Temperature sensor-based low-voltage electrical equipment safety monitoring method and system

Abstract

The application discloses a low-voltage electrical equipment safety monitoring method and system based on temperature sensing, relates to the technical field of low-voltage electrical circuit protection device control, and comprises the following steps: collecting multi-node real-time temperature data of low-voltage electrical equipment through a distributed temperature sensor array, determining an abnormal temperature change mode and a multi-stage early warning signal through thermal state analysis, triggering a circuit protection device to adjust the operating parameters of the low-voltage electrical equipment and forming a safety protection record log, constructing a fault prediction channel according to the safety protection record log, and finally generating a device health state evaluation report. The application solves the technical problems of single-point sensing monitoring lag, inaccurate abnormal mode judgment and rigid safety protection strategy in traditional low-voltage electrical equipment temperature monitoring, achieves the technical effects of comprehensive multi-node data collection by a distributed sensor array, accurate thermal state analysis and abnormal temperature change mode identification, and realizes accurate low-voltage electrical equipment temperature monitoring, effective protection and health prediction.

Description

Technical Field

[0001] This invention relates to the field of low-voltage electrical circuit protection device control technology, and in particular to a method and system for safety monitoring of low-voltage electrical equipment based on temperature sensing. Background Technology

[0002] Low-voltage electrical equipment is widely used in industrial production and daily life. Its operational safety directly affects the continuity of production activities and the safety of personnel and property. Abnormal temperature is a core cause of equipment failure and even fires, making accurate temperature monitoring and coordinated protection crucial. Current technologies for temperature monitoring and protection of low-voltage electrical equipment largely rely on single-point temperature sensors combined with fixed threshold-triggered circuit protection devices. These methods play a role in scenarios with simple structures and stable operation, but as equipment complexity increases and safety requirements rise, their limitations become apparent when applied to low-voltage electrical equipment operating in multiple nodes. Due to the multi-node thermal distribution characteristics and dynamic operating states of low-voltage electrical equipment, traditional methods cannot comprehensively collect temperature data from each node, easily missing local abnormal temperature changes. This leads to untimely triggering or malfunctioning of circuit protection devices, resulting in incomplete temperature data and inaccurate protection responses, failing to meet the needs for accurate monitoring and effective protection of low-voltage electrical equipment. Summary of the Invention

[0003] This application provides a method and system for safety monitoring of low-voltage electrical equipment based on temperature sensing, which solves the technical problems of lagging single-point sensing monitoring, inaccurate judgment of abnormal modes, and rigid safety protection strategies in traditional temperature monitoring of low-voltage electrical equipment.

[0004] The first aspect of this application provides a method for safety monitoring of low-voltage electrical equipment based on temperature sensing. The method includes: collecting real-time temperature data from multiple nodes of the low-voltage electrical equipment using a distributed temperature sensor array to obtain a temperature monitoring dataset; performing thermal state analysis based on the temperature monitoring dataset to obtain temperature change state parameters for temperature safety monitoring of the low-voltage electrical equipment, determining abnormal temperature change modes, wherein the abnormal temperature change modes include multi-level early warning signals; triggering a circuit protection device to dynamically adjust the operating parameters of the low-voltage electrical equipment according to the multi-level early warning signals, obtaining a safety protection log; and constructing a fault prediction channel based on the safety protection log to perform operational prediction and evaluation of the low-voltage electrical equipment, generating an equipment health status assessment report.

[0005] A second aspect of this application provides a low-voltage electrical equipment safety monitoring system based on temperature sensing. The system includes: a temperature monitoring dataset acquisition module, used to collect real-time temperature data from multiple nodes of the low-voltage electrical equipment via a distributed temperature sensor array to obtain a temperature monitoring dataset; an abnormal temperature change mode acquisition module, used to perform thermal state analysis based on the temperature monitoring dataset, obtain temperature change state parameters for temperature safety monitoring of the low-voltage electrical equipment, and determine abnormal temperature change modes, wherein the abnormal temperature change modes include multi-level early warning signals; a safety protection log acquisition module, used to dynamically adjust the operating parameters of the low-voltage electrical equipment according to the multi-level early warning signals triggering circuit protection devices to execute safety protection strategies, and obtain a safety protection log; and an equipment health status assessment report acquisition module, used to construct a fault prediction channel based on the safety protection log to perform operational prediction and assessment of the low-voltage electrical equipment, and generate an equipment health status assessment report.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] This application collects real-time temperature data from multiple nodes of low-voltage electrical equipment using a distributed temperature sensor array. Through thermal state analysis processing, including real-time temperature rise rate calculation, temperature distribution uniformity analysis, and determination of thermal equilibrium time parameters, it obtains temperature change state parameters and identifies abnormal temperature change patterns such as gradual, abrupt, and localized overheating concentrations. Combined with multi-level early warning signals, it triggers circuit protection devices to implement preventative, proactive, and emergency safety protection strategies to dynamically adjust equipment operating parameters. Furthermore, based on safety protection logs, it constructs a fault prediction channel, thereby achieving precise temperature safety monitoring and operational health assessment of low-voltage electrical equipment. This makes the safety management of low-voltage electrical equipment more reliable and efficient, achieving the technical effect of comprehensive multi-node data collection by a distributed sensor array, accurate thermal state analysis, and identification of abnormal temperature change patterns, enabling precise temperature monitoring, effective protection, and health prediction of low-voltage electrical equipment. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a flowchart illustrating the safety monitoring method for low-voltage electrical equipment based on temperature sensing provided in an embodiment of this application.

[0010] Figure 2 This is a schematic diagram of the structure of a low-voltage electrical equipment safety monitoring system based on temperature sensing provided in an embodiment of this application.

[0011] Figure labeling: Temperature monitoring dataset acquisition module 1, abnormal temperature change mode acquisition module 2, safety protection log acquisition module 3, equipment health status assessment report acquisition module 4. Detailed Implementation

[0012] This application provides a method and system for safety monitoring of low-voltage electrical equipment based on temperature sensing, which solves the technical problems of lagging single-point sensing monitoring, inaccurate judgment of abnormal modes, and rigid safety protection strategies in traditional temperature monitoring of low-voltage electrical equipment.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] Example 1, as Figure 1 As shown, a safety monitoring method for low-voltage electrical equipment based on temperature sensing is provided, wherein the method includes:

[0016] Step A100: Collect real-time temperature data of multiple nodes of low-voltage electrical equipment through a distributed temperature sensor array to obtain a temperature monitoring dataset.

[0017] In this application embodiment, low-voltage electrical equipment refers to electrical devices or components used in low-voltage power distribution and consumption systems to realize power distribution, control, protection or connection, and whose operating voltage conforms to the low-voltage range, typically AC 1000V and below, DC 1500V and below, conforming to the industry's general low-voltage standards.

[0018] Specifically, a distributed temperature sensor array is used. Based on the structural characteristics and heat distribution patterns of low-voltage electrical equipment, multiple monitoring points are deployed in the core operating areas of the low-voltage electrical equipment, such as the circuit nodes inside the cabinet, the periphery of the heat dissipation channel, and the connection points of electrical components. Taking a common low-voltage distribution cabinet as an example, temperature sensors can be set at key locations such as the main circuit circuit breaker, branch circuit contactor, and cable joints to form a multi-node monitoring network covering the main heat-generating and heat-dissipating areas of the equipment.

[0019] Next, the distributed temperature sensor array will collect data according to a preset acquisition cycle. This cycle can be dynamically adjusted by those skilled in the art based on the specific operating conditions of the low-voltage electrical equipment. For example, when the equipment is operating under high load, the acquisition cycle is set to 10 seconds per acquisition to ensure timely capture of rapid temperature changes; when the equipment is operating under low load, the acquisition cycle is set to 30 seconds per acquisition to reduce energy consumption while ensuring data validity. During each acquisition, each temperature sensor's monitoring point accurately records the current temperature value. For example, the first temperature sensor acquires a temperature of 45℃ at the main circuit breaker, the second temperature sensor acquires a temperature of 42℃ at the branch circuit contactor, and the third temperature sensor acquires a temperature of 48℃ at the cable joint. These temperature data from different monitoring points are aggregated in real time through the data transmission module, forming a raw data set containing three elements: monitoring point number, acquisition timestamp, and temperature value. The data transmission module is the core component connecting the distributed temperature sensor array and the data processing terminal. Its core function is to realize the real-time transmission and aggregation of multi-node temperature data, and it is typically deployed inside the cabinet of the low-voltage electrical equipment.

[0020] Then, the compiled raw dataset needs to undergo preliminary verification and integration to remove abnormal data caused by temporary malfunctions of temperature sensors. If a temperature sensor's single-sample temperature value exceeds twice the normal operating temperature range of low-voltage electrical equipment (typically -10℃ to 80℃), such as a fourth temperature sensor sampling 150℃, the data is considered abnormal and removed. After verification, the valid data is categorized and sorted according to the sampling timestamp and monitoring point number, ultimately constructing a complete temperature monitoring dataset. This dataset not only covers the real-time temperature values ​​of each monitoring point but also implicitly includes key information such as temperature differences between different monitoring points and the temperature change trend of the same monitoring node over time, providing comprehensive and accurate data support for subsequent thermal state analysis based on the temperature monitoring dataset.

[0021] By deploying a distributed temperature sensor array in key areas of low-voltage electrical equipment, dynamically adjusting the acquisition cycle according to operating conditions, and collecting temperature data from multiple nodes, a comprehensive and valid temperature monitoring dataset is obtained after verification and integration. This lays a reliable data foundation for subsequent thermal status analysis and safety monitoring of low-voltage electrical equipment.

[0022] Step A200: Perform thermal state analysis based on the temperature monitoring dataset to obtain temperature change state parameters for temperature safety monitoring of low-voltage electrical equipment, and determine abnormal temperature change modes, which include multi-level early warning signals.

[0023] Optionally, firstly, calculate the real-time temperature rise rate value by iterating through the temperature change data of multiple monitoring points based on the temperature monitoring dataset. Then, calculate the temperature difference and analyze and determine the temperature distribution uniformity index by combining the temperature monitoring data. Use the temperature data stability coefficient as the period endpoint to calculate the time period and determine the thermal equilibrium time parameter. Finally, integrate these three to obtain the temperature change state parameter. The specific steps are explained in detail in A210-A240.

[0024] After obtaining the temperature change parameters, historical temperature records of low-voltage electrical equipment are retrieved and combined with thermal equilibrium time parameters to set multi-level temperature rise thresholds including the first and second temperature rise thresholds. Then, based on the relationship between the real-time temperature rise rate and the above two thresholds and the temperature distribution uniformity index, the gradual and abrupt temperature rise modes are determined. Next, based on the real-time temperature rise rate, local temperature rise analysis is performed on multiple monitoring points to determine the local overheating concentration mode. The specific steps are explained in detail in A250-A280.

[0025] Next, the abnormal temperature change mode generates multi-level early warning signals. The gradual temperature rise mode requires traversing multiple monitoring points to determine the abnormal points and generating the first early warning signal in stages. The sudden temperature rise mode requires analyzing the impact of temperature change according to the real-time temperature rise rate and generating the second early warning signal in stages. The local overheating concentration mode requires extracting the transformed position of the target monitoring point and then performing early warning analysis according to the coordinates of the overheating hotspot to generate the third early warning signal. The specific steps are explained in detail in A291-A293.

[0026] Step A300: According to the multi-level early warning signal trigger circuit protection device, dynamically adjust the operating parameters of low-voltage electrical equipment to implement the safety protection strategy and obtain the safety protection record log.

[0027] In one embodiment of this application, a range of safe operating parameters is set according to multi-level early warning signals. The circuit protection device is triggered to start preventive, active, and emergency protection strategies in response to the first, second, and third early warning signals, and the operating parameters of the low-voltage electrical equipment are dynamically adjusted according to the range to generate corresponding operating adjustment parameters. The change trajectory of these operating adjustment parameters is recorded to generate a data set of equipment operating status. The data set is then sorted according to the time series to construct a safety protection record log. The specific steps are described in detail in A310-A360.

[0028] Step A400: Based on the safety protection log, construct a fault prediction channel to perform operational prediction and assessment on low-voltage electrical equipment, and generate an equipment health status assessment report.

[0029] Specifically, based on the safety protection log, the abnormal event sequence is determined by parsing according to the timestamp. This sequence is then traced back to the historical temperature data collected by the distributed temperature sensor array for spatiotemporal mapping to establish an event-temperature correlation database. Based on this database, fault analysis is performed to build a fault prediction channel. Through this channel, the operating status of low-voltage electrical equipment is predicted and evaluated, and an equipment health status assessment report is generated. The specific steps are explained in detail in A410-A440.

[0030] Furthermore, step A200 in the method provided in this application embodiment includes:

[0031] A210: Based on the temperature monitoring dataset, calculate the real-time temperature rise rate by traversing the temperature change data of multiple monitoring points.

[0032] A220: Based on multiple monitoring points and the temperature monitoring data, calculate the temperature difference, perform distribution analysis based on the temperature difference data, and determine the temperature distribution uniformity index.

[0033] A230: Using the temperature data stability coefficient as the end point of the cycle, calculate the time period based on the temperature monitoring data to determine the thermal equilibrium time parameter.

[0034] A240: The real-time temperature rise rate value, the temperature distribution uniformity index, and the thermal equilibrium time parameter are integrated to obtain the temperature change state parameter.

[0035] Specifically, the real-time temperature rise rate is first calculated based on the temperature monitoring dataset. Specifically, continuous time-series temperature data from all monitoring points are extracted from the dataset. For example, the temperature at the first monitoring point is 25℃ at 0 minutes, 29℃ at 2 minutes, and 32℃ at 4 minutes; the temperature at the second monitoring point is 26℃ at 0 minutes, 30℃ at 2 minutes, and 33℃ at 4 minutes. By iterating through this type of temperature change data for each monitoring point, the real-time temperature rise rate is calculated using the formula (temperature at the next moment - temperature at the previous moment) / time interval. For example, the temperature rise rate at the first monitoring point from 0 to 2 minutes is (29-25) / 2 = 2℃ / minute, and the temperature rise rate from 2 to 4 minutes is (32-29) / 2 = 1.5℃ / minute. Finally, the calculation results from all monitoring points are summarized to form a set of real-time temperature rise rate values ​​covering multiple nodes.

[0036] After calculating the real-time temperature rise rate, the temperature distribution uniformity index is further determined based on temperature monitoring data from multiple monitoring points. Since a single temperature value cannot reflect the temperature differences between different areas of the low-voltage electrical equipment, this step selects temperature data from all monitoring points at the same time stamp. For example, if the temperatures of the first to fifth monitoring points at a certain moment are 32℃, 33℃, 31℃, 32℃, and 33℃ respectively, the average temperature of all monitoring points at that time stamp is first calculated as (32+33+31+32+33) / 5 = 32.2℃. Then, the absolute value of the temperature difference between each monitoring point and the corresponding average temperature is calculated: |32-32.2| = 0.2℃ for the first monitoring point and |33-32.2| = 0.8℃ for the second monitoring point. The third monitoring point has a temperature of |31-32.2|=1.2℃, the fourth monitoring point has a temperature of |32-32.2|=0.2℃, and the fifth monitoring point has a temperature of |33-32.2|=0.8℃. Then, the absolute values ​​of these temperature differences are summed and divided by the number of monitoring points to obtain the temperature distribution uniformity index at that time point, which is (0.2+0.8+1.2+0.2+0.8) / 5=0.64. The smaller the value of the temperature distribution uniformity index, the more uniform the temperature distribution at each monitoring point. By performing the above analysis process on multiple time points, dynamic change data of the temperature distribution uniformity of low-voltage electrical equipment is formed.

[0037] Next, the temperature data stability coefficient is set as the criterion for determining the end point of the cycle. That is, when the temperature change at all monitoring points is less than 0.5℃ / 10 minutes within a certain time period, the temperature data is considered to have reached a stable state. This temperature data stability coefficient can be adjusted and set by those skilled in the art based on the actual conditions of the low-voltage electrical equipment. Starting from the moment the equipment is started, the temperature changes at each monitoring point are continuously tracked. For example, within 0-10 minutes after equipment startup, the temperature at the monitoring point gradually rises from 25℃ to 32℃. Within 10-20 minutes, the temperature fluctuates within the range of 32℃±0.3℃. After 20 minutes, within three consecutive 10-minute time periods, the temperature change at all monitoring points does not exceed 0.5℃. At this point, the temperature data stability coefficient meets the standard. This moment is taken as the end point of the cycle, and the time period from the start-up of the low-voltage electrical equipment to reaching a stable state is calculated, which is 20 minutes. This time period is the thermal equilibrium time parameter.

[0038] Finally, the calculated real-time temperature rise rate, temperature distribution uniformity index, and thermal equilibrium time parameter are integrated. For example, the real-time temperature rise rate is integrated to an average of 2℃ / minute, the temperature distribution uniformity index is integrated to an average of 0.64, and the thermal equilibrium time parameter is integrated to a total of 20 minutes, forming a comprehensive data set that includes three dimensions: temperature change trend, distribution difference, and stable duration. This set is the temperature change state parameter.

[0039] By traversing multiple monitoring points to calculate the real-time temperature rise rate, analyzing the temperature difference to determine the temperature distribution uniformity index, and determining the thermal equilibrium time parameter based on the temperature data stability coefficient, and integrating the three, a temperature change state parameter that can comprehensively reflect the thermal state of low-voltage electrical equipment is obtained, providing accurate and multi-dimensional data support for subsequent identification of abnormal temperature change patterns of equipment.

[0040] Furthermore, step A200 in the method provided in this application embodiment includes:

[0041] A250: Retrieve historical temperature records of low-voltage electrical equipment and combine them with thermal equilibrium time parameters to set multi-level temperature rise thresholds, which include a first temperature rise threshold and a second temperature rise threshold.

[0042] A260: When the real-time temperature rise rate value continuously exceeds the first temperature rise threshold but does not exceed the second temperature rise threshold, and the temperature distribution uniformity index is in the uniform range, a gradual temperature rise mode is determined.

[0043] A270: When the real-time temperature rise rate exceeds the second temperature rise threshold and the temperature distribution uniformity index is in a downward trend, a sudden temperature rise mode is determined.

[0044] A280: Based on the real-time temperature rise rate value, perform local temperature rise analysis on multiple monitoring points to determine the local overheating concentration mode.

[0045] Optionally, firstly, retrieve historical temperature records of the low-voltage electrical equipment. For example, over the past year, the average temperature rise rate of a certain low-voltage electrical equipment under normal operating conditions has remained stable between 0.6℃ / min and 0.9℃ / min. Simultaneously, combine this with the thermal equilibrium time parameter determined in step A230 above. Assuming the thermal equilibrium time parameter of a certain low-voltage electrical equipment is 20 minutes, meaning the equipment enters a stable thermal state 20 minutes after startup, comprehensively set multiple temperature rise thresholds. The first temperature rise threshold is set to 1℃ / min, slightly higher than the highest temperature rise rate under historical normal operating conditions, used to initially identify minor temperature anomalies. The second temperature rise threshold is set to 2℃ / min, much higher than the temperature rise range under historical normal operating conditions, serving as a key criterion for detecting sudden anomalies, used to identify rapid, high-risk temperature changes. This threshold setting, combining historical temperature records and thermal equilibrium time parameters, lays the foundation for subsequent accurate differentiation of temperature change patterns, especially for detecting sudden anomalies.

[0046] After setting multi-level temperature rise thresholds, the gradual temperature rise mode is determined based on the real-time temperature rise rate and temperature distribution uniformity index in the temperature change state parameters. For example, if the real-time temperature rise rate of the low-voltage electrical equipment remains at 1.2℃ / min for 15 minutes within a certain monitoring period, this value exceeds the first temperature rise threshold of 1℃ / min but does not reach the second temperature rise threshold of 2℃ / min. Simultaneously, by calculating the temperature distribution uniformity index at multiple monitoring points, it is found that the index remains stable at 0.5 within this monitoring period, and the set temperature distribution uniformity range is 0.7 or below. This index being within this range indicates that the temperature change trend at each monitoring point is consistent, with no significant local differences. In this case, it can be determined that the low-voltage electrical equipment is in a gradual temperature rise mode. Under this mode, the temperature change rate is slow and the temperature distribution is uniform, not belonging to the category of abrupt anomalies. By comparing the real-time temperature rise rate with the threshold and using the uniformity index as an auxiliary judgment, abnormal temperature changes can be identified, while abrupt risks can be eliminated, achieving non-abrupt anomaly screening in abrupt anomaly retrieval.

[0047] Next, addressing the core requirement of abrupt temperature rise detection, the focus is on comparing the real-time temperature rise rate with the second temperature rise threshold and analyzing the trend changes in the temperature distribution uniformity index to determine abrupt temperature rise patterns. For example, at a certain moment, the real-time temperature rise rate of low-voltage electrical equipment suddenly rises to 2.5℃ / min, significantly exceeding the second temperature rise threshold of 2℃ / min. Simultaneously, reviewing the temperature distribution uniformity index from the previous 5 minutes reveals that it rapidly increases from 0.4 (within the uniform range) to 1.3, exceeding the uniform range, and the uniformity shows a continuous downward trend. This indicates that the temperature differences between monitoring points of the equipment are rapidly widening, with some areas experiencing a sharp temperature increase, consistent with typical characteristics of abrupt temperature rise anomalies. In this case, it can be determined that the low-voltage electrical equipment is in an abrupt temperature rise pattern. This pattern, through the dual conditions of exceeding the second threshold and decreasing uniformity, accurately identifies high-risk abrupt temperature rise anomalies, avoiding missed or false detections due to relying solely on the rate threshold, and ensuring timely identification of abrupt temperature rise anomalies.

[0048] In addition to detecting abrupt changes in overall temperature, it is also necessary to focus on abrupt temperature changes in localized areas. Then, based on real-time temperature rise rates, local temperature rise analysis is performed on multiple monitoring points to identify localized overheating concentration patterns. For example, a low-voltage electrical equipment has five monitoring points. During a certain monitoring cycle, the real-time temperature rise rate of the third monitoring point reaches 1.9℃ / min, close to but not exceeding the second temperature rise threshold, while the real-time temperature rise rates of the other four monitoring points are stable at around 0.8℃ / min, within the historical normal operating temperature rise range. Analysis of the temperature rise rate difference between the third monitoring point and other monitoring points reveals that its temperature rise rate is more than twice that of the other monitoring points, and this difference continues to widen. In this case, it can be determined that the low-voltage electrical equipment is in a localized overheating concentration pattern. Although this pattern does not reach the second temperature rise threshold for overall abrupt changes, the abnormal temperature changes at local monitoring points are prominent, representing a localized abrupt change risk. By analyzing the real-time temperature rise rate of individual monitoring points separately and comparing multiple monitoring points, accurate detection of localized abrupt changes is achieved, compensating for the shortcomings of only focusing on overall temperature changes while ignoring localized abrupt changes.

[0049] By identifying three abnormal temperature change modes—gradual, abrupt, and localized overheating concentration—and especially by using a second temperature rise threshold plus uniformity decrease and local rate difference analysis, the system achieved accurate retrieval of abrupt anomalies, providing a clear and reliable basis for subsequent triggering of corresponding early warning signals for different abnormal modes.

[0050] Furthermore, step A200 in the method provided in this application embodiment further includes step A290, wherein step A290 includes:

[0051] A291: When the abnormal temperature change mode is the gradual temperature rise mode, multiple monitoring points are traversed to determine the abnormal point location, and the early warning level is classified according to the abnormal point location to generate the first early warning signal.

[0052] A292: When the abnormal temperature change mode is the sudden temperature rise mode, perform temperature change impact analysis based on the real-time temperature rise rate value, classify the early warning according to the impact range, and generate a second early warning signal.

[0053] A293: When the abnormal temperature change mode is the local overheating concentration mode, the target monitoring point is extracted and its location is transformed. Based on the coordinates of the overheating hot spot, an early warning analysis is performed to generate a third early warning signal.

[0054] Specifically, for a gradual temperature rise pattern, the system first iterates through all monitoring points covered by the distributed temperature sensor array. It extracts the real-time temperature rise rate and temperature distribution uniformity index for each monitoring point from the temperature change parameters. For example, if a low-voltage electrical device has 10 monitoring points, after iteration, it is found that the real-time temperature rise rate of the first, fifth, and seventh monitoring points consistently remains at 1.2℃ / min, exceeding the first temperature rise threshold of 1℃ / min but not reaching the second temperature rise threshold of 2℃ / min. All other monitoring points are within the normal range, thus identifying these three monitoring points as abnormal points. Subsequently, a warning level is determined based on the number of abnormal points: a single abnormal point is classified as Level 1 Warning, 2-4 abnormal points as Level 2 Warning, and 5 or more abnormal points as Level 3 Warning. Combining the current situation of the three abnormal points, a Level 2 Warning signal is finally generated, ensuring that the warning information clearly identifies the abnormal pattern and reflects the degree of abnormal spread.

[0055] When low-voltage electrical equipment is in a sudden temperature rise mode, the first step is to conduct a temperature change impact analysis based on the real-time temperature rise rate. This involves extracting the temperature change curves of each monitoring point and their correlation with the equipment circuits during the abnormal period from the temperature monitoring dataset. For example, if the real-time temperature rise rate of the low-voltage electrical equipment suddenly rises to 2.5℃ / minute, exceeding the second temperature rise threshold of 2℃ / minute, analysis reveals that this temperature change has caused a simultaneous and rapid temperature rise in the main circuit where monitoring points one through four are located, and the branch circuits where monitoring points five and six are located, while the backup circuits where monitoring points seven through ten are currently unaffected. Next, a warning level is established based on the number and importance of the affected circuits. A Level 2 warning is set if only a single non-core circuit is affected, while a Level 1 warning is set if multiple core circuits or the main circuit are affected. Combining the current status of the main circuit and branch circuits affected, a Level 1 warning signal is generated. This allows the warning information to intuitively reflect the threat level of sudden temperature changes to the operation of low-voltage electrical equipment, providing a basis for the formulation of subsequent proactive protection strategies.

[0056] For the localized overheating concentration pattern, the target monitoring point with significantly abnormal real-time temperature rise rate is first extracted from the temperature change parameters. For example, the real-time temperature rise rate of the sixth monitoring point reaches 1.8℃ / min, close to the second temperature rise threshold of 2℃ / min, while other monitoring points are stable at around 0.8℃ / min. The sixth monitoring point is then identified as the target monitoring point. Subsequently, the monitoring point's number is converted into its location information in the physical coordinate system of the low-voltage electrical equipment. Based on the design drawings of the low-voltage electrical equipment, it is assumed that the sixth monitoring point corresponds to the cable joint location on the right side of the middle layer inside the low-voltage electrical equipment cabinet, with coordinates (X, Y) in cm. Further, an early warning analysis is conducted by combining the coordinates of the overheating point with the location of key components of the low-voltage electrical equipment. A third-level warning (level two) is set if the overheating point is more than 5cm away from core components, including circuit breakers and contactors; a third-level warning (level one) is set if the overheating point is 5cm or less away from the core components. Measurements show that the distance between the coordinates of the sixth monitoring point and the adjacent circuit breaker is 3cm. Therefore, a third-level warning signal is generated, ensuring that the warning information accurately points to the abnormal location and reflects the potential threat of localized overheating to core components.

[0057] By designing differentiated early warning generation logic for different abnormal temperature change modes, the gradual temperature rise mode is classified according to the number of abnormal points, the sudden temperature rise mode is classified according to the scope of influence, and the local overheating concentration mode is classified according to the distance between the coordinates of the overheating hotspot and the core components, generating first, second, and third early warning signals respectively. This provides accurate and implementable instruction basis for subsequent circuit protection devices to execute appropriate safety protection strategies.

[0058] Furthermore, step A300 in the method provided in this application embodiment includes:

[0059] A310: Set the range of safe operating parameters based on the multi-level early warning signals.

[0060] A320: In response to the first warning signal, the circuit protection device is triggered to start a preventive protection strategy to dynamically adjust the operating parameters of the low-voltage electrical equipment according to the safe operating parameter range, and generate the first operating adjustment parameters.

[0061] A330: In response to the second warning signal, the circuit protection device is triggered to start the active protection strategy to dynamically adjust the operating parameters of the low-voltage electrical equipment according to the safe operating parameter range, and generate the second operating adjustment parameters.

[0062] A340: In response to the third warning signal, the circuit protection device is triggered to activate the emergency protection strategy to dynamically adjust the operating parameters of the low-voltage electrical equipment according to the safe operating parameter range, and generate the third operating adjustment parameters.

[0063] A350: Based on the first operating adjustment parameter, the second operating adjustment parameter, and the third operating adjustment parameter, the parameter change trajectory is recorded to generate a device operating status dataset.

[0064] A360: Sort the device operating status dataset according to time series to construct the security protection record log.

[0065] Specifically, the safe operating parameter range is first set based on multi-level warning signals: taking a low-voltage switchgear as an example, combined with its rated current of 30A and rated temperature of 60℃, the following parameters are set: when the first warning is triggered, the current must be ≤25A and the temperature ≤55℃; when the second warning is triggered, the current must be ≤20A and the temperature ≤50℃; when the third warning is triggered, the current must be ≤15A and the temperature ≤45℃. By defining the safety boundaries in a graded manner, the subsequent parameter adjustments are both based on a basis and match the risk level.

[0066] When the first warning signal is received, the circuit protection device is triggered to activate a preventative protection strategy. For example, if a feeder cabinet triggers the first warning level 2 due to a gradual temperature rise, and there are three abnormal monitoring points, the circuit protection device dynamically adjusts the operating parameters based on the safe operating parameter range. If the current operating current is 28A and the temperature is 58℃, the protection device gradually reduces the load distribution, stabilizing the current to 24A, and simultaneously activates auxiliary cooling to stabilize the temperature at 54℃. These adjusted parameters, such as the current of 24A and the temperature of 54℃, are recorded as the first operating adjustment parameters. This preventative adjustment provides gentle intervention in the early stages of an anomaly, preventing the temperature from worsening and maintaining equipment operation to the greatest extent possible, thus compensating for the shortcomings of not protecting without warning and shutting down when a problem occurs.

[0067] If the second warning signal is received, an active protection strategy is triggered. For example, if a capacitor bank triggers the second warning level one due to a sudden temperature rise, both its main circuit and branch circuits are affected by the temperature change. The device quickly adjusts according to the range of current ≤20A and temperature ≤50℃: at the current of 32A and temperature of 62℃, the circuit protection device immediately disconnects some unnecessary capacitor banks, causing the current to drop sharply to 19A. At the same time, the cooling fan starts at full power, reducing the temperature to 49℃ within 2 minutes. These parameters of 19A current and 49℃ temperature are stored as the second operating adjustment parameters. The active protection strategy intervenes proactively before the sudden risk spreads, improving response efficiency and preventing the fault from escalating from a sudden change to a larger fault.

[0068] When a third early warning signal is received, an emergency protection strategy is triggered. For example, if a busbar cabinet triggers the third early warning level 1 due to a localized overheating centralized mode, with the overheated spot only 3cm away from the circuit breaker, the device will intervene urgently based on the current ≤15A and temperature ≤45℃ range: At the current of 26A and temperature of 56℃, the protection device instantly activates the backup cooling circuit and limits the main bus current, reducing the current to 14A within 0.5 minutes and rapidly lowering the temperature to 44℃. These parameters of 14A current and 44℃ become the third operational adjustment parameters. The emergency protection strategy achieves millisecond-level response to localized high-risk overheating, solving the problems of slow response and weak measures when dealing with sudden localized overheating, and minimizing the risk of damage to core components.

[0069] Subsequently, based on the first, second, and third operational adjustment parameters, the parameter change trajectory was recorded to generate a device operation status dataset: the corresponding time adjustment curve of the current from 28A to 24A in the first operational adjustment, the heat dissipation process of the temperature from 58℃ to 54℃, the moment of the sudden drop of the current from 32A to 19A in the second operational adjustment, the duration of the rapid cooling from 62℃ to 49℃, and the moment of the emergency limit of the current from 26A to 14A in the third operational adjustment, the extreme cooling amplitude of the temperature from 56℃ to 44℃ were collected and integrated one by one according to the time point to form a device operation status dataset containing multiple parameters such as current and temperature changing over time, completely restoring the dynamic changes of parameters under each protection strategy.

[0070] Finally, the equipment operation status dataset is sorted by time series to construct a safety protection record log: based on timestamps, the information such as the first warning triggered at time T1 and the current adjustment taking 3 minutes, the second warning triggered at time T2 and the current adjustment taking 1 minute, and the third warning triggered at time T3 and the current adjustment taking 0.5 minutes are arranged in sequence to form a traceable safety protection record log, clearly presenting the entire process from the triggering of the warning to the completion of the parameter adjustment.

[0071] By setting safe operating parameter ranges according to multi-level early warning signals, and then triggering corresponding protection strategies to dynamically adjust operating parameters, record parameter change trajectories, and build logs according to time series in response to different early warning signals, dynamic control and traceable recording of the entire process of low-voltage electrical equipment from early warning to protection are realized. This solves the problems of fixed traditional protection strategies and lack of recording of the protection process, and provides accurate basis for subsequent fault analysis and equipment health assessment.

[0072] Furthermore, step A400 in the method provided in this application embodiment includes:

[0073] A410: Based on the security protection log, the sequence of abnormal events is determined by parsing the timestamps.

[0074] A420: The abnormal event sequence is traced back to the distributed temperature sensor array to collect historical temperature data and perform spatiotemporal mapping to establish an event-temperature correlation database.

[0075] A430: Based on the event-temperature correlation database, perform fault analysis and construct a fault prediction channel.

[0076] A440: The operating status of low-voltage electrical equipment is predicted and evaluated through the fault prediction channel, and a health status assessment report of the equipment is generated.

[0077] Specifically, the first step is to obtain the safety protection log constructed in the aforementioned steps. This log has integrated all the information on early warning triggers, protection strategy executions, and operating parameter adjustments of low-voltage electrical equipment in a time series. Each record is accompanied by a timestamp accurate to the second. These entries cover the time, type, handling measures, and parameter changes of the anomaly, providing complete data support for time-stamp-based analysis.

[0078] Next, the security protection logs are systematically analyzed according to their timestamps. First, all records in the logs are sorted from earliest to latest by timestamp, removing duplicate records caused by system caching. For example, two identical records with the same timestamp and the same warning event are excluded to ensure the accuracy of the parsing sequence. Then, for each sorted record, core information related to the abnormal event is extracted: including the timestamp, the type of warning signal triggered, the corresponding abnormal temperature change mode, the specific values ​​of the operating parameters before and after adjustment, and the type of security protection strategy executed. For example, when analyzing a particular record, the timestamp is extracted to the second, the warning signal type is the first warning, the abnormal temperature change mode is a gradual temperature rise mode, the current was 27A before adjustment and 23A after adjustment, the temperature was 57℃ before adjustment and 53℃ after adjustment, and the protection strategy is preventative protection. The core information for all records is extracted accordingly.

[0079] Then, valid abnormal events are filtered based on the extracted core information. The criteria for judging abnormal events are set as follows: as long as the record contains information on the triggering of a warning signal and the execution of the corresponding safety protection strategy, it is considered a valid abnormal event and is selected as an abnormal event entry to be processed. Entries in the log that only record normal operating parameters without warnings or protection actions are excluded. The filtered valid abnormal event entries are arranged sequentially by timestamp to form an abnormal event sequence. This sequence clearly presents the order of occurrence, specific types, and corresponding handling details of abnormal events in low-voltage electrical equipment within a specific time period.

[0080] Next, based on the timestamps of the abnormal event sequence, historical temperature data collected by the distributed temperature sensor array is extracted. The sequence is traversed to analyze the time period of the event occurrence and multiple time windows are set. The historical temperature data is mapped to these time windows according to the timestamp to build the event-temperature correlation. Then, the spatiotemporal data is indexed and integrated according to the relationship to establish the event-temperature correlation database. The specific steps are explained in detail in A421-A424.

[0081] Subsequently, historical abnormal temperature change parameters are extracted from the event-temperature correlation database. A temperature anomaly feature library is constructed through feature analysis. Temperature change trajectories are determined by temperature change search based on this database. The trajectories are mapped to the feature library to obtain the fault development time series. The fault is predicted according to the series to obtain initial fault prediction parameters and confidence values ​​are generated by confidence assessment. Finally, the initial parameters are selected and trained based on the parameter confidence values ​​to construct the fault prediction channel. The specific steps are explained in detail in A431-A435.

[0082] Finally, temperature anomalies in low-voltage electrical equipment are identified through the fault prediction channel to determine the temperature anomaly pattern. Based on this pattern, the fault prediction channel predicts the development of equipment faults and generates a fault development prediction trajectory. Real-time temperature data is matched with the temperature anomaly pattern to determine the real-time abnormal operating temperature parameters. The fault evolution is analyzed through the fault development prediction trajectory to determine the fault evolution path. The real-time abnormal operating temperature parameters and the fault evolution path are combined to evaluate the equipment operation and generate an equipment health index. The equipment operating status is then predicted based on the health index, and an equipment health status assessment report is constructed. The specific steps are explained in detail in A441-A446.

[0083] Furthermore, step A420 in the method provided in this application embodiment includes:

[0084] A421: Extract historical temperature data collected by the distributed temperature sensor array based on the timestamps of the abnormal event sequence.

[0085] A422: Traverse the sequence of abnormal events to analyze the time periods in which the events occurred, and set multiple time windows.

[0086] A423: Map the historical temperature data to the multiple time windows according to the timestamp for matching, and build an event-temperature correlation relationship.

[0087] A424: Based on the event-temperature correlation, perform spatiotemporal data indexing and integration to construct the event-temperature correlation database.

[0088] In one embodiment, historical temperature data is first extracted based on the timestamps in the abnormal event sequence: First, the precise timestamps of each event in the abnormal event sequence are sorted out, including the event trigger time and the time when the protection strategy is completed, so as to determine the time range for data extraction; then, the historical data storage module of the distributed temperature sensor array is called, and the historical temperature data collected by all monitoring points within this time range is extracted through the timestamp filtering function, ensuring that the extracted data not only covers the complete time period of the event, but also includes the basic temperature data before the event, providing a complete data source for subsequent analysis of the temperature change correlation before and after the event, and avoiding the loss of correlation information due to the narrow data extraction range.

[0089] Next, the sequence of abnormal events is traversed to analyze the time periods of the events, setting multiple time windows: Each abnormal event is analyzed individually using a per-event parsing approach, considering the time from the triggering of the warning signal to the adjustment of operating parameters to a safe range. This is combined with the lag effect of temperature changes to divide each event into multiple time windows. For example, for a gradual temperature rise event lasting 3 minutes, the time window is set as follows: 1 minute before the event to compare with the normal temperature state before the event; 3 minutes during the event to cover the entire process; and 2 minutes after the event to track the temperature recovery after the event is handled. Each window is labeled with a corresponding event identifier, such as the event number and abnormal mode type. Through multi-window division, accurate classification of temperature change data at different stages of the event is achieved.

[0090] Subsequently, historical temperature data is mapped to multiple time windows for matching based on timestamps. First, a dedicated data mapping rule is established for each time window; that is, the window interval to which the temperature data belongs is determined based on the timestamp. Then, the temperature data of all monitoring points within the same window are bound to the event identifier marked in that window. For example, the temperature data of each monitoring point within a window one minute before an event is associated with the corresponding gradual temperature rise event number and the first warning signal type, while simultaneously recording the location information of the monitoring points corresponding to the data. During the mapping process, data verification methods are used to check for timestamp mismatches or missing data, ensuring that the temperature data within each window can be accurately associated with the corresponding event information, forming an event-temperature correlation and clarifying the direct correspondence between events and temperature change data.

[0091] Finally, spatiotemporal data indexing and integration are performed based on the event-temperature correlation. Data indexes are established along both time and space dimensions. The time dimension sorts the correlated data by timestamp order to form a time index, while the spatial dimension categorizes temperature data within the same time window by monitoring point number to form a spatial index. A structured database construction tool is used to transform the correlated data with dual-dimensional indexes into structured storage tables. These tables include an event information table recording event number, anomaly mode, and time window range; a temperature data table recording monitoring point number, timestamp, and temperature value; and a correlation mapping table that links the first two tables through event number and time window, ultimately constructing the required event-temperature correlation database. A data query interface is also set up to support quick retrieval of corresponding temperature data by event identifier, time range, or monitoring point number, enabling efficient management and retrieval of correlated data.

[0092] By extracting historical temperature data by timestamp, setting time windows in stages, accurately mapping and constructing event-temperature correlations, and integrating data through spatiotemporal indexing, a structured event-temperature correlation database was built. This solved the problem of the disconnect between abnormal events and temperature data, and provided complete and traceable correlation data for subsequent fault analysis based on historical temperature change characteristics.

[0093] Furthermore, step A430 in the method provided in this application embodiment includes:

[0094] A431: Extract historical abnormal temperature change parameters from the historical temperature data, perform feature analysis based on the historical abnormal temperature change parameters, and construct a temperature anomaly feature library.

[0095] A432: Perform a temperature change search based on the event-temperature correlation database to determine the temperature change trajectory.

[0096] A433: Map the temperature change trajectory to a temperature anomaly feature library to obtain a fault development time series.

[0097] A434: Perform fault prediction according to the fault development time series, obtain initial fault prediction parameters, evaluate confidence level, and generate parameter confidence values.

[0098] A435: The initial fault prediction parameters are filtered and trained based on the confidence values ​​of the parameters to construct the fault prediction channel.

[0099] Optionally, firstly, extract historical abnormal temperature change parameters from historical temperature data, retrieve temperature data corresponding to all historical abnormal events from the event-temperature correlation database, focus on key parameters directly related to abnormal temperature changes, such as peak temperature rise rate, duration of temperature change, and maximum fluctuation range of temperature distribution uniformity under different abnormal modes, and remove abnormal parameter values ​​caused by temperature sensor errors through data cleaning methods to ensure that the extracted parameters truly reflect the abnormal temperature change characteristics of low-voltage electrical equipment.

[0100] Subsequently, feature analysis was performed on these historical abnormal temperature change parameters. The K-means clustering algorithm was used to group abnormal temperature changes with similar parameters into the same feature category. For example, parameters with a slow temperature rise rate and small uniformity fluctuations were clustered into the gradual temperature change feature category, while parameters with a sudden temperature rise rate and rapid uniformity decrease were clustered into the abrupt temperature change feature category. Each feature category was labeled with a corresponding abnormal mode label and typical fault type, such as poor contact or short circuit precursor. Finally, a temperature anomaly feature library containing multiple types of temperature change features and fault association information was constructed, providing a structured feature reference for subsequent fault identification.

[0101] After constructing the temperature anomaly feature database, temperature change searches are performed based on the event-temperature correlation database. The specific process is as follows: Using time-series retrieval, the temperature data of a specific monitoring point or the entire low-voltage electrical equipment within a specific time period are traversed in chronological order. For example, the temperature records before and after all abnormal events in a low-voltage distribution cabinet over the past three months are retrieved. Through data visualization and trajectory fitting techniques, discrete temperature data points are connected into a continuous temperature change curve. Key time nodes, such as the trigger time of warning signals and the execution time of protection strategies, are marked on the curve, along with corresponding event information, such as the type of abnormal mode and the adjustment of operating parameters. In this way, isolated temperature data is transformed into temperature change trajectories strongly correlated with events, clearly presenting the complete process of abnormal temperature changes from occurrence and development to control, providing intuitive trajectory evidence for subsequent fault development analysis.

[0102] Next, the temperature change trajectory is mapped to a temperature anomaly feature database. A cosine similarity feature matching algorithm is used to compare the similarity of the currently determined temperature change trajectory with the typical trajectories of each category in the temperature anomaly feature database, selecting the feature category with the highest matching degree. Historical fault development data corresponding to this feature category is retrieved, and key time nodes are extracted, such as the moment when the temperature change begins to deviate from the normal range, the moment the first temperature rise threshold is reached, the moment the protection strategy is triggered, and the moment the temperature change stabilizes. These time nodes are organized into a fault development time series in chronological order. This series not only contains time information but also associates the temperature change parameters and event status corresponding to each node. For example, at time T1: the temperature change begins to deviate, with a temperature rise rate of 0.8℃ / minute; at time T2: the first threshold is reached, triggering the first warning, thus providing a referable time dimension for fault prediction.

[0103] Subsequently, fault prediction is performed according to the fault development time series to obtain initial fault prediction parameters and conduct confidence assessment: An LSTM time series prediction model is used to construct the fault prediction model, using historical fault development time series as training samples. Real-time temperature change data of the current low-voltage electrical equipment and information on some existing fault nodes are input to predict key parameters for future fault development, such as the expected time to reach the second temperature rise threshold and the rate at which the fault's impact area expands. These parameters constitute the initial fault prediction parameters. To ensure parameter reliability, cross-validation is used for confidence assessment. Historical data is divided into training and test sets. The error between the LSTM time series prediction model's prediction results and the actual results in the test set is calculated to generate parameter confidence values. For example, if the error rate of a certain initial fault prediction parameter is 5%, the corresponding confidence value is 95%, thus quantifying the reliability of each initial fault prediction parameter.

[0104] Finally, the initial fault prediction parameters are screened and trained based on the parameter confidence values ​​to construct a fault prediction channel. The specific steps are as follows: A confidence threshold of 80% is set, and initial fault prediction parameters with confidence values ​​higher than this threshold are selected, while low-confidence parameters are removed to avoid affecting prediction accuracy. The selected reliable parameters are used as training samples to iteratively optimize the fault prediction model, adjusting the weights of the neural network and the lag order of the time series model, etc., so that the model can more accurately match temperature change characteristics and fault development patterns in subsequent predictions. After multiple rounds of screening and training, a set of processes and model combinations that can stably receive temperature change data and output reliable fault prediction results is formed, namely the fault prediction channel. This channel can realize automated processing from temperature change data input to fault prediction result output.

[0105] By extracting temperature variation parameters to build a feature library, searching for trajectory-related features, predicting parameters and evaluating confidence, and selecting and training optimized models, combined with data cleaning, cluster analysis, feature matching, cross-validation and other technical means, the problems of lack of structured feature support and low parameter reliability in fault prediction are solved. Finally, an accurate and stable fault prediction channel is built, providing a reliable technical path for early warning of faults in low-voltage electrical equipment.

[0106] Furthermore, step A440 in the method provided in this application embodiment includes:

[0107] A441: Based on the fault prediction channel, identify temperature anomalies in low-voltage equipment and determine the temperature anomaly mode.

[0108] A442: Based on the temperature anomaly pattern, the fault development of low-voltage electrical equipment is predicted through the fault prediction channel, and a fault development prediction trajectory is generated.

[0109] A443: Based on the temperature anomaly pattern, match the real-time temperature data to determine the real-time abnormal operating temperature parameters.

[0110] A444: The fault evolution path is determined by analyzing the fault development prediction trajectory.

[0111] A445: Based on the real-time abnormal operating temperature parameters and the fault evolution path, perform an operational assessment of the low-voltage electrical equipment and generate an equipment health index.

[0112] A446: Based on the equipment health index, predict the operating status of low-voltage electrical equipment and construct the equipment health status assessment report.

[0113] In one embodiment, temperature anomaly identification is first performed based on the fault prediction channel: A pre-stored temperature anomaly feature library within this channel is invoked, and real-time temperature change data collected by a distributed temperature sensor array is input into the channel. A cosine similarity feature matching algorithm is used to compare the data with typical features of three types of patterns in the temperature anomaly feature library: gradual, abrupt, and localized overheating concentration. The feature category with the highest matching degree is selected to determine the current temperature anomaly pattern. For example, if real-time data shows a slow temperature rise rate exceeding a first threshold and stable temperature distribution uniformity, it is matched as a gradual temperature rise pattern. This ensures that the identification result is associated with historical features, avoiding misjudgments caused by a single parameter and laying the foundation for subsequent accurate prediction.

[0114] Next, based on the established temperature anomaly pattern, the fault development is predicted through a fault prediction channel. This involves calling the LSTM time series prediction model within the channel, inputting the current anomaly pattern and real-time temperature change data, and simultaneously retrieving historical fault development data under similar anomaly patterns from an event-temperature correlation database as training references. The model learns the correlation between temperature change and fault in historical data to simulate the current fault's trend over time. For example, for abrupt temperature rise patterns, the model predicts the possible peak value of the temperature rise rate and the decrease in temperature distribution uniformity over a future period. These prediction nodes are then concatenated chronologically to generate a fault development prediction trajectory, visually presenting the dynamic process of the fault progressing from its current state to a potentially worsening state, thus overcoming the lack of dynamic prediction in existing technologies.

[0115] Then, based on the identified temperature anomaly patterns, real-time temperature data is matched to determine the real-time operational anomaly temperature parameters. The specific process is as follows: Using the current anomaly pattern as the screening criterion, key parameters directly related to this pattern are extracted from the real-time temperature change data of the distributed temperature sensor array. For example, for a localized overheating concentration pattern, the real-time temperature, temperature rise rate, and temperature difference data between the overheating hotspot monitoring points and surrounding monitoring points are extracted. Next, normal parameters unrelated to the current pattern, such as normal temperature values ​​of non-overheating hotspots, are removed to ensure that the finally determined real-time operational anomaly temperature parameters accurately correspond to the current anomaly pattern, avoiding irrelevant data from interfering with subsequent evaluations.

[0116] Subsequently, fault evolution analysis is conducted using the predicted fault development trajectory. Centered on this trajectory and combined with historical fault evolution cases stored in the fault prediction channel, data can be retrieved from the event-temperature correlation database. Trajectory backtracking and logical deduction methods are employed to analyze the current fault from its initial temperature change to potential chain reactions. For example, for a localized overheating concentration mode, trajectory analysis reveals that a sustained increase in the temperature of the overheated spot may first lead to aging of the local insulation layer, subsequently increasing contact resistance, and ultimately potentially developing into a short-circuit fault. These potential stages are then arranged chronologically to form a fault evolution path, clarifying the risk points at different stages of the fault and providing a logical basis for assessment.

[0117] Subsequently, based on real-time abnormal temperature parameters and fault evolution paths, the operation of low-voltage electrical equipment is evaluated. A quantitative weighted algorithm can be used to assign values ​​to the temperature exceedance value, temperature rise rate exceedance magnitude, and fault evolution path in the real-time abnormal temperature parameters. For example, a parameter exceedance magnitude weight of 60% and an evolution stage risk weight of 40% can be set. Those skilled in the art can score the parameter exceedance value according to its degree, such as 30 points for exceeding the limit by 3℃ and 70 points for exceeding the limit by 8℃. The evolution stage can be scored according to its risk level, such as 30 points for insulation aging and 60 points for poor contact. The weighted calculation yields the equipment health index, for example, 70×60%+60×40%=66 points. This transforms the qualitative assessment into a quantitative result, where a higher health index indicates a more severe current abnormality of the low-voltage electrical equipment and a higher potential risk of fault evolution.

[0118] Finally, based on the equipment health index, the operating status of low-voltage electrical equipment is predicted. A structured report generation tool is used, with the equipment health index at its core, integrating abnormal temperature patterns, fault development prediction trajectories, real-time abnormal operating temperature parameters, and fault evolution paths. The content is organized logically as follows: anomaly identification results – fault prediction trends – current abnormal parameters – potential evolution risks – quantitative health assessment – ​​future status prediction, thus constructing an equipment health status assessment report. For example, the report indicates the status level corresponding to a health index score of 66, such as a sub-healthy state, requiring inspection of localized overheating areas within 72 hours, predicting the potential risk stage within the next 24 hours without intervention, and attaching key data charts, such as temperature change trajectory diagrams, to ensure the report's completeness and guiding significance, providing a clear basis for the maintenance of low-voltage electrical equipment.

[0119] The above steps provide a systematic and accurate technical approach for predicting and assessing the operating status of low-voltage electrical equipment, effectively supporting equipment safety management and maintenance decisions.

[0120] In summary, the low-voltage electrical equipment safety monitoring method based on temperature sensing provided in this application has the following technical effects:

[0121] This application utilizes a distributed temperature sensor array deployed on low-voltage electrical equipment to collect temperature data. This data is then processed through abnormal temperature change pattern identification, multi-level early warning signal generation, and dynamic adjustment of safety protection strategies to obtain equipment operation and protection-related data. Combined with the construction of an event-temperature correlation database and the establishment of a fault prediction channel, the application performs operational prediction and assessment of the equipment and generates an equipment health status assessment report. This allows for precise monitoring of temperature anomalies and fault development in low-voltage electrical equipment, making safety monitoring more accurate and reliable. The application achieves the technical effect of comprehensively collecting multi-node data from the distributed sensor array, accurately analyzing thermal status, and identifying abnormal temperature change patterns, thereby realizing precise temperature monitoring, effective protection, and health prediction for low-voltage electrical equipment.

[0122] Example 2, as Figure 2 As shown, based on the same inventive concept as in Embodiment 1 above, this application provides a low-voltage electrical equipment safety monitoring system based on temperature sensing, the system comprising:

[0123] Temperature monitoring dataset acquisition module 1 is used to collect real-time temperature data of multiple nodes of low-voltage electrical equipment through a distributed temperature sensor array to obtain a temperature monitoring dataset.

[0124] Abnormal temperature change mode acquisition module 2 is used to perform thermal state analysis based on the temperature monitoring dataset, obtain temperature change state parameters for temperature safety monitoring of low-voltage electrical equipment, and determine abnormal temperature change modes, which include multi-level early warning signals.

[0125] The safety protection log acquisition module 3 is used to dynamically adjust the operating parameters of low-voltage electrical equipment according to the safety protection strategy executed by the multi-level early warning signal triggering circuit protection device, and obtain the safety protection log.

[0126] Equipment health status assessment report acquisition module 4 is used to construct a fault prediction channel based on the safety protection record log to perform operation prediction assessment on low voltage electrical equipment and generate an equipment health status assessment report.

[0127] Furthermore, the abnormal temperature change mode acquisition module 2 is used to perform the following steps:

[0128] The real-time temperature rise rate is calculated by traversing temperature change data from multiple monitoring points based on the temperature monitoring dataset. Temperature difference is calculated based on the temperature monitoring data from multiple monitoring points, and distribution analysis is performed to determine the temperature distribution uniformity index. The temperature data stability coefficient is used as the period endpoint, and the time period is calculated based on the temperature monitoring data to determine the thermal equilibrium time parameter. The real-time temperature rise rate, the temperature distribution uniformity index, and the thermal equilibrium time parameter are integrated to obtain the temperature change state parameter.

[0129] Furthermore, the abnormal temperature change mode acquisition module 2 is used to perform the following steps:

[0130] Historical temperature records of low-voltage electrical equipment are retrieved and combined with thermal equilibrium time parameters to set multi-level temperature rise thresholds, including a first temperature rise threshold and a second temperature rise threshold. When the real-time temperature rise rate continuously exceeds the first temperature rise threshold but does not exceed the second temperature rise threshold, and the temperature distribution uniformity index is in a uniform range, a gradual temperature rise mode is determined. When the real-time temperature rise rate exceeds the second temperature rise threshold, and the temperature distribution uniformity index is in a downward trend, a sudden temperature rise mode is determined. Based on the real-time temperature rise rate value, local temperature rise analysis is performed on multiple monitoring points to determine a local overheating concentration mode.

[0131] Furthermore, the abnormal temperature change mode acquisition module 2 is used to perform the following steps:

[0132] When the abnormal temperature change mode is the gradual temperature rise mode, multiple monitoring points are traversed to determine the abnormal location, and the early warning level is determined according to the abnormal location to generate a first early warning signal; when the abnormal temperature change mode is the sudden temperature rise mode, the temperature change impact is analyzed according to the real-time temperature rise rate value, and the early warning level is determined according to the impact range to generate a second early warning signal; when the abnormal temperature change mode is the local overheating concentration mode, the target monitoring point is extracted and its location is transformed, and the early warning analysis is performed according to the coordinates of the overheating hotspot to generate a third early warning signal.

[0133] Furthermore, the security protection log acquisition module 3 is used to perform the following steps:

[0134] The safe operating parameter range is set according to the multi-level early warning signals; in response to the first early warning signal, the circuit protection device is triggered to activate a preventive protection strategy to dynamically adjust the operating parameters of the low-voltage electrical equipment according to the safe operating parameter range, generating a first operating adjustment parameter; in response to the second early warning signal, the circuit protection device is triggered to activate an active protection strategy to dynamically adjust the operating parameters of the low-voltage electrical equipment according to the safe operating parameter range, generating a second operating adjustment parameter; in response to the third early warning signal, the circuit protection device is triggered to activate an emergency protection strategy to dynamically adjust the operating parameters of the low-voltage electrical equipment according to the safe operating parameter range, generating a third operating adjustment parameter; the parameter change trajectory is recorded based on the first operating adjustment parameter, the second operating adjustment parameter, and the third operating adjustment parameter to generate an equipment operating status dataset; the equipment operating status dataset is sorted according to time series to construct the safety protection record log.

[0135] Furthermore, the equipment health status assessment report acquisition module 4 is used to perform the following steps:

[0136] Based on the security protection log, the abnormal event sequence is determined by parsing the timestamps. The abnormal event sequence is then traced back to the distributed temperature sensor array and historical temperature data for spatiotemporal mapping to establish an event-temperature correlation database. Fault analysis is performed based on the event-temperature correlation database to construct a fault prediction channel. The operating status of low-voltage electrical equipment is predicted and evaluated through the fault prediction channel to generate a health status assessment report for the equipment.

[0137] Furthermore, the equipment health status assessment report acquisition module 4 is used to perform the following steps:

[0138] Historical temperature data collected by the distributed temperature sensor array is extracted based on the timestamps of the abnormal event sequence; the abnormal event sequence is traversed to analyze the event occurrence time period, and multiple time windows are set; the historical temperature data is mapped to the multiple time windows according to the timestamps for matching, and an event-temperature correlation is constructed; the event-temperature correlation database is constructed by integrating spatiotemporal data indexes according to the event-temperature correlation.

[0139] Furthermore, the equipment health status assessment report acquisition module 4 is used to perform the following steps:

[0140] Historical abnormal temperature change parameters are extracted from the historical temperature data. Feature analysis is performed based on these parameters to construct a temperature anomaly feature database. Temperature change search is conducted based on the event-temperature correlation database to determine the temperature change trajectory. The temperature change trajectory is mapped to the temperature anomaly feature database to obtain a fault development time series. Fault prediction is performed according to the fault development time series, and initial fault prediction parameters are obtained and their confidence is evaluated to generate parameter confidence values. The initial fault prediction parameters are then filtered and trained based on the parameter confidence values ​​to construct the fault prediction channel.

[0141] Furthermore, the equipment health status assessment report acquisition module 4 is used to perform the following steps:

[0142] Based on the fault prediction channel, temperature anomalies are identified in low-voltage equipment to determine temperature anomaly patterns. According to these temperature anomaly patterns, the fault prediction channel is used to predict the fault development of the low-voltage electrical equipment, generating a fault development prediction trajectory. Real-time temperature data is matched against these temperature anomaly patterns to determine real-time abnormal operating temperature parameters. Fault evolution analysis is performed using the fault development prediction trajectory to determine the fault evolution path. Based on the real-time abnormal operating temperature parameters and the fault evolution path, the operation of the low-voltage electrical equipment is evaluated to generate an equipment health index. Based on the equipment health index, the operating status of the low-voltage electrical equipment is predicted, and an equipment health status assessment report is constructed.

[0143] The temperature-sensing-based low-voltage electrical equipment safety monitoring system provided in this embodiment of the invention can execute the temperature-sensing-based low-voltage electrical equipment safety monitoring method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0144] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0145] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for safety monitoring of low-voltage electrical equipment based on temperature sensing, characterized in that, The method includes: A temperature monitoring dataset is obtained by collecting real-time temperature data from multiple nodes of low-voltage electrical equipment through a distributed temperature sensor array. Thermal state analysis is performed based on the temperature monitoring dataset to obtain temperature change state parameters for temperature safety monitoring of low-voltage electrical equipment, and abnormal temperature change modes are determined. The abnormal temperature change modes include multi-level early warning signals. According to the multi-level early warning signal trigger circuit protection device, the operating parameters of low-voltage electrical equipment are dynamically adjusted to implement the safety protection strategy, and a safety protection record log is obtained; Based on the safety protection log, a fault prediction channel is constructed to perform operational prediction and assessment of low-voltage electrical equipment, and an equipment health status assessment report is generated. The method for performing thermal state analysis based on the temperature monitoring dataset to obtain temperature change state parameters includes: The real-time temperature rise rate value is obtained by traversing the temperature change data of multiple monitoring points based on the temperature monitoring dataset. Temperature difference is calculated based on multiple monitoring points and the temperature monitoring data. Distribution analysis is performed based on the temperature difference data to determine the temperature distribution uniformity index. Using the temperature data stability coefficient as the end point of the cycle, the time period is calculated based on the temperature monitoring data to determine the thermal equilibrium time parameter; The temperature change state parameters are obtained by integrating the real-time temperature rise rate value, the temperature distribution uniformity index, and the thermal equilibrium time parameter. Among these methods, obtaining temperature change status parameters for temperature safety monitoring of low-voltage electrical equipment and identifying abnormal temperature change modes includes: By retrieving historical temperature records of low-voltage electrical equipment and combining them with thermal equilibrium time parameters, multi-level temperature rise thresholds are set, including a first temperature rise threshold and a second temperature rise threshold. When the real-time temperature rise rate value continuously exceeds the first temperature rise threshold but does not exceed the second temperature rise threshold, and the temperature distribution uniformity index is in the uniform range, a gradual temperature rise mode is determined. When the real-time temperature rise rate exceeds the second temperature rise threshold and the temperature distribution uniformity index is in a downward trend, a sudden temperature rise mode is determined. Based on the real-time temperature rise rate value, local temperature rise analysis is performed on multiple monitoring points to determine the local overheating concentration mode.

2. The method for safety monitoring of low-voltage electrical equipment based on temperature sensing as described in claim 1, characterized in that, The abnormal temperature change mode includes multi-level early warning signals, and the method includes: When the abnormal temperature change mode is the gradual temperature rise mode, multiple monitoring points are traversed to determine the abnormal point location, and the early warning level is classified according to the abnormal point location to generate the first early warning signal. When the abnormal temperature change mode is the sudden temperature rise mode, the temperature change impact is analyzed based on the real-time temperature rise rate value, the early warning level is classified according to the impact range, and a second early warning signal is generated. When the abnormal temperature change mode is the local overheating concentration mode, the target monitoring point is extracted and its location is transformed. Based on the coordinates of the overheating hot spot, an early warning analysis is performed to generate a third early warning signal.

3. The method for safety monitoring of low-voltage electrical equipment based on temperature sensing as described in claim 2, characterized in that, The method for dynamically adjusting the operating parameters of low-voltage electrical equipment according to the multi-level early warning signal trigger circuit protection device to execute the safety protection strategy and obtain a safety protection log includes: The range of safe operating parameters is set according to the multi-level early warning signals; In response to the first warning signal, the circuit protection device is triggered to activate a preventive protection strategy to dynamically adjust the operating parameters of the low-voltage electrical equipment according to the safe operating parameter range, and generate the first operating adjustment parameters; In response to the second warning signal, the circuit protection device is triggered to activate the active protection strategy and dynamically adjust the operating parameters of the low-voltage electrical equipment according to the safe operating parameter range, thereby generating the second operating adjustment parameters; In response to the third warning signal, the circuit protection device is triggered to activate the emergency protection strategy and dynamically adjust the operating parameters of the low-voltage electrical equipment according to the safe operating parameter range, thereby generating the third operating adjustment parameters; Based on the first operating adjustment parameter, the second operating adjustment parameter, and the third operating adjustment parameter, the parameter change trajectory is recorded to generate a device operating status dataset; The device operating status dataset is sorted according to time series to construct the security protection record log.

4. The method for safety monitoring of low-voltage electrical equipment based on temperature sensing as described in claim 1, characterized in that, Based on the aforementioned safety protection log, a fault prediction channel is constructed to perform operational prediction and assessment of low-voltage electrical equipment, generating an equipment health status assessment report. The method includes: Based on the security protection log, the sequence of abnormal events is determined by parsing the timestamps. The abnormal event sequence is traced back to the distributed temperature sensor array and historical temperature data for spatiotemporal mapping to establish an event-temperature correlation database. Fault analysis is performed based on the event-temperature correlation database to construct a fault prediction channel; The fault prediction channel is used to predict and assess the operating status of low-voltage electrical equipment, and a health status assessment report of the equipment is generated.

5. The method for safety monitoring of low-voltage electrical equipment based on temperature sensing as described in claim 4, characterized in that, The method involves tracing the abnormal event sequence back to the distributed temperature sensor array and performing a spatiotemporal mapping with historical temperature data to establish an event-temperature correlation database. Historical temperature data collected by the distributed temperature sensor array is extracted based on the timestamps of the abnormal event sequence. The sequence of abnormal events is traversed to analyze the time periods in which the events occur, and multiple time windows are set. The historical temperature data is mapped to the multiple time windows according to the timestamps to match them and construct an event-temperature correlation. The event-temperature correlation database is constructed by integrating spatiotemporal data indexes based on the event-temperature correlation relationship.

6. The method for safety monitoring of low-voltage electrical equipment based on temperature sensing as described in claim 4, characterized in that, Fault analysis is performed based on the aforementioned event-temperature correlation database, and a fault prediction channel is constructed. The method includes: Extract historical abnormal temperature change parameters from the historical temperature data, perform feature analysis based on the historical abnormal temperature change parameters, and construct a temperature anomaly feature library. Temperature change trajectory is determined by performing a temperature change search based on the event-temperature correlation database. The temperature change trajectory is mapped to a temperature anomaly feature database to obtain a fault development time series. Fault prediction is performed according to the fault development time series, initial fault prediction parameters are obtained, confidence is evaluated, and parameter confidence values ​​are generated. The initial fault prediction parameters are filtered and trained based on the confidence values ​​of the parameters to construct the fault prediction channel.

7. The method for safety monitoring of low-voltage electrical equipment based on temperature sensing as described in claim 4, characterized in that, The method for predicting and assessing the operational status of low-voltage electrical equipment through the fault prediction channel and generating an equipment health status assessment report includes: Based on the fault prediction channel, temperature anomalies are identified in low-voltage equipment to determine the temperature anomaly pattern. Based on the temperature anomaly pattern, the fault development of low-voltage electrical equipment is predicted through the fault prediction channel, and a fault development prediction trajectory is generated. Based on the temperature anomaly pattern, real-time temperature data is matched to determine the real-time abnormal operating temperature parameters. Fault evolution analysis is performed using the predicted fault development trajectory to determine the fault evolution path; The operation of low-voltage electrical equipment is evaluated based on the real-time abnormal operating temperature parameters and the fault evolution path to generate an equipment health index; Based on the equipment health index, the operating status of low-voltage electrical equipment is predicted, and an equipment health status assessment report is constructed.

8. A low-voltage electrical equipment safety monitoring system based on temperature sensing, characterized in that, For implementing the low-voltage electrical equipment safety monitoring method based on temperature sensing according to any one of claims 1-7, the system comprises: The temperature monitoring dataset acquisition module is used to collect real-time temperature data from multiple nodes of low-voltage electrical equipment through a distributed temperature sensor array to obtain a temperature monitoring dataset. An abnormal temperature change mode acquisition module is used to perform thermal state analysis based on the temperature monitoring dataset, obtain temperature change state parameters for temperature safety monitoring of low-voltage electrical equipment, and determine abnormal temperature change modes, wherein the abnormal temperature change mode includes multi-level early warning signals. The safety protection log acquisition module is used to dynamically adjust the operating parameters of low-voltage electrical equipment according to the multi-level early warning signal triggering circuit protection device to execute safety protection strategy, and to obtain the safety protection log. The equipment health status assessment report acquisition module is used to construct a fault prediction channel based on the safety protection record log to perform operational prediction assessment of low-voltage electrical equipment and generate an equipment health status assessment report.

Images