A two-sided switch cabinet thermal state internet of things monitoring grading method and system

Abstract

The application discloses a double-sided switch cabinet thermal state Internet of Things monitoring grading method and system, and relates to the technical field of switch cabinet thermal state monitoring and grading control. In view of the problem that the thermal coupling of front columns, rear columns and middle main bus bars of double-sided back-to-back switch cabinets causes the temperature rise sources of measuring points to be difficult to distinguish, a front column-middle main bus bar-rear column direction pairing measuring point group is established, temperature, environment benchmark and operation condition data are collected and aligned, and a corrected temperature rise is obtained; the temperature rise contribution of each candidate heat source is solved according to a thermal coupling response relationship, crosstalk contribution is stripped off, and the endogenous temperature rise of each regional source is reconstructed, so that the heat source attribution and thermal state grade are determined, and corresponding heat dissipation or maintenance information is output. The system comprises a direction pairing sensing component, an operation condition collecting component, a data processing and heat source attribution unit, a grading and heat dissipation processing unit and an Internet of Things linkage unit. The application is used for online thermal state monitoring, grading and directional processing of double-sided switch cabinets.

Description

Technical Field

[0001] This invention relates to the field of thermal status monitoring and hierarchical control technology for switchgear, specifically to a method and system for IoT-based thermal status monitoring and hierarchical control of double-sided back-to-back withdrawable switchgear. Background Technology

[0002] With the increasing number of circuits carried by power distribution systems and the increasing space integration in power distribution sites, withdrawable switchgear is gradually developing towards miniaturization, high density, and online operation and maintenance. Double-sided back-to-back withdrawable switchgear, by setting up front and rear functional units within the same cabinet and sharing a main busbar axis located in the middle of the cabinet, can increase circuit layout density with minimal increase in cabinet footprint. This is suitable for applications with limited installation space and a large number of power distribution circuits. As the number of functional units, conductive connections, and operating circuits within a single cabinet increases, the number of heat sources and the heat load within the unit space also increase accordingly, placing higher demands on online monitoring of the cabinet's thermal status and targeted heat dissipation measures.

[0003] In double-sided back-to-back withdrawable switchgear, the front functional units, rear functional units, and the intermediate main busbar are not thermally independent of each other. The heat generated by the front and rear functional units during operation can be transferred through heat conduction, heat radiation, and convection via the intermediate main busbar, the metal structural components of the cabinet, and the air inside the cabinet. The intermediate main busbar itself may also become an independent heat source due to changes in load current, connection resistance, or contact condition. Therefore, the temperature collected by thermal state measuring points located near the front, rear, or intermediate main busbar is usually the result of the combined effects of heat generated in the local area, heat transferred from other areas, heat generated by the intermediate main busbar, and ambient temperature. There is no longer a unique correspondence between the location of the measuring point and the actual location of the heat source.

[0004] Existing switchgear temperature monitoring technologies typically set up temperature monitoring points based on functional compartments, component locations, or conductive connection points within the cabinet. These points collect temperature information and compare it with preset thresholds, normal temperature ranges, or temperature control targets. When the temperature at a monitoring point reaches the corresponding condition, an alarm is output or the corresponding ventilation, exhaust, and heat dissipation devices are controlled. This method can improve the temperature measurement coverage inside the switchgear and can associate abnormal temperature information with the corresponding compartments or component locations.

[0005] For example, Chinese invention patent application CN117991842A discloses a method, apparatus, device, and storage medium for temperature control of switchgear. This method identifies the compartments within the switchgear to be controlled and the components within each compartment. The compartments are then divided according to the components. Based on the division, temperature monitoring points are determined, and temperature information from each monitoring point is acquired. A preset PID controller is then used to control the switchgear temperature based on this information. This solution, by further dividing the compartments and setting corresponding monitoring points, reduces unnecessary monitoring points and enables real-time temperature control of the switchgear.

[0006] However, the aforementioned existing technologies still process temperature information collected at each monitoring point, and their control logic is based on a definite correspondence between the monitoring point and the corresponding compartment or component. When this type of monitoring method is applied to a double-sided back-to-back switchgear where the front and rear rows share a common intermediate main busbar, the aforementioned correspondence is easily affected by the coupling of the front and rear thermal fields. For example, a temperature rise at the front row measuring points may be caused by the heat generated by the front row functional units themselves, or by the heat generated by the rear row functional units transferred through the intermediate main busbar or the air inside the cabinet, or by the heat generated by the intermediate main busbar itself; the rear row measuring points and the intermediate main busbar measuring points may also simultaneously contain heat contributions from multiple heat sources.

[0007] Without distinguishing the direction of heat source, directly judging anomalies and controlling heat dissipation in the corresponding area based on the original temperature of the measuring point can easily lead to misinterpreting crosstalk heat transferred from the opposite side as an anomaly on this side, and thus activating the local heat dissipation device. This approach may not only result in ineffective heat dissipation for areas that are not actual heat sources, but may also leave the actual heat-generating areas untreated. Therefore, in double-sided back-to-back switchgear, the problem faced by existing technologies is no longer simply determining whether the temperature of a certain monitoring point exceeds a threshold, but rather further determining where the temperature rise at that monitoring point originates, and whether the net thermal state of the area itself, after removing the thermal coupling effects of other areas, reaches an abnormal level.

[0008] The technical problem to be solved by this invention is: under the condition that the front and rear functional units of a double-sided back-to-back switchgear share the middle main busbar axis and the front and rear thermal fields are coupled with each other, how to distinguish the self-generated heat contribution, crosstalk heat contribution, and middle main busbar axis contribution of the temperature rise reflected by each thermal state measurement point based on the corresponding thermal state information of the front, rear, and middle main busbar axes, how to attribute abnormal heat sources to the front, rear, or middle main busbar axes, and how to obtain the net thermal state characterization quantity of the region after removing the thermal coupling effect between regions, so that the anomaly judgment, thermal state classification, and heat dissipation treatment are based on the net thermal state of the region rather than the original temperature of the measurement point. Summary of the Invention

[0009] To address the issue of thermal coupling between the front, rear, and middle main busbar shafts of double-sided back-to-back switchgear, where the temperature rise collected by a single thermal state measurement point simultaneously includes the heat contribution from the local area and the crosstalk heat contribution from other areas, existing methods that independently judge the original temperature based on the area where the measurement point is located are insufficient to determine the actual heat source, this invention provides a classification method and system for IoT monitoring of the thermal state of double-sided switchgear. This method distinguishes the temperature rise contribution of the front functional units, rear functional units, and middle main busbar shaft to each thermal state measurement point, assigns abnormal heat sources to the corresponding areas, and uses the net thermal state characterization of the local area after removing the thermal coupling influence between areas as the basis for thermal state classification and handling.

[0010] The core concept of this invention is to transform the direction of heat source from a passively included auxiliary factor in the regional temperature into a processing object that is paired for acquisition, thermal coupling identification, contribution decomposition, and independent attribution; first, the influence of the environment and normal operating conditions is corrected, then the contribution of candidate heat sources in the front, middle main bus axis and rear columns is decomposed, external crosstalk contribution is deducted from the temperature rise observed in each region, and finally, thermal state classification and directional treatment are carried out based on the endogenous temperature rise of the source.

[0011] A hierarchical method for monitoring the thermal status of double-sided switchgear using IoT, comprising: dividing the thermal status measurement points located in the front, middle, and rear functional units into multiple directional paired measurement point groups according to at least one of the hierarchical, circuit, or component correspondence relationships between the front functional units, the middle main busbar, and the rear functional units in the double-sided back-to-back switchgear; acquiring temperature data of each directional paired measurement point group within the same sampling time window, and acquiring the corresponding environmental reference temperature and operating condition data; and performing environmental reference correction on the temperature data based on the normal temperature rise baseline. The operating condition baseline is corrected to obtain the corrected temperature rise of the front row, the corrected temperature rise of the middle main bus axis, and the corrected temperature rise of the rear row. The thermal coupling response relationship between the front row candidate heat sources, the middle main bus axis candidate heat sources, and the rear row candidate heat sources and the paired measuring point groups in each direction is obtained, and the temperature rise contribution of each candidate heat source is solved accordingly. The crosstalk temperature rise contribution formed by other candidate heat sources is subtracted from the corrected temperature rise of each candidate region to obtain the source endogenous temperature rise characterization quantity of each candidate region. The region to which the abnormal heat source belongs is determined, and the thermal state level is determined according to the source endogenous temperature rise characterization quantity of the region to which it belongs. The corresponding heat dissipation treatment information or maintenance inspection information is output.

[0012] Its working mechanism is as follows: the front, middle main busbar axis, and rear measuring points of the double-sided switchgear are all mixed observations after multiple candidate heat sources have acted on them through different propagation paths. By pairing measuring point groups in different directions, multiple observations with physical correspondences are obtained. By correcting for environmental and operating conditions, normal common changes are weakened from the observations. Then, the contribution of each heat source is decomposed by using the differential response relationship from the candidate heat sources to each measuring point. This distinguishes the location of the measuring point from the actual heat source location, and establishes the thermal state classification on the self-heating state of the local area after removing external crosstalk.

[0013] As a further technical solution, the front and rear thermal state measuring points in the paired measuring point group in the same direction are respectively set at the corresponding front and rear functional units, and the middle main bus axis thermal state measuring point is set at the middle main bus axis of the corresponding layer between the front and rear functional units; the temperature data output by each thermal state measuring point is associated with the measuring point identifier, area identifier, layer identifier and timestamp, and is aligned to the same sampling time window through at least one of the following methods: unified sampling trigger, clock synchronization or data arrival time correction.

[0014] As a further technical solution, the normal temperature rise baseline is obtained through operating data with no known thermal anomalies, normal communication of thermal state measuring points, and identifiable ventilation status. The operating data is grouped according to functional unit type, operating load range, and ventilation status. The temperature rise of each thermal state measuring point relative to the ambient reference temperature of the same period is calculated, and the median, robust mean, or fitted value of the corresponding data group is used as the normal temperature rise baseline.

[0015] As a further technical solution, the thermal coupling response relationship is obtained through commissioning calibration events or historical high-confidence thermal response events, and includes at least the response delay, temperature rise response amplitude, and temperature rise attenuation relationship from the candidate heat source to each thermal state measurement point; the temperature rise contribution formed by the intermediate main busbar as an independent candidate heat source, together with the temperature rise contribution formed by the propagation of the preceding and subsequent candidate heat sources to other areas through the intermediate main busbar, cabinet structural components, or air inside the cabinet, is characterized by mutually independent heat source state parameters and thermal propagation response parameters.

[0016] As a further technical solution, before solving the temperature rise contribution of each candidate heat source, a thermal coupling response matrix is ​​constructed based on the thermal coupling response relationship, and it is determined whether it meets the preset identifiable conditions. When the preset identifiable conditions are not met, an unreliable attribution state is output, and no deterministic heat source attribution result for a single region is output.

[0017] As a further technical solution, a regularized least squares method with non-negative constraints is adopted to solve the temperature rise contribution of each candidate heat source based on the corrected temperature rise observation vector and the thermal coupling response matrix. The measurement point weights are set according to the measurement reliability of each thermal state measurement point, and the validity of the solution is judged based on the residual between the model reconstructed temperature rise and the actual corrected temperature rise.

[0018] As a further technical solution, based on the comprehensive contribution intensity and contribution ratio of each candidate heat source, combined with the degree of differentiation between different sources, reconstruction residuals and data integrity, the state of single heat source, multiple sources generating heat together or unreliable attribution is determined.

[0019] As a further technical solution, the intrinsic temperature rise characterization quantity of the abnormal heat source attribution area is compared with the classification conditions under the same environmental and operating condition benchmarks to determine the normal level, attention level, warning level, or abnormal level; when the thermal state measurement point fails, the operating condition data is missing, the thermal coupling response relationship fails, or the attribution is unreliable, the temperature over-limit protection based on the original temperature is retained.

[0020] Based on the same inventive concept, this invention also provides a double-sided switchgear thermal status IoT monitoring and grading system, including a direction pairing sensing component, an operating condition acquisition component, an IoT communication component, a memory, and a processor. The direction pairing sensing component includes a front-row thermal status sensor, a middle main busbar axis thermal status sensor, and a rear-row thermal status sensor; the operating condition acquisition component is used to acquire ambient reference temperature, operating load status, and ventilation status; the IoT communication component is used to receive temperature data and operating condition data with measurement point identifiers and timestamps; the memory is used to store measurement point configuration relationships, normal temperature rise baselines, thermal coupling response relationships, and grading conditions; the processor is used to perform time alignment, baseline correction, heat source contribution calculation, crosstalk removal, heat source attribution, thermal status grading, and output of handling information.

[0021] Furthermore, the IoT communication components include an edge gateway and a monitoring platform; the edge gateway is used for data reception, time alignment, local caching, and monitoring record encapsulation, while the monitoring platform is used to store measurement point configurations, model parameters, identifiable conditions, attribution conditions, hierarchical conditions, and historical monitoring records.

[0022] Beneficial effects Compared with the prior art, the present invention has the following beneficial effects: Firstly, by setting up a group of thermal state measurement points with corresponding front, middle main busbar axis and rear direction, and combining environmental and operating condition benchmark correction and thermal coupling response relationship, the present invention decomposes the temperature rise contribution of multiple candidate heat sources at each thermal state measurement point, which is beneficial to distinguish the self-generated heat contribution of this region and the crosstalk heat contribution of other regions, and assigns abnormal heat sources to the front functional unit, the rear functional unit or the middle main busbar axis.

[0023] Secondly, this invention deducts the crosstalk temperature rise contribution from other candidate heat sources from the corrected temperature rise of each candidate region, and uses the obtained source-endogenous temperature rise characterization quantity as the basis for thermal state classification and handling. This helps to reduce the situation of misjudging heat transfer from the opposite side as equipment malfunction on this side, and improves the pertinence of heat dissipation handling and maintenance inspection.

[0024] Third, this invention constrains the heat source attribution results by identifying conditions, source contribution ratio, reconstruction residuals and data integrity, and retains the original temperature over-limit protection when there is a measurement point failure, data missing, model failure or unreliable attribution of heat source. This helps to avoid outputting erroneous deterministic attribution results when observation conditions are insufficient, and maintains the safety and reliability of the thermal status monitoring process of double-sided switchgear. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the internal structure and thermal state sensing arrangement of the double-sided switch cabinet of the present invention. Figure 2 This is a schematic diagram of the directional pairing measurement point group of the present invention; Figure 3 This is a block diagram of the thermal state IoT monitoring and grading system of the present invention; Figure 4 This is a flowchart of the thermal state monitoring and heat source attribution method of the present invention; Figure 5 This is a schematic diagram showing the thermal coupling response relationship between the candidate heat source and the measuring points in each direction in this invention; Figure 6 This is a flowchart illustrating the single-source, multi-source, unreliable attribution, and secure rollback handling procedures of this invention.

[0026] Reference numerals in the attached diagram: 1—Double-sided switch cabinet, 2—Front row functional unit, 3—Rear row functional unit, 4—Middle main busbar shaft, 21—Front row thermal status sensing, 31—Rear row thermal status sensing, 41—Middle busbar shaft thermal status sensing, 5—Data processing and heat source attribution unit, 6—Grading and heat dissipation unit, 7—IoT linkage unit, 8—Ambient reference temperature sensor, 9—Operating condition acquisition component, 10—Edge gateway, 11—Monitoring platform, 12—Heat dissipation device, 13—Maintenance terminal. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0028] In this paper, "double-sided switchgear" refers to a switchgear in which the front and rear functional units are arranged back-to-back and share a central main busbar; "corrected temperature rise" refers to the temperature rise parameter obtained after the original temperature of the thermal state measurement point is corrected by environmental reference and normal operating condition baseline; "temperature rise contribution" refers to the temperature rise component formed by a candidate heat source at a certain thermal state measurement point through the corresponding heat propagation path; "source-endogenous temperature rise characterization quantity" refers to the temperature rise domain parameter used to characterize the self-heating state of a certain region after subtracting the crosstalk temperature rise contribution formed by other candidate heat sources in that region from the corrected temperature rise of that region. The source-endogenous temperature rise characterization quantity is not directly equivalent to heat expressed in joules or thermal power expressed in watts.

[0029] Example 1 This embodiment uses online thermal status monitoring and graded handling of double-sided back-to-back withdrawable switchgear as the best implementation method.

[0030] System composition: See Figure 1 and Figure 3 The double-sided switch cabinet 1 is equipped with a front functional unit 2, a rear functional unit 3, and an intermediate main busbar 4 located between the front functional unit 2 and the rear functional unit 3. Multiple front thermal status sensors 21 are installed at the front functional unit 2, multiple rear thermal status sensors 31 are installed at the rear functional unit 3, and multiple intermediate busbar thermal status sensors 41 are installed at the intermediate main busbar 4.

[0031] The front thermal status sensor 21, the rear thermal status sensor 31, and the middle busbar thermal status sensor 41 are respectively communicatively connected to the data processing and heat source attribution unit 5. The data processing and heat source attribution unit 5 is connected to the grading and heat dissipation handling unit 6, which is connected to at least one of the edge gateway 10, monitoring platform 11, heat dissipation device 12, and maintenance terminal 13 through the Internet of Things linkage unit 7.

[0032] The front thermal state sensor 21, the rear thermal state sensor 31, and the intermediate busbar shaft thermal state sensor 41 can employ contact temperature sensors, fiber optic temperature sensors, passive wireless temperature sensors, or other temperature sensing devices suitable for the insulation environment of the switchgear. An environmental reference temperature sensor 8 is installed at the air inlet of the cabinet, the external environment, or a location inside the cabinet where it is not directly exposed to local heat source radiation. The operating condition acquisition component 9 acquires the operating load status of the front functional unit 2 and the rear functional unit 3, the current carrying status of the intermediate main busbar shaft 4, and the ventilation status of the double-sided switchgear.

[0033] Establishment of direction pairing measurement point group: See Figure 2Based on the hierarchical, circuit, or component correspondences among the front functional unit 2, the middle main busbar shaft 4, and the rear functional unit 3, the corresponding front thermal state sensors 21, the middle busbar shaft thermal state sensors 41, and the rear thermal state sensors 31 are grouped into a directional pairing measurement point group. For double-sided switchgear with multiple drawer functional units in both the front and rear rows, the front functional units on the same layer, the corresponding middle main busbar shaft segment, and the rear functional units on the same layer are grouped into a directional pairing measurement point group; when there are multiple independent circuits on the same layer, sub-measurement point groups are further divided according to the circuit correspondence.

[0034] Data Acquisition and Time Alignment: The IoT linkage unit 7 acquires temperature data from each thermal state measurement point, and simultaneously acquires the ambient reference temperature, operating load status, ventilation status, and timestamp. Temperature data in paired measurement point groups in the same direction are aligned to the same sampling time window through at least one of the following methods: unified sampling trigger, clock synchronization, or data arrival time correction. For short-term missing data, interpolation can be performed based on the valid temperature values ​​before and after the missing data; for data with continuous missing time exceeding the allowable range, no interpolation is performed, and the corresponding thermal state measurement point is marked as invalid.

[0035] Normal operating condition baseline correction: Historical data of the double-sided switchgear under normal operating conditions, with no known thermal anomalies, normal communication of thermal state measurement points, identifiable ventilation status, and load changes within the normal operating range, are selected as normal operating sample data. These data are grouped according to functional unit type, load range, and ventilation status. For each thermal state measurement point, its temperature rise relative to the ambient reference temperature is calculated, and the median, robust mean, or fitted value of the corresponding data group is used as the normal temperature rise baseline for that measurement point under the corresponding operating conditions.

[0036] The corrected temperature rise at time t for the i-th thermal state measuring point is determined by the following formula: zᵢ(t)=Tᵢ(t)-T a (t)-bᵢ(x(t)) In the formula, Tᵢ(t) represents the original temperature of the i-th thermal state measurement point, and T a (t) represents the ambient reference temperature during the same period, x(t) represents the current operating load and ventilation status, bᵢ(x(t)) represents the normal temperature rise baseline of the measuring point under the corresponding operating conditions, and zᵢ(t) represents the corrected temperature rise after correction for the environment and operating conditions.

[0037] After processing with Equation (1), the normal temperature rise component formed by the overall change in ambient temperature, the increase in normal load, and the switching of ventilation status is weakened. The subsequent heat source contribution solution is mainly aimed at the temperature rise component that exceeds the baseline of normal operating conditions.

[0038] Thermal coupling contribution calculation: For each pair of measuring points in each direction, the corrected temperature rise of the front column, the middle main bus axis, and the rear column are combined into a corrected temperature rise observation vector. The candidate heat sources of the front column, the middle main bus axis, and the rear column are combined into a candidate heat source state vector. The thermal coupling relationship between the candidate heat sources and each thermal state measuring point is expressed by the following formula: z(t) = Hq(t) + ε(t) In the formula, z(t) represents the corrected temperature rise observation vectors of the front column, the middle main bus axis and the rear column, q(t) represents the state vectors of the front column candidate heat source, the middle main bus axis candidate heat source and the rear column candidate heat source, H represents the thermal coupling response relationship between the candidate heat source and each thermal state measurement point, and ε(t) represents the residual formed by measurement noise, unmodeled ventilation changes and model approximation errors.

[0039] In the thermal coupling response relationship, the diagonal parameters characterize the temperature rise effect of candidate heat sources on the measuring points in this region, while the off-diagonal parameters characterize the temperature rise effect caused by the propagation of candidate heat sources to other regions via the central main busbar axis 4, cabinet structural components, or air inside the cabinet. Under the constraint of non-negative candidate heat source states, the data processing and heat source attribution unit 5 solves for the contribution state of each candidate heat source based on the corrected temperature rise observation vector and the thermal coupling response relationship.

[0040] Based on the calculated contributions of each candidate heat source, the crosstalk temperature rise contributions of other candidate heat sources in the corresponding regions are subtracted from the corrected temperature rise of the front, middle main bus axis, and rear regions to obtain the source-endogenous temperature rise characterization of each candidate region: nⱼ(t)=max{0,zⱼ(t)-Σ k ≠ⱼcⱼ← k (t)} In the formula, zⱼ(t) represents the corrected temperature rise in candidate region j, and cⱼ← k (t) represents the crosstalk temperature rise contribution of candidate heat source k in candidate region j, and nⱼ(t) represents the intrinsic temperature rise characteristic of candidate region j after removing the influence of other candidate heat sources. The non-negative treatment in equation (3) is used to avoid measurement noise or model residuals from producing negative values ​​without positive heating physical meaning.

[0041] The data processing and heat source attribution unit 5 determines the attribution area of ​​the abnormal heat source based on the comprehensive contribution intensity of the leading candidate heat sources, the intermediate main bus axis candidate heat sources, and the trailing candidate heat sources. The grading and heat dissipation handling unit 6 compares the intrinsic temperature rise characteristic of the abnormal heat source attribution area with the grading conditions under the same environmental and operating condition benchmarks to determine the normal level, attention level, warning level, or abnormal level, and outputs the corresponding heat dissipation handling information or maintenance inspection information through the IoT linkage unit 7.

[0042] Example 2 This embodiment focuses on how to obtain the thermal coupling response relationship and how to avoid mixing the heat generated by the intermediate main bus shaft itself with the crosstalk heat propagated through the intermediate main bus shaft.

[0043] Candidate heat source setup: For each pair of measuring points in each direction, candidate heat sources are set up in the front row, the middle main busbar axis, and the rear row. The front row candidate heat sources characterize abnormal heating generated by the front functional units and their conductive connections; the rear row candidate heat sources characterize abnormal heating generated by the rear functional units and their conductive connections; and the middle main busbar axis candidate heat sources characterize abnormal heating generated by the middle main busbar axis or its connecting parts. The effect of heat propagation from the front or rear rows to other regions via the middle main busbar axis is not treated as the status of the middle main busbar axis candidate heat source, but rather as a cross-regional propagation parameter in the thermal coupling response relationship.

[0044] Commissioning Calibration Method: During the commissioning and debugging of the double-sided switchgear or when load adjustment is permitted, high-confidence thermal response events corresponding to the front, rear, and intermediate main busbar shafts are selected respectively. When calibrating the front-row candidate heat sources, within the safe operating range, the operating load of one front-row functional unit 2 is changed, while keeping the load of the corresponding rear-row functional unit 3 relatively stable. The response sequence, temperature rise amplitude, and attenuation process of the front-row thermal state sensor 21, the intermediate busbar shaft thermal state sensor 41, and the rear-row thermal state sensor 31 are recorded. When calibrating the rear-row candidate heat sources, the operating load of the rear-row functional unit 3 is changed, while keeping the load of the corresponding front-row functional unit 2 relatively stable. When calibrating the intermediate main busbar shaft candidate heat sources, events in which the current-carrying state of the intermediate main busbar shaft 4 changes significantly, while the loads of the corresponding front and rear local circuits remain relatively stable are selected.

[0045] For switchgear where active load adjustment is not permitted, high-confidence events can be selected from historical monitoring data. A high-confidence event must meet at least the following criteria: the load or corrected temperature rise change in one candidate area significantly precedes that in other areas; the operating load in other candidate areas remains relatively stable within the corresponding time period; the ventilation status does not change within the calibrated time period; the thermal status measurement data is complete; and no other known maintenance operations or cabinet door opening events have occurred.

[0046] Based on the above events, the response sequence, temperature rise response intensity, and attenuation characteristics of each candidate heat source to each thermal state measurement point are determined, and the thermal coupling response relationship in equation (2) is formed. The heat source state formed by the middle main bus axis as an independent candidate heat source is set separately from the non-diagonal response parameters corresponding to the propagation of heat from the front or rear columns through the middle main bus axis, and the two are not combined into the same parameter.

[0047] Model validity check: The calibration data is divided into parameter identification data and validation data. The thermal coupling response relationship is established using the parameter identification data, and then the model reconstruction results are checked using the validation data. The model validity check includes at least: whether the current number of valid measurement points meets the requirement of distinguishing three candidate heat sources; whether the response patterns formed by different candidate heat sources at each measurement point are distinguishable; whether the thermal coupling response matrix has complete column rank; whether the residual between the model reconstruction temperature rise and the actual corrected temperature rise is within the allowable range of the validation data; and whether the current data completeness rate meets the preset conditions.

[0048] If the preceding and following candidate heat sources exhibit highly similar response patterns across all valid thermal state measurement points, the current observation data is insufficient to distinguish between them. In this case, the output will not force the conclusion that either the preceding or following heat source is a single anomalous heat source, but will instead output an unreliable attribution state. If a calibration event characterizing the independent heating of the intermediate main busbar axis has not yet been obtained, a definitive result indicating that the intermediate main busbar axis is a single anomalous heat source will not be output for the time being.

[0049] The specific process of contribution solution: When the model is effective, the front, middle main bus axis and rear corrected temperature rise of the paired measurement point group in the same direction are input into the data processing and heat source attribution unit 5. The data processing and heat source attribution unit 5 solves the state of the three candidate heat sources in a constrained manner according to the thermal coupling relationship expressed by equation (2). During the solution process, the state of each candidate heat source is restricted to non-negative, the measurement point weights are set according to the sensor calibration state, communication quality and data noise, the change of heat source state in adjacent calculation cycles is subject to smoothing constraints, and the validity of the solution result is judged according to the residual between the model reconstruction result and the actual corrected temperature rise.

[0050] Taking the abnormal heating of the front functional unit 2 as an example, the front thermal state sensor 21 first shows a corrected temperature rise, and then the middle busbar axis thermal state sensor 41 and the rear thermal state sensor 31 may also show a temperature rise due to heat propagation. When using the present invention, the data processing and heat source attribution unit 5 calculates the temperature rise contribution of the front candidate heat source in the middle main busbar axis region and the rear region based on the thermal coupling response relationship from the front candidate heat source to the three directional measuring points; then, according to formula (3), the crosstalk temperature rise contribution is deducted from the corrected temperature rise of the middle main busbar axis region and the rear region. If the source endogenous temperature rise characterization of the middle main busbar axis region and the rear region after deduction does not reach the corresponding classification condition, the above region is not determined as an independent abnormal heat source region.

[0051] Example 3 This embodiment focuses on the methods for determining single heat sources, multiple sources generating heat together, and unreliable attribution states, as well as the IoT-linked handling and safe rollback process after model failure.

[0052] Single heat source attribution: The data processing and heat source attribution unit 5 determines the comprehensive contribution intensity of the top candidate heat sources, the middle main busbar axis candidate heat sources, and the bottom candidate heat sources based on the temperature rise contribution of each candidate heat source to all effective thermal state measurement points. Higher contribution weights are assigned to thermal state measurement points with higher measurement reliability, locations closer to the main conductive connection points, or more pronounced responses to the corresponding candidate heat sources.

[0053] When the contribution percentage of the largest source meets the attribution criteria, there is sufficient differentiation between the contribution percentages of the largest and second-largest sources, the reconstruction residual of the thermally coupled model is within the allowable range, the current number of valid measurement points and data integrity meet the requirements, and the heat source attribution results remain stable over multiple consecutive calculation periods, the candidate heat source with the largest comprehensive contribution intensity is identified as the single anomalous heat source. The attribution criteria and differentiation criteria are determined based on known single heat source calibration events, rather than using a fixed proportion without data sources.

[0054] Multi-source co-heating: When the combined contribution of two or three candidate heat sources meets the corresponding contribution conditions, and the thermal coupling model can distinguish the contribution of each candidate heat source, the multi-source co-heating state is output. For example, when abnormal heating occurs simultaneously in the front column functional unit 2 and the rear column functional unit 3, the source endogenous temperature rise and thermal state level of the front column region and the rear column region are determined respectively, and the corresponding handling information is output respectively, without forcibly selecting one region as the sole heat source.

[0055] Heat source attribution for the intermediate main busbar shaft: When abnormal heating occurs on the intermediate main busbar shaft 4 itself or at the busbar connection point, the intermediate busbar shaft thermal state sensor 41 generates a strong internal temperature rise. The front thermal state sensors 21 and the rear thermal state sensors 31 may also experience temperature rises due to the propagation of heat from the busbar to both sides. The data processing and heat source attribution unit 5 calculates the contribution of the candidate heat sources of the intermediate main busbar shaft to the crosstalk temperature rise of the front and rear measuring points based on the calibration response relationship of the candidate heat sources. After deducting the crosstalk contribution from the corrected temperature rise of the front and rear measuring points, if the internal temperature rise of the front and rear measuring points does not meet the corresponding classification conditions, the abnormal heat source is attributed to the intermediate main busbar shaft 4.

[0056] Unreliable Attribution Status: When the number of effective thermal status measurement points is insufficient, the response modes of candidate heat sources cannot be distinguished, the model reconstruction residual exceeds the allowable range, the operating load status or ventilation status is missing, the cabinet structure or sensor position has changed and the model has not been recalibrated, the heat source attribution results change repeatedly in adjacent calculation cycles, or the current operating condition exceeds the effective range of the thermal coupling response relationship, the system outputs an unreliable attribution status. In this status, the system continues to output measurement point temperature information and overall thermal status alarms, but does not output deterministic heat source attribution results for a single area.

[0057] Thermal Status Classification: Normal operating data is used to establish the normal distribution of endogenous temperature rise in the front, middle main busbar axis, and rear regions. Based on the normal distribution, allowable temperature rise conditions of the equipment, and maintenance requirements, classification conditions are set for normal, attention, warning, and abnormal levels. Under the normal level, the current sampling cycle and operating status are maintained; under the attention level, the sampling frequency of paired measurement point groups in the corresponding direction is increased; under the warning level, heat dissipation instructions and maintenance inspection prompts are output to the area belonging to the abnormal heat source; under the abnormal level, advanced alarm information is output, and information for restricting operation, shutdown inspection, or manual confirmation is generated according to the switchgear protection strategy.

[0058] IoT linkage: Edge gateway 10 receives data output from thermal status sensor and operating condition acquisition component 9, completes time alignment, data quality check and local caching, and performs at least one of corrected temperature rise calculation, heat source contribution solution or safety rollback judgment; monitoring platform 11 saves the measurement point configuration relationship, normal temperature rise baseline and version, thermal coupling response parameters, identifiable conditions, attribution conditions, classification conditions, historical monitoring records and handling records.

[0059] Safety Rollback: When a thermal state measurement point fails, operating condition data is missing, thermal coupling response relationship fails, or heat source cannot be reliably assigned, the system enters a safety rollback state. In the safety rollback state, temperature over-limit protection based on the original temperature is retained, selective cooling shutdown for a single assigned area is stopped, global cooling can be initiated or the existing cooling state can be maintained, and the model failure reason, invalid measurement point identification, and manual inspection information are output. Heat source assignment and targeted handling are restored only after data is recovered and the model's validity conditions are continuously met.

[0060] Through the above-mentioned safe retreat method, the heat source attribution function of the present invention serves as an enhancement to the original temperature safety protection function, without replacing or weakening the original switchgear temperature over-limit protection.

[0061] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A hierarchical method for IoT monitoring of the thermal status of a double-sided switchgear, characterized in that, include: The corresponding front thermal state measurement points, middle main busbar axis thermal state measurement points, and rear thermal state measurement points in the double-sided switchgear are divided into directional paired measurement point groups. The temperature of the measuring point within the same sampling time window is obtained, along with the corresponding ambient reference temperature, operating load status, and ventilation status. Based on the ambient reference temperature and the normal temperature rise baseline under the corresponding operating conditions, the temperature of the measuring point is corrected to obtain the front-row corrected temperature rise, the middle main busbar shaft corrected temperature rise, and the rear-row corrected temperature rise. Obtain the thermal coupling response relationship between the front candidate heat source, the middle main bus axis candidate heat source and the rear candidate heat source and each thermal state measurement point; Based on the thermal coupling response relationship and each corrected temperature rise, the temperature rise contribution of each candidate heat source is solved under the preset physical constraints. The source-generated temperature rise of each region is obtained by subtracting the crosstalk temperature rise contribution from other candidate heat sources from the corrected temperature rise of each region; the heat source attribution is determined based on the temperature rise contribution of each candidate heat source, the thermal status level is determined based on the source-generated temperature rise of the assigned region, and the corresponding heat dissipation treatment information or maintenance inspection information is output.

2. The method according to claim 1, characterized in that: In the same direction, the front and rear thermal state measuring points in the paired measuring point group are set in the corresponding front and rear functional units, respectively, and the thermal state measuring point of the middle main bus axis is set in the corresponding bus segment between the two. The data of each measuring point are associated with the measuring point identifier, area identifier, layer identifier and timestamp, and are aligned to the same sampling time window through unified sampling, clock synchronization or time correction.

3. The method according to claim 1, characterized in that: The normal temperature rise baseline is obtained through normal operation data. The normal operation data is grouped according to functional unit type, operating load range and ventilation status. The median, robust mean or fitted value of each thermal state measurement point relative to the ambient reference temperature of the same period is used as the normal temperature rise baseline under the corresponding operating conditions.

4. The method according to claim 1, characterized in that: The thermal coupling response relationship is obtained through commissioning calibration events or historical high-confidence thermal response events, and includes the response delay, temperature rise response amplitude and attenuation relationship from candidate heat sources to each thermal state measurement point; the heat source state parameters corresponding to the self-heating of the intermediate main bus shaft are set separately from the response parameters corresponding to the propagation of other candidate heat sources through the intermediate main bus shaft.

5. The method according to claim 1, characterized in that: A thermal coupling response matrix is ​​constructed based on the thermal coupling response relationship. When the number of effective thermal state measurement points is not less than the number of candidate heat sources, and the thermal coupling response matrix has a complete column rank, the current data is determined to meet the heat source identification condition; otherwise, an unreliable attribution state is output, and the output of single heat source attribution results is stopped.

6. The method according to claim 1, characterized in that: The front-column corrected temperature rise, the middle main bus axis corrected temperature rise, and the rear-column corrected temperature rise are used to form an observation vector, and the thermal coupling response relationship is used to form a response matrix. The temperature rise contribution of each candidate heat source is solved by regularized least squares method with non-negative constraints, and the validity of the solution result is judged by the residual between the model reconstructed temperature rise and the actual corrected temperature rise.

7. The method according to claim 1, characterized in that: Based on the temperature rise contribution of each candidate heat source to the effective thermal state measurement point, determine the comprehensive contribution intensity and source contribution ratio of each candidate heat source; when the contribution ratio of the largest source reaches the attribution condition and the difference between it and the contribution ratio of the second largest source reaches the distinction condition, determine the corresponding candidate heat source as an abnormal heat source; when multiple candidate heat sources simultaneously reach the common heating condition, output the multi-source common heating state; otherwise, output the unreliable attribution state.

8. The method according to claim 1, characterized in that: The source endogenous temperature rise of the assigned area is compared with the graded conditions under the corresponding operating conditions to determine the thermal state level. Based on the thermal state level, the sampling frequency is increased, heat dissipation commands are output, or maintenance and inspection information is output. When the measuring point is faulty, the operating condition data is missing, the thermal coupling response relationship fails, or the heat source cannot be reliably assigned, the temperature over-limit protection based on the original temperature is retained, and the selective heat dissipation control for a single area is stopped.

9. A hierarchical IoT monitoring system for the thermal status of a double-sided switchgear, characterized in that, The system includes a direction pairing sensing component, a working condition acquisition component, an IoT communication component, a memory, and a processor. The direction pairing sensing component includes a front thermal status sensor located in the front functional unit, a bus thermal status sensor located on the middle main busbar shaft, and a rear thermal status sensor located in the rear functional unit. These three corresponding sensors constitute a direction pairing measurement point group. The working condition acquisition component is used to acquire ambient reference temperature, operating load status, and ventilation status. The IoT communication component is used to receive measurement point temperature and operating condition data. The memory is used to store the measurement point configuration relationship, normal temperature rise baseline, thermal coupling response relationship, and grading conditions; The processor is used to perform time alignment and operating condition correction on the temperature measurement point, solve the temperature rise contribution of each candidate heat source according to the thermal coupling response relationship, obtain the source-induced temperature rise of each region after deducting the crosstalk temperature rise contribution, and output heat dissipation treatment information or maintenance inspection information according to the heat source attribution and source-induced temperature rise.

10. The system according to claim 9, characterized in that: The IoT communication components include an edge gateway and a monitoring platform. The edge gateway receives and aligns temperature data and operating condition data, and performs local caching and thermal state calculations. The monitoring platform stores measurement point configurations, normal temperature rise baselines, thermal coupling response parameters, attribution conditions, grading conditions, and historical records. When the measurement point data is incomplete or the current thermal coupling response relationship does not meet the identifiable conditions, the processor outputs an unreliable attribution status and retains temperature over-limit protection based on the original temperature.

Images