Heat dissipation control method and system for bus duct

Abstract

The invention relates to the field of bus duct heat dissipation control, in particular to a bus duct-oriented heat dissipation control method and system, which realize fundamental transformation from passive response to active prediction and have the core value of collaborative optimization of accurate prevention and economic energy conservation. The system constructs a model capable of accurately reflecting a bus duct temperature change rule under different working conditions by analyzing historical data, and extracts a window with a significant temperature rise trend from the model. The optimal advanced intervention time point of enhanced heat dissipation is calculated based on the thermal inertia principle, and precise regulation and control are started before temperature rise starts. Finally, the heat dissipation intensity is dynamically adjusted through an optimization algorithm, the real-time electricity price and equipment loss are comprehensively considered while it is ensured that the temperature rise slope is effectively restrained to be below a safety threshold value, and the heat dissipation cost is minimized. According to the whole set of scheme, traditional extensive heat dissipation based on a threshold value is upgraded into refined and preventive energy management based on prediction, and the energy economy is remarkably improved on the premise that long-term safe and stable operation of equipment is guaranteed.

Description

Technical Field

[0001] This application relates to the field of busbar heat dissipation control, specifically a heat dissipation control method and system for busbars. Background Technology

[0002] Busbar trunking, also known as busbar trunking systems or "bus bridges," is an electrical device used for the efficient and centralized distribution of high-power electrical energy. Heat dissipation is one of the most critical technical considerations in the design, selection, and application of busbar trunking. Its importance directly affects the system's safety, efficiency, lifespan, and cost.

[0003] Currently, busbar cooling primarily relies on passive cooling, which depends on heat exchange between the outer casing and the external air. However, for busbars installed in environments with high ambient temperatures or extremely poor ventilation (such as enclosed shafts or narrow mezzanines), this method is insufficient for effective cooling. Ordinary active cooling methods (such as air cooling and water cooling) are triggered when a certain temperature threshold is reached. During this process, the busbar has already undergone a short-term "thermal shock," which, over time, severely damages the lifespan of heat-generating components, thermal interface materials (such as silicone grease), solder joints, and water-cooled plate welds. Furthermore, cooling only begins after the heating trend has already established itself, resulting in poor short-term cooling performance. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a heat dissipation control method and system for bus trunking to solve the problems in the prior art.

[0005] To achieve the above objectives, this application adopts the following technical solution:

[0006] This application discloses a heat dissipation control method for busbar trunking, comprising the following steps:

[0007] Acquire temperature data samples from multiple historical periods of the busbar trunking, wherein the temperature data samples contain temperature values ​​at multiple sampling time points within one period;

[0008] The temperature data samples are analyzed for regularity to obtain a temperature change model, which includes a sub-model that represents the temperature values ​​at multiple time points within a variety of typical scenario cycles.

[0009] The temperature rise windows and temperature rise slopes of multiple temperature sampling points in various typical scenario cycles are extracted from the temperature change model, and target temperature rise windows and temperature sampling points corresponding to the target temperature rise windows are selected with temperature rise slopes greater than or equal to a preset slope threshold.

[0010] Determine the target temperature rise window corresponding to the current control cycle, set the enhanced heat dissipation time point and enhanced heat dissipation load based on the target temperature rise window, and perform heat dissipation control on the active heat dissipation device based on the enhanced heat dissipation time point and the enhanced heat dissipation load, wherein the enhanced heat dissipation time point is before the target temperature rise window, and the active heat dissipation device is an air-cooled device or a water-cooled device.

[0011] In one embodiment of this application, a temperature change model is obtained by performing regularity analysis on the temperature data samples, including:

[0012] Temperature data samples from multiple historical periods are preprocessed to obtain multiple preprocessed data samples. The preprocessing includes data completion, filtering, and normalization.

[0013] Unsupervised clustering based on dynamic time warping is performed on the multiple preprocessed data samples to obtain multiple data clusters; and scenario labeling is performed on the multiple data clusters to obtain multiple scenario clusters.

[0014] Calculate the average correlation coefficient of temperature data samples within each scenario cluster, and based on the average correlation coefficient, divide multiple scenario clusters into clusters with definite patterns, clusters with uncertain patterns, and clusters with no patterns.

[0015] For the clusters where the pattern is determined, all preprocessed data samples within the cluster are aligned, and the average temperature value at each time point is calculated. A mean model is also constructed based on the average temperature values ​​at multiple time points within the period.

[0016] For clusters with uncertain patterns, the preprocessed data samples are fitted with a Gaussian process model to obtain a Gaussian process model, wherein the Gaussian process model characterizes the predicted temperature values ​​at multiple time points and the confidence intervals of the predicted temperature values.

[0017] A temperature change model is constructed based on the mean model and the Gaussian process model.

[0018] In one embodiment of this application, the average correlation coefficient of temperature data samples within each scenario cluster is calculated, and multiple scenario clusters are divided into clusters with definite patterns, clusters with uncertain patterns, and clusters with no patterns based on the average correlation coefficient, including:

[0019] Calculate the correlation between any two temperature data samples within each scenario cluster. The relevance The mathematical expression is:

[0020]

[0021] In the formula, This represents one of the temperature data samples. This represents another temperature data sample. Representing temperature data samples The first in One value, Representing temperature data samples The average value in Representing temperature data samples The first in One value, Representing temperature data samples The average value in This indicates the number of time points in the temperature data sample;

[0022] Calculate all relevance The average value is used to obtain the average correlation coefficient;

[0023] Clusters of scenarios with an average correlation coefficient greater than the upper limit of the preset screening range are classified as clusters with a known pattern; clusters of scenarios with an average correlation coefficient falling within the preset screening range are classified as clusters with an uncertain pattern; and clusters of scenarios with an average correlation coefficient less than the lower limit of the preset screening range are classified as clusters with no pattern.

[0024] In one embodiment of this application, extracting the temperature rise window and the temperature rise slope of multiple temperature sampling points for various typical scenario cycles from the temperature change model includes:

[0025] Data fragments of multiple size time windows are obtained from the temperature change model based on a pre-constructed multi-size sliding window.

[0026] For a data segment of the mean model, a linear fit is performed on the data segment to obtain the segment slope;

[0027] For a data segment of a Gaussian process model, calculate the upper bound slope and the lower bound slope of the confidence interval respectively; and calculate the average of the upper bound slope and the lower bound slope to obtain the segment slope.

[0028] The slope of the segment of the mean model is compared with a preset initial slope threshold, and the lower bound slope of the Gaussian process model is compared with a preset initial slope threshold; and data segments with a slope greater than or equal to the initial slope threshold are selected as candidate data segments.

[0029] The time window of the candidate data segment is used as the candidate window, and consecutive candidate windows of the same scale are merged to obtain a merged window of multiple sizes.

[0030] By merging windows of different sizes, a temperature rise window is obtained;

[0031] Linear fitting is performed on the data segments within the temperature rise window to obtain the temperature rise slope.

[0032] In one embodiment of this application, merging windows of different sizes to obtain a temperature rise window includes:

[0033] Calculate the overlap of merged windows of different sizes , wherein the degree of overlap The mathematical expression is:

[0034]

[0035] In the formula, Indicates the first A merged window, Indicates the first A merged window, Display window The start time, Display window The start time, Display window End time, Display window End time;

[0036] The overlap is compared with a preset overlap threshold. If the overlap is greater than or equal to the preset overlap threshold, the merged windows of different sizes are merged to obtain a temperature rise window; otherwise, the merged windows of different sizes are removed.

[0037] In one embodiment of this application, setting an enhanced heat dissipation time point and an enhanced heat dissipation load based on the target temperature rise window includes:

[0038] Extract the start time of the target temperature rise window And based on the said start time Calculate the time point for enhanced heat dissipation The enhanced heat dissipation time point The mathematical expression is:

[0039]

[0040] In the formula, This indicates the response delay of the active cooling device. This represents the predicted temperature rise slope of the target temperature rise window. Indicates the basic heat dissipation balance slope. For the target temperature rise slope, The system thermal time constant characterizes the system's response speed to temperature changes;

[0041] Based on the enhanced heat dissipation time point An optimization model is constructed, wherein the objective of the optimization model is to minimize energy consumption, and the optimization model includes temperature constraints, temperature rise rate constraints, and physical constraints.

[0042] Solving the optimization model yields the enhanced heat dissipation load.

[0043] In one embodiment of this application, based on the enhanced heat dissipation time point Constructing an optimization model includes:

[0044] Construct constraints, wherein the constraints include temperature constraints, temperature rise rate constraints, and physical constraints;

[0045] The mathematical expression for the temperature constraint is:

[0046]

[0047] In the formula, Indicates that the busbar trunking is in Temperature at any moment This is the upper limit of the temperature. Indicates the end time of the target temperature rise window;

[0048] The mathematical expression for the temperature rise rate constraint is:

[0049]

[0050] The mathematical expression for the physical constraint is:

[0051]

[0052]

[0053] In the formula, Indicates the load on the active cooling equipment. This indicates the upper limit of the load change rate;

[0054] Construct an objective function with the goal of minimizing energy consumption, wherein the mathematical expression of the objective function is:

[0055]

[0056]

[0057]

[0058]

[0059] In the formula, Represents the total cost. Indicates energy consumption cost, This indicates the cost of equipment depreciation. This indicates a penalty for exceeding the limit. The rated power of the active cooling device. express Electricity price at any time Indicates the loss coefficient. Indicates the penalty coefficient. express The actual temperature rise slope at any given time.

[0060] In one embodiment of this application, solving the optimization model to obtain the enhanced heat dissipation load includes:

[0061] Time window for enhanced heat dissipation Discretization is performed to obtain multiple control time periods;

[0062] The optimization model is solved using a solver to obtain the enhanced heat dissipation load for multiple control time periods.

[0063] In one embodiment of this application, determining the target temperature rise window corresponding to the current control cycle includes:

[0064] The temperature change model is divided into multiple time periods for labeling various typical scenarios.

[0065] The label of the target scenario and the corresponding target sub-model are determined based on the current time and the time period of labels of multiple typical scenarios;

[0066] Extract the target temperature rise window from the target sub-model.

[0067] This application also provides a heat dissipation control system for busbar trunking, including:

[0068] The acquisition module is used to acquire temperature data samples of multiple historical periods of the bus trunking, wherein the temperature data samples contain temperature values ​​at multiple sampling time points within one period;

[0069] The pattern analysis module is used to perform pattern analysis on the temperature data sample to obtain a temperature change model, wherein the temperature change model includes a sub-model that represents the temperature values ​​corresponding to multiple time points within a variety of typical scenario cycles.

[0070] The temperature rise window extraction module is used to extract the temperature rise window and the temperature rise slope of multiple temperature sampling points in multiple typical scenario cycles from the temperature change model, and to filter out the target temperature rise window and the temperature sampling point corresponding to the target temperature rise window whose temperature rise slope is greater than or equal to a preset slope threshold.

[0071] A heat dissipation control module is used to set an enhanced heat dissipation time point and an enhanced heat dissipation load based on the target temperature rise window; and to perform heat dissipation control on the active heat dissipation device based on the enhanced heat dissipation time point and the enhanced heat dissipation load, wherein the enhanced heat dissipation time point is before the target temperature rise window, and the active heat dissipation device is an air-cooled device or a water-cooled device.

[0072] The beneficial effects of this application are as follows: This application presents a heat dissipation control method and system for busbar trunking, achieving a fundamental shift from passive response to proactive prediction. Its core value lies in the synergistic optimization of precise prevention and energy conservation. The system analyzes historical data to construct a model that accurately reflects the temperature change patterns of busbar trunking under different operating conditions, and extracts windows with significant temperature rise trends. Based on the principle of thermal inertia, it calculates the optimal early intervention point for enhanced heat dissipation, initiating precise control before the temperature rise begins. Finally, through optimized algorithms, the heat dissipation intensity is dynamically adjusted, ensuring that the temperature rise slope is effectively suppressed below a safe threshold while comprehensively considering real-time electricity prices and equipment losses to minimize heat dissipation costs. The entire solution upgrades traditional threshold-based, extensive heat dissipation to a prediction-based, refined, and preventative energy management system, significantly improving energy economy while ensuring the long-term safe and stable operation of equipment. Attached Figure Description

[0073] The present application will be further described below with reference to the accompanying drawings and embodiments:

[0074] Figure 1 This is an application scenario diagram of a heat dissipation control method for busbar trunking shown in one embodiment of this application;

[0075] Figure 2 This is a flowchart illustrating a heat dissipation control method for a busbar trunking according to one embodiment of this application;

[0076] Figure 3 This is a schematic diagram of the model building process in one embodiment of this application;

[0077] Figure 4 This is a structural diagram of a heat dissipation control system for a busbar trunking according to one embodiment of this application. Detailed Implementation

[0078] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0079] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the layers related to this application and are not drawn according to the actual number, shape and size of the layers in the actual implementation. In the actual implementation, the form, number and proportion of each layer can be arbitrarily changed, and the layer layout may also be more complex.

[0080] Numerous details are explored in the following description to provide a more thorough explanation of embodiments of this application; however, it will be apparent to those skilled in the art that embodiments of this application may be practiced without these specific details.

[0081] Figure 1 This is an application scenario diagram of a heat dissipation control method for busbar trunking shown in one embodiment of this application, such as... Figure 1 As shown, in this application, a cooling fan 130 (or water-cooled cooling) is installed on the side of the busbar trunking 110. Temperature sensors 120 are installed at key locations of the busbar trunking 110 (such as connection points (plug boxes, starting boxes, terminal boxes, etc.), inside plug units (sub-boxes / plug boxes), elbows, diameter changes or structural abrupt changes, and near high-power load output terminals, etc.). The overall temperature is the average value collected by multiple temperature sensors 120. The edge controller 140 sends the collected temperatures to the server 150 for storage and modeling. After modeling is completed, the server 150 outputs enhanced cooling command values ​​to the edge controller 140 according to a pre-planned time window. The cooling fan 130 is controlled by the edge controller 140. This enhances cooling of the busbar trunking 110 before the temperature rise window arrives, avoiding thermal shock.

[0082] In addition, in the above scenarios, the main heat dissipation solution is a supplement to the existing heat dissipation solution, and the basic heat dissipation is still provided by traditional convection cooling / fan base cooling.

[0083] Figure 2 This is a flowchart illustrating a heat dissipation control method for a busbar trunking system in one embodiment of this application, as shown below. Figure 2 As shown, a heat dissipation control method for busbar trunking in this application mainly includes the following steps:

[0084] S210, acquire temperature data samples of multiple historical periods of the bus trunking, wherein the temperature data samples contain temperature values ​​at multiple sampling time points within one period;

[0085] In this application, the temperature data samples are provided by Figure 1The scenario shown was captured during long-term operation. Specifically, due to seasonal temperature variations, at least one year of data collection is required. For busbars in similar scenarios, temperature data samples from other locations can be used as the temperature data samples for the current busbar.

[0086] Each historical cycle lasts one day. Due to the periodic changes in power equipment, the conductor current and heat generation in the busbar also exhibit corresponding periodic changes. This is the fundamental principle for subsequent periodic analysis and model construction.

[0087] S220, Perform regularity analysis on the temperature data sample to obtain a temperature change model, wherein the temperature change model includes a sub-model that characterizes the temperature values ​​corresponding to multiple time points within a variety of typical scenario cycles.

[0088] In this application, we first use the temperature data samples collected in the early stage to build a model, and use the temperature change model to reflect the relationship between temperature and time under various typical scenarios. Figure 3 This is a schematic diagram of the model building process in one embodiment of this application, such as... Figure 3 As shown, the specific model building process is as follows:

[0089] S221, preprocess the temperature data samples from multiple historical periods to obtain multiple preprocessed data samples, wherein the preprocessing includes data completion, filtering and normalization.

[0090] Specifically, time series interpolation (such as linear interpolation and spline interpolation) or patterns based on adjacent periods are used to fill in missing values ​​to ensure the continuity of the time series. Low-pass filtering (such as moving average and Savitzky-Golay filter) is used to remove high-frequency noise and preserve the low-frequency trend of temperature changes. Furthermore, a min-max normalization algorithm is used to map temperature values ​​to the [0,1] interval to eliminate dimensional differences between different busbars and different measurement channels.

[0091] S222, perform unsupervised clustering based on dynamic time warping on the multiple preprocessed data samples to obtain multiple data clusters; and perform scenario labeling on the multiple data clusters to obtain multiple scenario clusters;

[0092] Unsupervised clustering based on Dynamic Time Warping (DTW) calculates the minimum curved path distance between two time series, handling scaling and offsets along the time axis and capturing shape similarity rather than simple point-to-point distances. Based on the DTW distance matrix, hierarchical clustering or spectral clustering algorithms are used to automatically group temperature curves with similar shapes.

[0093] After obtaining the data clusters, semantic labels are assigned manually or semi-automatically based on the time characteristics (season, weekday / weekend), load characteristics, and environmental characteristics of each cluster.

[0094] DTW can identify similarities in temperature change patterns (such as temperature rise trends and peak positions) without being limited by absolute time alignment. It discovers typical operating scenarios for busbars in a data-driven manner, requiring no prior knowledge. The labeled scenario clusters become interpretable operating condition classifications, providing a basis for subsequent control strategy selection.

[0095] S223, calculate the average correlation coefficient of temperature data samples within each scenario cluster, and based on the average correlation coefficient, divide the multiple scenario clusters into clusters with definite patterns, clusters with uncertain patterns, and clusters with no patterns.

[0096] Specifically, it includes:

[0097] S2231, Calculate the correlation between any two temperature data samples within each scenario cluster. The relevance The mathematical expression is:

[0098]

[0099] In the formula, This represents one of the temperature data samples. This represents another temperature data sample. Representing temperature data samples The first in One value, Representing temperature data samples The average value in Representing temperature data samples The first in One value, Representing temperature data samples The average value in This indicates the number of time points in the temperature data sample;

[0100] Specifically, in this embodiment, the Pearson correlation coefficient is used to measure the linear similarity between temperature curves within a cluster, reflecting the consistency of shape and amplitude.

[0101] S2232, Calculate all relevance scores The average value is used to obtain the average correlation coefficient;

[0102] S2233, the scenario clusters with average correlation coefficients greater than the upper limit of the preset screening range are designated as clusters with certain patterns, the scenario clusters with average correlation coefficients falling within the preset screening range are designated as clusters with uncertain patterns, and the scenario clusters with average correlation coefficients less than the lower limit of the preset screening range are designated as clusters with no patterns.

[0103] The scenario clusters are divided into three categories by using preset thresholds (e.g., 0.8 / 0.6), corresponding to different modeling strategies. Specifically:

[0104] Clusters with a value greater than 0.8: Clusters with defined patterns;

[0105] Clusters with scenarios prior to 0.6-0.8: Clusters with uncertain patterns;

[0106] Clusters with a value less than 0.6: Irregular clusters.

[0107] S224, For the pattern-determined cluster, align all preprocessed data samples within the cluster, calculate the average temperature value at each time point, and construct a mean model based on the average temperature values ​​at multiple time points within the period;

[0108] Specifically, determining clusters based on patterns The average of all periodic data within the time frame is calculated to obtain a typical daily temperature template for this scenario. The mathematical expression for the mean model is:

[0109]

[0110] In the formula, For context indexing, This represents the predicted temperature value at time point t. Indicates the first A sample of temperature values ​​at time point t;

[0111] In addition, smooth interpolation (such as cubic splines) and fitting can be performed on the model to obtain smooth curves.

[0112] S225, For the uncertain cluster of regularity, the preprocessed data sample is fitted based on the Gaussian process model to obtain the Gaussian process model, wherein the Gaussian process model characterizes the temperature prediction values ​​at multiple time points and the confidence interval of the temperature prediction values.

[0113] Gaussian process regression (GPR) is used for modeling, providing predicted values ​​and their uncertainty ranges (confidence intervals). The mathematical expression of the Gaussian process model is:

[0114]

[0115] In the formula, This represents the predicted temperature at time t. This represents the mean function (the expected trend of temperature change under the standard scenario). This represents the covariance function (a measure of the correlation between temperatures at different time points).

[0116] The Gaussian process regression model described above does not directly output confidence intervals. Instead, it uses the covariance function to calculate the covariance between the predicted time point and all known historical time points, obtaining a vector. Historical data and confident covariance form a covariance matrix. Therefore, the uncertainty can be calculated using the following formula. (This can be understood as variance); utilizing uncertainty Confidence intervals can then be constructed. ,in, This represents the expected value of the mean function output. This is the range parameter.

[0117] S226, Construct a temperature change model based on the mean model and the Gaussian process model.

[0118] Finally, the mean model (deterministic) and Gaussian process model (probabilistic) are integrated into a unified model library according to scenario classification, and a unified calling interface is defined for all sub-models (input time point, output temperature prediction and uncertainty).

[0119] S230, extract the temperature rise window and temperature rise slope of multiple temperature sampling points for multiple typical scenario cycles from the temperature change model, and filter out the target temperature rise window and the temperature sampling point corresponding to the target temperature rise window whose temperature rise slope is greater than or equal to a preset slope threshold.

[0120] For different typical scenarios, the process of extracting the target temperature rise window includes:

[0121] S231, Based on the pre-constructed multi-size sliding window, extract data fragments of multiple size time windows from the temperature change model;

[0122] Temperature changes exhibit multi-timescale characteristics—rapid sudden temperature rises (minutes), regular load changes (hours), and long-term trends (hours). A single-size window cannot capture them all.

[0123] In this embodiment, data is analyzed independently within each window by sliding along the time axis at fixed steps, ensuring that no potential temperature rise segment is missed. Multiple window levels are typically set, such as 15 minutes (capturing sudden changes), 30 minutes (regular changes), and 60 minutes (long-term trends), to cover the entire spectrum. This ensures that both rapid events and slow trends are effectively detected.

[0124] S232, For the data segment of the mean model, perform linear fitting on the data segment to obtain the segment slope;

[0125] The mean model has filtered out random fluctuations and represents the most typical temperature change curve under this scenario. Linear regression on the time-temperature data points within the window shows a high goodness of fit (R²). The slope directly reflects the average temperature rise rate during this period.

[0126] Furthermore, linear fitting is an existing technical method, and its implementation process will not be elaborated here.

[0127] S233, For the data segment of the Gaussian process model, calculate the upper bound slope and the lower bound slope of the confidence interval respectively; and calculate the average of the upper bound slope and the lower bound slope to obtain the segment slope;

[0128] The Gaussian process model outputs a predicted distribution, with a mean, upper bound, and lower bound at each time point. In this embodiment, the slopes of the upper and lower bounds of the confidence interval are calculated:

[0129] Upper bound slope: Represents the worst-case scenario (fastest temperature rise);

[0130] Lower bound slope: Represents the best-case scenario (slowest temperature rise);

[0131] Average slope: Take the average of the two as the best estimate;

[0132] In this embodiment, the lower bound slope is used for screening to ensure that temperature rise can be detected even under the lower bound condition.

[0133] S234, compare the slope of the segment of the mean model with a preset initial slope threshold, and compare the lower bound slope of the Gaussian process model with a preset initial slope threshold; and select data segments with a slope greater than or equal to the initial slope threshold as candidate data segments;

[0134] The initial slope threshold is usually set higher than the final control threshold for the first round of screening to reduce computational burden. The purpose of the first round of screening is to quickly eliminate a large number of insignificant windows, reducing the burden of subsequent processing.

[0135] S235, the time window of the candidate data segment is used as the candidate window, and consecutive candidate windows of the same scale are merged to obtain a merged window of multiple sizes;

[0136] For windows of the same size, the physical temperature rise process is usually continuous, and the same event segmented by the sliding window should be merged. Window merging with short time intervals (e.g., 1-2 sampling points) is permitted to address detection fluctuations. The merged window must reach a minimum duration to avoid fragmentation.

[0137] By merging the data, the complete temperature rise process can be reconstructed, facilitating feature extraction. This also prevents the same physical event from being reported as multiple short events.

[0138] S236, merges windows of different sizes to obtain a temperature rise window. The specific merging process includes:

[0139] S2361, Calculate the overlap of merged windows of different sizes. , wherein the degree of overlap The mathematical expression is:

[0140]

[0141] In the formula, Indicates the first A merged window, Indicates the first A merged window, Display window The start time, Display window The start time, Display window End time, Display window End time;

[0142] S2362, compare the overlap with a preset overlap threshold, and when the overlap is greater than or equal to the preset overlap threshold, merge the merging windows of different sizes to obtain a temperature rise window; otherwise, remove the merging windows of different scales.

[0143] In the overlap calculation formula, the shorter window is used as the denominator to ensure that events detected by the smaller window are more sensitive.

[0144] For time windows of different scales, this application calculates the time overlap ratio of different scale windows to measure the probability that they detect the same event. If the overlap exceeds a threshold, it is considered that the same event has been detected, and fusion is performed. Single-scale detection may result in false alarms, while consistency across multiple scales ensures confirmation. Therefore, through multi-scale window verification, only events detected simultaneously by multiple scales are confirmed, improving reliability. Fusing the time boundaries of different scale windows yields more accurate event start and end times and can reduce the false alarm rate.

[0145] S237 performs linear fitting on the data segment within the temperature rise window to obtain the temperature rise slope.

[0146] Finally, a linear fit is performed on the fused complete temperature rise window (rather than the fragments) to obtain the final temperature rise slope. Then, the temperature rise slope is compared with a slope threshold to select the time window with rapid temperature rise as the target temperature rise window.

[0147] In determining the target temperature rise window, cross-validation at different time scales significantly improves detection reliability. Different strategies are employed for deterministic and probabilistic models, respecting their respective characteristics. Conservative strategies (such as lower bound slope selection) are used in key steps to ensure safety. A progressive process from fragmented to merged windows to fused windows gradually improves temporal accuracy. Fusing temporal information from different scales yields more accurate event boundaries. Finally, the slope is calculated over the complete event window to avoid fragmentation bias. This achieves an efficient and reliable conversion from temperature models to actionable temperature rise features, providing accurate input for subsequent predictive heat dissipation control.

[0148] S240, determine the target temperature rise window corresponding to the current control cycle, set the enhanced heat dissipation time point and enhanced heat dissipation load based on the target temperature rise window; and perform heat dissipation control on the active heat dissipation device based on the enhanced heat dissipation time point and the enhanced heat dissipation load, wherein the enhanced heat dissipation time point is before the target temperature rise window, and the active heat dissipation device is an air-cooled device or a water-cooled device.

[0149] In the specific control process, it is necessary to first determine the sub-model required for the current control cycle, and then extract the corresponding target temperature rise window from the sub-model to execute heat dissipation control. The specific process includes:

[0150] S241, the labels for various typical scenarios in the temperature change model are divided into multiple time periods;

[0151] S242, Determine the label of the target scenario and the corresponding target sub-model based on the current time and the time period of labels of multiple typical scenarios;

[0152] S243, Extract the target temperature rise window from the target sub-model.

[0153] Steps S241-S243 outline the process for scenario label matching and target model extraction. The busbar operating mode exhibits strong temporal regularity (season, date type, time period). By predefining the effective time period for each scenario label (e.g., the "summer peak workday" label corresponds to 9:00-17:00 on workdays from June to September), rapid scenario identification can be achieved. After matching the target scenario label, the pre-trained temperature change sub-model for that scenario is directly invoked, avoiding complex real-time calculations.

[0154] For example, if the current time is "July 15th (Wednesday) 14:30", the system automatically matches the "summer peak workday" scenario label and calls the corresponding sub-model. This model has learned the typical temperature rise pattern caused by the peak air conditioning load in the summer afternoon and can accurately predict the temperature changes in the next few hours.

[0155] S244, Extract the start time of the target temperature rise window. And based on the said start time Calculate the time point for enhanced heat dissipation The enhanced heat dissipation time point The mathematical expression is:

[0156]

[0157] In the formula:

[0158] This indicates the response delay of the active cooling device (device characteristics, such as the time it takes for the fan to start up to full speed (2-5 minutes)).

[0159] This represents the predicted temperature rise slope of the target temperature rise window (the slope extracted from the target temperature rise window).

[0160] This represents the basic heat dissipation equilibrium slope (the steady temperature rise slope when only the basic heat dissipation is applied).

[0161] The target temperature rise slope (the maximum allowable temperature rise rate for safe operation (e.g., 0.3°C / min));

[0162] The system thermal time constant (thermal inertia parameter of busbar, obtained through experiment or simulation (10-30 minutes)) is used to characterize the system's response speed to temperature changes.

[0163] The principle behind the above mathematical expression is as follows:

[0164] Assume that the temperature change follows a first-order system response:

[0165]

[0166] slope difference To eliminate part of the temperature rise slope (to avoid thermal shock);

[0167] Basic abilities The portion that has already been eliminated due to basic heat dissipation;

[0168] This indicates the proportion of the remaining items that need to be eliminated;

[0169] Temperature changes have inertia, and the heat dissipation effect is delayed. Ensure sufficient lead time for coverage;

[0170] This indicates that the slope of the temperature rise is from Down to The required time constant multiple. When near When the temperature reaches infinity, this term tends to infinity, which is consistent with physical reality (when the basic heat dissipation is close to equilibrium, the additional heat dissipation effect is limited).

[0171] The enhanced heat dissipation time calculated using the above formula can be used for predictive control, intervening before the temperature rise begins and transforming a passive response into proactive prevention. The more severe the temperature rise ( The larger the target or the more stringent the goal ( The smaller the value, the greater the lead time. Taking the maximum value provides double protection, preventing any delay factor from being underestimated.

[0172] For example:

[0173] Assuming the target temperature rise window begins

[0174] calculate .

[0175] If we take max(3,10.4) = 10.4min, then =15:00-10.4min=14:49:36.

[0176] Therefore, the system will activate enhanced cooling at 14:49 to prepare for the temperature rise starting at 15:00.

[0177] S245, based on the enhanced heat dissipation time point An optimization model is constructed, wherein the objective of the optimization model is to minimize energy consumption, and the optimization model includes temperature constraints, temperature rise rate constraints, and physical constraints.

[0178] Specifically, the mathematical expression for the optimization model is as follows:

[0179] The mathematical expression for the temperature constraint is:

[0180]

[0181] In the formula, Indicates that the busbar trunking is in Temperature at any moment This is the upper limit of the temperature. Indicates the end time of the target temperature rise window;

[0182] This is a hard safety constraint, ensuring that the busbar temperature never exceeds the maximum allowable temperature of the insulation material. It physically prevents serious faults such as insulation aging and thermal runaway. The constraint is effective throughout the entire control period, providing continuous protection.

[0183] The mathematical expression for the temperature rise rate constraint is:

[0184]

[0185] This is a quality control constraint that requires the actual temperature rise slope to not exceed the target value. It complements the temperature constraint, which prevents absolute exceedances, while the slope constraint prevents excessively rapid changes (even if the temperature does not reach the upper limit). Rapid temperature rise can cause mechanical thermal stress; this constraint ensures the long-term health of the equipment.

[0186] The mathematical expression for the physical constraint is:

[0187]

[0188]

[0189] In the formula, Indicates the load on the active cooling equipment. This indicates the upper limit of the load change rate;

[0190] These are equipment feasibility constraints. The first constraint ensures that the load is within the equipment's capacity (0 for off, 1 for full load). The second constraint limits the rate of load change to prevent electrical shocks, mechanical stress, and to control oscillations.

[0191] For example: if the active cooling device is a variable frequency fan. =0.7 indicates operation at 70% speed. Constraint This means that the rotational speed change per minute will not exceed 20% of the full range, preventing damage to the motor and transmission mechanism.

[0192] Construct an objective function with the goal of minimizing energy consumption, wherein the mathematical expression of the objective function is:

[0193]

[0194]

[0195]

[0196]

[0197] In the formula, Represents the total cost. Indicates energy consumption cost, This indicates the cost of equipment depreciation. This indicates a penalty for exceeding the limit. The rated power of the active cooling device. express Electricity price at any time Indicates the loss coefficient. Indicates the penalty coefficient. express The actual temperature rise slope at any given time.

[0198] Energy consumption cost Direct economic pursuit. To implement time-varying electricity pricing, the system should be guided to increase heat dissipation during off-peak hours and decrease it during peak hours. For example, if electricity prices are low at night, heat dissipation can be initiated earlier or increased; if prices are high at midday, the heat dissipation load should be reduced as much as possible.

[0199] Equipment depreciation cost Characterizes the economic efficiency of equipment lifespan. The square term penalizes drastic changes in load. Prevents frequent start-stop cycles that could shorten equipment lifespan.

[0200] Penalties for exceeding the limit This reflects a trade-off between safety and economy. Hard constraints are softened into penalties; specifically, occasional, minor exceedances are allowed (which are difficult to completely avoid in real-world control). However, exceedances are penalized quadratically: small exceedances have low costs, while large exceedances have dramatically increased costs.

[0201] The above cost function, by adjusting , These parameters allow us to find the optimal balance point in the three-dimensional space of economy, equipment lifespan, and safety.

[0202] S246, Solve the optimization model to obtain the enhanced heat dissipation load.

[0203] S2461, time window for enhanced heat dissipation Discretization is performed to obtain multiple control time periods;

[0204] Transform continuous-time optimization into a finite-dimensional mathematical programming problem. Control period. Discretized into N time periods, within each period It is approximately a constant.

[0205] S2462, the optimization model is solved using a solver to obtain the enhanced heat dissipation load for multiple control time periods.

[0206] In this application, quadratic programming (QP) or nonlinear programming (NLP) is used, and the interior point method, sequential quadratic programming, etc. are used to solve the above optimization model to obtain the enhanced heat dissipation load for each control time period.

[0207] This application presents a heat dissipation control method for busbar trunking, achieving a fundamental shift from passive response to proactive prediction. Its core value lies in the synergistic optimization of precise prevention and energy conservation. The system analyzes historical data to construct a model that accurately reflects the temperature variation patterns of busbar trunking under different operating conditions, extracting windows with significant temperature rise trends. Based on the principle of thermal inertia, it calculates the optimal early intervention point for enhanced heat dissipation, initiating precise control before the temperature rise begins. Finally, through optimized algorithms, the heat dissipation intensity is dynamically adjusted, ensuring that the temperature rise slope is effectively suppressed below a safe threshold while comprehensively considering real-time electricity prices and equipment losses to minimize heat dissipation costs. This entire solution upgrades traditional threshold-based, extensive heat dissipation to a prediction-based, refined, and preventative energy management system, significantly improving energy economy while ensuring the long-term safe and stable operation of equipment.

[0208] like Figure 4 As shown, this application also provides a heat dissipation control system for busbar trunking, including:

[0209] The acquisition module is used to acquire temperature data samples of multiple historical periods of the bus trunking, wherein the temperature data samples contain temperature values ​​at multiple sampling time points within one period;

[0210] The pattern analysis module is used to perform pattern analysis on the temperature data sample to obtain a temperature change model, wherein the temperature change model includes a sub-model that represents the temperature values ​​corresponding to multiple time points within a variety of typical scenario cycles.

[0211] The temperature rise window extraction module is used to extract the temperature rise window and the temperature rise slope of multiple temperature sampling points in multiple typical scenario cycles from the temperature change model, and to filter out the target temperature rise window and the temperature sampling point corresponding to the target temperature rise window whose temperature rise slope is greater than or equal to a preset slope threshold.

[0212] A heat dissipation control module is used to set an enhanced heat dissipation time point and an enhanced heat dissipation load based on the target temperature rise window; and to perform heat dissipation control on the active heat dissipation device based on the enhanced heat dissipation time point and the enhanced heat dissipation load, wherein the enhanced heat dissipation time point is before the target temperature rise window, and the active heat dissipation device is an air-cooled device or a water-cooled device.

[0213] This application presents a heat dissipation control system for busbar trunking, achieving a fundamental shift from passive response to proactive prediction. Its core value lies in the synergistic optimization of precise prevention and energy conservation. The system analyzes historical data to construct a model that accurately reflects the temperature variation patterns of busbar trunking under different operating conditions, extracting windows with significant temperature rise trends. Based on the principle of thermal inertia, it calculates the optimal early intervention point for enhanced heat dissipation, initiating precise control before the temperature rise begins. Finally, through optimized algorithms, it dynamically adjusts the heat dissipation intensity, ensuring that the temperature rise slope is effectively suppressed below a safe threshold while comprehensively considering real-time electricity prices and equipment losses to minimize heat dissipation costs. This entire solution upgrades traditional threshold-based, extensive heat dissipation to a refined, preventative energy management system based on prediction, significantly improving energy economy while ensuring the long-term safe and stable operation of equipment.

[0214] This embodiment also provides an electronic terminal, including: a processor and a memory;

[0215] The memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to cause the terminal to perform any of the methods in this embodiment.

[0216] As will be understood by those skilled in the art, the computer-readable storage medium described in this embodiment allows for the implementation of all or part of the steps in the above method embodiments by computer program-related hardware. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0217] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver, and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication between them. The memory is used to store computer programs, the communication interface is used to perform communication, and the processor and the transceiver are used to run the computer programs, so that the electronic terminal performs the steps of the above method.

[0218] In this embodiment, the memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.

[0219] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0220] In the above embodiments, although the present application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. The embodiments of the present application are intended to cover all such substitutions, modifications, and variations falling within the broad scope of the appended claims.

[0221] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A heat dissipation control method for busbar trunking, characterized in that, Including the following steps: Acquire temperature data samples from multiple historical periods of the busbar trunking, wherein the temperature data samples contain temperature values ​​at multiple sampling time points within one period; The temperature data samples are analyzed for regularity to obtain a temperature change model, which includes a sub-model that represents the temperature values ​​at multiple time points within a variety of typical scenario cycles. The temperature rise windows and temperature rise slopes of multiple temperature sampling points in various typical scenario cycles are extracted from the temperature change model, and target temperature rise windows and temperature sampling points corresponding to the target temperature rise windows are selected with temperature rise slopes greater than or equal to a preset slope threshold. Determine the target temperature rise window corresponding to the current control cycle, set the enhanced heat dissipation time point and enhanced heat dissipation load based on the target temperature rise window, and perform heat dissipation control on the active heat dissipation device based on the enhanced heat dissipation time point and the enhanced heat dissipation load, wherein the enhanced heat dissipation time point is before the target temperature rise window, and the active heat dissipation device is an air-cooled device or a water-cooled device.

2. The heat dissipation control method for a busbar trunking according to claim 1, characterized in that, A regularity analysis was performed on the temperature data samples to obtain a temperature change model, including: Temperature data samples from multiple historical periods are preprocessed to obtain multiple preprocessed data samples. The preprocessing includes data completion, filtering, and normalization. Unsupervised clustering based on dynamic time warping is performed on the multiple preprocessed data samples to obtain multiple data clusters; and scenario labeling is performed on the multiple data clusters to obtain multiple scenario clusters. Calculate the average correlation coefficient of temperature data samples within each scenario cluster, and based on the average correlation coefficient, divide multiple scenario clusters into clusters with definite patterns, clusters with uncertain patterns, and clusters with no patterns. For the clusters where the pattern is determined, all preprocessed data samples within the cluster are aligned, and the average temperature value at each time point is calculated. A mean model is also constructed based on the average temperature values ​​at multiple time points within the period. For clusters with uncertain patterns, the preprocessed data samples are fitted with a Gaussian process model to obtain a Gaussian process model, wherein the Gaussian process model characterizes the predicted temperature values ​​at multiple time points and the confidence intervals of the predicted temperature values. A temperature change model is constructed based on the mean model and the Gaussian process model.

3. The heat dissipation control method for a busbar trunking according to claim 1, characterized in that, Calculate the average correlation coefficient of temperature data samples within each scenario cluster, and based on the average correlation coefficient, divide multiple scenario clusters into clusters with definite patterns, clusters with uncertain patterns, and clusters with no patterns, including: Calculate the correlation between any two temperature data samples within each scenario cluster. The relevance The mathematical expression is: In the formula, This represents one of the temperature data samples. This represents another temperature data sample. Representing temperature data samples The first in One value, Representing temperature data samples The average value in Representing temperature data samples The first in One value, Representing temperature data samples The average value in This indicates the number of time points in the temperature data sample; Calculate all relevance The average value is used to obtain the average correlation coefficient; Clusters of scenarios with an average correlation coefficient greater than the upper limit of the preset screening range are classified as clusters with a known pattern; clusters of scenarios with an average correlation coefficient falling within the preset screening range are classified as clusters with an uncertain pattern; and clusters of scenarios with an average correlation coefficient less than the lower limit of the preset screening range are classified as clusters with no pattern.

4. The heat dissipation control method for a busbar trunking according to claim 2, characterized in that, The temperature rise windows and their slopes are extracted from multiple temperature sampling points across various typical scenario cycles from the temperature change model, including: Data fragments of multiple size time windows are obtained from the temperature change model based on a pre-constructed multi-size sliding window. For a data segment of the mean model, a linear fit is performed on the data segment to obtain the segment slope; For a data segment of a Gaussian process model, calculate the upper bound slope and the lower bound slope of the confidence interval respectively; and calculate the average of the upper bound slope and the lower bound slope to obtain the segment slope. The slope of the segment of the mean model is compared with a preset initial slope threshold, and the lower bound slope of the Gaussian process model is compared with a preset initial slope threshold; and data segments with a slope greater than or equal to the initial slope threshold are selected as candidate data segments. The time window of the candidate data segment is used as the candidate window, and consecutive candidate windows of the same scale are merged to obtain a merged window of multiple sizes. By merging windows of different sizes, a temperature rise window is obtained; Linear fitting is performed on the data segments within the temperature rise window to obtain the temperature rise slope.

5. The heat dissipation control method for a busbar trunking according to claim 4, characterized in that, By merging windows of different sizes, a temperature rise window is obtained, including: Calculate the overlap of merged windows of different sizes The degree of overlap The mathematical expression is: In the formula, Indicates the first A merged window, Indicates the first A merged window, Display window The start time, Display window The start time, Display window End time, Display window End time; The overlap is compared with a preset overlap threshold. If the overlap is greater than or equal to the preset overlap threshold, the merged windows of different sizes are merged to obtain a temperature rise window; otherwise, the merged windows of different sizes are removed.

6. The heat dissipation control method for a busbar trunking according to claim 1, characterized in that, Based on the target temperature rise window, the enhanced heat dissipation time point and enhanced heat dissipation load are set, including: Extract the start time of the target temperature rise window And based on the said start time Calculate the time point for enhanced heat dissipation The enhanced heat dissipation time point The mathematical expression is: In the formula, This indicates the response delay of the active cooling device. This represents the predicted temperature rise slope of the target temperature rise window. Indicates the basic heat dissipation balance slope. For the target temperature rise slope, The system thermal time constant characterizes the system's response speed to temperature changes; Based on the enhanced heat dissipation time point An optimization model is constructed, wherein the objective of the optimization model is to minimize energy consumption, and the optimization model includes temperature constraints, temperature rise rate constraints, and physical constraints. Solving the optimization model yields the enhanced heat dissipation load.

7. A heat dissipation control method for a busbar trunking according to claim 6, characterized in that, Based on the enhanced heat dissipation time point Constructing an optimization model includes: Construct constraints, wherein the constraints include temperature constraints, temperature rise rate constraints, and physical constraints; The mathematical expression for the temperature constraint is: In the formula, Indicates that the busbar trunking is in Temperature at any moment This is the upper limit of the temperature. Indicates the end time of the target temperature rise window; The mathematical expression for the temperature rise rate constraint is: The mathematical expression for the physical constraint is: In the formula, Indicates the load of the active cooling equipment. This indicates the upper limit of the load change rate; Construct an objective function with the goal of minimizing energy consumption, wherein the mathematical expression of the objective function is: In the formula, Represents the total cost. Indicates energy consumption cost, This indicates the cost of equipment depreciation. This indicates a penalty for exceeding the limit. The rated power of the active cooling device. express Electricity price at any time Indicates the loss coefficient. Indicates the penalty coefficient. express The actual temperature rise slope at any given time.

8. A heat dissipation control method for a busbar trunking according to claim 7, characterized in that, Solving the optimization model yields the enhanced heat dissipation load, including: Time window for enhanced heat dissipation Discretization is performed to obtain multiple control time periods; The optimization model is solved using a solver to obtain the enhanced heat dissipation load for multiple control time periods.

9. A heat dissipation control method for a busbar trunking according to claim 1, characterized in that, Determine the target temperature rise window corresponding to the current control cycle, including: The temperature change model is divided into multiple time periods for labeling various typical scenarios. The label of the target scenario and the corresponding target sub-model are determined based on the current time and the time period of labels of multiple typical scenarios; Extract the target temperature rise window from the target sub-model.

10. A heat dissipation control system for busbar trunking, characterized in that, include: The acquisition module is used to acquire temperature data samples of the busbar trunking for multiple historical periods, wherein the temperature data samples contain temperature values ​​at multiple sampling time points within one period. The pattern analysis module is used to perform pattern analysis on the temperature data sample to obtain a temperature change model, wherein the temperature change model includes a sub-model that represents the temperature values ​​corresponding to multiple time points within a variety of typical scenario cycles. The temperature rise window extraction module is used to extract the temperature rise window and the temperature rise slope of multiple temperature sampling points in multiple typical scenario cycles from the temperature change model, and to filter out the target temperature rise window and the temperature sampling point corresponding to the target temperature rise window whose temperature rise slope is greater than or equal to a preset slope threshold. A heat dissipation control module is used to set an enhanced heat dissipation time point and an enhanced heat dissipation load based on the target temperature rise window; and to perform heat dissipation control on the active heat dissipation device based on the enhanced heat dissipation time point and the enhanced heat dissipation load, wherein the enhanced heat dissipation time point is before the target temperature rise window, and the active heat dissipation device is an air-cooled device or a water-cooled device.

Images