Intelligent early warning and power distribution method for high-low voltage power distribution cabinet

Abstract

The present application relates to high and low voltage power distribution cabinet technical field, especially to a kind of high and low voltage power distribution cabinet's intelligent early warning and power distribution method.The method includes S1, signal acquisition;S2, establish LSTM early warning model;S3, low voltage side fast switching;S4, dynamic virtual impedance compensation;S5, fault recovery.The present application detects the travelling wave signal of high and low voltage power distribution cabinet high voltage side, introduces it into LSTM early warning model, predicts fault development time, thereby early warning to low voltage side, after receiving early warning signal, according to multistage load fast cut-off strategy, step by step cut off non-key load on production line, thereby avoid load in the process of fast voltage sudden drop downtime even burn, while starting dynamic virtual impedance compensation, prevent high voltage side voltage flash when low voltage side fast cut-off, and key load keeps production line basic operation by switching standby busbar.

Description

Technical Field

[0001] This invention relates to the field of high and low voltage switchgear technology, and in particular to an intelligent early warning and power distribution method for high and low voltage switchgear. Background Technology

[0002] High and low voltage switchgear is a power distribution device used in power supply systems for power distribution, control, metering, and cable connection. Conventional household electricity is generally low voltage. High voltage electricity is distributed to users after being stepped down by high and low voltage switchgear in substations. However, industrial electricity consumption is relatively large. To avoid losses during transportation and transfer, factories usually choose to directly connect to high voltage electricity and use their own high and low voltage switchgear for power distribution to meet actual production needs.

[0003] In traditional protection systems, when a high-voltage switchgear malfunctions, such as when insulation breakdown triggers a trip, the low-voltage side equipment will experience a rapid voltage drop, leading to shutdown. This could cause a sudden interruption or even burnout of a production line, affecting not only the normal operation of the factory production line but also potentially causing irreparable damage to the equipment. Summary of the Invention

[0004] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description and other accompanying drawings.

[0005] The purpose of this invention is to overcome the above-mentioned shortcomings and provide an intelligent early warning and power distribution method for high and low voltage switchgear. By detecting the traveling wave signal on the high-voltage side of the high and low voltage switchgear, the method incorporates it into an LSTM early warning model to predict the fault development time, thereby providing early warning for the low-voltage side. After receiving the early warning signal, the low-voltage side gradually cuts off non-critical loads on the production line according to a multi-level load rapid disconnection strategy, thereby preventing the loads from crashing or even burning out during rapid voltage drops. At the same time, dynamic virtual impedance compensation is activated to prevent voltage flashover on the high-voltage side when the low-voltage side is rapidly disconnected. Meanwhile, critical loads maintain the basic operation of the production line by switching to a backup bus.

[0006] This invention provides an intelligent early warning and power distribution method for high and low voltage switchgear, comprising:

[0007] S1. Signal Acquisition: Acquire the voltage traveling wave signal on the high-voltage side of the high- and low-voltage distribution cabinet. The sampling window size is 100. When the voltage change rate exceeds the threshold, an early warning detection is initiated. The voltage change rate threshold is calculated as (0.35 * peak voltage) / 1 microsecond.

[0008] S2. Establish an LSTM early warning model: Preprocess the voltage traveling wave signal, extract multiple features, and input them into the LSTM time series model. Establish the LSTM time series model through deep learning and output the predicted fault development time. ;

[0009] S3, Low-voltage side rapid switching: Within a certain period of time before a fault occurs on the high-voltage side, the multi-level load rapid disconnection strategy of the low-voltage cabinet is activated. Multiple switches control the disconnection of their corresponding production line loads. When the voltage drop exceeds the threshold, the switch is switched to the backup bus.

[0010] S4. Dynamic Virtual Impedance Compensation: Calculate the inrush current when the low-voltage side is cut off, and activate dynamic virtual impedance compensation on the high-voltage side to prevent voltage flashover on the high-voltage side when the low-voltage side is quickly cut off.

[0011] S5. Fault Recovery: After the fault in the high-voltage switchgear is cleared, the low-voltage switchgear automatically restores the load according to priority.

[0012] In some embodiments, in step S2, the LSTM temporal model specifically includes an input layer, a first LSTM layer, a second LSTM layer, a fully connected layer, and an output layer, wherein the first LSTM layer has 64 units, the second LSTM layer has 32 units, the fully connected layer has 16 neurons, the output layer has 1 output neuron, a dropout layer is set between the first LSTM layer and the second LSTM layer with a dropout rate of 20%, and the linear activation function of the fully connected layer is ReLU.

[0013] In some embodiments, in step S2, the input features include traveling wave steepness, polarity difference ratio, and high-frequency attenuation coefficient:

[0014] Traveling wave steepness, i.e., maximum voltage change rate , representing the initial fault intensity;

[0015] Polarity difference ratio, i.e. / Positive wave amplitude represents the fault phase identification degree;

[0016] The high-frequency attenuation coefficient, which is the spectral energy above 500kHz / spectral energy above 100kHz, represents the degree of insulation degradation.

[0017] In some embodiments, in step S3, the fault development time is predicted based on the output of the LSTM time series model in step S2. Establish a judgment mechanism:

[0018] when The system sends a pre-protection command to the low-voltage switchgear to warn of a possible fault on the high-voltage side and to pre-activate the dynamic virtual impedance compensation step to prevent voltage flashover on the high-voltage side when the low-voltage side is quickly disconnected.

[0019] when The low-voltage switchgear officially launched a multi-level load rapid cut-off strategy, with multiple switches controlling the cut-off of their corresponding production line loads.

[0020] At the same time, additional early warning conditions are set when... and When the signal is interrupted, a cut-off signal is sent directly to the low-voltage switchgear without waiting for t to be less than 50ms.

[0021] This indicates the slope of the fault development. If it exceeds the threshold, it means that the fault is deteriorating rapidly. Therefore, there is no need to wait for the low-voltage cabinet to respond; a cut-off signal can be sent directly.

[0022] In some embodiments, the multi-level load fast disconnection strategy sets priorities for each level of load, including non-critical loads, semi-critical loads, and critical loads. The specific correspondence between the control switches and each level of load is as follows:

[0023] For non-critical loads, the interruption time is less than 50ms, controlled by solid-state switches (SSRs).

[0024] Semi-critical loads, controlled by magnetic latching relays, can be interrupted for 100-200ms;

[0025] Critical loads are controlled uninterruptedly by STS static switches;

[0026] When a voltage drop of >40% is detected on the high-voltage side, the STS static switch connected to the critical load is switched to the backup bus power supply.

[0027] In some embodiments, the priority of each load level can be dynamically adjusted based on the real-time load rate P:

[0028]

[0029] in, , The weighting coefficient for each equipment is set according to the actual production line conditions. For real-time load power, For the maximum allowable power of the line, This is the per-unit value of the power change rate.

[0030] In some embodiments, the specific steps of dynamic virtual impedance compensation in step S4 are as follows:

[0031] S41. Real-time calculation of load inrush current on the low-voltage side. :

[0032]

[0033] in, This is the peak value of the inrush current. It is an exponentially decaying function. The decay time constant, The amplitude of the oscillation component. The oscillation frequency is... The initial phase angle;

[0034] S42, High-voltage side based on load inrush current Dynamically adjust the virtual impedance of the SVG :

[0035]

[0036] Where K is the adaptive proportional coefficient, which automatically adjusts according to the system's short-circuit capacity. Here, represents the system short-circuit capacity, and represents the system setpoint. Here, THD is the harmonic suppression coefficient, and THD is the total harmonic distortion.

[0037] The specific formula for calculating the adaptive scaling factor K is as follows:

[0038]

[0039] in, Here, represents the system short-circuit capacity, and represents the system setpoint. This serves as the system's baseline capacity and also as the system's setpoint.

[0040] The specific formula for calculating the total harmonic distortion (THD) is as follows:

[0041]

[0042] in, For fundamental current, It is the second harmonic. Let h be the harmonic current.

[0043] In some embodiments, step S42 involves limiting the virtual impedance magnitude:

[0044]

[0045] Reference impedance :

[0046]

[0047] in, The system's rated voltage. This is the baseline capacity.

[0048] In some embodiments, the specific process of switching to backup bus power supply via STS static switch is as follows: First, it is detected whether the main bus voltage drops by more than 40%. If the drop exceeds the threshold, the phase of the backup bus is detected to obtain the phase difference with the main bus. Adjust the thyristor conduction angle to make the phase difference This enables seamless switching of critical loads to the backup bus.

[0049] In some embodiments, in step S5, recovery conditions are set such that when the main bus voltage returns to normal and lasts for more than 5 seconds, critical loads are switched to the main bus first, and then semi-critical loads and non-critical loads are restored in sequence. If the voltage is abnormal, the recovery is paused.

[0050] By adopting the above technical solution, the beneficial effects of the present invention are:

[0051] This invention detects the traveling wave signal on the high-voltage side of the high- and low-voltage distribution cabinet and incorporates it into an LSTM early warning model to predict the fault development time, thereby providing early warning for the low-voltage side. After receiving the early warning signal, the low-voltage side gradually disconnects non-critical loads on the production line according to a multi-level load rapid disconnection strategy, thereby preventing the loads from crashing or even burning out during rapid voltage drops. At the same time, dynamic virtual impedance compensation is activated to prevent voltage flashover on the high-voltage side when the low-voltage side is rapidly disconnected, while critical loads maintain the basic operation of the production line by switching to a backup bus.

[0052] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.

[0053] Undoubtedly, such and other objects of the present invention will become more apparent after the following detailed description of the preferred embodiments, which are illustrated in various accompanying drawings and figures.

[0054] To make the above and other objects, features and advantages of the present invention more apparent and understandable, one or more preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0055] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0056] In the accompanying drawings, the same parts use the same reference numerals, and the drawings are schematic and not necessarily drawn to actual scale.

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only one or more embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on such drawings without creative effort.

[0058] Figure 1 This is a schematic diagram of the overall process of intelligent early warning and power distribution in some embodiments of the present invention;

[0059] Figure 2 This is a schematic diagram of the LSTM early warning model structure in some embodiments of the present invention;

[0060] Figure 3 This is a schematic diagram of the collaborative control process between the high and low voltage distribution cabinets in some embodiments of the present invention;

[0061] Figure 4 This is a schematic diagram showing the correspondence between the low-voltage side load and the control switch in some embodiments of the present invention;

[0062] Figure 5 This is a schematic diagram of the gradual load recovery process after fault clearing in some embodiments of the present invention. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0064] Furthermore, in the description of this invention, it should be understood that the terms "center," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0065] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral unit; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. However, specifying a direct connection indicates that the two main bodies are not connected through a transitional structure, but rather formed as a whole through a connecting structure. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0066] In this invention, unless otherwise expressly specified and limited, the first feature "on" or "below" the second feature may be in direct contact with the first and second features, or indirect contact through an intermediate medium. In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0067] Reference Figure 1 , Figure 1 This is a schematic diagram of the overall process of intelligent early warning and power distribution in some embodiments of the present invention.

[0068] According to some embodiments of the present invention, the present invention provides an intelligent early warning and power distribution method for high and low voltage switchgear, comprising:

[0069] S1. Signal Acquisition: Acquire the voltage traveling wave signal on the high-voltage side of the high- and low-voltage distribution cabinet. The sampling window size is 100. When the voltage change rate exceeds the threshold, an early warning detection is activated. The voltage change rate threshold is calculated as (0.35 * peak voltage) / 1 microsecond; for example, the 10kV threshold is 0.35 * 14.14kV / 1. =4.95kV / The advantage of using traveling wave signals for early warning is that they have a certain resistance to external interference during transmission. Even when a fault occurs on the high-voltage side, the traveling wave signal can still transmit stably, which is beneficial for real-time monitoring and prediction of the fault's occurrence time. However, the disadvantage is that the traveling wave signal lacks sufficient discrimination for specific fault types, thus it cannot accurately predict specific fault types. Therefore, when it is introduced into the LSTM early warning model, the final output is the predicted fault development time. Rather than the specific type of fault.

[0070] Reference Figure 2 , Figure 2 This is a schematic diagram of the LSTM early warning model structure in some embodiments of the present invention.

[0071] S2. Establish an LSTM early warning model: Preprocess the voltage traveling wave signal, extract multiple features, and input them into the LSTM time series model. Establish the LSTM time series model through deep learning and output the predicted fault development time. ;

[0072] The LSTM temporal model specifically includes an input layer, a first LSTM layer, a second LSTM layer, a fully connected layer, and an output layer. The first LSTM layer has 64 units, the second LSTM layer has 32 units, the fully connected layer has 16 neurons, and the output layer has one output neuron. A dropout layer with a 20% dropout rate is placed between the first and second LSTM layers. The fully connected layer uses ReLU as its linear activation function. Multiple LSTM layers are used to extract higher-level temporal features, improving the output's prediction of fault development time. More accurate; the LSTM layer specifically includes an input gate, a forget gate, and an output gate. The input gate determines how many new features to store, the forget gate determines how many historical fault patterns to retain, and the output gate is used to control the current output amount; the Dropout layer inserted between the first LSTM layer and the second LSTM layer randomly masks 20% of neurons to prevent rote memorization of training data during the learning process and improve the generalization ability to new fault scenarios.

[0073] Input features include traveling wave steepness, polarity difference ratio, and high-frequency attenuation coefficient:

[0074] Traveling wave steepness, i.e., maximum voltage change rate , representing the initial fault intensity;

[0075] Polarity difference ratio, i.e. The positive wave amplitude represents the fault phase identification degree; if the polarity difference ratio is >0.6, it indicates a one-way grounding fault.

[0076] The high-frequency attenuation coefficient, which is the energy of the spectrum above 500kHz / the energy of the spectrum above 100kHz, represents the degree of insulation degradation. The larger the high-frequency attenuation coefficient, the more random the discharge, which indirectly indicates that the insulation degradation is more severe.

[0077] It is understandable that voltage V, positive wave amplitude, negative wave amplitude, and spectral energy can all be obtained during the traveling wave waveform detection process. The specific formulas and calculation methods for obtaining parameter values ​​from the waveform will not be described in detail here; please refer to existing technologies.

[0078] The three input features—traveling wave steepness, polarity difference ratio, and high-frequency attenuation coefficient—are combined over a continuous period of time to form an input matrix, which is then input into the LSTM time series model to output the predicted fault development time. .

[0079] Reference Figure 3 , Figure 3 This is a schematic diagram of the collaborative control process between the high and low voltage distribution cabinets in some embodiments of the present invention.

[0080] S3, Low-voltage side rapid switching: Within a certain period of time before a fault occurs on the high-voltage side, the multi-level load rapid disconnection strategy of the low-voltage cabinet is activated. Multiple switches control the disconnection of their corresponding production line loads. When the voltage drop exceeds the threshold, the switch is switched to the backup bus.

[0081] Based on the fault development time predicted by the LSTM time series model output in step S2. Establish a judgment mechanism:

[0082] when The system sends a pre-protection command to the low-voltage switchgear to warn of a possible fault on the high-voltage side and to pre-activate the dynamic virtual impedance compensation step to prevent voltage flashover on the high-voltage side when the low-voltage side is quickly disconnected.

[0083] when The low-voltage switchgear officially launched a multi-level load rapid cut-off strategy, with multiple switches controlling the cut-off of their corresponding production line loads.

[0084] At the same time, additional early warning conditions are set when... and When the signal is interrupted, a cut-off signal is sent directly to the low-voltage switchgear without waiting for t to be less than 50ms.

[0085] This indicates the slope of the fault development. If it exceeds the threshold, it means that the fault is deteriorating rapidly. Therefore, there is no need to wait for the low-voltage cabinet to respond; a cut-off signal can be sent directly.

[0086] Reference Figure 4 , Figure 4 This is a schematic diagram showing the correspondence between the low-voltage side load and the control switch in some embodiments of the present invention.

[0087] In the multi-level load quick disconnection strategy, priorities are set for each level of load, including non-critical loads, semi-critical loads, and critical loads. The specific correspondence between control switches and each level of load is as follows:

[0088] For non-critical loads, the interruption time is less than 50ms, controlled by solid-state switches (SSRs).

[0089] Semi-critical loads, controlled by magnetic latching relays, can be interrupted for 100-200ms;

[0090] Critical loads are controlled by STS static switches and are uninterrupted; when a voltage drop of >40% is detected on the high-voltage side, the STS static switch connected to the critical load is switched to the backup bus for power supply.

[0091] STS static transfer switch is a power supply two-to-one automatic switching system. Under normal operating conditions, when the main bus is within the normal voltage range, the load is always connected to the main bus. When the main bus fails, the load automatically switches to the backup bus. After the main bus returns to normal, the load automatically switches back to the main bus, providing dual bus power supply to the load. At the same time, phase synchronization is very important for STS static transfer switches because it can ensure that there is no voltage distortion or current surge when switching between two buses. Ideally, the output of the static switch will not be interrupted due to switching, except for a maximum interruption of about 0.2ms near the zero crossing of the AC current. To achieve this, static transfer switches usually have a synchronization range. Within this range, the static transfer switch can ensure that the phase difference between the two buses is kept within a very small range, thereby achieving seamless switching.

[0092] The specific process of switching to standby bus power supply via STS static switch is as follows: First, check if the main bus voltage drops by more than 40%. If the drop exceeds the threshold, check the standby bus phase to obtain the phase difference with the main bus. The STS static switch adjusts the thyristor conduction angle, and the phase-locked loop controller controls the phase difference. Excess energy is released into the capacitor, enabling a seamless switch of critical loads to the backup bus.

[0093] The priority of each load level can be dynamically adjusted based on the real-time load rate P:

[0094]

[0095] in, , The weighting coefficient for each equipment is set according to the actual production line conditions. For real-time load power, For the maximum allowable power of the line, This is the per-unit value of the power change rate;

[0096] The correspondence between load levels, priorities, and execution switches is shown in Table 1:

[0097] Table 1

[0098]

[0099] It is understandable that when high and low voltage distribution cabinets are initially installed, execution switches with corresponding priorities are configured for each load according to the production conditions of the basic production line, such as air conditioners - solid-state switches (SSR), production auxiliary equipment - magnetic latching relays, etc. As the production line is actually running, the priorities of some loads change. The above-mentioned dynamic adjustment range of priority levels does not contradict the specific type of execution switch. The dynamic adjustment of priority levels only affects the order in which they are prioritized for power-off.

[0100] S4. Dynamic Virtual Impedance Compensation: Calculate the inrush current when the low-voltage side is cut off, and activate dynamic virtual impedance compensation on the high-voltage side to prevent voltage flashover on the high-voltage side when the low-voltage side is quickly cut off.

[0101] The specific steps of dynamic virtual impedance compensation are as follows:

[0102] S41. Real-time calculation of load inrush current on the low-voltage side. :

[0103]

[0104] in, This is the peak value of the inrush current. It is an exponentially decaying function. The decay time constant, The amplitude of the oscillation component. The oscillation frequency is... The initial phase angle;

[0105] Peak inrush current The specific calculation formula is as follows:

[0106]

[0107]

[0108] in, For the low-voltage side current, the set window period ensures that the measured low-voltage side current covers at least 5 oscillation cycles, thereby improving the peak value of the inrush current. Accuracy; such as oscillation frequency Then the window period ;

[0109] decay time constant The decay time constant is obtained by extracting the envelope of the low-voltage side current variation curve, fitting the envelope to an exponential function, and solving it using the least squares method. This is achieved in a stable high- and low-voltage switchgear. The value is generally constant, ranging from 10 to 80 ms; correspondingly, the oscillation component can be obtained by subtracting the exponentially decaying component from the original current, and the amplitude of the oscillation component can be obtained by performing an FFT on the oscillation component. and oscillation frequency In a stable high and low voltage switchgear, the amplitude of the oscillation component and oscillation frequency It is usually a fixed value, such as Typically stable between 100-400Hz, it's understandable that the load inrush current is calculated in real time. The parameters vary greatly depending on the actual high and low voltage distribution cabinet model and configuration. Here, we only briefly describe the calculation method and do not go into the specific calculation process for specific models. The specific parameter values ​​are obtained through Rogowski coil and ADC sampling in the low voltage distribution cabinet.

[0110] S42, High-voltage side based on load inrush current Dynamically adjust the virtual impedance of the SVG :

[0111]

[0112] Where K is the adaptive proportional coefficient, which automatically adjusts according to the system's short-circuit capacity. Here, represents the system short-circuit capacity, and represents the system setpoint. Here, THD is the harmonic suppression coefficient, and the current variation rate is preferred. Accuracy is improved by obtaining the result through a three-point numerical differentiation method;

[0113] The specific formula for calculating the adaptive scaling factor K is as follows:

[0114]

[0115] in, The system short-circuit capacity is the system setpoint, determined based on the current short-circuit capacity of the high-voltage busbar. This is the system baseline capacity, also the system setpoint, typically taken as 100 MVA; when the system short-circuit capacity... When the value is large, the value of K is small, reducing the virtual impedance, which affects the system's short-circuit capacity. When the load is small, the K value is large, which enhances the virtual impedance and allows the compensation strength to automatically adapt to changes in the power distribution load.

[0116] The specific formula for calculating the total harmonic distortion (THD) is as follows:

[0117]

[0118] in, For fundamental current, It is the second harmonic. The current is the harmonic current of order h;

[0119] For harmonic suppression coefficient, Set segmented value retrieval rules: When THD < 8% A value of 0.05 is used for slight compensation of the total harmonic distortion (THD); when 8% ≤ THD ≤ 25%, A value of 0.07 is used for moderate compensation of the total harmonic distortion (THD); when THD > 25%, A value of 0.10 is used to enhance compensation for the total harmonic distortion (THD).

[0120] Virtual impedance amplitude limit:

[0121]

[0122] Reference impedance :

[0123]

[0124] in, The system's rated voltage. The virtual impedance amplitude is set at 0.01 times the base impedance and 0.30 times the base impedance. The dynamic virtual impedance enables adaptive adjustment of strong system weak compensation and weak system strong compensation, solving the problem of overcompensation or undercompensation that occurs when the power distribution load changes in traditional fixed compensation.

[0125] Reference Figure 5 , Figure 5 This is a schematic diagram of the gradual load recovery process after fault clearing in some embodiments of the present invention.

[0126] S5. Fault Recovery: After the fault in the high-voltage switchgear is cleared, the low-voltage switchgear automatically restores the load according to priority.

[0127] Set recovery conditions: when the main bus voltage returns to normal and lasts for more than 5 seconds, critical loads will be switched to the main bus first, and then semi-critical loads and non-critical loads will be restored in sequence. If the voltage is abnormal, the recovery will be paused.

[0128] It should be understood that the embodiments disclosed herein are not limited to the specific processing steps or materials disclosed herein, but should be extended to equivalent substitutions of such features as understood by those skilled in the art. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0129] The term "embodiment" in this specification refers to a specific feature or characteristic described in connection with an embodiment that is included in at least one embodiment of the invention. Therefore, phrases or "embodiments" appearing in various places throughout the specification do not necessarily refer to the same embodiment.

[0130] Furthermore, the described features or characteristics can be incorporated into one or more embodiments in any other suitable manner. In the above description, specific details, such as thickness, quantity, etc., are provided to provide a comprehensive understanding of embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented without the aforementioned specific details or may be implemented using other methods, components, materials, etc.

Claims

1. An intelligent early warning and power distribution method for high and low voltage switchgear, characterized in that, include S1. Signal Acquisition: Acquire the voltage traveling wave signal on the high-voltage side of the high- and low-voltage distribution cabinet. The sampling window size is 100. ; When the voltage change rate exceeds the threshold, an early warning detection is initiated. The voltage change rate threshold is calculated as (0.35 * peak voltage) / 1 microsecond. S2. Establish an LSTM early warning model: Preprocess the voltage traveling wave signal, extract multiple features, and input them into the LSTM time series model. Establish the LSTM time series model through deep learning and output the predicted fault development time. ; Input features include traveling wave steepness, polarity difference ratio, and high-frequency attenuation coefficient: Traveling wave steepness, i.e., maximum voltage change rate , representing the initial fault intensity; Polarity difference ratio, i.e. / Positive wave amplitude represents the fault phase identification degree; The high-frequency attenuation coefficient, which is the spectral energy above 500kHz / spectral energy above 100kHz, represents the degree of insulation degradation. Fault development time predicted based on LSTM time series model output Establish a judgment mechanism: when The system sends a pre-protection command to the low-voltage switchgear to warn of a possible fault on the high-voltage side and to pre-activate the dynamic virtual impedance compensation step to prevent voltage flashover on the high-voltage side when the low-voltage side is quickly disconnected. when The low-voltage switchgear officially launched a multi-level load rapid cut-off strategy, with multiple switches controlling the cut-off of their corresponding production line loads. At the same time, additional early warning conditions are set when... and When the signal is interrupted, a cut-off signal is sent directly to the low-voltage switchgear without waiting for t to be less than 50ms. This indicates the slope of the fault development. If it exceeds the threshold, it means that the fault is accelerating and deteriorating. Therefore, there is no need to wait for the low-voltage cabinet to respond. A cut-off signal is sent directly. S3, Low-voltage side rapid switching: Within a certain period of time before a fault occurs on the high-voltage side, the multi-level load rapid disconnection strategy of the low-voltage cabinet is activated. Multiple switches control the disconnection of their corresponding production line loads. When the voltage drop exceeds the threshold, the switch is switched to the backup bus. S4. Dynamic Virtual Impedance Compensation: Calculate the inrush current when the low-voltage side is cut off, and activate dynamic virtual impedance compensation on the high-voltage side to prevent voltage flashover on the high-voltage side when the low-voltage side is quickly cut off. S5. Fault Recovery: After the fault in the high-voltage switchgear is cleared, the low-voltage switchgear automatically restores the load according to priority.

2. The intelligent early warning and power distribution method for high and low voltage distribution cabinets according to claim 1, characterized in that, In step S2, the LSTM temporal model specifically includes an input layer, a first LSTM layer, a second LSTM layer, a fully connected layer, and an output layer. The first LSTM layer has 64 units, the second LSTM layer has 32 units, the fully connected layer has 16 neurons, and the output layer has 1 output neuron. A dropout layer with a dropout rate of 20% is set between the first and second LSTM layers, and the linear activation function of the fully connected layer is ReLU.

3. The intelligent early warning and power distribution method for high and low voltage distribution cabinets according to claim 2, characterized in that, In the multi-level load quick disconnection strategy, priorities are set for each level of load, including non-critical loads, semi-critical loads, and critical loads. The specific correspondence between control switches and each level of load is as follows: For non-critical loads, the interruption time is less than 50ms, controlled by solid-state switches (SSRs). Semi-critical loads, controlled by magnetic latching relays, can be interrupted for 100-200ms; Critical loads are controlled uninterruptedly by STS static switches; When a voltage drop of >40% is detected on the high-voltage side, the STS static switch connected to the critical load is switched to the backup bus power supply.

4. The intelligent early warning and power distribution method for high and low voltage distribution cabinets according to claim 3, characterized in that, The priority of each load level can be dynamically adjusted based on the real-time load rate P: in, , The weighting coefficient for each equipment is set according to the actual production line conditions. For real-time load power, For the maximum allowable power of the line, This is the per-unit value of the power change rate.

5. The intelligent early warning and power distribution method for high and low voltage distribution cabinets according to claim 1, characterized in that, In step S4, the specific steps for dynamic virtual impedance compensation are as follows: S41. Real-time calculation of load inrush current on the low-voltage side. : in, This is the peak value of the inrush current. It is an exponentially decaying function. The decay time constant, The amplitude of the oscillation component. The oscillation frequency is... The initial phase angle; S42, High-voltage side based on load inrush current Dynamically adjust the virtual impedance of the SVG : Where K is the adaptive proportional coefficient, which automatically adjusts according to the system's short-circuit capacity. Here, represents the system short-circuit capacity, and represents the system setpoint. Here, THD is the harmonic suppression coefficient, and THD is the total harmonic distortion. The specific formula for calculating the adaptive scaling factor K is as follows: in, Here, represents the system short-circuit capacity, and represents the system setpoint. This serves as the system's baseline capacity and also as the system's setpoint. The specific formula for calculating the total harmonic distortion (THD) is as follows: in, For fundamental current, It is the second harmonic. Let h be the harmonic current.

6. The intelligent early warning and power distribution method for high and low voltage distribution cabinets according to claim 5, characterized in that, The virtual impedance magnitude limitation is involved in step S42: Reference impedance : in, The system's rated voltage. This is the baseline capacity.

7. The intelligent early warning and power distribution method for high and low voltage distribution cabinets according to claim 3, characterized in that, The specific process of switching to standby bus power supply via STS static switch is as follows: First, check if the main bus voltage drops by more than 40%. If the drop exceeds the threshold, check the standby bus phase to obtain the phase difference with the main bus. Adjust the thyristor conduction angle to make the phase difference This enables seamless switching of critical loads to the backup bus.

8. The intelligent early warning and power distribution method for high and low voltage distribution cabinets according to claim 1, characterized in that, In step S5, recovery conditions are set. When the main bus voltage returns to normal and lasts for more than 5 seconds, critical loads are switched to the main bus first, and then semi-critical loads and non-critical loads are restored in sequence. If the voltage is abnormal, the recovery is paused.

Images