A hyperconverged pole-mounted switchgear

Abstract

This invention discloses a hyper-converged pole-mounted switchgear, belonging to the technical field of switchgear. The device includes a switch body, a power extraction unit, and a feeder terminal. The switch body comprises a deeply fused, solid-sealed pole with integrated casting, housing a vacuum interrupter, isolating switch contacts, voltage and current sensors. The power extraction unit supplies power to the deeply fused, solid-sealed pole via a separable spring-loaded electrical contact structure. The feeder terminal is embedded in the switch housing and is operated under energization via a dedicated operating lever. This invention also provides a transient electromagnetic interference suppression module to eliminate electromagnetic interference in the device; a CLC topology power extraction unit adaptable to multiple operating conditions; and a status monitoring module that integrates multi-source data for fault diagnosis, solving the problems of low integration degree, weak electromagnetic interference resistance, poor power extraction reliability, and low intelligence in existing primary and secondary fused pole-mounted switches. This device can provide intelligent power supply for 10kV distribution networks, improving reliability and reducing operation and maintenance costs.

Description

Technical Field

[0001] This invention relates to a power grid switching device, and more particularly to a hyper-converged pole-mounted switching device. Background Technology

[0002] With the advancement of the construction of new power systems, the distribution network is gradually transforming from a power network that simply receives and distributes electricity to users into a power network that integrates and interacts with power sources, grids, loads, and storage, and is flexibly coupled with the upper-level power grid.

[0003] The integrated primary and secondary pole-mounted switch is an advanced distribution network automation device. As a key piece of equipment in my country's 10kV distribution network, it achieves intelligent monitoring and fault handling of distribution network operation through deep integration of traditional primary equipment (such as circuit breakers) and secondary equipment (such as control and protection devices). This not only improves the reliability and security of the distribution network but also provides strong technical support for the intelligentization and automation of the power grid. Its functions are mainly reflected in the following aspects: 1) Intelligent sensing: Integrated sensors and intelligent terminals enable the equipment to monitor the line operating status in real time and perform data analysis; 2) Fault identification and handling: When a fault occurs, the equipment can quickly and accurately identify the fault type and isolate and handle it, reducing the scope and duration of power outages; 3) High-reliability power supply: Automated control strategies ensure the continuity and reliability of power supply.

[0004] Currently, there is a large number of primary and secondary fusion devices. The application of these devices is not only related to the safety and reliability of power supply, but also an important part of the new energy system and a key carrier for achieving the "dual carbon" goal.

[0005] However, existing primary and secondary fusion column switches still have the following prominent problems:

[0006] 1. Low integration: Each component needs to be installed, wired and debugged independently, which is a complicated process; the equipment is large in size and there are many connection points between components, resulting in a high failure rate; the installation efficiency is low and the technical skills required of the operators are high.

[0007] 2. Poor signal stability in strong electromagnetic environments: Signals are susceptible to transient electromagnetic interference, leading to inaccurate equipment status monitoring data and poor reliability of control signals, which affects the stability of power distribution network operation.

[0008] 3. Low power supply reliability: The power supply is low and it relies on the backup battery to drive the switch. The backup battery has a short lifespan and is prone to failure. Furthermore, there is a lack of power supply control strategies that can adapt to the three operating conditions of "normal operation, overvoltage operation, and secondary short circuit operation", which further increases the equipment failure rate.

[0009] 4. Low level of intelligence: The condition monitoring and fault diagnosis technologies are not perfect, and the data clustering does not distinguish the importance of each dimension (such as current, voltage, and temperature), resulting in poor clustering effect; the fault location relies only on a single fault traveling wave signal, and the location accuracy of high-resistance faults (such as 700Ω transition resistors) is low; intelligent fault decision-making based on the fusion of multi-source heterogeneous information (traveling wave, injected signal, μPMU data) has not been realized. Summary of the Invention

[0010] Purpose of the invention: The purpose of this invention is to provide a hyper-converged pole-mounted switchgear. Through multi-component hyper-convergence, adaptive electromagnetic interference suppression, multi-state adaptive self-powering, and intelligent fault diagnosis technologies, the device achieves miniaturization, high reliability, low maintenance, and intelligent operation, solving the problems of low integration degree of primary and secondary converged pole-mounted switches, weak anti-electromagnetic interference capability, poor power supply reliability, and low level of intelligence.

[0011] Technical solution: The hyper-converged pole-mounted switchgear includes a switch body, a power supply unit, and a feeder terminal, all of which are housed within the switchgear enclosure;

[0012] The switch body includes a deeply integrated solid-sealed pole, a circuit breaker mechanism, and a disconnecting switch mechanism. The deeply integrated solid-sealed pole integrates a vacuum interrupter, the contact of the disconnecting switch, and a measuring sensor. The vacuum interrupter is connected in series with the contact of the disconnecting switch, and the connection between the vacuum interrupter and the circuit breaker mechanism, and the connection between the contact of the disconnecting switch and the disconnecting switch mechanism, control the opening and closing of the main circuit of the power distribution network.

[0013] The power supply unit is used to supply power to the hyper-converged pole-mounted switchgear;

[0014] The feeder terminal is detachably connected to the switch body via a detachable connector to form a secondary circuit. The secondary circuit is used to transmit the monitoring and control signals required by the feeder terminal, and to supply power at the same time.

[0015] Optionally, the hyperconverged pole-mounted switchgear further includes a transient electromagnetic interference (EMI) suppression module, used to eliminate electromagnetic interference in the device using adaptive transient EMI suppression technology, wherein the adaptive transient EMI suppression technology includes:

[0016] Based on the set wavelet basis function, the length of the signal to be processed, the sampling frequency, and the bandwidth of transient electromagnetic interference, the number of wavelet packet decomposition layers, i.e. the decomposition depth of the wavelet packet number, is determined.

[0017] Based on the relationship between the decomposition depth and the maximum resolution of the frequency band after wavelet packet decomposition, the integration interval is obtained, and the energy distribution of the wavelet packet decomposition process is calculated, including the variance σ of the portion entering the low-pass filter after decomposition. l 2and the variance σ entering the high-pass filter section h 2 ;

[0018] Based on the variance σ of the original signal x 2 and the variance σ after wavelet packet decomposition l 2 σ h 2 The energy concentration degree G and sub-power ratio η are calculated, and then irregular decomposition is performed according to the energy concentration situation. Based on the threshold, all nodes where the interference is located are subjected to the first interference suppression process, while other nodes maintain the original wavelet packet coefficients unchanged. The first interference suppression process includes setting the wavelet packet coefficients above the threshold to zero.

[0019] The signal after the interference suppression processing is subjected to a second interference suppression processing using an LMS adaptive filter and / or an adaptive notch filter; wherein the second interference suppression processing includes: after the signal is input into the LMS adaptive filter, the filter coefficients are weighted and iterated through an adaptive filtering algorithm to minimize the error between the output signal and the desired signal; the adaptive notch filter dynamically adjusts the filter parameters through weighting coefficients to interfere with phase and amplitude changes.

[0020] The design of wavelet basis functions ensures the accurate decomposition of distribution network signals (such as 10kV voltage signals and 50Hz current signals), distinguishes normal signals from transient electromagnetic interference (such as high-frequency interference caused by lightning strikes), and thus solves the problem of signal instability in strong electromagnetic environments.

[0021] Optionally, when setting the threshold, both electromagnetic interference and noise signals are considered to be eliminated simultaneously. The calculation steps are as follows:

[0022] Analyze the variance of the noise signal after wavelet packet decomposition and perform Gaussian noise modeling;

[0023] Calculate the false alarm probability based on the probability density function of the Gaussian distribution;

[0024] The threshold in the wavelet packet domain is calculated based on the noise signal variance and the false alarm probability.

[0025] Optionally, the irregular decomposition includes:

[0026] If the energy concentration degree G is greater than or equal to the preset threshold Th_G, the corresponding node continues to be decomposed; if the energy concentration degree G is less than or equal to Th_G, the decomposition stops.

[0027] If the sub-power ratio η is greater than or equal to the preset threshold Th_η, the interference is located in the right sub-node; if the sub-power ratio η is less than or equal to Th_η, the interference is located in the left sub-node.

[0028] The preset threshold Th_G is calculated based on the noise variance and the false alarm probability. The preset threshold Th_η = 1 + ε, where ε represents the tolerance.

[0029] Optionally, the .in, Let P be the standard deviation of the noise S(t). fa This represents the probability of a false alarm.

[0030] Optionally, the LMS adaptive filter performs the following steps when employing the adaptive filtering algorithm:

[0031] The error signal of the filtering system is calculated using the following formula:

[0032]

[0033] Among them, X j Let be the signal after wavelet packet j-level decomposition, W be the filter weighting coefficients, and d be the signal after wavelet packet j-level decomposition. j The desired output signal of the system;

[0034] According to W j+1 =W j +2μe j Update the filter tap coefficients, where E[·] represents the expectation, μ is the step size factor, and the following requirements are met:

[0035]

[0036] Among them, X j,k Let N be the signal of the k-th frequency band after the wavelet packet is decomposed at level j, and N be the actual number of frequency bands after the wavelet packet is decomposed at level j.

[0037] Optionally, the power extraction unit adopts a CLC capacitor topology, and the parameters of the power extraction capacitor and inductor are matched according to three operating modes: normal operation, overvoltage operation, and secondary short-circuit operation. The power extraction unit also includes a Vienna rectifier and an LLC resonant converter. The output terminal of the Vienna rectifier is electrically connected to the input terminal of the LLC resonant converter. A voltage-current dual closed-loop PI control strategy is used to adjust the phase of the input current to synchronize with the voltage. Constant voltage or constant current output is achieved through voltage loop and current loop sliding mode controllers.

[0038] Optionally, it also includes a condition monitoring and fault diagnosis module, which performs the following operations: preprocessing the monitoring data and calculating the lower quartiles of the data in each dimension. and upper quartiles The interquartile relative deviation coefficients are obtained using the following formula. :

[0039]

[0040] The interquartile relative deviation coefficient of each dimension of data is used as the weight of that dimension, i.e.

[0041]

[0042] Substituting the following weighted Euclidean distance formula, we can complete the improved SOM algorithm clustering:

[0043]

[0044] in, For the i-th dimension of the monitoring data, the feature value is taken. Let be the weight value of the i-th dimension corresponding to the SOM neuron. For the weights of each dimension, The total dimension.

[0045] Optionally, since the power grid uses three-phase lines and there is electromagnetic coupling between the lines, the condition monitoring and fault diagnosis module also includes a distribution network fault location subunit, which is used to convert the coupled three-phase voltage and current signals into sequence equations without coupling relationships for the three-phase injected distribution network using the phase mode transformation formula, and to perform fault traveling wave location based on the sequence equations.

[0046] Optionally, the phase mode transformation formula adopts the Karrenbauer phase mode transformation formula to decouple the sequence component into 0-mode and 1,2-mode components.

[0047] The phase mode transformation formula is:

[0048]

[0049] The inverse transform formula is:

[0050] .

[0051] Optionally, the status monitoring and fault diagnosis module further includes a distribution network fault location subunit, which is used to assist in determining the fault location by utilizing the traveling wave generated when the fault transfer device is closed, based on the fault traveling wave ranging.

[0052] Optionally, the status monitoring and fault diagnosis module further includes a fault branch identification and ranging subunit, which is used to determine the fault location by using the traveling wave generated when the fault transfer device is closed, based on the fault traveling wave ranging.

[0053] Optionally, the power supply unit adopts an integrated power supply PT, which is connected to a deeply fused solid-sealed pole through a separable spring electrical contact structure.

[0054] Optionally, the feeder terminal is a miniaturized plug-in feeder terminal (FTU).

[0055] Optionally, the switch body and the integrated power supply PT are mounted on the switch box; the FTU is embedded in the switch box and is operated under power via a dedicated operating lever.

[0056] Optionally, the measuring sensor includes a current sensor and a voltage sensor.

[0057] Optionally, the current sensor is an LPCT type current sensor.

[0058] Optionally, the voltage sensor is an electronic voltage sensor.

[0059] Optionally, the deeply fused solid-sealed pole is integrally cast using the APG process.

[0060] Optionally, the circuit breaker body, integrated power supply PT, current sensor, voltage sensor and disconnector are prefabricated, assembled and tested in the factory, and then installed as a whole in the switch box, which adopts IP67 protection design.

[0061] Optionally, the outside of the switch box is provided with a mechanical isolation break indicator and an electrical isolation break indicator, which are linked to the disconnect switch mechanism.

[0062] Optionally, the electrical contact resistance of the separable connector is ≤50mΩ.

[0063] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:

[0064] 1. Enhanced integration and reduced operation and maintenance costs: The integrated solid-sealed pole formed by APG process eliminates mechanical connection points between components, significantly reducing the number of failure points compared to traditional split-type equipment; the fully prefabricated and separable spring-type electrical contact structure reduces equipment installation time and volume, and solves the problem of cumbersome independent installation, wiring, and debugging of each component of the pole-mounted switchgear, greatly reducing installation difficulty and operation and maintenance costs.

[0065] 2. Enhanced electromagnetic interference resistance and signal stability: The transient electromagnetic interference suppression module forms a two-stage interference suppression mechanism. The second interference suppression process is further divided into: first, using the sparsity of the wavelet packet domain for coarse suppression at the wavelet packet domain threshold, and then performing fine suppression. Specifically, LMS focuses on time-varying interference, while the notch filter focuses on narrowband interference. This invention, through the organic combination of irregular decomposition, threshold suppression, and LMS filtering, achieves an interference suppression rate of over 95%, and significantly improves the signal-to-noise ratio, avoiding inaccurate monitoring data and unreliable control signals under strong electromagnetic environments, thus ensuring the stable operation of the power distribution network.

[0066] 3. Improve power supply reliability and extend equipment life: The CLC capacitor topology is compatible with various operating conditions and has high conversion efficiency; the multi-mode switching design without backup battery eliminates battery failure points, effectively extending the overall life of the equipment and solving the problem of low power supply reliability of PT.

[0067] 4. Enhance intelligence and accurately locate faults: Improved SOM algorithm significantly increases clustering accuracy; fusion of fault traveling wave and closing traveling wave ranging methods enables precise fault location; multi-source information fusion diagnosis enables accurate fault identification; accuracy is significantly improved compared to traditional single signal diagnosis, enabling intelligent monitoring of equipment status and rapid fault handling. Attached Figure Description

[0068] Figure 1 This is a diagram of the deep-integrated solid-state electrode of the present invention;

[0069] Figure 2 This is a connection diagram of the integrated power supply PT according to an embodiment of the present invention;

[0070] Figure 3 This is an assembly diagram of a miniaturized plug-in FTU according to an embodiment of the present invention;

[0071] Figure 4 This is a flowchart of the wavelet packet adaptive anti-interference algorithm according to an embodiment of the present invention;

[0072] Figure 5 This is a block diagram of the d-axis current inner loop controller in an embodiment of the present invention;

[0073] Figure 6 This is a block diagram of the voltage outer loop control system for the d-axis in an embodiment of the present invention;

[0074] Figure 7 This is a comparison chart of the positioning results between the traditional two-end ranging method and the integrated ranging method of this invention. Detailed Implementation

[0075] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0076] The present invention provides a hyper-converged pole-mounted switchgear, comprising a switch body, a power supply unit, and a feeder terminal, wherein the switch body comprises a deeply integrated solid-sealed pole, a circuit breaker mechanism, and a disconnecting switch mechanism.

[0077] In one embodiment, the deeply fused solid-sealed electrode post is integrally cast using the APG process, such as... Figure 1 As shown on the right side, it integrates a vacuum interrupter, the contacts of a disconnecting switch, a current sensor, and a voltage sensor, such as... Figure 1 The left side is shown.

[0078] In one embodiment, the power supply unit employs an integrated power supply PT, and the feeder terminal employs a miniaturized plug-in feeder terminal (FTU). For example... Figure 2 As shown, the integrated power supply PT uses a separable spring electrical contact structure to connect with a deeply fused solid-sealed pole. The FTU is detachably connected to the switch body via the separable connector electrical contact structure. The circuit breaker body, integrated power supply PT, current sensor, voltage sensor, and disconnector are prefabricated, assembled, and tested in the factory, and then installed as a whole in the switch box. The housing adopts an IP67 protection design. Figure 3 As shown.

[0079] The present invention provides a hyperconverged pole-mounted switchgear, which further includes a transient electromagnetic interference suppression module for using adaptive transient electromagnetic interference suppression technology to eliminate electromagnetic interference in the device.

[0080] like Figure 4 As shown, the adaptive transient electromagnetic interference suppression technology includes:

[0081] Wavelet basis functions were designed based on the principles of similarity, regularity, symmetry, compact support, vanishing moments, and discrete form. In this embodiment, by comparing the performance advantages and disadvantages of various wavelet basis functions, the 5th-order Daubechies wavelet basis function was finally selected.

[0082] The decomposition depth of the wavelet packet is determined based on the length of the signal to be processed, the sampling frequency, and the bandwidth of the transient electromagnetic interference.

[0083] Based on the relationship between the decomposition depth and the maximum resolution of the frequency band after wavelet packet decomposition, the integration interval is obtained, and the energy distribution of the wavelet packet decomposition process is calculated, including the variance σ of the portion entering the low-pass filter after decomposition. l 2 and the variance σ entering the high-pass filter section h 2 ;

[0084] Based on the variance σ of the original signal x 2 and the variance σ after wavelet packet decomposition l 2 σ h 2 Calculate the energy concentration degree G and the sub-power ratio η, then perform irregular decomposition according to the energy concentration situation, and perform first interference suppression processing on all nodes where the interference is located based on the threshold.

[0085] The signal after the interference suppression process is subjected to a second interference suppression process using an LMS adaptive filter and / or an adaptive notch filter.

[0086] In one implementation, the transient electromagnetic interference suppression module specifically performs the following adaptive transient electromagnetic interference suppression technology process:

[0087] Let the power spectral density function of the original signal be P. xx (ω), with variance σ x 2 The part that enters the low-pass filter after decomposition is denoted as l, and its variance is σ. l 2 The portion entering the high-pass filter is denoted as h, and its variance is σ. h 2 The energy distribution of the wavelet packet decomposition process can then be expressed as:

[0088]

[0089] The integration interval in the above formula can be obtained from the following formula:

[0090]

[0091] Where M is the decomposition level of the wavelet packet, that is, the decomposition depth of the wavelet packet, and B... W This represents the maximum bandwidth resolution after wavelet packet decomposition.

[0092] Define the energy concentration degree G and the subband power ratio η, where G represents the energy concentration of each layer during decomposition; η represents the power ratio of the high-frequency part to the low-frequency part in the subband, and the expressions are as follows:

[0093]

[0094] Then, based on the energy concentration degree G and the sub-power ratio η, irregular decomposition is performed according to the set threshold, and the first interference suppression processing is then applied to the node where the interference is located. Considering that in addition to electromagnetic interference, there are other noise signals, this embodiment models these noise signals as zero-mean Gaussian noise signals. Therefore, when performing the first interference suppression processing, the variance of the noise signal after wavelet packet transform needs to be considered.

[0095] Assume the variance of the noise S(t) is σ 2 Its power spectral density is S s (ω)=σ 2 The first-order wavelet packet is equivalent to the filter g. o (n), the first-order decomposition coefficient is W1, and the decomposed noise signal is denoted as W1S, with variance σ1. 2 The filter's frequency response is G0(ω), and g0(n) is a unit impulse response sequence. Then we have...

[0096]

[0097] In the formula, n is an integer representing the nth sample point in the unit impulse response sequence; ω represents the angular frequency. The W1S power spectral density is...

[0098]

[0099] The autocorrelation function of W1S is

[0100]

[0101] In the formula, τ represents the time offset. Since the mean of zero-mean noise remains zero after filtering, we have:

[0102]

[0103] Therefore, the noise signal, after wavelet packet decomposition, still follows a mean of zero and a variance of σ. 2 The probability density function of W1S is a Gaussian distribution. W1S for

[0104]

[0105] In the formula, x represents the possible values ​​of the noise signal W1S after first-level wavelet packet decomposition. Let Th_G be the threshold in the wavelet packet domain, and P fa Let be the probability of a false alarm, then we have

[0106]

[0107] Therefore, the threshold is obtained as

[0108] .

[0109] Then, based on the energy concentration degree G and the sub-power ratio η, the complete decomposition tree structure of the wavelet packet can be optimized into an irregular decomposition according to the energy concentration situation, which greatly improves the computational efficiency and signal fidelity. The specific decomposition process is as follows:

[0110] If the value of G is greater than or equal to the preset threshold Th_G, it indicates that there are still many interference components in the high-frequency components, and the node continues to be decomposed; if the value of G is less than Th_G, it indicates that the energy difference between the low-frequency components and the high-frequency components after the wavelet packet decomposition of the node is not significant, and the sub-band contains only noise power without interference power, so the node will not continue to be decomposed to the next layer after reaching this layer.

[0111] η can be used to determine the location of the interference in a child node. If η is greater than or equal to a preset threshold Th_η, the interference is located in the right child node; if η is less than or equal to Th_η, the interference is located in the left child node.

[0112] In one implementation, the first interference suppression process involves setting all wavelet packet coefficients exceeding a preset threshold Th_G in the node where the interference occurs to zero, while maintaining the original wavelet packet coefficients in other nodes, thereby achieving the effect of electromagnetic interference elimination. The preset threshold Th_G is based on... The calculation yielded that, Let P be the standard deviation of the noise S(t). fa The false alarm probability is given by the preset threshold Th_η = 1 + ε, where ε represents the tolerance.

[0113] Then, the signal after the first interference suppression process is input into the LMS adaptive filter to calculate the error signal of the filtering system:

[0114]

[0115] Among them, X j Let be the signal after wavelet packet j-level decomposition, W be the filter weighting coefficients, and d be the signal after wavelet packet j-level decomposition. j The desired output signal of the system;

[0116] According to W j+1 =W j +2μe j Update filter tap coefficients W j E[·] represents the expectation, μ is the step size factor, and the following requirements are satisfied:

[0117]

[0118] Among them, X j,k Let N be the signal of the k-th frequency band after the wavelet packet is decomposed at level j, and N be the actual number of frequency bands after the wavelet packet is decomposed at level j.

[0119] The power extraction unit adopts a CLC capacitor topology. First, the parameters of the power extraction capacitor and inductor are matched according to three operating modes: normal operation, overvoltage operation, and secondary short-circuit operation. In the pole-mounted switch power supply system, a Vienna rectifier is used on the rectifier side, and an LLC converter is used on the secondary side to achieve efficient power conversion and energy management. The voltage and current dual closed-loop PI control strategy quickly responds to the phase deviation between current and voltage, adjusting the phase of the input current in a timely manner to synchronize it with the voltage. Voltage and current loop sliding mode controllers construct the charging strategy. Based on the good dynamic performance of sliding mode control, various complex charging commands are tracked and implemented by changing the voltage or current reference values ​​to achieve constant voltage or constant current output.

[0120] Under steady-state conditions, the three-phase AC grid-side current i after coordinate transformation d i q Respectively with the command current i d * i q* By making comparisons and then adjusting using proportional-integral methods, the reference voltage vector V is further obtained. d V q The value of is determined, and the duration of action of each basic voltage vector is also determined. The above process can be described as follows:

[0121]

[0122] In the formula, K ip K iI These represent the proportional gain and integral gain of the inner current loop, respectively. d * i q * These are the rectifier grid-side current command signals in the synchronous rotating coordinate system dq, and e. d *(s) and e q *(s) represent the reference values ​​of the dq-axis voltage on the grid side in the rotating coordinate system, s is the integral coefficient of the transfer function of the integrator, and v d With v q These are the feedback values ​​of the three-phase grid-side voltages along the dq axis. A coefficient of is added. The decoupling mechanism is designed to compensate for the coupling between the d-axis and q-axis variables in the dq synchronous rotating coordinate system.

[0123] Combining the above three formulas, we get:

[0124]

[0125] Among them, e d and e q This represents the grid-side voltage in a rotating coordinate system. The system's d-axis reference is i. d * It can be obtained through the voltage outer loop proportional-integral control element, and it is related to the active power and proportional to the current flowing through the load. Reactive power is determined by the q-axis reference current i. q * To control it.

[0126] In the dq rotating coordinate system, the input current can be decoupled and controlled separately on the d-axis and q-axis. After decoupling control, the d-axis and q-axis become independent. Due to the symmetry of the two current inner loop structures, the d-axis is chosen as an example to illustrate the design of the PI regulator.

[0127] To control the output active power and DC-side output voltage of the Vienna rectifier, the output quantity of the outer loop of the DC output voltage can be used as the reference value for the d-axis current. The outer loop is designed using a PI control algorithm. In a two-phase synchronous rotating coordinate system, the state-space average model of the Vienna rectifier is as follows:

[0128]

[0129] In the formula, e d and e q i represents the grid-side voltage in the rotating coordinate system. d and i q d represents the grid-side current in the rotating coordinate system. d and d q For S a S b S c Variables in a synchronously rotating coordinate system.

[0130] Due to the coupling effect between the d-axis and q-axis components, they are prone to mutual interference during adjustment, increasing the design complexity of the control system. Therefore, a decoupling-based dual closed-loop control strategy is adopted to improve control accuracy and stability. Under the condition that the VIENNA rectifier achieves unity power factor operation, the core of the control lies in the synchronous grid-side current and voltage. By performing decoupling control in the dq rotating coordinate system and setting the q-axis current reference value to zero, reactive power components are effectively suppressed, thereby simplifying the control structure and improving the system's stability and dynamic response performance. Assuming i... ref As a current reference, K vP For proportional gain, K vI U is the integral coefficient. dc * represents the DC component reference value, U dc i is the DC component feedback value. qref with i qref The dq-axis component is the current reference.

[0131]

[0132] Current i after coordinate transformation d i q and the reference current value i respectively dref i qref By comparison, and after adjusting the proportional-integral parameters, the reference voltage U is obtained. d U q The value of . The formula is as follows:

[0133]

[0134] In the formula, K iP K iI These represent the proportional gain and integral gain of the inner current loop, respectively. dref i qref These are the signals for the grid-side current command in the synchronous rotating coordinate system dq. The system incorporates a decoupling compensation stage to improve the independence and speed of current control.

[0135] get:

[0136]

[0137] In the VIENNA rectifier system, active power is controlled by the d-axis current target value, while reactive power is controlled by the q-axis current target value. The input current can be decoupled into non-interfering d-axis and q-axis components, achieving independent control for each. This embodiment uses the d-axis as an example to illustrate its decoupled PI control strategy through a decoupled controller. The decoupled controller consists of two parts: an inner current loop controller and an outer current loop control system, as described below. Figure 5 , Figure 6 As shown.

[0138] In the current inner loop controller, T s For the current loop sampling operation period, 1 / (T) s s+1) represents the inertial element corresponding to the system's sampling delay, and k PWM / (0.5T s s+1) represents the inertial element corresponding to the PWM wave output delay. To simplify the analysis, the disturbance caused by the grid-side voltage is ignored, and the transfer function of the PI regulator is changed to a form with zeros and poles, as follows:

[0139]

[0140] Because the output active power of the VIENNA rectifier is related to the output DC side voltage U o Closely related. In order to achieve U o The system enables rapid and accurate adjustment, utilizing an outer voltage loop control system to ensure the stability of the DC output waveform. The output of the outer voltage loop is set to the reference value of the d-axis current, thereby ensuring the output voltage U... o It can accurately track the required reference voltage. The outer voltage loop employs a PI control strategy. Its transfer function is shown below:

[0141]

[0142] The aforementioned inner-loop current controller employs a proportional-integral regulator, which can be approximated as a control element with first-order inertial characteristics. Assume U... m Let C be the carrier coefficient of the PWM stage and C be the output filter capacitor. Then, the transfer function from the PWM control signal to the output capacitor current is 3U. m / 2V0, the transfer function from output capacitor current to output voltage is 2 / SC, resulting in Figure 6 The voltage loop closed-loop transfer function model is shown. Through this design, the inner current loop and the outer voltage loop form a complete closed-loop control system, effectively coordinating their operation and ensuring the stable operation and precise control of the VIENNA rectifier.

[0143] In one embodiment, the hyperconverged pole-mounted switchgear further includes a condition monitoring and fault diagnosis module. This module is used to: firstly, perform data preprocessing, data transformation, and data normalization to construct a complete dataset; then, propose an improved self-organizing feature map algorithm (SOM), assigning different weights to each dimension according to its importance, whereby the weighted Euclidean distance is expressed as:

[0144]

[0145] in, For the i-th dimension of the monitoring data, the feature value is taken. Let be the weight value of the i-th dimension corresponding to the SOM neuron. For the weights of each dimension, The total dimension is [value]. Weighted Euclidean distance is used to distinguish the importance of different dimensional attributes. Dimensions with lower importance have lower weights, which to some extent reduces the impact of excessively high dimensionality on poor clustering results. This invention uses the quartile relative deviation coefficient weighting method to assign weights.

[0146] Since the weighted method based on the coefficient of variation is limited by the mean not being zero, the "interquartile range coefficients" of each variable can be considered as their weights. If the lower quartiles of the data... and upper quartiles Given, then This reflects the central tendency of the data, while This reflects the deviation of the data, therefore a measure of relative deviation is defined:

[0147]

[0148] This is called the interquartile range (IVR). Calculating the IVR for each dimension of data allows us to use it as the weight for that dimension.

[0149]

[0150] When performing clustering, the appropriate use of weighted Euclidean distance can reflect the different roles of each variable in the data and can have a better effect on improving the clustering results.

[0151] In one embodiment, the condition monitoring and fault diagnosis module further includes a distribution network fault location subunit. This subunit employs a distribution network fault location method based on three-phase injection, comprising: using a phase mode transformation formula to convert it into an uncoupled sequence equation, i.e., line mode and zero mode, and then performing fault traveling wave location based on the sequence equation.

[0152] In this embodiment, the phase mode transformation formula adopts the Karrenbauer phase mode transformation formula to decouple the sequence component into 0-mode and 1,2-mode components.

[0153] The phase mode transformation formula is:

[0154]

[0155] The inverse transform formula is:

[0156] .

[0157] Furthermore, the condition monitoring and fault diagnosis module also includes a distribution network fault location subunit. This unit is based on the master-slave combined high-resistance fault location method of fault phase transfer. On the basis of fault traveling wave ranging, it uses the traveling wave generated when the fault transfer device is closed to assist in determining the fault location, thereby improving the accuracy of high-resistance fault location in the distribution network.

[0158] Furthermore, the condition monitoring and fault diagnosis module also includes a fault branch identification and location subunit. This unit, based on fault phase transfer and voltage injection fault branch identification and location methods, improves the accuracy of high-resistance fault location in the distribution network by utilizing the traveling wave generated when the fault transfer device closes, in addition to fault traveling wave location measurement. The fault transfer device is a high-power power electronic switching device capable of actively and quickly transferring fault current from the main circuit to a predetermined path. Its core function is not simply to "trip" the circuit after detecting a fault, but rather to first "accept" the fault current and then "disconnect" it at the control moment, thereby actively generating a controllable and measurable traveling wave signal.

[0159] In one embodiment, the fault transfer device employs a hybrid DC circuit breaker. Unlike traditional mechanical circuit breakers, the hybrid DC circuit breaker includes: an ultra-fast mechanical switch (UFD) to carry the normal operating current with extremely low losses during conduction; and a power electronic switch (such as an IGBT) connected in parallel with the mechanical switch, which rapidly (on the order of microseconds) turns on during a fault to provide a transfer path for the fault current, and then automatically and rapidly turns off after the mechanical switch opens, ultimately completely interrupting the fault current.

[0160] Traditional traveling wave ranging relies on the traveling wave generated instantaneously upon the occurrence of a fault. However, this method can sometimes result in low ranging accuracy or even failure due to weak traveling wave signals, complex waveforms, or significant interference. The active traveling wave ranging method used in this invention has the following basic process:

[0161] (1) Fault occurrence: A permanent fault occurs on the line (such as lightning strikes causing insulator breakdown).

[0162] (2) Device start-up: The protection system detects a fault and starts the fault transfer device (such as a hybrid circuit breaker).

[0163] (3) Initial interruption: The circuit breaker operates quickly to clear the fault current. The line enters a state without current, but the fault point still exists (permanent fault).

[0164] (4) Active closing: In order to accurately locate, the ranging system will command the fault transfer device to close again (usually close for a very short time, such as a few hundred microseconds and then disconnect again).

[0165] (5) Generating an auxiliary traveling wave: This active closing operation is equivalent to suddenly reapplying a voltage step to the faulty line. This operation will generate a traveling wave with very certain intensity, polarity, and timing, which will be reflected after propagating to the fault point.

[0166] (6) Precise measurement: The ranging equipment at the device will accurately capture this "actively generated" traveling wave and the traveling wave reflected back from the fault point. Since the closing time is known and the starting point of the traveling wave is very clear, the distance to the fault point can be calculated with extreme precision by calculating the time difference Δt between the arrival of the two traveling waves: distance = (wave velocity * Δt) / 2.

[0167] (7) Reconnection: After the distance measurement is completed, the device is quickly disconnected again to isolate the fault.

[0168] Therefore, the present invention has the following advantages:

[0169] High signal quality: The active-generated traveling wave has a more regular start time, amplitude, and waveform than the traveling wave generated when the fault occurs naturally, making it easier to identify.

[0170] Strong anti-interference capability: It avoids the contamination of traveling wave signals by complex electromagnetic transient processes (such as electric arcs and multiple breakdowns) that may accompany the initial stage of a fault.

[0171] Suitable for permanent faults: Especially suitable for accurately locating permanent faults after the circuit breaker trips for the first time, providing precise coordinates for line inspection and maintenance.

[0172] Figure 7 This figure compares the positioning results of the conventional two-end ranging method with the integrated ranging method of the present invention. Main line 1 and main line 2 in the figure use the integrated ranging positioning method of the present invention, while main line 3, branch line 1, and branch line 2 use the conventional two-end ranging method. As can be seen from the figure, the ranging positioning method of the present invention has a smaller ranging error.

[0173] Analysis reveals that traditional two-end ranging methods calculate the location distance of the fault point by combining the arrival time of the traveling wave at both ends with the wave velocity. The most significant factor affecting ranging accuracy is the accuracy of the wave velocity. However, in existing traditional two-end ranging methods, the wave velocity is calculated in advance using line parameters. The actual traveling wave velocity is not only closely related to sampling frequency, temperature, and earth resistivity, but also extremely difficult to obtain accurately from line parameters, leading to a significant discrepancy between the theoretically calculated wave velocity and the actual wave velocity.

[0174] The integrated ranging and positioning method of this invention incorporates closing traveling waves into the traditional dual-end ranging method. Switchgear upstream and downstream of the fault is closed separately. The distribution network fault location subunit can calculate the actual wave velocity based on the fault reflection wave time and closing time of the closing traveling wave, thereby reducing the impact of theoretically calculated waveforms on ranging accuracy.

[0175] The fault transfer device used in this invention is not only a protection device, but also an active measurement signal generator: through controlled "closing-opening" operation, a traveling wave is artificially and precisely generated, thereby solving the measurement problem caused by the uncontrollable signal source in traditional traveling wave ranging, and greatly improving the reliability and accuracy of fault ranging.

[0176] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Although this application discloses preferred embodiments as above, it is not intended to limit this application. Any changes or modifications made by those skilled in the art without departing from the scope of the technical solution of this application using the disclosed technical content are equivalent to equivalent implementation cases and all fall within the scope of the technical solution.

Claims

1. A hyperconverged pole-mounted switchgear, characterized in that, The switch body, power supply unit, and feeder terminal are all housed within the switch box. The switch body includes a deeply integrated solid-sealed pole, a circuit breaker mechanism, and a disconnecting switch mechanism. The deeply integrated solid-sealed pole integrates a vacuum interrupter, the contact of the disconnecting switch, and a measuring sensor. The vacuum interrupter is connected in series with the contact of the disconnecting switch, and the connection between the vacuum interrupter and the circuit breaker mechanism, and the connection between the contact of the disconnecting switch and the disconnecting switch mechanism, control the opening and closing of the main circuit of the power distribution network. The power supply unit is used to supply power to the hyper-converged pole-mounted switchgear; The feeder terminal is detachably connected to the switch body via a detachable connector to form a secondary circuit. The secondary circuit is used to transmit the monitoring and control signals required by the feeder terminal, and to supply power at the same time. The device also includes a transient electromagnetic interference suppression module, used to eliminate electromagnetic interference in the device using adaptive transient electromagnetic interference suppression technology, which includes: Based on the set wavelet basis function, the length of the signal to be processed, the sampling frequency, and the bandwidth of transient electromagnetic interference, the number of wavelet packet decomposition layers, i.e. the decomposition depth of the wavelet packet number, is determined. According to the relationship between the decomposition depth and the maximum resolution of the frequency band after wavelet packet decomposition, the integral interval is obtained, and the energy distribution of the wavelet packet decomposition process is calculated, including the variance σ of the part entering the low-pass filter after decomposition l 2 and the variance σ of the part entering the high-pass filter h 2 ; Based on the variance σ of the original signal x 2 and the variance σ after wavelet packet decomposition l 2 σ h 2 The energy concentration degree G and sub-power ratio η are calculated, and then irregular decomposition is performed according to the energy concentration situation. Based on a threshold, all nodes where interference is located are subjected to the first interference suppression process, while other nodes maintain their original wavelet packet coefficients unchanged. The first interference suppression process includes setting the wavelet packet coefficients above the threshold to zero. The irregular decomposition includes: if the energy concentration degree G is greater than or equal to a preset threshold Th_G, the corresponding node continues to be decomposed; if the energy concentration degree G is less than Th_G, the decomposition stops; if the sub-power ratio η is greater than or equal to a preset threshold Th_η, the interference is located in the right sub-node; if the sub-power ratio η is less than Th_η, the interference is located in the left sub-node. The preset threshold Th_G is calculated based on the noise variance and false alarm probability, and the preset threshold Th_η = 1 + ε, where ε represents the tolerance. The signal after the interference suppression processing is subjected to a second interference suppression processing using an LMS adaptive filter and / or an adaptive notch filter; wherein the second interference suppression processing includes: after the signal is input into the LMS adaptive filter, the filter coefficients are weighted and iterated through an adaptive filtering algorithm to minimize the error between the output signal and the desired signal; the adaptive notch filter dynamically adjusts the filter parameters through weighting coefficients to interfere with phase and amplitude changes.

2. The hyperconverged pole-mounted switchgear according to claim 1, characterized in that, When setting the threshold, both electromagnetic interference and noise signals are considered for elimination. The calculation steps are as follows: Analyze the variance of the noise signal after wavelet packet decomposition and perform Gaussian noise modeling; Calculate the false alarm probability based on the probability density function of the Gaussian distribution; The threshold in the wavelet packet domain is calculated based on the noise signal variance and the false alarm probability.

3. The hyperconverged pole-mounted switchgear according to claim 1, characterized in that, The power extraction unit includes a CLC capacitor topology structure for parameter matching of the power extraction capacitor and inductor according to three operating modes: normal operation, overvoltage operation, and secondary short-circuit operation. The power extraction unit also includes a Vienna rectifier and an LLC resonant converter. The output terminal of the Vienna rectifier is electrically connected to the input terminal of the LLC resonant converter. A voltage-current dual closed-loop PI control strategy is used to adjust the phase of the input current to synchronize with the voltage. Constant voltage or constant current output is achieved through voltage loop and current loop sliding mode controllers.

4. The hyperconverged pole-mounted switchgear according to any one of claims 1-3, characterized in that, The device also includes a condition monitoring and fault diagnosis module for monitoring the status of the pole-mounted switch: The monitoring data was preprocessed, and the lower quartiles of each dimension of the data were calculated. and upper quartiles Calculate the interquartile range coefficients using the following formula. : The interquartile relative deviation coefficient of each dimension of data is used as the weight of that dimension, that is: ， Substituting the following weighted Euclidean distance formula, we can complete the improved SOM algorithm clustering: in, For the i-th dimension of the monitoring data, the feature value is taken. Let be the weight value of the i-th dimension corresponding to the SOM neuron. For the weights of each dimension, The total dimension.

5. The hyperconverged pole-mounted switchgear according to claim 4, characterized in that, The condition monitoring and fault diagnosis module also includes a distribution network fault location subunit, which is used to convert the coupled three-phase voltage and current signals into sequence equations without coupling relationships for three-phase injected distribution networks using phase mode transformation formulas, and to perform fault traveling wave location based on the sequence equations.

6. The hyperconverged pole-mounted switchgear according to claim 5, characterized in that, The condition monitoring and fault diagnosis module also includes a distribution network fault location subunit, which is used to determine the fault location by using the traveling wave generated when the fault transfer device is closed, based on the fault traveling wave ranging.

7. The hyperconverged pole-mounted switchgear according to claim 5, characterized in that, The status monitoring and fault diagnosis module also includes a fault branch identification and ranging subunit, which is used to determine the fault location by using the traveling wave generated when the fault transfer device is closed, based on the fault traveling wave ranging.

8. The hyperconverged pole-mounted switchgear according to claim 1, characterized in that, The deep-integrated solid-sealed pole adopts an integrated casting process; the switch box adopts an IP67 protection design, and its outer side is equipped with mechanical isolation break indicator and electrical isolation break indicator linked to the disconnecting switch mechanism.

Images