Power grid line monitoring device and monitoring method based on sensing and computing intelligent SOC chip

Abstract

The invention discloses a power grid line monitoring device based on an SOC chip. The power grid line monitoring device comprises the SOC chip, a power supply unit, a current sampling circuit, a voltage sampling circuit, an HRF radio frequency front end peripheral circuit, a temperature and humidity sensor and a Bluetooth communication unit. The SOC chip is connected with the power supply unit, the current sampling circuit, the voltage sampling circuit, the HRF front-end peripheral circuit, the temperature and humidity sensor and the Bluetooth communication unit. The invention further discloses a monitoring method of the power grid line monitoring device based on the sensing and computing intelligent SOC chip. According to the invention, the technical problems of multi-parameter data asynchronization, low fault judgment accuracy and lack of intelligent prediction capability caused by limited MCU computing power and module discrete architecture of the existing power grid line monitoring device are solved.

Description

Technical Field

[0001] This invention relates to the field of smart grid line fault monitoring and location technology, and in particular to a power grid line monitoring device and monitoring method based on a sensing, computing, communication, and positioning intelligent SOC chip. Background Technology

[0002] Smart grid distribution network line faults exhibit characteristics of "multiple causes and complex scenarios." High-altitude areas experience frequent lightning strikes due to decreased air insulation strength, humid environments easily trigger intermittent grounding, and aging cables with hidden faults require early warning. Existing technologies primarily rely on an MCU combined with a discrete sensor architecture, which has the following limitations: First, using an MCU + discrete sensors, with the MCU performing threshold discrimination and external ADCs and temperature / humidity sensors connected via SPI or I / O... 2 The C-interface communication suffers from several drawbacks: high data latency, poor synchronization of multiple parameters, limited MCU computing power (resulting in an inability to run deep learning models), high false positive rate, and lack of predictive capabilities. Secondly, a patent application (CN119414292A) discloses a method and system for identifying faults in power distribution lines, which relies on a single-parameter MCU for fault identification and uses a zero-sequence current threshold. This method fails to distinguish between low-current grounding and lightning strike faults, resulting in a high false positive rate. It also lacks an NPU and multi-parameter processing capabilities, predictive functionality, and adaptability. Thirdly, while data is collected at the terminal side and uploaded to the cloud for analysis, it cannot function when communication is interrupted, and the terminal side lacks NPU preprocessing capabilities, leading to insufficient real-time performance. All of these solutions, due to their discrete hardware, asynchronous data, and computing power bottlenecks, cannot meet the demands of smart grids for multi-parameter fusion, real-time simulation, and scenario generalization. Therefore, there is an urgent need to propose a power grid line monitoring device and method based on a sensing-computing-connection-position-intelligent SOC chip to address the technical problems of asynchronous multi-parameter data, low fault identification accuracy, and lack of intelligent predictive capabilities caused by the limited computing power of MCUs and the discrete module architecture of existing power grid line monitoring devices. Summary of the Invention

[0003] The main objective of this invention is to propose a power grid line monitoring device and its monitoring method based on a sensing and computing intelligent SOC chip, aiming to solve the technical problems of existing power grid line monitoring devices, such as asynchronous multi-parameter data, low fault identification accuracy, and lack of intelligent prediction capabilities, caused by the limited computing power of MCUs and the discrete module architecture.

[0004] To achieve the above objectives, the present invention provides a power grid line monitoring device based on a sensing-computing-communication-positioning intelligent SOC chip, wherein the power grid line monitoring device based on the sensing-computing-communication-positioning intelligent SOC chip includes: an SOC chip, a power supply unit, a current sampling circuit, a voltage sampling circuit, an HRF radio frequency front-end peripheral circuit, a temperature and humidity sensor, and a Bluetooth communication unit; the SOC chip is connected to the power supply unit, the current sampling circuit, the voltage sampling circuit, the HRF radio frequency front-end peripheral circuit, the temperature and humidity sensor, and the Bluetooth communication unit respectively;

[0005] The power supply unit is used to draw power from the power distribution line and to supply power to the device;

[0006] The current sampling unit is used to convert the current of the phase line into a first voltage signal that the SOC chip can recognize, and to connect the first voltage signal to the ADC input interface of the SOC chip.

[0007] The voltage sampling unit is used to convert the phase line electric field into a second voltage signal that the SOC chip can recognize, and to connect the second voltage signal to the ADC input interface of the SOC chip.

[0008] The HRF radio frequency front-end peripheral circuit is used to switch the radio frequency transmission / reception mode;

[0009] The temperature and humidity sensor is used to collect the temperature and humidity of the surrounding environment of the device, and convert them into a third voltage signal and a fourth voltage signal that can be recognized by the SOC chip, and then connect the third voltage signal and the fourth voltage signal to the ADC input interface of the SOC chip.

[0010] In one preferred embodiment, the power module has a built-in rechargeable backup power supply;

[0011] Under normal conditions, the power module is used to convert the high-voltage AC power of the power distribution line into low-voltage DC power to supply power to the device and backup power supply.

[0012] In the event of a power outage, the backup power supply of the power module provides power to the device.

[0013] In one preferred embodiment, the SOC chip includes a processing core, a storage unit, a power management unit, a clock unit, a DMA controller, and an NPU; the processing core is connected to the storage unit, the power management unit, the clock unit, the DMA controller, and the NPU via a system bus.

[0014] In one preferred embodiment, the SOC chip includes a wide-range ADC circuit unit with at least four channels, the wide-range ADC circuit unit being connected to a current sampling circuit, a voltage sampling circuit, and a temperature and humidity sensor, respectively.

[0015] In one preferred embodiment, the power grid line monitoring device based on the sensing, computing, communication, and positioning intelligent SOC chip further includes a first frequency selective filtering circuit; the first frequency selective filtering circuit is connected to the SOC chip and is used to filter the BeiDou satellite signals received by the antenna.

[0016] In one preferred embodiment, the SOC chip includes a BeiDou baseband processing unit, which is connected to a first frequency selective filtering circuit.

[0017] In one preferred embodiment, the SOC chip includes an HRF baseband processing unit, which is connected to the HRF radio frequency front-end peripheral circuit.

[0018] In one preferred embodiment, the HRF radio frequency front-end peripheral circuit includes a transmit / receive switching circuit, a second frequency selective filtering circuit, a power amplifier circuit, and an antenna; the antenna is connected to the second frequency selective filtering circuit and the power amplifier circuit respectively, the second frequency selective filtering circuit and the power amplifier circuit are both connected to the transmit / receive switching circuit, and the transmit / receive switching circuit is connected to the SOC chip.

[0019] A monitoring method for a power grid line monitoring device based on a sensing and computing intelligent SOC chip includes the following steps:

[0020] S1. The system powers on, starts the main thread, configures the operating parameters of each unit circuit, initializes the database, and monitors the status of the sub-threads.

[0021] S2. Construct the first sub-thread: BeiDou positioning and time synchronization;

[0022] Periodically retrieve current latitude, longitude, altitude, and clock information;

[0023] Read the location information from the device's shared database and determine whether the latitude, longitude, and altitude information in the current shared database are consistent with the BeiDou positioning. If they are inconsistent, report the event and update the shared database.

[0024] Read the device's RTC clock and determine whether the current RTC clock is consistent with the BeiDou time synchronization; if not, synchronize the device's RTC clock with the BeiDou clock information.

[0025] S3. Construct a second sub-thread to obtain data from adjacent devices;

[0026] The Bluetooth communication unit is networked with neighboring devices to obtain time-stamped current and voltage data from neighboring devices and update the shared database.

[0027] S4. Construct a third sub-thread for line fault identification and prediction;

[0028] The system continuously collects current, voltage, temperature, humidity, and altitude data from local and adjacent devices. After preprocessing, the data is updated to the shared database. The NPU is invoked to perform fault identification and prediction for the local phase line based on the five-dimensional data acquired by the device. The system also performs fault identification and prediction for the three-phase line based on the five-dimensional data acquired by the device and the voltage and current of adjacent phases. The system determines whether there are any events that need to be reported. If so, the system performs event reporting and updates the shared database.

[0029] S5. Construct the fourth child thread for uplink communication;

[0030] Using the HRF mode, a communication channel is established with the upper-layer device, and the communication of the upper-layer device is responded to in real time to determine whether the shared database needs to be updated. If so, the shared database is updated.

[0031] S6. Construct the fifth sub-thread for local communication, status monitoring, and indication;

[0032] The system responds in real time to maintenance interface communication, monitors the status acquisition interface, outputs status indicators, and determines whether the shared database needs to be updated. If so, the shared database is updated.

[0033] One preferred embodiment, step S4, specifically includes:

[0034] S41. Data acquisition and preliminary verification: acquire current, voltage, temperature, humidity and altitude data of this phase, acquire current and voltage data of adjacent phases through Bluetooth communication unit, and remove invalid values.

[0035] S42. Standardize data formats, unify data sampling frequency, align timelines, and convert data types;

[0036] S43. Denoising: A time-domain denoising algorithm is used to reduce data redundancy.

[0037] S44. Feature extraction: Parallel processing of multi-dimensional data including current, voltage, temperature, humidity, and altitude; performing harmonic analysis and trend feature extraction.

[0038] S45, fault identification, traveling wave head identification, accelerates the operation of deep learning models, performs model inference through matrix operations, and matches potential hidden dangers, faults, and traveling wave head feature libraries;

[0039] S46. Output the result. Receive the recognition result and determine whether to trigger an alarm or report an event based on the preset algorithm.

[0040] In the above technical solution of the present invention, the power grid line monitoring device based on the Sensing Computing and Communication Intelligent SOC chip includes: an SOC chip, a power supply unit, a current sampling circuit, a voltage sampling circuit, an HRF radio frequency front-end peripheral circuit, a temperature and humidity sensor, and a Bluetooth communication unit; the SOC chip is connected to the power supply unit, the current sampling circuit, the voltage sampling circuit, the HRF radio frequency front-end peripheral circuit, the temperature and humidity sensor, and the Bluetooth communication unit respectively; the power supply unit is used to draw power from the power distribution line and supply power to the device; the current sampling unit is used to convert the current of the phase line into a first voltage signal that the SOC chip can recognize, and connect the first voltage signal to the ADC input interface of the SOC chip; the voltage sampling unit is used to convert the phase line electric field into a second voltage signal that the SOC chip can recognize, and connect the second voltage signal to the ADC input interface of the SOC chip; the HRF radio frequency front-end peripheral circuit is used to switch the radio frequency transmission / reception mode; the temperature and humidity sensor is used to collect the temperature and humidity of the surrounding environment of the device, and convert them into a third voltage signal and a fourth voltage signal that the SOC chip can recognize, and connect the third voltage signal and the fourth voltage signal to the ADC input interface of the SOC chip. This invention solves the technical problems of existing power grid line monitoring devices, such as asynchronous multi-parameter data, low fault identification accuracy, and lack of intelligent prediction capabilities, caused by the limited computing power of MCUs and the modular architecture. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of a power grid line monitoring device based on a sensing and computing intelligent SOC chip according to an embodiment of the present invention;

[0043] Figure 2 This is a schematic diagram of a SOC chip according to an embodiment of the present invention;

[0044] Figure 3 This is a first schematic diagram of the monitoring method of the power grid line monitoring device based on the sensing and computing intelligent SOC chip according to an embodiment of the present invention;

[0045] Figure 4 This is a second schematic diagram of the monitoring method of the power grid line monitoring device based on the sensing and computing intelligent SOC chip according to an embodiment of the present invention.

[0046] The realization of the objective, functional characteristics and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Furthermore, in this invention, descriptions involving "first," "second," etc., are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.

[0049] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0050] See Figures 1-2 According to one aspect of the present invention, the present invention provides a power grid line monitoring device based on a sensing-computing-communication-intelligent SOC chip, wherein the power grid line monitoring device based on the sensing-computing-communication-intelligent SOC chip includes: an SOC chip, a power supply unit, a current sampling circuit, a voltage sampling circuit, an HRF radio frequency front-end peripheral circuit, a temperature and humidity sensor, and a Bluetooth communication unit; the SOC chip is connected to the power supply unit, the current sampling circuit, the voltage sampling circuit, the HRF radio frequency front-end peripheral circuit, the temperature and humidity sensor, and the Bluetooth communication unit respectively;

[0051] The power supply unit is used to draw power from the power distribution line and to supply power to the device;

[0052] The current sampling unit is used to convert the current of the phase line into a first voltage signal that the SOC chip can recognize, and to connect the first voltage signal to the ADC input interface of the SOC chip.

[0053] The voltage sampling unit is used to convert the phase line electric field into a second voltage signal that the SOC chip can recognize, and to connect the second voltage signal to the ADC input interface of the SOC chip.

[0054] The HRF radio frequency front-end peripheral circuit is used to switch the radio frequency transmission / reception mode, amplify the weak HRF communication signal output by the SOC chip, convert the electrical signal into electromagnetic wave radiation and transmit it through the antenna, or receive spatial electromagnetic waves and convert them into electrical signals, and filter the HRF operating frequency band signal input to the SOC chip.

[0055] The temperature and humidity sensor is used to collect the temperature and humidity of the surrounding environment of the device, and convert them into a third voltage signal and a fourth voltage signal that can be recognized by the SOC chip, and then connect the third voltage signal and the fourth voltage signal to the ADC input interface of the SOC chip.

[0056] Specifically, in this embodiment, the power grid line monitoring device based on the Sensing Computing and Positioning Intelligent SOC chip further includes: a memory, a status indicator circuit, and a local maintenance interface; the memory, status indicator circuit, and local maintenance interface are all connected to the SOC chip; the memory is used to store event records, as well as voltage, current, temperature, humidity, and altitude waveform data. When the storage capacity is less than 1GB, NAND Flash with parallel communication is used; when the storage requirement is greater than 1GB, eMMC memory with an SDIO interface is used; the status indicator circuit is used to indicate the working status of the device itself and the operating status of the monitored line, converting the high and low levels output by the SOC chip into the on / off state of the indicator light or the flipping of the indicator plate on the power grid line monitoring device body, indicating the status of the tested line and the device itself in real time; the local maintenance interface uses Bluetooth communication and / or RS485 communication, and is used to maintain the device; the local maintenance interface includes a USB interface and an RS232 interface. The USB interface is mainly used for program upgrades and large data copying, and the RS232 interface is mainly used for user interaction, including function debugging, parameter reading and writing, etc.

[0057] Specifically, in this embodiment, the power module has a built-in rechargeable backup power supply;

[0058] Under normal conditions, the power module is used to convert the high-voltage AC power of the power distribution line into low-voltage DC power to supply power to the device and backup power supply.

[0059] In the event of a power outage, the backup power supply of the power module provides power to the device.

[0060] The power supply unit provides a stable and reliable power supply for the SOC chip and peripheral devices. Power extraction and current sampling utilize a clamp-on current transformer with a double-winding iron core and an insulated shell. Two independent windings are wound on the iron core: the sampling winding outputs a small current of tens of mA, specifically for current sampling of the line; the power supply winding draws power from the electromagnetic energy of the line to power the entire device. The two windings are isolated by multiple layers of insulation material to prevent leakage or electromagnetic coupling interference between the windings. The current transformer has a spring-loaded fastening device, allowing it to be securely fastened to the cable under test. The power output from the power-taking winding passes through a transient suppressor, rectifier bridge, overcurrent protection, and DC-DC converter before supplying power to each circuit unit. The transient suppressor is an SMBJ22CA with a clamping voltage of approximately 26V and bidirectional protection. The rectifier bridge is a GBU1006 rectifier bridge (10A / 600V) with a surge current withstand capability >200A. The overcurrent protection uses a self-resetting fuse with an operating current of 3A. The DC-DC converter uses a multi-stage TPS54339 power supply to output the power required by the unit circuits at 5.0V and 3.3V respectively, with an efficiency >90%. The power supply unit also has a built-in 300F / 5.5V supercapacitor. When the circuit under test is energized, the power-taking circuit supplies power to the device while charging the supercapacitor; when the circuit under test is de-energized, the supercapacitor supplies power to the device.

[0061] Specifically, in this embodiment, the SOC chip includes a processing core, a storage unit, a power management unit, a clock unit, a DMA controller, and an NPU. The processing core is connected to the storage unit, power management unit, clock unit, DMA controller, and NPU via a system bus. The NPU is a neural network processing unit. The processing core is responsible for executing instructions, performing logical operations, and scheduling tasks. It includes a general-purpose computing core, an instruction cache (I-Cache), and a data cache (D-Cache). It interacts with the NPU, wide-range ADC circuit unit, BeiDou baseband processing unit, HRF baseband processing unit, and peripherals via the system bus. The processing core uses a dual-core Cortex A7 processor with a 1GHz clock speed, a 64kB L1 instruction cache, a 64kB L1 data cache, and a 256kB L2 cache. The storage unit uses 256MB of DDR3 memory and 2kB of RAM. OTP storage space; the storage unit provides high-speed temporary and persistent data storage, reducing external storage access latency; it includes volatile SDRAM for temporary storage of runtime program variables, stack data, and intermediate calculation results, a key resource for ensuring efficient CPU operation and real-time response; non-volatile OTP stores startup code, with cache synchronized with main memory via the system bus, and the DMA controller directly moves data, reducing CPU intervention; the power management unit manages the SOC's internal power supply, including power monitoring, linear voltage regulation, and ADC reference source; the linear voltage regulator is used for internal power regulation and allocation, with a built-in LDO generating 0.9V for the computing core, 1.0V for the NPU, and 1.5V for the DDR; the ADC reference source is the reference voltage for the ADC, ensuring ADC accuracy, with a reference voltage of 3.3V; the clock unit provides a synchronous clock signal and accurate... The timing control includes timers, pulse width modulation signals, and a real-time clock (RTC). The clock unit includes one 64-bit timer, two 32-bit timers, one dedicated watchdog timer, eight PWM channels, and one RTC clock. The DMA controller supports memory-to-memory, peripheral-to-memory, and memory-to-peripheral transfers, reducing CPU load. The watchdog timer (WDT) prevents software deadlocks, and its reset time is programmable. The reset system consists of modules such as power-on reset (POR), undervoltage detection (BOR), and the watchdog timer (WDT). Its function is to ensure reliable reset of the chip in the event of power failure, program crashes, or external interference, returning each module to its initial state and ensuring stable system operation. The DMA controller is a 16-channel DMA, supporting 8 / 16 / 32 / 64-bit data widths and supporting peripheral-to-peripheral, memory-to-memory, memory-to-peripheral, and peripheral-to-memory data transfers.

[0062] Specifically, in this embodiment, the NPU adopts a fixed-point arithmetic array, balancing computing power and power consumption. The computing density matches the 0.2 TOPS requirement, and it includes 8-16 basic multiply-accumulate (MAC) units, supporting parallel computation of core AI operators such as convolution, pooling, and fully connected layers, adapting to lightweight neural network inference. It is configured with 64KB L1 cache and 256KB L2 cache for temporary storage of weights and feature maps, reducing the frequency of interaction with external DDR and lowering memory access power consumption. It supports feature filtering and parameter fusion, outputting structured multi-dimensional inference results for easy processing core direct calls. The core supports embedded optimization frameworks such as TensorFlow Lite, PyTorchMobile, and ONNX Runtime Mini, allowing deployment of lightweight models like MobileNet, YOLOv8-nano, and SSD-Lite without complex porting. It supports model quantization tools (INT8 quantization), which can convert trained floating-point models into fixed-point models, adapting to the NPU computing core while reducing model storage footprint and inference power consumption.

[0063] Specifically, in this embodiment, the SOC chip includes a wide-range ADC circuit unit with at least four channels. The wide-range ADC circuit unit is connected to a current sampling circuit, a voltage sampling circuit, and a temperature and humidity sensor, respectively. The wide-range ADC circuit unit includes seven high-precision ADCs. The wide-range ADC circuit unit is a 24-bit Σ-Δ ADC with an ENOB ≥ 16 bits, a sampling rate ≥ 1MHz, a reference source of 3.3V, and supports synchronous sampling. The conversion results of the wide-range ADC circuit unit are stored in a designated memory unit via DMA. The computing core only needs to perform simple configuration on it during the startup phase, and no further control is required. The calculation results can be obtained through shared memory.

[0064] Specifically, in this embodiment, the SOC chip includes a BeiDou baseband processing unit, which is connected to a first frequency selective filtering circuit; the power grid line monitoring device based on the Sensing Computing-Tongwei Smart SOC chip further includes a first frequency selective filtering circuit; the first frequency selective filtering circuit is connected to the SOC chip and is used to filter the BeiDou satellite signals received by the antenna; the first frequency selective filtering circuit amplifies, filters, and down-converts the weak radio frequency signals received by the antenna, converting them into intermediate frequency signals or digital signals suitable for processing by the BeiDou baseband processing unit in the SOC chip; the BeiDou baseband processing unit further processes the intermediate frequency signals or digital signals output by the first frequency selective filtering circuit, including signal acquisition, tracking, demodulation and decoding of navigation messages, and extracting information such as position, speed, and time from the satellite signals; the first frequency selective filtering circuit supports B1I / B3I / GPS L1 / GLONASS L1, with a latitude and longitude positioning error of <2.5 meters and an altitude positioning error of <10 meters.

[0065] Specifically, in this embodiment, the SOC chip includes an HRF baseband processing unit, which is connected to the HRF radio frequency front-end peripheral circuit. The HRF radio frequency front-end peripheral circuit includes a transmit / receive switching circuit, a second frequency selective filtering circuit, a power amplifier circuit, and an antenna. The antenna is connected to both the second frequency selective filtering circuit and the power amplifier circuit, which are both connected to the transmit / receive switching circuit, which is connected to the SOC chip. The radio frequency front-end is responsible for signal conversion and conditioning. During reception, it performs frequency selective filtering on the weak radio frequency signal captured by the antenna to remove interference noise, and then amplifies it through a low-noise amplifier. The signal is down-converted to an intermediate frequency (IF) or baseband signal. During transmission, the baseband signal from the baseband front-end is up-converted to the radio frequency (RF) band, and the power is boosted by a power amplifier circuit to meet the requirements of long-distance transmission. Finally, it is radiated out through the antenna. The baseband front-end focuses on signal processing and protocol parsing. During reception, it demodulates, decodes, and filters the signal output from the RF front-end to recover the original data. During transmission, the data to be transmitted is encoded and modulated to be converted into a baseband signal suitable for RF front-end processing. The transmit / receive switching circuit accurately switches the transmit and receive states under the coordinated control of the baseband and RF front-ends to avoid transmit and receive signal conflicts and ensure smooth communication. The HRF radio frequency supports the 470MHz~510MHz frequency band, with a single-frequency transmit power of 20dBm (adjustable) and a receive sensitivity of -110dBm.

[0066] Specifically, in this embodiment, the various module circuits in the SOC chip are connected via a system bus or a peripheral bus. The system bus clock is 400MHz, and the peripheral bus clock is 100MHz, supporting 8 / 16 / 32-bit wide transmission. The system bus serves as a high-speed backbone for rapid data transfer between the processing core, storage unit, power management unit, clock unit, system control, NPU, wide-range ADC circuit unit, Beidou baseband processing unit, HRF baseband processing unit, and peripheral bus. The peripheral bus connects to external memory, USB, UART, SPI, and I... 2 Low-speed interfaces such as C and GPIO are interconnected through a bridge to achieve layered collaboration between internal high-speed, high-bandwidth data exchange and external low-speed, low-bandwidth data interaction.

[0067] Specifically, in this embodiment, the current sampling circuit uses a current transformer as its core to convert the current flowing through the tested circuit into a voltage signal that the SOC can recognize. The converted voltage signal is then connected to the ADC input interface of the SOC chip. The current sampling circuit consists of a current transformer, a load circuit, and a filter network. The sampling current transformer shares the same iron core as the power-drawing current transformer. The metering section has a turns ratio of 16000:1 and a rated current of 800A. The load resistor is a precision resistor with a 10Ω±0.1% accuracy and a temperature drift of less than 25ppm / ℃, which converts the current signal into a voltage signal. The filter network uses a second-order Sallen-Key low-pass filter with a cutoff frequency of 10kHz and a stopband attenuation of >40dB / dec.

[0068] Specifically, in this embodiment, the voltage sampling circuit converts the phase line voltage into a voltage signal that the SOC chip can recognize, employing a non-invasive voltage measurement method. The circuit consists of a sensing probe and a sensing unit circuit. The sensing probe, acting as a signal front end, is wrapped in a ring around the cable under test. The inner copper foil induces an electromotive force through electric field coupling, while the outer copper foil shielding layer blocks external electric field interference. The sensing unit includes a compensation circuit and an amplification circuit. The sampling signal compensation circuit consists of picofarad-level capacitors and megaohm-level resistors to compensate for signal attenuation caused by environmental changes and leakage current. The amplification circuit converts the weak induced current into a measurable voltage signal, which is then connected to the ADC input interface of the SOC chip. The SOC chip obtains the current cable voltage value based on the voltage value sampled by the ADC according to a preset algorithm.

[0069] Specifically, in this embodiment, the temperature and humidity sensor converts ambient temperature and humidity into voltage signals that the SOC chip can recognize. The sensor uses the CHT8336 chip from Shanghai Shensiling Microelectronics Technology Co., Ltd. as its core. Temperature and humidity are output via two independent analog voltage outputs, requiring no digital communication protocol and directly connecting to the SOC chip's ADC input interface. It can measure a temperature range of -40℃ to 125℃ with an accuracy of ±0.5℃, and a humidity range of 0% RH to 100% with an accuracy of ±2% RH. It is factory calibrated and requires no software compensation. The SOC chip obtains the current ambient temperature based on the voltage value sampled by the ADC, according to a preset temperature-voltage correspondence table, and the current ambient humidity based on a preset humidity-voltage correspondence table.

[0070] Specifically, in this embodiment, the HRF radio frequency front-end peripheral circuit is used to realize the transmission and reception of wireless signals at a frequency of 470MHz. The power amplifier circuit adopts an RF5110 amplifier, a dedicated 470MHz PA, a three-stage amplification architecture, a gain of 30dB, a saturated output power of +30dBm, a linearity OIP3 >40dBm, and an efficiency >50%. The transmit / receive switching circuit adopts an HMC1047 single-pole double-throw switch, a 470MHz frequency band, a switching time <300ns, an insertion loss <0.3dB, and an isolation >45dB. The antenna frequency range is 470~486MHz, the input impedance is 50 ohms, the VSWR is <2.0, and the gain is >2dBi. If the power grid line monitoring device is installed in an unshielded environment, a short rod antenna can be used, with the antenna body fixed to the outer shell of the power grid line monitoring device. If it is installed in a shielded environment (e.g., inside a sealed metal box), a long feeder antenna is required, with the antenna body placed outside the shielded environment.

[0071] Specifically, in this embodiment, the first frequency-selective filtering circuit is connected to the antenna, which is responsible for receiving signals transmitted by BeiDou satellites. The RF filter suppresses out-of-band interference (such as spurious signals and noise) and improves receiving sensitivity. The filtered signal is input to the RF front-end of the BeiDou positioning unit of the SOC. The first frequency-selective filtering circuit includes an RF front-end circuit, a bandpass filter, and a matching network circuit. The RF front-end uses a MAX2769 multi-band GNSS RF front-end, integrating an LNA, mixer, and IF amplifier. The bandpass filter uses a ceramic filter with a center frequency of 1575.42MHz, a bandwidth of 20MHz, and an insertion loss of <2dB. The matching network circuit uses a π-type network, composed of inductors, capacitors, and resistors. The antenna has a frequency range of 1561-1576MHz, an input impedance of 50Ω, a VSWR of <2.0, and a gain of >32dBi. If the power grid line monitoring device is installed in an unshielded environment, an active patch antenna can be used. If it is installed in a shielded environment (e.g., inside a sealed metal box), a long feeder antenna must be used, placing the antenna body outside the shielded environment.

[0072] Specifically, in this embodiment, the Bluetooth communication unit is used to communicate with external devices equipped with Bluetooth interfaces. The Bluetooth communication unit includes a Bluetooth chip and an antenna. The Bluetooth chip is an nRF52832 model, which supports Bluetooth 5.0 and Bluetooth Low Energy protocols, has excellent power consumption advantages, and a UART interface. The antenna structure is formed on the PCB by etching metal layers, which has high integration, low cost, and is compatible with Bluetooth frequency bands. Signal transmission and reception are achieved by optimizing the shape (such as inverted F type or monopole type). Bluetooth supports near-field configuration and debugging via a mobile APP, which is convenient for on-site operation and maintenance. It also supports communication with nearby Bluetooth-enabled devices.

[0073] Specifically, in this embodiment, the SOC chip and the device's hardware circuitry are combined to provide the hardware foundation for the functional implementation of the power grid line monitoring device. Taking a medium-voltage overhead line monitoring scenario as an example, a group of power grid line monitoring devices is deployed approximately every 1 kilometer, with each group containing 3 devices, one per phase. The device is clamped onto the cable under test via a clamp-on current transformer and an electric field sensing probe. The HRF and BeiDou antenna are installed on the top of the device, and the temperature and humidity sensors are arranged on the outside of the device casing. The device draws power from the cable under test via the clamp-on current transformer and obtains latitude, longitude, altitude, and clock information accurate to nanoseconds in real time through the BeiDou circuit unit. The current sampling circuit plus the SOC's built-in ADC is used to obtain... The current value of the cable under test is obtained through an electric field sampling circuit and the voltage value of the cable under test is obtained through the SOC's built-in ADC. The ambient temperature and humidity values ​​are obtained through a temperature and humidity sensor and the SOC's built-in ADC. The current and voltage values ​​of two adjacent monitoring devices are obtained through a Bluetooth unit circuit. Each device calls its built-in NPU to use the five-dimensional data of current, voltage, temperature, humidity, and altitude of the cable under test to complete the identification of four types of faults: short circuit, overload, lightning strike, and aging. Using the three-phase current and voltage data of the cable under test plus those of the two adjacent devices, ground fault identification and line fault prediction are completed. The memory can store application programs, fault identification and prediction results, current and voltage waveforms, and event logs. The device communicates with the upper-level equipment through the HRF channel. The upper-level equipment can send commands and push models to the device, and the device can report events, upload fault data, and current and voltage waveforms to the upper-level equipment.

[0074] See Figures 3-4 According to one aspect of the present invention, the present invention provides a monitoring method for a power grid line monitoring device based on a sensing and computing intelligent SOC chip, comprising the following steps:

[0075] S1. The system powers on, starts the main thread, configures the operating parameters of each unit circuit, initializes the database, and monitors the status of the sub-threads.

[0076] S2. Construct the first sub-thread: BeiDou positioning and time synchronization;

[0077] Periodically retrieve current latitude, longitude, altitude, and clock information;

[0078] Read the location information from the device's shared database and determine whether the latitude, longitude, and altitude information in the current shared database are consistent with the BeiDou positioning. If they are inconsistent, report the event and update the shared database.

[0079] Read the device's RTC clock and determine whether the current RTC clock is consistent with the BeiDou time synchronization; if not, synchronize the device's RTC clock with the BeiDou clock information.

[0080] S3. Construct a second sub-thread to obtain data from adjacent devices;

[0081] The Bluetooth communication unit is networked with neighboring devices to obtain time-stamped current and voltage data from neighboring devices and update the shared database.

[0082] S4. Construct a third sub-thread for line fault identification and prediction;

[0083] The system continuously collects current, voltage, temperature, humidity, and altitude data from local and adjacent devices. After preprocessing, the data is updated to the shared database. The NPU is invoked to perform fault identification and prediction for the local phase line based on the five-dimensional data acquired by the device. The system also performs fault identification and prediction for the three-phase line based on the five-dimensional data acquired by the device and the voltage and current of adjacent phases. The system determines whether there are any events that need to be reported. If so, the system performs event reporting and updates the shared database.

[0084] S5. Construct the fourth child thread for uplink communication;

[0085] Using the HRF mode, a communication channel is established with the upper-layer device, and the communication of the upper-layer device is responded to in real time to determine whether the shared database needs to be updated. If so, the shared database is updated.

[0086] S6. Construct the fifth sub-thread for local communication, status monitoring, and indication;

[0087] The system responds in real time to maintenance interface communication, monitors the status acquisition interface, outputs status indicators, and determines whether the shared database needs to be updated. If so, the shared database is updated.

[0088] One preferred embodiment, step S4, specifically includes:

[0089] S41. Data acquisition and preliminary verification: acquire current, voltage, temperature, humidity and altitude data of this phase, acquire current and voltage data of adjacent phases through Bluetooth communication unit, and remove invalid values.

[0090] S42. Standardize data formats, unify data sampling frequency, align timelines, and convert data types;

[0091] S43. Denoising: A time-domain denoising algorithm is used to reduce data redundancy.

[0092] S44. Feature extraction: Parallel processing of multi-dimensional data including current, voltage, temperature, humidity, and altitude; performing harmonic analysis and trend feature extraction.

[0093] S45, fault identification, traveling wave head identification, accelerate the operation of deep learning models, such as CNN and lightweight Transformer, perform model inference through matrix operations, and match potential hidden dangers, faults, and traveling wave head feature libraries;

[0094] S46. Output the result. Receive the recognition result and determine whether to trigger an alarm or report an event based on the preset algorithm.

[0095] Specifically, in this embodiment, by employing a highly integrated SOC chip, the power grid line monitoring device of the present invention can replace the original circuit system composed of multiple discrete chips, significantly reducing the peripheral circuits, while greatly improving the accuracy of line fault identification and prediction.

[0096] Specifically, in this embodiment, the present invention uses a highly integrated SOC chip with integrated multi-functional modules to replace the multiple discrete chips used in traditional power grid line monitoring devices for current sampling, voltage sampling, BeiDou timing and positioning, HRF communication, management, and other functions. Data between modules is exchanged through an on-chip high-speed bus. This integrated design fundamentally solves the technical problems of poor synchronization of multiple parameters and high inference delay caused by discrete modules and low speed of external interfaces. At the same time, it simplifies the circuit structure and solves the problems of poor signal integrity, slow communication speed between chips, high power consumption, space congestion, poor heat dissipation, and insufficient electrical isolation that exist in discrete device solutions.

[0097] Specifically, in this embodiment, the SOC chip integrates an NPU, which can perform multi-parameter intelligent inference. It receives current, voltage, temperature, humidity, and altitude data through the chip's internal high-speed bus, runs a CNN-RNN model, and has a low fault identification misjudgment rate. It can predict faults a certain time in advance, and its accuracy is significantly improved compared to the MCU solution. The NPU represents a leap from "threshold triggering" to "intelligent inference" compared to the MCU, and has the ability to generalize scenarios and predict faults, meeting the needs of complex power distribution networks.

[0098] Specifically, in this embodiment, the device's uplink communication uses an HRF radio frequency front-end peripheral circuit, communicating with the processing core within the SOC via an on-chip bus. Compared to the low-speed inter-chip interface communication of discrete chip solutions, this reduces latency by two orders of magnitude. Simultaneously, positioning and timing functions are introduced to facilitate daily inspections, maintenance, management, and fault location. Determining the accurate location of the power grid line monitoring device optimizes operation and maintenance plans and resource allocation, improving operational efficiency. The introduction of BeiDou precise time synchronization ensures accurate synchronization of all parameters of this device and the time of each device, enhancing fault location and isolation efficiency. Furthermore, it provides a unified time reference for power grid operation status analysis, ensuring power supply reliability.

[0099] Specifically, in this embodiment, a Bluetooth communication unit is introduced. For overhead line monitoring scenarios, three independent single-phase monitoring devices share current and voltage data with nanosecond-level time scales, enabling accurate identification of end-side grounding faults.

[0100] Specifically, in this embodiment, the SOC chip integrates SDRAM and Flash storage and has scalable storage capabilities. By reserving expansion interfaces or employing specific storage expansion technologies, users can expand the data storage capacity of the power grid line monitoring device according to actual needs, thereby meeting the demand for large-scale power data storage in different application scenarios. This scalable storage design improves the adaptability and flexibility of the power grid line monitoring device.

[0101] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. All equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A power grid line monitoring device based on a sensing and computing intelligent SOC chip, characterized in that, include: SOC chip, power supply unit, current sampling circuit, voltage sampling circuit, HRF radio frequency front-end peripheral circuit, temperature and humidity sensor and Bluetooth communication unit; The SOC chip is connected to the power supply unit, current sampling circuit, voltage sampling circuit, HRF radio frequency front-end peripheral circuit, temperature and humidity sensor and Bluetooth communication unit respectively. The power supply unit is used to draw power from the power distribution line and to supply power to the device; The current sampling unit is used to convert the current of the phase line into a first voltage signal that the SOC chip can recognize, and to connect the first voltage signal to the ADC input interface of the SOC chip. The voltage sampling unit is used to convert the phase line electric field into a second voltage signal that the SOC chip can recognize, and to connect the second voltage signal to the ADC input interface of the SOC chip. The HRF radio frequency front-end peripheral circuit is used to switch the radio frequency transmission / reception mode; The temperature and humidity sensor is used to collect the temperature and humidity of the surrounding environment of the device, and convert them into a third voltage signal and a fourth voltage signal that can be recognized by the SOC chip, and then connect the third voltage signal and the fourth voltage signal to the ADC input interface of the SOC chip.

2. The power grid line monitoring device based on the sensing and computing intelligent SOC chip according to claim 1, characterized in that, The power module has a built-in rechargeable backup power supply. Under normal conditions, the power module is used to convert the high-voltage AC power of the power distribution line into low-voltage DC power to supply power to the device and backup power supply. In the event of a power outage, the backup power supply of the power module provides power to the device.

3. The power grid line monitoring device based on a sensing and computing intelligent SOC chip according to any one of claims 1-2, characterized in that, The SOC chip includes a processing core, a storage unit, a power management unit, a clock unit, a DMA controller, and an NPU; the processing core is connected to the storage unit, the power management unit, the clock unit, the DMA controller, and the NPU via a system bus.

4. The power grid line monitoring device based on a sensing and computing intelligent SOC chip according to any one of claims 1-2, characterized in that, The SOC chip includes a wide-range ADC circuit unit with at least four channels, which is connected to a current sampling circuit, a voltage sampling circuit, and a temperature and humidity sensor, respectively.

5. The power grid line monitoring device based on a sensing and computing intelligent SOC chip according to any one of claims 1-2, characterized in that, The power grid line monitoring device based on the sensing, computing, communication, and positioning intelligent SOC chip also includes a first frequency selective filtering circuit; the first frequency selective filtering circuit is connected to the SOC chip and is used to filter the Beidou satellite signals received by the antenna.

6. The power grid line monitoring device based on the sensing and computing intelligent SOC chip according to claim 5, characterized in that, The SOC chip includes a BeiDou baseband processing unit, which is connected to a first frequency selective filter circuit.

7. The power grid line monitoring device based on a sensing and computing intelligent SOC chip according to any one of claims 1-2, characterized in that, The SOC chip includes an HRF baseband processing unit, which is connected to the HRF radio frequency front-end peripheral circuit.

8. The power grid line monitoring device based on a sensing and computing intelligent SOC chip according to any one of claims 1-2, characterized in that, The HRF radio frequency front-end peripheral circuit includes a transmit / receive switching circuit, a second frequency selective filtering circuit, a power amplifier circuit, and an antenna; the antenna is connected to the second frequency selective filtering circuit and the power amplifier circuit respectively, and the second frequency selective filtering circuit and the power amplifier circuit are both connected to the transmit / receive switching circuit, which is connected to the SOC chip.

9. A monitoring method for a power grid line monitoring device based on a sensing and computing intelligent SOC chip as described in any one of claims 1-8, characterized in that, Includes the following steps: S1. The system powers on, starts the main thread, configures the operating parameters of each unit circuit, initializes the database, and monitors the status of the sub-threads. S2. Construct the first sub-thread: BeiDou positioning and time synchronization; Periodically retrieve current latitude, longitude, altitude, and clock information; Read the location information from the device's shared database and determine whether the latitude, longitude, and altitude information in the current shared database are consistent with the BeiDou positioning. If there is a discrepancy, report the event and update the shared database; Read the device's RTC clock and determine whether the current RTC clock is consistent with BeiDou time synchronization; If they are inconsistent, the device's RTC clock will be synchronized with the BeiDou clock information; S3. Construct a second sub-thread to obtain data from adjacent devices; The Bluetooth communication unit is networked with neighboring devices to obtain time-stamped current and voltage data from neighboring devices and update the shared database. S4. Construct a third sub-thread for line fault identification and prediction; The system continuously collects current, voltage, temperature, humidity, and altitude data from local and adjacent devices. After preprocessing, the data is updated to the shared database. The NPU is invoked to perform fault identification and prediction for the local phase line based on the five-dimensional data acquired by the device. The system also performs fault identification and prediction for the three-phase line based on the five-dimensional data acquired by the device and the voltage and current of adjacent phases. The system determines whether there are any events that need to be reported. If so, the system performs event reporting and updates the shared database. S5. Construct the fourth child thread for uplink communication; Using the HRF mode, a communication channel is established with the upper-layer device, and the communication of the upper-layer device is responded to in real time to determine whether the shared database needs to be updated. If so, the shared database is updated. S6. Construct the fifth sub-thread for local communication, status monitoring, and indication; The system responds in real time to maintenance interface communication, monitors the status acquisition interface, outputs status indicators, and determines whether the shared database needs to be updated. If so, the shared database is updated.

10. The monitoring method of the power grid line monitoring device based on the sensing and computing intelligent SOC chip according to claim 9, characterized in that, Step S4 specifically includes: S41. Data acquisition and preliminary verification: acquire current, voltage, temperature, humidity and altitude data of this phase, acquire current and voltage data of adjacent phases through Bluetooth communication unit, and remove invalid values. S42. Standardize data formats, unify data sampling frequency, align timelines, and convert data types; S43. Denoising: A time-domain denoising algorithm is used to reduce data redundancy. S44. Feature extraction: Parallel processing of multi-dimensional data including current, voltage, temperature, humidity, and altitude; performing harmonic analysis and trend feature extraction. S45, fault identification, traveling wave head identification, accelerates the operation of deep learning models, performs model inference through matrix operations, and matches potential hidden dangers, faults, and traveling wave head feature libraries; S46. Output the result. Receive the recognition result and determine whether to trigger an alarm or report an event based on the preset algorithm.

Images