Voltage transformer protection and self-healing method and system based on multi-source information fusion

Abstract

The invention discloses a voltage transformer protection and self-healing method and system based on multi-source information fusion. According to the method, multi-source operation data of electricity, temperature, air pressure, vibration and the like inside and outside the switch cabinet are synchronously collected through multiple types of sensors, and standardized feature vectors are formed after preprocessing and feature extraction. And the protection device analyzes the vector data in real time based on a multi-dimensional preset threshold value, and automatically triggers a protection action and cuts off the fault voltage transformer when an abnormality is detected. For a 10-35kV single-bus segmented system, after a fault segment voltage transformer is cut off, secondary voltage of the fault segment voltage transformer is switched to normal segment power supply through a switching device; and in subsequent continuous monitoring, if the operating parameters of the fault section are recovered to the safe range, the system automatically executes back switching to recover the operation of the original equipment. According to the invention, full-process automatic closed loop from fault identification, isolation and standby switching to system self-healing is realized, and the power supply continuity, the protection accuracy and the operation and maintenance intelligent level are obviously improved.

Description

Technical Field

[0001] This invention relates to the field of safety assessment for power transmission and transformation projects, and more specifically, to a method and system for the protection and self-healing of voltage transformers based on multi-source information fusion. Background Technology

[0002] Voltage transformers, as key measurement and control devices in power systems, are directly related to the safety, stability, and power quality of the power grid. Traditional protection methods mainly rely on single electrical quantity threshold judgments, which have limitations such as limited monitoring dimensions, delayed fault response, and lack of self-recovery capabilities, making them unsuitable for the comprehensive perception and rapid self-healing requirements of equipment status in high-reliability power scenarios. With the development of multi-source information fusion and intelligent decision-making technologies, constructing a protection and self-healing system that can integrate multi-dimensional information such as electrical, temperature, and mechanical data, and achieve automatic fault isolation and intelligent equipment switching, is of great significance for improving the resilience of power grid operation, preventing cascading failures, and promoting the intelligent operation and maintenance transformation of power equipment.

[0003] Existing voltage transformer protection technologies typically rely on single electrical parameters (such as voltage or current thresholds) for fault diagnosis, lacking comprehensive perception of multi-dimensional information such as temperature, air pressure, and grounding current. This incomplete information can easily lead to misjudgments or missed diagnoses. Furthermore, traditional protection devices often require manual intervention for fault diagnosis and equipment replacement after triggering, resulting in slow response times and disruptions to power supply continuity. Existing systems also lack effective self-healing mechanisms, failing to automatically switch to backup equipment and monitor recovery status after a fault occurs, leading to prolonged power outages and increased maintenance costs. Simultaneously, existing methods employ relatively simple data preprocessing and feature fusion techniques, making them ill-suited for handling interference signals under complex operating conditions, further limiting the accuracy and reliability of the protection. Summary of the Invention

[0004] The main objective of this invention is to provide a voltage transformer protection and self-healing method and system based on multi-source information fusion, to at least solve the problems of inaccurate fault diagnosis and lack of backup mechanisms in existing technologies. Achieving unmanned maintenance and rapid recovery of voltage transformers is of great significance for ensuring the safe and stable operation of high-reliability power scenarios such as smart grids and new energy integration.

[0005] To achieve the above objectives, a method and system for voltage transformer protection and self-healing based on multi-source information fusion is provided.

[0006] In a first aspect, the present invention provides a voltage transformer protection and self-healing method based on multi-source information fusion, the method comprising:

[0007] Multiple sensors are used to collect real-time multi-source data of the switch cabinet's interior and exterior. The real-time multi-source data is preprocessed to obtain standardized multi-source data, and feature vectors are calculated based on the standardized multi-source data.

[0008] The protection device receives feature vectors, presets multiple protection thresholds, and compares the feature vectors with the multiple protection thresholds to trigger protection actions.

[0009] The protection device detects the three voltage values ​​of the second voltage transformer and presets a normal voltage threshold range. If the three voltage values ​​of the second voltage transformer are within the normal voltage threshold range, the voltage switching device is activated to replace the first voltage transformer with the second voltage transformer. The first voltage transformer is the voltage transformer that is initially in operation, and the second voltage transformer is the initially idle standby voltage transformer.

[0010] The system continuously collects real-time multi-source data from both inside and outside the switchgear after the voltage switching device is activated. It then uses this real-time multi-source data to calculate new feature vectors. When all new feature vectors are within multiple protection thresholds, the system shuts down the voltage switching device to allow the second voltage transformer to exit the operating state and the first voltage transformer to re-enter the operating state.

[0011] Specifically, multiple sensors are used to collect real-time multi-source operational data from both inside and outside the switchgear, including:

[0012] The three-phase voltage and zero-sequence voltage are collected using the secondary winding of the voltage transformer; the primary current is collected using a multi-stage amplified current transformer; the temperature data of the voltage transformer is collected using a temperature sensor; the internal air pressure of the voltage transformer is collected using a pressure sensor; the grounding current and resonant current are collected using a zero-sequence current transformer; the thermal distribution data outside the switchgear is collected using an infrared detector; the vibration data outside the switchgear is collected using a vibration sensor; and the smoke data outside the switchgear is collected using a smoke sensor.

[0013] Specifically, preprocessing real-time multi-source data yields standardized multi-source data, including:

[0014] Low-pass filtering, signal isolation, and time step unification are performed on the three-phase voltage, zero-sequence voltage, and primary side current.

[0015] Perform data loss compensation, outlier removal, and data smoothing on temperature data and internal air pressure;

[0016] Common-mode suppression, small-signal amplification, and power frequency notch filtering are applied to grounding current and resonant current.

[0017] Specifically, feature vectors are calculated based on standardized multi-source data, including:

[0018] The effective values ​​of three-phase voltage, zero-sequence voltage, primary current, temperature, internal air pressure, grounding current, phase angle between resonant current and zero-sequence voltage, thermal distribution, vibration, and smoke concentration are calculated based on standardized multi-source data.

[0019] Specifically, the protection device receives a feature vector, presets multiple protection thresholds, and compares the feature vector with the multiple protection thresholds to trigger a protection action, including:

[0020] Preset overvoltage threshold, undervoltage threshold, zero-sequence overvoltage threshold, overcurrent threshold, temperature upper limit threshold, temperature change threshold, air pressure upper limit threshold, grounding current threshold, phase threshold, thermal distribution threshold, vibration threshold, and smoke concentration threshold;

[0021] When any of the three-phase voltage effective values ​​exceeds the overvoltage threshold for more than 100ms, the protection device controls the voltage transformer to take off.

[0022] When the effective value of all three-phase voltages is less than the low voltage threshold for more than 1 second, the protection device controls the voltage transformer to take off.

[0023] When the effective value of the zero-sequence voltage is greater than the zero-sequence overvoltage threshold for more than 500ms, the protection device controls the voltage transformer to take off.

[0024] When the effective value of the primary current first exceeds the overcurrent threshold and lasts for more than 200ms, the current-sensitive harmonic suppressor is activated to eliminate resonance.

[0025] When the current-sensitive harmonic suppressor has been activated and the effective value of the primary current is greater than the overcurrent threshold for more than 200ms, the protection device controls the voltage transformer to exit operation.

[0026] Specifically, the method includes receiving feature vectors through a protection device, presetting multiple protection thresholds, comparing the feature vectors with the multiple protection thresholds to trigger a protection action, and also includes:

[0027] When the effective value of the temperature data exceeds the upper temperature threshold, the protection device controls the voltage transformer to stop operating.

[0028] When the effective value of the internal air pressure exceeds the upper limit threshold, the protection device controls the voltage transformer to stop operating.

[0029] When the effective value of the grounding current is greater than the grounding current threshold for more than 300ms, the protection device controls the voltage transformer to take off.

[0030] When the phase angle is greater than the phase threshold, the protection device controls the voltage transformer to take off.

[0031] When the effective value of the thermal distribution exceeds the thermal distribution threshold, the protection device controls the voltage transformer to shut down and marks it as a suspected fire.

[0032] When the effective value of vibration exceeds the vibration threshold, the protection device controls the voltage transformer to stop operating and marks the abnormal vibration.

[0033] When the effective value of the smoke concentration is greater than the smoke concentration threshold, the protection device controls the voltage transformer to stop operating and marks the abnormal smoke.

[0034] Specifically, the protection device detects the three-phase voltage values ​​of the second voltage transformer, presets a normal voltage threshold range, and if the three-phase voltage values ​​of the second voltage transformer are within the normal voltage threshold range, the voltage switching device is activated to replace the first voltage transformer with the second voltage transformer, including:

[0035] The secondary winding of the second voltage transformer is used to collect the three-phase voltage values ​​of the second voltage transformer and the bus voltage value.

[0036] A preset normal voltage threshold range is set. When the three voltage values ​​are within the normal voltage threshold range and the bus voltage value is not zero, the voltage switching device is activated after the protection device controls the first voltage transformer to take off.

[0037] The voltage switching device controls the second voltage transformer to enter the operating mode to replace the first voltage transformer.

[0038] Specifically, real-time multi-source operational data from both inside and outside the switchgear is continuously collected after the voltage switching device is activated. New feature vectors are calculated using this real-time multi-source operational data, including:

[0039] The system continuously collects the effective values ​​of three-phase voltage, zero-sequence voltage, primary side current, temperature data, internal air pressure, grounding current, phase angle, thermal distribution, vibration, and smoke concentration after the voltage switching device is activated.

[0040] The new effective values ​​of three-phase voltage, zero-sequence voltage, bus voltage, primary current, temperature, internal air pressure, grounding current, phase angle, thermal distribution, vibration, and smoke concentration are calculated based on real-time multi-source data.

[0041] Specifically, the real-time multi-source data and multiple protection thresholds after the voltage switching device is activated are compared to ensure that all components inside the switchgear return to normal operation. After normal operation is restored, the voltage switching device is shut down to restore the voltage transformer to its normal operating state, including:

[0042] When the effective values ​​of the new three-phase voltages are all between the overvoltage threshold and the undervoltage threshold;

[0043] When the effective value of the new primary current is less than the overcurrent threshold;

[0044] When the valid value of the new temperature data is less than the upper temperature threshold;

[0045] When the effective value of the new internal air pressure is less than the upper limit threshold of the air pressure;

[0046] When the effective value of the new grounding current is less than the grounding current threshold;

[0047] When the new phase angle is less than the phase threshold;

[0048] When the effective value of the new thermal distribution is less than the thermal distribution threshold;

[0049] When the effective value of the new vibration is less than the vibration threshold;

[0050] When the effective value of the new smoke concentration is less than the smoke concentration threshold;

[0051] When the bus voltage value is not zero;

[0052] When all the above conditions are met, the protection device controls the first voltage transformer to restart and controls the second voltage transformer to shut down.

[0053] Secondly, this invention provides a voltage transformer protection and self-healing system based on multi-source information fusion. The system applies the method described in the first part, and the system includes:

[0054] The multi-source data acquisition and processing unit is used to acquire real-time multi-source data from inside and outside the switch cabinet using multiple sensors, preprocess the real-time multi-source data to obtain standardized multi-source data, and calculate feature vectors based on the standardized multi-source data.

[0055] The fault detection unit is connected to the multi-source data acquisition and processing unit. The fault detection unit is used to receive feature vectors through the protection device, preset multiple protection thresholds, and compare the feature vectors with the multiple protection thresholds to trigger protection actions.

[0056] The voltage transformer switching unit is connected to the fault detection unit. The voltage transformer switching unit is used to detect the three voltage values ​​of the second voltage transformer through the protection device. The normal voltage threshold range is preset. If the three voltage values ​​of the second voltage transformer are within the normal voltage threshold range, the voltage switching device is activated to replace the first voltage transformer with the second voltage transformer.

[0057] The voltage transformer self-healing unit is connected to the voltage transformer switching unit. The voltage transformer self-healing unit is used to continuously collect real-time multi-source operating data inside and outside the switch cabinet after the voltage switching device is activated. It uses the real-time multi-source operating data to calculate new feature vectors. When the new feature vectors are all within multiple protection thresholds, the voltage switching device is shut down so that the second voltage transformer exits the operating state and the first voltage transformer re-enters the operating state.

[0058] This application provides a voltage transformer protection and self-healing method and system based on multi-source information fusion. This method comprehensively collects multi-dimensional operational data, including electrical, temperature, and air pressure data, performs standardized processing and feature extraction, and constructs a multi-threshold collaborative intelligent fault judgment mechanism. When abnormalities such as overvoltage, overcurrent, or overheating are detected, the system automatically triggers protection actions and performs uninterrupted switching after verifying the status of backup equipment online. Subsequently, the system continuously monitors the operating environment, and automatically performs a back-off operation after the original fault conditions are eliminated and all parameters return to normal. This completes the entire closed-loop management process from fault identification, equipment isolation, backup commissioning to system self-healing without manual intervention, significantly improving the power supply reliability and intelligent operation and maintenance level of the power system. Attached Figure Description

[0059] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0060] Figure 1 A flowchart illustrating a voltage transformer protection and self-healing method based on multi-source information fusion provided for this application;

[0061] Figure 2 A connection diagram of a voltage transformer protection device based on a multi-source information fusion-based voltage transformer protection and self-healing method provided in this application;

[0062] Figure 3 A logic diagram of the voltage transformer protection action provided in this application for a voltage transformer protection and self-healing method based on multi-source information fusion;

[0063] Figure 4 This application provides a connection diagram for a voltage transformer protection and self-healing system based on multi-source information fusion.

[0064] The above figures include the following reference numerals:

[0065] 10. Switchgear; 11. Infrared detector; 12. Vibration sensor; 13. Smoke sensor; 14. Wireless temperature transmitter; 15. High voltage sensor; 16. Busbar; 17. Current-sensitive harmonic suppressor; 18. Small busbar; 20. Voltage transformer; 21. Embedded wireless temperature sensor; 22. Current transformer; 23. Pressure sensor; 24. Zero-sequence current transformer; 25. Backup voltage transformer; 26. Switch; 30. Protection device; 40. Voltage switching device; 50. Bus tie device. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0067] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein.

[0068] In this invention, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0069] This application provides a voltage transformer protection and self-healing method and system based on multi-source information fusion. This method integrates multiple sensors to collect and fuse multi-dimensional operating data such as voltage, current, temperature, and air pressure in real time. After standardization processing and feature extraction, the data is dynamically compared with preset multiple protection thresholds to achieve accurate and rapid fault diagnosis and protection action triggering. Simultaneously, the method innovatively introduces an online verification and automatic switching mechanism for backup voltage transformers. Under the premise of ensuring the integrity of backup equipment, a voltage switching device enables hot replacement of operating equipment, ensuring power supply continuity. After the fault is cleared, the system continuously monitors key parameters and automatically performs a switchback operation once the operating environment fully returns to normal. Ultimately, this forms a complete closed-loop self-healing process covering "real-time monitoring - intelligent diagnosis - seamless switching - state recovery," greatly enhancing the reliability, resilience, and automation level of voltage transformer operation and maintenance.

[0070] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0071] Figure 1 A flowchart illustrating a voltage transformer protection and self-healing method based on multi-source information fusion provided in this application is shown below. Figure 1 As shown in this embodiment, a voltage transformer protection and self-healing method based on multi-source information fusion is provided. The method includes:

[0072] Multiple sensors are used to collect real-time multi-source data of the switch cabinet's interior and exterior. The real-time multi-source data is preprocessed to obtain standardized multi-source data, and feature vectors are calculated based on the standardized multi-source data.

[0073] The protection device receives feature vectors, presets multiple protection thresholds, and compares the feature vectors with the multiple protection thresholds to trigger protection actions.

[0074] The protection device detects the three voltage values ​​of the second voltage transformer and presets a normal voltage threshold range. If the three voltage values ​​of the second voltage transformer are within the normal voltage threshold range, the voltage switching device is activated to replace the first voltage transformer with the second voltage transformer. The first voltage transformer is the voltage transformer that is initially in operation, and the second voltage transformer is the initially idle standby voltage transformer.

[0075] The system continuously collects real-time multi-source data from both inside and outside the switchgear after the voltage switching device is activated. It then uses this real-time multi-source data to calculate new feature vectors. When all new feature vectors are within multiple protection thresholds, the system shuts down the voltage switching device to allow the second voltage transformer to exit the operating state and the first voltage transformer to re-enter the operating state.

[0076] This application provides a voltage transformer protection and self-healing method based on multi-source information fusion. First, multiple sensors are integrated to collect real-time multi-source data such as electrical, temperature, and air pressure data within the switchgear. After preprocessing and feature extraction, standardized feature vectors are formed. Second, the protection device analyzes the feature vectors in real-time according to preset multi-dimensional thresholds and automatically triggers protection actions when an anomaly is detected. Subsequently, the system verifies the status of the backup voltage transformer online. If its voltage value is within the normal range, a switching device enables uninterrupted replacement of the faulty device with the backup device. Finally, the system continuously monitors the operating data after replacement, and automatically performs a reverting operation once all parameters return to normal, thus forming a complete intelligent closed loop of "monitoring-protection-switching-recovery".

[0077] This method significantly improves the comprehensiveness and accuracy of fault diagnosis through multi-source information fusion and multi-dimensional threshold collaborative judgment, effectively avoiding the limitations of single-parameter protection. The introduction of online verification and automatic switching mechanisms for backup equipment enables seamless power supply connection in fault conditions, significantly improving the reliability and continuity of system power supply. The fully automated closed-loop design reduces reliance on manual intervention, lowers operation and maintenance costs and response time, and demonstrates an effective combination of intelligent equipment operation and maintenance and system self-healing capabilities, providing an innovative solution for the stable operation of critical power system equipment.

[0078] Specifically, multiple sensors are used to collect real-time multi-source operational data from both inside and outside the switchgear, including:

[0079] The three-phase voltage and zero-sequence voltage are collected using the secondary winding of the voltage transformer; the primary current is collected using a multi-stage amplified current transformer; the temperature data of the voltage transformer is collected using a temperature sensor; the internal air pressure of the voltage transformer is collected using a pressure sensor; the grounding current and resonant current are collected using a zero-sequence current transformer; the thermal distribution data outside the switchgear is collected using an infrared detector; the vibration data outside the switchgear is collected using a vibration sensor; and the smoke data outside the switchgear is collected using a smoke sensor.

[0080] This application provides a voltage transformer protection and self-healing method based on multi-source information fusion. This method comprehensively utilizes various sensing devices such as the secondary winding of the voltage transformer, multi-stage amplified current transformer, temperature sensor, pressure sensor, and zero-sequence current transformer to simultaneously collect key physical quantities such as three-phase voltage, zero-sequence voltage, primary side current, body temperature, internal air pressure, grounding current, and resonant current, thereby achieving comprehensive real-time monitoring of the electrical, thermal, and mechanical status of the equipment.

[0081] Figure 2 A connection diagram of a voltage transformer protection device based on a multi-source information fusion-based voltage transformer protection and self-healing method provided in this application is shown below. Figure 2 As shown, this embodiment provides a voltage transformer protection device, which includes:

[0082] Switchgear 10, Infrared detector 11, Vibration sensor 12, Smoke sensor 13, Wireless temperature transmitter 14, High voltage sensor 15, Busbar 16, Current-sensitive harmonic suppressor 17, Small busbar 18, Voltage transformer 20, Embedded wireless temperature sensor 21, Current transformer 22, Pressure sensor 23, Zero-sequence current transformer 24, Backup voltage transformer 25, Switch 26, Protection device 30, Voltage switching device 40 and Bus tie device 50.

[0083] The switchgear 10 has multiple units. Infrared detector 11, vibration sensor 12, smoke sensor 13, wireless temperature transmitter 14, current-sensitive harmonic suppressor 17, voltage transformer 20 and backup voltage transformer 25 are installed inside the switchgear 10. Busbar 16 passes through the interior of multiple switchgear 10 units. High voltage sensor 15 is installed on busbar 16. Embedded wireless temperature sensor 21, current transformer 22, pressure sensor 23 and zero-sequence current transformer 24 are installed inside voltage transformer 20. Protection device 30 is used to control voltage switching device 40 to switch voltage transformer 20 and backup voltage transformer 25.

[0084] When an abnormal situation such as overvoltage or overcurrent occurs on bus 16, the bus tie device 50 will disconnect the abnormal bus 16 and connect the normal bus 16 to replace the abnormal bus 16.

[0085] When the secondary voltage on the small busbar 18 is abnormal, the voltage switching device 40 controls the disconnection of the abnormal small busbar 18 and connects the normal small busbar 18 to replace the abnormal small busbar 18.

[0086] When resonance is detected, the current-sensitive harmonic suppressor 17 begins its harmonic suppression operation. When the current-sensitive harmonic suppressor 17 fails to resolve the resonance, or when any of the abnormal conditions detected by the infrared detector 11, vibration sensor 12, smoke sensor 13, wireless temperature transmitter 14, high-voltage sensor 15, wireless temperature sensor 21, current transformer 22, pressure sensor 23, and zero-sequence current transformer 24 occur, the protection device 30 controls the switch 26 to open, disconnecting the voltage transformer 20 from the bus 16; simultaneously, the voltage switching device 40 controls the voltage transformer 20 to go offline, and controls the standby voltage transformer 25 to take over the operation of the voltage transformer 20.

[0087] When the resonance disappears and the data detected by the infrared detector 11, vibration sensor 12, smoke sensor 13, wireless temperature transmitter 14, high voltage sensor 15, wireless temperature sensor 21, current transformer 22, pressure sensor 23, and zero-sequence current transformer 24 are all normal, the protection device 30 controls the switch 26 to close, so that the voltage transformer 20 is reconnected to the bus 16; at the same time, the voltage switching device 40 controls the standby voltage transformer 25 to go offline and controls the voltage transformer 20 to start working again.

[0088] This method breaks through the limitations of traditional protection that relies on only a single electrical quantity. By integrating heterogeneous information from multiple sources such as electrical, temperature, and air pressure, it can identify complex potential risks, including insulation overheating, internal faults, and resonance anomalies, earlier and more comprehensively. This provides a solid data foundation for subsequent intelligent diagnosis and precise protection, and greatly improves the depth of state perception and the reliability of fault early warning.

[0089] Specifically, preprocessing real-time multi-source data yields standardized multi-source data, including:

[0090] Low-pass filtering, signal isolation, and time step unification are performed on the three-phase voltage, zero-sequence voltage, and primary side current.

[0091] Perform data loss compensation, outlier removal, and data smoothing on temperature data and internal air pressure;

[0092] Common-mode suppression, small-signal amplification, and power frequency notch filtering are applied to grounding current and resonant current.

[0093] This application provides a voltage transformer protection and self-healing method based on multi-source information fusion. A key aspect of this method is the implementation of layered and categorized refined preprocessing for the diverse real-time operational data collected. Specifically, for core electrical quantities such as three-phase voltage, zero-sequence voltage, and primary-side current, low-pass filtering, signal isolation, and time synchronization are employed to ensure their accuracy and synchronization. For state quantities such as temperature and internal air pressure, data loss compensation, outlier removal, and smoothing are performed to improve data integrity and stability. For weak or heavily interfered signals such as grounding current and resonant current, common-mode rejection, small-signal amplification, and power frequency notch filtering are applied to effectively extract characteristic information.

[0094] In the low-pass filtering, signal isolation, and time-step unification processing of three-phase voltage, zero-sequence voltage, and primary-side current, a second-order Butterworth low-pass filter is used with a cutoff frequency set to 500Hz (10 times higher than the power frequency) to prevent high-frequency noise aliasing. Signal isolation uses a precision isolation amplifier (such as ISO124) to achieve 3000Vrms electrical isolation and eliminate common-mode interference. Time-step unification unifies the time steps of three-phase voltage, zero-sequence voltage, and primary-side current. Data loss compensation, outlier removal, and data smoothing are performed on temperature data and internal air pressure. Data loss compensation uses the exponentially weighted moving average (EWMA) method to predict missing values. Outlier removal is based on the physical constraint of temperature change (maximum rate of change 5℃ / minute). Data smoothing uses a Savitzky-Golay filter (window length = 5, polynomial order = 2) for smoothing. Common-mode rejection, small-signal amplification, and power-frequency notch filtering are performed on ground current and resonant current. Common-mode rejection is achieved through a differential amplifier circuit, with a common-mode rejection ratio (CMRR) > 100dB. Small-signal amplification utilizes an instrumentation amplifier (such as AD620) to achieve a dynamic range of 0.1mA-10A. Power frequency notch filtering eliminates 50Hz power frequency interference using a digital notch filter while preserving the resonant frequency component.

[0095] This method employs differentiated processing strategies tailored to the physical characteristics and noise sources of different data types, transforming raw, mixed, multi-source heterogeneous data into a high-quality, comparable, and clearly defined standardized data stream. This not only lays a solid foundation for subsequent accurate feature calculations and reliable threshold comparisons but also significantly improves the anti-interference capability, measurement accuracy, and decision reliability of the entire intelligent protection system from the data source.

[0096] Specifically, feature vectors are calculated based on standardized multi-source data, including:

[0097] The effective values ​​of three-phase voltage, zero-sequence voltage, primary current, temperature, internal air pressure, grounding current, phase angle between resonant current and zero-sequence voltage, thermal distribution, vibration, and smoke concentration are calculated based on standardized multi-source data.

[0098] This application provides a voltage transformer protection and self-healing method based on multi-source information fusion. This method extracts a set of key characteristic quantities that comprehensively characterize the operating state of the voltage transformer from preprocessed standardized multi-source data. Specifically, it includes calculating the RMS values ​​of the three-phase voltage, zero-sequence voltage, primary current, temperature, internal air pressure, grounding current, and the phase angle between the resonant current and the zero-sequence voltage.

[0099] The formula for calculating the effective value of three-phase voltage is:

[0100]

[0101] in, Represents phase voltage effective value , Represents phase voltage The A momentary sample value, This represents the number of sampling points within one power frequency cycle. Indicates the sampling point index.

[0102] The formula for calculating the effective value of zero-sequence voltage is:

[0103]

[0104] in, Indicates the effective value of the zero-sequence voltage. , and This represents the instantaneous sampled value of the three-phase voltage.

[0105] The formula for calculating the effective value of the primary current is:

[0106]

[0107] in, This represents the effective value of the primary current. Indicates the second side One current sample value, Indicates the overall magnification factor. , Indicates the current transformer ratio. This represents the total voltage gain of a multistage amplifier.

[0108] The formula for calculating the effective value of temperature data is:

[0109]

[0110] in, Indicates the effective value of temperature. Indicates the first Each temperature sample value, Indicates the length of the sliding window (usually) (corresponding to a 10-second window).

[0111] The formula for calculating the effective value of internal air pressure is:

[0112]

[0113] in, Indicates the effective value of the internal air pressure. Indicates the first Each air pressure sample value Indicates the length of the sliding window (usually) (corresponding to a 5-second window).

[0114] The formula for calculating the effective value of grounding current is:

[0115]

[0116] in, Indicates the effective value of the grounding current. Indicates the zero-sequence current transformer. One grounding current sampling value.

[0117] The formula for calculating the phase angle between the resonant current and the zero-sequence voltage is:

[0118]

[0119] in, Indicates the phase angle. Represents the zero-sequence voltage. A momentary sample value, Represents the resonant current. A momentary sample value, This represents the FFT frequency index corresponding to the fundamental frequency. , This indicates the fundamental frequency (50Hz). Indicates the sampling frequency (4000 Hz), with an imaginary unit. , Take the complex phase angle function.

[0120] The formula for calculating the effective value of thermal distribution is:

[0121]

[0122] in, Indicates the effective value of thermal distribution. Represents the pixel coordinates of an infrared image. Temperature value or grayscale value at that location This indicates the number of pixels in the horizontal direction of the infrared image. This indicates the number of pixels in the vertical direction of the infrared image.

[0123] The formula for calculating the effective value of vibration is:

[0124]

[0125] in, Indicates the effective value of vibration. Indicates the first One acceleration sample value, This indicates the number of vibration sampling points in the window (usually corresponding to a 0.1-1 second time window).

[0126] The formula for calculating the effective value of smoke concentration is:

[0127]

[0128] in, This indicates the effective value of smoke concentration. Indicates the first Each smoke concentration sample value, Indicates the length of the sliding window (usually) ).

[0129] This method transforms raw waveforms or sampled values ​​from different sensors, reflecting different dimensions of conditions (such as electrical, thermal, mechanical, and insulation), into a series of feature parameters with clear physical meaning, facilitating quantitative analysis and threshold comparison. This feature vector construction process realizes the transformation of multi-source heterogeneous information into structured and standardized decision-making basis, providing an efficient and unified input interface for subsequent rapid and accurate comprehensive fault judgment and intelligent protection decision-making. It serves as a key bridge connecting front-end sensing and back-end intelligent decision-making.

[0130] Specifically, the protection device receives a feature vector, presets multiple protection thresholds, and compares the feature vector with the multiple protection thresholds to trigger a protection action, including:

[0131] Preset overvoltage threshold, undervoltage threshold, zero-sequence overvoltage threshold, overcurrent threshold, temperature upper limit threshold, temperature change threshold, air pressure upper limit threshold, grounding current threshold, phase threshold, thermal distribution threshold, vibration threshold, and smoke concentration threshold;

[0132] When any of the three-phase voltage effective values ​​exceeds the overvoltage threshold for more than 100ms, the protection device controls the voltage transformer to take off.

[0133] When the effective value of all three-phase voltages is less than the low voltage threshold for more than 1 second, the protection device controls the voltage transformer to take off.

[0134] When the effective value of the zero-sequence voltage is greater than the zero-sequence overvoltage threshold for more than 500ms, the protection device controls the voltage transformer to take off.

[0135] When the effective value of the primary current first exceeds the overcurrent threshold and lasts for more than 200ms, the current-sensitive harmonic suppressor is activated to eliminate resonance.

[0136] When the current-sensitive harmonic suppressor has been activated and the effective value of the primary current is greater than the overcurrent threshold for more than 200ms, the protection device controls the voltage transformer to exit operation.

[0137] Specifically, the method includes receiving feature vectors through a protection device, presetting multiple protection thresholds, comparing the feature vectors with the multiple protection thresholds to trigger a protection action, and also includes:

[0138] When the effective value of the temperature data exceeds the upper temperature threshold, the protection device controls the voltage transformer to stop operating.

[0139] When the effective value of the internal air pressure exceeds the upper limit threshold, the protection device controls the voltage transformer to stop operating.

[0140] When the effective value of the grounding current is greater than the grounding current threshold for more than 300ms, the protection device controls the voltage transformer to take off.

[0141] When the phase angle is greater than the phase threshold, the protection device controls the voltage transformer to take off.

[0142] When the effective value of the thermal distribution exceeds the thermal distribution threshold, the protection device controls the voltage transformer to shut down and marks it as a suspected fire.

[0143] When the effective value of vibration exceeds the vibration threshold, the protection device controls the voltage transformer to stop operating and marks the abnormal vibration.

[0144] When the effective value of the smoke concentration is greater than the smoke concentration threshold, the protection device controls the voltage transformer to stop operating and marks the abnormal smoke.

[0145] This application provides a voltage transformer protection and self-healing method based on multi-source information fusion. This method achieves protection by pre-setting multi-dimensional thresholds covering electrical, thermal, mechanical, and insulation states (including over / under voltage, overcurrent, zero-sequence overvoltage, upper temperature limit, upper air pressure limit, grounding current, and resonant phase threshold), and comparing the real-time calculated feature vectors with these thresholds one by one. The system sets differentiated delay criteria (e.g., 100ms, 200ms, 300ms, 500ms, 1s), and only executes protection actions (controlling the voltage transformer to exit operation) after the abnormal characteristic quantity persists for more than the set time, thereby effectively distinguishing between transient interference and real faults.

[0146] This method constructs a multi-dimensional, hierarchical, and time-delayed intelligent protection criterion system. It breaks through the limitations of traditional single-electrical-quantity protection, significantly improving the comprehensiveness and accuracy of fault identification by integrating multiple physical quantities for comprehensive judgment; it introduces differentiated time-delay logic to enhance anti-interference capabilities and effectively prevent false tripping; simultaneously, it incorporates non-electrical quantities (such as temperature and air pressure) into the hard protection criteria, achieving early and direct protection against potential faults such as internal overheating and insulation degradation, greatly improving the reliability and early warning capabilities of the protection system.

[0147] Specifically, the protection device detects the three-phase voltage values ​​of the second voltage transformer, presets a normal voltage threshold range, and if the three-phase voltage values ​​of the second voltage transformer are within the normal voltage threshold range, the voltage switching device is activated to replace the first voltage transformer with the second voltage transformer, including:

[0148] The secondary winding of the second voltage transformer is used to collect the three-phase voltage values ​​of the second voltage transformer and the bus voltage value.

[0149] A preset normal voltage threshold range is set. When the three voltage values ​​are within the normal voltage threshold range and the bus voltage value is not zero, the voltage switching device is activated after the protection device controls the first voltage transformer to take off.

[0150] The voltage switching device controls the second voltage transformer to enter the operating mode to replace the first voltage transformer.

[0151] This application provides a voltage transformer protection and self-healing method based on multi-source information fusion. After the main voltage transformer fails, the protection device automatically verifies the health status of the standby voltage transformer online. The specific process is as follows: the three-phase voltage values ​​of the standby equipment are collected in real time through its secondary windings and compared with a preset normal voltage threshold range; only when all phase voltages are within the normal range does the system determine that the standby equipment is available, and then immediately activates the voltage switching device to put the standby voltage transformer into operation to replace the faulty equipment.

[0152] This method ensures that only backup devices in normal working condition are put into the system by directly and in real time monitoring the output voltage of backup devices. This fundamentally avoids secondary failures or switching failures caused by abnormalities in the backup devices themselves, greatly improving the reliability of the entire system's redundancy backup and the success rate of switching operations. It is a key guarantee for achieving seamless power supply and high system availability.

[0153] Specifically, real-time multi-source operational data from both inside and outside the switchgear is continuously collected after the voltage switching device is activated. New feature vectors are calculated using this real-time multi-source operational data, including:

[0154] The system continuously collects the effective values ​​of three-phase voltage, zero-sequence voltage, primary side current, temperature data, internal air pressure, grounding current, phase angle, thermal distribution, vibration, and smoke concentration after the voltage switching device is activated.

[0155] The new effective values ​​of three-phase voltage, zero-sequence voltage, bus voltage, primary current, temperature, internal air pressure, grounding current, phase angle, thermal distribution, vibration, and smoke concentration are calculated based on real-time multi-source data.

[0156] This application provides a voltage transformer protection and self-healing method based on multi-source information fusion. This method initiates a continuous follow-up monitoring process after the standby voltage transformer is put into operation. This process continues to synchronously collect all key characteristic quantities related to fault diagnosis, including the RMS values ​​of three-phase voltage, zero-sequence voltage, primary current, temperature data, internal air pressure, grounding current, and the phase angle between the resonant current and the zero-sequence voltage.

[0157] The method for calculating the new eigenvector is the same as the method for calculating the eigenvector.

[0158] This method continuously collects multi-source data consistent with the original fault criteria after equipment switching. The system can assess in real time whether the overall operating status of the switchgear has truly returned to normal, and whether the system is stable after the backup equipment is put into operation. This provides continuous and reliable real-time data support for subsequent judgments on whether the original fault has been eliminated and whether the conditions for safe switching back to the original equipment are met. It is a necessary step to ensure that the entire self-healing logic is rigorous and safe, and ultimately achieves complete system recovery.

[0159] Specifically, the real-time multi-source data and multiple protection thresholds after the voltage switching device is activated are compared to ensure that all components inside the switchgear return to normal operation. After normal operation is restored, the voltage switching device is shut down to restore the voltage transformer to its normal operating state, including:

[0160] When the effective values ​​of the new three-phase voltages are all between the overvoltage threshold and the undervoltage threshold;

[0161] When the effective value of the new primary current is less than the overcurrent threshold;

[0162] When the valid value of the new temperature data is less than the upper temperature threshold;

[0163] When the effective value of the new internal air pressure is less than the upper limit threshold of the air pressure;

[0164] When the effective value of the new grounding current is less than the grounding current threshold;

[0165] When the new phase angle is less than the phase threshold;

[0166] When the effective value of the new thermal distribution is less than the thermal distribution threshold;

[0167] When the effective value of the new vibration is less than the vibration threshold;

[0168] When the effective value of the new smoke concentration is less than the smoke concentration threshold;

[0169] When the bus voltage value is not zero;

[0170] When all the above conditions are met, the protection device controls the first voltage transformer to restart and controls the second voltage transformer to shut down.

[0171] This application provides a voltage transformer protection and self-healing method based on multi-source information fusion. This method establishes a comprehensive system status assessment standard. This standard requires that after the standby equipment is put into operation, all key characteristic parameters—including three-phase voltage, primary current, temperature, internal air pressure, grounding current, and resonant phase angle—must be continuously monitored and confirmed to have recovered to their respective preset safety threshold ranges (e.g., voltage within the normal range, current below the overcurrent threshold, etc.). Only when all these conditions are simultaneously met will the protection device automatically perform a switchback operation, that is, re-engage the original voltage transformer and deactivate the standby equipment.

[0172] Figure 3 A logic diagram illustrating the voltage transformer protection operation of a voltage transformer protection and self-healing method based on multi-source information fusion provided in this application is shown below. Figure 3 The diagram shows the logic for the protection action of a voltage transformer, which includes:

[0173] When excessive voltage, excessive current, excessive temperature, excessive pressure, busbar overvoltage, PT overcurrent, PT internal overtemperature, atmospheric overpressure, infrared overtemperature, flue gas, or severe vibration are detected outside the PT cabinet, the protection trips and the voltage switching device is activated to switch the voltage transformer.

[0174] When the busbar is detected to have voltage, temperature, or pressure recovery, the protection is triggered to close, and the voltage switching device is started to restore the voltage transformer.

[0175] This method does not simply switch back based on time or a single electrical quantity, but requires that all abnormal conditions that triggered the initial protection action must be completely eliminated to ensure that the overall system operating environment has fully returned to normal. This "AND" logic judgment based on multi-dimensional verification minimizes the risk of blind switching when potential faults have not been completely eliminated, ensuring the safety, reliability, and ultimate success of the entire self-healing process.

[0176] Figure 4A schematic diagram of a voltage transformer protection and self-healing system based on multi-source information fusion is provided for this application, as shown below. Figure 4 As shown, this embodiment provides a voltage transformer protection and self-healing system based on multi-source information fusion. The system includes:

[0177] The multi-source data acquisition and processing unit is used to acquire real-time multi-source data from inside and outside the switch cabinet using multiple sensors, preprocess the real-time multi-source data to obtain standardized multi-source data, and calculate feature vectors based on the standardized multi-source data.

[0178] The fault detection unit is connected to the multi-source data acquisition and processing unit. The fault detection unit is used to receive feature vectors through the protection device, preset multiple protection thresholds, and compare the feature vectors with the multiple protection thresholds to trigger protection actions.

[0179] The voltage transformer switching unit is connected to the fault detection unit. The voltage transformer switching unit is used to detect the three voltage values ​​of the second voltage transformer through the protection device. The normal voltage threshold range is preset. If the three voltage values ​​of the second voltage transformer are within the normal voltage threshold range, the voltage switching device is activated to replace the first voltage transformer with the second voltage transformer.

[0180] The voltage transformer self-healing unit is connected to the voltage transformer switching unit. The voltage transformer self-healing unit is used to continuously collect real-time multi-source operating data inside and outside the switch cabinet after the voltage switching device is activated. It uses the real-time multi-source operating data to calculate new feature vectors. When the new feature vectors are all within multiple protection thresholds, the voltage switching device is shut down so that the second voltage transformer exits the operating state and the first voltage transformer re-enters the operating state.

[0181] This application provides a voltage transformer protection and self-healing system based on multi-source information fusion. The system consists of four functional units forming a closed-loop workflow: First, the multi-source data acquisition and processing unit is responsible for acquiring and processing various sensor data to generate feature vectors; second, the fault detection unit analyzes the feature vectors based on multi-dimensional thresholds and triggers protection actions; subsequently, the voltage transformer switching unit verifies the status of the standby equipment online and performs automatic switching; finally, the voltage transformer self-healing unit continuously monitors system parameters and automatically performs a back-switching operation after confirming that the system has fully recovered to normal, thereby restoring the original equipment operation.

[0182] This system, through modular and streamlined hardware and functional design, transforms complex intelligent protection and self-healing logic into a stable and reliable engineering implementation. Each unit has a clear division of labor and works collaboratively to sequentially complete the entire process of status perception, fault diagnosis, seamless switching, and system recovery, forming a highly automated closed loop from data acquisition to final execution. This systematic design not only ensures the high reliability and implementability of the entire solution but also enhances the system's maintainability and scalability through clear unit division, providing a robust system-level solution for realizing intelligent operation and maintenance of voltage transformers and proactive self-healing of the power grid.

Claims

1. A voltage transformer protection and self-healing method based on multi-source information fusion, characterized in that, The protection and self-healing methods include: Multiple sensors are used to collect real-time multi-source data of the switch cabinet's interior and exterior. The real-time multi-source data is preprocessed to obtain standardized multi-source data, and feature vectors are calculated based on the standardized multi-source data. The feature vector is received by the protection device, multiple protection thresholds are preset, and the feature vector is compared with the multiple protection thresholds to trigger a protection action. The protection device detects the three voltage values ​​of the second voltage transformer and presets a normal voltage threshold range. If the three voltage values ​​of the second voltage transformer are within the normal voltage threshold range, the voltage switching device is activated to replace the first voltage transformer with the second voltage transformer. The first voltage transformer is the voltage transformer that is initially in operation, and the second voltage transformer is the standby voltage transformer that is initially idle. The system continuously collects real-time multi-source data from the inside and outside of the switch cabinet after the voltage switching device is activated. It then calculates new feature vectors using the real-time multi-source data. When all the new feature vectors are within the range of multiple protection thresholds, the system shuts down the voltage switching device to allow the second voltage transformer to exit the operating state and the first voltage transformer to re-enter the operating state.

2. The voltage transformer protection and self-healing method based on multi-source information fusion according to claim 1, characterized in that, The method of using multiple sensors to collect real-time multi-source operational data from both inside and outside the switchgear includes: The three-phase voltage and zero-sequence voltage are collected using the secondary winding of the voltage transformer; the primary current is collected using a multi-stage amplified current transformer; the temperature data of the voltage transformer is collected using a temperature sensor; the internal air pressure of the voltage transformer is collected using a pressure sensor; the grounding current and resonant current are collected using a zero-sequence current transformer; the thermal distribution data outside the switchgear is collected using an infrared detector; the vibration data outside the switchgear is collected using a vibration sensor; and the smoke data outside the switchgear is collected using a smoke sensor.

3. The voltage transformer protection and self-healing method based on multi-source information fusion according to claim 2, characterized in that, The preprocessing of the real-time multi-source data to obtain standardized multi-source data includes: The three-phase voltage, the zero-sequence voltage, and the primary-side current are subjected to low-pass filtering, signal isolation, and time-step unification processing. Data loss compensation, outlier removal, and data smoothing are performed on the temperature data and the internal air pressure. Common-mode suppression, small-signal amplification, and power frequency notch filtering are applied to the grounding current and the resonant current.

4. The voltage transformer protection and self-healing method based on multi-source information fusion according to claim 1, characterized in that, The step of calculating the feature vector based on the standardized multi-source data includes: The effective values ​​of three-phase voltage, zero-sequence voltage, primary current, temperature data, internal air pressure, grounding current, phase angle between the resonant current and the zero-sequence voltage, thermal distribution, vibration, and smoke concentration are calculated based on the standardized multi-source data.

5. The voltage transformer protection and self-healing method based on multi-source information fusion according to claim 4, characterized in that, The process of receiving the feature vector through a protection device, presetting multiple protection thresholds, and comparing the feature vector with the multiple protection thresholds to trigger a protection action includes: Preset overvoltage threshold, undervoltage threshold, zero-sequence overvoltage threshold, overcurrent threshold, temperature upper limit threshold, temperature change threshold, air pressure upper limit threshold, grounding current threshold, phase threshold, thermal distribution threshold, vibration threshold, and smoke concentration threshold; When any one of the three-phase voltage RMS values ​​exceeds the overvoltage threshold for more than 100ms, the protection device controls the voltage transformer to stop operating. When the effective values ​​of all three phase voltages are all less than the low voltage threshold for more than 1 second, the protection device controls the voltage transformer to stop operating. When the effective value of the zero-sequence voltage is greater than the zero-sequence overvoltage threshold for more than 500ms, the protection device controls the voltage transformer to stop operating. When the effective value of the primary current is greater than the overcurrent threshold for the first time and lasts for more than 200ms, the current-sensitive harmonic suppressor is activated to eliminate resonance. When the current-sensitive harmonic suppressor has been activated and the effective value of the primary current is greater than the overcurrent threshold for more than 200ms, the protection device controls the voltage transformer to stop operating.

6. The voltage transformer protection and self-healing method based on multi-source information fusion according to claim 5, characterized in that, The step of receiving the feature vector through a protection device, presetting multiple protection thresholds, comparing the feature vector with the multiple protection thresholds to trigger a protection action further includes: When the effective value of the temperature data is greater than the upper temperature threshold, the protection device controls the voltage transformer to stop operating; When the effective value of the internal air pressure is greater than the upper limit threshold of the air pressure, the protection device controls the voltage transformer to stop operating; When the effective value of the grounding current is greater than the grounding current threshold for more than 300ms, the protection device controls the voltage transformer to stop operating. When the phase angle is greater than the phase threshold, the protection device controls the voltage transformer to stop operating; When the effective value of the thermal distribution is greater than the thermal distribution threshold, the protection device controls the voltage transformer to stop operating and marks it as a suspected fire. When the effective value of the vibration is greater than the vibration threshold, the protection device controls the voltage transformer to stop operating and marks the abnormal vibration. When the effective value of the smoke concentration is greater than the smoke concentration threshold, the protection device controls the voltage transformer to stop operating and marks the abnormal smoke.

7. The voltage transformer protection and self-healing method based on multi-source information fusion according to claim 1, characterized in that, The process involves detecting the three voltage values ​​of the second voltage transformer through the protection device, setting a preset normal voltage threshold range, and if the three voltage values ​​of the second voltage transformer are within the normal voltage threshold range, then activating the voltage switching device to replace the first voltage transformer with the second voltage transformer. This includes: The secondary winding of the second voltage transformer is used to collect the three voltage values ​​of the second voltage transformer and the bus voltage value. The normal voltage threshold range is preset. When the three voltage values ​​are within the normal voltage threshold range and the bus voltage value is not zero, the voltage switching device is activated after the protection device controls the first voltage transformer to take off. The voltage switching device controls the second voltage transformer to enter the operating mode to replace the first voltage transformer.

8. The voltage transformer protection and self-healing method based on multi-source information fusion according to claim 5, characterized in that, The continuous acquisition of real-time multi-source operational data from both inside and outside the switchgear after the voltage switching device is activated, and the calculation of new feature vectors using this real-time multi-source operational data, includes: The effective values ​​of the three-phase voltage, the zero-sequence voltage, the primary current, the temperature data, the internal air pressure, the grounding current, the phase angle, the thermal distribution, the vibration, and the smoke concentration are continuously collected after the voltage switching device is activated. Based on the real-time multi-source data, calculate the new effective values ​​of three-phase voltage, zero-sequence voltage, bus voltage, primary current, temperature, internal air pressure, grounding current, phase angle, thermal distribution, vibration, and smoke concentration.

9. A voltage transformer protection and self-healing method based on multi-source information fusion according to claims 7 and 8, characterized in that, The process of comparing real-time multi-source operational data after the voltage switching device is activated with multiple protection thresholds to ensure that all components inside the switchgear return to normal operation, and then shutting down the voltage switching device to restore the voltage transformer to its normal operating state after normal operation is restored, includes: When all the effective values ​​of the new three-phase voltages are between the overvoltage threshold and the undervoltage threshold; When the effective value of the new primary current is less than the overcurrent threshold; When the valid value of the new temperature data is less than the upper temperature threshold; When the effective value of the new internal air pressure is less than the upper limit threshold of the air pressure; When the effective value of the new grounding current is less than the grounding current threshold; When the new phase angle is less than the phase threshold; When the effective value of the new thermal distribution is less than the thermal distribution threshold; When the new effective value of vibration is less than the vibration threshold; When the effective value of the new smoke concentration is less than the smoke concentration threshold; When the bus voltage value is not zero; When all the above conditions are met, the protection device controls the first voltage transformer to restart and controls the second voltage transformer to shut down.

10. A voltage transformer protection and self-healing system based on multi-source information fusion, characterized in that, The protection and self-healing system employs the method described in any one of claims 1 to 9, wherein the protection and self-healing system comprises: A multi-source data acquisition and processing unit is used to acquire real-time operating multi-source data inside and outside the switch cabinet using multiple sensors, perform preprocessing on the real-time operating multi-source data to obtain standardized multi-source data, and calculate the feature vector based on the standardized multi-source data. A fault detection unit is connected to the multi-source data acquisition and processing unit. The fault detection unit is used to receive the feature vector through the protection device, preset multiple protection thresholds, and compare the feature vector with the multiple protection thresholds to trigger the protection action. A voltage transformer switching unit is connected to the fault detection unit. The voltage transformer switching unit is used to detect the three voltage values ​​of the second voltage transformer through the protection device, preset the normal voltage threshold range, and if the three voltage values ​​of the second voltage transformer are within the normal voltage threshold range, the voltage switching device is activated to replace the first voltage transformer with the second voltage transformer. A voltage transformer self-healing unit is connected to the voltage transformer switching unit. The voltage transformer self-healing unit is used to continuously collect real-time multi-source operating data inside and outside the switch cabinet after the voltage switching device is activated. The unit uses the real-time multi-source operating data to calculate the new feature vector. When the new feature vector is within multiple protection thresholds, the unit shuts down the voltage switching device so that the second voltage transformer exits the operating state and the first voltage transformer re-enters the operating state.

Images