Intelligent temperature measurement system for cable intermediate joint

Abstract

The invention provides an intelligent temperature measurement system for a cable intermediate joint, which relates to the technical field of power equipment monitoring and comprises a temperature measurement sensing unit, a remote communication unit, an intelligent early warning and diagnosis unit, a multi-protocol compatible adaptation unit, a power supply unit and a local display control unit. The temperature measurement sensing unit fixes a plurality of sensors through a self-adaptive clamping mechanism, so that lossless temperature measurement of cables of multiple specifications and mechanism state monitoring are realized; the remote communication unit adopts dual-mode transmission and hardware encryption to guarantee stable transmission and safety of data; the intelligent early warning unit integrates multi-dimensional data through a CNN-LSTM fusion model, and outputs temperature prediction and fault diagnosis in advance; the multi-protocol adaptation unit realizes compatible butt joint with a power grid system, the power supply unit adopts the combination of solar energy and energy storage to guarantee stable power supply, and the local display control unit supports local data viewing and audible and visual alarm. The system overcomes the defects of real-time performance, adaptability, compatibility and the like in the prior art, and can early warn faults in advance, reduce power failure loss and reduce operation and maintenance cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power equipment monitoring, in particular to an intelligent temperature measurement system for cable intermediate joints. BACKGROUND

[0002] In a power transmission network, a cable intermediate joint is a key component for connecting cable lines, and its operating state directly determines the reliability of the power system. Due to the structural characteristics of the cable intermediate joint, such as the splicing of the insulating layer and the connection of the conductor, local overheating may occur due to increased contact resistance, insulation aging, and sealing failure during long-term operation, which may cause insulation breakdown, joint burning, and even cable fires, resulting in large-scale power outages and causing significant losses to industrial production and residents' lives.

[0003] The existing temperature monitoring technology for cable intermediate joints has significant defects such as poor real-time performance, poor installation and adaptability, limited communication, and poor system compatibility. The existing technology can only solve a single technical problem and cannot meet the complex requirements of high-precision acquisition and multi-specification adaptation, so there is an urgent need for an intelligent temperature measurement system for cable intermediate joints that optimizes the entire process. SUMMARY

[0004] The main purpose of the present application is to provide an intelligent temperature measurement system for cable intermediate joints to solve the composite defects of the prior art.

[0005] To solve the above technical problems, the technical solution adopted by the present application is: an intelligent temperature measurement system for cable intermediate joints, comprising a temperature measurement and perception unit, a remote communication unit, an intelligent early warning and diagnosis unit, a multi-protocol compatible adaptation unit, a power supply unit, and an on-site display and control unit. In a preferred embodiment, the temperature measurement and perception unit comprises a temperature sensor, an adaptive clamping mechanism, a mechanical state monitor, and a connecting cable, wherein: The temperature sensor uses a platinum resistance sensor and adopts a 3-wire connection method to eliminate the influence of lead resistance on measurement accuracy, and is used to collect the temperature of the core heating area of the cable intermediate joint in real time. The adaptive clamping mechanism is made of shape memory alloy material and can adjust the clamping force. The inner side is pasted with a high-temperature resistant silicone rubber pad, which is suitable for cables of different specifications and diameters, and is used to non-destructively wrap the cable to avoid damaging the cable skin. The mechanical state monitor uses a micro vibration sensor to monitor the mechanical wear state of the adaptive clamping mechanism and to warn of the risk of adhesion failure by capturing changes in vibration amplitude. The connecting cable uses a temperature-resistant shielded cable with waterproof connectors at both ends to resist industrial electromagnetic interference and harsh environmental corrosion, and to achieve stable data transmission between the temperature sensor, the mechanical state monitor, and the subsequent units.

[0006] In the preferred embodiment, the remote communication unit includes an edge computing core, a dual-mode communication module, a data encryption module, and a communication antenna, wherein: The edge computing core uses a microcontroller to execute temperature drift compensation algorithms and data processing; The dual-mode communication module is used to automatically switch transmission modes according to communication quality, ensuring stable data transmission in complex scenarios. The data encryption module uses a hardware encryption chip that stores a key for encrypting transmitted data. The communication antennas employ omnidirectional and suction cup antennas to enhance communication signal strength, increase transmission distance, and improve anti-interference capabilities.

[0007] In the preferred embodiment, the intelligent early warning and diagnosis unit includes a multi-dimensional data acquisition module, a CNN-LSTM fusion prediction model, and an early warning and diagnosis output module, wherein: The multi-dimensional data acquisition module is used to integrate temperature characteristics, mechanical characteristics, electrical characteristics, environmental characteristics, time characteristics, and system characteristics to provide comprehensive input data for the model; The CNN-LSTM fusion prediction model uses a CNN feature extraction and LSTM time modeling architecture to extract key local features and time trends of the data, and outputs temperature prediction curves and failure probabilities. The early warning and diagnostic output module is used to trigger graded early warnings based on model results, diagnose four types of problems: sensor failure, mechanical wear, connector overheating, and system failure, and output the fault type.

[0008] In the preferred embodiment, the multi-protocol compatibility adapter unit includes a core chip and a protocol driver library, wherein: The core chip uses a microcontroller to execute protocol identification and data conversion logic; The protocol driver library has built-in industrial protocol drivers, covering a variety of standard protocols and vendor protocols, to adapt to the communication needs of different power systems. In the preferred embodiment, the power supply unit includes a main power supply module, a charging management module, and a protection module, wherein: The main power supply module includes solar panels and batteries, which provide constant power to the system; The charging management module uses an MPPT solar controller to optimize solar charging efficiency and extend battery life. The protection module integrates overcharge, over-discharge, overcurrent, and reverse connection protection functions to ensure the stable operation of the power supply system.

[0009] In the preferred embodiment, the local display control unit includes a housing, a display module, a button module, and an audible and visual alarm module, wherein: The enclosure is made of IP67 rated plastic to protect internal components from dust and water immersion. The display module uses a touch screen with a built-in photosensitive sensor and supports automatic brightness adjustment, which is used to intuitively display real-time temperature, historical curves, alarm information, etc. The button module includes multiple waterproof physical buttons for on-site interface switching, parameter modification, and system reset; The audible and visual alarm module includes indicator lights and a buzzer, which are used to immediately trigger an alarm when the temperature exceeds the threshold. This invention provides an intelligent temperature measurement system for cable joints. It utilizes a high-precision platinum resistance sensor in the temperature sensing unit to collect the temperature of the core heating area of ​​the joint; a shape memory alloy adaptive clamping mechanism to achieve non-destructive fixing of cables of various specifications; a micro-vibration sensor to monitor the mechanism's status; an edge computing core in the remote communication unit to process data; a dual-mode communication module to intelligently switch transmission modes; and a hardware encryption chip to ensure data security. An intelligent early warning and diagnosis unit integrates multi-dimensional data, and a CNN-LSTM fusion prediction model outputs temperature predictions and fault probabilities. A multi-protocol compatible adaptation unit connects to existing power grid systems. The power supply unit uses solar panels and lithium batteries for power supply, and a charging management module and protection module ensure stability. A protective shell, an automatically dimming touchscreen, and an audible and visual alarm module in the local display control unit support local maintenance. This system achieves uninterrupted high-precision temperature measurement, stable data transmission, early fault warning, and compatibility with power grid systems and local maintenance. It solves the problems of poor real-time performance, insufficient adaptability, limited communication, poor compatibility, and delayed early warning in existing technologies, providing support for reducing overheating accidents and power outage losses, and lowering maintenance costs. Attached Figure Description

[0010] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a schematic diagram of an intelligent temperature measurement system for cable intermediate joints according to the present invention; In the diagram: Temperature sensing unit 1, Temperature sensor 11, Adaptive clamping mechanism 12, Mechanical condition monitor 13, Connecting cable 14, Remote communication unit 2, Edge computing core 21, Dual-mode communication module 22, Data encryption module 23, Communication antenna 24, Intelligent early warning and diagnosis unit 3, Multi-dimensional data acquisition module 31, CNN-LSTM fusion prediction model 32, Early warning and diagnosis output module 33, Multi-protocol compatible adaptation unit 4, Core chip 41, Protocol driver library 42, Power supply unit 5, Main power supply module 51, Charging management module 52, Protection module 53, Local display control unit 6, Display module 61, Audible and visual alarm module 62, Local microcontroller 63. Detailed Implementation

[0011] Example 1 like Figure 1As shown, an intelligent temperature measurement system for cable joints is used in this embodiment for temperature monitoring of cable joints in urban power grids. The ambient temperature range is -10℃ to 40℃, and there are no strong electromagnetic interference sources. It needs to achieve 24-hour uninterrupted monitoring, remote operation and maintenance, and interfacing with the existing SCADA system of the power grid.

[0012] A smart temperature measurement system for cable joints includes a temperature sensing unit 1, a remote communication unit 2, an intelligent early warning and diagnosis unit 3, a multi-protocol compatible adapter unit 4, a power supply unit 5, and a local display and control unit 6. In the preferred embodiment, the temperature sensing unit 1 includes a temperature sensor 11, an adaptive clamping mechanism 12, a mechanical condition monitor 13, and a connecting cable 14, wherein: In the preferred embodiment, the temperature sensor 11 is a PT1000-BA grade platinum resistance sensor with a measurement range of -50℃ to 200℃ and a measurement accuracy of ±0.1℃. The lead wire is made of polytetrafluoroethylene insulated silver-plated copper wire, and a 3-wire connection method is used to eliminate the influence of lead wire resistance on measurement accuracy. The sensor probe shell is made of 316 stainless steel with a corrosion resistance rating greater than IP65. To collect real-time temperature data of the core heating area of ​​cable joints, a total of four sensors are deployed: two sensors at the conductor connection point and two sensors on the surface of the insulation layer at each cable joint. The distance between the measuring points is half the circumference of the cable. The sensors at the conductor connection point are aligned with the center of the joint crimping point, and the sensors on the surface of the insulation layer are placed at the junction of the insulation layer and the sheath at both ends of the joint, covering the critical area most prone to overheating. Each temperature sensor 11 is welded to the core wire of the connecting cable 14 via a 3-wire terminal block. The weld is encapsulated with high-temperature heat-shrink tubing to prevent oxidation and moisture. The other end of the connecting cable 14 is connected to the analog signal acquisition interface of the local display control unit 6 via an M12 waterproof connector. The local display control unit 6 provides a constant excitation current to the sensor through a constant current source circuit to realize the conversion and acquisition of temperature signals. After the data is processed by the DLSTM fusion ISSA drift compensation algorithm, the measurement accuracy is improved and the accuracy attenuation caused by continuous long-term operation is reduced. In the preferred embodiment, the main body of the adaptive clamping mechanism 12 is made of Ti-50Ni-10Cu type shape memory alloy, and the whole is a semi-encircling structure with an adjustable opening size to adapt to 110kV cable intermediate joints; a 2mm thick HT-870 type high temperature resistant silicone pad is pasted on the inside, and the surface is provided with anti-slip texture to enhance the tightness of the fit with the cable sheath and avoid damage to the cable outer sheath. For non-destructive fixing and multi-size adaptation of sensors, each temperature sensor 11 is fixed to the sensor mounting base of the clamping mechanism by M3 bolts; In the preferred embodiment, the mechanical condition monitor 13 uses an MS801 miniature vibration sensor to convert acceleration into a 0-3.3V analog voltage signal. The sensor is used to monitor the mechanical wear and fit status of the adaptive clamping mechanism 12 in real time. It is embedded in the mounting groove on the side of the main body of the clamping mechanism and fixed with epoxy resin. The signal output line of the sensor is soldered to the spare core wire of the connecting cable 14 and connected to the analog signal acquisition interface of the local display control unit 6. The local display control unit 6 provides a stable 3.3V power supply to the sensor and periodically samples to acquire vibration data. In the preferred embodiment, the connecting cable 14 is a KFFP-3×0.75 type high-temperature shielded cable with oxygen-free copper conductor, polytetrafluoroethylene insulation layer, tin-plated copper mesh shielding layer, and polytetrafluoroethylene outer sheath. Both ends are equipped with M12 type IP67 waterproof connectors, and the overall sealing and insulation components are all heat-resistant to more than 160℃. For stable data transmission between temperature sensing unit 1 and local display control unit 6, resisting industrial electromagnetic interference; one end of the cable is connected to temperature sensor 11 and vibration sensor through M12 waterproof connector, and the other end is connected to cable interface of local display control unit 6 through the same waterproof connector.

[0013] In the preferred embodiment, the remote communication unit 2 includes an edge computing core 21, a dual-mode communication module 22, a data encryption module 23, and a communication antenna 24, wherein: In the preferred embodiment, the edge computing core 21 uses an STM32H743VIT6 microcontroller with a power supply voltage of 3.3V and an operating temperature range of -40℃ to 85℃. It is used to execute lightweight temperature drift compensation algorithms, data hierarchical compression and transmission control logic. It is connected to the dual-mode communication module 22 through the SPI interface to realize data transmission and reception control; it communicates with the local display control unit 6 through the UART interface to obtain the raw data of the temperature sensing unit 1; it is connected to the data encryption module 23 through the I2C interface to realize the encryption processing of transmitted data; at the same time, it has a built-in DMA controller for high-speed data transmission and reduces CPU usage. In the preferred embodiment, the dual-mode communication module 22 consists of a SIM8200CE 4G / 5G module and an SX1278 LoRa module. The SIM8200CE module supports TD-LTE / FDD-LTE / 5G NR networks, with downlink speeds ≥1Gbps and uplink speeds ≥100Mbps. It connects to the edge computing core 21 via a Mini PCIe interface. The SX1278 module supports unlicensed industrial frequency bands, has a maximum transmit power of 20dBm, a transmission distance of up to 5km, and connects to the edge computing core 21 via an SPI interface. The dual-mode communication module 22 is used to intelligently switch transmission modes according to communication quality to maintain stable data transmission in complex scenarios. It is connected to the edge computing core 21 through the Mini PCIe interface and the SPI interface respectively to receive compressed and encrypted monitoring data. The antenna interface of the module is connected to the communication antenna 24 through the radio frequency cable. The edge computing core 21 monitors the signal strength RSSI, packet loss rate PLR, and latency in real time. When RSSI ≥ -85dBm and PLR ≤ 1%, 4G / 5G mode transmission is enabled. When RSSI < -85dBm or PLR > 5%, it switches to LoRa mode to prevent data loss. In the preferred embodiment, the data encryption module 23 uses an AES128C hardware encryption chip, which supports the AES-256 symmetric encryption algorithm. The key is stored in a one-time programmable area within the chip and is connected to the edge computing core 21 via an I2C interface. Used to perform hardware encryption on transmitted data to prevent data from being tampered with. The encryption chip communicates with the edge computing core 21 through the I2C interface. The edge computing core 21 sends the data frame to be transmitted to the encryption chip. After the chip performs encryption processing according to the pre-stored AES-256 key, it returns the encrypted data. In the preferred embodiment, the communication antenna 24 includes a 4G / 5G omnidirectional fiberglass antenna and a LoRa suction cup antenna. The 4G / 5G antenna model is TA-500, with a gain of 5dBi, an operating frequency band of 700MHz-2700MHz, vertical polarization, an interface type of SMA-J, and a material of fiberglass. LoRa antenna model LA-300, gain 3dBi, operating frequency band 433MHz±10MHz, vertical polarization; To enhance communication signal strength and expand transmission coverage, the 4G / 5G antenna is connected to the antenna interface of the SIM8200CE module of the dual-mode communication module 22 via the SMA-J interface and is installed on a metal bracket near the monitoring point; the LoRa antenna is connected to the antenna interface of the SX1278 module of the dual-mode communication module 22 via the SMA-J interface and is installed on a wall bracket inside the cable tunnel.

[0014] In the preferred embodiment, the intelligent early warning and diagnosis unit 3 includes a multi-dimensional data acquisition module 31, a CNN-LSTM fusion prediction model 32, and an early warning and diagnosis output module 33, wherein: In the preferred embodiment, the multi-dimensional data acquisition module 31 consists of a data interface board and a signal conditioning circuit, used to integrate six types of monitoring data and provide input data for the prediction model. The data interface board uses a custom PCB board, integrating 16 analog signal input channels, 4 digital signal input channels and 2 Ethernet interfaces; the signal conditioning circuit uses the AD8421 amplifier and the filtering circuit is a second-order low-pass filter. The six categories of monitoring data are temperature characteristics, mechanical characteristics, electrical characteristics, environmental characteristics, temporal characteristics, and system characteristics, among which: The temperature characteristics are derived from the real-time temperature, 24-hour temperature change rate, and temperature extreme values ​​detected by the four PT1000 sensors in temperature sensing unit 1. The mechanical characteristics are derived from the amplitude, peak factor, and kurtosis detected by the vibration sensor in temperature sensing unit 1; The electrical characteristics are derived from the real-time cable load current and three-phase current imbalance detected by the power grid transformer, and are collected through the Ethernet interface; Environmental characteristics are derived from the ambient temperature and relative humidity detected by temperature and humidity sensors; The time characteristics are derived from the connector's operating years, cumulative power-on time, and last maintenance time stored in the remote monitoring backend. The system features are derived from the sensor communication status, power supply voltage, and signal quality of the communication module in the local display control unit 6 and the remote communication unit 2; The data acquisition module connects to the remote communication unit 2 via an Ethernet interface to receive monitoring data; it also interfaces with the power grid SCADA system via another Ethernet interface to collect electrical data such as load current; the data is filtered and amplified by the signal conditioning circuit and stored locally for use by the prediction model. In the preferred embodiment, the CNN-LSTM fusion prediction model 32 is deployed on a server in the remote monitoring backend to extract local features and time trends of the data, and output a 72-hour temperature prediction curve and fault probability, wherein: The six types of features are normalized using Min-Max to construct a 60×60 feature matrix input layer with 60 time-dimensional sampling points and 60 feature dimensions; The feature extraction module includes: a first convolutional layer with 32 3×3 filters, stride 1, output 60×60×32; a first pooling layer with 2×2 max pooling, stride 2, output 30×30×32; a second convolutional layer with 64 3×3 filters, stride 1, output 30×30×64; a second pooling layer with 2×2 max pooling, stride 2, output 15×15×64; and a flattening layer that converts the pooling output into a one-dimensional vector. The LSTM time modeling module includes one LSTM hidden layer with 64 nodes, Dropout set to 0.2, and an output dimension of 64; the first fully connected layer has 128 nodes; and the second fully connected layer has 2 nodes, which output the 72-hour temperature prediction and the fault probability, respectively. The training data consisted of historical data on 110kV cable joints in the local area, including those under normal operation, overheating faults, mechanical wear, and sensor drift. Each data set contained 90 days of continuous sampling. The optimizer used a weight decay setting of 0.001, an initial learning rate of 0.001, and a decay of 10% every 10 rounds. The loss function was a weighted sum of temperature prediction and cross-entropy. Iterative training was conducted until the training set loss and the test set loss were reduced to 0.001. The model calls the cached data of the multi-dimensional data acquisition module 31 through the API interface, reads batch data every 5 minutes for incremental training, and updates the model parameters in real time; the prediction curve and fault probability output by the model are transmitted to the early warning and diagnosis output module 33 through the network, which provides an alarm 68 hours earlier than the traditional threshold alarm method, thus avoiding the escalation of the fault. In the preferred embodiment, the early warning diagnosis output module 33 consists of an alarm logic processing unit, a remote push interface, and a local linkage interface, which is used to trigger graded early warnings based on the model output and output fault diagnosis results. The alarm logic processing unit uses an STM32F407VET6 microcontroller, which has built-in alarm threshold configuration and fault judgment logic. When the predicted temperature is ≥80℃, a general alarm is triggered, and an alarm SMS is sent via the remote push interface and pushed to the operation and maintenance team via the APP; when the predicted temperature is ≥120℃, a critical alarm is triggered, and the alarm is simultaneously pushed to the operation and maintenance team and the dispatch center, and an emergency operation and maintenance work order is generated. When a sensor malfunction is determined by the temperature data exceeding the range or the data not being updated for more than 10 minutes, the sensor output is abnormal; mechanical wear is determined by the vibration amplitude ≥0.1g and lasting for at least 5 minutes; connector overheating is determined by the temperature change rate ≥0.5℃ / h and lasting for at least 1 hour; system malfunction is determined by communication failures at least 3 consecutive transmissions or the power supply voltage being less than 10V or greater than 14V. The module receives the output data of the CNN-LSTM fusion prediction model 32 through the API interface, generates alarm information and diagnostic results after logical processing, communicates with the power grid operation and maintenance platform through the remote push interface, and sends the alarm signal to the local display control unit 6 through the local linkage interface to trigger the audible and visual alarm.

[0015] In the preferred embodiment, the multi-protocol compatibility adapter unit 4 includes a core chip 41 and a protocol driver library 42, wherein: In the preferred embodiment, the core chip 41 is an STM32F767IGT6 microcontroller, which supports multiple communication interfaces such as Ethernet, UART, and SPI, and operates at a power supply voltage of 3.3V. It is used to execute protocol identification and data format conversion logic. It connects to the remote communication unit 2 and the SCADA system switch via the Ethernet interface, receives monitoring data from the remote communication unit 2, converts it into a SCADA system-compatible format, and then sends it to the SCADA system. It connects to the storage area via the SPI interface to read protocol configurations and connects to the local maintenance terminal via the UART interface. In the preferred embodiment, the protocol driver library 42 has built-in industrial protocol drivers that support the MMS / GOOSE protocol and Modbus-RTU protocol required in this embodiment, wherein: The IEC 61850 protocol driver supports data reading, data reporting, alarm push, and GOOSE message transmission, mapping temperature data to the TTTR logical node of the SCADA system and alarm information to the ALM logical node of the SCADA system. The Modbus-RTU protocol driver supports RTU mode, which maps temperature data to Modbus registers. The protocol driver library 42 is used to adapt to the communication protocol of the SCADA system. It is stored in the flash memory of the core chip 41 and is called and executed by the core chip 41. It receives messages from the SCADA system through the Ethernet interface. After the classification model built into the driver library analyzes the message characteristics, it identifies the protocol type, loads the corresponding driver, converts the monitoring data into a protocol-compatible format, and sends it to the SCADA system through the Ethernet interface of the core chip 41.

[0016] In the preferred embodiment, the power supply unit 5 includes a main power supply module 51, a charging management module 52, and a protection module 53, wherein: In the preferred embodiment, the main power supply module 51 consists of an SM10-12 type monocrystalline silicon solar panel and an LP1250 type 12V lithium iron phosphate battery. The solar panel has dimensions of 300mm×400mm×30mm, an open-circuit voltage of 18V, and a maximum output power of 10W. The lithium iron phosphate battery has a capacity of 5Ah, a charge-discharge cycle life of more than 1000 cycles, an operating temperature range of -20℃ to 60℃, and is equipped with a battery management system that supports overcharge, over-discharge, and overcurrent protection functions. The solar panel is connected to the input terminal of the charging management module 52 via the MC4 waterproof connector. The output DC power is processed by the charging management module 52 to charge the lithium iron phosphate battery. The lithium iron phosphate battery is connected to the output terminal of the charging management module 52 via the positive and negative terminals to provide 12V DC power to each unit of the system. The solar panel can charge more than 60Wh per day on sunny days, which meets the system's daily power consumption and replenishes the lithium iron phosphate battery. When the battery is fully charged, it can supply power to the system for about 160 hours on its own, enabling power supply during long periods of no sunlight on cloudy days. In the preferred embodiment, the charging management module 52 uses a CN3791 MPPT solar controller with an input voltage range of 6V-30V and an output voltage of 13.8V, and has reverse connection protection and short circuit protection functions. Used to optimize solar charging efficiency and extend battery life; its input terminal is connected to the solar panel through the MC4 connector to receive the output current of the solar panel; its output terminal is connected to the battery of the main power supply module 51 through the terminal block to charge the battery; the power output interface supplies power to each unit of the system through a DC-DC converter. In the preferred embodiment, the protection module 53 consists of a DW01 battery protection chip and an 8205A MOSFET, used to provide overcharge, over-discharge, overcurrent, and reverse connection protection for the power supply system. The protection chip and MOSFET are soldered on the PCB of the charging management module 52 and connected in series between the main power supply module 51 and the load. When the battery voltage is detected to be overcharged, over-discharged or overcurrent, the DW01 chip controls the 8205A MOSFET to turn off, cutting off the charging or discharging circuit. The reverse protection diode is connected in parallel to the power input terminal to prevent short circuit or reverse connection from damaging the equipment.

[0017] In the preferred embodiment, the local display control unit 6 includes a display module 61, an audible and visual alarm module 62, and a local microcontroller 63, wherein: In the preferred embodiment, the main body shell of the local display control unit 6 is made of ABS-67 type IP67 protection level ABS plastic shell, with an impact resistance level of IK08, and is made of flame-retardant ABS. It is equipped with a waterproof sealing ring, and the inlet uses an M12 waterproof connector to protect against dust intrusion and water immersion, and to protect the internal components from the harsh outdoor environment. The internal components are fixed to the shell with screws. In the preferred embodiment, the display module 61 uses an ILI9341 touchscreen with a brightness of 300 cd / m² and a viewing angle ≥120°; it has a built-in BH1750 photosensitive sensor for local display of monitoring data and system status; the touchscreen is connected to the local microcontroller 63 via an SPI interface to receive display data sent by the local microcontroller 63, and touch operation signals are fed back to the local microcontroller 63 via an I2C interface; the photosensitive sensor is connected to the local microcontroller 63 via an I2C interface to collect ambient light intensity in real time and automatically adjust the screen brightness according to the light intensity; In the preferred embodiment, the audible and visual alarm module 62 includes an LED indicator and a buzzer, which are used to trigger a local alarm when the temperature exceeds the threshold or when the system fails. The LED indicator and the buzzer are connected to the local microcontroller 63 through a transistor driving circuit. After receiving the alarm signal from the early warning diagnostic output module 33, the local microcontroller 63 controls the LED indicator to flash and the buzzer to sound. In the preferred embodiment, the local microcontroller 63 uses an STM32F407VET6 chip, which supports communication interfaces such as UART, SPI, and I2C. It is used to handle on-site interactive logic, data display control, and alarm triggering. It is connected to the remote communication unit 2 through the UART interface to receive monitoring data and alarm information. It is connected to the display module 61 through the SPI interface to send display data and receive touch operations. It is connected to the audible and visual alarm module 62 through the GPIO interface to control alarm output. It is connected to the photosensitive sensor through the I2C interface to collect light intensity data.

[0018] The preferred embodiment also includes a working method for an intelligent temperature measurement system for cable joints: S1. Fix the temperature sensor 11 of the temperature sensing unit 1 to the conductor connection part and the surface of the insulation layer of the cable intermediate joint through the adaptive clamping mechanism 12, and connect the shielded cable to the local display control unit 6; install the solar panel and communication antenna 24, input the basic information of the equipment through the local display control unit 6, the system automatically connects to the network to complete bidirectional authentication, and the multi-protocol adaptation unit automatically identifies the docking protocol and completes the parameter configuration. S2. Temperature sensor 11 collects temperature data, mechanical condition monitor 13 collects vibration data, and simultaneously collects multi-dimensional data such as ambient temperature and humidity and power supply voltage. Temperature sensing unit 1 performs moving average filtering and drift compensation on temperature data. S3, the remote communication unit 2 performs hierarchical compression and encryption on the pre-processed data, and automatically switches modes according to the communication quality to transmit the data to the intelligent early warning and diagnosis unit 3; S4. The intelligent early warning unit analyzes multi-dimensional data through the CNN-LSTM fusion prediction model 32, outputs temperature prediction curve and fault probability, and triggers local sound and light alarm and remote work order push when the threshold is exceeded. S5, the multi-protocol compatible adaptation unit 4 synchronizes the processed temperature data and alarm information to the existing system to achieve data sharing; S6. Maintenance personnel can view data and fault information through the local display control unit 6 and locate the problem according to the troubleshooting suggestions. After the fault is repaired, the alarm can be cleared by pressing the button. The system automatically collects data to verify the repair effect and synchronizes it to the background to update the work order status.

[0019] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be defined as the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. An intelligent temperature measurement system for cable joints, characterized in that: It includes a temperature sensing unit, a remote communication unit, an intelligent early warning and diagnostic unit, a multi-protocol compatible adapter unit, a power supply unit, and a local display and control unit; The temperature sensing unit includes a temperature sensor, an adaptive clamping mechanism, a mechanical condition monitor, and a connecting cable; The remote communication unit includes an edge computing core, a dual-mode communication module, a data encryption module, and a communication antenna; The intelligent early warning and diagnosis unit includes a multi-dimensional data acquisition module, a CNN-LSTM fusion prediction model, and an early warning and diagnosis output module; The multi-protocol compatibility adapter unit includes the core chip and the protocol driver library; The power supply unit includes a main power supply module, a charging management module, and a protection module; The local display control unit includes a display module, an audible and visual alarm module, and a local microcontroller.

2. The intelligent temperature measurement system for cable joints according to claim 1, characterized in that: The adaptive clamping mechanism is wrapped around the cable joint and can be adjusted to fit different specifications of cables by adjusting the opening size. It is used to fix the temperature sensor and mechanical condition monitor and avoid damage to the cable sheath. The temperature sensor is fixed on the adaptive clamping mechanism, and the probe is aimed at the conductor connection part of the cable intermediate joint to collect the temperature of the core heating area of ​​the joint. The mechanical condition monitor is embedded in the adaptive clamping mechanism and connected to the adaptive clamping mechanism through a fastener to monitor the mechanical wear of the adaptive clamping mechanism; One end of the connecting cable is electrically connected to the temperature sensor and mechanical condition monitor via a waterproof connector, and the other end is connected to the signal interface of the local display control unit via a waterproof connector to realize data transmission between the temperature sensing unit and the local display control unit.

3. The intelligent temperature measurement system for cable joints according to claim 1, characterized in that: The edge computing core is connected to the dual-mode communication module, the local display control unit, and the data encryption module to receive raw data, execute data processing algorithms, and control data transmission. The dual-mode communication module is electrically connected to the communication antenna and the intelligent early warning and diagnostic unit, and switches the transmission mode according to the communication quality and transmits data to the intelligent early warning and diagnostic unit. The data encryption module interacts with the edge computing core, receives the data to be transmitted from the edge computing core, encrypts it, and sends the encrypted data back to the edge computing core. The communication antenna is connected to the dual-mode communication module to enhance the communication signal strength.

4. The intelligent temperature measurement system for cable joints according to claim 1, characterized in that: The multi-dimensional data acquisition module connects to the remote communication unit, external power grid transformer and environmental monitoring equipment via Ethernet interface, collects and integrates various feature data, and transmits the integrated data to the CNN-LSTM fusion prediction model; The CNN-LSTM fusion prediction model is deployed in the remote monitoring backend. It calls the cached data of the multi-dimensional data acquisition module through the interface, analyzes the data characteristics and time trends, and outputs the temperature prediction results and fault probability to the early warning and diagnosis output module. The early warning and diagnostic output module connects to the CNN-LSTM fusion prediction model via a network interface, and to the operation and maintenance platform via an interface; it also connects to the local display and control unit via a local linkage interface, triggering graded early warnings and transmitting alarms based on the model output.

5. The intelligent temperature measurement system for cable joints according to claim 1, characterized in that: The core chip connects to the remote communication unit and external power grid system via an Ethernet interface, and to the local maintenance terminal via other interfaces to receive monitoring data from the remote communication unit and perform protocol identification and data format conversion. The protocol driver library is stored in the flash memory of the core chip and interacts with the core chip. It is used to load the corresponding protocol driver after the core chip recognizes the external system protocol.

6. The intelligent temperature measurement system for cable joints according to claim 1, characterized in that: The solar panel of the main power supply module is connected to the input terminal of the charging management module via a waterproof connector, and the energy storage component of the main power supply module is connected to the output terminal of the charging management module to convert solar energy into electrical energy and store it. The charging management module is connected to the protection module via a DC-DC converter to optimize solar charging efficiency and convert electrical energy into a voltage suitable for each unit before transmitting it to the protection module. The protection module is connected in series between the charging management module and each system unit, and is connected to the power supply interface of each unit to protect the power supply circuit and stabilize the output power.

7. The intelligent temperature measurement system for cable joints according to claim 1, characterized in that: The display module is connected to the local microcontroller via an interface and is fixed on the housing of the local display control unit. It receives display data from the local microcontroller and feeds back touch operation signals to the local microcontroller. The audible and visual alarm module is connected to the local microcontroller via a drive circuit and is embedded in a reserved hole on the surface of the housing. It is used to receive control signals from the local microcontroller and trigger audible and visual alarms. The local microcontroller is connected to the photosensitive sensors of the audible and visual alarm module, the remote communication unit, and the display module. It is used to receive monitoring data and alarm signals from the remote communication unit, process the on-site interactive logic, and control the operation of each module.

8. The intelligent temperature measurement system for cable intermediate joints according to claim 2, characterized in that: The temperature sensor uses a multi-wire wiring method. The leads are fixed to the core wire of the connecting cable by welding, and the welding joint is sealed with a high-temperature resistant encapsulation component. The adaptive clamping mechanism also has a protective pad on the inside, and the opening size can be adjusted by bolts to accommodate cables of different diameters; The mechanical condition monitor is fixed in the mounting groove with epoxy resin adhesive, and the signal output line is soldered to the spare core wire of the connecting cable.

9. The intelligent temperature measurement system for cable joints according to claim 3, characterized in that: The edge computing core has a built-in direct memory access controller, which reduces CPU usage; The dual-mode communication module automatically switches transmission modes based on signal strength and packet loss rate parameters. When the communication quality is good, it adopts high-speed communication mode; when the communication quality is poor, it switches to long-distance communication mode.

10. The intelligent temperature measurement system for cable joints according to claim 4, characterized in that: The early warning and diagnostic output module has built-in alarm threshold configuration logic. When the predicted temperature reaches the first threshold, alarm information is only pushed to the operation and maintenance team; when the second threshold is reached, alarm information is simultaneously pushed to the dispatch center and an operation and maintenance work order is generated. The CNN-LSTM fusion prediction model reads batches of data at preset intervals for incremental training and updates model parameters in real time to improve prediction accuracy.

Images