Line load capacity prediction system and method based on node temperature monitoring

Abstract

The invention belongs to monitoring equipment and monitoring methods in the power industry, and particularly relates to a line load capacity prediction system and method based on node temperature monitoring. The system comprises a 4G communication module, a full-duplex general synchronous / asynchronous serial transceiver module and a temperature sensor, and is characterized in that a hardware microcontroller is also arranged in the line load capacity prediction system based on node temperature monitoring, and the hardware microcontroller comprises a controller STM32L452U, a hardware power supply T1, a voltage stabilizer VR1, hardware interfaces JP1-JP5, hardware keys SW1 and SW2, and hardware common terminals CN1-CN4; according to the prediction method, a controller U, a hardware power supply, a hardware interface, a hardware key and a hardware common end jointly form a low-power-consumption embedded microcontroller, the low-power-consumption embedded microcontroller is used for controlling data acquisition, data processing and data uploading of a sensor, and the temperature of the sensor is predicted through hardware design, a communication module, temperature acquisition and other parts. The real-time load condition of a line in the system and the environment temperature in the micrometeorological monitoring device are collected, and the purpose of predicting the temperature of the wire in real time is achieved.

Description

Technical Field

[0001] The present invention belongs to monitoring equipment and methods in the power industry, and particularly relates to a line load capacity prediction system and method based on node temperature monitoring. Background Art

[0002] At present, with the rapid development of the national economy and the increasing improvement of people's living standards, the electricity consumption is also increasing rapidly. Consequently, the scale of the power system is also increasing. The load capacity of the transmission line is mainly reflected in the line temperature. Once the temperature is too high, it will pose a huge challenge to the stable operation of the power system. With the increase in the number of transmission lines, it is extremely laborious to measure the temperatures of a large number of line nodes manually with hand-held devices, and it is impossible to timely grasp the nodes with too high temperature. Once the transmission line cannot be timely detected and processed due to too high temperature, it will affect the safety of the transmission line, and further affect industrial and agricultural production and people's lives, causing economic losses and inconvenience to people's lives. Maintenance requires manpower, material resources and financial resources. Summary of the Invention

[0003] The purpose of the present invention is to overcome the above technical deficiencies and provide a line load capacity prediction system and method based on node temperature monitoring that can control the line temperature in a timely manner, has a reasonable structure, and is safe and reliable to use.

[0004] The technical solution adopted by the present invention to solve the technical problem is that the line load capacity prediction system based on node temperature monitoring includes: a 4G communication module, a full-duplex universal synchronous / asynchronous serial transceiver module, and a temperature sensor. Its characteristics are that there is also a hardware microcontroller in the line load capacity prediction system based on node temperature monitoring. The hardware microcontroller includes a controller STM32L452U, a hardware power supply T1 and a voltage regulator VR1, hardware interfaces JP1~JP5, hardware keys SW1, SW2, and hardware common terminals CN1~CN4; their connection relationships are as follows: The 2nd pin of the controller U is connected to the 2nd pin of the key SW2 in the hardware key circuit, one end of the resistor R4, and one end of the capacitor C13. The 3rd and 4th pins of the controller U are respectively connected to the 2nd pin, 1st pin of the crystal oscillator X2, and one ends of the capacitors C6 and C5. The other ends of the capacitors C5 and C6 are grounded. The 14th pin of the controller U is connected to the E port of the 4G communication module in the hardware power supply circuit and one end of the resistor R2. The 16th and 17th pins of the controller U are respectively connected to the 3rd pin and 2nd pin of JP1 in the hardware interface circuit. The 21st pin of the controller U is connected to the 2nd pin of the light-emitting diode in the hardware key circuit. The 23rd pin of the controller U is connected to the 3rd pin of JP5 in the hardware interface circuit. The 41st pin of the controller U is connected to the 2nd pin of JP4 in the hardware interface circuit and one end of the resistor R3. The 42nd and 43rd pins of the controller U are respectively connected to the 1st pin and 3rd pin of JP2 in the hardware interface circuit. The 46th and 49th pins of the controller U are respectively connected to the 2nd pin and 1st pin of JP3 in the hardware interface circuit. The 56th pin of the controller U is connected to the 4th pin of JP5 in the hardware interface circuit. The 15th, 20th, 22nd, 24th, 25th, 26th, 27th, and 28th pins in the controller U are respectively connected to the 10th, 9th, 8th, 7th, 6th, 5th, 4th, and 3rd pins of the interface CN1 in the hardware common terminal circuit. The 33rd, 34th, 35th, 36th, 37th, and 38th pins in the controller U are respectively connected to the 1st, 2nd, 3rd, 4th, 5th, and 6th pins of the interface CN4 in the hardware common terminal circuit. The 44th and 45th pins in the controller U are respectively connected to the 4th and 3rd pins of the common terminal circuit interface CN2. The 50th, 55th, 57th, 61st, and 62nd pins in the controller U are respectively connected to the 7th, 6th, 5th, 4th, and 3rd pins of the interface CN3 in the common terminal circuit; The 5th and 6th pins in the controller U are respectively connected to the 2nd pin of the crystal oscillator X1, one end of the capacitor C1 and the 1st pin of the crystal oscillator X1, one end of the capacitor C2. The other ends of the capacitors C1 and C2 are grounded. The 1st pin of the controller U is connected to the 3.3V power supply and one end of the transformer FB1. The 13th pin of the controller U is connected to one ends of the capacitors C3 and C4 and the other end of the transformer FB1. The other ends of the capacitors C3 and C4 are grounded. The 32nd, 64th, 48th, and 19th pins of the controller U are connected to 3.3 V power supply, pin 7 of controller U is connected to pin 2 of button SW1 in the hardware button circuit, one end of resistor R6, and one end of capacitor C12. Pin 60 of controller U is connected to one end of resistor R1, and the other end of resistor R1 is grounded. Pin 12 of controller U is connected to pins 31, 63, 47, and 18 of controller U and grounded; Port G of the 4G communication module in the hardware power supply circuit is connected to pins 5, 6, 7, and 8 of power supply T1, 4G + 5V power supply, and one ends of capacitors C10 and C11. The other ends of capacitors C10 and C11 are grounded. Port S is connected to pins 1, 2, and 3 of power supply T1, +5V power supply, and the other end of resistor R2. Pin 1 of voltage regulator VR1 is connected to one end of capacitor C7 and +5V power supply. Pin 3 of voltage regulator VR1 is connected to one ends of capacitors C8 and C9 and 3.3V power supply. Pin 2 of voltage regulator VR1 is connected to the other ends of capacitors C7, C8, and C9 and ground; In the hardware interface circuit, pin 4 of interface JP1, pin 4 of JP3, and pin 2 of JP5 are connected to 3.3V power supply. Pin 4 of interface JP2 is connected to 4G + 5V power supply. Pin 3 of JP4 is connected to the other end of resistor R3 and 3.3V power supply. Pin 1 of interface JP1, pin 2 of JP2, pin 3 of JP3, pin 1 of JP4, and pin 1 of JP5 are grounded; In the hardware button circuit, pin 1 of button SW1 is connected to the other end of capacitor C12 and ground. The other end of resistor R6 is connected to 3.3V power supply. Pin 1 of button SW2 is connected to the other end of capacitor C13 and ground. The other end of resistor R4 is connected to 3.3V power supply. Pin 1 of light-emitting diode LD1 is connected to one end of resistor R5, and the other end of resistor R5 is connected to 3.3V power supply; In the hardware common terminal circuit, pin 2 of interface CN1, pin 2 of CN2, and pin 2 of CN3 are connected to 3.3V power supply. Pin 1 of CN1, pin 1 of CN2, and pin 1 of CN3 are grounded;. The line load capacity prediction method for node temperature monitoring is composed of a low-power embedded microcontroller composed of a controller U, a hardware power supply, a hardware interface, a hardware button, and a hardware common terminal, which is used to control sensor data acquisition, data processing, and data upload. A low-voltage difference regulator VR1 is used to stabilize the input power supply to 3.3V to power the microcontroller, and a MOS tube T1 is used to control the power-on status of the 4G communication module to reduce system power consumption; first, data acquisition is performed, and temperature sensors are installed at key nodes of the transmission line such as wire joints and insulators to collect the surface temperature of the wire in real time. Meteorological monitoring devices are deployed along the transmission line to collect ambient temperature, wind speed, humidity, sunshine intensity, and other micro data. Meteorological data, use load monitoring current transformers and voltage transformers to collect real-time load data of current, voltage and power of the line. Micro-meteorological data is collected every 5 minutes to ensure the real-time and continuity of the data. Load data is collected every 1 minute to capture the rapid changes in load. The collected data is uploaded to the cloud server in real time through the 4G communication module, and the data is stored and processed centrally; the collected raw data is cleaned, outliers and missing values ​​are removed, and the sliding window method is used to smooth the data. When the temperature of a node in the transmission line exceeds the limit and the line load capacity has problems, the system will use the light-emitting diode LD1 to sound and light alarm, and the system will monitor and process it in time.

[0005] The beneficial effects of the present invention are as follows: the line load capacity prediction system based on node temperature monitoring is to collect the real-time load situation of the line in the system and the ambient temperature in the micro-meteorological monitoring device through hardware design, communication module, temperature collection and other parts, so as to achieve the purpose of real-time prediction of the conductor temperature, and set the early warning threshold. Once the early warning range is reached, the line equipment owner and relevant full-time personnel can be notified at the first time, and the operation and maintenance personnel can be arranged to carry out infrared temperature measurement work in a targeted manner at the first time to ensure that the temperature of the transmission line is controllable and under control in real time. It ensures the safe operation of the transmission line, guarantees the electricity demand of industrial and agricultural production and people's lives, and saves manpower, material and financial resources for repairing the transmission line. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The following is a detailed description of the embodiments with reference to the accompanying drawings.

[0007] Figure 1 It is the hardware block diagram of the line load capacity prediction system for node temperature monitoring.

[0008] Figure 2 yes Figure 1 Schematic diagram of the hardware controller circuit in the microcontroller.

[0009] Figure 3 yes Figure 1 Schematic diagram of the hardware power supply circuit in the microcontroller.

[0010] Figure 4 yesFigure 1 Schematic diagram of the hardware interface circuit in the microcontroller.

[0011] Figure 5 is Figure 1 Schematic diagram of the hardware key circuit in the microcontroller.

[0012] Figure 6 is Figure 1 Schematic diagram of the hardware common terminal circuit in the microcontroller.

[0013] Figure 7 is Figure 1 Block diagram of the node temperature prediction model for the load nodes of the medium - power line.

[0014] Figure 8 is the block diagram of the line load capacity prediction system for node temperature monitoring.

[0015] In the figure: 1 - Node temperature monitoring and prediction system; 1 - 1 - 4G communication module; 1 - 2 - Full - duplex universal synchronous / asynchronous serial transceiver module; 1 - 3 - Microcontroller; 1 - 4 - Temperature sensor; 2 - Server; 3 - Node temperature; 4 - Solar power supply. Specific implementation mode

[0016] Example, referring to the appendix Figure 1 , in the node temperature monitoring and prediction system 1 of the line load capacity prediction system for node temperature monitoring, the temperature sensor 1 - 4 in the node temperature monitoring and prediction system 1 is unidirectionally input - connected to the microcontroller 1 - 3 and the node temperature 3. The microcontroller 1 - 3 is bidirectionally input - connected to the solar power supply 4. The microcontroller 1 - 3 is unidirectionally input - connected to the full - duplex universal synchronous / asynchronous serial transceiver module 1 - 2. The full - duplex universal synchronous / asynchronous serial transceiver module 1 - 2 is unidirectionally input - connected to the 4G communication module 1 - 1. The 4G communication module 1 - 1 is unidirectionally input - connected to the server 2. Referring to the appendix Figures 2 - 6 , the microcontroller 1 - 3 includes a hardware controller STM32L452U, a hardware power supply STS7PF30LT1, a voltage regulator XC6206P332MRVR1, crystal oscillators X1, X2, a hardware interface USB - TTLJP1, 4GJP2, ST - LinK JP3, DS18B20JP4, OLEDJP5, hardware keys SW - PB SW1, SW2, a light - emitting diode Green LD1, resistors R1~R6, capacitors C1~C13, hardware common terminal interfaces Header10XICN1, HEADER6CN4, Header CN2, HEADER7CN3; their connection relationship is: the 2nd pin of the hardware controller U is connected to Figure 5One end of pin 2 of button SW2, one end of 10K resistor R4, and one end of 0.1UF capacitor C13 in the hardware button circuit are respectively connected to pins 3 and 4 of controller U, and the other ends of pins 2 and 1 of 32.768 crystal oscillator X2, and one ends of capacitors 4.3PFC6 and C5. The other ends of capacitors C5 and C6 are grounded, and pin 14 of controller U is connected to Figure 3 Port E of the 4G communication module in the hardware power supply circuit, one end of 4.7 resistor R2, and pins 16 and 17 of controller U are respectively connected to pins 3 and 2 of USB-TTL JP1 in the hardware interface circuit. Pin 21 of controller U is connected to Figure 5 Pin 2 of Green light-emitting diode LD1 in the hardware button circuit, and pin 23 of controller U is connected to Figure 4 Pin 3 of OLED JP5 in the hardware interface circuit, and pin 41 of controller U is connected to Figure 4 Pin 2 of DS18B20 JP4 and one end of resistor R3 in the hardware interface circuit, and pins 42 and 43 of controller U are respectively connected to Figure 4 Pins 1 and 3 of 4GJP2 in the hardware interface circuit, and pins 46 and 49 of controller U are respectively connected to Figure 4 Pins 2 and 1 of ST - LinK JP3 in the hardware interface circuit, and pin 56 of controller U is connected to Figure 4 Pin 4 of OLEDJP5 in the hardware interface circuit, and pins 15, 20, 22, 24, 25, 26, 27, and 28 of controller U are respectively connected to Figure 6 Pins 10, 9, 8, 7, 6, 5, 4, and 3 of interface Header10XICN1 in the hardware common terminal circuit, and pins 33, 34, 35, 36, 37, and 38 of controller U are respectively connected to Figure 6 Pins 1, 2, 3, 4, 5, and 6 of interface HEADER6CN4 in the hardware common terminal circuit, and pins 44 and 45 of controller U are respectively connected to Figure 6 Pins 4 and 3 of interface HeaderCN2 in the hardware common terminal circuit interface, and pins 50, 55, 57, 61, and 62 of controller U are respectively connected to Figure 6 Pins 7, 6, 5, 4, and 3 of interface HEADER7CN3 in the hardware common terminal circuit; pins 5 and 6 of controller U are respectively connected to pin 2 of 8MHZ crystal oscillator X1, one end of 20PF capacitor C1 and pin 1 of crystal oscillator X1, one end of 20PF capacitor C2. The other ends of capacitors C1 and C2 are grounded. Pin 1 of controller U is connected to the 3.3V power supply and one end of transformer FB1. Pin 13 of controller U is connected to one ends of capacitors 0.1uF C3 and 1uF C4, and the other end of 100uHBead transformer FB1. The other ends of capacitors C3 and C4 are grounded. Pins 32, 64, 48, and 19 of controller U are connected to the 3.3V power supply. Pin 7 of controller U is connected to Figure 5One end of pin 2 of the SW-PB button SW1, one end of the 10K resistor R6, and one end of the 0.1uF capacitor C12 in the hardware button circuit; one end of the 10K resistor R1 is connected to pin 60 of the controller U, and the other end of the resistor R1 is grounded; pin 12 of the controller U is connected to pins 31, 63, 47, and 18 of the controller U and grounded; pins 32, 64, 48, and 19 of the controller U are connected to the 3.3V power supply; Figure 3 Port G of the 4G communication module in the hardware power supply circuit is connected to pins 5, 6, 7, and 8 of the STS7PF30LT1 power supply T1, the 4G + 5V power supply, one end of the 1uF capacitor C10, and one end of the 0.1uF C11; the other ends of the capacitors C10 and C11 are grounded; port S is connected to pins 1, 2, and 3 of the power supply T1, the +5V power supply, and the other end of the 4.7K resistor R2; pin 1 of the XC6206P332MR voltage regulator power supply VR1 is connected to one end of the 1uF capacitor C7 and the +5V power supply; pin 3 of the voltage regulator power supply VR1 is connected to one end of the capacitors 0.1uF C8, 0.1uF C9, and the 3.3V power supply; pin 2 of the voltage regulator power supply VR1 is connected to the other ends of the capacitors C7, C8, and C9 and ground; Figure 4 Pin 4 of the interface JP1, pin 4 of the interface JP3, and pin 2 of the interface JP5 in the hardware interface circuit are connected to the 3.3V power supply; pin 4 of the interface JP2 is connected to the 4G + 5V power supply; pin 3 of the interface JP4 is connected to the other end of the resistor R3 and the 3.3V power supply; pin 1 of the interface JP1, pin 2 of the interface JP2, pin 3 of the interface JP3, pin 1 of the interface JP4, and pin 1 of the interface JP5 are grounded; Figure 3 Pin 1 of the button SW1 in the hardware button circuit is connected to the other end of the 0.1uF capacitor C12 and ground; the other end of the resistor R6 is connected to the 3.3V power supply; pin 1 of the button SW2 is connected to the other end of the capacitor C13 and ground; the other end of the resistor R4 is connected to the 3.3V power supply; pin 1 of the light-emitting diode LD1 is connected to one end of the 1K resistor R5, and the other end of the resistor R5 is connected to the 3.3V power supply; Figure 6 Pin 2 of the interface CN1, pin 2 of the interface CN2, and pin 2 of the interface CN3 in the hardware common terminal circuit are connected to the 3.3V power supply; pin 1 of the interface CN1, pin 1 of the interface CN2, and pin 1 of the interface CN3 are grounded.

[0017] Refer to the appendix Figure 7 In the block diagram of the power line load monitoring model, empirical data, meteorological data, heat balance models, etc. are unidirectionally input to connect to the neural network model, and on-line monitoring data is unidirectionally input to connect to the neural network model; the neural network model is unidirectionally input to connect to the prediction of steady-state and transient load capabilities, and the neural network model is unidirectionally input to connect to the prediction of node temperature status.

[0018] Refer to the appendix Figure 8 In the block diagram of the power line load capacity prediction system, all three temperature acquisition terminals are unidirectionally input to connect to the load analysis system, and the load analysis system is connected to the early warning information and empirical data, meteorological data, heat balance models, etc. on both sides.

[0019] The prediction method of the line load capacity system for node temperature monitoring consists of a controller U, a hardware power supply, a hardware interface, a hardware button, and a hardware common terminal, which together form a low-power embedded microcontroller 1-3. It is used to control the data acquisition, data processing, and data uploading of temperature sensors 1-4. A low-dropout regulator VR1 is used to stabilize the input power supply to 3.3V to power the microcontroller 1-3, and an MOS transistor T1 is used to control the power-on situation of the 4G communication module 1-1 to reduce the system power consumption. First, data acquisition is carried out. Temperature sensors 1-4 are installed at key nodes of the transmission line, such as wire joints and insulators, to collect the surface temperature of the wire in real time. A meteorological monitoring device is deployed along the transmission line to collect micro-meteorological data such as ambient temperature, wind speed, humidity, and solar radiation intensity. A load monitoring device uses current transformers and voltage transformers to collect real-time load data of the line, such as current, voltage, and power. Among them, the micro-meteorological data is collected every 5 minutes to ensure the real-time and continuity of the data, and the load data is collected every 1 minute to capture the rapid changes of the load. The collected data is uploaded to the cloud server 2 in real time through the 4G communication module, and the data is centrally stored and processed. The original collected data is cleaned to remove outliers and missing values, and the sliding window method is used to smooth the data. When the temperature of a certain node on the transmission line exceeds the limit and the line load capacity has problems, the system will give an audible and visual alarm through the light-emitting diode LD1 and monitor and process it in time through this system.

[0020] The specific method process is as follows: 1. Data acquisition (1)Data source: Temperature sensors: Deployed at key nodes of the transmission line, such as wire joints and insulators, to collect the surface temperature of the wire in real time. The sensor type is a single-bus digital temperature sensor 1-4, such as DS18B20, which has the characteristics of high precision and low power consumption.

[0021] Micro-meteorological monitoring device: Deployed along the transmission line to collect micro-meteorological data such as ambient temperature, wind speed, humidity, and solar radiation intensity.

[0022] Load monitoring device: Collect real-time load data of the line, namely current, voltage, power, etc., through current transformers and voltage transformers.

[0023] (2)Collection time interval: Temperature data and micro-meteorological data: Collected every 5 minutes to ensure the real-time and continuity of the data. Load data: Collected every 1 minute to capture the rapid changes of the load.

[0024] (3)Data transmission: The collected data is uploaded to the cloud server 2 in real time through the 4G communication module 1-1 to ensure the centralized storage and processing of the data.

[0025] 2. Data Processing (1)Data Cleaning: Clean the collected raw data, removing outliers such as abnormal temperature data caused by sensor failures and missing values.

[0026] Use the sliding window method to smooth the data and reduce noise interference.

[0027] (2)Data Normalization: Normalize data with different dimensions such as temperature, load, and ambient temperature to the range of 0 to 1 for subsequent model processing.

[0028] (3)Feature Extraction: Extract features related to line temperature, including: current temperature, historical temperature trend; current load, historical load trend; ambient temperature, wind speed, humidity, solar radiation intensity and other micro-meteorological data.

[0029] 3. Model Construction (1)Neural Network Selection: Refer to Appendix Figure 5 Adopt the Long Short-Term Memory Network (LSTM) as the core prediction model. LSTM is a special type of Recurrent Neural Network (RNN) that is good at processing time series data and can capture long-term dependencies between temperature, load, and meteorological data.

[0030] (2)Model Input: Historical temperature data, load data, and micro-meteorological data.

[0031] Model Output: Predicted values of line temperature for a future period such as 1 hour, 6 hours, and 24 hours.

[0032] (3)Model Training: Use historical data such as temperature, load, and meteorological data for the past year to train the LSTM model offline.

[0033] (4)Loss Function: Mean Squared Error (MSE), which is used to measure the difference between the predicted temperature and the actual temperature.

[0034] (5)Optimization Algorithm: Adam optimizer, with the learning rate set to 0.001.

[0035] (6)Model Validation: Use the cross-validation method to evaluate the prediction accuracy of the model and ensure the generalization ability of the model across different time periods and different lines.

[0036] 4. Prediction Analysis (1)Online Learning: After the system is deployed, the LSTM model supports online learning. After each new data is collected, the model dynamically updates the weight parameters to adapt to the latest change trend of the line temperature.

[0037] (2)Trigger conditions for online learning: When the prediction error exceeds a preset threshold such as 5%, the model update is automatically triggered.

[0038] Fine-tune the model regularly, such as once a week, to ensure its adaptability and accuracy.

[0039] (3)Temperature prediction: Based on the current and historical data, the model predicts the change trend of the line temperature in the next 1 hour, 6 hours, and 24 hours.

[0040] The prediction results are displayed in a visual form, facilitating the maintenance personnel to intuitively understand the future trend of the line temperature.

[0041] 5. Early warning mechanism Setting of early warning thresholds: (1)According to the design specifications of the line and historical operation data, set the temperature early warning thresholds. For example: Level 1 early warning: The temperature exceeds 70°C, which is 80% of the maximum designed temperature of the line.

[0042] Level 2 early warning: The temperature exceeds 80°C, which is 90% of the maximum designed temperature of the line.

[0043] Level 3 early warning: The temperature exceeds 90°C, which is 100% of the maximum designed temperature of the line.

[0044] (2)Trigger conditions for early warning: When the predicted temperature exceeds the early warning threshold, the system automatically triggers an early warning.

[0045] The early warning level is dynamically adjusted according to the severity of the predicted temperature.

[0046] (3)Early warning notification: Push early warning information via text message, email, or mobile App to notify the owner of the line equipment and relevant full-time personnel.

[0047] The early warning information includes: early warning level, predicted temperature, prediction time, location of the affected line, etc.

[0048] Load capacity assessment (1)Dynamic assessment model: Based on the predicted temperature data, combined with the physical parameters of the line such as wire material, cross-sectional area, heat dissipation conditions, etc., construct a dynamic load capacity assessment model.

[0049] (2)Model output: The maximum allowable load of the line at present and in the future for a period of time.

[0050] (3)Load scheduling suggestions: Provide load scheduling suggestions for the power dispatching center based on the evaluation results to avoid overloading of lines.

[0051] (4)Optimization of maintenance plans: Optimize the maintenance plan of the line according to the temperature prediction and load capacity evaluation results, and give priority to the maintenance of high-risk lines.

[0052] 7. System implementation (1)Hardware platform: The low-power embedded microprocessor STM32L452U is responsible for data acquisition and processing. The 4G communication module 1-1 realizes remote data transmission.

[0053] (2)Software platform: The cloud server 2 is responsible for data storage, model training and predictive analysis.

[0054] The front-end interface displays the temperature prediction results, warning information and load capacity evaluation report.

[0055] 8. Technical advantages Real-time performance: Ensure the real-time performance and accuracy of temperature prediction through high-frequency data acquisition and online learning.

[0056] Intelligence: Use the LSTM neural network to automatically capture the complex laws of temperature changes and reduce manual intervention.

[0057] Warning accuracy: Ensure the timeliness and effectiveness of warning information based on the dynamic threshold and multi-level warning mechanism.

[0058] Load optimization: Provide a scientific basis for the safe operation of the power system through dynamic load capacity evaluation.

Claims

1. A line load capacity prediction system based on node temperature monitoring, comprising a 4G communication module (1-1), a full-duplex universal synchronous / asynchronous serial transceiver module (1-2), and a temperature sensor (1-4), characterized in that: The line load capacity prediction system based on node temperature monitoring is also provided with a hardware microcontroller (1-3), which includes a controller STM32L452U, a hardware power supply T1 and a voltage regulator VR1, hardware interfaces JP1~JP5, hardware buttons SW1, SW2, and hardware common terminals CN1~CN4; their connection relationship is: the 2nd foot of the controller U is connected to the 2nd foot of the button SW2 in the hardware button circuit, one end of the resistor R4, and one end of the capacitor C13, and the 3rd and 4th feet of the controller U are respectively connected to the crystal Pin 2 and 1 of the oscillator X2, one end of the capacitors C6 and C5, and the other end of the capacitors C5 and C6 are grounded. Pin 14 of the controller U is connected to the E port of the 4G communication module in the hardware power supply circuit and one end of the resistor R2. Pins 16 and 17 of the controller U are respectively connected to pins 3 and 2 of JP1 in the hardware interface circuit. Pin 21 of the controller U is connected to pin 2 of the light-emitting diode in the hardware key circuit. Pin 23 of the controller U is connected to pin 3 of JP5 in the hardware interface circuit. Pin 41 of the controller U is connected to pin 2 of JP4 in the hardware interface circuit and resistor R3. At one end, the 42nd and 43rd pins of the controller U are connected to the 1st and 3rd pins of JP2 in the hardware interface circuit respectively, the 46th and 49th pins of the controller U are connected to the 2nd and 1st pins of JP3 in the hardware interface circuit respectively, the 56th pin of the controller U is connected to the 4th pin of JP5 in the hardware interface circuit, the 15th, 20th, 22nd, 24th, 25th, 26th, 27th, 28th pins of the controller U are connected to the 10th, 9th, 8th, 7th, 6th, 5th, 4th, 3rd pins of the interface CN1 in the hardware common circuit respectively, the 33rd, 34th, 35th, 36th, 37th pins of the controller U are connected to the 4th pin of JP5 in the hardware interface circuit respectively. Pins 1, 2, 3, 4, 5, and 6 of interface CN4 in the hardware common circuit are connected respectively; pins 44 and 45 of controller U are connected respectively to pins 4 and 3 of interface CN2 in the common circuit; pins 50, 55, 57, 61, and 62 of controller U are connected respectively to pins 7, 6, 5, 4, and 3 of interface CN3 in the common circuit; pins 5 and 6 of controller U are connected respectively to pin 2 of crystal oscillator X1, one end of capacitor C1, pin 1 of crystal oscillator X1, and one end of capacitor C2; the other ends of capacitors C1 and C2 are grounded; pin 1 of controller U is connected to 3.3V power supply and one end of transformer FB1; pin 13 of controller U is connected to one end of capacitors C3 and C4, and the other end of transformer FB1; the other ends of capacitors C3 and C4 are grounded; pins 32, 64, 48, and 19 of controller U are connected to 3.3 V power supply, the 7th pin of the controller U is connected to the 2nd pin of the button SW1 in the hardware button circuit, one end of the resistor R6, and one end of the capacitor C12, the 60th pin of the controller U is connected to one end of the resistor R1, and the other end of the resistor R1 is grounded, the 12th pin of the controller U is connected to the 31st, 63rd, 47th, and 18th pins of the controller U and grounded; the G port of the 4G communication module (1-1) in the hardware power supply circuit is connected to the 5th, 6th, 7th, and 8th pins of the power supply T1 and the 4G+5V power supply, one end of the capacitors C10 and C11, and the other end of the capacitors C10 and C11 is grounded, the S port is connected to the 1st, 2nd, and 3rd pins of the power supply T1, the +5V power supply, and the resistor R2 The other end of the voltage regulator VR1, pin 1 of the voltage regulator VR1 is connected to one end of the capacitor C7 and the +5V power supply, the 3rd pin of the voltage regulator VR1 is connected to one end of the capacitors C8 and C9 and the 3.3V power supply, the 2nd pin of the voltage regulator VR1 is connected to the other end of the capacitors C7, C8 and C9 and the ground; the 4th pin of the interface JP1, the 4th pin of JP3 and the 2nd pin of JP5 in the hardware interface circuit are connected to the 3.3V power supply, the 4th pin of the interface JP2 is connected to the 4G+5V power supply, the 3rd pin of JP4 is connected to the other end of the resistor R3 and the 3.3V power supply, the 1st pin of the interface JP1, the 2nd pin of JP2, the 3rd pin of JP3, the 1st pin of JP4 and the 1st pin of JP5 are grounded; the 1st pin of the button SW1 in the hardware button circuit is connected to the other end of the capacitor C12 and the ground, the other end of the resistor R6 is connected to the 3.3V power supply, and the 1st pin of the button SW2 is connected to the capacitor C13 The other end of the resistor R4 is connected to the 3.3V power supply, the pin 1 of the light-emitting diode LD1 is connected to one end of the resistor R5, and the other end of the resistor R5 is connected to the 3.3V power supply; the 2nd pin of the interface CN1, the 2nd pin of CN2, and the 2nd pin of CN3 in the hardware common circuit are connected to the 3.3V power supply, and the 1st pin of CN1, the 1st pin of CN2, and the 1st pin of CN3 are grounded.

2. The line load capacity prediction system based on node temperature monitoring according to claim 1 is characterized in that: The line load capacity prediction method for node temperature monitoring is a low-power embedded microcontroller (1-3) composed of a controller U, a hardware power supply, a hardware interface, a hardware button, and a hardware common terminal, which is used to control sensor data acquisition, data processing, and data upload. A low-voltage difference regulator VR1 is used to stabilize the input power supply to 3.3V to power the microcontroller (1-3), and a MOS tube T1 is used to control the power-on status of a 4G communication module (1-1) to reduce system power consumption; firstly, data acquisition is performed, and a temperature sensor (1-4) is installed at key nodes of a transmission line such as a conductor joint. , insulators, etc., collect the surface temperature of the conductor in real time, deploy a meteorological monitoring device along the transmission line to collect micro-meteorological data such as ambient temperature, wind speed, humidity, and sunshine intensity, and use a load monitoring current transformer and a voltage transformer to collect the real-time load data of the line's current, voltage, and power. The micro-meteorological data is collected every 5 minutes to ensure the real-time and continuity of the data, and the load data is collected every 1 minute to capture the rapid changes in the load. The collected data is uploaded to the cloud server (2) in real time through the 4G communication module (1-1), and the data is centrally stored and processed; The collected raw data is cleaned, outliers and missing values ​​are removed, and the data is smoothed using the sliding window method. When the temperature at a certain node of the transmission line (3) exceeds the limit and the line load capacity has a problem, the system will use the light-emitting diode LD1 to sound and light alarm, and the system will monitor and handle it in time.

Images