal node for elimination. Based on the combined effect of the shielded wire potential distribution and the core wire-shielded wire coupling feedback on the complex potential of the core wire tap node, the parameters of the transmission model are expressed as core wire-side equivalent parameters, including core wire-side equivalent series parameters, core wire-side equivalent mutual admittance parameters, and core wire-side equivalent external reference body admittance parameters. Construct the equivalent parameter model of the core wire side based on the relevant parameters of the core wire side.

[0019] Optionally, a phased identification strategy is adopted to sequentially identify parameters. , , , , and ,include: Phase 1: Under the aforementioned capacitance constraint, the equivalent leakage inductance on the core wire side is used. Equivalent external reference capacitance on the core wire side For the parameters to be identified, the parameters are... and Set the value to 0, construct the error objective function between the calculated ratio difference of the equivalent parameter model on the core wire side and the measured ratio difference data at multiple frequency points and multiple tap positions, and solve for the error objective function with the goal of minimizing the error objective function. and And determined by the capacitance constraint. ; Phase Two: Utilize the gains from Phase One , and Substituting the equivalent parameter model of the core wire side, calculate the calculated ratio difference of the equivalent parameter model of the core wire side in the first stage; subtract the measured ratio difference data from the calculated ratio difference of the first stage to obtain the residual ratio difference; define the loss class combination parameters: and The residual ratio difference is then fitted with the values ​​of the loss-type combination parameters to obtain the solution. and ; Phase 3: Fixed gains from Phase 1 , , The second phase of acquisition , The equivalent AC resistance of the core wire side As parameters to be identified, the measured voltage proportional error data at multiple frequency points and multiple tap positions are substituted into the equivalent parameter model on the core wire side for determination. The equivalent mutual conductance on the core side is obtained from the relationship in the second stage. Equivalent external reference body conductance of the core wire side .

[0020] Optionally, the complex coefficient matrix containing the compensation capacitor The matrix elements are constructed as follows: For the line, number The column, with diagonal elements as follows: The off-diagonal elements are: ,in, , Indicates when Time to take Otherwise take ; Right-hand vector The first element in the matrix The elements are:

[0021] in, The external reference potential is usually taken as the ground potential; Indicates when Time to take Otherwise take ; Indicates when Time to take Otherwise take When the external reference body is at ground potential: .

[0022] Optionally, after the compensation capacitor is connected, the first The complex voltage deviation of each compensation calculation node is:

[0023] in, To compensate for the capacitor after connection Complex potential of equivalent nodes on each core wire side; For the first The ideal following potential of each node The real part of the complex voltage deviation corresponds to the proportional error of the tap voltage amplitude, and the imaginary part corresponds to the phase error of the tap voltage.

[0024] Optionally, when the structure and error distribution of the autotransformer satisfy a symmetrical relationship, the number of compensation calculation nodes to be determined can be reduced by using half-side calculation and mirrored access: Let the number of internal compensation nodes be: The number of independently calculated half-side compensation nodes is: Let the vector of the independent half-side compensation capacitor be: ,in, , indicating the first The compensation capacitance value required for each half-side compensation node; The variable to be determined is from the complete set of compensation capacitors. Simplified to half-side independent compensation capacitor vector By solving The theoretical value of half of the compensation capacitor is obtained, and the compensation capacitor values ​​and access terminals of all compensation nodes are obtained through mirror relationship.

[0025] Compared with the prior art, the above-described technical solutions conceived in this invention can achieve the following beneficial effects: 1. This invention provides an error compensation method for an autotransformer inductive voltage divider. First, a transmission model of the autotransformer inductive voltage divider is established. Then, by eliminating elements at the shielding node, the combined effect of the complex shielding structure is converted into equivalent parameters on the core wire side. A staged parameter identification method is used to solve for six equivalent parameters sequentially. Combined with port frequency sweep total capacitance constraints and multi-stage error fitting, stable identification of model parameters is achieved. This enables the construction of a complete transmission model reflecting the shielded structure under broadband conditions. Then, with the equivalent parameters known, a compensation capacitor is connected between the shielding compensation node and the end reference tap. A compensation calculation model including the compensation capacitor branch is constructed, calculating the required compensation capacitor size for each compensation node and determining the connection node and end of the compensation capacitor. This transforms the error analysis results into an implementable hardware compensation scheme. The compensation capacitor obtained by this scheme is used for compensation, applying the compensation effect to the shielding potential boundary, avoiding direct participation of the compensation branch in the core wire main proportional network, and reducing disturbances to the original turns ratio and tap current distribution. This invention addresses the challenges of existing passive compensation techniques, which struggle to simultaneously address both ratio and phase errors, and the complexity and implementation difficulties of active compensation techniques. It eliminates the need for complex auxiliary voltage dividers, active drivers, or feedback control networks, enabling compensation using passive capacitors. This method is suitable for pre-fabricated or structurally defined autotransformer voltage dividers and achieves a balance between amplitude and phase errors in the tap voltage.

[0026] 2. This invention provides an error compensation method for an autotransformer inductive voltage divider. Based on the circuit structure of the autotransformer inductive voltage divider with a shielded wire structure, when actually connecting the compensation capacitor, the calculated error is... The compensation capacitor corresponding to each compensation capacitor value is connected to the shielding wire node. This adjusts the shielding potential distribution, reduces the impact of non-ideal shielding potential distribution and parasitic capacitive coupling on the core wire tap recombination potential, helps reduce tap voltage ratio error, and suppresses the angle difference degradation problem that may occur when directly compensating the core wire main network. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating an error compensation method for an autotransformer inductive voltage divider provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the self-coupling inductive voltage divider transmission model provided in the embodiment of the present invention; Figure 3 This is a schematic diagram of the equivalent model of the core wire side of the self-coupling inductive voltage divider provided in the embodiment of the present invention; Figure 4 This is a schematic diagram of the compensation capacitor connection scheme in an error compensation method for an autotransformer voltage divider provided in an embodiment of the present invention; Figure 5This is a diagram illustrating the error compensation effect of an autotransformer inductive voltage divider at 3000Hz, provided in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0029] The following description, in conjunction with a preferred embodiment, illustrates the content involved in the above embodiments.

[0030] For autotransformer voltage dividers with shielding structures, a hardware error compensation method is needed that is suitable for wide-band, high-accuracy applications. This method should reduce tap voltage errors caused by parasitic capacitance and non-ideal shielding potential distribution without directly disturbing the main proportional network of the core wire or adding complex active drive structures, while also taking into account ratio and phase difference performance.

[0031] To address the problems existing in the prior art, this invention provides an error compensation method for an autotransformer inductive voltage divider, such as... Figure 1 As shown, it includes: S1. For an autotransformer with a shielded wire structure, establish a corresponding transmission model; wherein, the transmission model includes a core wire segmented series impedance branch, a shielded wire segmented series impedance branch, a core wire-shielded wire coupling admittance branch, a shielded wire external reference body admittance branch, and a shielded wire node distributed admittance branch. S2. Eliminate the shielded wire nodes as internal nodes to simplify the transmission model into a core-side equivalent parameter model. The core-side equivalent parameter model includes equivalent series branches, equivalent inter-node admittance branches, and equivalent node external reference body admittance branches. The core-side equivalent parameter model is represented by six equivalent parameters on the core side, including the equivalent leakage inductance on the core side. Equivalent AC resistance on the core wire side Equivalent mutual capacitance on the core wire side Equivalent external reference capacitance on the core wire side Equivalent mutual conductance on the core wire side Equivalent external reference body conductance of the core wire side ; S3. Obtain measured voltage proportional error data at multiple frequency points and multiple tap positions, and use a staged identification strategy to sequentially identify the parameters. , , , , and The equivalent parameter model of the target core wire side with determined parameters is obtained; S4, in the equivalent parameter model on the target core wire side... Equivalent node on each core wire side A compensation capacitor branch connected to the end reference tap is configured to construct a compensation calculation model; wherein, the equivalent node on the core wire side The compensation calculation node is used to calculate the complex potential and tap error before and after compensation; the compensation calculation model includes four types of branches: equivalent series branches between equivalent nodes on adjacent core wire sides, with admittance denoted as... The equivalent mutual admittance branches between equivalent nodes on the core wire side are denoted as follows: The equivalent admittance branch of the core wire side equivalent node to the external reference body, the admittance is denoted as The compensation capacitor branch between the equivalent node on the core side and the end reference tap, and the admittances of the equivalent node on the core side and the two ends are respectively denoted as... and ; S5. Based on the current balance relationship of the equivalent nodes on each core wire side in the compensation calculation model, the node equations after connecting the compensation capacitor are obtained:

[0032] in: It is a complex coefficient matrix containing compensation capacitors. The right-hand vector is formed by the end reference potential, the external reference volume potential, and the compensation branch. This refers to the collection of all compensation capacitors; During the simulation, all compensation capacitors are set to zero to obtain the equivalent node complex potential on the core side in the uncompensated state. The compensation capacitors are then substituted into the above node equation to obtain the equivalent node complex potential on the core side after compensation. S6. Solve for the set of compensation capacitors with the objective of minimizing the sum of squared complex voltage deviations at all equivalent nodes on the compensated core side:

[0033] The compensation capacitance values ​​between each compensation calculation node and the end reference tap are obtained, and the connection point of the compensation capacitor is determined at the same time.

[0034] To address the tap voltage error issues caused by parasitic capacitance, non-ideal shielding potential distribution, and core-shield coupling feedback in wide-band, high-accuracy proportional transfer of autotransformers, this application provides an error compensation method for autotransformers. This method is based on the core-shield node structure and the core-side equivalent parameter model. A compensation capacitor branch is introduced into the core-side equivalent parameter model, and the required compensation capacitor size and its connection point for each compensation node are calculated. For autotransformers with a core-shield structure, the compensation capacitor is preferably connected between the shielding node corresponding to the core-side equivalent node and the end reference tap to adjust the shielding potential distribution and avoid the compensation branch directly participating in the core-side main proportional network.

[0035] like Figure 2 As shown, the autotransformer inductive voltage divider is divided along the tap direction into... Each segment corresponds to Each core wire node and Each shielded wire node. The core wire node is denoted as: The shielded wire node is denoted as: . and For the end boundary node of the core wire, to This is an internal tap node for the core wire; and For the end boundary node of the shielded wire, to For internal nodes of the shielding wire.

[0036] The complex potential vector of the core node is defined as:

[0037] The complex potential vector of a shielded wire node is defined as:

[0038] in, Indicates the core wire node Complex potential relative to the reference point, Indicates shielded wire node Complex potential relative to the reference point, .

[0039] In actual working connection, both ends of the core wire and the shield wire are connected to the same end reference tap, thus satisfying: , .in, For the first End reference tap reset potential For the first The reference tap at end number is reset to its original potential. The input voltage is: .

[0040] The target communication frequency is The corresponding angular frequency is: The input voltage between the two reference taps is:

[0041] in, and This is the end reference potential under actual operating conditions, which can be grounded at one end, suspended at both ends, symmetrically suspended, or other end potential forms. The angular frequency corresponding to the target AC operating frequency is:

[0042] The branch connection relationships of the transmission model are as follows: First, segmented series impedance branches of the core wires. Adjacent core wire nodes. and The segmented series impedance branch connecting the core wires has the following branch impedance: .in, For the first Equivalent AC resistance of the core wire segment For the first Equivalent leakage inductance of the segment core wire. This branch is used to represent the segment leakage impedance of the proportional winding and its resulting series voltage drop effect.

[0043] Second, the shielded wire is segmented into series impedance branches. Adjacent shielded wire nodes... and The shielded wires connecting the segments form a series impedance branch, and the branch impedance is: .in, For the first Equivalent AC resistance of the shielded wire segment For the first The equivalent leakage inductance of the shielded wire. This branch is used to represent the impedance of the shielded wire itself along the line, so that the shielded wire is no longer an ideal equipotential body under broadband conditions.

[0044] Third, the core-shield coupling admittance branch. Core node. Shielding wire nodes at corresponding positions The admittance of the branch connecting the core wire and shield wire is: .in, For the first The core wire-shield coupling capacitor at each location. This corresponds to the loss conductance. This branch represents the capacitive and lossy coupling between the core wire and the shield wire, and is the path through which the shield wire potential shift is fed back to the core wire tap potential.

[0045] Fourth, the shielded wire external reference body admittance branch. Shielded wire node. The admittance of the external reference body branch of the shielded wire connecting to ground, the casing, or other external reference bodies is: .in, The equivalent capacitance of the shielded wire node to the external reference body. This corresponds to the loss conductivity. This branch is used to represent the capacitive leakage and loss effects between the shielded wire and the external environment.

[0046] Fifth, the admittance branches are distributed between shielded nodes. Different shielded wire nodes... and The distributed admittance branches between the shielded nodes are given by: .in, Distributed capacitance between shielded wire nodes. This corresponds to the loss conductance. This branch is used to represent the distributed coupling effect between different nodes inside the shielding structure.

[0047] Since the compensation objective of this invention is to reduce the voltage error at the core wire tap nodes, and the effect of the shielding wire-related branches on the core wire tap potential is jointly reflected through the shielding wire potential distribution and the core wire-shielding wire coupling feedback, the shielding wire nodes are treated as internal nodes for elimination in the compensation calculation, simplifying the transmission model to a core wire-side equivalent parameter model. The core wire-side equivalent parameter model translates the combined effect of the shielding wire network on the complex potential of the core wire taps to the core wire side. Because the proportional winding of an autotransformer inductive voltage divider typically employs a symmetrical stranded and uniformly segmented structure, similar branches can be represented using unified equivalent parameters. For example... Figure 3 As shown, the equivalent parameter model on the core wire side includes the following three types of equivalent branches: The equivalent series branch on the core wire side is represented as follows:

[0048]

[0049] The equivalent mutual admittance branch on the core wire side is represented as: ; The equivalent external reference volume admittance branch on the core wire side is represented as follows: ; in, and The equivalent series effect on the core wire side is jointly characterized; and The equivalent mutual admittance between core wire tap nodes is jointly characterized; and The effect of the core wire tap node on the equivalent admittance of the external reference body is jointly characterized.

[0050] The correspondence between the above equivalent parameters and the transmission model is as follows:

[0051]

[0052]

[0053]

[0054]

[0055]

[0056] in, , Represents the complex admittance matrix on the shield side. The inverse matrix,

[0057] Therefore, the equivalent parameters on the core wire side to be determined are:

[0058] For any set of candidate parameters All of these can be substituted into the model to calculate the error values ​​that correspond one-to-one with the measured frequency points and tap positions. Therefore, the essence of parameter identification is to find a set of equivalent parameters that make the model calculation error and the multi-frequency measured tap error as consistent as possible at the corresponding frequency points and tap positions.

[0059] As can be seen from the equivalent circuit relationship, the branch currents generated by the mutual admittance between tap nodes and the admittance of tap nodes to ground or to the casing need to pass through the equivalent series leakage impedance to form the tap potential shift. Therefore, the main error term is determined by the product of the equivalent series leakage impedance and the equivalent distributed admittance. In this scheme, instead of fitting all six parameters as free variables at once, this method can reduce the coupling effect between leakage inductance, capacitance, resistance, and conductance, and reduce the problems of strong parameter correlation, non-unique identification results, and unclear physical meaning.

[0060] For equivalent mutual admittance branches between tap nodes:

[0061] For tapped nodes to ground or equivalent admittance branches to the casing:

[0062] The above expression shows that the main trend of error change is... , The decision is made. Both types of capacitors will affect the main bandwidth error curve. If directly... and Simultaneously, when fitted together with other parameters as free variables, different capacitor combinations can easily produce similar error curves, leading to mutual compensation of capacitor parameters. To reduce this correlation, this embodiment first uses port sweep impedance to obtain the total equivalent capacitance. And establish capacitance constraints based on the energy equivalence relationship under ideal voltage divider potential.

[0063] Optionally, the total equivalent capacitance of the port Obtained through the following methods: Short-circuit the input and output ends of the core wire and the shield wire respectively, and use an impedance analyzer to perform frequency sweep impedance measurement on the port of the autotransformer to obtain the port complex impedance. Then, the port complex admittance is obtained from the port complex impedance. Within a selected frequency range, the port complex admittance is expressed as a parallel combination of the port equivalent loss conductance and the port total equivalent capacitance, and the port total equivalent capacitance is determined by the slope of the change of the imaginary part of the port complex admittance relative to the angular frequency. .

[0064] Specifically, with the core wire and shield wire maintained in actual working connection, a frequency sweep impedance measurement is performed on the voltage divider port to obtain the port complex impedance:

[0065] in, , For the first Each sweep frequency point Where is the number of frequency points. The port complex admittance is:

[0066] Within a selected frequency range, the port complex admittance can be approximated as:

[0067] in, The total equivalent capacitance is obtained by fitting the port sweep frequency impedance data. Let be the equivalent loss conductance at the port. In actual calculations, it can be obtained from the slope of the imaginary part of the port admittance with respect to the angular frequency. :

[0068] The above-mentioned total equivalent capacitance is the overall capacitance effect seen from the core wire side port under actual connection conditions. It already includes the equivalent mutual capacitance between core wire tap nodes, the equivalent capacitance of core wire tap nodes to ground or to the shell, and the capacitance effect of the shielding structure after being referred to the core wire side.

[0069] Let the input voltage be Under ideal partial pressure conditions, the first The potential of each core wire tap node is:

[0070] According to the equivalent relationship of electric field energy storage, the total equivalent capacitance satisfies:

[0071] in, For the first The factor for the equivalent capacitance branch of each node to ground or to the casing is typically used. Internal tap nodes are usually calculated using... Whether end nodes are fully included can be adjusted based on port frequency sweep wiring, boundary node processing methods, and the actual structure. This invention takes... Substitute We can obtain:

[0072] in, The mutual capacitance energy weighting coefficient, This is the weighting coefficient for energy relative to ground or to the casing capacitor.

[0073] This constraint makes and They no longer participate in the fitting as completely free parameters, thereby reducing the mutual compensation between capacitance parameters.

[0074] Optionally, the phased identification strategy includes the following steps: Phase 1: Under the aforementioned capacitance constraint, the equivalent leakage inductance on the core wire side is used. Equivalent external reference capacitance on the core wire side For the parameters to be identified, the parameters are... and Set the value to 0, construct the error objective function between the calculated ratio difference of the equivalent parameter model on the core wire side and the measured ratio difference data at multiple frequency points and multiple tap positions, and solve for the error objective function with the goal of minimizing the error objective function. and And determined by the capacitance constraint. Among them, the equivalent mutual capacitance and equivalent external reference volume capacitance Both factors affect the total equivalent capacitance and tap error at the port. If both are used as free parameters for fitting, parameter compensation can easily occur. This invention establishes a total capacitance constraint based on the energy equivalence under ideal voltage divider potentials, and... and Constrained within the same total equivalent capacitance relationship, thereby improving the stability of capacitance parameter identification.

[0075] Specifically, after the total equivalent capacitance constraint is established, the first stage is used to determine the leakage inductance and capacitance parameters that dominate the broadband error variation in the reduced-order model on the core wire side.

[0076] Dominant capacitive error refers to the error caused by equivalent mutual capacitance under wide bandwidth conditions. Equivalent capacitance to ground or to case With equivalent leakage inductance The main tap ratio difference curve formed by the combined effect. Loss parameters. , The influence of AC resistance coupled with it is not identified as a free parameter in this stage, but is left to be processed in the subsequent residual error stage.

[0077] Let the set of taps participating in the fitting be: , No. The frequency points are , No. The measured difference of each tap is The model calculation ratio is The objective function for the first stage is:

[0078] in:

[0079] For the first The weight of each frequency point This is the frequency-weighted exponent. When... When the frequency points have the same weight; when At that time, the weight of high-frequency points increases.

[0080] Phase Two: Utilize the gains from Phase One , and Substituting the equivalent parameter model of the core wire side, calculate the calculated ratio difference of the equivalent parameter model of the core wire side in the first stage; subtract the measured ratio difference data from the calculated ratio difference of the first stage to obtain the residual ratio difference; define the loss class combination parameters: and The residual ratio difference is then fitted with the values ​​of the loss-type combination parameters to obtain the solution. and .

[0081] Specifically, the parameters obtained in the first stage are substituted into the model, and the conductivity loss term is not introduced for the time being, to obtain the first-stage model ratio difference:

[0082] Subtracting the measured ratio difference from the first-stage model ratio difference yields the unexplained ratio difference residue from the first stage:

[0083] This residual mainly corresponds to the influence of loss-related parameters on the ratio difference. As can be seen from the aforementioned equivalent circuit relationship, the conductivity loss term mainly appears in combination form in the ratio difference residual; therefore, it is not directly identified simultaneously in this stage. , , Instead, we first define two loss class combination parameters: ,

[0084] Therefore, the second stage identifies the loss correction amount by making it consistent with the residual ratio difference from the first stage. , Construct the objective function:

[0085] in, For the first The residual scale at each frequency point is used to normalize the residual amplitude at different frequencies, and can be taken as follows: , To prevent excessively small residuals from causing overly strong normalization, a lower limit value is set.

[0086] Phase 3: Fixed gains from Phase 1 , , The second phase of acquisition , The equivalent AC resistance of the core wire side As parameters to be identified, the measured voltage proportional error data at multiple frequency points and multiple tap positions are substituted into the equivalent parameter model on the core wire side for determination. The equivalent mutual conductance on the core side is obtained from the relationship in the second stage. Equivalent external reference body conductance of the core wire side .

[0087] Specifically, first, fix the parameters obtained in the first two stages, so as to As the parameter to be identified. For any candidate The corresponding conductivity parameter is determined by the following formula: , .

[0088] Substitute the above parameters into the model to calculate the first... The frequency point, the first Model complex error at each tap position:

[0089] The fitting objective function is:

[0090] in, The number of frequency points. This represents the set of taps participating in the fitting. The objective function incorporates information on both ratio difference and angle difference, and therefore can be used to determine... And determine based on the results of the second phase. .

[0091] Through the above process, the six equivalent parameters required for the error calculation model are finally obtained. , , , , , .

[0092] The phased parameter identification method used in this embodiment can reduce the coupling effect between leakage inductance, capacitance, resistance and conductance, and reduce the problems of strong parameter correlation, non-unique identification results and unclear physical meaning.

[0093] In constructing the transmission model, the impedance along the shielding wire, the admittance of the shielding wire to the external reference body, the distributed admittance between shielding nodes, and the core-shield coupling feedback are incorporated into the error calculation. This allows for a more complete reflection of the residual tap error formation mechanism of the shielded wire structure under broadband conditions. During simplification, the complex shielding wire network is converted into an equivalent parameter model on the core side through shielding node elimination. Furthermore, the complex original branch parameters of the shielded wire structure are merged into six equivalent parameters on the core side, reducing the difficulty of model application and making the model parameters more suitable for identification through measured data. A total capacitance constraint is established based on the energy equivalence relationship under ideal voltage divider potentials, and the equivalent mutual capacitance is included. and equivalent external reference volume capacitance By constraining the parameters within the same total equivalent capacitance relationship, the stability of capacitance parameter identification is improved. A staged parameter identification method is adopted, solving for six equivalent parameters sequentially. First, leakage inductance-capacitance parameters, which dominate broadband errors, are identified under the total capacitance constraint. Then, loss-related combined parameters are identified using the ratio difference residual. Finally, AC resistance and conductance parameters are obtained by decoupling through complex error data. This approach enables the construction of a complete transmission model that reflects the shielded structure under broadband conditions, simplifies model expression, improves the stability and physical interpretability of parameter acquisition, and provides a foundation for subsequent capacitance compensation calculations.

[0094] Based on the known equivalent parameter model and parameters of the core wire side, the correspondence between the compensation calculation nodes and the actual compensation nodes is determined. For example... Figure 4 As shown, in the equivalent parameter model on the target core wire side, the first... Equivalent node on each core wire side Configure a compensation capacitor branch connected to the end reference tap to construct a compensation calculation model. The calculation nodes in the compensation calculation model use the core-side equivalent nodes, still denoted as: , No. Equivalent node on each core wire side The complex potential is still denoted as .

[0095] No. The ideal following potential of the equivalent node on each core wire side is determined according to the ideal proportional relationship between the reference potentials at both ends:

[0096] in: .

[0097] Core wire side equivalent node These are nodes in the compensation calculation model, used to calculate the complex potential and tap error before and after compensation; the actual compensation node is the node where the compensation capacitor is actually connected in the hardware.

[0098] For an autotransformer with a core-shielded wire structure, the equivalent node on the core side is... With shielded wire node Corresponding in structural location. The first [unit / line] calculated in the equivalent model on the core wire side. The compensation amount is preferably selected based on the actual connection to the shielded wire node. Up. At this moment, It is the equivalent node on the core wire side. The ideal proportional potential is also the shielding wire node. The target shielding potential that should be followed.

[0099] In the compensation calculation, the first The node and the first The compensation capacitance between the reference taps at the ends is denoted as: Its admittance is: . No. The node and the first The compensation capacitance between the reference taps at the ends is denoted as: Its admittance is: .

[0100] After adding the compensation capacitor branch to the equivalent model on the core wire side, the compensation calculation model includes four types of branches. In the Dog-Akin node equations, a sign convention is adopted where nodal self-admittance is positive and mutual admittance is negative; The complex coefficient matrix containing the compensation capacitor The matrix elements are constructed as follows: For the line, number The column, with diagonal elements as follows: The off-diagonal elements are: ,in, , Indicates when Time to take Otherwise take ; Right-hand vector The first element in the matrix The elements are:

[0101] in, The external reference potential is usually taken as the ground potential; Indicates when Time to take Otherwise take ; Indicates when Time to take Otherwise take When the external reference body is at ground potential: .

[0102] Therefore, the equivalent node complex potential on the core wire side after compensation is: .

[0103] During the simulation calculation, before compensation, all compensation capacitors are set to zero: The equivalent node complex potential on the core side in the uncompensated state can be obtained; after compensation, the capacitance to be compensated is substituted into the above node equation to obtain the equivalent node complex potential on the core side after compensation.

[0104] The process of solving for the compensation capacitor includes: No. The ideal following potential for each node is:

[0105] After the compensation capacitor is connected, the first The complex voltage deviation at each node is defined as: ;in, To compensate for the capacitor after connection The complex potential of the equivalent node on each core wire side. The real part of the complex voltage deviation corresponds to the proportional error of the tap voltage amplitude, and the imaginary part corresponds to the phase error of the tap voltage.

[0106] To avoid the problem of improved ratio difference but deteriorated angle difference that may occur when directly compensating the main proportional network of the core wire, this invention aims to minimize the sum of squares of the complex voltage deviations of all internal nodes and solves for the set of compensation capacitors: This yields the theoretical values ​​of the compensation capacitance between each compensation node and the end reference tap, and simultaneously determines the size of the compensation capacitance and the connection point, i.e., determines: , .

[0107] After obtaining the capacitance value and connection port of each compensation capacitor, when actually connecting the compensation capacitor, based on the corresponding relationship between the equivalent node on the core wire side and the shield wire node in the structural position, the calculated value of the first compensation capacitor is... The compensation capacitor corresponding to each compensation capacitor value is connected to the shielding wire node. superior.

[0108] In one specific embodiment, a ten-segment autotransformer is used as an example to illustrate the determination and connection process of the compensation capacitor of the present invention. This embodiment is only used to illustrate the compensation calculation and hardware connection method of the present invention and does not constitute a limitation on the scope of protection of the present invention. For autotransformers with other numbers of segments, other end reference potential forms, or other target compensation frequencies, the number of nodes, end reference potentials, and compensation calculation parameters can be adjusted in the same way.

[0109] In this embodiment, the autotransformer inductive voltage divider is divided into ten segments along the tap direction, namely:

[0110] The corresponding core wire nodes are c0, c1, ..., c 10 The shielding wire nodes are s0, s1, ..., s 10 Among them, c0 and c 10 For the reference nodes at the ends of the core wires, s0 and s 10 C1 to C9 are the reference nodes at the end of the shielded wire; C1 to C9 are the equivalent calculation nodes on the core wire side; and S1 to S9 are the corresponding actual shielded wire compensation nodes.

[0111] In actual operation, both ends of the core wire and the shield wire are connected to the same end reference taps, namely end reference tap 0 and end reference tap 10. The complex potentials of the two end reference taps are denoted as V0 and V, respectively. 10 The input voltage is:

[0112] This embodiment does not limit V0 and V. 10 The specific absolute potential relative to the external reference body is only required to be the end reference potential under actual working conditions.

[0113] The target compensation frequency is set as follows:

[0114] The corresponding angular frequency is:

[0115] At the target compensation frequency, compensation calculations are performed using pre-obtained equivalent parameters on the core wire side.

[0116] In this embodiment, the equivalent parameters used for compensation calculation are:

[0117]

[0118]

[0119]

[0120] in, The equivalent leakage inductance on the core wire side, This is the equivalent AC resistance on the core wire side. This is the equivalent mutual capacitance on the core wire side. This is the equivalent external reference capacitance on the core wire side. The above parameters are known input quantities for compensation calculations and are not required to be equivalent to a specific segment of a real conductor or a specific actual capacitor element.

[0121] For the i-th core wire side equivalent calculation node c i Its ideal follower potential is:

[0122] in: .

[0123] In the equivalent model on the core wire side, two optional compensation capacitor branches are set for the i-th computing node: a compensation capacitor connected to the reference tap at end number 0. And the compensation capacitor connected to the 10th end reference tap. Their corresponding admittances are as follows:

[0124]

[0125] Substituting the above capacitor compensation branch into the equivalent node equation on the core wire side, the complex potential of the node after compensation is calculated, and the compensation capacitor is solved with the objective of minimizing the sum of squares of the complex voltage deviations of each internal node. The compensation capacitor connection scheme obtained after the compensation calculation is shown in the table below, and the compensation effect is as follows: Figure 5 As shown.

[0126] Table 1. Compensation Capacitor Connection Scheme for Ten-Segment Autotransformer Inductive Voltage Divider

[0127] As shown in Table 1, the compensation capacitors in this embodiment are symmetrically distributed. At shielded wire nodes s1 to s4 near the 0th end reference tap, the compensation capacitors are connected between the corresponding shielded wire nodes and the 0th end reference tap; at shielded wire nodes s6 to s9 near the 10th end reference tap, the compensation capacitors are connected between the corresponding shielded wire nodes and the 10th end reference tap. The center node s5 is a symmetrical balance position and no compensation capacitor is connected therein in this embodiment.

[0128] Furthermore, in an alternative embodiment, when the structure and error distribution of the autotransformer satisfy a symmetrical relationship, the number of compensation calculation nodes to be determined is reduced by using half-side calculation and mirrored access: Let the number of internal compensation nodes be: The number of independently calculated half-side compensation nodes is: Let the vector of the independent half-side compensation capacitor be: ,in, , indicating the first The compensation capacitance value required for each half-side compensation node; The variable to be determined is from the complete set of compensation capacitors. Simplified to half-side independent compensation capacitor vector By solving We obtain the theoretical value of half of the compensation capacitor, and then obtain the compensation capacitor values ​​and connection points of all compensation nodes through mirror relationships.

[0129] Furthermore, the embodiments of the present invention can adapt to different end reference potential conditions, including one end grounded, both ends suspended, symmetrically suspended, or other end potential forms; at the same time, the compensation effect can be calculated by substituting the nominal capacitance, which makes it easy to modify the actual compensation scheme according to the component tolerance, withstand voltage level and installation conditions.

[0130] This invention provides an error compensation method for an autotransformer inductive voltage divider. First, a transmission model of the autotransformer inductive voltage divider is established. Then, by eliminating elements at the shielding node, the combined effect of the complex shielding structure is converted into equivalent parameters on the core wire side. A staged parameter identification method is used to solve for six equivalent parameters sequentially. Combined with port frequency sweep total capacitance constraints and multi-stage error fitting, stable identification of model parameters is achieved. This constructs a complete transmission model reflecting the shielded structure under broadband conditions. Then, with the equivalent parameters known, a compensation capacitor is connected between the shielding compensation node and the end reference tap, constructing a compensation calculation model including the compensation capacitor branch. The required compensation capacitor size for each compensation node is calculated, and the connection node and connection end of the compensation capacitor are determined, thus transforming the error analysis results into an implementable hardware compensation scheme. The compensation capacitor obtained by this scheme is used for compensation, applying the compensation effect to the shielding potential boundary, avoiding direct participation of the compensation branch in the core wire main proportional network, and reducing disturbances to the original turns ratio and tap current distribution. Without the need for complex auxiliary voltage dividers, active drive sources, or feedback control networks, compensation can be achieved using passive capacitor elements, making it suitable for autotransformer inductive voltage dividers that have already been manufactured or whose structure is already defined.

[0131] Furthermore, the embodiments of the present invention are also applicable to autotransformer inductive voltage dividers without shielded wire structures. For autotransformer inductive voltage dividers without independent shielded wire nodes, the actual compensation node can be a core wire proportional tap. .

[0132] For autotransformer voltage dividers with shielded wires, the compensation capacitor can preferably be connected to the shielded wire compensation node; for autotransformer voltage dividers without shielded wires, the compensation capacitor can be connected to the core wire proportional tap, tap-to-tap node, or other equivalent compensation node, thereby adapting to the compensation requirements under different error formation paths.

[0133] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for error compensation of an autotransformer inductive voltage divider, characterized in that, include: S1. For an autotransformer with a shielded wire structure, establish a corresponding transmission model; wherein, the transmission model includes a core wire segmented series impedance branch, a shielded wire segmented series impedance branch, a core wire-shielded wire coupling admittance branch, a shielded wire external reference body admittance branch, and a shielded wire node distributed admittance branch. S2. Eliminate the shielded wire nodes as internal nodes to simplify the transmission model into a core-side equivalent parameter model. The core-side equivalent parameter model includes equivalent series branches, equivalent inter-node admittance branches, and equivalent node external reference body admittance branches. The core-side equivalent parameter model is represented by six equivalent parameters on the core side, including the equivalent leakage inductance on the core side. Equivalent AC resistance on the core wire side Equivalent mutual capacitance on the core wire side Equivalent external reference capacitance on the core wire side Equivalent mutual conductance on the core wire side Equivalent external reference body conductance of the core wire side ; S3. Obtain measured voltage proportional error data at multiple frequency points and multiple tap positions, and use a staged identification strategy to sequentially identify the parameters. , , , , and The equivalent parameter model of the target core wire side with determined parameters is obtained; S4, in the equivalent parameter model on the target core wire side... Equivalent node on each core wire side A compensation capacitor branch connected to the end reference tap is configured to construct a compensation calculation model; wherein, the equivalent node on the core wire side The compensation calculation node is used to calculate the complex potential and tap error before and after compensation; the compensation calculation model includes four types of branches: equivalent series branches between equivalent nodes on adjacent core wire sides, with admittance denoted as... The equivalent mutual admittance branches between equivalent nodes on the core wire side are denoted as follows: The equivalent admittance branch of the core wire side equivalent node to the external reference body, the admittance is denoted as The compensation capacitor branch between the equivalent node on the core side and the end reference tap, and the admittances of the equivalent node on the core side and the two ends are respectively denoted as... and ; S5. Based on the current balance relationship of the equivalent nodes on each core wire side in the compensation calculation model, the node equations after connecting the compensation capacitor are obtained: in: It is a complex coefficient matrix containing compensation capacitors. The right-hand vector is formed by the end reference potential, the external reference volume potential, and the compensation branch. This refers to the collection of all compensation capacitors; During the calculation, all compensation capacitors are set to zero to obtain the equivalent node complex potential on the core side in the uncompensated state; the compensation capacitor to be determined is substituted into the above node equation to obtain the equivalent node complex potential on the core side after compensation. S6. Solve for the set of compensation capacitors with the objective of minimizing the sum of squared complex voltage deviations at all equivalent nodes on the compensated core side: The compensation capacitance values ​​between each compensation calculation node and the end reference tap are obtained, and the connection point of the compensation capacitor is determined at the same time.

2. The method as described in claim 1, characterized in that, Also includes: When actually connecting the compensation capacitor, based on the corresponding relationship between the equivalent node on the core wire side and the shield wire node in terms of structural position, the calculated first... The compensation capacitor corresponding to each compensation capacitor value is connected to the shielding wire node. superior.

3. The method as described in claim 1, characterized in that, The establishment of a corresponding transmission model for an autotransformer with a shielded wire structure includes: The proportional winding in the autotransformer is divided along the tap direction into... Section, obtained Individual core wire nodes The shielded wire is divided into Each shielded wire node ;in, and Given the boundary nodes, to This is the tap node for the core wire to be determined. and For the boundary node of the shielding line, to For internal nodes of the shielding wire; Based on the proportional winding segmented series leakage impedance between adjacent core wire nodes, the segmented series impedance of the shield wire between adjacent shield wire nodes, the core wire-shield coupling admittance between core wire nodes and shield wire nodes at the same location, the external reference volume admittance of the shield wire node, and the distributed admittance between shield wire nodes of different shield wire nodes, corresponding branches are constructed; the transmission model including each branch is expressed as: ； ； in, The core wire is segmented and connected in series with impedance. AC resistance for segmented core wires. Leakage inductance is segmented for the core wire. The admittance is segmented and connected in series for the core wire. The shielded wire is segmented and connected in series with impedance. Segmented AC resistance for the shielded wire, For segmented inductance of shielded wire, For the segmented series admittance of the shielded wire, For chip-screen local coupling admittance. The coupling conductance between the core wire and the shield wire. This refers to the coupling capacitance between the core wire and the shield wire. For the external reference body admittance of the shielded wire, The equivalent conductivity of the shielded wire node to the external reference body. The equivalent capacitance of the shielded wire node to the external reference body; To shield the distribution of guides between nodes, To shield the equivalent mutual conductance between nodes. This is to shield the equivalent mutual capacitance between nodes.

4. The method as described in claim 3, characterized in that, The step of eliminating elements by treating the shielded wire nodes as internal nodes and simplifying the transmission model into an equivalent parameter model on the core wire side includes: The shielded wire node is used as an internal node for elimination. Based on the combined effect of the shielded wire potential distribution and the core wire-shielded wire coupling feedback on the complex potential of the core wire tap node, the parameters of the transmission model are expressed as core wire-side equivalent parameters, including core wire-side equivalent series parameters, core wire-side equivalent mutual admittance parameters, and core wire-side equivalent external reference body admittance parameters. Construct the equivalent parameter model of the core wire side based on the relevant parameters of the core wire side.

5. The method as described in claim 4, characterized in that, The method employs a phased identification strategy, sequentially identifying parameters. , , , , and ,include: Phase 1: Under the aforementioned capacitance constraint, the equivalent leakage inductance on the core wire side is used. Equivalent external reference capacitance on the core wire side For the parameters to be identified, the parameters are... and Set the value to 0, construct the error objective function between the calculated ratio difference of the equivalent parameter model on the core wire side and the measured ratio difference data at multiple frequency points and multiple tap positions, and solve for the error objective function with the goal of minimizing the error objective function. and And determined by the capacitance constraint. ; Phase Two: Utilize the gains from Phase One , and Substituting the equivalent parameter model of the core wire side, calculate the calculated ratio difference of the equivalent parameter model of the core wire side in the first stage; subtract the measured ratio difference data from the calculated ratio difference of the first stage to obtain the residual ratio difference; define the loss class combination parameters: and The residual ratio difference is then fitted with the values ​​of the loss-type combination parameters to obtain the solution. and ; Phase 3: Fixed gains from Phase 1 , , The second phase of acquisition , The equivalent AC resistance of the core wire side As parameters to be identified, the measured voltage proportional error data at multiple frequency points and multiple tap positions are substituted into the equivalent parameter model on the core wire side for determination. The equivalent mutual conductance on the core side is obtained from the relationship in the second stage. Equivalent external reference body conductance of the core wire side .

6. The method as described in claim 5, characterized in that, The complex coefficient matrix containing the compensation capacitor The matrix elements are constructed as follows: For the line, number The column, with diagonal elements as follows: The off-diagonal elements are: ,in, , Indicates when Time to take Otherwise take ; Right-hand vector The first element in the matrix The elements are: in, The external reference potential is usually taken as the ground potential; Indicates when Time to take Otherwise take ; Indicates when Time to take Otherwise take When the external reference body is at ground potential: .

7. The method as described in claim 6, characterized in that, After the compensation capacitor is connected, the first The complex voltage deviation of each compensation calculation node is: in, To compensate for the capacitor after connection Complex potential of equivalent nodes on each core wire side; For the first The ideal following potential of each node The real part of the complex voltage deviation corresponds to the proportional error of the tap voltage amplitude, and the imaginary part corresponds to the phase error of the tap voltage.

8. The method as described in claim 1, characterized in that, When the structure and error distribution of the autotransformer satisfy a symmetrical relationship, the number of compensation calculation nodes to be determined is reduced by using half-side calculation and mirror connection. Let the number of internal compensation nodes be: The number of independently calculated half-side compensation nodes is: Let the vector of the independent half-side compensation capacitor be: ,in, , indicating the first The compensation capacitance value required for each half-side compensation node; The variable to be determined is from the complete set of compensation capacitors. Simplified to half-side independent compensation capacitor vector By solving The theoretical value of half of the compensation capacitor is obtained, and the compensation capacitor values ​​and access terminals of all compensation nodes are obtained through mirror relationship.

Images