A camera recognition method and system for electrical control and protection using permalloy current transformers.

By using permalloy current transformer modules for electrical control and protection, and combining visual and electrical information in a two-way manner, the problem of the separation between visual information and electrical quantities in the existing technology is solved. This enables fast and reliable protection actions and fault evidence preservation, avoiding equipment damage and malfunctions.

CN123073740APending Publication Date: 2026-11-10STATE GRID FUJIAN ELECTRIC POWER CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611161212.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-03
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In existing overhead line protection systems for power distribution networks, visual information and electrical quantity information are disconnected and cannot be linked. The CT energy harvesting system cannot be dynamically adjusted, protection actions are delayed, video evidence of fault moments is easily lost, power supply reliability is insufficient, and it is difficult to distinguish between fault current and normal load current, leading to equipment damage and maloperation or failure to operate.

Method used

The electrical control and protection are achieved by using a permalloy current transformer module. The camera intelligent recognition module analyzes the line image, and the electrical parameter analysis module monitors the electrical parameters. The electromagnetic induction of the permalloy current transformer is used to obtain electrical energy to power the camera and control module. Through magnetic core saturation characteristic detection and supercapacitor power supply strategy, the two-way fusion and dynamic adjustment of visual and electrical information are realized to ensure the reliability and speed of protection actions.

Benefits of technology

It achieves bidirectional fusion of visual information and CT energy harvesting system, improves protection response speed, reduces transformer impact damage, ensures continuity of video recording and communication during faults, and achieves protection effect of zero false tripping and zero failure to trip.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123073740A_ABST
    Figure CN123073740A_ABST
Patent Text Reader

Abstract

This invention proposes a camera-based recognition method and system for electrical control and protection using permalloy current transformers. The method employs a permalloy current transformer module to sample the target line, a camera-based intelligent recognition module to analyze the line channel image, and an electrical analysis module to analyze the electrical parameters of the target line. When abnormal electrical parameters occur in the target line, the control and protection module controls the background maintenance device module to perform operations including load reduction, tripping, and alarming, based on the type of abnormality. This invention also includes a permalloy current transformer module, a camera-based intelligent recognition module, a control and protection module, an electrical parameter analysis module, a core saturation feature detection module, a background maintenance device module, and a wireless communication module. The control and protection module collects data from the electrical parameter analysis module, the permalloy current transformer module, the core saturation feature detection module, and the camera-based intelligent recognition module and performs logical judgments.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention proposes a camera recognition method and system for electrical control and protection using permalloy current transformers, which relates to the field of intelligent electrical technology. Background Technology

[0002] In the field of overhead line protection for power distribution networks, the main shortcomings of currently widely used short-circuit fault detection and protection technologies include the following:

[0003] Visual information and electrical quantity information are completely separated, making it impossible to achieve coordinated protection.

[0004] While video surveillance devices can identify potential safety hazards in passageways in advance (such as cranes entering the safe distance from power lines), their identification results are only used for alarm purposes and do not participate in protection and control. Although the threat is visually foreseen, electrical protection systems cannot utilize this information to prepare in advance or accelerate their actions, resulting in a significant waste of information resources.

[0005] Visual information cannot be used to control the CT energy harvesting system, thus missing the opportunity to improve protective performance.

[0006] In existing solutions, the CT energy harvesting system only passively completes the functions of power supply and current detection. Its working state (such as load size and magnetic core operating point) remains unchanged and cannot be dynamically adjusted according to visually predicted external threat information. It is also impossible to use visual lead to optimize the saturation response characteristics of the CT magnetic core.

[0007] The rich information about CT energy harvesting power sources has not been explored and utilized:

[0008] Current solutions treat CT power extraction only as a means of supplying power, neglecting the rich electrical characteristics inherent in the extraction voltage itself. When abnormal disturbances occur in the line current, the ripple of the extraction voltage will be the first to respond. This potential of "power supply as sensor" is completely overlooked in existing technologies.

[0009] Delayed protection action causes impact damage to the transformer.

[0010] Traditional overcurrent protection relies on the current amplitude reaching the set value to activate, and it typically takes 10-40ms from the occurrence of a fault to the tripping of the circuit breaker. During this period, a huge short-circuit current has already flowed through equipment such as transformers, causing electrodynamic shocks and thermal damage, thus shortening the equipment's lifespan.

[0011] A single electrical criterion cannot effectively distinguish between fault current and normal load current.

[0012] Normal operations such as starting large-capacity motors and transformer inrush current can also generate a large current rise. Traditional overcurrent protection and existing di / dt identification methods are difficult to reliably distinguish between normal load changes and short-circuit faults, which can easily lead to false tripping or failure to trip.

[0013] Video evidence of a malfunction is easily lost.

[0014] When a malfunction causes a power outage in the CT scanner, the existing video surveillance system immediately shuts down the camera, making it impossible to save crucial video evidence from the moment the malfunction occurs. This severely impacts subsequent fault analysis and liability determination.

[0015] The power supply reliability of CT power harvesting cameras is insufficient in the event of a failure.

[0016] Under fault current, the output voltage of the CT may drop sharply or even be interrupted due to deep saturation of the magnetic core, causing the device's own power supply to collapse and the tripping actuator and communication module to malfunction. Summary of the Invention

[0017] In view of this, in order to fill the gaps and deficiencies in the existing technology, this invention proposes a camera recognition method and system for electrical control and protection using permalloy current transformers, in order to solve the problems of high structural cost, complex control, insufficient attitude compensation and poor end-connection reliability in existing automatic charging devices.

[0018] This invention proposes a camera recognition method and system for electrical control and protection using permalloy current transformers, comprising the following:

[0019] According to a first aspect of the present invention, the present invention proposes a camera recognition method for electrical control and protection using permalloy current transformers, characterized in that a permalloy current transformer module is used to sample the target line, a camera intelligent recognition module is used to analyze the line channel image, an electrical analysis module is used to analyze the electrical parameters of the target line, and when the electrical parameters of the target line are abnormal, the control and protection module controls the background operation and maintenance device module to perform operations on the line, including load reduction, tripping, and alarm, according to the type of abnormality.

[0020] Furthermore, the method also includes the following:

[0021] Step S1: The operation of the permalloy transformer module includes installing the permalloy transformer energy harvesting module of the permalloy transformer module on the overhead conductor of the target line, obtaining electrical energy from the conductor through electromagnetic induction, and then powering the camera intelligent recognition module, control and protection module, electrical parameter analysis module, magnetic core saturation characteristic detection module and wireless communication module after rectification, filtering and power conversion.

[0022] Step S2: The camera intelligent recognition module continuously captures images of the line channel and transmits the data to the control and protection module for real-time image analysis;

[0023] Step S3: The control and protection module actively prepares and couples the signal based on the visual prediction signal from the camera intelligent recognition module.

[0024] Step S4: Use the electrical parameter analysis module to monitor the energy harvesting voltage ripple characteristics;

[0025] Step S5: Use the electrical parameter analysis module to detect and analyze the rate of change of current;

[0026] Step S6: Continuously analyze the output voltage waveform of the secondary side of the permalloy transformer using a core saturation characteristic detection module.

[0027] Step S7: Use the control and protection module to perform four-level criterion time correlation and comprehensive logic judgment;

[0028] Step S8: Use the control and protection module to perform tripping and evidence preservation.

[0029] Further, step S2 includes the following:

[0030] In step S2, the camera intelligent recognition module uses a camera to continuously capture images of the line channel and identify whether there are any targets in the channel that could threaten the safety of the line. If a target is identified, it further determines whether the target has entered the line's safe distance range. If the target enters the safe distance range, it outputs a visual prediction signal, records the timestamp of the target's appearance, and starts the target's motion trajectory prediction to predict the time window when the target will touch the conductor in the future.

[0031] Further, step S3 includes the following:

[0032] In step S3, if the camera intelligent recognition module confirms that there is an external threat, the control and protection module actively reduces the equivalent impedance of the energy harvesting load through the permalloy current transformer module, so that the static operating point of the permalloy core is artificially moved upward and enters the edge of the shallow saturation region.

[0033] Further, step S4 includes the following:

[0034] In step S4, the electrical parameter analysis module is used to monitor the characteristics of the tapped voltage ripple, including real-time acquisition of the instantaneous value of the DC voltage after the secondary side of the transformer is rectified at a high sampling rate greater than 10kHz; calculation of the peak-to-peak value of the voltage ripple and its voltage change rate; and recording the timestamp when an abnormal drop in the tapped voltage is detected and the peak-to-peak value of the ripple increases sharply.

[0035] Further, step S5 includes the following:

[0036] In step S5, the electrical parameter analysis module continuously samples the secondary output of the permalloy transformer at a high frequency greater than 10kHz; then it calculates the instantaneous current value and its rate of change in real time; and compares the rate of change of current with a preset first threshold: if the rate of change of current is less than the first threshold, it is determined to be a normal load change and no subsequent criteria are triggered; if the rate of change of current is greater than the first threshold, it is determined to be an abnormal current rise, and a high rate of change signal is output and a timestamp is recorded.

[0037] Further, step S6 includes the following:

[0038] In step S6, the magnetic core saturation characteristic detection module continuously analyzes the output voltage waveform of the secondary side of the permalloy transformer: it monitors the peak change rate and harmonic distortion rate of the voltage waveform in real time, extracts the ratio of the third harmonic amplitude to the fundamental amplitude, and then calculates the saturation depth index; when the saturation depth index is detected to be greater than 0.7, it outputs a magnetic saturation confirmation signal and records the timestamp.

[0039] Further, step S7 includes the following:

[0040] In step S7, the control and protection module performs correlation analysis on the four signals and their timestamps from steps S2, S4, S5, and S6 to determine whether an alarm signal or a trip signal needs to be output.

[0041] Further, step S8 includes the following:

[0042] The control and protection module sends a trip signal to the circuit breaker operating mechanism through the switch output interface, and the circuit breaker operates to isolate the fault; the fault time, judgment parameters, and video before and after the fault are completely transmitted to the power distribution master station through the wireless communication module to form an unalterable and complete fault report.

[0043] According to a second aspect of the present invention, the present invention provides a camera recognition system for electrical control and protection using a permalloy current transformer, for executing a camera recognition method for electrical control and protection using a permalloy current transformer as described in any one of the present invention, characterized in that it includes the following:

[0044] The permalloy current transformer module is used to sample the target line and power the camera intelligent recognition module, control and protection module, electrical parameter analysis module, magnetic core saturation characteristic detection module and wireless communication module;

[0045] The electrical parameter analysis module is used to monitor the characteristics of the energy harvesting voltage ripple and detect and analyze the rate of change of current.

[0046] The magnetic core saturation characteristic detection module is used to continuously analyze the output voltage waveform of the secondary side of the permalloy transformer and calculate the saturation depth index.

[0047] The camera intelligent recognition module is used to analyze images of the line channel;

[0048] The supercapacitor backup power supply and power management module includes an energy storage supercapacitor bank and a charge and discharge management circuit. When the permalloy transformer module is drawing power normally, the power extraction circuit charges it. When deep core saturation is detected or the power extraction voltage of the permalloy transformer module drops sharply, the system automatically switches to supercapacitor power supply to ensure that the camera intelligent recognition module, control and protection module and communication module continue to work during the fault.

[0049] The background operation and maintenance device module includes a large-capacity energy storage capacitor and its charging and discharging management circuit specifically for circuit breaker tripping, as well as a switch output interface for outputting tripping signals and pre-charging control signals to the circuit breaker operating mechanism.

[0050] The control and protection module is used to collect data from the electrical parameter analysis module, the permalloy transformer module, the magnetic core saturation characteristic detection module, and the camera intelligent recognition module, and to make logical judgments to ultimately decide whether to output an alarm signal or a trip signal.

[0051] The present invention has the following advantages:

[0052] 1. Achieve deep bidirectional fusion of visual information and CT energy harvesting system, rather than one-way alarm.

[0053] In existing solutions, video recognition results are only used for alarms and do not participate in protection control. This application is the first to apply visual prediction signals in reverse to the CT energy harvesting system—by actively adjusting the energy harvesting load to reduce the magnetic saturation threshold, the visual information is upgraded from "only for viewing" to "both for viewing and for control," realizing bidirectional physical coupling between vision and electricity, which is a breakthrough at the principle level.

[0054] 2. Fully explore the sensing potential of CT power sources to add early warning criteria at zero cost.

[0055] Existing solutions only consider CT power extraction as a power supply method, ignoring the fault information contained in the power extraction voltage ripple. This application utilizes the voltage sampling channel of the existing power management circuit to add "abnormal power extraction ripple" as a precursor criterion at zero hardware cost, detecting fault disturbances earlier than di / dt and improving the response speed of the protection.

[0056] 3. Completely resolves the issues of power outages and video loss in the event of a CT camera malfunction.

[0057] Existing solutions immediately power down the camera when a CT scan is interrupted due to a fault, resulting in the loss of critical video evidence. This application employs a two-stage supercapacitor power supply strategy—"visual prediction triggering pre-charging + magnetic saturation confirmation triggering switching"—to ensure uninterrupted recording and communication during a fault, achieving a perfect balance between protective actions and evidence preservation.

[0058] 4. The protection action speed is significantly improved, reducing transformer impact damage.

[0059] Traditional overcurrent protection requires waiting for the current to reach the set value before it can operate (10-40ms). This application proactively lowers the magnetic saturation threshold through a second-level advance of visual prediction, allowing the permalloy current transformer (CT) to enter deep saturation several milliseconds earlier than the silicon steel CT. Combined with the sub-millisecond response of the energy harvesting ripple and the millisecond-level discrimination of di / dt, the tripping decision is completed before the fault current reaches its peak value. This significantly reduces the peak short-circuit current that the transformer can withstand and extends the equipment life.

[0060] 5. Achieve dual protection of "zero false alarms and zero refusal to operate".

[0061] Four-level criterion series confirmation: No tripping occurs when visual prediction indicates presence but electrical quantities are normal (to avoid false tripping); if visual prediction misses a detection, electrical quantity criteria can still trigger tripping (to avoid failure to trip); no tripping occurs when di / dt is high but the energy extraction ripple is normal and the magnetic core is not deeply saturated (to distinguish between normal load changes and faults). This four-level protection ensures the reliability and selectivity of the protection.

[0062] 6. The saturation criterion has been upgraded from "binarization" to "continuous quantization," resulting in more precise judgment.

[0063] Existing solutions only determine "saturation / unsaturation". This application calculates the saturation depth index SDI (a continuous value from 0 to 1), which enables graded early warning and graded actions, making the protection strategy more refined and flexible.

[0064] 7. Achieve microsecond-level time synchronization and fusion of multi-source signals

[0065] Compared to the shortcomings of existing solutions where video and electrical quantities are independent and out of time, this application achieves a time synchronization accuracy of better than 1ms by sharing a clock source, enabling visual events and electrical events to be accurately correlated, and providing a physical basis for multi-criteria fusion. Attached Figure Description

[0066] Figure 1 This is a schematic diagram of the steps of the present invention.

[0067] Figure 2 This is a schematic diagram of the overall structure of the present invention.

[0068] Figure 3 This is a schematic diagram of the overall process of the protection method of the present invention.

[0069] Figure 4 This is a schematic diagram of the four-level criterion time correlation and timing logic of the present invention.

[0070] Figure 5 This is a schematic diagram of the supercapacitor two-stage backup power supply switching circuit of the present invention. Detailed Implementation

[0071] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings.

[0072] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0073] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention; as used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise; furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0074] like Figures 1 to 5 As shown, this invention proposes a camera recognition method and system for electrical control and protection using permalloy current transformers, comprising the following:

[0075] According to a first aspect of the present invention, the present invention provides a camera identification method for electrical control and protection using a permalloy current transformer, characterized in that it includes the following:

[0076] Step S1: The operation of the permalloy transformer module includes installing the permalloy transformer energy harvesting module of the permalloy transformer module on the overhead conductor of the target line, obtaining electrical energy from the conductor through electromagnetic induction, and then powering the camera intelligent recognition module, control and protection module, electrical parameter analysis module, magnetic core saturation characteristic detection module and wireless communication module after rectification, filtering and power conversion.

[0077] Step S2: The camera intelligent recognition module continuously captures images of the line channel and transmits the data to the control and protection module for real-time image analysis;

[0078] Step S3: The control and protection module actively prepares and couples the signal based on the visual prediction signal from the camera intelligent recognition module.

[0079] Step S4: Use the electrical parameter analysis module to monitor the energy harvesting voltage ripple characteristics;

[0080] Step S5: Use the electrical parameter analysis module to detect and analyze the rate of change of current;

[0081] Step S6: Continuously analyze the output voltage waveform of the secondary side of the permalloy transformer using a core saturation characteristic detection module.

[0082] Step S7: Use the control and protection module to perform four-level criterion time correlation and comprehensive logic judgment;

[0083] Step S8: Use the control and protection module to perform tripping and evidence preservation.

[0084] Furthermore, in relevant embodiments of the present invention, each step includes the following:

[0085] Step S1: Normal operation of the device:

[0086] The permalloy CT energy harvesting module is mounted on the overhead power line of the distribution network. It obtains electrical energy from the conductor through electromagnetic induction, and after rectification, filtering, and DC-DC conversion, it powers the three high-definition cameras of the camera intelligent recognition module, the edge intelligent processing unit of the control and protection module, the control and protection module of the control and protection module, and the communication module. The three cameras continuously acquire images of the line channel at a standard frame rate (30fps), the edge intelligent unit performs routine target detection on the images, and the various electrical quantity monitoring units monitor the line status in real time.

[0087] Step S2: Visual Prediction (second-level scale):

[0088] The camera intelligent recognition module uses three high-definition cameras to continuously capture images of the line channel, and the edge intelligent processing unit performs real-time analysis of the images:

[0089] Step S21: Identify whether there are any targets in the passage that may threaten the safety of the line, such as cranes or large vehicles;

[0090] Step S22: If a target is detected, further determine whether the target has entered the safe distance range of the line;

[0091] Step S23: If the target enters the safe distance range, output the "visual prediction signal", record the target's appearance timestamp Tvision, and start the target's motion trajectory prediction to estimate the possible time window when the target touches the wire.

[0092] Step S3: Active preparation and coupling modulation based on visual prediction signals

[0093] In response to the "visual prediction signal" output in step S2, the control and protection module performs the following two proactive modulation actions in advance, before any change in electrical quantity:

[0094] Step S31: Dynamic reallocation of power supply priority:

[0095] The control and protection module immediately reduced the camera frame rate from 30fps to 5fps (only retaining key frame captures to significantly save power).

[0096] Suspend non-core communication heartbeats and redundant computing tasks;

[0097] All the saved electrical energy will be prioritized to charge the trip storage capacitor and supercapacitor backup power supply to ensure that the trip actuator has sufficient reserve energy before a fault occurs.

[0098] Step S32: Dynamic modulation of CT energy harvesting load (actively reducing magnetic saturation threshold):

[0099] The control and protection module adjusts the equivalent input impedance of the DC-DC converter after the CT power harvesting circuit through the PWM signal;

[0100] According to the CT equivalent circuit principle, the smaller the energy extraction load impedance (the greater the power extracted), the easier it is for the magnetic core to approach the saturation region.

[0101] Under the premise that visual prediction confirms that there is an external threat, the device actively adjusts the energy harvesting load (reduces the equivalent impedance), so that the static operating point of the permalloy core is artificially moved upward, approaching the edge of the shallow saturation region.

[0102] This allows the saturation inflection point, which originally required a large fault current to trigger, to be actively "pulled down." When the subsequent short-circuit current arrives, the magnetic core will instantly jump from the "shallow saturation" state to the "deep saturation" region.

[0103] Step S4: Monitoring of energy harvesting voltage ripple characteristics (precursor-level auxiliary criterion) (see...) Figure 5 (Time sequence relationship)

[0104] After the active modulation is completed in step S3, the electrical parameter analysis module immediately starts high-sensitivity ripple monitoring of the permalloy CT energy harvesting voltage:

[0105] Step S41: The electrical parameter analysis module acquires the instantaneous DC voltage value after rectification of the secondary side of the CT in real time at a high sampling rate (≥10kHz);

[0106] Step S42: The electrical parameter analysis module calculates the peak-to-peak value of the voltage ripple, Vpp, and its rate of change, dVpp / dt.

[0107] Step S43: When an abnormal drop in the energy harvesting voltage and a sharp surge in the ripple peak-to-peak value are detected (this phenomenon originates from the output voltage distortion caused by the rapid change of CT magnetic flux at the initial stage of fault current), even if di / dt has not exceeded the high current threshold at this time, the "energy harvesting abnormal warning signal" is output in advance, and the electrical parameter analysis module records the timestamp Tripple.

[0108] Step S5: Current change rate detection (millisecond scale)

[0109] The electrical parameter analysis module continuously performs high-speed sampling (≥10kHz) on the secondary side output of the permalloy CT:

[0110] Step S51: Calculate the instantaneous current value I(t) and its change rate di / dt in real time;

[0111] Step S52: Compare di / dt with the preset first threshold K1 (typical value 50A / ms):

[0112] If di / dt<K1, it is determined as normal load change, and subsequent criteria are not triggered;

[0113] If di / dt≥K1, it is determined as abnormal current rise, the "high change rate signal" is output, and the timestamp Tdi / dt is recorded.

[0114] Step S6: Permalloy CT magnetic saturation depth detection (microsecond-level confirmation) (see Figure 5 timing relationship)

[0115] The magnetic core saturation feature detection module continuously analyzes the output voltage waveform on the secondary side of the permalloy CT:

[0116] Step S61: Monitor the peak change rate and harmonic distortion rate of the voltage waveform in real time, and extract the amplitude ratio H3 / H1 of the third harmonic to the fundamental wave;

[0117] Step S62: Calculate the saturation depth index SDI (Saturation Depth Index):

[0118] SDI=α·(H3 / H1)+β·(peak clipping rate), where α and β are preset weight coefficients, and the value range of SDI is 0~1:

[0119] SDI<0.3: Unsaturated;

[0120] 0.3≤SDI<0.7: Shallow saturation (warning zone);

[0121] SDI≥0.7: Deep saturation (confirmation zone);

[0122] Step S63: Since the visual prediction in step 2b has actively moved the static operating point of the magnetic core to the edge of the shallow saturation region, once the short-circuit current starts to rise, the magnetic core will quickly cross the shallow saturation region and enter the deep saturation region.

[0123] Step S64: When deep saturation (SDI≥0.7) is detected, output a "magnetic saturation confirmation signal" and record the timestamp Tsat;

[0124] Step S65: At the instant the magnetic saturation confirmation signal is generated, the control and protection module immediately performs a power-off protection switch—switching the device power supply from the CT energy harvesting main circuit to the pre-charged supercapacitor backup power supply from step 2a, ensuring that the camera's recording and communication functions are not affected in the event of a subsequent CT voltage loss or deep saturation causing an interruption in energy harvesting (see...). Figure 5 ).

[0125] Step S7: Fourth-level criterion time correlation and comprehensive logical judgment (see...) Figure 5 )

[0126] The control and protection module performs correlation analysis on the four signals and their timestamps: step S2 (visual prediction), step S4 (abnormal energy harvesting ripple), step S5 (high di / dt change rate), and step S6 (deep core saturation), and executes the following comprehensive judgment logic:

[0127] Logic 1 (Active Protection - Level 4 Confirmation Full-Connection Trip):

[0128] Condition A: The visual prediction signal exists (Tvision has recorded it), and the trajectory prediction shows that the collision risk window has been entered;

[0129] Condition B: An abnormal energy harvesting ripple warning signal exists (Tripple has been recorded);

[0130] Condition C: A high rate of change signal exists (Tdi / dt has been recorded), and di / dt ≥ K1;

[0131] Condition D: Magnetic saturation confirmation signal exists (Tsat has been recorded), and SDI≥0.7;

[0132] Time-related conditions:

[0133] Tripple precedes Tdi / dt (ripple disturbance precedes large current change).

[0134] |Tdi / dt-Tsat|<ΔT1 (ΔT1 is the preset time window, typically 5ms);

[0135] Tvision precedes all electrical quantity signals (visual prediction at the second level comes first);

[0136] Judgment: If all four conditions are met simultaneously and the time correlation is valid, the short circuit fault is confirmed as caused by external force damage, and a trip signal is output.

[0137] Logic 2 (Backup Protection – Electrical Quantity Confirmation Trip, to Address Visual Missed Detection):

[0138] If any two or three of conditions B, C, and D are met simultaneously (condition A may not exist, possibly due to visual omissions caused by nighttime, inclement weather, or intelligent recognition failure), and the time correlation is established, it is determined to be an internal short circuit fault, and a trip signal is output to ensure no missed detections.

[0139] Logic 3 (No tripping – visual prediction exists but electrical quantities are normal):

[0140] Condition A exists (visual prediction signal), but conditions B, C, and D are not met → the target enters a safe distance but does not touch the conductor, and no effective fault current is generated → no tripping, continuous tracking and monitoring.

[0141] Logic 4 (No tripping – di / dt high but not a fault):

[0142] If condition C exists (high di / dt), but condition B does not exist (normal energy extraction ripple), and condition D does not exist (the magnetic core is not deeply saturated, SDI<0.3), and condition A does not exist, it is determined to be a normal operation such as starting a large-capacity motor or transformer inrush current, and the circuit breaker will not trip to avoid false tripping.

[0143] Step S8: Tripping Execution and Evidence Preservation (see...) Figure 5 )

[0144] Step S81: The control and protection module sends a trip signal to the circuit breaker operating mechanism through the switch output interface, and the circuit breaker operates to isolate the fault;

[0145] Step S82: Since step 5 has switched to supercapacitor power supply, even if the main CT power supply is interrupted due to a fault, the three cameras will continue to work and forcibly record 2 seconds of video evidence before and after the fault at the highest frame rate (60fps).

[0146] Step S83: The fault time, parameters of the four-level judgment criteria (visual recognition result, energy harvesting ripple curve, di / dt curve, SDI saturation depth curve), and videos before and after the fault are completely transmitted to the power distribution master station through the 4G / 5G wireless communication module to form an unalterable and complete fault report.

[0147] In addition to the above, the present invention also has related embodiments, including the following:

[0148] In one embodiment of the present invention, the permalloy current transformer module includes the following:

[0149] The permalloy current transformer module includes a permalloy magnetization characteristic model, comprising the following:

[0150] The magnetization curve of permalloy exhibits nonlinear characteristics, including magnetic saturation and hysteresis. The relationship between its magnetization M and magnetic field strength H can be expressed as:

[0151] ;

[0152] Where Ms is the saturation magnetization, Hc is the coercivity, and H0 is a characteristic parameter related to the initial permeability of the material. For 1J85 permalloy, the initial permeability μ≥80,000 and the saturation magnetic induction Bs≈0.75T.

[0153] The corresponding relationship between magnetic induction intensity B and magnetic field strength H is:

[0154] ;

[0155] Where μ0 is the vacuum permeability.

[0156] The permalloy current transformer module includes an equivalent circuit model of the CT energy harvesting circuit, comprising the following:

[0157] The equivalent circuit of the permalloy CT energy harvesting module can be modeled as follows:

[0158] ;

[0159] Where e(t) is the induced electromotive force on the secondary side, N is the number of turns in the secondary winding, Φ(t) is the magnetic flux of the magnetic core, and A is the effective cross-sectional area of ​​the magnetic core.

[0160] The relationship between the secondary output voltage u(t) and the load impedance ZL is as follows:

[0161] ;

[0162] ZCT represents the secondary impedance of CT.

[0163] The permalloy current transformer module includes the relationship between the energy harvesting voltage ripple and the magnetic flux change, including the following:

[0164] The ripple in the energy harvesting voltage originates from the instability of the rate of change of magnetic flux. When an abnormal disturbance occurs in the line current, the rate of change of magnetic flux Φ(t) changes abruptly, manifested as a sudden increase in the peak-to-peak value Vpp of the energy harvesting voltage ripple. The ripple component of the instantaneous value u(t) of the energy harvesting voltage can be expressed as:

[0165] ;

[0166] in This represents the DC component (moving average) of the energy extraction voltage. The peak-to-peak value of the ripple is defined as:

[0167] ;

[0168] Ripple change rate:

[0169] ;

[0170] When the fault current first occurs, the rapid change in magnetic flux causes the output voltage to distort, Vpp increases sharply and dVpp / dt rises sharply. This phenomenon occurs sub-milliseconds earlier than the change in current amplitude, which constitutes the physical basis of the unique "energy harvesting voltage ripple precursor criterion" of this invention.

[0171] The permalloy current transformer module includes a quantitative calculation model for the core saturation depth index (SDI), comprising the following:

[0172] The output voltage waveform of the secondary side of the CT will be distorted when the magnetic core is saturated. This is mainly manifested as peak clipping and increased harmonic content, especially a significant increase in the third harmonic component.

[0173] Let the secondary output voltage be u(t), and perform its Fourier decomposition:

[0174] ;

[0175] Where Un is the amplitude of the nth harmonic and ω is the fundamental angular frequency.

[0176] The ratio of the third harmonic amplitude to the fundamental frequency amplitude:

[0177] ;

[0178] Peak clipping rate is defined as:

[0179] ;

[0180] Where Upeak_actual is the actual peak value of the waveform, and Upeak_ideal is the theoretical peak value when unsaturated.

[0181] The Saturation Depth Index (SDI) is defined as follows:

[0182]

[0183] Where α and β are preset weighting coefficients (α+β=1), calibrated experimentally. The SDI value ranges from 0 to 1.

[0184] SDI < 0.3: Unsaturated;

[0185] 0.3≤SDI<0.7: Shallow saturation (warning zone);

[0186] SDI≥0.7: Deep saturation (confirmation zone).

[0187] The permalloy current transformer module includes the effect of dynamic modulation of the energy harvesting load on the core operating point, including the following:

[0188] The relationship between the energy-harvesting load impedance ZL and the core operating point is as follows:

[0189] ;

[0190] Where I1 is the primary current, le is the effective length of the magnetic circuit, and μ(H) is the permeability related to the magnetic field strength. f(ZL) is the modulation function of the load impedance on the core operating point. The smaller the load impedance (the greater the power drawn), the higher the core operating point Bop, and the closer it is to the saturation region.

[0191] By adjusting the equivalent input impedance of the DC-DC converter using a PWM signal, dynamic control of the core's oscillation range (ZL) can be achieved, thereby actively adjusting the core's static operating point. When the visual prediction signal is triggered, the equivalent impedance is actively reduced to shift the core's operating point upward, creating conditions for rapid entry into deep saturation.

[0192] In one embodiment of the present invention, the camera intelligent recognition module includes the following:

[0193] The camera intelligent recognition module includes camera visual prediction and edge intelligent processing algorithms.

[0194] Among them, visual target detection and recognition algorithms

[0195] Image data collected by three high-definition cameras is preprocessed and then input into a lightweight convolutional neural network model deployed on an edge intelligent processing unit for real-time analysis.

[0196] (1) Image preprocessing: Perform preprocessing operations such as size normalization (e.g., compression to 320×320 pixels), color space conversion, and grayscale stretching on the acquired RGB images to enhance image contrast and adapt to different lighting conditions.

[0197] (2) Target detection model: A lightweight target detection network (such as the improved YOLO-Nano or MobileNet-SSD) is adopted, and its core structure includes:

[0198] Backbone network: Depthwise separable convolutional layers extract multi-scale feature maps, significantly reducing the number of parameters;

[0199] Feature pyramid: It integrates shallow high-resolution features with deep semantic features to improve the detection capability of small targets;

[0200] Detection head: For each grid cell, predicts the bounding box coordinates (x,y,w,h), confidence score cc, and class probability pp.

[0201] The loss function is:

[0202] Where S is the number of grid cells, and B is the number of bounding boxes predicted for each grid cell. Indicates whether the i-th grid and j-th bounding box are responsible for predicting the target.

[0203] Among them, the target motion trajectory prediction algorithm

[0204] Upon detecting threatening targets such as cranes or large vehicles, the edge intelligence unit initiates target tracking and trajectory prediction:

[0205] Target tracking: Kalman filtering or kernel correlation filtering (KCF) algorithms are used to continuously track the target and maintain the target's state vector.

[0206] ;

[0207] Where (xt,yt) is the target center position, (vx,t,vy,t) is the velocity component, and (wt,ht) is the target size.

[0208] (2) Trajectory prediction: Based on historical trajectory data, multinomial extrapolation or Long Short-Term Memory (LSTM) networks are used to predict the future location of the target.

[0209] ;

[0210] Where F is the state transition matrix and wt is the process noise.

[0211] (3) Collision risk assessment: Collision detection is performed between the predicted trajectory and the spatial position of the conductor. When the predicted target is about to enter the safe distance range of the line,

[0212] Output the "visual prediction signal" and estimate the touch time window [Ttouch_min, Ttouch_max].

[0213] The lightweight deployment of edge intelligent inference engines

[0214] The edge intelligence processing unit employs model quantization (compressing 32-bit floating-point numbers into 8-bit integers) and pruning techniques to compress the model size to a level suitable for embedded devices. The inference engine utilizes hardware acceleration instructions from the microcontroller (such as DSP instruction sets) to optimize convolution operations, achieving an end-to-end inference latency of less than 100ms.

[0215] Among them, the power supply priority dynamic reallocation algorithm

[0216] The power supply priority dynamic reallocation algorithm in this invention draws on the idea of ​​priority-based dynamic resource scheduling in multi-agent task allocation, and performs real-time optimized allocation of the device's limited power when the visual prediction signal is triggered.

[0217] 1. Power supply task modeling

[0218] Each power-consuming module of the device is modeled as an independent power supply task set T={T1,T2,...,Tn}, and each task Ti has the following attributes:

[0219] Power consumption requirement Pi (unit: mW);

[0220] Current priority pi∈[0,1];

[0221] Power supply status si∈{0,1} (0 indicates power outage / frequency reduction, 1 indicates normal power supply);

[0222] Minimum sustaining power consumption Pi,min.

[0223] The power consumption parameters of each power module are as follows:

[0224] Table 1 Power Consumption Parameter Table

[0225]

[0226] Among them, the power supply priority dynamic reallocation algorithm

[0227] Normal operating mode (no visual prediction signal):

[0228] Each module is powered normally according to the preset priority, and the total power consumption meets the following requirements:

[0229]

[0230] The PCT_av smart label represents the total power that the CT can currently provide for energy harvesting.

[0231] Warning mode (after visual prediction signal is triggered):

[0232] The control and protection module performs power redistribution, with the objective function being:

[0233]

[0234] Constraints:

[0235]

[0236] Pi,alloc≥Pi,min (for critical tasks)

[0237] si∈{0,1},i=1,2,...,n

[0238] Where wi is the task weight (dynamically adjusted in warning mode: the trip charging has the highest weight, and the camera has the lowest weight), and Psuper_reserve is the reserve power that the supercapacitor can release.

[0239] Specific scheduling strategy:

[0240] (1) The camera frame rate was reduced from 30fps to 5fps, and the power consumption was reduced to Pcam_5≈⅙Pcam_30;

[0241] (2) Suspend non-core communication heartbeat packets (power consumption reduced to Pcomm_hb≈0.1Pcomm);

[0242] (3) Suspend non-urgent redundant computing tasks;

[0243] (4) All the electricity saved:

[0244] △P = (Pcam_30 − Pcam_5) + (Pcomm − Pcomm_hb) + Predundant; Prioritized allocation to:

[0245] Charging of tripped energy storage capacitor: allocation ratio ηtrip⋅ΔP, ηtrip=0.6;

[0246] Supercapacitor backup power charging: allocation ratio ηsuper⋅ΔP, ηsuper=0.4.

[0247] 3. Feedback mechanism with dynamic priority updates

[0248] The state of charge (SOC) of the supercapacitor and the energy storage status of the tripped capacitor are fed back to the dispatcher in real time.

[0249]

[0250]

[0251] When SOCsuper≥0.9 and Etrip≥0.95Etrip_max, the system can appropriately restore some non-core functions to achieve dynamic balance of power supply resources.

[0252] In one embodiment of the present invention, Figure 2 This is a block diagram of the overall structure of the device of the present invention, showing the connection relationships between the permalloy CT energy harvesting module, the three-channel camera module, various electrical quantity detection units (energy harvesting ripple monitoring unit, electrical parameter analysis module, magnetic core saturation characteristic detection module), the CT energy harvesting load dynamic modulation unit, the control and protection module, the supercapacitor backup power supply and trip energy storage capacitor, and the wireless communication module. Solid arrows in the diagram indicate the direction of energy / current signal flow, and dashed arrows indicate the direction of control signal flow.

[0253] In one embodiment of the present invention, Figure 3 The flowchart illustrates the overall protection process of this invention, showing the complete protection flow from step 0 to step 7. Step 2 (active preparation and coupling modulation triggered by visual prediction, including 2a power supply redistribution and 2bCT load modulation) and step 5 (supercapacitor switching triggered by magnetic saturation depth detection) are key improvements of this invention, marked with dashed boxes. Step 6 includes four comprehensive judgment logic branches (logic one to logic four), each leading to either tripping or not tripping.

[0254] In one embodiment of the present invention, Figure 4 This is a four-level criterion time correlation and timing logic diagram. The horizontal axis is the time axis, and the vertical axis is arranged with 5 signal lines (visual prediction Tvision, energy harvesting ripple warning Tripple, high di / dt change rate Tdi / dt, magnetic saturation confirmation Tsat, and comprehensive criterion output). The trigger time of each signal is marked with a solid circle, and the signal duration interval is represented by a line segment. The constraint range of time windows ΔT1 and ΔT2 is marked below, and the comprehensive logic judgment result is at the bottom.

[0255] In one embodiment of the present invention, Figure 5 The block diagram of the two-stage backup power supply switching circuit for supercapacitors shows that the main power supply (CT power) is divided into two paths (trip capacitor group and supercapacitor group) through the charging management circuit, and each is connected to the load through switches K1 and K2. The trigger logic of the two-stage switching is marked in detail in two text boxes below: the first stage is triggered by the visual prediction signal for pre-charging, and the second stage is triggered by the magnetic saturation confirmation signal for power switching.

[0256] Figure 2 In this configuration, the permalloy CT energy harvesting module is mounted on an overhead conductor. Its secondary outputs are connected to the CT energy harvesting load dynamic modulation unit, the energy harvesting voltage ripple monitoring unit, the electrical parameter analysis module, the magnetic core saturation characteristic detection module, and the power management module, respectively. The output of the CT energy harvesting load dynamic modulation unit is connected to a DC-DC converter circuit, which supplies power to each power-consuming module. The power management module is connected to the supercapacitor backup power supply and the trip energy storage capacitor, respectively, to realize charging and discharging management and power switching control. The three-camera module is connected to the edge intelligent processing unit. The edge intelligent processing unit, the energy harvesting voltage ripple monitoring unit, the electrical parameter analysis module, and the magnetic core saturation characteristic detection module are all connected to the high-precision time synchronization module to obtain a unified timestamp. The processing results of each unit are collected to the control and protection module for comprehensive judgment logic calculation. The control and protection module is connected to the circuit breaker operating mechanism through a switch output interface and to the power distribution master station through a wireless communication module.

[0257] Figure 3In the diagram, dashed box A represents step 2, "Active preparation and coupling modulation triggered by visual prediction," which includes two parallel sub-steps: 2a, "Power redistribution (frame rate 30→5fps, priority charging)" and 2b, "CT load modulation (PWM load adjustment, actively lowering the saturation threshold)." Dashed box B represents step 5, "Supercapacitor switching triggered by magnetic saturation depth detection." The comprehensive logic judgment in step 6 includes four branches: Logic 1 (four-level full confirmation → tripping), Logic 2 (electrical quantity confirmation → tripping, preventing missed detection), Logic 3 (visual prediction only → no tripping), and Logic 4 (only di / dt high → no tripping, distinguishing load fluctuations).

[0258] Figure 4 In the diagram, the horizontal axis represents time (milliseconds), and the vertical axis displays five signal trajectories: the visual prediction signal Tvision, the energy harvesting ripple warning signal Tripple, the high rate of change of di / dt signal Tdi / dt, the magnetic saturation confirmation signal Tsat, and the comprehensive criterion output. Tvision triggers earliest (second-level lead time), marked with a solid dot; Tripple triggers next (sub-millisecond level), marked with a solid dot; Tdi / dt triggers next (millisecond level), marked with a solid dot; and Tsat triggers latest (microsecond level), marked with a solid dot. ΔT1 represents the time interval from Tripple to Tdi / dt (≤5ms), and ΔT2 represents the time interval from Tdi / dt to Tsat (≤5ms). The bottom shows the trip signal output by the comprehensive criterion when all four conditions A, B, C, and D are met.

[0259] Figure 5 In the system, the main power supply is a permalloy CT energy harvesting module, which is split into two paths after passing through the charging management circuit: the upper path is connected to the trip energy storage capacitor bank via switch K1 (dedicated to trip execution), and the lower path is connected to the supercapacitor bank via switch K2 (used to maintain recording and communication during faults). The load side consists of the camera, control and protection module, and communication module. The upper right side is labeled "Level 1 (Pre-charge Trigger): Visual prediction signal Tvision → Control and protection module enables charging management circuit, charging the trip capacitor and supercapacitor simultaneously at maximum power"; the lower right side is labeled "Level 2 (Switching Trigger): Magnetic saturation confirmation signal Tsat (SDI≥0.7) → Control and protection module closes switch K2, supercapacitor bank is connected to the load, maintaining 60fps recording and communication."

[0260] According to a second aspect of the present invention, the present invention provides a camera recognition system for electrical control and protection using a permalloy current transformer, used in the camera recognition method for electrical control and protection using a permalloy current transformer as described in any one of the present invention, characterized in that it includes the following:

[0261] The permalloy current transformer module is used to sample the target line and power the camera intelligent recognition module, control and protection module, electrical parameter analysis module, magnetic core saturation characteristic detection module and wireless communication module;

[0262] The electrical parameter analysis module is used to monitor the characteristics of the energy harvesting voltage ripple and detect and analyze the rate of change of current.

[0263] The magnetic core saturation characteristic detection module is used to continuously analyze the output voltage waveform of the secondary side of the permalloy transformer and calculate the saturation depth index.

[0264] The camera intelligent recognition module is used to analyze images of the line channel;

[0265] The supercapacitor backup power supply and power management module includes an energy storage supercapacitor bank and a charge and discharge management circuit. When the permalloy transformer module is drawing power normally, the power extraction circuit charges it. When deep core saturation is detected or the power extraction voltage of the permalloy transformer module drops sharply, the system automatically switches to supercapacitor power supply to ensure that the camera intelligent recognition module, control and protection module and communication module continue to work during the fault.

[0266] The background operation and maintenance device module includes a large-capacity energy storage capacitor and its charging and discharging management circuit specifically for circuit breaker tripping, as well as a switch output interface for outputting tripping signals and pre-charging control signals to the circuit breaker operating mechanism.

[0267] The control and protection module is used to collect data from the electrical parameter analysis module, the permalloy transformer module, the magnetic core saturation characteristic detection module, and the camera intelligent recognition module, and to make logical judgments to ultimately decide whether to output an alarm signal or a trip signal.

[0268] Furthermore, in relevant embodiments of the present invention, each module includes the following:

[0269] The permalloy current transformer module includes the following:

[0270] The module consists of a cut-and-lock magnetic core and secondary windings made of permalloy (1J85 material) with high initial permeability (μi≥80,000). The saturation magnetic induction intensity of the magnetic core is Bs≈0.75T. The module is mounted on the overhead conductor of the distribution network and simultaneously performs four functions: (1) drawing power from the conductor to supply power to the entire device; (2) detecting the line current in real time; (3) outputting a voltage waveform signal reflecting the working state of the magnetic core; and (4) using the ripple characteristics of its energy extraction voltage as a signal source for auxiliary judgment of fault precursors.

[0271] The camera intelligent recognition module includes the following:

[0272] It integrates three 2-megapixel high-definition cameras—one for front view, one for rear view, and one for downward view—to monitor the direction of the conductor's movement, the rear of the conductor, and the channel directly below it, achieving panoramic monitoring of the conductor's passage.

[0273] The control and protection module includes the following:

[0274] The control and protection module includes an edge intelligent processing unit, which is used to deploy a lightweight convolutional neural network model to perform real-time analysis of three video images, identify targets that may threaten line safety such as cranes and large vehicles, predict the target's motion trajectory, and output a visual prediction signal.

[0275] The electrical parameter analysis module includes a CT power harvesting load dynamic modulation unit, a power harvesting voltage ripple monitoring unit, and a current sampling and di / dt calculation unit. The CT power harvesting load dynamic modulation unit is connected between the permalloy CT power harvesting module and the subsequent DC-DC conversion circuit. It is controlled by the main control processor and adjusts the equivalent input impedance of the power harvesting circuit through PWM signals to achieve dynamic adjustment of the static operating point of the magnetic core.

[0276] The power harvesting voltage ripple monitoring unit is connected after the CT power harvesting rectifier filter circuit. It collects the instantaneous value of DC voltage at a high sampling rate (≥10kHz), calculates the peak-to-peak value and rate of change of voltage ripple in real time, and outputs a power harvesting abnormality warning signal.

[0277] The current sampling and di / dt calculation unit is used to perform high-speed sampling (sampling rate ≥ 10kHz) of the output signal of the secondary side of the permalloy CT, and to calculate the instantaneous current value I(t) and its rate of change di / dt in real time.

[0278] The magnetic core saturation characteristic detection module is used to analyze the output voltage waveform of the secondary side of the permalloy CT in real time, extract the waveform peak change rate and harmonic distortion rate (especially focusing on the third harmonic component), and calculate the saturation depth index SDI.

[0279] The high-precision time synchronization module provides a unified clock reference for camera image acquisition, CT current sampling, energy harvesting ripple sampling, and di / dt calculation, ensuring that the timestamp accuracy of all data is better than 1ms.

[0280] The supercapacitor backup power supply and power management module includes an energy storage supercapacitor bank and a charge and discharge management circuit. When the main CT is receiving power normally, the power supply circuit charges it. When the magnetic core is detected to be deeply saturated (SDI≥0.7) or the CT power supply voltage drops suddenly, the power supply automatically switches to the supercapacitor to ensure that the camera, main control processor and communication module continue to work during the fault.

[0281] The back-end operation and maintenance device module includes a tripping energy storage capacitor and a switch output interface. The tripping energy storage capacitor and switch output interface include a large-capacity energy storage capacitor and its charging and discharging management circuit specifically for circuit breaker tripping, as well as a switch output interface that outputs tripping signals and pre-charging control signals to the circuit breaker operating mechanism.

[0282] The wireless communication module uploads early warning information, fault data, and video evidence to the power distribution master station via 4G / 5G networks.

[0283] The connection relationships between the components are as follows:

[0284] The secondary outputs of the permalloy current transformer module are connected to: the CT power harvesting load dynamic modulation unit, the power harvesting voltage ripple monitoring unit, the current sampling and di / dt calculation unit, the magnetic core saturation characteristic detection unit, and the power management module, respectively.

[0285] The output of the CT power harvesting load dynamic modulation unit of the electrical parameter analysis module is connected to the DC-DC converter circuit, which supplies power to each power module.

[0286] The power management module is connected to the supercapacitor backup power supply and the tripped energy storage capacitor respectively to realize charge and discharge management and power switching control.

[0287] The three-camera module of the camera intelligent recognition module is connected to the edge intelligent processing unit;

[0288] The edge intelligent processing unit, the energy harvesting voltage ripple monitoring unit, the current sampling and di / dt calculation unit, and the magnetic core saturation characteristic detection unit are all connected to the high-precision time synchronization module to obtain a unified timestamp.

[0289] The processing results of each unit are collected by the main control processor for comprehensive judgment logic calculation;

[0290] The control and protection module is connected to the circuit breaker operating mechanism via a digital output interface and to the power distribution master station via a wireless communication module.

[0291] The above are preferred embodiments of the present invention. Any changes made to the technical solution of the present invention that do not exceed the scope of the technical solution of the present invention shall fall within the protection scope of the present invention.

Claims

1. A camera recognition method using permalloy current transformers for electrical control and protection, characterized in that, The target line is sampled using a permalloy current transformer module, the line channel image is analyzed using a camera intelligent recognition module, and the electrical parameters of the target line are analyzed using an electrical analysis module. When the electrical parameters of the target line are abnormal, the control and protection module controls the background operation and maintenance device module to perform operations on the line, including load reduction, tripping, and alarm, according to the type of abnormality.

2. The camera recognition method for electrical control and protection using a permalloy current transformer according to claim 1, characterized in that, The method also includes the following: Step S1: The operation of the permalloy transformer module includes installing the permalloy transformer energy harvesting module of the permalloy transformer module on the overhead conductor of the target line, obtaining electrical energy from the conductor through electromagnetic induction, and then powering the camera intelligent recognition module, control and protection module, electrical parameter analysis module, magnetic core saturation characteristic detection module and wireless communication module after rectification, filtering and power conversion. Step S2: The camera intelligent recognition module continuously captures images of the line channel and transmits the data to the control and protection module for real-time image analysis; Step S3: The control and protection module actively prepares and couples the signal based on the visual prediction signal from the camera intelligent recognition module. Step S4: Use the electrical parameter analysis module to monitor the characteristics of the energy harvesting voltage ripple; Step S5: Use an electrical parameter analysis module to detect and analyze the rate of change of current; Step S6: Continuously analyze the output voltage waveform of the secondary side of the permalloy transformer using a core saturation characteristic detection module. Step S7: Use the control and protection module to perform four-level criterion time correlation and comprehensive logic judgment; Step S8: Use the control and protection module to perform tripping and evidence preservation.

3. The camera recognition method for electrical control and protection using a permalloy current transformer according to claim 2, characterized in that, Step S2 includes the following: In step S2, the camera intelligent recognition module uses a camera to continuously capture images of the line channel and identify whether there are any targets in the channel that could threaten the safety of the line. If a target is detected, further determine whether the target has entered the safe distance range of the line; If the target enters the safe distance range, output a visual prediction signal, record the timestamp of the target's appearance, and start the target's motion trajectory prediction to predict the time window when the target will touch the guide wire in the future.

4. The camera recognition method for electrical control and protection using a permalloy current transformer according to claim 3, characterized in that, Step S3 includes the following: In step S3, if the camera intelligent recognition module confirms that there is an external threat, the control and protection module actively reduces the equivalent impedance of the energy harvesting load through the permalloy current transformer module, so that the static operating point of the permalloy core is artificially moved upward and enters the edge of the shallow saturation region.

5. The camera recognition method for electrical control and protection using a permalloy current transformer according to claim 4, characterized in that, Step S4 includes the following: In step S4, the electrical parameter analysis module is used to monitor the characteristics of the tapped voltage ripple, including real-time acquisition of the instantaneous value of the DC voltage after the secondary side of the transformer is rectified at a high sampling rate greater than 10kHz; calculation of the peak-to-peak value of the voltage ripple and its voltage change rate; and recording the timestamp when an abnormal drop in the tapped voltage is detected and the peak-to-peak value of the ripple increases sharply.

6. The camera recognition method for electrical control and protection using a permalloy current transformer according to claim 5, characterized in that, Step S5 includes the following: In step S5, the electrical parameter analysis module continuously samples the secondary output of the permalloy transformer at a high frequency greater than 10kHz; then it calculates the instantaneous current value and its rate of change in real time; and compares the rate of change of current with a preset first threshold: if the rate of change of current is less than the first threshold, it is determined to be a normal load change and no subsequent criteria are triggered; if the rate of change of current is greater than the first threshold, it is determined to be an abnormal current rise, and a high rate of change signal is output and a timestamp is recorded.

7. The camera recognition method for electrical control and protection using a permalloy current transformer according to claim 6, characterized in that, Step S6 includes the following: In step S6, the magnetic core saturation characteristic detection module continuously analyzes the output voltage waveform of the secondary side of the permalloy transformer: it monitors the peak change rate and harmonic distortion rate of the voltage waveform in real time, extracts the ratio of the third harmonic amplitude to the fundamental amplitude, and then calculates the saturation depth index; when the saturation depth index is detected to be greater than 0.7, it outputs a magnetic saturation confirmation signal and records the timestamp.

8. The camera recognition method for electrical control and protection using a permalloy current transformer according to claim 7, characterized in that, Step S7 includes the following: In step S7, the control and protection module performs correlation analysis on the four signals and their timestamps from steps S2, S4, S5, and S6 to determine whether an alarm signal or a trip signal needs to be output.

9. A camera recognition method for electrical control and protection using a permalloy current transformer according to claim 8, characterized in that, Step S8 includes the following: The control and protection module sends a trip signal to the circuit breaker operating mechanism through the switch output interface, and the circuit breaker operates to isolate the fault; the fault time, judgment parameters, and video before and after the fault are completely transmitted to the power distribution master station through the wireless communication module to form an unalterable and complete fault report.

10. A camera recognition system employing a permalloy current transformer for electrical control and protection, used to execute a camera recognition method employing a permalloy current transformer for electrical control and protection as described in any one of claims 1 to 9, characterized in that, Includes the following: The permalloy current transformer module is used to sample the target line and power the camera intelligent recognition module, control and protection module, electrical parameter analysis module, magnetic core saturation characteristic detection module and wireless communication module; The electrical parameter analysis module is used to monitor the characteristics of the energy harvesting voltage ripple and detect and analyze the rate of change of current. The magnetic core saturation characteristic detection module is used to continuously analyze the output voltage waveform of the secondary side of the permalloy transformer and calculate the saturation depth index. The camera intelligent recognition module is used to analyze images of the line channel; The supercapacitor backup power supply and power management module includes an energy storage supercapacitor bank and a charge and discharge management circuit. When the permalloy transformer module is drawing power normally, the power extraction circuit charges it. When deep core saturation is detected or the power extraction voltage of the permalloy transformer module drops sharply, the system automatically switches to supercapacitor power supply to ensure that the camera intelligent recognition module, control and protection module and communication module continue to work during the fault. The background operation and maintenance device module includes a large-capacity energy storage capacitor and its charging and discharging management circuit specifically for circuit breaker tripping, as well as a switch output interface for outputting tripping signals and pre-charging control signals to the circuit breaker operating mechanism. The control and protection module is used to collect data from the electrical parameter analysis module, the permalloy transformer module, the magnetic core saturation characteristic detection module, and the camera intelligent recognition module, and to make logical judgments to ultimately decide whether to output an alarm signal or a trip signal.