A passive magnetic-based power transmission line live-line state active early warning system

Abstract

This invention belongs to the field of power equipment condition monitoring and fault early warning, and provides an active early warning system for the energized state of transmission lines based on passive magnetic sensing. The system includes a passive magnetic sensor array module, a self-powered energy management module, an edge intelligent processing and early warning central module, and a multimodal early warning and communication execution module. Through the precise coordination of these modules, the invention achieves fully automated monitoring of the energized state of transmission lines, from passive sensing, energy autonomy, and edge intelligent analysis to active early warning and networked reporting, providing strong technical support for the safe and stable operation of transmission lines.

Description

Technical Field

[0001] This invention belongs to the field of power equipment condition monitoring and fault early warning, specifically a passive magnetic transmission line energized condition active early warning system. Background Technology

[0002] With the rapid development of smart grids, the need for real-time, online, and non-contact monitoring of the operating status of high-voltage transmission lines is becoming increasingly urgent.

[0003] In the prior art, such as Chinese Patent Publication No. CN105425131A, a transmission line safety early warning system and method are disclosed, relating to the field of transmission line early warning technology. This system addresses the problem that existing detection systems are too bulky to be applied during inspection operations. The transmission line safety early warning system includes: a signal acquisition device, a signal processing device, a control device, and an alarm device. The signal processing device is signal-connected to the signal acquisition device and is used to analyze the amplitude of the signal acquired by the signal acquisition device and the proportion of the signal among all acquired signals. The control device is used to determine whether the amplitude and proportion of the signal analyzed by the signal processing device exceed amplitude thresholds and proportion thresholds, respectively. When the control device determines that the amplitude of the signal analyzed by the signal processing device exceeds the amplitude threshold and the proportion of the signal exceeds the proportion threshold, the control device controls the alarm device to sound an alarm. This transmission line safety early warning system is applied during transmission line inspections.

[0004] However, such technologies mostly rely on preset fixed thresholds and can only alarm after a fault occurs or when the limit is severely exceeded. This is a passive response and cannot achieve early warning of faults. In addition, most online monitoring devices require external power supply or regular battery replacement. In harsh environments such as the fields and mountains where power transmission lines are located, the reliability of their energy supply and maintenance costs become key bottlenecks restricting their large-scale deployment.

[0005] To address these issues, those skilled in the art have proposed an active early warning system for the energized state of transmission lines based on passive magnetism. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides an active early warning system for the energized state of transmission lines based on passive magnetism, thereby solving the problems of delayed early warning and reliance on external power supply in existing technologies.

[0007] A passive magnetic transmission line energized active early warning system includes a passive magnetic sensor array module, a self-powered energy management module, an edge intelligent processing and early warning hub module, and a multi-modal early warning and communication execution module. The passive magnetic sensor array module serves as the system's sensing terminal, used to sense the full vector information of the spatial static magnetic field and alternating magnetic field generated by the current in the transmission line with high precision in a completely non-contact and power-free manner. The self-powered energy management module is coupled to the passive magnetic sensor array module in terms of energy flow, and is used to efficiently harvest microwatt to milliwatt-level energy from the power frequency alternating magnetic field environment radiated by the conductor, and convert it into a continuous and stable main power supply for the system. The edge intelligent processing and early warning central module serves as the computing and decision-making core of the system. It is powered by the self-powered energy management module and is used to process and fuse multi-source sensor signals. Through embedded intelligent algorithms, it diagnoses the electrical and mechanical status of the line in real time and generates hierarchical early warning decisions. The multimodal early warning and communication execution module serves as the system's execution and interaction unit, used to accurately execute instructions issued by the edge intelligent processing and early warning central module.

[0008] Furthermore, the passive magnetic sensor array module includes a sensing unit, a signal conditioning unit, and a synchronous sampling and buffering unit; The sensing unit consists of at least three high-sensitivity magnetoresistive sensor chips in an orthogonal three-dimensional configuration guaranteed by mechanical structure, integrated on a rigid printed circuit board, forming a unit for accurately measuring the three components of a single-point magnetic field in space. The sensor array consists of a sensor group precisely distributed along a preset rigid baseline parallel to the direction of the conductor, forming a two-dimensional measurement array, which calculates the magnetic field gradient tensor through multi-point spatial measurement. The signal conditioning unit is electrically connected to the sensing unit. The signal conditioning unit includes a low-noise amplifier designed for microvolt-level signals, a bandpass filter circuit to suppress high-frequency and power frequency harmonic interference, and a high-resolution analog-to-digital converter, used to perform primary amplification, frequency domain filtering, and digital conversion on the original magnetic signal. The synchronous sampling and buffering unit has an embedded high-precision clock source. It sends a global sampling trigger pulse to all sensing units through a digital bus to control them to perform nanosecond-level time synchronization data acquisition. The acquired data frames are temporarily stored in a high-speed buffer memory, waiting for the main processor to read them.

[0009] Furthermore, the self-powered energy management module includes an energy capture submodule, an energy management submodule, and an energy storage submodule; The energy harvesting submodule employs a magnetoelectric composite transducer, which consists of a high-permeability ferrite core, a piezoelectric ceramic sheet bonded to the point of maximum strain in the core, and an auxiliary induction coil wound on the core. It is used to convert the alternating magnetic field energy of the conductor into electrical energy through the synergy of the magnetostrictive-piezoelectric effect and the electromagnetic induction effect. The energy management submodule includes a maximum power point tracking circuit, a rectifier bridge, a DC-DC converter, and an intelligent power management integrated circuit, which are used to efficiently rectify, regulate, and distribute the captured unstable electrical energy. The energy storage submodule consists of supercapacitors, which are used to store excess energy and provide peak power support when the magnetic field is weak, ensuring that the system can still complete at least one complete early warning reporting cycle even without an external field.

[0010] Furthermore, the edge intelligent processing and early warning central module includes a data preprocessing and feature extraction submodule, a core intelligent algorithm submodule, and an early warning decision submodule; The data preprocessing and feature extraction submodule is used to remove outliers and perform digital filtering on the input digital magnetic field signal, and to calculate the effective value of the current, the content of each harmonic, the three-phase imbalance, the magnetic field gradient, and the conductor spatial sag parameters based on the magnetic field distribution inversion in real time. The core intelligent algorithm submodule integrates a two-level analysis model, which includes a physical rule-based state filter and a machine learning-based risk predictor. The state filter based on physical rules applies a modified Biot-Savart law model, combined with conductor geometric parameters, to invert the conductor current in real time. The formula is as follows: ; in The spatial magnetic field vector measured by the passive magnetic sensor array module. Let be the vector of geometric position parameters of the wire relative to the sensor array. Represents the inversion function; The machine learning-based risk predictor uses a lightweight time-series deep learning model that has been pruned and quantized. It takes a multi-dimensional time series constructed from the feature parameters output by the previous stage as input, and learns the long-short-term dependencies and normal patterns in the sequence through its internal attention mechanism. It outputs a risk probability score between 0 and 1 to quantify the probability of a failure occurring within a specific time window in the future. The early warning decision submodule embeds a rule engine-based decision tree, which receives risk probability scores, whether each feature parameter exceeds a dynamic threshold, and auxiliary environmental data. Through weighted fusion and logical judgment, it outputs specific early warning levels and readable fault type predictions.

[0011] Furthermore, the machine learning model in the core intelligent algorithm submodule is deployed on an edge computing microcontroller that integrates a hardware neural network accelerator.

[0012] Furthermore, the multimodal early warning and communication execution module includes an early warning hierarchical response submodule and a communication submodule; The warning logic matrix embedded in the warning classification response submodule is a multi-dimensional lookup table that maps warning levels, fault types and preset actions one by one. The communication submodule integrates a dual-mode communication chip in its hardware and runs a customized protocol stack on its software. It supports automatically saving the breakpoint when the transmission is interrupted and resuming transmission after the connection is restored. At the same time, it can dynamically select the optimal communication channel and rate according to the quality of the wireless signal on site.

[0013] Furthermore, the active early warning system for the energized state of the transmission line based on passive magnetism also includes an auxiliary environmental perception and coordination module. The auxiliary environmental perception and coordination module adopts an independent waterproof package and is electrically connected to the edge intelligent processing and early warning central module through a waterproof connector and cable. The auxiliary environmental perception and coordination module includes a contact temperature measurement unit and a micro-vibration sensing unit. The contact temperature measurement unit uses a digital temperature sensor and is tightly attached to the hot spot by insulating thermally conductive silicone grease and metal clamps. The micro-vibration sensing unit uses a triaxial MEMS accelerometer to monitor the vibration spectrum of the tower at a sampling rate of not less than 1 kHz. The data from the auxiliary environmental perception and collaboration module is analyzed by the edge intelligent processing and early warning center module in collaboration with magnetic perception data through data fusion algorithms, in order to distinguish the different essences of similar appearances.

[0014] Furthermore, the active early warning system for the energized state of transmission lines based on passive magnetism adopts an adaptive low-power collaborative mechanism, and the specific working steps are as follows: Step 1: The signal conditioning unit and synchronous sampling unit in the passive magnetic sensor array module periodically wake up and sample with an extremely low duty cycle of less than 1%. Step 2: Once the sampled magnetic field data is preliminarily determined to be normal, only the data is stored in non-volatile memory, while other major modules of the system remain in deep sleep. Step 3: When the instantaneous rate of change of the magnetic field exceeds the first-level soft threshold dynamically calculated from historical data, the edge intelligent processing and early warning central module is immediately woken up via the GPIO hardware interrupt line; Step 4: The awakened edge intelligent processing and early warning central module dynamically adjusts the CPU frequency and operating voltage according to the amount of data to be analyzed and the algorithm requirements. After completing the task in the shortest possible time, it actively sends instructions to the power management chip to make the system quickly return to deep sleep state.

[0015] Furthermore, the active early warning system for the energized state of transmission lines based on passive magnetism adopts an integrated conformal packaging and heat dissipation structure, specifically as follows: All electronic modules are integrated into a housing consisting of a die-cast aluminum alloy body and an alumina ceramic radio frequency transparent window using a high-density assembly process. The housing is filled with a high thermal conductivity and high insulation silicone gel and is divided into three electromagnetic shielding isolation chambers for radio frequency, analog and digital by metal partitions. The magnetoelectric composite transducer of the self-powered energy management module has its ferrite core directly exposed at the bottom of the housing to maximize the coupling of the magnetic field of the conductor, while also being waterproof and sealed. The outer shell is precision-machined with airfoil-shaped heat dissipation fins, and the high thermal conductivity of the exposed ferrite core is used as an additional heat path to efficiently conduct the heat generated by the high-power chip inside to the entire shell for convection and radiation heat dissipation.

[0016] Furthermore, the active early warning system for the energized state of the transmission line based on passive magnetism operates as a distributed network node. After the node is powered on, it broadcasts detection messages through the communication submodule and self-organizes into a wireless mesh network based on signal strength and link quality. The network uses a distributed consensus algorithm to dynamically elect a cluster head node, which is responsible for aggregating the periodic reports and early warning information of its cluster member nodes and communicating with the remote monitoring center cloud platform.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention employs a passive magnetic sensor array module to accurately sense the full vector information of the static and alternating magnetic fields generated by the current in the transmission line in a completely non-contact and power-free manner. Combined with a self-powered energy management module, it captures and converts electrical energy from the magnetic field of the conductor, realizing the system's energy self-sufficiency and long-term maintenance-free operation. This fundamentally solves the problem of traditional online monitoring devices relying on external power supply or periodic battery replacement in harsh environments such as the field and high mountains, and significantly reduces deployment and maintenance costs.

[0018] 2. This invention integrates a physical inversion model based on the modified Biot-Savart law and a lightweight time-series deep learning model through an edge intelligent processing and early warning central module. This enables real-time inversion and feature extraction of multi-dimensional states such as line current, sag, and harmonics. It can also dynamically learn normal patterns based on historical data and perform probabilistic scoring of fault risks within a specific future time window. This upgrades the early warning mode from the traditional passive alarm after threshold exceedance to early proactive warning of abnormal conditions, significantly improving fault prevention capabilities.

[0019] 3. This invention ensures long-term reliable operation and intelligent collaboration of the system under extreme environments by adopting an adaptive low-power collaborative mechanism, an integrated conformal packaging heat dissipation structure, and distributed mesh network self-organization technology. The system sleeps with an extremely low duty cycle and only wakes up the core processor when an anomaly is detected. Combined with efficient heat dissipation design and electromagnetic shielding, the stability of electronic components is guaranteed. Multiple nodes can autonomously form a network and dynamically elect cluster heads to achieve data aggregation and communication self-healing, which greatly enhances the flexibility of system deployment and the robustness of the overall network. Attached Figure Description

[0020] Figure 1 This is a diagram of the overall system architecture of the present invention; Figure 2 This is an architectural diagram of the passive magnetic sensing array module in this invention; Figure 3 This is an architecture diagram of the self-powered energy management module in this invention; Figure 4 This is an architecture diagram of the edge intelligent processing and early warning hub module in this invention; Figure 5 This is an architecture diagram of the auxiliary environment perception and collaboration module in this invention. Detailed Implementation

[0021] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0022] As attached Figure 1 To be continued Figure 5 As shown: This invention provides an active early warning system for the energized state of transmission lines based on passive magnetism, comprising a passive magnetic sensor array module, a self-powered energy management module, an edge intelligent processing and early warning hub module, and a multimodal early warning and communication execution module; The passive magnetic sensor array module serves as the sensing terminal of the system, used to sense the full vector (magnitude and direction) information of the static magnetic field and alternating magnetic field generated by the current in the transmission line with high precision in a completely non-contact and power-free manner. The self-powered energy management module is coupled with the passive magnetic sensor array module in terms of energy flow. It is used to efficiently harvest microwatt to milliwatt-level energy from the power frequency alternating magnetic field environment radiated by the conductor and convert it into a continuous and stable main power supply for the system, thereby achieving energy autonomy. The edge intelligent processing and early warning central module serves as the computing and decision-making core of the system. Powered by the self-powered energy management module, it processes and integrates multi-source sensor signals, diagnoses the electrical and mechanical status of the line in real time through embedded intelligent algorithms, and generates hierarchical early warning decisions. The multimodal early warning and communication execution module serves as the system's execution and interaction unit. It is used to accurately execute the instructions issued by the edge intelligent processing and early warning central module, realizing a multi-level response from local intuitive alarms to remote data reporting.

[0023] As shown above, this system achieves accurate non-contact magnetic field information acquisition through a passive magnetic sensor array module, and collects energy from the magnetic field radiated by the conductor through a self-powered energy management module, providing a continuous and stable power supply for the entire system, thus realizing true energy autonomy. The edge intelligent processing and early warning central module acts as the brain, intelligently analyzing the acquired magnetic field signals and predicting fault risks. Finally, the multi-modal early warning and communication execution module performs hierarchical early warning and remote reporting. The closed-loop workflow enables the monitoring and early warning of transmission line status to achieve a fundamental transformation from external dependence and passive response to self-sufficiency and proactive early warning, providing core hardware and algorithm support for building an intelligent and reliable transmission line status monitoring network.

[0024] It should be further explained in this embodiment that the passive magnetic sensor array module includes a sensing unit, a signal conditioning unit, and a synchronous sampling and buffering unit; The sensing unit consists of at least three high-sensitivity magnetoresistive sensor chips (preferably tunnel magnetoresistive or anisotropic magnetoresistive types) in an orthogonal three-dimensional configuration guaranteed by mechanical structure, integrated on a rigid printed circuit board, forming a unit for accurately measuring the three components of a single-point magnetic field in space. The sensor array consists of multiple sensor groups precisely distributed along a preset rigid baseline parallel to the direction of the conductor, forming a one-dimensional or two-dimensional measurement array. The magnetic field gradient tensor is calculated through multi-point spatial measurement. The signal conditioning unit is electrically connected to the sensing unit. The signal conditioning unit includes a low-noise amplifier designed for microvolt-level signals, a bandpass filter circuit to suppress high-frequency and power frequency harmonic interference, and a high-resolution analog-to-digital converter, which is used to perform primary amplification, frequency domain filtering and digital conversion on the original magnetic signal. The synchronous sampling and buffering unit has an embedded high-precision clock source. It sends global sampling trigger pulses to all sensing units through a digital bus to control them to perform nanosecond-level time-synchronized data acquisition. The acquired data frames are temporarily stored in a high-speed buffer memory, waiting for the main processor to read them.

[0025] It should be further explained in this embodiment that the self-powered energy management module includes an energy capture submodule, an energy management submodule, and an energy storage submodule; The energy harvesting submodule uses a magnetoelectric composite transducer, which consists of a high-permeability ferrite core, a piezoelectric ceramic sheet attached to the point of maximum strain in the core, and an auxiliary induction coil wound on the core. It is used to convert the alternating magnetic field energy of the conductor into electrical energy through the synergy of the magnetostrictive-piezoelectric effect and the electromagnetic induction effect. The energy management submodule includes a maximum power point tracking circuit, a rectifier bridge, a DC-DC converter, and an intelligent power management integrated circuit, which are used to efficiently rectify, regulate, and distribute captured unstable electrical energy. The energy storage submodule consists of a supercapacitor and / or a low-temperature lithium iron phosphate battery, used to store excess energy and provide peak power support when the magnetic field is weak, ensuring that the system can still complete at least one complete early warning reporting cycle even without an external field.

[0026] It should be further explained in this embodiment that the edge intelligent processing and early warning center module includes a data preprocessing and feature extraction submodule, a core intelligent algorithm submodule, and an early warning decision submodule; The data preprocessing and feature extraction submodule is used to remove outliers and perform digital filtering on the input digital magnetic field signal, and to calculate the effective value of the current, the content of each harmonic, the three-phase imbalance, the magnetic field gradient, and the conductor spatial sag parameters based on the magnetic field distribution inversion in real time. The core intelligent algorithm submodule integrates a two-level analysis model, which includes a physical rule-based state filter and a machine learning-based risk predictor. Based on a physical rule-based state filter, and applying a modified Biot-Savart law model, combined with conductor geometric parameters, the conductor current is inverted in real time. The formula is as follows: ; in The spatial magnetic field vector measured by the passive magnetic sensing array module is given by an array consisting of N sensor groups. It is a vector containing 3N components, that is: ; This is a vector of geometric position parameters of the conductor relative to the sensor array. This includes, but is not limited to, the initial three-dimensional distance between the conductor and the sensor. The parameters of the catenary equation of the conductor, the tower height, the span, and other a priori or spatial structural parameters obtained through installation and calibration define the physical constraints generated by the magnetic field. This represents the inversion function or inversion process, whose physical essence is solving the inverse problem of the Biot-Savart law. The formula for the forward problem is... That is, given the current I and the geometry G, the magnetic field B can be calculated, and the inversion can be performed. Given the measured values ​​of B and the geometric parameters G, the task is to calculate the current I. In practical numerical calculations, this is usually transformed into an optimization problem: finding the optimal current value. This results in the theoretical magnetic field calculated from the current value using the forward problem formula: Compared with the measured magnetic field Minimize the error norm between them, that is: This model, by integrating an extended Kalman filter or a least-squares adaptive algorithm, can estimate and compensate online for minute sensor displacements caused by thermal expansion and contraction, structural loosening, etc. (Slow drift of parameters), thereby achieving long-term high-precision measurement; Machine learning-based risk predictors employ lightweight time-series deep learning models (such as causal convolutional networks or miniature Transformers) that have been pruned and quantized for compression. They take multidimensional time series constructed from the current, sag, and other feature parameters output by the previous stage as input. Through its internal gating or attention mechanism, it learns the long-term and short-term dependencies and normal patterns in the sequence and outputs a risk probability score between 0 and 1 to quantify the likelihood of a fault (such as overload, wire breakage, or insulation breakdown) occurring within a specific future time window. The early warning decision submodule embeds a fuzzy inference system or a decision tree based on a rule engine. It receives risk probability scores, whether each feature parameter exceeds the dynamic threshold, and auxiliary environmental data. Through weighted fusion and logical judgment, it outputs specific early warning levels (such as normal, attention, warning, and severe) and readable fault type predictions (such as "suspected joint overheating" and "abnormal increase in sag").

[0027] As shown above, the passive magnetic sensing array module, through an orthogonal three-dimensional sensor group forming a measurement array and coupled with high-precision synchronous sampling, achieves accurate calculation of the spatial magnetic field gradient tensor, providing high-quality raw data for subsequent current inversion. The self-powered energy management module uses a magnetoelectric composite transducer, synergistically utilizing the magnetostrictive-piezoelectric effect and electromagnetic induction effect, significantly improving the efficiency of energy harvesting from weak alternating magnetic fields. The edge intelligent processing and early warning center module, through a two-level analysis model of physical rule inversion and machine learning prediction, can not only accurately reconstruct the conductor's operating current, but also quantitatively assess future fault risks, forming a complete intelligent link from perception to analysis to decision-making. The collaborative work of these three core modules constitutes the technical foundation for the system's high-precision perception, efficient energy autonomy, and intelligent early warning decision-making.

[0028] It should be further explained in this embodiment that the machine learning model in the core intelligent algorithm submodule is deployed on an edge computing microcontroller that integrates a hardware neural network accelerator; the model adopts an online incremental learning strategy to dynamically update the operating data marked as normal in the current period to its training set, and periodically fine-tunes the model in the background, so that the baseline model can adapt to the seasonal changes in line load and the aging drift of equipment.

[0029] It should be further explained in this embodiment that the multimodal early warning and communication execution module includes an early warning hierarchical response submodule and a communication submodule; The early warning logic matrix embedded in the early warning classification response submodule is a multi-dimensional lookup table that maps early warning levels, fault types, and preset actions one by one. Caution level: The system only internally records event characteristics, and the tri-color LED indicator on the control housing flashes slowly green at a low frequency; Warning level: The yellow indicator light flashes frequently, and a concise event message containing time, location, risk score and key characteristic values ​​is sent to the remote monitoring platform via LoRaWAN protocol or NB-IoT network; Severe level: Triggers a constant red LED and drives a high-decibel piezoelectric buzzer to sound. At the same time, it activates a dual-channel strong guarantee communication mode, which synchronously uploads an encrypted data packet containing the complete original data frame, analysis process and diagnostic conclusion through a low-power wide area network and a backup 4G Cat.1 cellular network, and initiates a retransmission mechanism until a server confirmation receipt is received. The communication submodule integrates a dual-mode communication chip in hardware and runs a custom protocol stack in software. It supports automatic saving of breakpoints when transmission is interrupted and resuming transmission after connection is restored. At the same time, it can dynamically select the optimal communication channel and rate based on the on-site wireless signal quality (RSSI / SNR).

[0030] It should be further explained in this embodiment that the active early warning system for the energized state of transmission lines based on passive magnetism also includes an auxiliary environmental perception and coordination module. The auxiliary environmental perception and coordination module adopts an independent waterproof package and is electrically connected to the edge intelligent processing and early warning central module through a waterproof connector and cable. The auxiliary environmental perception and coordination module includes a contact temperature measurement unit and a micro-vibration sensing unit. The contact temperature measurement unit uses a platinum resistance or digital temperature sensor. It is tightly attached to the wire tension clamp, splicing tube and other hot spots that are prone to overheating by insulating thermally conductive silicone grease and metal clamps, so as to achieve temperature measurement with an accuracy of ±1℃. The micro-vibration sensing unit uses a triaxial MEMS accelerometer to monitor the vibration spectrum of the tower or conductor body at a sampling rate of not less than 1 kHz. The data from the auxiliary environmental perception and collaboration module is analyzed in collaboration with the magnetic sensing data by the edge intelligent processing and early warning center module through data fusion algorithms (such as Kalman filtering or DS evidence theory) to distinguish the different natures of similar phenomena, such as "normal increase in conductor sag due to high environmental temperature" and "abnormal sag due to mechanical damage to the conductor".

[0031] It should be further explained in this embodiment that the active early warning system for the energized state of transmission lines based on passive magnetism adopts an adaptive low-power collaborative mechanism, and the specific working steps are as follows: Step 1: The signal conditioning unit and synchronous sampling unit in the passive magnetic sensor array module periodically wake up and sample at an extremely low duty cycle of less than 1% (e.g., 10 milliseconds activated per second). Step 2: Once the sampled magnetic field data is preliminarily judged (e.g., the difference between the sampled data and the previous period's data is less than a threshold) and confirmed to be without abnormalities, the data is stored only in non-volatile memory, while other major modules of the system remain in deep sleep. Step 3: When the instantaneous rate of change of the magnetic field is detected ( When the absolute value exceeds the first-level soft threshold dynamically calculated from historical data, the edge intelligent processing and early warning central module is immediately woken up via the GPIO hardware interrupt line. Step 4: The awakened edge intelligent processing and early warning central module dynamically adjusts the CPU frequency and operating voltage according to the amount of data to be analyzed and the algorithm requirements. After completing the task in the shortest possible time, it actively sends instructions to the power management chip to make the system quickly return to deep sleep state.

[0032] It should be further explained in this embodiment that the active early warning system for the energized state of transmission lines based on passive magnetism adopts an integrated conformal packaging and heat dissipation structure, specifically as follows: All electronic modules are integrated into a housing consisting of a die-cast aluminum alloy body and an alumina ceramic radio frequency transparent window using a high-density assembly process. The housing is filled with a high thermal conductivity and high insulation silicone gel and is divided into three electromagnetic shielding isolation chambers for radio frequency, analog and digital by metal partitions. The magnetoelectric composite transducer part of the self-powered energy management module has its ferrite core directly exposed at the bottom of the housing or embedded in a non-metallic side window to maximize the coupling of the magnetic field of the conductor, while also being waterproof and sealed. The exterior of the casing is precision-machined with airfoil-shaped heat dissipation fins, and the high thermal conductivity of the exposed ferrite core is used as an additional heat path to efficiently conduct the heat generated by the high-power chip (such as the 4G communication module) inside to the entire casing for convection and radiation heat dissipation.

[0033] Further explanation of this embodiment is that the active early warning system for the energized state of transmission lines based on passive magnetism operates as a distributed network node. After multiple nodes are powered on, they broadcast detection messages through the communication submodule and self-organize into a wireless mesh network based on signal strength and link quality. The network uses a distributed consensus algorithm (such as a simplified variant of Raft) to dynamically elect a cluster head node. This cluster head node is responsible for aggregating the periodic reports and early warning information of its cluster member nodes and communicating with the remote monitoring center cloud platform in a unified manner, thereby realizing the functions of network coverage extension, data aggregation, and routing self-healing in the event of a single node failure.

[0034] As shown above, the core intelligent algorithm submodule is deployed on an edge microcontroller with a hardware neural network accelerator and supports online incremental learning, enabling the model to adapt to changes in line load and equipment aging drift, maintaining long-term early warning accuracy. The auxiliary environmental perception and collaboration module effectively distinguishes similar abnormal phenomena caused by different root causes by introducing temperature and vibration data and fusing them with magnetic data, improving the accuracy of fault diagnosis. The adaptive low-power collaboration mechanism and the integrated conformal package heat dissipation structure ensure the system's ultra-low power operation and physical reliability in harsh outdoor environments from the software scheduling and hardware design levels, respectively. These enhanced features together improve the system's practicality, intelligence, and environmental adaptability in actual complex working conditions.

[0035] Working principle: First, the passive magnetic sensor array module periodically wakes up with an extremely low duty cycle. Through its orthogonally distributed high-sensitivity magnetoresistive sensor group, it synchronously collects the full vector information of the spatial magnetic field around the transmission line. The collected microvolt-level raw magnetic signal is amplified, filtered and digitized by a signal conditioning unit consisting of a low-noise amplifier, a bandpass filter and a high-resolution analog-to-digital converter. Next, the digitized magnetic field data is sent to the edge intelligent processing and early warning central module. The data preprocessing unit of this module first removes outliers and filters the data. Then, using a physical inversion model based on the modified Biot-Savart law, combined with known conductor geometric parameters, it calculates key state variables such as conductor current and sag in real time. At the same time, the lightweight time-series deep learning model in the core intelligent algorithm submodule analyzes the time series of these state variables, learns their normal patterns, and outputs a risk probability score representing the possibility of future failures. The early warning decision submodule integrates this score, whether each feature parameter exceeds the dynamic threshold, and temperature and vibration data from the auxiliary environmental perception module, and generates specific early warning levels and failure type predictions through fuzzy reasoning or a rule engine. Meanwhile, the self-powered energy management module is always working: its magnetoelectric composite transducer continuously captures energy from the power frequency alternating magnetic field of the conductor, and after maximum power point tracking, rectification and voltage stabilization, it supplies power to each module of the system and stores excess energy in the supercapacitor, ensuring that the system can still complete at least one complete early warning reporting cycle when the magnetic field is weak or there is no magnetic field. Then, the early warning decision is sent to the multimodal early warning and communication execution module, which performs preset actions according to the early warning level: from flashing local LED indicators and sounding the buzzer, to sending encrypted early warning messages to the remote monitoring center through LoRaWAN, NB-IoT or 4G networks. The communication submodule has the ability to resume interrupted transmission and adapt to the channel, ensuring reliable information reporting. Finally, in scenarios where multiple nodes are deployed, each node, after being powered on, self-organizes into a wireless mesh network through its communication module and dynamically elects a cluster head node. The cluster head node is responsible for aggregating the data of all nodes in the area and communicating with the remote cloud platform in a unified manner, thereby achieving the coverage extension of the monitoring network and efficient data backhaul.

[0036] The embodiments of the present invention are given for the purposes of illustration and description. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Any changes, modifications, substitutions and variations made by those skilled in the art to the above embodiments within the scope of the present invention should be included within the protection scope of the present invention.

Claims

1. A passive magnetic based active warning system for live condition of power transmission line, characterized in that: It includes a passive magnetic sensor array module, a self-powered energy management module, an edge intelligent processing and early warning hub module, and a multi-modal early warning and communication execution module; The passive magnetic sensor array module serves as the system's sensing terminal, used to sense the full vector information of the spatial static magnetic field and alternating magnetic field generated by the current in the transmission line with high precision in a completely non-contact and power-free manner. The self-powered energy management module is coupled to the passive magnetic sensor array module in terms of energy flow, and is used to efficiently harvest microwatt to milliwatt-level energy from the power frequency alternating magnetic field environment radiated by the conductor, and convert it into a continuous and stable main power supply for the system. The edge intelligent processing and early warning central module serves as the computing and decision-making core of the system. It is powered by the self-powered energy management module and is used to process and fuse multi-source sensor signals. Through embedded intelligent algorithms, it diagnoses the electrical and mechanical status of the line in real time and generates hierarchical early warning decisions. The multimodal early warning and communication execution module serves as the system's execution and interaction unit, used to accurately execute instructions issued by the edge intelligent processing and early warning central module.

2. A passive magnetic based power line electrification status active early warning system as claimed in claim 1, characterized in that: The passive magnetic sensor array module includes a sensing unit, a signal conditioning unit, and a synchronous sampling and buffering unit. The sensor unit is composed of at least three high-sensitivity magnetoresistance sensor chips in a mechanically-structured orthogonal three-dimensional configuration, integrated on a rigid printed circuit board, to form a sensor group for precisely measuring three components of a magnetic field at a single point in space The sensor group is precisely distributed along a preset rigid baseline parallel to the direction of the conductor, to form a two-dimensional measurement array, and the magnetic field gradient tensor is solved through multi-point measurement in space. The signal conditioning unit is electrically connected to the sensing unit. The signal conditioning unit includes a low-noise amplifier designed for microvolt-level signals, a bandpass filter circuit to suppress high-frequency and power frequency harmonic interference, and a high-resolution analog-to-digital converter, used to perform primary amplification, frequency domain filtering, and digital conversion on the original magnetic signal. The synchronous sampling and buffering unit has an embedded high-precision clock source. It sends a global sampling trigger pulse to all sensing units through a digital bus to control them to perform nanosecond-level time synchronization data acquisition. The acquired data frames are temporarily stored in a high-speed buffer memory, waiting for the main processor to read them.

3. The active early warning system for energized transmission lines based on passive magnetism as described in claim 1, characterized in that: The self-powered energy management module includes an energy capture submodule, an energy management submodule, and an energy storage submodule; The energy harvesting submodule employs a magnetoelectric composite transducer, which consists of a high-permeability ferrite core, a piezoelectric ceramic sheet bonded to the point of maximum strain in the core, and an auxiliary induction coil wound on the core. It is used to convert the alternating magnetic field energy of the conductor into electrical energy through the synergy of the magnetostrictive-piezoelectric effect and the electromagnetic induction effect. The energy management submodule includes a maximum power point tracking circuit, a rectifier bridge, a DC-DC converter, and an intelligent power management integrated circuit, which are used to efficiently rectify, regulate, and distribute the captured unstable electrical energy. The energy storage submodule consists of supercapacitors, which are used to store excess energy and provide peak power support when the magnetic field is weak, ensuring that the system can still complete at least one complete early warning reporting cycle even without an external field.

4. The active early warning system for energized transmission lines based on passive magnetism as described in claim 1, characterized in that: The edge intelligent processing and early warning central module includes a data preprocessing and feature extraction submodule, a core intelligent algorithm submodule, and an early warning decision submodule; The data preprocessing and feature extraction submodule is used to perform threshold removal and digital filtering on the input digital magnetic field signal, and to calculate the effective value of the current, the content of each harmonic, the three-phase imbalance, the magnetic field gradient, and the conductor spatial sag parameters based on the magnetic field distribution inversion in real time. The core intelligent algorithm submodule integrates a two-level analysis model, which includes a physical rule-based state filter and a machine learning-based risk predictor. The state filter based on physical rules applies a modified Biot-Savart law model, combined with conductor geometric parameters, to invert the conductor current in real time. The formula is as follows: ； in The spatial magnetic field vector measured by the passive magnetic sensor array module. Let be the vector of geometric position parameters of the wire relative to the sensor array. Represents the inversion function; The machine learning-based risk predictor uses a lightweight time-series deep learning model that has been pruned and quantized. It takes a multi-dimensional time series constructed from the feature parameters output by the previous stage as input, and learns the long-short-term dependencies and normal patterns in the sequence through its internal attention mechanism. It outputs a risk probability score between 0 and 1 to quantify the probability of a failure occurring within a specific time window in the future. The early warning decision submodule embeds a rule engine-based decision tree, which receives risk probability scores, whether each feature parameter exceeds a dynamic threshold, and auxiliary environmental data. Through weighted fusion and logical judgment, it outputs specific early warning levels and readable fault type predictions.

5. The active early warning system for energized transmission lines based on passive magnetism as described in claim 4, characterized in that: The machine learning model in the core intelligent algorithm submodule is deployed on an edge computing microcontroller that integrates a hardware neural network accelerator.

6. The active early warning system for energized transmission lines based on passive magnetism as described in claim 1, characterized in that: The multimodal early warning and communication execution module includes an early warning hierarchical response submodule and a communication submodule; The warning logic matrix embedded in the warning classification response submodule is a multi-dimensional lookup table that maps warning levels, fault types and preset actions one by one. The communication submodule integrates a dual-mode communication chip in its hardware and runs a customized protocol stack on its software. It supports automatically saving the breakpoint when the transmission is interrupted and resuming transmission after the connection is restored. At the same time, it can dynamically select the optimal communication channel and rate according to the quality of the wireless signal on site.

7. The active early warning system for energized transmission lines based on passive magnetism as described in claim 1, characterized in that: The active early warning system for the energized state of transmission lines based on passive magnetism also includes an auxiliary environmental perception and coordination module. The auxiliary environmental perception and coordination module is independently waterproof and is electrically connected to the edge intelligent processing and early warning central module through a waterproof connector and cable. The auxiliary environmental perception and coordination module includes a contact temperature measurement unit and a micro-vibration sensing unit. The contact temperature measurement unit uses a digital temperature sensor and is tightly attached to the hot spot by insulating thermally conductive silicone grease and metal clamps. The micro-vibration sensing unit uses a triaxial MEMS accelerometer to monitor the vibration spectrum of the tower at a sampling rate of not less than 1 kHz. The data from the auxiliary environmental perception and collaboration module is analyzed by the edge intelligent processing and early warning center module in collaboration with magnetic perception data through data fusion algorithms, in order to distinguish the different essences of similar appearances.

8. The active early warning system for energized transmission lines based on passive magnetism as described in claim 1, characterized in that: The active early warning system for the energized state of transmission lines based on passive magnetism adopts an adaptive low-power collaborative mechanism, and its specific working steps are as follows: Step 1: The signal conditioning unit and synchronous sampling unit in the passive magnetic sensor array module periodically wake up and sample with an extremely low duty cycle of less than 1%. Step 2: Once the sampled magnetic field data is preliminarily determined to be normal, only the data is stored in non-volatile memory, while other major modules of the system remain in deep sleep. Step 3: When the instantaneous rate of change of the magnetic field exceeds the first-level soft threshold dynamically calculated from historical data, the edge intelligent processing and early warning central module is immediately woken up via the GPIO hardware interrupt line; Step 4: The awakened edge intelligent processing and early warning central module dynamically adjusts the CPU frequency and operating voltage according to the amount of data to be analyzed and the algorithm requirements. After completing the task in the shortest possible time, it actively sends instructions to the power management chip to make the system quickly return to deep sleep state.

9. The active early warning system for the energized state of a transmission line based on passive magnetism as described in claim 1, characterized in that: The active early warning system for the energized state of transmission lines based on passive magnetism adopts an integrated conformal packaging and heat dissipation structure, specifically: All electronic modules are integrated into a housing consisting of a die-cast aluminum alloy body and an alumina ceramic radio frequency transparent window using a high-density assembly process. The housing is filled with a high thermal conductivity and high insulation silicone gel and is divided into three electromagnetic shielding isolation chambers for radio frequency, analog and digital by metal partitions. The magnetoelectric composite transducer of the self-powered energy management module has its ferrite core directly exposed at the bottom of the housing to maximize the coupling of the magnetic field of the conductor, while also being waterproof and sealed. The outer shell is precision-machined with airfoil-shaped heat dissipation fins, and the high thermal conductivity of the exposed ferrite core is used as an additional heat path to efficiently conduct the heat generated by the high-power chip inside to the entire shell for convection and radiation heat dissipation.

10. The active early warning system for energized transmission lines based on passive magnetism as described in claim 1, characterized in that: The active early warning system for energized transmission lines based on passive magnetism operates as a distributed network node. After the node is powered on, it broadcasts detection messages through the communication submodule and self-organizes into a wireless mesh network based on signal strength and link quality. The network uses a distributed consensus algorithm to dynamically elect a cluster head node, which is responsible for aggregating the periodic reports and early warning information of its cluster members and communicating with the remote monitoring center cloud platform.

Images