A method and device for intelligent temperature control of a computer network control cabinet

Abstract

This invention discloses an intelligent temperature control method and device for a computer network control cabinet, relating to the field of equipment status monitoring technology. The method includes: obtaining heat source status indicators; configuring work tasks for a future preset time window via a user terminal, processing the work tasks through a control cabinet twin model, and obtaining characteristic values ​​of the heat source status indicators; obtaining the layout topology of the control cabinet's heat source equipment, the layout topology of the control cabinet's heat dissipation equipment, the cabinet's external temperature, and the cabinet's external humidity; using multiple parameters as constraints, retrieving samples of control cabinets in the same mode when the heat dissipation equipment is not activated, performing mode analysis, and obtaining a predicted value for the cabinet's internal temperature; performing heat dissipation control optimization based on the predicted internal temperature, external temperature, and external humidity to obtain pre-configured heat dissipation parameters; and performing temperature control based on the pre-configured heat dissipation parameters. This invention solves the technical problems of severe lag and poor control results in the temperature control of computer network control cabinets in existing technologies.

Description

Technical Field

[0001] This invention relates to the field of equipment condition monitoring technology, specifically to an intelligent temperature control method and device for a computer network control cabinet. Background Technology

[0002] Computer network control cabinets, as key equipment in industrial automation, integrate PLC control and network communication functions, undertaking important control and data interaction tasks. During actual operation, the electronic components inside the control cabinet generate a significant amount of heat. Traditional temperature control methods generally employ a reactive response mode, i.e., monitoring the cabinet temperature through temperature sensors and only initiating cooling strategies when the temperature exceeds a preset threshold. This reactive control strategy has significant drawbacks. From temperature exceeding the limit to the cooling device's response and then to temperature reduction, a certain amount of time is required. During this period, the high-temperature environment may have already caused performance damage or lifespan reduction to precision electronic components such as the PLC. Furthermore, because it fails to consider future changes in the equipment's workload, it cannot predictively adjust the cooling strategy, resulting in severe temperature control lag. Summary of the Invention

[0003] This application provides an intelligent temperature control method and device for computer network control cabinets, which addresses the technical problems of severe lag and poor control results in the temperature control of computer network control cabinets in the prior art.

[0004] In view of the above problems, this application provides an intelligent temperature control method and device for a computer network control cabinet.

[0005] In a first aspect, this application provides an intelligent temperature control method for a computer network control cabinet, the method comprising:

[0006] Obtain heat source status indicators, wherein the heat source status indicators characterize the working status indicators of non-heat dissipation equipment that affect the temperature inside the cabinet;

[0007] Through the user terminal, the work tasks for a future preset time window are configured, and the work tasks are processed through the control cabinet twin model to obtain the characteristic values ​​of the heat source status index.

[0008] Obtain the layout topology of the control cabinet's heat source equipment, the layout topology of the control cabinet's heat dissipation equipment, the cabinet's external temperature, and the cabinet's external humidity;

[0009] Using the characteristic values ​​of the heat source status index, the layout topology of the heat source equipment in the control cabinet, the layout topology of the heat dissipation equipment in the control cabinet, the outside temperature and the outside humidity of the cabinet as constraints, samples of the same mode control cabinet when the heat dissipation equipment is not started are retrieved, and mode analysis is performed to obtain the predicted value of the cabinet temperature.

[0010] Based on the predicted temperature inside the cabinet, the temperature outside the cabinet, and the humidity outside the cabinet, heat dissipation control optimization is performed to obtain pre-configured heat dissipation parameters;

[0011] When the work task is executed and the external temperature and humidity of the cabinet do not change, temperature regulation is performed based on the pre-configured heat dissipation parameters.

[0012] Secondly, this application provides an intelligent temperature control device for a computer network control cabinet, comprising:

[0013] The status acquisition module is used to obtain heat source status indicators, wherein the heat source status indicators characterize the working status indicators of non-heat dissipation equipment that affect the temperature inside the cabinet.

[0014] The heat source index acquisition module is used to configure the work tasks for a future preset time window through the user terminal, process the work tasks through the control cabinet twin model, and obtain the characteristic values ​​of the heat source status index.

[0015] The control cabinet parameter acquisition module is used to obtain the layout topology of the control cabinet heat source equipment, the layout topology of the control cabinet heat dissipation equipment, the cabinet outside temperature, and the cabinet outside humidity.

[0016] The temperature prediction module is used to retrieve samples of the same mode control cabinet when the heat source status index feature value, the layout topology of the heat source equipment of the control cabinet, the layout topology of the heat dissipation equipment of the control cabinet, the outside temperature of the cabinet and the outside humidity of the cabinet as constraints, perform mode analysis, and obtain the predicted value of the temperature inside the cabinet.

[0017] The heat dissipation control optimization module is used to perform heat dissipation control optimization based on the predicted temperature inside the cabinet, the temperature outside the cabinet, and the humidity outside the cabinet to obtain pre-configured heat dissipation parameters.

[0018] The temperature control module is used to perform temperature control based on the pre-configured heat dissipation parameters when the work task is executed and the external temperature and humidity of the cabinet do not change.

[0019] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0020] This application proposes an intelligent temperature control method and device for a computer network control cabinet. By constructing a digital twin model of the control cabinet to predict the characteristic values ​​of the heat source state, and combining multi-constraint condition retrieval of same-modal samples for mode analysis, the predictability and control effect of temperature control are significantly improved. Compared with traditional methods, the technical solution provided in this application significantly overcomes the hysteresis defect of the post-response mode, realizing the transformation from passive reaction to active prediction, and achieving the technical effects of implementing preventive control before the temperature rises, reducing the response delay of heat dissipation equipment, and reducing the risk of equipment overheating. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating an intelligent temperature control method for a computer network control cabinet, as provided in an embodiment of this application.

[0023] Figure 2 This is a schematic diagram of the structure of an intelligent temperature control device for a computer network control cabinet provided in an embodiment of this application.

[0024] The components represented by each number in the attached diagram are explained below:

[0025] Status acquisition module 100, heat source index acquisition module 200, control cabinet parameter acquisition module 300, temperature prediction module 400, heat dissipation control optimization module 500, and temperature regulation module 600. Detailed Implementation

[0026] This application provides an intelligent temperature control method and device for computer network control cabinets, which addresses the technical problems of severe lag and poor control results in the temperature control of computer network control cabinets in the prior art.

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0028] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0029] Example 1, as Figure 1 As shown, this application provides an intelligent temperature control method for a computer network control cabinet, wherein the method includes:

[0030] S10: Obtain heat source status indicators, wherein the heat source status indicators characterize the working status indicators of non-heat dissipation equipment that affect the temperature inside the cabinet.

[0031] In the process of temperature control in computer network control cabinets, existing methods often fail to accurately identify which equipment operating status parameters truly have a significant impact on the cabinet temperature. Traditional methods typically treat all equipment status equally or rely solely on experience to select monitoring indicators, resulting in a lack of precise data foundation for temperature control.

[0032] Step S10 in the method provided in this application embodiment includes:

[0033] Obtain the set of operating status indicators for non-heat-dissipating equipment inside the cabinet;

[0034] Extract the first working status indicator from the set of working status indicators;

[0035] Based on the first working status index, a correlation analysis is performed on the temperature inside the cabinet to obtain the correlation coefficient of the first working status index.

[0036] Specifically, based on the first working status index, a correlation analysis is performed on the cabinet temperature to obtain the correlation coefficient of the first working status index, including:

[0037] When the first working status indicator is a quantitative indicator, with the first working status indicator as the only independent variable and the cabinet temperature as the dependent variable, Pearson correlation analysis is performed to obtain the correlation coefficient of the first working status indicator.

[0038] When the first working status indicator is a type indicator, the first working status indicator is switched to the only control fluctuation quantity. Several cabinet temperature fluctuation quantities are collected, the average value is calculated, and the correlation coefficient of the first working status indicator is obtained.

[0039] When the correlation coefficient of the first working status indicator meets the first preset correlation coefficient range, the first working status indicator is added to the heat source status indicator, wherein the first preset correlation coefficient ranges of the type indicator and the quantitative indicator are different.

[0040] After traversing the set of working status indicators, the heat source status indicators are obtained.

[0041] In this embodiment, a set of operating status indicators for non-heat-dissipating devices within a computer network control cabinet is obtained. The computer network control cabinet contains various devices, such as CPUs, memory, and switches. The set of operating status indicators for these non-heat-dissipating devices includes indicators such as CPU utilization, CPU power consumption, memory utilization, and network throughput.

[0042] Extract the first working status indicator from the set of working status indicators and use it as the current processing object. For example, the first working status indicator is CPU utilization.

[0043] Based on the first operating status indicator, a correlation analysis is performed on the cabinet temperature to obtain the correlation coefficient of the first operating status indicator. For example, when the first operating status indicator is a quantitative indicator, with CPU utilization as the sole independent variable and cabinet temperature as the dependent variable, a Pearson correlation analysis is performed on CPU utilization and cabinet temperature. The Pearson correlation coefficient is calculated by obtaining CPU utilization sequences and corresponding temperature sequences for the same time period to obtain the correlation coefficient of the first operating status indicator. The value range of the correlation coefficient of the first operating status indicator is [-1, 1], where a positive value indicates a positive correlation between the two indicators, and a negative value indicates a negative correlation. The larger the absolute value of the correlation coefficient of the first operating status indicator, the stronger the correlation between the first operating status indicator and temperature.

[0044] When the first working status indicator is a type indicator, such as the first working status indicator being the equipment working mode, the switching of the first working status indicator is used as the sole control fluctuation quantity. When the equipment switches to a certain mode, the temperature change within a certain period of time, such as 30 minutes, is selected, and after multiple recordings, the average temperature change is calculated as the correlation coefficient.

[0045] When the correlation coefficient of the first working state indicator meets the first preset correlation coefficient range, the first working state indicator is added to the heat source state indicator. The first preset correlation coefficient ranges for the type indicator and the quantitative indicator are different. For example, the first preset correlation coefficient range for the quantitative indicator can be set to an absolute value greater than or equal to 0.6, and the first preset correlation coefficient range for the type indicator can be set to greater than or equal to 3℃ and less than or equal to 15℃.

[0046] After traversing the set of working status indicators, the heat source status indicators are obtained.

[0047] By systematically analyzing the correlation between the working status indicators of non-heat-dissipating equipment inside the cabinet and the temperature inside the cabinet, we can scientifically screen out the heat source status indicators that truly have a significant impact on the temperature, so as to establish an accurate and reliable heat source assessment index system and provide accurate data input for subsequent temperature prediction.

[0048] S20: Configure the work tasks for a future preset time window through the user terminal, process the work tasks through the control cabinet twin model, and obtain the characteristic values ​​of the heat source status index.

[0049] Existing temperature control methods lack the ability to predict future workloads, failing to anticipate the types and amounts of tasks the equipment will perform, resulting in control strategies lagging behind actual heat load changes. This passive response mode makes it difficult for the cooling system to prepare before temperatures rise, leading not only to poor control but also increasing the risk of equipment overheating.

[0050] Step S20 in the method provided in this application embodiment includes:

[0051] Obtain the historical work log of the computer network control cabinet, wherein the historical work log includes an array of work type records, an array of work quantity records, and labels that identify the record values ​​of work status indicators;

[0052] The work type record array and the workload record array are placed in the input node of a fully connected neural network, and the labels that identify the work status index record values ​​are placed in the output node of the fully connected neural network to train the control cabinet twin model.

[0053] From the work tasks, extract the work type array and the workload array, and perform analysis through the control cabinet twin model to obtain the predicted values ​​of the work status indicators;

[0054] Based on the heat source status index, feature values ​​of the heat source status index are extracted from the predicted values ​​of the working status index.

[0055] In this embodiment, the historical work log of the computer network control cabinet is obtained. The historical work log includes an array of work type records, an array of workload records, and tags that identify the recorded values ​​of work status indicators. The array of work type records contains the types of work performed by the computer network control cabinet, such as "data backup" and "network monitoring." The array of workload records contains the workload corresponding to the work type, such as the amount of data backed up and the duration of network monitoring.

[0056] The work type record array and the work quantity record array are placed in the input nodes of a fully connected neural network, and the labels identifying the work status index record values ​​are placed in the output nodes of the fully connected neural network to train the control cabinet twin model. Specifically, the number of input layer nodes is set to the total dimension of the work type and work quantity features, for example, 20 nodes based on the work type and work quantity features. The hidden layer contains 64 nodes, using the ReLU activation function, and the output layer is used to output the predicted work status index. A supervised training method is used, with the work type index record values ​​and the work status index record values ​​as input and the work status index as the target output. The Adam optimizer and mean squared error loss function are used for training until the prediction error of the control cabinet twin model on the validation set tends to stabilize. For example, the control cabinet twin model is considered to be successfully trained when the accuracy of the predicted work status index is all above 90%.

[0057] From the work tasks, extract the work type array and the workload array, and perform analysis through the control cabinet twin model to obtain the predicted values ​​of the work status indicators.

[0058] Based on the heat source status index, feature values ​​of the heat source status index are extracted from the predicted values ​​of the working status index.

[0059] By transforming future work tasks into heat source state characteristics through digital twin models, an accurate mapping from work plans to heat load prediction is achieved, and the ability to quantitatively predict future thermal states is provided. Early perception of impending heat load changes can create conditions for implementing preventative temperature control.

[0060] S30: Obtain the layout topology of the control cabinet heat source equipment, the layout topology of the control cabinet heat dissipation equipment, the cabinet outside temperature, and the cabinet outside humidity.

[0061] Traditional temperature control methods often overlook the complex impact of equipment layout topology and external environmental factors on heat dissipation, relying solely on simple feedback control based on the cabinet's internal temperature. This simplistic approach makes the control strategy unable to adapt to the differences in heat conduction characteristics under different equipment layouts, nor can it effectively cope with the impact of changes in the external environment.

[0062] In this embodiment, the layout topology of the heat source devices and the heat dissipation devices of the control cabinet are obtained based on the control cabinet design information. The layout topology refers to the arrangement information of the heat source devices and heat dissipation devices within the control cabinet design layout. This layout information affects the heat source distribution and airflow within the cabinet, thereby further influencing the internal temperature. Temperature and humidity sensors are used to obtain the internal temperature and external humidity of the control cabinet, enabling subsequent temperature prediction and optimization to fully consider the equipment layout characteristics and the influence of the external environment, significantly improving the adaptability and accuracy of the control strategy.

[0063] S40: Using the characteristic values ​​of the heat source status index, the layout topology of the heat source equipment in the control cabinet, the layout topology of the heat dissipation equipment in the control cabinet, the outside temperature of the cabinet, and the outside humidity of the cabinet as constraints, retrieve the same mode control cabinet samples when the heat dissipation equipment is not started, perform mode analysis, and obtain the predicted value of the cabinet temperature.

[0064] Existing methods mostly rely on simple empirical judgments for regulation, making it difficult to extract temperature changes under the influence of complex, multi-factor factors for temperature prediction. This inaccuracy in prediction directly leads to a lack of reliable basis for the formulation of subsequent regulation strategies, affecting the overall regulation effect.

[0065] Step S40 in the method provided in this application embodiment includes:

[0066] The system records the sample heat source status index, sample heat source equipment layout topology, sample heat dissipation equipment layout topology, sample cabinet external temperature, sample cabinet external humidity and cabinet internal temperature when the heat dissipation equipment is not started.

[0067] When the deviation between the recorded value of the sample heat source status index and the characteristic value of the heat source status index is less than or equal to the corresponding index deviation threshold, and the similarity between the sample heat source equipment layout topology and the control cabinet heat source equipment layout topology is greater than or equal to the first topology similarity threshold, and the similarity between the sample heat dissipation equipment layout topology and the control cabinet heat dissipation equipment layout topology is greater than or equal to the second topology similarity threshold, and the deviation between the sample cabinet outside temperature and the cabinet outside temperature is less than or equal to the cabinet outside temperature deviation threshold, and the deviation between the sample cabinet outside humidity and the cabinet outside humidity is less than or equal to the cabinet outside humidity deviation threshold, the recorded value of the cabinet inside temperature is added to the same modal control cabinet sample.

[0068] When the number of samples of the same modal control cabinet is greater than or equal to the preset fitting number, mode analysis is performed to obtain the predicted value of the temperature inside the cabinet.

[0069] When the number of samples from the same modal control cabinet is greater than or equal to the preset fitting number, a mode analysis is performed to obtain the predicted temperature value inside the cabinet, including:

[0070] Based on the temperature deviation threshold, cluster analysis is performed on the set of cabinet internal temperature records of the same mode control cabinet sample to obtain multiple clusters of cabinet internal temperature records.

[0071] Delete the temperature record values ​​of clusters whose number of temperature record values ​​is less than or equal to the cluster number threshold from the set of temperature record values ​​inside the cabinet, and obtain the selected set of temperature record values ​​inside the cabinet.

[0072] The average value of the selected set of temperature records inside the cabinet is calculated to obtain the predicted temperature value inside the cabinet.

[0073] In this embodiment of the application, the same method as steps S10 to S30 is used to obtain the recorded values ​​of the sample heat source status index, the sample heat source equipment layout topology, the sample heat dissipation equipment layout topology, the sample cabinet external temperature, the sample cabinet external humidity, and the cabinet internal temperature of the sample control cabinet when the heat dissipation equipment is not started.

[0074] When the deviations between the recorded values ​​of the sample heat source status indicators and the characteristic values ​​of the heat source status indicators are all less than or equal to the corresponding indicator deviation thresholds, and the similarity between the sample heat source equipment layout topology and the control cabinet heat source equipment layout topology is greater than or equal to the first topology similarity threshold, and the similarity between the sample heat dissipation equipment layout topology and the control cabinet heat dissipation equipment layout topology is greater than or equal to the second topology similarity threshold, and the deviation between the sample cabinet exterior temperature and the cabinet exterior temperature is less than or equal to the cabinet exterior temperature deviation threshold, and the deviation between the sample cabinet exterior humidity and the cabinet exterior humidity is less than or equal to the cabinet exterior humidity deviation threshold, it indicates that the operating conditions, layout topology, and exterior conditions of the sample control cabinet are similar to those of the current control cabinet. The status of the sample control cabinet has high reference value for optimizing the current control cabinet, and the recorded values ​​of the cabinet interior temperature are added to the sample of the same modal control cabinet.

[0075] Among them, the corresponding index deviation threshold refers to the deviation threshold set in advance for the characteristic values ​​of multiple heat source status indicators. For example, if the deviation threshold of CPU utilization is set to 5%, and the deviation value is less than or equal to the deviation threshold, it means that the sample heat source status indicators are closer to the current computer network control cabinet, and the status of the sample control cabinet has greater reference value.

[0076] The first topological similarity threshold and the second topological similarity threshold are thresholds set to characterize the topological similarity between the sample control cabinet and the current control cabinet layout. The topological similarity is calculated using a graph similarity algorithm. The first topological similarity threshold is exemplarily set to 85%, and the second topological similarity threshold is exemplarily set to 90%.

[0077] Among them, the external temperature deviation value and the external humidity deviation value are thresholds set to characterize the deviation between the sample control cabinet and the current control cabinet environment. For example, the external temperature deviation threshold is set to 1℃ and the external humidity deviation threshold is set to 2%. The external temperature and external humidity are collected by temperature sensors.

[0078] Information on multiple control cabinets is obtained and filtered for addition. When the number of control cabinet samples of the same mode is greater than or equal to the preset fitting number, mode analysis is performed to obtain the predicted value of the temperature inside the cabinet.

[0079] Specifically, based on a temperature deviation threshold, cluster analysis is performed on the set of internal temperature records for control cabinet samples of the same modality to obtain multi-cluster internal temperature records. For example, a cluster analysis algorithm such as the K-means algorithm is used to perform cluster analysis on the set of internal temperature records for control cabinet samples of the same modality. A temperature deviation threshold is set to remove temperature data that deviates significantly; for example, the temperature deviation threshold can be set to 12℃ to remove outliers that deviate more than 12℃ from the cluster center, thus reducing the impact of occasional extreme data.

[0080] From the set of cabinet temperature records, delete the temperature records of clusters whose number of temperature records is less than or equal to a cluster quantity threshold, thus obtaining a selected set of cabinet temperature records. For example, the cluster quantity threshold can be set to 20 to remove sample data with too small a data volume, thereby further filtering to make the cabinet temperature data in the set of cabinet temperature records more representative.

[0081] The mean value of the selected set of cabinet internal temperature records is calculated, and all cabinet internal temperature records in the selected set are used as the predicted cabinet internal temperature value.

[0082] By retrieving historical samples with the same modalities under multiple constraints and performing mode analysis, accurate temperature prediction based on big data analysis was achieved. High-precision predicted cabinet temperatures were obtained by fully utilizing similar operating condition information from historical data, providing solid data support for subsequent optimization of control strategies.

[0083] S50: Based on the predicted temperature inside the cabinet, the temperature outside the cabinet, and the humidity outside the cabinet, perform heat dissipation control optimization to obtain pre-configured heat dissipation parameters.

[0084] Traditional temperature control methods often employ fixed control parameters when formulating heat dissipation strategies, failing to dynamically optimize these parameters based on factors such as temperature and environmental conditions. This rigid control approach can lead to either excessive or insufficient heat dissipation.

[0085] Step S50 in the method provided in this application embodiment includes:

[0086] Obtain the target temperature inside the cabinet;

[0087] Using the predicted temperature inside the cabinet, the temperature outside the cabinet, the humidity outside the cabinet, the target temperature inside the cabinet, the layout topology of the heat source equipment in the control cabinet, and the layout topology of the heat dissipation equipment in the control cabinet as constraints, retrieve the set of records of heat dissipation control parameters;

[0088] Outlier analysis is performed on the set of records of heat dissipation control parameters to obtain the set of outlier eigenvalues;

[0089] Extract the heat dissipation control parameter record array corresponding to the minimum value of the outlier eigenvalue in the outlier eigenvalue set, and set it as the pre-configured heat dissipation parameter.

[0090] In this embodiment of the application, the target temperature inside the cabinet is obtained. For example, according to the technical manual, the recommended operating temperature of the current control cabinet is 25°C, so the target temperature inside the cabinet can be set to 25°C.

[0091] Constrained by the predicted internal temperature, external temperature, external humidity, target internal temperature, and the layout topology of the control cabinet's heat source equipment and heat dissipation equipment, a set of heat dissipation control parameter record arrays is retrieved. For example, multiple heat dissipation control parameter record arrays are retrieved if the internal temperature is within ±1℃ of the predicted internal temperature, the external temperature is within ±1℃ of the current external temperature, the external humidity is within ±1%, the similarity between the control cabinet's heat source equipment layout topology and the current control cabinet equipment layout topology is greater than or equal to 85%, and the similarity between the control cabinet's heat dissipation equipment layout topology and the current control cabinet's heat dissipation equipment layout topology is greater than or equal to 90%, and these arrays are added to the heat dissipation control parameter record array set.

[0092] Outlier analysis is performed on the set of records of heat dissipation control parameters, for example, by using a local outlier detection algorithm to obtain a set of outlier feature values.

[0093] An array of heat dissipation control parameters corresponding to the minimum outlier eigenvalues ​​is selected as the pre-configured heat dissipation parameters. The minimum outlier eigenvalue indicates that the corresponding heat dissipation control parameters are more stable and reliable, and are more likely to have good optimization effects.

[0094] By performing heat dissipation control optimization, intelligent optimization of control parameters is achieved, obtaining the optimal heat dissipation parameters for specific predicted temperatures and environmental conditions, ensuring that the control parameters can meet the cooling requirements without causing energy waste.

[0095] S60: When the work task is executed and the external temperature and humidity of the cabinet do not change, temperature regulation is performed based on the pre-configured heat dissipation parameters.

[0096] Step S60 in the method provided in this application embodiment further includes:

[0097] When the work task is not performed, or when either the external temperature or the external humidity changes, routine temperature control is performed.

[0098] In this embodiment of the application, when the work task is executed and the external temperature and humidity of the cabinet do not change, temperature regulation is performed based on the pre-configured heat dissipation parameters.

[0099] When a task is not performed, or when either the external temperature or humidity changes, the normal temperature control is applied.

[0100] By applying pre-configured heat dissipation parameters at appropriate times, a complete closed loop from prediction to execution is achieved. When conditions are met, the optimal control parameters are directly applied, which significantly improves the foresight and effectiveness of the control response, realizes preventive temperature control, and maximizes the value of early prediction and optimization.

[0101] Example 2, as Figure 2 As shown, based on the same inventive concept as the intelligent temperature control method for a computer network control cabinet provided in Embodiment 1, this embodiment of the invention also provides an intelligent temperature control device for a computer network control cabinet, comprising:

[0102] The status acquisition module 100 is used to acquire heat source status indicators, wherein the heat source status indicators characterize the working status indicators of non-heat dissipation equipment that affect the temperature inside the cabinet.

[0103] The heat source index acquisition module 200 is used to configure the work tasks for a future preset time window through the user terminal, process the work tasks through the control cabinet twin model, and obtain the characteristic values ​​of the heat source status index.

[0104] The control cabinet parameter acquisition module 300 is used to obtain the layout topology of the control cabinet heat source equipment, the layout topology of the control cabinet heat dissipation equipment, the cabinet outside temperature and the cabinet outside humidity.

[0105] The temperature prediction module 400 is used to retrieve samples of the same mode control cabinet when the heat dissipation equipment is not started, and perform mode analysis to obtain the predicted value of the cabinet temperature, based on the characteristic value of the heat source status index, the layout topology of the heat source equipment of the control cabinet, the layout topology of the heat dissipation equipment of the control cabinet, the outside temperature of the cabinet and the outside humidity of the cabinet.

[0106] The heat dissipation control optimization module 500 is used to perform heat dissipation control optimization based on the predicted value of the temperature inside the cabinet, the temperature outside the cabinet, and the humidity outside the cabinet to obtain pre-configured heat dissipation parameters;

[0107] The temperature control module 600 is used to perform temperature control based on the pre-configured heat dissipation parameters when the work task is executed and the external temperature and humidity of the cabinet do not change.

[0108] In one embodiment, the status acquisition module 100 is further configured to:

[0109] Obtain the set of operating status indicators for non-heat-dissipating equipment inside the cabinet;

[0110] Extract the first working status indicator from the set of working status indicators;

[0111] Based on the first working status index, a correlation analysis is performed on the temperature inside the cabinet to obtain the correlation coefficient of the first working status index.

[0112] Specifically, based on the first working status index, a correlation analysis is performed on the cabinet temperature to obtain the correlation coefficient of the first working status index, including:

[0113] When the first working status indicator is a quantitative indicator, with the first working status indicator as the only independent variable and the cabinet temperature as the dependent variable, Pearson correlation analysis is performed to obtain the correlation coefficient of the first working status indicator.

[0114] When the first working status indicator is a type indicator, the first working status indicator is switched to the only control fluctuation quantity. Several cabinet temperature fluctuation quantities are collected, the average value is calculated, and the correlation coefficient of the first working status indicator is obtained.

[0115] When the correlation coefficient of the first working status indicator meets the first preset correlation coefficient range, the first working status indicator is added to the heat source status indicator, wherein the first preset correlation coefficient ranges of the type indicator and the quantitative indicator are different.

[0116] After traversing the set of working status indicators, the heat source status indicators are obtained.

[0117] In one embodiment, the heat source index acquisition module 200 is further configured to:

[0118] Obtain the historical work log of the computer network control cabinet, wherein the historical work log includes an array of work type records, an array of work quantity records, and labels that identify the record values ​​of work status indicators;

[0119] The work type record array and the workload record array are placed in the input node of a fully connected neural network, and the labels that identify the work status index record values ​​are placed in the output node of the fully connected neural network to train the control cabinet twin model.

[0120] From the work tasks, extract the work type array and the workload array, and perform analysis through the control cabinet twin model to obtain the predicted values ​​of the work status indicators;

[0121] Based on the heat source status index, feature values ​​of the heat source status index are extracted from the predicted values ​​of the working status index.

[0122] In one embodiment, the temperature prediction module 400 is further configured to:

[0123] The system records the sample heat source status index, sample heat source equipment layout topology, sample heat dissipation equipment layout topology, sample cabinet external temperature, sample cabinet external humidity and cabinet internal temperature when the heat dissipation equipment is not started.

[0124] When the deviation between the recorded value of the sample heat source status index and the characteristic value of the heat source status index is less than or equal to the corresponding index deviation threshold, and the similarity between the sample heat source equipment layout topology and the control cabinet heat source equipment layout topology is greater than or equal to the first topology similarity threshold, and the similarity between the sample heat dissipation equipment layout topology and the control cabinet heat dissipation equipment layout topology is greater than or equal to the second topology similarity threshold, and the deviation between the sample cabinet outside temperature and the cabinet outside temperature is less than or equal to the cabinet outside temperature deviation threshold, and the deviation between the sample cabinet outside humidity and the cabinet outside humidity is less than or equal to the cabinet outside humidity deviation threshold, the recorded value of the cabinet inside temperature is added to the same modal control cabinet sample.

[0125] When the number of samples of the same modal control cabinet is greater than or equal to the preset fitting number, mode analysis is performed to obtain the predicted value of the temperature inside the cabinet.

[0126] When the number of samples from the same modal control cabinet is greater than or equal to the preset fitting number, a mode analysis is performed to obtain the predicted temperature value inside the cabinet, including:

[0127] Based on the temperature deviation threshold, cluster analysis is performed on the set of cabinet internal temperature records of the same mode control cabinet sample to obtain multiple clusters of cabinet internal temperature records.

[0128] Delete the temperature record values ​​of clusters whose number of temperature record values ​​is less than or equal to the cluster number threshold from the set of temperature record values ​​inside the cabinet, and obtain the selected set of temperature record values ​​inside the cabinet.

[0129] The average value of the selected set of temperature records inside the cabinet is calculated to obtain the predicted temperature value inside the cabinet.

[0130] In one embodiment, the heat dissipation control optimization module 500 is further configured to:

[0131] Obtain the target temperature inside the cabinet;

[0132] Using the predicted temperature inside the cabinet, the temperature outside the cabinet, the humidity outside the cabinet, the target temperature inside the cabinet, the layout topology of the heat source equipment in the control cabinet, and the layout topology of the heat dissipation equipment in the control cabinet as constraints, retrieve the set of records of heat dissipation control parameters;

[0133] Outlier analysis is performed on the set of records of heat dissipation control parameters to obtain the set of outlier eigenvalues;

[0134] Extract the heat dissipation control parameter record array corresponding to the minimum value of the outlier eigenvalue in the outlier eigenvalue set, and set it as the pre-configured heat dissipation parameter.

[0135] In one embodiment, the temperature control module 600 is further configured to:

[0136] When the work task is not performed, or when either the external temperature or the external humidity changes, routine temperature control is performed.

[0137] In summary, the embodiments of this application have at least the following technical effects:

[0138] This application proposes an intelligent temperature control method and device for a computer network control cabinet. By constructing a digital twin model of the control cabinet to predict the characteristic values ​​of heat source states, and combining multi-constraint conditions to retrieve same-modal samples for mode analysis, the predictability and control effect of temperature control are significantly improved. Specifically, by analyzing the correlation between the working state of the equipment and the temperature inside the cabinet, key heat source state indicators are accurately identified; then, the control cabinet twin model is used to process future work tasks and predict the characteristic values ​​of heat source states; based on this, using the characteristic values ​​of heat source states, equipment layout topology, and external environmental parameters as multiple constraints, the same-modal cases when the heat dissipation equipment is not started in historical samples are retrieved, and the accurate predicted value of the temperature inside the cabinet is obtained through mode analysis; finally, heat dissipation control optimization is performed based on the predicted temperature and environmental parameters to obtain the optimal pre-configured heat dissipation parameters, and control is implemented in advance during task execution. Compared with traditional methods, the technical solution provided in this application significantly overcomes the lag defect of the post-response mode, realizes the transformation from passive reaction to active prediction, and achieves the technical effects of implementing preventive control before the temperature rises, reducing the response delay of heat dissipation equipment, and reducing the risk of equipment overheating.

[0139] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0140] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0141] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for intelligent temperature control of a computer network control cabinet, characterized in that, include: Obtain heat source status indicators, wherein the heat source status indicators characterize the working status indicators of non-heat dissipation equipment that affect the temperature inside the cabinet; Through the user terminal, the work tasks for a future preset time window are configured, and the work tasks are processed through the control cabinet twin model to obtain the characteristic values ​​of the heat source status index. Obtain the layout topology of the control cabinet's heat source equipment, the layout topology of the control cabinet's heat dissipation equipment, the cabinet's external temperature, and the cabinet's external humidity; Using the characteristic values ​​of the heat source status index, the layout topology of the heat source equipment in the control cabinet, the layout topology of the heat dissipation equipment in the control cabinet, the outside temperature and the outside humidity of the cabinet as constraints, samples of the same mode control cabinet when the heat dissipation equipment is not started are retrieved, and mode analysis is performed to obtain the predicted value of the cabinet temperature. Based on the predicted internal temperature, external temperature, and external humidity, heat dissipation control optimization is performed to obtain pre-configured heat dissipation parameters, including: Obtain the target temperature inside the cabinet; Using the predicted temperature inside the cabinet, the temperature outside the cabinet, the humidity outside the cabinet, the target temperature inside the cabinet, the layout topology of the heat source equipment in the control cabinet, and the layout topology of the heat dissipation equipment in the control cabinet as constraints, retrieve the set of records of heat dissipation control parameters; Outlier analysis is performed on the set of records of heat dissipation control parameters to obtain the set of outlier eigenvalues; Extract the heat dissipation control parameter record array corresponding to the minimum value of the outlier eigenvalue in the outlier eigenvalue set, and set it as the pre-configured heat dissipation parameter; When the work task is executed and the external temperature and humidity of the cabinet do not change, temperature regulation is performed based on the pre-configured heat dissipation parameters.

2. The method as described in claim 1, characterized in that, Obtain heat source status indicators, including: Obtain the set of operating status indicators for non-heat-dissipating equipment inside the cabinet; Extract the first working status indicator from the set of working status indicators; Based on the first working status index, a correlation analysis is performed on the temperature inside the cabinet to obtain the correlation coefficient of the first working status index. When the correlation coefficient of the first working status indicator meets the first preset correlation coefficient range, the first working status indicator is added to the heat source status indicator, wherein the first preset correlation coefficient ranges of the type indicator and the quantitative indicator are different. After traversing the set of working status indicators, the heat source status indicators are obtained.

3. The method as described in claim 2, characterized in that, Based on the first working status index, a correlation analysis is performed on the cabinet temperature to obtain the correlation coefficient of the first working status index, including: When the first working status indicator is a quantitative indicator, with the first working status indicator as the only independent variable and the cabinet temperature as the dependent variable, Pearson correlation analysis is performed to obtain the correlation coefficient of the first working status indicator. When the first working status indicator is a type indicator, the first working status indicator is switched to the only control fluctuation quantity. Several cabinet temperature fluctuation quantities are collected, the average value is calculated, and the correlation coefficient of the first working status indicator is obtained.

4. The method as described in claim 1, characterized in that, Through the user terminal, work tasks within a preset time window are configured. These tasks are then processed using a control cabinet twin model to obtain characteristic values ​​of the heat source status indicators, including: Obtain the historical work log of the computer network control cabinet, wherein the historical work log includes an array of work type records, an array of work quantity records, and labels that identify the record values ​​of work status indicators; The work type record array and the workload record array are placed in the input node of a fully connected neural network, and the labels that identify the work status index record values ​​are placed in the output node of the fully connected neural network to train the control cabinet twin model. From the work tasks, extract the work type array and the workload array, and perform analysis through the control cabinet twin model to obtain the predicted values ​​of the work status indicators; Based on the heat source status index, feature values ​​of the heat source status index are extracted from the predicted values ​​of the working status index.

5. The method as described in claim 4, characterized in that, Using the characteristic values ​​of the heat source status index, the layout topology of the heat source equipment in the control cabinet, the layout topology of the heat dissipation equipment in the control cabinet, the external temperature and the external humidity of the cabinet as constraints, samples of control cabinets in the same mode when the heat dissipation equipment is not activated are retrieved, and mode analysis is performed to obtain the predicted value of the internal temperature of the cabinet, including: The system records the sample heat source status index, sample heat source equipment layout topology, sample heat dissipation equipment layout topology, sample cabinet external temperature, sample cabinet external humidity and cabinet internal temperature when the heat dissipation equipment is not started. When the deviation between the recorded value of the sample heat source status index and the characteristic value of the heat source status index is less than or equal to the corresponding index deviation threshold, and the similarity between the sample heat source equipment layout topology and the control cabinet heat source equipment layout topology is greater than or equal to the first topology similarity threshold, and the similarity between the sample heat dissipation equipment layout topology and the control cabinet heat dissipation equipment layout topology is greater than or equal to the second topology similarity threshold, and the deviation between the sample cabinet outside temperature and the cabinet outside temperature is less than or equal to the cabinet outside temperature deviation threshold, and the deviation between the sample cabinet outside humidity and the cabinet outside humidity is less than or equal to the cabinet outside humidity deviation threshold, the recorded value of the cabinet inside temperature is added to the same modal control cabinet sample. When the number of samples of the same modal control cabinet is greater than or equal to the preset fitting number, mode analysis is performed to obtain the predicted value of the temperature inside the cabinet.

6. The method as described in claim 5, characterized in that, When the number of samples from the same modal control cabinet is greater than or equal to the preset fitting number, a mode analysis is performed to obtain the predicted temperature value inside the cabinet, including: Based on the temperature deviation threshold, cluster analysis is performed on the set of cabinet internal temperature records of the same mode control cabinet sample to obtain multi-cluster cabinet internal temperature records. Delete the temperature record values ​​of clusters whose number of temperature record values ​​is less than or equal to the cluster number threshold from the set of temperature record values ​​inside the cabinet, and obtain the selected set of temperature record values ​​inside the cabinet. The average value of the selected set of temperature records inside the cabinet is calculated to obtain the predicted temperature value inside the cabinet.

7. The method as described in claim 1, characterized in that, Also includes: When the work task is not performed, or when either the external temperature or the external humidity changes, routine temperature control is performed.

8. An intelligent temperature control device for a computer network control cabinet, characterized in that, An intelligent temperature control method for implementing a computer network control cabinet according to any one of claims 1 to 7, the device comprising: The status acquisition module is used to obtain heat source status indicators, wherein the heat source status indicators characterize the working status indicators of non-heat dissipation equipment that affect the temperature inside the cabinet. The heat source index acquisition module is used to configure the work tasks for a future preset time window through the user terminal, process the work tasks through the control cabinet twin model, and obtain the characteristic values ​​of the heat source status index. The control cabinet parameter acquisition module is used to obtain the layout topology of the control cabinet heat source equipment, the layout topology of the control cabinet heat dissipation equipment, the cabinet outside temperature, and the cabinet outside humidity. The temperature prediction module is used to retrieve samples of the same mode control cabinet when the heat source status index feature value, the layout topology of the heat source equipment of the control cabinet, the layout topology of the heat dissipation equipment of the control cabinet, the outside temperature of the cabinet and the outside humidity of the cabinet as constraints, perform mode analysis, and obtain the predicted value of the cabinet temperature. A heat dissipation control optimization module is used to perform heat dissipation control optimization based on the predicted temperature inside the cabinet, the temperature outside the cabinet, and the humidity outside the cabinet, to obtain pre-configured heat dissipation parameters, including: Obtain the target temperature inside the cabinet; Using the predicted temperature inside the cabinet, the temperature outside the cabinet, the humidity outside the cabinet, the target temperature inside the cabinet, the layout topology of the heat source equipment in the control cabinet, and the layout topology of the heat dissipation equipment in the control cabinet as constraints, retrieve the set of records of heat dissipation control parameters; Outlier analysis is performed on the set of records of heat dissipation control parameters to obtain the set of outlier eigenvalues; Extract the heat dissipation control parameter record array corresponding to the minimum value of the outlier eigenvalue in the outlier eigenvalue set, and set it as the pre-configured heat dissipation parameter; The temperature control module is used to perform temperature control based on the pre-configured heat dissipation parameters when the work task is executed and the external temperature and humidity of the cabinet do not change.

Images