A multi-scene self-adaptive passive wireless temperature measurement method for power equipment based on RFID
By adaptively configuring RFID temperature measurement tags and dynamically optimizing communication parameters, combined with machine learning models, the problems of signal interference and temperature identification in multiple scenarios of power equipment have been solved, achieving efficient and reliable temperature monitoring and early warning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING YINGHUADA POWER ELECTRONICS ENG TECH CO LTD
- Filing Date
- 2026-03-13
- Publication Date
- 2026-06-12
AI Technical Summary
Existing passive wireless RFID temperature measurement solutions struggle to cope with signal interference, low accuracy in temperature fluctuation identification, insufficient scenario adaptability, and lack of self-learning and optimization capabilities in various power equipment scenarios, resulting in signal attenuation, data loss, high false alarm and false alarm rates, and performance degradation.
An RFID-based adaptive passive wireless temperature measurement method for power equipment in multiple scenarios is adopted. By collecting and classifying scene features, the structure and communication parameters of the temperature measurement tag are adaptively configured. Combined with a machine learning model, real-time optimization is performed to achieve accurate adaptation of the tag to the scene and data processing.
It improves signal stability and data transmission reliability in complex scenarios, enhances the accuracy of temperature identification and early warning precision, reduces deployment costs, and has self-learning capabilities to maintain excellent performance over the long term.
Smart Images
Figure CN122192548A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power equipment condition monitoring, and in particular to an RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment. Background Technology
[0002] Electrical equipment generates heat due to the current effect during operation, and excessively high temperatures are one of the main signs of equipment failure. Therefore, real-time temperature monitoring of critical nodes in electrical equipment (such as cable heads, contacts, and busbars) is crucial for preventing accidents and ensuring the safe and stable operation of the power grid. RFID passive wireless temperature measurement technology has been widely used in the field of power temperature monitoring in recent years due to its advantages such as requiring no batteries, being maintenance-free, and offering flexible installation.
[0003] However, the power industry has a complex and diverse range of application scenarios, and power equipment in different scenarios varies significantly in terms of installation space, electromagnetic environment, high voltage level, and degree of metal interference. For example, the internal space of ring main units is extremely narrow, with dense metal components and a high voltage level, which places extremely high demands on the size, insulation performance, and anti-metal interference capabilities of temperature measurement tags; transformers of wind power equipment are usually installed at long distances and have complex on-site electromagnetic interference, posing challenges to communication distance and signal stability; and cables are densely distributed in low-voltage cabinets, and temperature fluctuations are frequent, requiring high accuracy in temperature identification.
[0004] Existing passive wireless RFID temperature measurement solutions mostly employ fixed hardware configurations and uniform processing logic, making it difficult to address the diverse needs across various scenarios. Their main shortcomings are: (1) The signal interference response capability is weak, and fixed communication parameters are prone to signal attenuation and data loss in complex electromagnetic or metallic environments; (2) The accuracy of temperature fluctuation identification is low, and it is difficult to distinguish between normal fluctuations and abnormal heating with a unified threshold, resulting in a high false alarm and false alarm rate; (3) Insufficient scene adaptability; a single label design is difficult to maintain high accuracy in all scenarios. (4) Lacking self-learning and optimization capabilities, long-term performance may decline due to equipment aging or environmental changes.
[0005] Therefore, there is an urgent need in this field for an adaptive temperature measurement method that can intelligently sense the environment, dynamically adjust parameters, accurately identify anomalies, and have continuous optimization capabilities.
[0006] Therefore, this invention proposes an RFID-based adaptive passive wireless temperature measurement method for power equipment in multiple scenarios. Summary of the Invention
[0007] The purpose of this invention is to address the shortcomings of existing technologies by proposing an RFID-based adaptive passive wireless temperature measurement method for power equipment in multiple scenarios.
[0008] To achieve the above objectives, the present invention adopts the following technical solution: An RFID-based adaptive passive wireless temperature measurement method for power equipment in multiple scenarios includes the following steps: S1. Scene Feature Acquisition and Classification: Acquire scene parameters of the target power equipment; input the scene parameters into a pre-trained scene classification model and output the corresponding scene type; S2, Adaptive Configuration of Temperature Measurement Tag: Based on the scene type identified by S1, select the corresponding installation structure for the RFID passive temperature measurement tag, configure the thickness of the high voltage resistant insulation layer, select the material of the anti-metal shielding layer, and set the gain of the loop antenna to achieve precise adaptation of the tag structure and electrical parameters to the specific scene. S3. Dynamic optimization of communication parameters of the data collector: The data collector periodically sends a detection signal to the temperature measurement tag through its antenna module and obtains the signal strength indicator RSSI and data transmission success rate in real time; when the RSSI is lower than the preset first threshold or the data transmission success rate is lower than the preset second threshold, the communication parameter dynamic adjustment strategy is triggered; after adjustment, the communication quality is re-evaluated until the preset communication quality target is met, forming a closed-loop optimization. S4. Differentiated processing of temperature measurement data: After preprocessing, the temperature data collected by the temperature measurement tag is compared with the temperature threshold model of the corresponding scenario type. The temperature threshold model defines the normal operating temperature range, the warning temperature threshold, and the alarm temperature threshold. Based on the comparison result, it is determined whether the device is in a normal, warning, or alarm state, and the corresponding data recording frequency and alarm response mechanism are triggered. S5. Remote monitoring and feedback optimization: The remote monitoring platform receives and stores temperature data, equipment status and communication parameters. It uses historical data to establish a temperature trend prediction model through the Long Short-Term Memory (LSTM) algorithm to predict future temperature changes. It also collects on-site feedback data periodically and iteratively optimizes the scenario classification model, temperature threshold model and communication parameter adjustment strategy.
[0009] Preferably: In step S1: The scenario parameters include basic equipment parameters, environmental parameters, and heat point parameters; The scenario types include at least ring main unit plug scenario, switch cabinet contact scenario, transformer busbar scenario, and low-voltage switchgear cable scenario; the scenario classification model is a neural network model based on the error backpropagation algorithm, which is trained by a large number of labeled scenario parameter samples.
[0010] Preferably, the basic parameters of the equipment include the equipment type and the high voltage level. The environmental parameters include installation space dimensions, electromagnetic interference intensity, and distribution of metallic environment. The parameters of the heating point include the location of the heating point, historical heating fault records, and normal operating temperature range.
[0011] Preferably, in step S2, the selection of the installation structure is as follows: For ring main unit plugging scenarios, select a plug-type installation part with an inner diameter specification of M12-1.25; For switchgear contact applications, a snap-on mounting part made of elastic alloy material is selected; For transformer busbar scenarios, a hook-type mounting part that is fastened with bolts is selected; For low-voltage switchgear cable scenarios, choose anti-metal temperature measurement tags that can be secured with cable ties. The configuration of the high-voltage resistant insulation layer thickness is based on the high-voltage level of the scenario: When the high voltage level is ≤10KV, the insulation layer thickness is 3mm; When 10KV < high voltage level ≤ 35KV, the insulation layer thickness is 5mm; When the high voltage level is >35KV, the insulation layer thickness is 8mm; The configuration of the anti-metal shielding layer material and the loop antenna gain is based on the electromagnetic interference intensity and the distribution of the metallic environment: When the electromagnetic interference intensity is ≤60dBμV / m and the metal components are sparse, a common nano-silver shielding layer is used, and the antenna gain is 3dBi. When the electromagnetic interference intensity is less than 60dBμV / m and less than 100dBμV / m and the metal components are relatively dense, a nano-silver enhanced shielding layer is used, and the antenna gain is 4-5dBi. When the electromagnetic interference intensity is >100dBμV / m and the metal components are densely packed, a nano-silver-carbon fiber composite shielding layer is used, and the antenna gain is 6dBi.
[0012] Preferably, in step S3, the dynamic adjustment strategy for communication parameters specifically includes: RF output power adjustment: The power is adjusted in stages from 5dBm to 30dBm. When -80dBm < RSSI ≤ -70dBm, the power is increased by 5dBm. When -90dBm < RSSI ≤ -80dBm, increase the power by 10dBm; When RSSI ≤ -90dBm, the power will be increased to the maximum value of 30dBm; Antenna access quantity adjustment: Initially connect 1 antenna. If the communication quality still does not meet the standard after adjusting the power, gradually increase the number of antennas, up to a maximum of 4, and prioritize adding antennas in areas with weak signal coverage. Tag signal transmission cycle adjustment: dynamically adjusted within the range of 0.5 seconds to 5 seconds. When the signal interference is small, a long cycle of 3-5 seconds is used to reduce power consumption, and when the signal interference is large, a short cycle of 0.5-1 seconds is used to improve the probability of data reception. The communication quality target is RSSI ≥ -70dBm and data transmission success rate ≥ 98%.
[0013] Preferably, in step S4, the preprocessing includes outlier removal and fluctuating data smoothing; Abnormal data removal involves directly removing data that exceeds the operating range of the temperature sensor on the temperature measurement tag. Fluctuation data smoothing employs a moving average filtering algorithm, the mathematical expression of which is: ,in This represents the average temperature value calculated after filtering, in degrees Celsius; n represents the size of the sliding window, ranging from 5 to 10 data points. This represents the i-th raw temperature data value within the sliding window, in degrees Celsius.
[0014] Preferably, in step S4, the temperature threshold model sets different thresholds for different scene types, as follows: For the ring main unit plugging scenario (12KV), the normal operating temperature range is -40℃ to +85℃, the warning threshold is set to 90℃, and the alarm threshold is set to 100℃. For switchgear contact scenarios (35KV), the normal operating temperature range is -40℃ to +90℃, the warning threshold is set to 95℃, and the alarm threshold is set to 105℃. For transformer busbar scenarios (110KV), the normal operating temperature range is -40℃ to +95℃, the warning threshold is set to 100℃, and the alarm threshold is set to 110℃. For low-voltage switchgear cable scenarios (0.4KV), the normal operating temperature range is -40℃ to +75℃, the warning threshold is set to 80℃, and the alarm threshold is set to 90℃.
[0015] Preferably, in step S5, the temperature trend prediction model is constructed using the Long Short-Term Memory (LSTM) network algorithm. The input of the model is a series of temperature data over a past period, including historical temperature data for the last 1 hour, 3 hours, and 24 hours, and the output is the predicted temperature values for the next 1 hour and 3 hours. Its iterative optimization is carried out regularly. The optimization is based on factors including the error between the temperature measurement data collected on site and the data calibrated manually, the accuracy of the early warning and alarm signals, and the changes in scene parameters. The internal weight parameters of the scene classification model and the temperature threshold model are adjusted through optimization algorithms, and the trigger threshold and adjustment range of the communication parameter adjustment strategy are fine-tuned.
[0016] A system for implementing an RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment, characterized in that it includes: Multiple RFID passive temperature measurement tags, whose installation structure, insulation layer thickness, shielding layer material and antenna gain can be configured according to instructions or preset rules; At least one data acquisition unit, with a built-in antenna module and processor, is used to perform dynamic optimization steps for the data acquisition unit's communication parameters and to communicate with the temperature measurement tag and the remote monitoring platform; The remote monitoring platform is equipped with a scene classification model, a temperature threshold model, a temperature trend prediction model, and a feedback optimization algorithm. It is used to perform the scene feature acquisition and classification, temperature measurement data differentiation processing, and remote monitoring and feedback optimization steps, and to send configuration instructions to the data collector.
[0017] A computer-readable storage medium storing a computer program thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of a multi-scenario adaptive passive wireless temperature measurement method for power equipment based on RFID.
[0018] The beneficial effects of this invention are as follows: Intelligent anti-interference: By dynamically optimizing communication parameters through a closed loop, the problem of signal instability in complex scenarios is fundamentally solved, resulting in extremely high data transmission reliability.
[0019] Precise identification: The combination of scenario-based threshold models and advanced filtering algorithms enables the keen detection of abnormal heating, greatly improving the accuracy of early warning.
[0020] Flexible adaptation: The temperature measurement tags are configured from multiple dimensions, enabling a single method to flexibly adapt to a wide variety of power equipment scenarios. This approach is highly versatile and reduces deployment costs.
[0021] Continuous evolution: By introducing machine learning models and feedback mechanisms, the system is made capable of self-learning, becoming "smarter" with use and maintaining excellent performance over the long term.
[0022] Maximizing the advantages of passive operation: While fully leveraging the advantages of RFID's passive and maintenance-free operation, intelligent methods are used to compensate for its potential lack of reliability in complex environments. Attached Figure Description
[0023] Figure 1This is an overall flowchart of an RFID-based adaptive passive wireless temperature measurement method for power equipment in multiple scenarios proposed in this invention. Figure 2 This invention presents a flowchart of the dynamic optimization process for communication parameters of a data acquisition device in a multi-scenario adaptive passive wireless temperature measurement method for power equipment based on RFID. Figure 3 This invention presents a flowchart of the temperature data differentiation processing for a multi-scenario adaptive passive wireless temperature measurement method for power equipment based on RFID. Figure 4 This is a flowchart illustrating the specific implementation of the RFID-based adaptive passive wireless temperature measurement method for power equipment in a ring main unit (12KV) scenario. Detailed Implementation
[0024] The technical solution of the present invention will be further described in detail below with reference to specific embodiments.
[0025] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "setting" should be interpreted broadly. For example, they can refer to a fixed connection or setting, a detachable connection or setting, or an integral connection or setting. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.
[0026] Example 1: Ring Main Unit Blockage Scenario (12KV) This embodiment takes the 12KV ring main unit, which is commonly used in urban power distribution networks, as the application object. Its internal space is narrow, metal components are densely packed, and the electromagnetic environment is complex.
[0027] Scene feature collection and classification Data collection parameters: Basic equipment parameters: Equipment type: Ring main unit; High voltage level: 12KV.
[0028] Environmental parameters: Installation space dimensions: 30cm (length) × 20cm (width) × 15cm (height); Electromagnetic interference intensity: 80 dBμV / m (detected value); Metal environment distribution: Dense metal partitions and supports inside the cabinet, and all metal conductors within 10cm of the temperature measuring point.
[0029] Heating point parameters: Location of heating point: A-phase cable connector; Historical fault record: The temperature at this point once rose to 110℃ due to poor contact; Normal operating temperature range: -10℃ ~ +65℃ (based on historical operating data).
[0030] Scene classification: The above parameter vector is input into a pre-trained BP neural network scene classification model (this model has been trained with more than 1000 samples, with an accuracy of >99%). The model outputs that the scene is "ring network cabinet blockage scene" and gives a confidence level of 98.5%.
[0031] Temperature tag adaptive configuration Installation structure compatibility: Select the temperature measuring label with "plug-type installation part" and its internal thread specification is M12-1.25. It can be directly screwed in to replace the original cable head plug of the ring main unit. The installation is convenient and the contact is good.
[0032] Insulation and shielding parameter matching: Insulation layer: The high voltage level is 12KV (meeting the condition of 10KV < level ≤ 35KV), so the thickness of the high voltage resistant insulation layer is configured as 5mm.
[0033] Shielding layer and antenna: The electromagnetic interference intensity is 80 dBμV / m (satisfying the condition of 60dBμV / m < interference ≤ 100dBμV / m) and the metal components are densely packed, so a "nano-silver enhanced anti-metal shielding layer" is selected and the gain of the loop antenna is adjusted to 4 dBi.
[0034] Dynamic optimization of data acquisition communication parameters Initial state: The collector operates with a single antenna, the RF power is set to 15dBm, and the tag signal transmission period is 1 second.
[0035] Quality monitoring and triggering: Initial communication monitoring showed that the RSSI (Signal Strength Indicator) was -75dBm (below the -70dBm threshold), and the data transmission success rate was 92% (below the 95% threshold). The system determined that the communication quality was substandard and immediately triggered dynamic parameter adjustment.
[0036] Dynamic adjustment strategy: Power adjustment: When RSSI is in the range of -80dBm < RSSI ≤ -70dBm, the RF power will be increased by 5dBm to 20dBm according to the strategy.
[0037] Antenna adjustment: After power adjustment and retesting, RSSI rose to -68dBm, but the success rate was only 96%, still not fully meeting the standard. The system then activated the second antenna to create spatial diversity and enhance signal reception capability.
[0038] Periodic adjustment: To cope with minor interference, the tag signal transmission period is slightly shortened to 0.8 seconds to increase the data sampling density.
[0039] Closed-loop evaluation: After the above adjustments and re-monitoring, RSSI stabilized at -65dBm, and the data transmission success rate improved to 99.5%, fully meeting the preset high-quality communication target (RSSI ≥ -70dBm and success rate ≥ 98%). Parameter optimization is complete.
[0040] Temperature measurement data differentiation processing Threshold model application: Call the dedicated temperature threshold model for "ring network cabinet plugging scenario (12KV)": normal range -40℃~+85℃, warning threshold 90℃, alarm threshold 100℃.
[0041] Data preprocessing: Five temperature data points were continuously collected from the tag: 88℃, 91℃, 89℃, 95℃, and 87℃. First, all data were within the sensor's effective range (-40℃ to 125℃), and no anomalies were discarded. Then, a moving average filtering algorithm with a sliding window size of n=5 was used for smoothing. T avg =(88+91+89+95+87) / 5=90℃ Status Assessment and Response: The calculated average temperature of 90℃ has reached the warning threshold. The system immediately classifies the status as "warning" and sends a warning signal to the remote monitoring platform. At the same time, it automatically increases the data acquisition frequency to 0.5 seconds / time to strengthen the monitoring of the hot spot.
[0042] Remote monitoring and feedback optimization Real-time monitoring: The platform interface highlights the temperature of the A-phase cable head of the ring main unit at 90℃ (warning status) and plots a trend curve showing the temperature rising from 75℃ to 90℃ over the past hour.
[0043] Trend Prediction: The LSTM temperature trend prediction model analyzes historical data from the past 1 hour and 3 hours, predicting that the temperature at this point may rise to 98℃ within the next hour, approaching the alarm threshold. The platform issues an early warning alert: "Abnormal temperature trend, please monitor carefully."
[0044] Closed-loop feedback: After receiving the report, maintenance personnel conducted an on-site inspection, confirmed the issue as a loose connection, and addressed it promptly. One month later, the system collected data on this incident: the average error between the temperature measurement data and subsequent manual calibration data was ≤ ±0.8℃, indicating the early warning was accurate. Based on this feedback, the system fine-tuned the weights of the "ring mains blockage scenario" classification model, making it more accurate in identifying similar features.
[0045] Example 2: Wind turbine transformer busbar scenario (35KV) This embodiment focuses on the busbar connection point of an outdoor box-type transformer in a wind farm, characterized by long installation distance, large ambient temperature difference, and complex electromagnetic interference.
[0046] Scene feature collection and classification Data Acquisition Parameters: Equipment Type: Box-type transformer; High Voltage Level: 35KV. Environmental Parameters: Installation distance (from data collector to tag): approximately 15 meters; Electromagnetic interference intensity: 65 dBμV / m (affected by the fan frequency converter); Metallic environment distribution: The busbar itself is a large metal conductor, surrounded by a transformer oil tank. Heat Point Parameters: Heat point location: Low-voltage side busbar connection; Normal operating temperature range: -30℃~+80℃.
[0047] Scene classification: The model identifies it as a "transformer busbar scene".
[0048] Temperature tag adaptive configuration Installation structure: The "hook-type mounting part" is selected and is firmly fastened to the surface of the busbar with stainless steel bolts.
[0049] Insulation and shielding: Class 35KV, insulation layer thickness 5mm. Electromagnetic interference 65dBμV / m, but considering the needs of long-distance outdoor communication, a nano-silver enhanced shielding layer is used, and the antenna gain is configured to 5dBi to compensate for path loss.
[0050] Dynamic optimization of data acquisition communication parameters Initial challenge: Initial communication RSSI was only -85dBm, with a success rate of 90%.
[0051] Optimization process: First, the power was increased by 10dBm (to 25dBm), improving the RSSI to -72dBm. Then, a directional antenna was added to align with the transformer substation, further improving the RSSI to -66dBm, but the success rate fluctuated. Finally, the signal transmission period was shortened from 2 seconds to 1 second, stabilizing the success rate at 98.5%.
[0052] Temperature measurement data differentiation processing Threshold model: The "Transformer busbar scenario (35KV)" model is applied: normal range -40℃~+95℃, warning threshold 100℃, alarm threshold 110℃.
[0053] Data processing: Due to diurnal variations in ambient temperature, data fluctuates significantly. A large sliding window of n=10 is used for smoothing, effectively filtering out normal fluctuations. On a certain day, the average temperature was found to consistently exceed 100℃, triggering an alert.
[0054] Remote monitoring and feedback optimization The platform's predictive model indicated that temperatures would continue to rise, allowing maintenance personnel to remotely adjust the fan load and prevent downtime. The feedback data was used to optimize long-distance communication strategies and the temperature threshold for this scenario.
[0055] Example 3: Switchgear contact scenario (10KV) This embodiment is for the vacuum circuit breaker contacts of a 10KV high-voltage switchgear in a factory power distribution room.
[0056] Scene feature collection and classification Data Acquisition Parameters: Equipment Type: Switchgear; High Voltage Level: 10KV. Environmental Parameters: Compact internal space; Electromagnetic Interference Intensity: 55 dBμV / m (relatively low); Metallic Environment Distribution: Contacts are located within the arc-extinguishing chamber, surrounded by metallic shielding. Heating Point Parameters: Heating Point Location: Stationary contact.
[0057] Scene classification: The model identifies it as "switch cabinet contact scene".
[0058] Temperature tag adaptive configuration Installation structure: The "snap-on mounting part" is selected, which uses its elastic alloy material to directly snap and fix it to the guide rod of the contact, without the need for drilling or welding.
[0059] Insulation and shielding: 10KV rating, 3mm thick insulation layer. Low electromagnetic interference, using ordinary nano-silver shielding layer, antenna gain 3dBi.
[0060] Dynamic optimization of data acquisition communication parameters Communication quality was good from the start (RSSI = -60dBm, success rate 99.5%). To save energy, the system automatically reduced the RF power slightly from the initial 15dBm to 12dBm and extended the tag signal transmission cycle to 3 seconds.
[0061] Temperature measurement data differentiation processing Threshold model: Apply the "switch cabinet contact scenario (10KV)" model: warning threshold 90℃, alarm threshold 100℃.
[0062] After a short circuit fault was closed, the temperature spiked instantly. Even after smoothing, the average temperature still quickly reached 105°C, immediately triggering an alarm.
[0063] Remote monitoring and feedback optimization The platform issued an audible and visual alarm, indicating the exact location of the fault. Subsequent data analysis optimized the model's response speed to instantaneous high-current heating.
[0064] Example 4: Low-voltage switchgear cable scenario (0.4KV) This embodiment focuses on the cable outlet of a low-voltage distribution cabinet in a commercial building's power distribution room, characterized by dense cables and frequent temperature fluctuations.
[0065] Scene feature collection and classification Data Acquisition Parameters: Equipment Type: Low-voltage switchgear; High-voltage level: 0.4KV. Environmental Parameters: Dense cabling; Electromagnetic interference intensity: 45 dBμV / m; Metallic environment distribution: Mainly cabinet and cable armor. Heat Point Parameters: Heat point location: T-junction of phase B main cable.
[0066] Scene classification: The model identifies it as "low-voltage cabinet cable scene".
[0067] Temperature tag adaptive configuration Installation structure: Use "anti-metal temperature measurement tags" with insulating adhesive on the back, and use nylon cable ties to tightly bind them to the cable surface.
[0068] Insulation and shielding: Low grade, the insulation layer mainly serves a physical protection function, with a thickness of 1.5mm. Weak interference, using a common shielding layer.
[0069] Temperature measurement data differentiation processing Threshold model: Apply the "low-voltage cabinet cable scenario (0.4KV)" model: warning threshold 80℃, alarm threshold 90℃.
[0070] Due to load variations, the temperature fluctuates frequently between 55℃ and 75℃. The system employs a moving average filter with n=5, combined with trend analysis (such as a continuous rise in the average temperature over multiple consecutive periods) to identify anomalies, effectively avoiding false alarms caused by normal fluctuations.
[0071] Feedback optimization Long-term data shows that the normal summer temperature at this point is higher than normal. The system automatically fine-tunes the summer warning threshold to 85℃, which is more in line with the actual operating conditions.
[0072] Example 5: Transformer bushing scenario for electric arc furnace in a metallurgical plant (110KV) This embodiment is designed for the extremely harsh environment of the metallurgical industry, characterized by strong electromagnetic interference, high temperature, and high dust.
[0073] Scene feature collection and classification Data Acquisition Parameters: Equipment Type: Electric Arc Furnace Transformer; High Voltage Level: 110KV. Environmental Parameters: Extremely high electromagnetic interference intensity, reaching 115 dBμV / m; High ambient temperature. Heating Point Parameters: Heating Point Location: High-voltage bushing guide rod.
[0074] Scene classification: The model identifies it as "Transformer busbar scene (extreme interference subclass)".
[0075] Temperature tag adaptive configuration Insulation and shielding: Class 110KV, insulation layer thickness 8mm. Extremely strong electromagnetic interference; employs a "nano-silver-carbon fiber composite shielding layer"; antenna gain configured to a maximum of 6dBi.
[0076] Dynamic optimization of data acquisition communication parameters Initial challenge: Initial communication is completely interrupted.
[0077] Optimization process: The system decisively increased the RF power to the maximum value of 30dBm and simultaneously activated 4 antennas to attempt communication from different directions. After multiple attempts and closed-loop adjustments, the optimal antenna combination and orientation were finally found, and the tag transmission period was set to the shortest 0.5 seconds. Data was successfully captured in the gaps of strong interference, and the success rate was barely maintained at the target of 98%.
[0078] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.
Claims
1. A multi-scenario adaptive passive wireless temperature measurement method for power equipment based on RFID, characterized in that, Includes the following steps: S1. Scene Feature Acquisition and Classification: Acquire scene parameters of the target power equipment; input the scene parameters into a pre-trained scene classification model and output the corresponding scene type; S2, Adaptive Configuration of Temperature Measurement Tag: Based on the scene type identified by S1, select the corresponding installation structure for the RFID passive temperature measurement tag, configure the thickness of the high voltage resistant insulation layer, select the material of the anti-metal shielding layer, and set the gain of the loop antenna to achieve precise adaptation of the tag structure and electrical parameters to the specific scene. S3. Dynamic optimization of communication parameters of the data collector: The data collector periodically sends a detection signal to the temperature measurement tag through its antenna module and obtains the signal strength indicator RSSI and data transmission success rate in real time; when the RSSI is lower than a preset first threshold or the data transmission success rate is lower than a preset second threshold, the dynamic adjustment strategy of communication parameters is triggered. After adjustments, the communication quality is reassessed until the preset communication quality target is met, thus forming a closed-loop optimization. S4. Differentiated processing of temperature measurement data: After preprocessing, the temperature data collected by the temperature measurement tag is compared with the temperature threshold model of the corresponding scenario type. The temperature threshold model defines the normal operating temperature range, the warning temperature threshold, and the alarm temperature threshold. Based on the comparison result, it is determined whether the device is in a normal, warning, or alarm state, and the corresponding data recording frequency and alarm response mechanism are triggered. S5. Remote monitoring and feedback optimization: The remote monitoring platform receives and stores temperature data, equipment status and communication parameters. It uses historical data to establish a temperature trend prediction model through the Long Short-Term Memory (LSTM) algorithm to predict future temperature changes. It also collects on-site feedback data periodically and iteratively optimizes the scenario classification model, temperature threshold model and communication parameter adjustment strategy.
2. The RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment according to claim 1, characterized in that, In step S1: The scenario parameters include basic equipment parameters, environmental parameters, and heat point parameters; The scenario types include at least ring main unit plug scenario, switch cabinet contact scenario, transformer busbar scenario, and low-voltage cabinet cable scenario.
3. The RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment according to claim 2, characterized in that: The basic parameters of the equipment include the equipment type and high voltage level. The environmental parameters include installation space dimensions, electromagnetic interference intensity, and distribution of metallic environment. The parameters of the heating point include the location of the heating point, historical heating fault records, and normal operating temperature range.
4. The RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment according to claim 1, characterized in that, In step S2, the selection of the installation structure is specifically as follows: For ring main unit plugging scenarios, select a plug-type installation part with an inner diameter specification of M12-1.25; For switchgear contact applications, a snap-on mounting part made of elastic alloy material is selected; For transformer busbar scenarios, a hook-type mounting part that is fastened with bolts is selected; For low-voltage switchgear cable scenarios, choose anti-metal temperature measurement tags that can be secured with cable ties. The configuration of the high-voltage resistant insulation layer thickness is based on the high-voltage level of the scenario: When the high voltage level is ≤10KV, the insulation layer thickness is 3mm; When 10KV < high voltage level ≤ 35KV, the insulation layer thickness is 5mm; When the high voltage level is >35KV, the insulation layer thickness is 8mm; The configuration of the anti-metal shielding layer material and the loop antenna gain is based on the electromagnetic interference intensity and the distribution of the metallic environment: When the electromagnetic interference intensity is ≤60dBμV / m and the metal components are sparse, a common nano-silver shielding layer is used, and the antenna gain is 3dBi. When the electromagnetic interference intensity is less than 60dBμV / m and less than 100dBμV / m and the metal components are relatively dense, a nano-silver enhanced shielding layer is used, and the antenna gain is 4-5dBi. When the electromagnetic interference intensity is >100dBμV / m and the metal components are densely packed, a nano-silver-carbon fiber composite shielding layer is used, and the antenna gain is 6dBi.
5. The RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment according to claim 1, characterized in that, The communication parameter dynamic adjustment strategy in step S3 specifically includes: RF output power adjustment: The power is adjusted in stages from 5dBm to 30dBm. When -80dBm < RSSI ≤ -70dBm, the power is increased by 5dBm. When -90dBm < RSSI ≤ -80dBm, increase the power by 10dBm; When RSSI ≤ -90dBm, the power will be increased to the maximum value of 30dBm; Antenna access quantity adjustment: Initially connect 1 antenna. If the communication quality still does not meet the standard after adjusting the power, gradually increase the number of antennas, up to a maximum of 4, and prioritize adding antennas in areas with weak signal coverage. Tag signal transmission cycle adjustment: dynamically adjusted within the range of 0.5 seconds to 5 seconds. When the signal interference is small, a long cycle of 3-5 seconds is used to reduce power consumption, and when the signal interference is large, a short cycle of 0.5-1 seconds is used to improve the probability of data reception. The communication quality target is RSSI ≥ -70dBm and data transmission success rate ≥ 98%.
6. The RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment according to claim 1, characterized in that, In step S4, preprocessing includes outlier removal and fluctuation smoothing. Abnormal data removal involves directly removing data that exceeds the operating range of the temperature sensor on the temperature measurement tag. Fluctuation data smoothing employs a moving average filtering algorithm, the mathematical expression of which is: ,in This represents the average temperature value calculated after filtering, in degrees Celsius; n represents the size of the sliding window, ranging from 5 to 10 data points. This represents the i-th raw temperature data value within the sliding window, in degrees Celsius.
7. The RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment according to claim 1, characterized in that, In step S4, the temperature threshold model sets different thresholds for different scene types, as follows: For the ring main unit plugging scenario (12KV), the normal operating temperature range is -40℃ to +85℃, the warning threshold is set to 90℃, and the alarm threshold is set to 100℃. For switchgear contact scenarios (35KV), the normal operating temperature range is -40℃ to +90℃, the warning threshold is set to 95℃, and the alarm threshold is set to 105℃. For transformer busbar scenarios (110KV), the normal operating temperature range is -40℃ to +95℃, the warning threshold is set to 100℃, and the alarm threshold is set to 110℃. For low-voltage switchgear cable scenarios (0.4KV), the normal operating temperature range is -40℃ to +75℃, the warning threshold is set to 80℃, and the alarm threshold is set to 90℃.
8. The RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment according to claim 1, characterized in that, In step S5, the temperature trend prediction model is constructed using the Long Short-Term Memory (LSTM) network algorithm. The input of the model is the temperature data of a past time series, including historical temperature data of the last 1 hour, 3 hours and 24 hours, and the output is the temperature prediction value for the next 1 hour and 3 hours. Its iterative optimization is carried out regularly. The optimization is based on factors including the error between the temperature measurement data collected on site and the data calibrated manually, the accuracy of the early warning and alarm signals, and the changes in scene parameters. The internal weight parameters of the scene classification model and the temperature threshold model are adjusted through optimization algorithms, and the trigger threshold and adjustment range of the communication parameter adjustment strategy are fine-tuned.
9. A system for implementing the method according to any one of claims 1 to 8, characterized in that, include: Multiple RFID passive temperature measurement tags, whose installation structure, insulation layer thickness, shielding layer material and antenna gain can be configured according to instructions or preset rules; At least one data acquisition unit, with a built-in antenna module and processor, is used to perform dynamic optimization steps for the data acquisition unit's communication parameters and to communicate with the temperature measurement tag and the remote monitoring platform; The remote monitoring platform is equipped with a scene classification model, a temperature threshold model, a temperature trend prediction model, and a feedback optimization algorithm. It is used to perform the scene feature acquisition and classification, temperature measurement data differentiation processing, and remote monitoring and feedback optimization steps, and to send configuration instructions to the data collector.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the RFID-based multi-scenario adaptive passive wireless temperature measurement method for power equipment as described in any one of claims 1 to 8.