Bus duct temperature field real-time monitoring and early warning method based on multi-sensor fusion

Abstract

The invention discloses a bus duct temperature field real-time monitoring and early warning method based on multi-sensor fusion. The method comprises the following steps: step 1, arranging and initializing a multi-type sensor array; 2, collecting and preprocessing data of a multi-source sensor; 3, reconstructing a temperature field based on hierarchical fusion; 4, calculating a dynamic early warning threshold value; 5, temperature field monitoring and grading early warning are carried out; step 6, verification and data updating after early warning; multi-type sensor array three-dimensional arrangement is adopted, a hierarchical fusion algorithm is combined, comprehensive and accurate reconstruction of a bus duct temperature field is achieved, the problems that a single sensor is limited in monitoring range and poor in data reliability are solved, the temperature monitoring precision is improved to be within + / -0.2 DEG C, and a local overheating area can be accurately positioned; the dynamic early warning threshold value is calculated based on the environment temperature and the load current, the defects that a fixed threshold value is prone to false early warning and early warning omission are overcome, and monitoring requirements under different operation working conditions can be met.

Description

Technical Field

[0001] This invention belongs to the field of power equipment monitoring technology, specifically a method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion. Background Technology

[0002] Busbar trunking, as a key device for transmitting electrical energy in power systems, is widely used in large-scale power applications such as high-rise buildings and industrial plants. During operation, factors such as increased contact resistance, overload operation, and poor ventilation can easily lead to localized overheating. If not monitored and warned in time, this may cause serious safety accidents such as fires and power outages.

[0003] Existing methods for monitoring busbar temperature mostly employ a single type of sensor (such as thermocouples or infrared sensors) for single-point monitoring, which has the following drawbacks:

[0004] First, the monitoring range is limited and cannot fully reflect the overall temperature field distribution of the busbar trunking, which may lead to the omission of potential local overheating hazards.

[0005] Secondly, it has weak anti-interference ability. In the electromagnetic interference, dust and vibration environment of industrial sites, the monitoring data has large errors and insufficient reliability.

[0006] Third, the warning threshold is fixed and does not take into account the impact of dynamic factors such as ambient temperature and load current on the normal operating temperature of the busbar trunking, which can easily lead to false or missed warnings.

[0007] To address these issues, some technologies have attempted to employ multi-sensor deployments. However, these methods merely calculate average values ​​or compare thresholds of the monitoring data, failing to achieve deep data fusion and fully leverage the advantages of different sensor types. Consequently, monitoring accuracy and early warning precision still need improvement. Therefore, developing a method capable of comprehensively and accurately monitoring the temperature field of busbar trunking and providing reliable early warnings has become an urgent need in the field of power equipment safety monitoring. Summary of the Invention

[0008] In view of the above situation and to overcome the shortcomings of the prior art, the present invention provides a method for real-time monitoring and early warning of busbar temperature field by multi-sensor fusion, which effectively solves the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion, comprising the following steps:

[0010] Step 1: Deployment and Initialization of Multi-Type Sensor Arrays

[0011] A three-dimensional sensor array consisting of thermocouple sensors, fiber Bragg grating sensors, and infrared temperature sensors is deployed in the key monitoring areas of the busbar trunking. Specifically, two thermocouple sensors and two fiber Bragg grating sensors are deployed at each of the conductor connection points and branch joints of the busbar trunking, with a spacing of 5 cm. Three infrared temperature sensors are deployed per square meter on the surface of the busbar trunking shell, with the sensor probes at a vertical distance of 8 to 12 cm from the shell surface. One ambient temperature sensor and one current sensor are deployed in the environment in which the busbar trunking is located. The ambient temperature sensor is deployed 1 to 1.5 meters above the busbar trunking, and the current sensor is connected in series at the busbar trunking inlet. All sensors are initialized and calibrated, and the sensor data acquisition frequency is set to 10 Hz. Data transmission adopts a dual-mode redundant transmission method of RS-485 bus and wireless WiFi.

[0012] Step 2: Multi-source sensor data acquisition and preprocessing

[0013] The sensor array collects temperature data, ambient temperature data, and busbar load current data at each monitoring point of the busbar trunking in real time, with a collection period of 100 milliseconds. The Grubbs criterion is used to remove abnormal data, the temperature data of the same monitoring point is smoothed, the infrared temperature measurement data is distance corrected, and all preprocessed data is uniformly converted into Celsius units, with timestamps accurate to the millisecond level.

[0014] Step 3: Temperature field reconstruction based on hierarchical fusion

[0015] A hierarchical fusion algorithm is used to fuse the preprocessed multi-source data. The first layer is the fusion of data from the same type of sensor, the second layer is the fusion of data from different types of sensors, and the third layer is the reconstruction of the three-dimensional temperature field of the busbar based on the fused data using the Kriging interpolation algorithm.

[0016] Step 4: Calculation of Dynamic Early Warning Threshold

[0017] A baseline temperature model is established based on ambient temperature data and load current data. Level 1, Level 2, and Level 3 warning thresholds are set, and the thresholds are verified and corrected by combining historical busbar operation data.

[0018] Step 5: Temperature Field Monitoring and Graded Early Warning

[0019] The fused temperature values ​​of each monitoring point in the reconstructed temperature field are compared with the dynamic early warning threshold in real time, and the corresponding level of early warning measures are activated based on the comparison results.

[0020] Step Six: Verification and Data Update After the Warning

[0021] After troubleshooting, staff recorded the results, verified the monitoring data, stored the relevant data for this warning in the database, and updated the historical operation database.

[0022] Preferably, the specific operations for sensor initialization calibration in step one include: calibrating the thermocouple sensor in a 0°C ice-water mixture with an error correction range of ±0.1°C; calibrating the fiber optic grating sensor in a 25°C standard environment with a wavelength drift correction value of 0.02 nm; calibrating the infrared temperature sensor aligned with a 25°C standard blackbody with a measurement error controlled within ±0.3°C; and calibrating the ambient temperature sensor and current sensor using a standard temperature source and a standard current source, respectively, with the current measurement error controlled within ±0.5 amperes.

[0023] Preferably, the specific operations of data preprocessing in step two include: when the data deviation exceeds 3 times the standard deviation, it is judged as abnormal data and removed, and supplementary data is collected for the monitoring point; the temperature data of the same monitoring point collected by the thermocouple sensor and the fiber optic grating sensor are smoothed by the moving average algorithm, and the sliding window size is set to 5 data points; the shell temperature data collected by the infrared temperature sensor is corrected for distance using the formula "corrected temperature = measured temperature - 0.05 × (actual distance - 10)", where the actual distance is in centimeters.

[0024] Preferably, the specific operations of layered fusion in step three include:

[0025] First layer: Weighted average fusion is used for multiple thermocouple sensor data in the same area, with the weight of sensors with small calibration errors set to 0.6 and those with large errors set to 0.4; Median fusion is used for multiple fiber Bragg grating sensor data in the same area; Regional mean fusion is used for infrared temperature sensor data, dividing the busbar casing into multiple 10 cm × 10 cm areas and calculating the mean of all infrared sensor data in each area.

[0026] The second layer: Based on the DS evidence theory, thermocouple fusion data and fiber Bragg grating fusion data of the same monitoring location are fused, with the confidence level of thermocouple data set at 0.8 and the confidence level of fiber Bragg grating data set at 0.9; the fused conductor temperature data is correlated and fused with the infrared shell temperature data of the corresponding area, with the mapping coefficient of copper busbar trunking set at 0.92 and the mapping coefficient of aluminum busbar trunking set at 0.88;

[0027] The third layer: The three-dimensional temperature field of the busbar is reconstructed using the Kriging interpolation algorithm, and the interpolation grid accuracy is set to 2 cm × 2 cm × 2 cm.

[0028] Preferably, the specific operations for calculating the dynamic early warning threshold in step four include:

[0029] S1. Establish a reference temperature model: Reference temperature = 25 degrees Celsius + 0.01 × load current, where the load current is in amperes; when the ambient temperature deviates from 25 degrees Celsius, the correction is 0.8 × (ambient temperature - 25 degrees Celsius).

[0030] S2. Set warning levels and corresponding thresholds: Level 1 warning threshold = base temperature + 30 degrees Celsius; Level 2 warning threshold = base temperature + 45 degrees Celsius; Level 3 warning threshold = base temperature + 60 degrees Celsius; When the load current exceeds 80% of the rated current, the warning threshold for each level is lowered by 5 degrees Celsius; When the ambient temperature exceeds 35 degrees Celsius, the warning threshold for each level is further lowered by 3 degrees Celsius.

[0031] S3. Threshold Verification and Correction: If the actual temperature is close to the threshold but no fault occurs for three consecutive times, the corresponding warning threshold will be raised by 2 degrees Celsius; if the temperature does not reach the threshold but abnormal overheating occurs, the corresponding warning threshold will be lowered by 3 degrees Celsius.

[0032] Preferably, the specific operations for tiered early warning in step five include:

[0033] When the temperature at the monitoring point reaches the Level 1 warning threshold, the Level 1 warning is activated: a yellow warning signal is issued, warning data is recorded, and staff are reminded to conduct inspections via a local audible and visual alarm. The inspection cycle is shortened to once every 30 minutes.

[0034] When the temperature at the monitoring point reaches the level 2 warning threshold, the level 2 warning is activated: an orange warning signal is issued, and in addition to performing the level 1 warning operation, a warning message is sent to the operation and maintenance management platform, and an inspection work order is automatically generated, requiring staff to arrive at the site within 15 minutes to troubleshoot the fault.

[0035] When the temperature at the monitoring point reaches the Level 3 warning threshold, the Level 3 warning is activated: a red warning signal is issued, and in addition to performing the Level 2 warning operation, the power control system is linked to issue a trip warning suggestion and activate the emergency cooling device until the temperature drops below the Level 2 warning threshold.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] 1. This invention employs a three-dimensional array of multiple types of sensors, combined with a hierarchical fusion algorithm, to achieve a comprehensive and accurate reconstruction of the temperature field of the busbar trunking. This solves the problems of limited monitoring range and poor data reliability of a single sensor, improving the temperature monitoring accuracy to within ±0.2 degrees Celsius, and enabling precise location of local overheated areas.

[0038] 2. This invention calculates dynamic early warning thresholds based on ambient temperature and load current, overcoming the shortcomings of fixed thresholds that are prone to false and missed early warnings. The accuracy of early warnings is improved by more than 80%, and it can adapt to the monitoring needs under different operating conditions.

[0039] 3. This invention employs multiple safeguards such as dual-mode redundant transmission, abnormal data removal, and smoothing processing to improve the reliability and stability of data transmission. It can still work stably in complex industrial environments such as electromagnetic interference and vibration, and has strong adaptability.

[0040] 4. This invention sets up a graded early warning mechanism and links it with emergency response measures, which can take targeted countermeasures according to the severity of overheating. At the same time, it optimizes the model through data updates to form a closed-loop management system, which effectively improves the operational safety of busbar trunking and reduces the accident rate. Attached Figure Description

[0041] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0042] In the attached diagram:

[0043] Figure 1 This is a schematic diagram of a method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion according to the present invention. Detailed Implementation

[0044] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0045] Depend on Figure 1 This invention relates to a method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion, comprising the following steps:

[0046] Step 1: Deployment and Initialization of Multi-Type Sensor Arrays

[0047] A three-dimensional sensor array consisting of thermocouple sensors, fiber Bragg grating sensors, and infrared temperature sensors is deployed in the key monitoring areas of the busbar trunking. Specifically, two thermocouple sensors and two fiber Bragg grating sensors are deployed at each of the conductor connection points and branch joints of the busbar trunking, with a spacing of 5 cm between them. Three infrared temperature sensors are deployed per square meter on the surface of the busbar trunking shell, with the sensor probes at a vertical distance of 8 to 12 cm from the shell surface. One ambient temperature sensor and one current sensor are deployed in the environment in which the busbar trunking is located. The ambient temperature sensor is deployed 1 to 1.5 meters above the busbar trunking, and the current sensor is connected in series at the inlet end of the busbar trunking.

[0048] Perform initial calibration on all sensors: calibrate the thermocouple sensor in a 0°C ice-water mixture with an error correction range of ±0.1°C; calibrate the fiber optic grating sensor in a 25°C standard environment with a wavelength drift correction of 0.02 nm; calibrate the infrared temperature sensor aligned with a 25°C standard blackbody with a measurement error controlled within ±0.3°C; calibrate the ambient temperature sensor and current sensor using standard temperature and current sources respectively, with current measurement errors controlled within ±0.5 amperes; after initialization, set the sensor data acquisition frequency to 10 Hz, and use a dual-mode redundant transmission method of RS485 bus and wireless WiFi for data transmission.

[0049] Step 2: Multi-source sensor data acquisition and preprocessing

[0050] The sensor array collects temperature data, ambient temperature data, and busbar load current data at each monitoring point in the busbar trunking in real time, with a collection period of 100 milliseconds.

[0051] The collected multi-source data underwent preprocessing: Outlier data was removed using the Grubbs criterion; data with a deviation exceeding three times the standard deviation were identified as outlier and removed, and data for that monitoring point was supplemented. Temperature data from the same monitoring point collected by thermocouple and fiber optic sensors were smoothed using a moving average algorithm with a sliding window size of five data points. Temperature data from the outer casing collected by the infrared temperature sensor was corrected for distance; the measured value was adjusted based on the actual distance between the sensor and the casing using the following formula: Corrected temperature = Measured temperature - 0.05 × (Actual distance - 10), where the actual distance is in centimeters. All preprocessed data were converted to degrees Celsius, with timestamps accurate to milliseconds.

[0052] Step 3: Temperature field reconstruction based on hierarchical fusion

[0053] A hierarchical fusion algorithm is used to fuse preprocessed multi-source data to achieve accurate reconstruction of the busbar temperature field.

[0054] First layer: Data fusion of similar sensors. Multiple thermocouple sensor data from the same area are fused using a weighted average method. The weights are allocated based on the sensor calibration accuracy, with sensors with smaller calibration errors having a weight of 0.6 and those with larger errors having a weight of 0.4. Multiple fiber Bragg grating sensor data from the same area are fused using the median value to eliminate the influence of extreme values. Infrared temperature sensor data are fused using the regional mean value. The busbar casing is divided into multiple 10cm × 10cm regions, and the mean value of all infrared sensor data within each region is calculated.

[0055] The second layer: Cross-type sensor data fusion. Based on DS evidence theory, thermocouple fusion data and fiber Bragg grating fusion data from the same monitoring location are fused. The confidence level of thermocouple data is set at 0.8, and the confidence level of fiber Bragg grating data is set at 0.9. The fused temperature value is calculated through evidence combination rules. The fused conductor temperature data is then correlated with the infrared shell temperature data of the corresponding area to establish a mapping relationship between conductor temperature and shell temperature. The mapping coefficient is determined according to the busbar material: the mapping coefficient is 0.92 for copper busbars and 0.88 for aluminum busbars.

[0056] The third layer: Temperature field reconstruction. Based on the fused temperature data from each monitoring point, the three-dimensional temperature field of the busbar is reconstructed using the Kriging interpolation algorithm. The interpolation grid accuracy is set to 2 cm × 2 cm × 2 cm, generating a temperature field distribution cloud map to clearly identify the location of the highest temperature point and the temperature gradient distribution.

[0057] Step 4: Calculation of Dynamic Early Warning Threshold

[0058] The dynamic warning threshold is calculated based on ambient temperature data and load current data. The specific steps are as follows:

[0059] S1. Establish a reference temperature model: Reference temperature = 25 degrees Celsius + 0.01 × load current, where the load current is in amperes; when the ambient temperature deviates from 25 degrees Celsius, the reference temperature is corrected by 0.8 × (ambient temperature - 25 degrees Celsius).

[0060] S2. Set warning levels and corresponding thresholds: Level 1 warning threshold = base temperature + 30 degrees Celsius; Level 2 warning threshold = base temperature + 45 degrees Celsius; Level 3 warning threshold = base temperature + 60 degrees Celsius. When the load current exceeds 80% of the rated current, the warning threshold for each level is lowered by 5 degrees Celsius; when the ambient temperature exceeds 35 degrees Celsius, the warning threshold for each level is further lowered by 3 degrees Celsius.

[0061] S3. Threshold Verification and Correction: Based on the historical operating data of the bus trunking, the calculated dynamic threshold is verified. If the actual temperature is close to the threshold but no fault occurs for three consecutive times, the corresponding warning threshold will be raised by 2 degrees Celsius. If the temperature does not reach the threshold but abnormal overheating occurs, the corresponding warning threshold will be lowered by 3 degrees Celsius.

[0062] Step 5: Temperature Field Monitoring and Graded Early Warning

[0063] The fused temperature values ​​of each monitoring point in the reconstructed temperature field of step three are compared in real time with the dynamic early warning threshold calculated in step four:

[0064] When the temperature at the monitoring point reaches the first-level warning threshold, the first-level warning is activated: a yellow warning signal is issued, the warning time, location, and corresponding ambient temperature and load current data are recorded, and the staff are reminded to conduct inspections through a local audible and visual alarm, with the inspection cycle shortened to once every 30 minutes.

[0065] When the temperature at the monitoring point reaches the level 2 warning threshold, the level 2 warning is activated: an orange warning signal is issued, and in addition to performing the level 1 warning operation, a warning message is sent to the operation and maintenance management platform, including a temperature field distribution cloud map and abnormal area location information. An inspection work order is automatically generated, requiring staff to arrive at the site within 15 minutes to troubleshoot the fault.

[0066] When the temperature at the monitoring point reaches the Level 3 warning threshold, the Level 3 warning is activated: a red warning signal is issued, and in addition to performing the Level 2 warning operation, the power control system is linked to issue a trip warning suggestion, reminding staff to take emergency power-off measures within 5 minutes. At the same time, the emergency cooling device is activated until the temperature drops below the Level 2 warning threshold.

[0067] Step Six: Verification and Data Update After the Warning

[0068] After troubleshooting based on the warning information, staff record the cause of the fault and the handling results, and verify the monitoring data: if the temperature returns to below the reference temperature after troubleshooting, the warning is deemed valid; if no fault is found but the temperature remains high, the working status of the sensor array is rechecked, and abnormal sensors are replaced and calibrated.

[0069] All data related to this early warning (including monitoring data, early warning threshold, early warning level, and fault handling results) will be stored in the database, and the historical operation database of the bus trunking will be updated to provide data support for the subsequent optimization of dynamic early warning thresholds and the improvement of temperature field models. The update cycle is within 24 hours after each early warning is handled.

[0070] Example: Taking the monitoring of a copper busbar trunking in an industrial plant as an example, the rated current of the busbar trunking is 1000 amperes. The method of this invention is used for real-time monitoring and early warning of the temperature field. The specific steps are as follows:

[0071] Step 1: Deployment and Initialization of Multi-Type Sensor Arrays

[0072] Two thermocouple sensors and two fiber optic grating sensors are installed at each of the three conductor connection points and two branch joint points of the busbar trunking, with a spacing of 5 cm. The surface area of ​​the busbar trunking shell is 10 square meters, and 30 infrared temperature sensors are installed there, with the probes 10 cm vertically away from the shell. An ambient temperature sensor is installed 1.2 meters above the busbar trunking, and a current sensor is connected in series at the inlet end.

[0073] All sensors were calibrated. Thermocouple sensors were calibrated in a 0°C ice-water mixture with an error correction of ±0.1°C; fiber optic grating sensors were calibrated in a 25°C standard environment with a wavelength drift correction of 0.02 nm; infrared temperature sensors were calibrated against a 25°C standard blackbody with a measurement error controlled within ±0.3°C; ambient temperature and current sensors were calibrated using standard temperature and current sources, respectively, with a current measurement error of ±0.5 amperes. The data acquisition frequency was set to 10 Hz, employing dual-mode transmission via RS-485 bus and wireless WiFi.

[0074] Step 2: Multi-source sensor data acquisition and preprocessing

[0075] Data was collected at 100-millisecond intervals. At a certain moment, the thermocouple data for connection point 1 was 58.2°C and 58.5°C, the fiber optic grating data was 58.3°C and 58.4°C, the infrared casing temperature was 45.6°C, the ambient temperature was 30°C, and the load current was 850 amperes. Outliers were removed using the Grubbs criterion, and no outliers were found in this data set. A moving average (window size 5) was applied to the thermocouple data, yielding a value of 58.35°C. The infrared data, due to its 10-centimeter distance, required no correction. All data were converted to degrees Celsius, and the timestamps were accurate to milliseconds.

[0076] Step 3: Temperature field reconstruction based on hierarchical fusion

[0077] The first layer of fusion: The weighted average of thermocouple data (weights 0.6 and 0.4) yields 58.3 degrees Celsius, and the median fusion of fiber Bragg grating data yields 58.35 degrees Celsius. The second layer of fusion: Based on DS evidence theory, thermocouple and fiber Bragg grating data are fused to obtain a fused temperature of 58.33 degrees Celsius. The temperature mapping coefficient between the copper busbar conductor and the outer shell is 0.92, and the associated infrared outer shell temperature is 45.6 degrees Celsius, verifying the rationality of the fusion. The third layer of fusion: Using the Kriging interpolation algorithm, a three-dimensional temperature field is reconstructed with a 2 cm × 2 cm × 2 cm grid to generate a temperature field cloud map, determining that the connection point is the current highest temperature point.

[0078] Step 4: Calculation of Dynamic Early Warning Threshold

[0079] The reference temperature is 25 + 0.01 × 850 = 33.5 degrees Celsius, the ambient temperature is 30 degrees Celsius, the correction is 0.8 × (30 - 25) = 4 degrees Celsius, and the corrected reference temperature is 33.5 + 4 = 37.5 degrees Celsius. The load current of 850 amps exceeds 80% of the rated current (800 amps), so the warning thresholds at all levels are lowered by 5 degrees Celsius. The ambient temperature of 30 degrees Celsius does not exceed 35 degrees Celsius, so no further adjustment is needed.

[0080] The first-level warning threshold = 37.5 + 30 - 5 = 62.5 degrees Celsius;

[0081] The level 2 warning threshold = 37.5 + 45 - 5 = 77.5 degrees Celsius;

[0082] The threshold for a Level 3 warning is 37.5 + 60 - 5 = 92.5 degrees Celsius.

[0083] Step 5: Temperature Field Monitoring and Graded Early Warning

[0084] The current fusion temperature of 58.33 degrees Celsius is lower than the first-level warning threshold of 62.5 degrees Celsius, so there is no warning and monitoring continues. If the load current rises to 900 amps and the ambient temperature rises to 36 degrees Celsius, the base temperature is recalculated as 25 + 0.01 × 900 + 0.8 × (36 - 25) = 43.8 degrees Celsius. The warning threshold is lowered by 5 + 3 = 8 degrees Celsius, and the first-level warning threshold becomes 43.8 + 30 - 8 = 65.8 degrees Celsius. If the monitored temperature rises to 66 degrees Celsius at this time, the first-level warning is activated, a yellow warning signal is issued, relevant data is recorded, and the inspection cycle is shortened to 30 minutes.

[0085] Step Six: Verification and Data Update After the Warning

[0086] Staff found no faults during their inspection and determined that the temperature rise was a normal result of increased load. The results were recorded. The monitoring data, warning thresholds, and results were stored in the database, the historical operation database was updated, and the subsequent threshold calculation model was optimized.

[0087] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0088] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion, comprising the following steps: Step 1: Deployment and Initialization of Multi-Type Sensor Arrays A three-dimensional sensor array consisting of thermocouple sensors, fiber Bragg grating sensors, and infrared temperature sensors is deployed in the key monitoring areas of the busbar trunking. Specifically, two thermocouple sensors and two fiber Bragg grating sensors are deployed at each of the conductor connection points and branch joints of the busbar trunking, with a spacing of 5 cm. Three infrared temperature sensors are deployed per square meter on the surface of the busbar trunking shell, with the sensor probes at a vertical distance of 8 to 12 cm from the shell surface. One ambient temperature sensor and one current sensor are deployed in the environment in which the busbar trunking is located. The ambient temperature sensor is deployed 1 to 1.5 meters above the busbar trunking, and the current sensor is connected in series at the busbar trunking inlet. All sensors are initialized and calibrated, and the sensor data acquisition frequency is set to 10 Hz. Data transmission adopts a dual-mode redundant transmission method of RS-485 bus and wireless WiFi. Step 2: Multi-source sensor data acquisition and preprocessing The sensor array collects temperature data, ambient temperature data, and busbar load current data at each monitoring point of the busbar trunking in real time, with a collection period of 100 milliseconds. The Grubbs criterion is used to remove abnormal data, the temperature data of the same monitoring point is smoothed, the infrared temperature measurement data is distance corrected, and all preprocessed data is uniformly converted into Celsius units, with timestamps accurate to the millisecond level. Step 3: Temperature field reconstruction based on hierarchical fusion A hierarchical fusion algorithm is used to fuse the preprocessed multi-source data. The first layer is the fusion of data from the same type of sensor, the second layer is the fusion of data from different types of sensors, and the third layer is the reconstruction of the three-dimensional temperature field of the busbar based on the fused data using the Kriging interpolation algorithm. Step 4: Calculation of Dynamic Early Warning Threshold A baseline temperature model is established based on ambient temperature data and load current data. Level 1, Level 2, and Level 3 warning thresholds are set, and the thresholds are verified and corrected by combining historical busbar operation data. Step 5: Temperature Field Monitoring and Graded Early Warning The fused temperature values ​​of each monitoring point in the reconstructed temperature field are compared with the dynamic early warning threshold in real time, and the corresponding level of early warning measures are activated based on the comparison results. Step Six: Verification and Data Update After the Warning After troubleshooting, staff recorded the results, verified the monitoring data, stored the relevant data for this warning in the database, and updated the historical operation database.

2. The method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion as described in claim 1, characterized in that: The specific operations for sensor initialization calibration in step one include: calibrating the thermocouple sensor in a 0°C ice-water mixture with an error correction range of ±0.1°C; calibrating the fiber optic grating sensor in a 25°C standard environment with a wavelength drift correction of 0.02 nm; calibrating the infrared temperature sensor aligned with a 25°C standard blackbody with a measurement error controlled within ±0.3°C; and calibrating the ambient temperature sensor and current sensor using standard temperature and current sources respectively, with the current measurement error controlled within ±0.5 amperes.

3. The method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion as described in claim 1, characterized in that: The specific operations of data preprocessing in step two include: when the data deviation exceeds 3 times the standard deviation, it is judged as abnormal data and removed, and supplementary data is collected for the monitoring point; the temperature data of the same monitoring point collected by thermocouple sensor and fiber optic grating sensor are smoothed by moving average algorithm, and the sliding window size is set to 5 data points; the shell temperature data collected by infrared temperature sensor is corrected for distance using the formula "corrected temperature = measured temperature - 0.05 × (actual distance - 10)", where the actual distance is in centimeters.

4. The method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion as described in claim 1, characterized in that: The specific operations of layered fusion in step three include: First layer: Weighted average fusion is used for multiple thermocouple sensor data in the same area, with the weight of sensors with small calibration errors set to 0.6 and those with large errors set to 0.4; Median fusion is used for multiple fiber Bragg grating sensor data in the same area; Regional mean fusion is used for infrared temperature sensor data, dividing the busbar casing into multiple 10 cm × 10 cm areas and calculating the mean of all infrared sensor data in each area. The second layer: Based on the DS evidence theory, thermocouple fusion data and fiber Bragg grating fusion data of the same monitoring location are fused, with the confidence level of thermocouple data set at 0.8 and the confidence level of fiber Bragg grating data set at 0.9; the fused conductor temperature data is correlated and fused with the infrared shell temperature data of the corresponding area, with the mapping coefficient of copper busbar trunking set at 0.92 and the mapping coefficient of aluminum busbar trunking set at 0.88; The third layer: The three-dimensional temperature field of the busbar is reconstructed using the Kriging interpolation algorithm, and the interpolation grid accuracy is set to 2 cm × 2 cm × 2 cm.

5. The method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion according to claim 1, characterized in that: The specific steps for calculating the dynamic early warning threshold in step four include: S1. Establish a reference temperature model: Reference temperature = 25 degrees Celsius + 0.01 × load current, where the load current is in amperes; when the ambient temperature deviates from 25 degrees Celsius, the correction is 0.8 × (ambient temperature - 25 degrees Celsius). S2. Set warning levels and corresponding thresholds: Level 1 warning threshold = base temperature + 30 degrees Celsius; Level 2 warning threshold = base temperature + 45 degrees Celsius; Level 3 warning threshold = base temperature + 60 degrees Celsius; When the load current exceeds 80% of the rated current, the warning threshold for each level is lowered by 5 degrees Celsius; When the ambient temperature exceeds 35 degrees Celsius, the warning threshold for each level is further lowered by 3 degrees Celsius. S3. Threshold Verification and Correction: If the actual temperature is close to the threshold but no fault occurs for three consecutive times, the corresponding warning threshold will be raised by 2 degrees Celsius; if the temperature does not reach the threshold but abnormal overheating occurs, the corresponding warning threshold will be lowered by 3 degrees Celsius.

6. The method for real-time monitoring and early warning of busbar temperature field using multi-sensor fusion according to claim 1, characterized in that: The specific operations for tiered early warning in step five include: When the temperature at the monitoring point reaches the Level 1 warning threshold, the Level 1 warning is activated: a yellow warning signal is issued, warning data is recorded, and staff are reminded to conduct inspections via a local audible and visual alarm. The inspection cycle is shortened to once every 30 minutes. When the temperature at the monitoring point reaches the level 2 warning threshold, the level 2 warning is activated: an orange warning signal is issued, and in addition to performing the level 1 warning operation, a warning message is sent to the operation and maintenance management platform, and an inspection work order is automatically generated, requiring staff to arrive at the site within 15 minutes to troubleshoot the fault. When the temperature at the monitoring point reaches the Level 3 warning threshold, the Level 3 warning is activated: a red warning signal is issued, and in addition to performing the Level 2 warning operation, the power control system is linked to issue a trip warning suggestion and activate the emergency cooling device until the temperature drops below the Level 2 warning threshold.

Images