Insulated bus duct and hotspot temperature anomaly early warning method and system thereof

Abstract

The application discloses a kind of hot spot temperature anomaly early warning methods of insulated bus duct, belong to the online monitoring technical field of electric power equipment.For the problem that existing bus duct temperature monitoring only relies on fixed absolute threshold, early warning is seriously lagging behind, the real-time temperature data of bus duct connector is obtained, the change rate of temperature with time is calculated, including short-term change rate and long-term change rate;When long-term change rate is positive and exceeds the preset aging rate threshold, a first early warning signal is generated;At the same time, according to real-time temperature data and long-term change rate, the predicted threshold below the absolute temperature threshold is dynamically calculated, and a second early warning signal is generated when the real-time temperature reaches the predicted threshold.The application combines trend analysis with dynamic threshold, realizes early identification of early signs of insulation aging and prior prediction of thermal runaway risk, effectively solves the problem of early warning lag in the prior art, and significantly improves the operation reliability and operation efficiency of bus duct power supply system.

Description

Technical Field

[0001] This invention relates to the field of online monitoring technology for power equipment, and in particular to an insulated busbar trunking and its hot spot temperature anomaly early warning method and system. Background Technology

[0002] Existing intelligent busbar temperature monitoring technologies, such as the patent filed by Baoshi Electric in 2016, achieve real-time monitoring of hotspot temperatures by embedding temperature sensors within the connectors. However, these existing technologies generally employ alarm logic based on absolute temperature thresholds, meaning the system only issues an alarm when the monitored temperature exceeds a preset, fixed safety value. This "threshold-triggered" mechanism exhibits significant lag, essentially constituting a "post-event alarm."

[0003] However, in real-world engineering scenarios, the temperature rise at busbar connections due to factors such as insulation aging and increased contact resistance is a slow and continuous process. When the temperature passively climbs to a preset absolute threshold, the insulation material has often already suffered irreversible thermal damage, even approaching the edge of thermal runaway, leaving maintenance personnel with an extremely limited window of opportunity for safe intervention. Therefore, existing technologies cannot provide effective early warnings by predicting temperature trends in the early stages of potential fault development, failing to meet the high reliability requirements of modern power supply systems for "pre-emptive awareness" and "preventive maintenance" of equipment status. Summary of the Invention

[0004] Therefore, it is necessary to provide a method and system for early warning of abnormal temperatures in insulated busbar trunking and its hot spots, addressing the technical problem of how to effectively identify and warn of insulation performance failure in its early stages.

[0005] A method for early warning of abnormal hot spot temperature in an insulated busbar trunking includes:

[0006] Obtain real-time temperature data of the busbar connectors;

[0007] Based on the real-time temperature data, calculate the rate of temperature change over time, which includes both short-term and long-term rates of change.

[0008] When the long-term rate of change is positive and exceeds a preset aging rate threshold, a first warning signal is generated;

[0009] Based on the real-time temperature data and the long-term rate of change, a predicted threshold below the absolute temperature threshold is dynamically calculated. When the real-time temperature data reaches the predicted threshold, a second warning signal is generated.

[0010] In one embodiment, calculating the rate of change of temperature over time includes:

[0011] Using a sliding time window as the unit, a linear fit is performed on the real-time temperature data within the sliding time window, and the slope obtained from the fit is used as the long-term rate of change.

[0012] In one embodiment, the step of dynamically calculating a predicted threshold below an absolute temperature threshold includes:

[0013] Obtain the thermal aging characteristic curves of the busbar insulation material;

[0014] Based on the thermal aging characteristic curve, the current temperature value, and the long-term rate of change, predict the time required for the temperature to reach the absolute temperature threshold.

[0015] The prediction threshold is determined based on the required time and the thermal aging characteristic curve.

[0016] In one embodiment, the step further includes:

[0017] The acquired real-time temperature data is filtered to eliminate instantaneous noise.

[0018] In one embodiment, the first warning signal is used to indicate that the aging process of the insulation material is accelerating, and the second warning signal is used to indicate that immediate maintenance intervention is required.

[0019] A hot spot temperature anomaly early warning system for insulated busbar trunking includes:

[0020] The temperature acquisition module is used to acquire real-time temperature data of the busbar trunking connectors;

[0021] The data processing module is used to calculate the rate of change of temperature over time based on the real-time temperature data, wherein the rate of change includes short-term rate of change and long-term rate of change.

[0022] The early warning analysis module is used to generate a first early warning signal when the long-term change rate is positive and exceeds a preset aging rate threshold; and to dynamically calculate a predicted threshold below the absolute temperature threshold based on the real-time temperature data and the long-term change rate, and to generate a second early warning signal when the real-time temperature data reaches the predicted threshold.

[0023] The warning output module is used to output the first warning signal and / or the second warning signal.

[0024] In one embodiment, the data processing module is specifically used to perform linear fitting on the real-time temperature data within the sliding time window in units of sliding time window, and use the slope obtained from the fitting as the long-term rate of change.

[0025] In one embodiment, the early warning analysis module is specifically used to obtain the thermal aging characteristic curve of the busbar insulation material, predict the time required for the temperature to reach the absolute temperature threshold based on the thermal aging characteristic curve, the current temperature value and the long-term rate of change, and determine the predicted threshold based on the required time and the thermal aging characteristic curve.

[0026] An insulated busbar trunking system includes:

[0027] Busbar trunking body;

[0028] A temperature sensor embedded in the connector; and,

[0029] Hot spot temperature anomaly early warning system for insulated busbar trunking as described in any of the above embodiments.

[0030] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for early warning of hot spot temperature anomalies in insulated busbar trunking as described in any of the above embodiments.

[0031] The aforementioned insulated busbar trunking and its hotspot temperature anomaly early warning method and system construct a dual early warning model of "trend + threshold" by acquiring real-time temperature data and calculating the rate of temperature change over time. First, by monitoring the long-term rate of change, early signs of accelerated aging of the insulation material can be identified even before the temperature reaches the danger threshold. When the long-term rate of change is positive and exceeds a preset aging rate threshold, a first early warning signal is generated. This mechanism achieves advanced "attention" level early warning, providing maintenance personnel with valuable time for planned inspections and preventing the fault from deteriorating to an uncontrollable state. Second, by dynamically calculating and predicting the threshold based on real-time temperature data and the long-term rate of change, the problem of fixed thresholds being unable to adapt to changes in operating conditions is solved. When the real-time temperature reaches this dynamically predicted threshold, a second early warning signal is generated. This dynamic threshold comprehensively considers the current temperature level and the rate of increase, ensuring that an emergency alarm is issued in advance within the predicted "safe handling time window," avoiding false alarms caused by normal load fluctuations and guaranteeing timely intervention when the fault accelerates. In summary, by combining trend analysis with dynamic thresholds, this invention achieves a technological leap from "passive response" to "active prediction," significantly improving the timeliness and accuracy of early warnings and effectively ensuring the operational reliability of the busbar power supply system. Attached Figure Description

[0032] Figure 1 This is a flowchart illustrating a method for early warning of abnormal hot spot temperature in an insulated busbar trunking in one embodiment.

[0033] Figure 2 This is an overall flowchart of a hotspot temperature anomaly early warning method for an insulated busbar trunking in one embodiment;

[0034] Figure 3 This is a timing diagram of a hotspot temperature anomaly early warning method for an insulated busbar trunking in one embodiment;

[0035] Figure 4 This is a linear fitting graph of a hotspot temperature anomaly early warning method for an insulated busbar trunking in one embodiment;

[0036] Figure 5 This is a schematic diagram of the thermal aging characteristic curve of the insulating material in an early warning method for hot spot temperature anomalies in an insulated busbar trunking according to one embodiment.

[0037] Figure 6 This is a schematic diagram of the early warning state transition of an early warning method for hot spot temperature anomalies in an insulated busbar trunking in one embodiment;

[0038] Figure 7 This is a functional block diagram of a hotspot temperature anomaly early warning system for an insulated busbar trunking in one embodiment;

[0039] Figure 8 This is a functional block diagram of a computer device in one embodiment. Detailed Implementation

[0040] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below. In the description of the present invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention.

[0041] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0042] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0043] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0044] It should be noted that when an element is referred to as being "fixed to" or "set on" another element, it can be directly on the other element or there may be an intervening element. When an element is considered to be "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "upper," "lower," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0045] Please see Figure 1 and Figure 2 This invention provides a method for early warning of hot spot temperature anomalies in insulated busbar trunking, aiming to solve the problem of warning lag caused by relying solely on absolute temperature thresholds for alarms in existing technologies. Its core concept lies in constructing a dual-layer protection mechanism combining "trend warning" and "dynamic threshold warning" through deep coupling of dynamic tracking of temperature change trends and material aging characteristics. The method 10 for early warning of hot spot temperature anomalies in insulated busbar trunking includes the following steps:

[0046] Step S101: Obtain real-time temperature data of the busbar connector;

[0047] Step S102: Calculate the rate of change of temperature over time based on the real-time temperature data, whereby the rate of change includes both short-term and long-term rates of change.

[0048] Step S103: When the long-term rate of change is positive and exceeds the preset aging rate threshold, a first warning signal is generated;

[0049] Step S104: Based on the real-time temperature data and the long-term rate of change, dynamically calculate a predicted threshold that is lower than the absolute temperature threshold. When the real-time temperature data reaches the predicted threshold, generate a second warning signal.

[0050] See Figure 2 The figure shows the overall flowchart of the early warning method of the present invention. The method begins with step S101: data acquisition and preprocessing. Specifically, a high-precision temperature sensor, such as an NTC thermistor or a digital temperature sensor chip, is pre-embedded inside the key connector of the busbar trunking. This sensor collects real-time temperature data of the surface of the metal conductor or insulation layer of the connector at a fixed sampling frequency (e.g., once per second or once per minute) and transmits this data to the data processing unit located on a local monitoring terminal or a cloud server. To ensure the accuracy of subsequent analysis, the data processing unit first preprocesses the raw temperature data, including but not limited to removing outliers that are significantly outside the reasonable range, and smoothing the data using methods such as moving average or Kalman filtering to eliminate measurement noise caused by electromagnetic interference or instantaneous load fluctuations.

[0051] The method then proceeds to the core analysis phase, step S102: calculating the rate of temperature change over time. This step does not simply calculate the difference between the current temperature and the previous temperature, but rather calculates two indices with different physical meanings in parallel: the short-term rate of change and the long-term rate of change. The short-term rate of change, k... t This reflects temperature fluctuations over a very short period, such as between two adjacent sampling points, and is used to capture sudden thermal shocks. The long-term rate of change, K, is one of the key features of this invention. It calculates the overall trend of temperature change by analyzing a longer time window, such as a temperature data sequence over the past 30 minutes to 2 hours. This long-term rate of change, K, effectively filters out short-term spikes caused by normal load fluctuations, and truly reflects the slow, continuous evolution of the contact resistance or insulation performance of the connector.

[0052] In step S103, based on the rate of change calculated in step S102, the first level of early warning logic is executed: aging rate early warning. The system has a pre-set "aging rate threshold" derived from the allowable aging rate of insulating materials such as polyester film and epoxy resin. When the long-term rate of change K calculated in step S102 is positive (i.e., the temperature is rising) and its value exceeds the pre-set aging rate threshold, the system determines that the insulating material is aging at a rate exceeding the normal range. At this time, although the current temperature may be far from the danger level, the system will immediately generate and output a first early warning signal. This signal is usually a "caution" or "attention" level alert, designed to remind maintenance personnel that the busbar connection may have early hidden dangers such as poor contact, slight overload, or poor heat dissipation, requiring planned inspection.

[0053] Next, in step S104, the second level of early warning logic is executed: dynamic threshold early warning. This invention abandons the traditional fixed absolute temperature threshold. Instead, the system dynamically calculates a "predicted threshold" based on the current real-time temperature value obtained in step S101 and the long-term rate of change K calculated in step S102. The calculation logic for this predicted threshold is as follows: based on the thermal aging characteristic curve of the busbar insulation material, which typically conforms to the Arrhenius equation (that is, for every certain temperature increase, the insulation life decreases exponentially), and combined with the current temperature rise rate K, it is predicted that without intervention, the temperature will reach a recognized dangerous absolute temperature value (e.g., 130°C) within a preset "safe handling time window" (e.g., 30 minutes, 1 hour, or 2 hours). This "predicted threshold" is dynamic and always lower than that fixed dangerous absolute temperature value. When the real-time monitored temperature data reaches this dynamically calculated predicted threshold, the system determines that the fault has entered an accelerated deterioration stage and immediate intervention is necessary. At this time, the system will generate and output a second early warning signal, which is at the "alarm" level, prompting maintenance personnel to immediately arrange emergency repairs.

[0054] Finally, the system outputs the aforementioned first and / or second warning signals to maintenance personnel through various means such as human-machine interface, SMS push, and audible and visual alarms.

[0055] Thus, by acquiring real-time temperature data and calculating the rate of temperature change over time, a dual early warning model of "trend + threshold" was constructed. First, by monitoring the long-term rate of change, early signs of accelerated aging of insulation materials can be identified even before the temperature reaches a dangerous threshold. When the long-term rate of change is positive and exceeds a preset aging rate threshold, a first early warning signal is generated. This mechanism achieves advanced "attention" level early warning, providing maintenance personnel with valuable time for planned inspections and preventing the fault from deteriorating to an uncontrollable state. Second, by dynamically calculating and predicting the threshold based on real-time temperature data and the long-term rate of change, the problem of fixed thresholds being unable to adapt to changes in operating conditions is solved. When the real-time temperature reaches this dynamically predicted threshold, a second early warning signal is generated. This dynamic threshold comprehensively considers the current temperature level and the rate of increase, ensuring that an emergency alarm is issued in advance within the predicted "safe handling time window," avoiding false alarms caused by normal load fluctuations and guaranteeing timely intervention when the fault accelerates.

[0056] See Figure 3 This figure, presented as a time-series diagram, more intuitively demonstrates the superiority of the method of this invention over the traditional threshold method. The horizontal axis represents time, and the vertical axis represents temperature. The curve T_real represents the actual monitored temperature curve, which goes through two stages: a slow rise (stage A) and an accelerated rise (stage B). The dashed line T_threshold_fixed represents the traditional fixed absolute temperature threshold. It can be seen that when the temperature rises slowly in stage A, the traditional method does not trigger any alarms. An alarm is only triggered when the temperature reaches T_threshold_fixed in the later stages of stage B, by which time the insulation material is already close to damage. In contrast, the method of this invention, in the early stages of stage A, triggers the first warning signal (level one warning) by monitoring the long-term rate of change K to be positive and exceeding the threshold, achieving proactive "attention." Subsequently, in the early stages of stage B, the threshold T_threshold_dynamic is dynamically calculated and predicted. When the temperature reaches this dynamic threshold, the second warning signal (level two warning) is triggered, allowing for emergency intervention Δt earlier than the traditional method, thus gaining a valuable time window for fault handling.

[0057] In summary, by combining "trend analysis" with "material properties", this invention achieves a leap from "passive response" to "active prediction", fundamentally solving the problem of delayed early warning in existing technologies.

[0058] Within the framework of early warning based on the rate of temperature change, accurately and robustly calculating the "long-term rate of change" is one of the key factors determining the reliability of the early warning. This invention provides a preferred implementation method, which combines a "sliding time window" with "linear fitting" to obtain the long-term rate of change, thereby filtering out short-term interference and truly reflecting the long-term evolution trend of temperature. The following will be discussed in conjunction with the appendix... Figure 4 The linear fitting diagram illustrates the preferred embodiments of the present invention.

[0059] Specifically, during the continuous input of real-time temperature data streams, the system maintains a first-in, first-out (FIFO) data queue, the length of which is the size of the sliding time window. The window length must balance sensitivity to trends and resistance to interference. For example, if the window is too short (e.g., 5 minutes), the system may misinterpret short-term load fluctuations as long-term trends, leading to frequent false alarms; if the window is too long (e.g., 4 hours), the response to the true trend will be too slow, failing to provide early warning. Through engineering practice, this invention preferably sets the sliding time window length between 30 minutes and 2 hours. For example, a 1-hour window can be used for scenarios with relatively stable loads, such as data centers; a 30-minute window can be used for scenarios with large load fluctuations, such as factories.

[0060] When each new temperature data point is collected and stored in the queue, the system performs mathematical processing on all data points (time t, temperature T) within that window. Specifically, it uses the least squares method to perform univariate linear regression fitting and calculates the slope K of the fitted line. This slope K is the overall rate of temperature change within that time window.

[0061] Figure 4 The mathematical principles behind this process are illustrated. The scatter plot represents temperature data points collected within a sliding window, and the straight line L is the best-fit line obtained using the least squares method. The equation of the fitted line can be expressed as T = K˙t + b, where K is the slope, representing the average rate of temperature change within the window, expressed in °C / minute or °C / hour. Essentially, the fitting process involves finding a straight line that minimizes the sum of the squares of the perpendicular distances from all data points to that line.

[0062] The long-term rate of change K calculated in this way has a clear physical meaning: K>0 indicates that the temperature generally shows an upward trend within the window, K<0 indicates a downward trend, and the larger |K| is, the more drastic the change. Since the fitting process uses all data points within the window, rather than relying only on the first and last points, the impact of abnormal fluctuations of a single or a few data points (such as instantaneous sensor noise) on K is minimized, thus ensuring the stability and reliability of the K value.

[0063] Of course, linear fitting is not the only implementation of this invention. In other feasible embodiments of this invention, to capture the nonlinear characteristics of temperature changes or further improve prediction accuracy, more complex algorithms can be used to calculate long-term rates of change or predict future temperatures. For example:

[0064] Exponential smoothing: The Holt-Winters exponential smoothing model can be applied. This model can capture not only the trend term of temperature (i.e., the long-term rate of change) but also seasonal fluctuations (e.g., periodic changes in load at different times of the day). In this way, the system can establish a baseline of "normal temperature change pattern," and when the actual rate of temperature change deviates significantly from this baseline, an early warning can be triggered, thereby achieving more intelligent anomaly detection.

[0065] Machine learning prediction: A Long Short-Term Memory (LSTM) network model can be trained. This model is a special type of recurrent neural network, particularly adept at processing time-series data. By inputting historical data such as temperature, load current, and ambient temperature over a past period (e.g., 24 hours), the LSTM model can learn the complex patterns of temperature changes and accurately predict temperature trends over a future period. In this case, the "long-term rate of change" is no longer a simple linear slope, but rather a trend characteristic implied by a series of predicted values ​​output by the model. An early warning is triggered when the predicted future temperature value or its rate of change exceeds a safe range. This method can handle more complex operating conditions and has higher prediction accuracy, but requires a large amount of historical data for model training.

[0066] The differential method combined with filtering involves first performing first-order difference calculations on the temperature data to obtain a series of instantaneous rates of change. Then, a low-pass filter (such as a Butterworth filter) is applied to these instantaneous rates of change to remove high-frequency fluctuations and retain low-frequency trend components, thus obtaining a smoothed long-term rate of change. This method is very mature in the field of signal processing, requires minimal computation, and is easy to implement on embedded devices.

[0067] All the above alternative solutions share the same core objective as this invention: to obtain stable indicators that accurately reflect the long-term temperature evolution trend, thereby providing reliable input for subsequent early warning models. Regardless of the specific algorithm used, the technical effect is the same: through in-depth processing of the raw data, the early warning system's ability to identify real fault trends and its resistance to interference are improved.

[0068] After successfully obtaining the long-term temperature change rate K, how to use this information to construct a "predictive" early warning threshold is another core aspect that distinguishes this invention from existing technologies. This invention provides a preferred implementation method, namely, dynamically calculating the predictive threshold based on the inherent thermal aging characteristics of the busbar insulation material. The following will be combined with the appendix... Figure 5 The thermal aging characteristic curve illustrates another preferred embodiment of the present invention.

[0069] First, the system pre-stores the thermal aging characteristic curves of the insulating materials used (e.g., polyester film, cross-linked polyethylene, or epoxy resin commonly used in busbar trunking). These curves are typically based on the Arrhenius equation or engineering data provided by the material supplier. Figure 5 This characteristic curve is illustrated schematically. As can be seen from the graph, the lifespan of the insulating material (vertical axis, logarithmic scale) is inversely correlated with the operating temperature (horizontal axis). The higher the temperature, the faster the insulation performance of the material deteriorates, and the shorter the lifespan. Specifically, the curve can be expressed as: L = A˙e B / T Where L is the lifetime, T is the absolute temperature, and A and B are material-related constants. This curve is essentially a "stress-life" model for insulating materials, revealing the exponential relationship between temperature and aging rate.

[0070] Based on this, the calculation process for the dynamic threshold is as follows:

[0071] 1. Obtain key parameters: The system acquires the current temperature value T in real time. now And the long-term rate of change K calculated by the aforementioned method.

[0072] 2. Predicting the time to reach the dangerous temperature: The system has a pre-defined, recognized absolute temperature value T that will cause the insulation material to fail rapidly. danger (For example, this value might be set to 130°C for common insulating materials.) Based on the current temperature T now Given the current rate of change K, the temperature can be estimated from T by simple linear extrapolation. now Rise to T danger Required time Δt: Δt = (T danger -T now ) / K.

[0073] 3. Determine the safety handling time window: The system also has a preset "safety handling time window" T. window For example, 30 minutes. This time window is set to take into account the reasonable time required from system alarm to the arrival of maintenance personnel on site and completion of the handling.

[0074] 4. Dynamically calculate the prediction threshold: Now, the key step lies in adjustment. The system does not automatically predict the threshold when the temperature reaches T. danger Instead of only calling the police when the alarm is triggered, it is necessary to ensure that there is a T after the alarm is triggered. window The system needs to process the data within a given timeframe. Therefore, the system needs to solve for a temperature value T. predict So that the change rate K from T predict Rise to T danger The required time is exactly equal to T window That is: T predict = T danger -K×T window This T predictIt refers to a dynamic prediction threshold.

[0075] 5. Trigger an early warning: When the real-time temperature T now Reaching or exceeding this dynamically calculated T predict When this happens, the system will trigger a level-two warning signal.

[0076] The advantage of this approach lies in its consideration of the dynamic nature of actual operating conditions. For example, when K is large (temperature rises rapidly), the calculated T... predict If the temperature is low, the system will issue an emergency warning at a lower temperature to ensure sufficient response time. Conversely, if K is small (temperature rises slowly), then T... predict The threshold will be relatively high, preventing the system from issuing unnecessary emergency alerts prematurely and avoiding a waste of operational resources. This contrasts sharply with the traditional fixed threshold method.

[0077] In addition to the method based on simple linear extrapolation described above, other embodiments of the present invention can also be extended to calculate other more accurate dynamic thresholds as follows:

[0078] 1. Based on a material lifetime accumulation model: The concept of "equivalent aging time" or "lifetime consumption" can be introduced. The system no longer only focuses on whether the temperature has reached a certain point, but continuously calculates how much lifetime the insulation material has consumed. According to the thermal aging characteristic curve, the higher the temperature, the more lifetime is consumed per unit time. The system can calculate the accumulated lifetime consumption in real time. When it is predicted that the remaining lifetime will be in the future T... window An alert is triggered when the time limit expires. This method can more accurately reflect the cumulative damage caused by thermal aging of materials, and is especially suitable for scenarios where the temperature fluctuates at high temperatures for extended periods.

[0079] 2. Combining Ambient Temperature and Load Current: To more accurately predict temperature trends, data from ambient temperature sensors and load current transformers can be incorporated into the model. By analyzing historical data, a multivariate regression model or neural network model can be established to correlate ambient temperature, load current, and connector temperature. When the load current surges or the ambient temperature rises, the model can more accurately predict future temperature changes at the connector, thereby dynamically adjusting the prediction threshold and reducing false alarms and missed alarms.

[0080] 3. Adaptive Threshold Adjustment: This feature combines machine learning methods to allow the system to automatically learn normal temperature variation patterns under different operating conditions. The system will adjust the currently predicted temperature threshold. predict Compare with the normal value under the same historical period and load conditions. If the currently calculated T... predict If the temperature is significantly lower than the historical normal level (meaning an abnormally high temperature), the system will issue an early warning even if the absolute value has not yet reached the threshold calculated based on material properties, thereby identifying some unconventional and sudden failure modes.

[0081] These extended solutions of preferred embodiments are all further optimizations and enrichments of the specific implementation methods under the guidance of the core idea of ​​this invention—dynamic early warning based on material aging characteristics and real-time trends—aiming to make the early warning system more intelligent, accurate, and adaptable to different scenarios.

[0082] It is understandable that in early warning methods based on dynamic trend analysis, the quality of the raw data directly determines the accuracy of all subsequent analyses. Any abnormal data points originating from electromagnetic interference, sensor noise, or communication interruptions, if used directly to calculate the long-term rate of change without processing, may lead to distorted calculation results, resulting in false alarms or missed alarms. Therefore, this invention introduces a crucial data preprocessing step after acquiring real-time temperature data: filtering the raw data to eliminate transient noise.

[0083] This data preprocessing step is preferably performed in real time within a data processing unit (e.g., an intelligent monitoring module embedded in the busbar or an edge computing gateway). Various implementation methods can be chosen to suit different hardware platforms and performance requirements.

[0084] A preferred and efficient implementation is to use a moving average filtering method. This method maintains a data buffer of a fixed length (e.g., 5 or 10 sampling points). Whenever a new raw temperature data T is generated... raw (n) When data is collected, the system calculates the arithmetic mean of the new data and the first N-1 data in the buffer, which is taken as the effective temperature value T at the current moment. filtered (n):

[0085]

[0086] This method is simple to implement and computationally inexpensive, making it ideal for running on resource-constrained embedded microcontrollers. Its filtering effect increases with the window length N, but correspondingly, the response to temperature changes becomes slightly delayed. In practice, N can be set to 3-5 to strike a balance between noise suppression and response speed.

[0087] Another, more preferred implementation is to use a first-order low-pass digital filter, also known as the exponentially weighted moving average method. Its calculation formula is:

[0088]

[0089] Where α is the filter coefficient, and its value ranges from 0 to 1. The output T of this filter filtered (n) represents the current input Traw(n) and the previous output T. filteredThe filter is a weighted average of (n-1). A larger α value makes the filter more sensitive to rapid changes but weaker at filtering noise; a smaller α value results in stronger filtering but a slower response. By appropriately setting the value of α (e.g., 0.1 or 0.2), spike noise caused by electromagnetic interference can be effectively smoothed out while preserving the true trend of temperature changes. This method is widely used in industrial process control because it eliminates the need to maintain a long data buffer, thus saving significant memory.

[0090] In addition to the two linear filtering methods mentioned above, for more complex noise types that may occur, the preferred embodiment of the technical solution of the present invention can also be extended to employ other more advanced signal processing techniques:

[0091] 1. Kalman Filtering: If the system can obtain other information related to temperature changes, such as load current or ambient temperature, a state-space model can be constructed, and Kalman filtering can be applied. Kalman filtering is an optimal estimator that can recursively estimate the system state (i.e., the true temperature) using the system's dynamic model and multiple noisy observations. Its filtering effect is far superior to simple moving averages, and it can simultaneously filter out noise and predict temperature changes, but its implementation is relatively complex. For applications like busbar trunking, which have extremely high requirements for long-term reliability, deploying Kalman filtering on the monitoring server or cloud platform side is feasible.

[0092] 2. Outlier Removal: Besides noise, the raw data may also contain "outliers" caused by sensor malfunctions or communication interruptions, such as a temperature reading far outside the normal range. The preprocessing stage needs to be able to identify and remove these outliers. One approach is to set a "reasonableness threshold" based on common sense physics. For example, any reading that changes by more than 50°C within 1 second, or any reading below -50°C or above 200°C, will be directly deemed invalid and replaced with the previous valid value or through interpolation.

[0093] 3. Multi-sensor data fusion: In a busbar system, there may be multiple connectors, each with multiple temperature measurement points. The preprocessing stage can fuse data from multiple sensors at the same or adjacent locations. For example, the average or median of temperature data from different phases at the same connector can be calculated to reduce the impact of single sensor failure or localized heat sources.

[0094] Regardless of the specific preprocessing method used, the ultimate goal remains the same: to provide the core trend analysis module with a clean, reliable temperature time series that accurately reflects the thermal state of the busbar connections. This seemingly basic step is the first line of defense ensuring the stable and accurate operation of the entire early warning system. Its technical effect is directly reflected in a significant improvement in the reliability of the subsequently calculated long-term rate of change K and dynamic prediction threshold, thereby fundamentally reducing the system's false alarm and false negative rates.

[0095] It is important to note that the "trend + threshold" dual early warning model constructed in this invention outputs not only a single alarm signal, but also multi-level early warning information with clear direction and varying degrees of urgency. This hierarchical early warning mechanism can transform complex analysis results into intuitive operation and maintenance instructions, significantly improving the usability of early warning information and operational efficiency.

[0096] like Figure 6 As shown in the warning state transition diagram, specifically, when the warning model makes different judgments, the system generates and outputs warning signals of corresponding levels. This invention preferably sets two main warning levels, but this scheme can be extended to more levels.

[0097] The first warning signal, also known as the "attention" level signal, is generated when the monitored long-term rate of change K is positive (temperature rise) and the value of K exceeds the preset aging rate threshold. This signal is usually issued before the temperature reaches a dangerous level, but it reveals a crucial early characteristic—the aging process of the insulation material is accelerating. This signal is intended to alert maintenance personnel to a developing potential problem with the equipment. For example, when the system issues this signal, it can display the busbar connector icon in yellow on the monitoring backend interface, along with the message: "Attention: Connector A temperature continues to rise, aging rate exceeds normal range. It is recommended to schedule an inspection soon, focusing on checking the torque of the connecting bolts and heat dissipation." This signal directs troubleshooting towards early causes such as "poor contact," "minor overload," or "deteriorating ventilation and heat dissipation," preventing maintenance personnel from conducting aimless troubleshooting after receiving the alarm.

[0098] The second warning signal, or "alarm" level signal, is generated when the real-time monitored temperature reaches the dynamically calculated predicted threshold. This signal means that without intervention, the temperature will reach or exceed the dangerous value within the preset safe handling time window. This is a more urgent signal, indicating that the fault has entered a phase of accelerated deterioration and immediate action is required. When the system issues this signal, the icon on the interface will turn red, and an audible and visual alarm will sound. Simultaneously, an emergency notification will be pushed via SMS or the app: "Emergency Alarm: Connector B temperature is abnormal, has reached the predicted threshold, and is expected to reach a dangerous temperature in 30 minutes. Please arrange maintenance immediately!" This signal not only indicates the level of urgency but also provides a quantified time window, enabling maintenance personnel to rationally prioritize tasks and prepare necessary tools, thereby effectively preventing insulation breakdown or fire accidents caused by delayed handling.

[0099] To further improve the accuracy of operation and maintenance decisions, this invention can also correlate early warning signals with preliminary fault cause analysis. For example, the system can combine different characteristics of short-term and long-term change rates to perform preliminary fault type classification and output more detailed early warning information:

[0100] Mode 1: Load surge type failure. If a short-term rate of change k is detected... t A sharp increase in the long-term rate of change (K) that had previously remained stable within the normal range may indicate that the busbar has experienced a sudden overload event such as a short circuit or the startup of high-power equipment. In this case, even if the long-term rate of change has not yet exceeded the threshold, the system can still issue a "load shock" warning, suggesting that the downstream load be checked.

[0101] Mode 2: Slowly increasing contact resistance indicates a major fault. If the long-term rate of change K is consistently positive, while the short-term rate of change k... t If the temperature is relatively stable and the ratio of temperature to load current shows a gradually increasing trend, it can be basically determined that the contact resistance at the connector is slowly increasing due to oxidation, loosening, or other reasons. The first warning signal issued at this point clearly indicates "check the connector," providing maintenance personnel with a precise direction for troubleshooting.

[0102] Mode 3: Heat dissipation failure. If the long-term rate of change K remains positive, and the ambient temperature sensor data is normal, but the temperature difference between the busbar trunking casing and the connector is decreasing, this may indicate that the ventilation and heat dissipation channels inside the busbar trunking are blocked or the cooling fan is malfunctioning. In this case, the warning signal can prompt "Check the heat dissipation system".

[0103] Therefore, this invention elevates "data monitoring" to "intelligent diagnosis" by transforming cold, hard temperature data into "attention" and "alarm" signals with clear business implications, and further associating these signals with possible causes of failure. This tiered early warning and cause-related association directly reduces the data analysis workload of maintenance personnel, enabling them to respond quickly and accurately based on the level and implications of the warning signals. This allows them to focus their valuable time and energy on the most critical fault points, significantly improving overall maintenance efficiency.

[0104] It is worth mentioning that, to achieve the above method, this invention also provides a hotspot temperature anomaly early warning system for insulated busbar trunking. This system adopts a modular design, with each module interacting with the others through internal interfaces to jointly complete the entire process from data acquisition to early warning output.

[0105] See Figure 7 The system includes four core functional modules: temperature acquisition module 110, data processing module 120, early warning analysis module 130, and early warning output module 140.

[0106] The temperature acquisition module 110 can be physically comprised of multiple high-precision digital temperature sensors, such as the DS18B20 or more advanced MEMS temperature sensors, embedded within the busbar connector. This module is responsible for sampling the temperature at a preset frequency (e.g., 1Hz) and converting the raw analog or digital signal into a standard temperature value. To ensure reliable data transmission, this module can upload real-time data to the monitoring system via RS-485 bus, CAN bus, or wireless communication networks (such as LoRa or ZigBee).

[0107] The data processing module 120 is the system's "signal conditioning" unit. It receives the raw temperature data stream from the temperature acquisition module 110. This module integrates the aforementioned data preprocessing algorithms, such as moving average filtering or first-order low-pass filtering, to eliminate electromagnetic interference and sensor noise during the acquisition process. The filtered "clean" temperature data is organized into a time series and stored in memory. More importantly, this module is the core for calculating the rate of temperature change. It maintains a sliding time window and performs mathematical operations on the temperature series within the window, such as least-squares linear fitting, to calculate the short-term and long-term rates of change K in parallel. The output of the data processing module 120 is a set of structured data, including: the current filtered effective temperature value, the short-term rate of change, and the long-term rate of change K.

[0108] The early warning analysis module 130 is the system's "decision center." This module receives structured data from the data processing module 120 and executes the core logic of this invention. First, it has a built-in preset "aging rate threshold" and a "thermal aging characteristic curve" model for the busbar insulation material.

[0109] The early warning analysis module 130 performs the following judgment:

[0110] 1. Trend Early Warning: Compare the long-term rate of change K with the aging rate threshold. If K > threshold, then the first-level early warning condition is met.

[0111] 2. Dynamic threshold warning: Obtain the current temperature value T now and the long-term rate of change K. Based on the preset danger temperature value T. danger and safety handling time window T window Using formula T predict =T danger -K×T window Dynamically calculate the prediction threshold. Compare T. now With T predict If T now ≥T predict If so, it is determined that the conditions for a Level II warning have been met.

[0112] Based on the above logic, the early warning analysis module 130 will generate a corresponding early warning instruction, carrying the early warning level and a preliminary fault cause label (such as "aging rate too fast", "temperature is about to exceed the limit", etc.).

[0113] The early warning output module 140 serves as the "interface" for interaction between the system and maintenance personnel. It receives early warning commands from the early warning analysis module 130 and converts them into specific, perceptible signals. For example, this module can drive the audible and visual alarms in the monitoring room, turning the corresponding device icons on the monitoring screen yellow (Level 1 warning) or red (Level 2 warning), and sending detailed early warning information (including device location, warning level, current temperature, rate of change, and recommended measures) to the mobile phones of designated maintenance personnel via SMS gateway or mobile application push interface.

[0114] The advantages of this modular design lie in its high flexibility and scalability. First, the functions of each module are cohesive, with clearly defined responsibilities, facilitating independent development, testing, and maintenance. Second, the standardized interfaces between modules allow for flexible system deployment. For example, for a single busbar, all modules can be integrated into a single intelligent monitoring terminal. For large projects, the temperature acquisition module 110 can be distributed on-site, while the data processing, early warning analysis, and output modules can be deployed on a cloud server for centralized monitoring and unified management. This flexible deployment capability from local to cloud is a key technical advantage of the system architecture of this invention, enabling it to adapt to various application scenarios, from single devices to complex power distribution networks.

[0115] In the data processing module 120 of the system, the specific implementation method for calculating the long-term rate of change K is one of the key technical points determining the system performance. This invention provides a preferred implementation scheme, namely, in this module, using a "sliding time window combined with least squares linear fitting" method to accurately and stably calculate K.

[0116] Specifically, the data processing module 120 internally maintains a data structure, typically a first-in-first-out (FIFO) circular buffer, to store temperature data for a recent period. The size of this buffer is determined by a preset sliding time window length. For example, if the window length is set to 1 hour and the sampling frequency is 1 time / minute, the buffer capacity is 60 data points. Whenever a new temperature data point (t...) is generated... n ,T n The data is received from the temperature acquisition module 110, and the data processing module 120 performs the following operations:

[0117] 1. Update Buffer: Pushes new data points into the buffer and pops up the oldest data point in the window.

[0118] 2. Perform linear fitting: For all N data points (t) in the buffer... i ,T i For each i=1…N, perform a univariate linear regression. According to the least squares principle, the slope K (i.e., the long-term rate of change) of the fitted line is calculated using the following formula:

[0119]

[0120] In embedded systems, to reduce computation and avoid floating-point operations, the time coordinates can be normalized. For example, the time of the first point in the window can be set to 0, and the time of the remaining points can be set to the offset relative to that point, thereby simplifying the calculation.

[0121] Output: The calculated K value, along with the current temperature T. n Together, they are transmitted to the early warning analysis module 130.

[0122] The advantage of this implementation method lies in its clear mathematical principles and the definite physical meaning of the calculation results—namely, the average rate of temperature change per unit time. Furthermore, because it utilizes all the data within the window, the influence of individual noise points on the K value is effectively suppressed, ensuring the stability of the K value.

[0123] In other feasible embodiments of the present invention, in order to adapt to different computing platforms or scenarios with higher real-time requirements, the above computing method can be extended and optimized in various ways:

[0124] Recursive Least Squares (RLS): Standard linear fitting requires recalculating all data within the window each time, which can be computationally intensive when the window is large. RLS can be optimized by using the recursive formula to quickly update the K value at each new data point, utilizing the previous calculation results and the current data, without needing to retain all historical data. This significantly reduces computational complexity and storage requirements, which is particularly important for systems deployed on resource-constrained embedded devices.

[0125] Data Compression and Approximate Calculation: For scenarios with extremely high sampling frequencies (e.g., once per second), maintaining a one-minute window requires 60 data points. Lossy compression of the original data can be performed; for example, averaging the data every 10 seconds before storing it in the window. This significantly reduces the number of data points that need to be processed within the window while still preserving trend information. Simultaneously, during the fitting process, numerical calculation methods (such as table lookup) can be used to replace complex multiplication and division operations, further improving computational efficiency.

[0126] Robust Regression: Standard linear regression is highly sensitive to outliers. If there are significant noise points (outliers) within the window caused by occasional disturbances, they can severely impact the fitting results. Robust regression methods, such as those using the Huber loss function or the RANSAC algorithm, can automatically identify and reduce the weight of these outliers during the fitting process. This makes the calculated K value more robust, truly reflecting the core trend of temperature rather than being swayed by extreme values.

[0127] Dynamic window length adjustment: A mechanism can be designed so that the length of the sliding window is no longer fixed. For example, when the system detects sharp fluctuations in the short-term rate of change, the window length can be automatically shortened to improve the response speed to rapid changes. Conversely, when the system is in a steady state, the window length is increased to obtain a smoother and more stable long-term trend. This adaptive window adjustment strategy enables the system to maintain optimal performance under different operating conditions.

[0128] Although the specific algorithms and optimization methods of the above implementation methods differ, they are all essentially deepening and enhancing the internal functions of the data processing module 120. The ultimate technical effect they achieve is consistent: to efficiently, accurately, and robustly extract the key long-term rate of change K from the raw temperature data stream, providing a solid data foundation for the decision-making accuracy of the entire early warning system.

[0129] In the early warning analysis module 130 of the system described in this invention, calculating the dynamic prediction threshold is one of the core functions. This module not only needs to perform logical judgments but also needs to have built-in material models and algorithms to achieve advanced prediction of the future. This invention provides a preferred embodiment in which the dynamic threshold is calculated based on the thermal aging characteristic model of insulating materials and the Arrhenius equation in this module.

[0130] Specifically, the model parameters of the insulating materials used are pre-stored in the internal data structure or configuration file of the early warning analysis module 130. These parameters are usually in the form of lookup tables or formula coefficients. For example, according to the Arrhenius equation, the relationship between the aging rate r of the material and the temperature T (Kelvin) can be approximated as:

[0131]

[0132] E a Here, R is the activation energy, R is the gas constant, and A is the frequency factor. For the early warning analysis module 130, it does not directly concern itself with the absolute aging rate, but rather with the cumulative effect of temperature. Therefore, this model can be converted into a "lifespan consumption" model.

[0133] When performing dynamic threshold calculation, the early warning analysis module 130 will perform the following operations:

[0134] 1. Input reception: Obtain the current temperature T from the data processing module 120. now And the long-term rate of change K.

[0135] 2. Calculate the temperature-time curve: Assume that the current temperature change trend remains unchanged for a period of time, that is, the temperature T changes with time t as T(t) = T now +K˙t.

[0136] 3. Integral calculation of equivalent aging time: defined at a constant reference temperature T ref (For example, at the rated operating temperature of a busbar) the time required for the material to lose its entire lifespan is L. refAt a varying temperature T(t), the relationship between the equivalent lifetime dL consumed within a small time interval dt and dt is determined by the Arrhenius equation. The system can integrate in real time to calculate the consumption of the material's remaining lifetime from the current moment. Furthermore, the system can solve for a predicted temperature T. predict This means that starting from this temperature point, if the temperature is increased at the current rate of change K, in the future T... window The equivalent life consumed within a given time period is equal to the remaining life of the material.

[0137] 4. Output result: The final T obtained predict It serves as a dynamic prediction threshold for comparison with real-time temperature.

[0138] Compared to simple linear extrapolation, this method more accurately considers the nonlinear cumulative effect of temperature on insulation material damage, and can more scientifically balance "timely early warning" and "avoiding premature alarms".

[0139] In addition to this continuous integration-based method, the system of the present invention can also be extended to other more intelligent and scenario-based dynamic threshold calculation schemes:

[0140] 1. Dynamic Threshold Model Based on Historical Big Data: The early warning analysis module 130 can connect to a cloud database that stores massive amounts of historical operating data for all busbars of the same model under normal and fault conditions. By performing data mining and machine learning training on this data, a "dynamic threshold model" can be constructed. This model is no longer a simple formula, but a complex machine learning model (such as a random forest or gradient boosting tree). The model's input includes features such as current temperature, rate of change, load current, ambient temperature, and equipment operating years; the output is a dynamic, historically validated optimal early warning threshold. This approach offers the highest accuracy and can learn the optimal early warning strategy under various complex operating conditions.

[0141] 2. Dynamic Threshold Based on Digital Twin: A high-fidelity "digital twin" model can be established for each busbar. This model simulates the heat conduction and electro-thermal coupling processes of the physical busbar in real time in virtual space. The early warning analysis module 130 can call this digital twin model, input the current operating conditions, and let the model predict the temperature field distribution over a future period. The output accuracy of the digital twin model is much higher than that of simple linear extrapolation. When the model predicts that the hot spot temperature of the connector will be at T... window The system can trigger an early warning when the dangerous value is reached within a certain time. This represents the most advanced predictive maintenance technology in the industrial field.

[0142] 3. Dynamic Threshold with Multi-Parameter Coupling: The early warning analysis module 130 can incorporate data from more dimensions to dynamically adjust the threshold. For example, it can combine load current data. If the current is not increasing while the temperature continues to rise, it indicates that the contact resistance is increasing. In this case, the weight of the K value should be amplified, and the dynamic threshold should be lowered accordingly for earlier warning. Conversely, if the current and temperature rise simultaneously, it may just be a normal load change, and the dynamic threshold can be appropriately increased to avoid false alarms.

[0143] These extended solutions, whether based on precise materials physics models or machine learning models based on massive amounts of data, all aim to create a dynamic decision-making center within the system's early warning analysis module 130 that can "keep pace with the times" and "predict the future." The direct technical effect is that the system is no longer a passive numerical comparator, but an expert system capable of making accurate and intelligent decisions by combining the equipment's inherent physical characteristics with external environmental conditions.

[0144] It is worth mentioning that this invention can ultimately be integrated into a new type of insulated busbar trunking product, enabling the product itself to possess advanced hotspot temperature anomaly early warning capabilities, achieving a leap from "passively accepting faults" to "actively reporting risks." The physical structure of this product represents a further technological upgrade to existing intelligent busbar trunking systems.

[0145] Specifically, this new type of insulated busbar trunking mainly consists of three parts: the busbar trunking body, the embedded temperature sensor, and the integrated early warning system.

[0146] The busbar trunking body has the same structure as a conventional insulated busbar trunking, including conductive bars, insulating material, a metal shell, and connectors for electrical connections between sections. The improvement of this invention mainly lies in the connector portion.

[0147] The temperature sensor embedded in the connector forms the physical sensing layer of this invention. During manufacturing, a high-precision, high-reliability temperature sensor, such as a PT100 platinum resistance temperature sensor or a digital integrated temperature sensor chip, is securely fixed to the critical hotspot locations of the connector through pre-embedding or potting. For example, it can be attached tightly to the surface of the conductive busbar at the connector or embedded inside insulating material. The sensor's leads are routed to the intelligent monitoring terminal in the busbar trunking via shielded cables. This embedded design ensures that the sensor can directly and accurately sense the true internal temperature of the connector, avoiding interference from external ambient temperature.

[0148] The integrated early warning system is the "brain" of this product. It is typically installed as a standalone, high-protection intelligent monitoring terminal at the beginning box or section of the busbar trunking. This terminal integrates the aforementioned data processing module 120, early warning analysis module 130, and early warning output module 140. It is connected to multiple embedded temperature sensors via cables to collect data in real time. The terminal carries a microcontroller (MCU) and corresponding embedded software, implementing the core algorithm of this invention—calculating the rate of temperature change, constructing a "trend + threshold" dual early warning model, and outputting tiered early warning signals.

[0149] The final form of this product makes the early warning function a "plug-and-play" feature of the busbar trunking. After installation and power-on, the system begins to operate automatically. When early potential problems occur, the intelligent monitoring terminal can display early warning information on its built-in screen or upload the early warning signal to the user's host computer monitoring system via a communication interface (such as RS-485 or Ethernet).

[0150] In other feasible embodiments of the invention, this integrated busbar product can be extended and optimized in various ways:

[0151] Wireless Passive Sensor Integration: To further enhance the ease of installation and reliability of the product, embedded temperature sensors can utilize wireless passive technology. For example, surface acoustic wave (SAW) temperature sensors can be used. These sensors require no battery power or connecting cables, and are read wirelessly via radio frequency signals. During the busbar trunking manufacturing process, SAW sensor tags are embedded inside the connector. The accompanying reader antenna can be mounted on the exterior of the busbar trunking housing, allowing for non-contact temperature data reading. This approach completely eliminates the installation costs and potential points of failure associated with sensor cables, making it particularly suitable for large busbar trunking systems with long distances and multiple nodes.

[0152] Full communication interface support: The intelligent monitoring terminal can integrate multiple industrial communication protocols, such as Modbus RTU, Modbus TCP / IP, PROFINET, IEC61850, etc. This enables the new bus trunking products to seamlessly connect to various building automation systems, power monitoring systems, or industrial control systems, achieving higher-level system linkage and data sharing.

[0153] Integrated Molding Process: To improve the overall integrity and protection level (IP rating) of the product, the intelligent monitoring terminal can be integrated with the busbar trunking housing. For example, the terminal circuit board can be encapsulated within an explosion-proof cavity in a metal casing, leaving only the communication interface and status indicator lights exposed. This design enables the product to be used in harsh environments with extremely high protection requirements, such as marine engineering and chemical plants.

[0154] Scalable edge computing capabilities: The intelligent monitoring terminal can be designed with enhanced edge computing capabilities. It can not only run the early warning algorithm of this invention, but also act as an edge node to aggregate, analyze, and store data from multiple busbar units under its jurisdiction. For example, it can perform temperature data comparison and analysis between multiple busbar sections, identifying abnormal sections with excessive temperature differences, thereby achieving more macroscopic fault diagnosis.

[0155] By deeply integrating the early warning system into the busbar trunking product, this invention not only provides a method but also a physical product with inherent intelligence. The technical benefits of this product are obvious: it internalizes complex monitoring and early warning technologies into the product's inherent functions, reducing system integration costs and technical barriers for users, while simultaneously improving the intelligence level and operational reliability of the entire power distribution system.

[0156] The core solution of this invention can be implemented not only as a hardware system but also as a computer program product. That is, the method described in this invention—namely, the series of logical steps of "acquiring real-time temperature data, calculating the rate of change, generating a first warning signal based on the long-term rate of change, and generating a second warning signal based on a dynamic prediction threshold"—is written into computer-executable code and stored on a computer-readable storage medium. When this storage medium is installed on a general-purpose computer device or a dedicated embedded controller, such as… Figure 8 As shown, when executed by its processor, the method of the present invention can be implemented.

[0157] This implementation method offers exceptional flexibility and versatility. It means that the technical solution of this invention no longer depends on a specific hardware platform, but can be integrated into various existing systems as a software module or software library.

[0158] Specifically, the computer-readable storage medium can take various physical forms. For example, it can be an embedded non-volatile memory, such as a Flash chip or EEPROM, pre-installed in the intelligent monitoring terminal of the busbar trunking at the factory. The firmware program is burned into it before the terminal leaves the factory. When the terminal is powered on, its MCU automatically loads and executes the program code in the storage medium, thereby enabling the terminal to have the early warning function described in this invention.

[0159] It can also be a CD-ROM, DVD, or a USB flash drive. Maintenance personnel can install the program from these media onto a standard industrial computer or server. This computer connects to the busbar temperature sensors in the field via a data acquisition card or communication interface. After running the program, it can function as a centralized monitoring server, providing unified early warning analysis for all busbars.

[0160] It can also be virtual storage space on a cloud server. The program code can be deployed on a cloud platform. The on-site busbars send data to the cloud via an IoT gateway. A server instance in the cloud loads and executes the program code, enabling parallel processing and analysis of massive amounts of device data. The analysis results are then pushed back to the user terminal through the cloud platform's service interface.

[0161] In other feasible embodiments of the present invention, the functionality of the computer program product can be expanded in many ways to adapt to different deployment environments and performance requirements:

[0162] Software as a Service (SaaS) Model: A SaaS platform can be built based on the programs stored on this storage medium. Users do not need to purchase any servers or software; they only need to open an account for their busbar equipment to remotely use the early warning service of this invention via a webpage or APP. All data storage, analysis, and early warning push are completed in the cloud, and users pay only as needed, greatly reducing users' initial investment and operation and maintenance costs.

[0163] Edge computing version and cloud computing version: The program stored on the storage medium can be developed in two versions. One version is a lightweight edge computing version, with highly optimized code, small memory footprint, and low power consumption, for deployment on smart gateways or embedded devices in the field to achieve low-latency local alerts. The other version is a powerful cloud computing version, deployed on server clusters, capable of processing massive amounts of data from thousands of devices and utilizing more sophisticated machine learning models for deep analysis and prediction.

[0164] Open API Interface: The program code can be designed with standard application programming interfaces (APIs) reserved. This allows other third-party systems (such as building automation systems and power dispatching systems) to easily call the early warning function of this invention through the API, obtain real-time analysis results, or configure early warning parameters. This openness is key to the widespread application of this invention's technical solution.

[0165] Visual configuration tool: Stored alongside the program on the medium, a visual configuration tool may also be included. Users can use this tool to intuitively set parameters such as the sliding time window length, aging rate threshold, hazardous temperature value, and safe handling time window without directly modifying the code. This design allows the system to flexibly adapt to different models of busbars with different insulation materials, as well as scenarios with varying operation and maintenance requirements.

[0166] By using a computer-readable storage medium, the method of this invention is freed from being confined to a "dedicated device" and transformed into a "universal capability." Its technical advantage lies in significantly lowering the barrier to technology adoption, enabling any device with basic computing capabilities—whether a simple microcontroller or a powerful cloud server—to quickly obtain the advanced early warning capabilities based on trend analysis and material properties provided by this invention by running the program on the medium.

[0167] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0168] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for early warning of abnormal hot spot temperature in an insulated busbar trunking, characterized in that, include: Obtain real-time temperature data of the busbar connectors; Based on the real-time temperature data, calculate the rate of temperature change over time, including both short-term and long-term rates of change. When the long-term rate of change is positive and exceeds a preset aging rate threshold, a first warning signal is generated; Based on the real-time temperature data and the long-term rate of change, a predicted threshold below the absolute temperature threshold is dynamically calculated. When the real-time temperature data reaches the predicted threshold, a second warning signal is generated.

2. The method for early warning of abnormal hot spot temperature in insulated busbar trunking according to claim 1, characterized in that, The calculation of the rate of change of temperature over time includes: Using a sliding time window as the unit, a linear fit is performed on the real-time temperature data within the sliding time window, and the slope obtained from the fit is used as the long-term rate of change.

3. The method for early warning of abnormal hot spot temperature in insulated busbar trunking according to claim 1, characterized in that, The step of dynamically calculating a predicted threshold below the absolute temperature threshold includes: Obtain the thermal aging characteristic curves of the busbar insulation material; Based on the thermal aging characteristic curve, the current temperature value, and the long-term rate of change, predict the time required for the temperature to reach the absolute temperature threshold. The prediction threshold is determined based on the required time and the thermal aging characteristic curve.

4. The method for early warning of abnormal hot spot temperature in insulated busbar trunking according to claim 1, characterized in that, It also includes the following steps: The acquired real-time temperature data is filtered to eliminate instantaneous noise.

5. The method for early warning of abnormal hot spot temperature in insulated busbar trunking according to claim 1, characterized in that, The first warning signal is used to indicate that the aging process of the insulation material is accelerating, and the second warning signal is used to indicate that immediate maintenance intervention is required.

6. A hotspot temperature anomaly early warning system for an insulated busbar trunking, characterized in that, include: The temperature acquisition module is used to acquire real-time temperature data of the busbar connection head; The data processing module is used to calculate the rate of change of temperature over time based on the real-time temperature data, wherein the rate of change includes short-term rate of change and long-term rate of change. The early warning analysis module is used to generate a first early warning signal when the long-term change rate is positive and exceeds a preset aging rate threshold; and to dynamically calculate a predicted threshold below the absolute temperature threshold based on the real-time temperature data and the long-term change rate, and to generate a second early warning signal when the real-time temperature data reaches the predicted threshold. The warning output module is used to output the first warning signal and / or the second warning signal.

7. The hot spot temperature anomaly early warning system for insulated busbar trunking according to claim 6, characterized in that, The data processing module is specifically used to perform linear fitting on the real-time temperature data within the sliding time window in units of sliding time window, and to use the slope obtained from the fitting as the long-term rate of change.

8. The hot spot temperature anomaly early warning system for insulated busbar trunking according to claim 6, characterized in that, The early warning analysis module is specifically used to obtain the thermal aging characteristic curve of the busbar insulation material, predict the time required for the temperature to reach the absolute temperature threshold based on the thermal aging characteristic curve, the current temperature value, and the long-term change rate, and determine the predicted threshold based on the required time and the thermal aging characteristic curve.

9. An insulated busbar trunking system, characterized in that, include: Busbar trunking body; A temperature sensor embedded in the connector; and, The hot spot temperature anomaly early warning system for insulated busbar trunking as described in any one of claims 6-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the hot spot temperature anomaly early warning method for insulated busbar trunking as described in any one of claims 1-5.

Images