Weak electric security temperature control type cabinet cooling method and system

Abstract

The present application relates to the technical field of cabinet heat management, and discloses a temperature control type cabinet cooling method and system for weak current security and protection. The method comprises collecting the temperature inside the cabinet and constructing a temperature sequence with a time stamp; constructing a current three-dimensional heat distribution map according to the sequence and the layout coordinates of the cabinet; inputting the heat distribution map into a heat conduction simulation unit based on material thermal properties and fluid dynamics parameters, and calculating to generate a predictive three-dimensional heat distribution map at a future time; extracting and marking the heat areas to be intervened that exceed the threshold according to the prediction map; generating a preliminary cooling scheme based on the geometric center, volume of each heat area and the spatial relationship with the cooling unit, and forming a global cooling strategy after integration combined with the upper limit of the total power; and finally decomposing into control instructions for each cooling unit and executing power adjustment. The method realizes the advance prediction and spatial accurate regulation of the cabinet heat state, overcomes the hysteresis of the traditional method, and improves the cooling energy efficiency.

Description

A temperature-controlled cabinet cooling method and system for low-voltage security Technical Field

[0001] This invention relates to the field of cabinet thermal management technology, specifically to a temperature-controlled cabinet cooling method and system for low-voltage security. Background Technology

[0002] The high-density electronic equipment operating within low-voltage security cabinets generates a significant amount of concentrated heat. Currently, widely used cooling methods rely on temperature sensors installed in a few fixed locations within the cabinet. When a sensor reading exceeds a preset threshold, the control system activates the associated cooling equipment. This method is essentially a delayed response, intervening only after localized overheating has occurred. Because it relies on data from only a limited number of monitoring points, the system cannot obtain complete three-dimensional temperature field information within the cabinet and lacks the ability to predict future heat accumulation and diffusion paths. This results in a lack of foresight in thermal management and creates control blind spots and response delays due to insufficient information.

[0003] Existing cooling resource allocation methods are also relatively crude. Systems typically treat the entire rack space as a uniform target for cooling, or simply enhance cooling of the area near the alarm-triggered sensor. This approach ignores the non-uniformity of heat distribution in space, as well as the spatial geometric relationship between different cooling units and specific heat sources. As a result, the output of cooling airflow or cold energy lacks spatial directionality, easily leading to insufficient cooling of local hotspots and excessive cooling of other areas, resulting in resource misallocation, low overall cooling efficiency, and high energy consumption.

[0004] There is a need for a temperature control method that can predict the future heat distribution inside the cabinet in advance and, based on this, achieve precise and dynamic allocation of cooling resources in three-dimensional space, in order to overcome the lag of passive response and improve the energy efficiency and intelligence level of the cooling system. Summary of the Invention

[0005] The purpose of this invention is to provide a temperature-controlled cabinet cooling method and system for low-voltage security, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a temperature-controlled cabinet cooling method for low-voltage security systems, the method comprising:

[0007] Collect real-time temperature information from designated monitoring points inside the low-voltage security cabinet;

[0008] The real-time temperature information is converted into a temperature data sequence containing timestamps;

[0009] Based on the temperature data sequence and the pre-stored rack space layout coordinates, a three-dimensional heat distribution map is constructed within the current time window;

[0010] The three-dimensional heat distribution map is input into the heat conduction simulation unit, which performs calculations based on preset material thermal property parameters and fluid dynamic parameters to generate a predictive three-dimensional heat distribution map for future moments.

[0011] Based on the predicted future three-dimensional thermal distribution map, spatial regions that exceed a preset temperature threshold are extracted and marked as thermal regions to be intervened in.

[0012] Based on the geometric center location, volume parameters, and spatial relationship with the internal cooling units of the heat-affected areas, a preliminary cooling resource allocation scheme is generated for each heat-affected area.

[0013] Integrate the preliminary cooling resource allocation schemes for all heat areas to be intervened, and combine them with the total available cooling power limit of the low-voltage security cabinet to generate a global cooling strategy instruction;

[0014] The global cooling strategy instructions are decomposed into specific control instructions for different cooling units inside the rack.

[0015] The specific control command is executed to adjust the power of the designated cooling unit within the low-voltage security cabinet.

[0016] Preferably, the step of constructing a three-dimensional heat distribution map within the current time window based on the temperature data sequence and pre-stored rack space layout coordinates includes:

[0017] The temperature data sequence is subjected to time-series filtering to eliminate random temperature measurement noise;

[0018] Each temperature value in the temperature data sequence that has undergone time-series filtering is bound to the corresponding monitoring point spatial coordinates in the pre-stored cabinet spatial layout coordinates to form temperature point data with spatial coordinates.

[0019] Based on all temperature point data with spatial coordinates, a spatial interpolation algorithm is used to estimate the temperature of spatial locations inside the cabinet where no monitoring points are set, generating continuous temperature field data covering the entire internal space of the cabinet.

[0020] The continuous temperature field data is mapped onto a three-dimensional mesh model based on the coordinates of the cabinet spatial layout, thereby forming a three-dimensional thermal distribution map within the current time window.

[0021] Preferably, the step of inputting the three-dimensional heat distribution map into the heat conduction simulation unit, wherein the heat conduction simulation unit performs calculations based on preset material thermal property parameters and fluid dynamic parameters, including:

[0022] Obtain the power parameters and spatial location information of the main heat-generating components inside the low-voltage security cabinet;

[0023] The power parameters of the main heat-generating components are input to the heat conduction simulation unit as an internal steady-state heat source;

[0024] The thermal conductivity coefficient of the structural shell material of the low-voltage security cabinet and the convective heat transfer coefficient of the air inside the cabinet are used as the boundary condition parameters of the heat conduction simulation unit.

[0025] The three-dimensional heat distribution map within the current time window is used as the initial temperature field condition for the heat conduction simulation unit;

[0026] The heat conduction simulation unit is invoked to perform iterative calculations based on the internal steady-state heat source, the boundary condition parameters, and the initial temperature field conditions, according to a set future time step.

[0027] In each iteration of the calculation, the temperature value of each grid node in the three-dimensional mesh model is updated until a preset future time is reached, and the final predictive three-dimensional heat distribution map of the future time is output.

[0028] Preferably, the step of generating a preliminary cooling resource allocation scheme for each heat-affected area based on its geometric center location, volume parameters, and spatial relationship with the internal cooling units of the rack includes:

[0029] Spatial projection is performed from the geometric center of the heat-affected area to the plane where the cooling unit inside the cabinet is located, and at least one of the nearest cooling units is identified as the associated cooling unit.

[0030] Calculate the ratio between the volume parameters of the thermal region to be intervened and the rated heat dissipation capacity of the associated cooling unit, and use it as the basic cooling demand intensity index;

[0031] Based on the spatial distance between the associated cooling unit and the geometric center of the heat region to be intervened, the basic cooling demand intensity index is attenuated and compensated to obtain the corrected cooling demand intensity index.

[0032] Based on the revised cooling demand intensity index, the initial power ratio to be allocated to the associated cooling unit is determined, thereby forming the preliminary cooling resource allocation scheme.

[0033] Preferably, the preliminary cooling resource allocation scheme integrating all heat zones to be intervened, combined with the total available cooling power limit of the low-voltage security cabinet, generates a global cooling strategy instruction, including:

[0034] Summarize the initial power proportion allocated to each associated cooling unit in all preliminary cooling resource allocation schemes;

[0035] For the same associated cooling unit involved in multiple preliminary cooling resource allocation schemes, the total requested power ratio of the associated cooling unit is obtained by summing up all the initial power ratios allocated to it.

[0036] The total requested power ratio of each associated cooling unit is compared with the maximum allowed operating power ratio of the associated cooling unit. If the total requested power ratio exceeds the maximum allowed operating power ratio, the final power ratio of the associated cooling unit is set to its maximum allowed operating power ratio, and the overflow power difference is recorded.

[0037] Based on the total available cooling power limit of the low-voltage security cabinet, the final power ratio of all associated cooling units is normalized and adjusted to ensure that the sum of the actual operating power of all associated cooling units does not exceed the total available cooling power limit.

[0038] Based on the adjusted final power ratio of each associated cooling unit, a global cooling strategy instruction containing the target power value of each cooling unit is generated.

[0039] Preferably, the processing of the overflowed power difference includes:

[0040] When the requested power of an associated cooling unit exceeds its maximum allowable operating power ratio, resulting in an overflow power difference, the heat region to be intervened corresponding to the associated cooling unit is identified.

[0041] Identify other cooling units that are spatially adjacent to the thermal area to be intervened and whose power request has not reached the upper limit, and use them as auxiliary cooling units;

[0042] Based on the spatial distance between the auxiliary cooling unit and the area to be intervened in the heat, as well as the remaining available power capacity of the auxiliary cooling unit, a portion or all of the overflow power difference is allocated to the auxiliary cooling unit, and the total requested power ratio of the auxiliary cooling unit is updated.

[0043] Preferably, after generating the global cooling strategy instructions and before executing specific control instructions, a cooling strategy verification step is also included:

[0044] Construct a simplified real-time thermal feedback verification model;

[0045] The global cooling strategy command is input into the simplified real-time thermal feedback verification model to simulate the cooling effect within a shortened future time window;

[0046] Obtain the verification three-dimensional heat distribution map output by the simplified model after the simulation is completed;

[0047] Determine whether there are still spatial regions exceeding the preset temperature threshold in the three-dimensional thermal distribution map used for verification;

[0048] If it exists, the power value of the corresponding cooling unit in the global cooling strategy instruction is fine-tuned according to the verification result to generate a corrected global cooling strategy instruction.

[0049] Preferably, the construction of the simplified real-time thermal feedback verification model includes:

[0050] Based on the three-dimensional heat distribution map within the current time window and the global cooling strategy instructions, a local three-dimensional mesh sub-model focusing on the heat region to be intervened in and its adjacent space is established.

[0051] Ignore temperature changes outside the boundary of the local 3D mesh sub-model and treat the boundary temperature as constant;

[0052] The power value of the cooling unit located within or within the local three-dimensional mesh sub-model in the global cooling strategy instruction is converted into the equivalent heat dissipation rate in the local three-dimensional mesh sub-model.

[0053] Using a larger time step than the heat conduction simulation unit, rapid iterative calculations are performed on the local three-dimensional mesh sub-model to obtain a verification three-dimensional heat distribution map at the end of the shortened future time window.

[0054] Preferably, the collection of real-time temperature information at designated monitoring points inside the low-voltage security cabinet includes:

[0055] Simultaneously activate multiple temperature sensors arranged at different depths and heights inside the low-voltage security cabinet;

[0056] The multiple temperature sensors are driven by synchronous sampling pulses to acquire temperature readings at each monitoring point at the same physical moment.

[0057] The acquired temperature readings are packaged with the corresponding temperature sensor identifiers and transmitted to the data processing node via redundant data links to form a raw temperature information set.

[0058] Preferably, the present invention also includes a temperature-controlled cabinet cooling system for low-voltage security, the system including a main controller and a distributed temperature sensor array and a controllable cooling unit array connected in communication with it, the main controller being configured to execute the temperature-controlled cabinet cooling method for low-voltage security as described above.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] A three-dimensional thermal distribution map is constructed based on real-time acquired temperature data sequences and the rack space layout, and then input into a heat conduction simulation unit that integrates pre-set material thermal properties and fluid dynamic parameters for calculation. This process simulates the conduction, convection, and diffusion of heat within the specific structure, materials, and airflow field of the rack using a physical model. It can predict future temperature field changes, enabling early identification and warning of potential overheating areas. This transforms the cooling system's operating mode from passively remedying overheating at monitoring points to proactively intervening before abnormal temperature increases occur, eliminating the lag window in temperature control and improving the foresight and reliability of thermal safety management.

[0061] Based on the predicted heat distribution map, the geometric center and volume parameters of the heat load define its spatial location and scale. This, combined with the spatial orientation of each cooling unit, generates a preliminary allocation scheme, which is then globally integrated and optimized under total power constraints. This mechanism drives the cooling units to adjust their power output and airflow angle according to the spatial attributes of the heat zone. Cooling airflow or cooling capacity is directed precisely to specific spatial coordinates where overheating is imminent, rather than indiscriminately filling the entire cabinet cavity. This achieves targeted deployment and on-demand distribution of cooling resources in three-dimensional space, avoiding cooling blind spots and resource waste. Under the same cooling power, it achieves better overall temperature uniformity and hotspot suppression capabilities, or reduces the overall energy consumption of the cooling system while maintaining the same heat dissipation effect. Attached Figure Description

[0062] Figure 1 is a schematic diagram of the working principle of the temperature-controlled cabinet cooling method for low-voltage security according to the present invention;

[0063] Figure 2 is a flowchart for constructing a three-dimensional heat distribution map;

[0064] Figure 3 is a flowchart of generating a preliminary cooling resource allocation plan;

[0065] Figure 4 shows the dynamic correlation curve between total cooling power and average cabinet temperature.

[0066] Figure 5 shows the simulated distribution of thermal conduction in the three-dimensional temperature field of the cross-section of the low-voltage security cabinet. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Referring to Figure 1, this invention provides a temperature-controlled cabinet cooling method for low-voltage security systems. The method includes: collecting real-time temperature information from designated monitoring points inside the low-voltage security cabinet, the information originating from temperature sensors pre-deployed within the cabinet; converting the real-time temperature information into a temperature data sequence containing timestamps, the conversion process being completed by a data acquisition module, the timestamps marking the acquisition time of each temperature value; constructing a three-dimensional heat distribution map within the current time window based on the temperature data sequence and pre-stored cabinet spatial layout coordinates, the pre-stored cabinet spatial layout coordinates defining the specific location of each monitoring point within the three-dimensional space of the cabinet; inputting the three-dimensional heat distribution map into a heat conduction simulation unit, the heat conduction simulation unit performing calculations based on preset material thermal property parameters and fluid dynamics parameters to generate a predictive three-dimensional heat distribution map for future moments, the preset parameters being stored in the system's configuration database; and extracting spatial regions exceeding a preset temperature threshold from the predictive future three-dimensional heat distribution map, marking these spatial regions as heat regions to be intervened in, the preset temperature threshold being set according to the safe operating temperature of the equipment within the cabinet. Based on the geometric center location, volume parameters, and spatial relationship with the cooling units inside the cabinet of the heat-affected area to be intervened, a preliminary cooling resource allocation scheme is generated for each heat-affected area. The location information of the cooling units inside the cabinet is predefined. The preliminary cooling resource allocation schemes for all heat-affected areas are integrated, and combined with the upper limit of the total available cooling power of the low-voltage security cabinet, a global cooling strategy instruction is generated. The upper limit of the total available cooling power is determined by the cabinet's power supply and cooling system specifications. The global cooling strategy instruction is decomposed into specific control instructions for different cooling units inside the cabinet. The decomposition process is completed based on the identification mapping of the cooling units. The specific control instructions are executed to adjust the power of the designated cooling units within the low-voltage security cabinet. Power adjustment is achieved by adjusting the output of the cooling unit drive circuit.

[0069] Example 1: Referring to Figure 2, the temperature data sequence is subjected to time-series filtering to eliminate random temperature measurement noise. The time-series filtering uses a moving average algorithm. Each temperature value in the time-series filtered temperature data sequence is bound to the corresponding monitoring point spatial coordinates in the pre-stored cabinet spatial layout coordinates, forming temperature point data with spatial coordinates. The binding operation is performed through the data association module. Based on all temperature point data with spatial coordinates, a spatial interpolation algorithm is used to estimate the temperature of the spatial locations inside the cabinet where no monitoring points are set, generating continuous temperature field data covering the entire internal space of the cabinet. The spatial interpolation algorithm uses the radial basis function method. The continuous temperature field data is mapped to a three-dimensional mesh model established based on the cabinet spatial layout coordinates, thereby forming a three-dimensional thermal distribution map within the current time window. The three-dimensional mesh model consists of a finite number of mesh cells. The power parameters and spatial location information of the main heat-generating components inside the low-voltage security cabinet are obtained. This information is read from the equipment configuration file. The power parameters of the main heat-generating components are input as internal steady-state heat sources to the heat conduction simulation unit. The thermal conductivity coefficient of the structural shell material of the low-voltage security cabinet and the convective heat transfer coefficient of the air inside the cabinet are used as boundary condition parameters for the heat conduction simulation unit. The three-dimensional heat distribution map within the current time window is used as the initial temperature field condition for the heat conduction simulation unit. The heat conduction simulation unit is invoked and iteratively calculated according to the set future time step based on the internal steady-state heat source, the boundary condition parameters, and the initial temperature field condition. The heat conduction simulation unit uses the finite volume method to solve the energy equation. In each iteration, the temperature value of each grid node in the three-dimensional mesh model is updated until the preset future time is reached, and the final predictive three-dimensional heat distribution map of the future time is output.

[0070] In practical implementation, the temperature data sequence undergoes time-series filtering to eliminate random temperature measurement noise. This filtering employs a first-order lag filtering algorithm, whose output is obtained by weighting the filtering result from the previous moment with the sampled value at the current moment. Each temperature value in the time-series filtered temperature data sequence is bound to the corresponding spatial coordinates of a monitoring point in a pre-stored rack space layout coordinate system. This binding operation is performed by looking up a predefined coordinate mapping table, thus forming temperature point data with spatial coordinates. Based on all temperature point data with spatial coordinates, a spatial interpolation algorithm is used to estimate the temperature of unmonitored locations within the rack, generating continuous temperature field data covering the entire rack interior. In some embodiments, the spatial interpolation algorithm uses an inverse distance weighting method. The inverse distance weighting method estimates the temperature by calculating a weighted average of the distances between the unknown point and all known temperature point data; the weighting function is inversely proportional to the power of the distance. A specific formula for this algorithm is expressed as:

[0071]

[0072] Where: symbol Indicates the spatial location to be estimated Temperature value at, symbol Indicates the first Temperature values ​​of temperature points with spatial coordinates, symbol Indicates position To the The Euclidean distance between the spatial coordinates of each temperature point, sign... It is the power-law parameter of the distance and is a constant greater than zero, with the sign... This represents the total number of temperature point data points with spatial coordinates involved in the interpolation. Continuous temperature field data is mapped onto a 3D mesh model based on the spatial layout coordinates of the server rack, thus creating a three-dimensional thermal distribution map within the current time window. The mapping process sets the temperature value of each mesh node to the estimated temperature value corresponding to that node's spatial location in the continuous temperature field data.

[0073] The power parameters and spatial location information of the main heat-generating components inside the low-voltage security cabinet are obtained. This information is understood to be read from the equipment management database, and the power parameters include rated power or real-time monitored power. The power parameters of the main heat-generating components are input as internal steady-state heat sources into the heat conduction simulation unit. Optionally, the internal steady-state heat sources are defined as the heat generation rate of the corresponding grid region in the three-dimensional mesh model based on the spatial location information of the heat-generating components. The thermal conductivity coefficient of the cabinet's outer shell material and the convective heat transfer coefficient of the air inside the cabinet are used as boundary condition parameters for the heat conduction simulation unit. In some embodiments, these material thermal properties and fluid dynamics parameters are stored in the system configuration file. The three-dimensional heat distribution map within the current time window is used as the initial temperature field condition for the heat conduction simulation unit.

[0074] The heat conduction simulation unit is invoked to perform iterative calculations based on the internal steady-state heat source, boundary condition parameters, and initial temperature field conditions, according to a set future time step. The heat conduction simulation unit uses the finite difference method to solve the three-dimensional unsteady-state heat conduction equation. In each iteration, the temperature value of each grid node in the three-dimensional mesh model is updated, and the calculation is updated to consider the combined effects of heat conduction, convection heat transfer, and internal heat sources. The iterative calculation continues until the preset future time is reached, outputting the final predictive three-dimensional heat distribution map for the future time.

[0075] Example 2: Referring to Figure 3, a spatial projection is made from the geometric center of the area to be intervened in the heat zone to the plane containing the cooling units inside the cabinet. At least one nearest cooling unit is identified as the associated cooling unit. The spatial projection uses an orthogonal projection method. The ratio between the volume parameter of the area to be intervened in the heat zone and the rated heat dissipation capacity of the associated cooling unit is calculated as the basic cooling demand intensity index. Based on the spatial distance between the associated cooling unit and the geometric center of the area to be intervened in the heat zone, the basic cooling demand intensity index is attenuated to obtain a corrected cooling demand intensity index. The attenuation compensation follows an inverse relationship with the distance. Based on the corrected cooling demand intensity index, the initial power ratio to be allocated to the associated cooling unit is determined, thus forming the preliminary cooling resource allocation scheme. The initial power ratio is obtained through a linear mapping function.

[0076] In practical implementation, a spatial projection is made from the geometric center of the area to be intervened in the heat zone to the plane where the cooling units inside the cabinet are located, determining the nearest cooling unit as the associated cooling unit. In practice, a vertical projection method is used, and the plane where the cooling units inside the cabinet are located is typically the inner wall of the cabinet where fans or air conditioning vents are installed. The ratio between the volumetric parameter of the area to be intervened in the heat zone and the rated heat dissipation capacity of the associated cooling unit is calculated as the basic cooling demand intensity index. It can be understood that the volumetric parameter is calculated from the corresponding spatial region of the three-dimensional heat distribution map, and the rated heat dissipation capacity is an inherent performance parameter of the cooling unit. The basic cooling demand intensity index is attenuated and compensated based on the spatial distance between the associated cooling unit and the geometric center of the area to be intervened in the heat zone to obtain a corrected cooling demand intensity index. This attenuation compensation reflects the physical law that the cooling effect weakens with increasing distance. This attenuation compensation relationship can be implemented through a function containing a distance variable; in some embodiments, an exponential attenuation model is used.

[0077] In some embodiments, when multiple cooling units exist within a set distance tolerance range after spatial projection from the geometric center of the area to be intervened in the heat region, all multiple cooling units can be identified as associated cooling units. A corrected cooling demand intensity index is calculated for each associated cooling unit. It can be understood that different associated cooling units differ in spatial distance. The calculated corrected cooling demand intensity index will differ depending on the location. Based on the corrected cooling demand intensity index of each associated cooling unit, an initial power ratio is allocated, thus forming a preliminary cooling resource allocation scheme for a given heat zone corresponding to multiple associated cooling units. In practical implementation, spatial distance... The calculation is based on the Euclidean distance formula in a three-dimensional coordinate system. Attenuation coefficient. The value can be calibrated using prior fluid simulations or experimental data. Optionally, the calibration process aims to make the cooling resource allocation scheme more consistent with the actual heat dissipation characteristics.

[0078] Example 3: Summarize the initial power ratios allocated to each associated cooling unit in all preliminary cooling resource allocation schemes. The summarization operation is completed in the data aggregation module. For the same associated cooling unit involved in multiple preliminary cooling resource allocation schemes, all the initial power ratios allocated to it are superimposed to obtain the total requested power ratio of the associated cooling unit. The total requested power ratio of each associated cooling unit is compared with the maximum allowable operating power ratio of the associated cooling unit. If the total requested power ratio exceeds the maximum allowable operating power ratio, the final power ratio of the associated cooling unit is set to its maximum allowable operating power ratio, and the overflow power difference is recorded. According to the total available cooling power limit of the low-voltage security cabinet, the final power ratio of all associated cooling units is normalized to ensure that the sum of the actual operating power of all associated cooling units does not exceed the total available cooling power limit. The normalization adjustment adopts a scaling algorithm. Based on the adjusted final power ratio of each associated cooling unit, a global cooling strategy instruction containing the target power value of each cooling unit is generated. When the requested power of an associated cooling unit exceeds its maximum allowable operating power ratio, resulting in an overflow power difference, the heat area to be intervened corresponding to the associated cooling unit is identified. Other cooling units that are spatially adjacent to the thermal area to be intervened and whose power request has not reached the upper limit are identified as auxiliary cooling units. Spatial proximity is determined based on the grid distance in the three-dimensional mesh model. Based on the spatial distance between the auxiliary cooling unit and the thermal area to be intervened, as well as the remaining available power capacity of the auxiliary cooling unit, a portion or all of the overflow power difference is allocated to the auxiliary cooling unit, and the total requested power ratio of the auxiliary cooling unit is updated.

[0079] In practice, the initial power ratio allocated to each associated cooling unit in all preliminary cooling resource allocation schemes is summarized. This summarization is done by iterating through the list of preliminary cooling resource allocation schemes corresponding to all areas requiring intervention. For the same associated cooling unit involved in multiple preliminary cooling resource allocation schemes, all its allocated initial power ratios are summed to obtain the total requested power ratio for the associated cooling unit. This summarization calculation is performed separately for each associated cooling unit, and the formula is as follows:

[0080]

[0081] Where: symbol Indicates the first The proportion of the total requested power of each associated cooling unit, symbol Indicates the first The preliminary cooling resource allocation plan is the first one. The initial power allocation to each associated cooling unit, symbol It involves the first The total number of preliminary cooling resource allocation schemes for each associated cooling unit.

[0082] The total requested power ratio of each associated cooling unit is compared with the maximum allowable operating power ratio of the associated cooling unit. In some embodiments, the maximum allowable operating power ratio is a set upper limit of the safe operating range of the cooling unit. If the total requested power ratio of the associated cooling unit exceeds its maximum allowable operating power ratio, the final power ratio of the associated cooling unit is set to its maximum allowable operating power ratio, and the overflow power difference is recorded. Calculated as ,symbol Indicates the first The maximum allowable operating power ratio of each associated cooling unit. If the total requested power ratio of the associated cooling units does not exceed the maximum allowable operating power ratio, the final power ratio of the associated cooling units is temporarily set to the total requested power ratio, and the overflow power difference is zero.

[0083] The final power ratio of all associated cooling units is normalized based on the total available cooling power limit of the low-voltage security cabinet. This normalization ensures that the sum of the actual operating power of all associated cooling units does not exceed the total available cooling power limit. In some embodiments, the normalization is achieved through a scaling factor, which is equal to the ratio of the total available cooling power limit to the sum of the final power ratios of all associated cooling units. Based on the adjusted final power ratios of each associated cooling unit, a global cooling strategy instruction containing the target power value for each cooling unit is generated. The target power value is obtained by multiplying the final power ratio by the rated power of the cooling unit.

[0084] When an associated cooling unit's requested power exceeds its maximum allowable operating power ratio, resulting in an overflow power difference, the corresponding heat region to be intervened for that associated cooling unit is identified. Optionally, this identification process is completed by querying the mapping relationship between the preliminary cooling resource allocation scheme and the associated cooling units. Other cooling units spatially adjacent to the heat region to be intervened for power requests not exceeding the upper limit are identified as auxiliary cooling units. Spatial proximity is determined based on Manhattan distance or Euclidean distance thresholds in the 3D mesh model. Based on the spatial distance between the auxiliary cooling unit and the heat region to be intervened for, and the remaining available power capacity of the auxiliary cooling unit, a portion or all of the overflow power difference is allocated to the auxiliary cooling unit, and the total requested power ratio of the auxiliary cooling unit is updated. The remaining available power capacity is the difference between the maximum allowable operating power ratio of the auxiliary cooling unit and its current total requested power ratio. In specific implementations, the allocation of the overflow power difference can be weighted based on the relative proximity and remaining capacity of the auxiliary cooling units. After updating the total requested power ratio of the auxiliary cooling unit, it is necessary to re-execute the comparison with the maximum allowable operating power ratio and subsequent normalization adjustment steps. Optionally, this process can be iterated until no new overflow power difference is generated or a preset number of iterations is reached.

[0085] Example 4: After generating the global cooling strategy command and before executing the specific control command, a cooling strategy verification step is performed. A simplified real-time thermal feedback verification model is constructed, which focuses on the rapid evaluation of heat distribution. The global cooling strategy command is input into the simplified real-time thermal feedback verification model to simulate the cooling effect within a shortened future time window. The verification three-dimensional heat distribution map output by the simplified model after the simulation is completed is obtained. It is determined whether there are still spatial regions exceeding the preset temperature threshold in the verification three-dimensional heat distribution map. If so, the power values ​​of the corresponding cooling units in the global cooling strategy command are fine-tuned according to the verification results to generate a corrected global cooling strategy command. Based on the three-dimensional heat distribution map within the current time window and the global cooling strategy command, a local three-dimensional mesh sub-model focusing on the heat region to be intervened in and its adjacent space is established. Temperature changes outside the boundary of the local three-dimensional mesh sub-model are ignored, and the boundary temperature is considered constant. The power values ​​of the cooling units located within or within the effective range of the local three-dimensional mesh sub-model in the global cooling strategy command are converted into the equivalent heat dissipation rate in the local three-dimensional mesh sub-model. Using a larger time step than the heat conduction simulation unit, rapid iterative calculations are performed on the local three-dimensional mesh sub-model to obtain a verification three-dimensional heat distribution map at the end of the shortened future time window.

[0086] In the specific implementation, the cooling strategy verification steps included after generating the global cooling strategy command and before executing specific control commands are described. A simplified real-time thermal feedback verification model is constructed, which is understood to be used to quickly evaluate the cooling effect of the global cooling strategy command before final execution. Based on the 3D thermal distribution map within the current time window and the global cooling strategy command, a local 3D mesh sub-model focusing on the thermal region to be intervened in and its adjacent space is established. The local 3D mesh sub-model is a spatial subset of the complete 3D mesh model.

[0087] In some embodiments, constructing a simplified real-time thermal feedback verification model involves simplifying the settings of key parameters. Optional simplification methods are illustrated in the following parameter comparison table, see Table 1:

[0088] Table 1: Comparison of Element Parameters between Simplified Validation Model and Full Heat Conduction Simulation Model

[0089]

[0090] The global cooling strategy command is input into a simplified real-time thermal feedback verification model to simulate the cooling effect within a shortened future time window. The length of the shortened future time window is shorter than the future time window used by the heat conduction simulation unit for prediction. After the simulation is completed, a verification 3D thermal distribution map is obtained from the simplified model output. This map reflects the predicted temperature field in a local area at the end of the shortened future time window after applying the cooling power corresponding to the global cooling strategy command.

[0091] Ignoring temperature variations outside the boundaries of the local 3D mesh sub-model, the boundary temperature is treated as constant. In practice, the boundary temperature value is taken from the temperature value at the corresponding boundary location on the 3D heat distribution map within the current time window. The power value of the cooling unit located within or within the effective range of the local 3D mesh sub-model in the global cooling strategy command is converted into the equivalent heat dissipation rate in the local 3D mesh sub-model. This conversion can be understood as based on the principle of energy conservation; the heat removed by the cooling unit per unit time is equivalent to applying a negative heat source term at the corresponding location in the local 3D mesh sub-model. Equivalent heat dissipation rate The calculation formula is expressed as follows:

[0092]

[0093] Where: symbol Represents the equivalent heat dissipation rate, symbol The efficiency coefficient of a cooling unit is a dimensionless coefficient that converts input electrical power into effective heat dissipation. (Symbol: ) This indicates the target electrical power value assigned to a specific cooling unit in the global cooling strategy command.

[0094] Using a larger time step than the heat conduction simulation unit, rapid iterative calculations are performed on the local 3D mesh sub-model to obtain the verification 3D heat distribution map at the end of the shortened future time window. The larger time step reduces the number of iterations required to reach the specified simulation time. It is determined whether there are still spatial regions exceeding a preset temperature threshold in the verification 3D heat distribution map. In some embodiments, this determination is completed by scanning the temperature values ​​of all mesh nodes in the verification 3D heat distribution map. If spatial regions exceeding the preset temperature threshold exist, the power values ​​of the corresponding cooling units in the global cooling strategy command are fine-tuned based on the verification results, generating a corrected global cooling strategy command. Optionally, the fine-tuning operation increases the power values ​​of cooling units affecting the hottest areas by a preset step size. If the temperature of all regions in the verification 3D heat distribution map does not exceed the preset temperature threshold, the global cooling strategy command does not need to be corrected.

[0095] Referring to Figure 4, the actual operation of the low-voltage security temperature-controlled cabinet cooling method reveals the dynamic changes and correlations of total cooling power, average temperature before cooling, and average temperature after cooling over time. Specifically, the fluctuation of total cooling power (blue curve) shows a significant positive correlation with the change of average temperature before cooling (red curve): when the average temperature before cooling increases, the total cooling power increases accordingly to enhance heat dissipation; while the average temperature after cooling (green curve) shows a lagging downward trend with the adjustment of total cooling power, reflecting the effect of the cooling strategy on cabinet temperature control. Further analysis shows that within the fluctuation period of the time step, the peak value of total cooling power and the peak value of average temperature before cooling are highly consistent in time sequence, reflecting the real-time adaptation of the cooling resource allocation scheme to changes in heat load; at the same time, the fluctuation amplitude of the average temperature after cooling is smaller than that of the average temperature before cooling, indicating that the global cooling strategy command can effectively suppress drastic changes in cabinet temperature, verifying the optimization effect of the cooling strategy verification step on the control effect. This curve visually demonstrates the dynamic coupling relationship between cooling power regulation and rack temperature changes, providing data support for subsequent parameter optimization of cooling strategies.

[0096] Example 5: Simultaneously activate multiple temperature sensors arranged at different depths and heights inside the low-voltage security cabinet. The activation signal is issued by the central controller. Synchronous sampling pulses drive the multiple temperature sensors to acquire temperature readings at each monitoring point at the same physical moment. The acquired temperature readings are packaged with the corresponding temperature sensor identifiers and transmitted to the data processing node via a redundant data link to form a raw temperature information set. The redundant data link uses a dual-channel transmission protocol.

[0097] In practice, multiple temperature sensors located at different depths and heights within the low-voltage security cabinet are simultaneously activated. The activation process is accomplished by the central control unit sending a unified digital signal to all temperature sensors, ensuring that all sensors transition from dormant to operational status at the same time. A synchronous sampling pulse drives the multiple temperature sensors to acquire temperature readings at each monitoring point at the same physical moment. This synchronous sampling pulse is an electrical signal with a precise period and rising edge timing, generated by the clock module within the central control unit and broadcast to all temperature sensors.

[0098] In some embodiments, the timing accuracy of the synchronous sampling pulse is crucial to ensuring data simultaneity. Optionally, the clock module employs a highly stable crystal oscillator as the time reference. It is understood that each temperature sensor immediately performs an analog-to-digital conversion upon receiving the rising edge of the synchronous sampling pulse, converting the analog signal from the sensed thermistor into a digital temperature reading. The acquired temperature reading is then packaged with the corresponding temperature sensor identifier, a unique digital code pre-assigned to each sensor to distinguish the data source in subsequent processing. The packaging operation is performed locally on the microcontroller of each temperature sensor, generating a data packet containing an identifier field and a temperature data field.

[0099] The raw temperature information is transmitted to the data processing node via redundant data links to form a raw data set. These redundant data links employ a dual-channel transmission protocol, such as simultaneously using a wired RS-485 bus and a wireless ZigBee module at the physical layer. In practice, the data processing node verifies and selects the best data packets received from the same temperature sensor via both channels to ensure data integrity and reliability. At the data processing node, the raw temperature information set is organized into a timestamped data structure, where each record is associated with a temperature sensor identifier and its acquired temperature reading.

[0100] Period of synchronous sampling pulse The sampling period is directly related to the time resolution of the temperature data sequence, and its value is set according to monitoring requirements. It can be understood that the time interval between the rising edges of two synchronous sampling pulses defines the sampling period. In one implementation, to ensure the capture of temperature change trends within a short time window, the sampling period... The calculation needs to consider thermal inertia and signal processing speed, and their relationship can be expressed as:

[0101]

[0102] Where: symbol Indicates the period of the synchronous sampling pulse, symbol The smallest observable time constant representing the temperature change of a typical heat-generating component within a low-voltage security cabinet, denoted by the symbol [symbol missing]. This represents the maximum processing time required from the completion of analog-to-digital conversion by the sensor to the readiness of the data packet. In some embodiments, the temperature sensor identification is written to the sensor's non-volatile memory during system initialization via a hardware DIP switch or software configuration. The dual-channel transmission protocol execution mechanism in the redundant data link is as follows: the data processing node listens to two communication ports in parallel, immediately records the valid data packet received from either port, and compares it with the data packet of the same identification arriving later from the other port. The version with the correct checksum or better signal quality is selected and stored in the original temperature information set. Optionally, the signal quality is determined by the received signal strength indication or the bit error rate.

[0103] Referring to Figure 5, a three-dimensional mesh model presents the spatial distribution characteristics of the temperature field at the Z=75cm cross-section of the server rack. In the figure, the X and Y axes correspond to the planar coordinates of the server rack cross-section (unit: cm), the Z-axis is the height coordinate of the server rack space at that cross-section, and the color scale on the right quantifies the temperature range (24.0℃~34.8℃). From the distribution pattern, the central region of the cross-section exhibits a clear high-temperature accumulation zone (corresponding to the red-orange range of 31.2℃~34.8℃ in the color scale), gradually transitioning to a low-temperature region towards the periphery (corresponding to the blue-purple range of 24.0℃~29.4℃ in the color scale). This characteristic conforms to the physical law in heat conduction simulation that "heat-generating components are steady-state heat sources, and heat diffuses into the surrounding space." The generation of this distribution map is based on spatial interpolation calculations of pre-stored server rack space layout coordinates and temperature point data. It is the result of iterative calculations by the heat conduction simulation unit under the initial temperature field conditions and can be directly used for subsequent extraction of the heat-affected area and formulation of cooling strategies.

[0104] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0105] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A temperature-controlled cabinet cooling method for low-voltage security systems, characterized in that, include: Collect real-time temperature information from designated monitoring points inside the low-voltage security cabinet; The real-time temperature information is converted into a temperature data sequence containing timestamps; based on the temperature data sequence and the pre-stored rack space layout coordinates, a three-dimensional heat distribution map is constructed within the current time window; The three-dimensional heat distribution map is input into the heat conduction simulation unit, which performs calculations based on preset material thermal property parameters and fluid dynamic parameters to generate a predictive three-dimensional heat distribution map for future moments. Based on the predicted future three-dimensional thermal distribution map, spatial regions that exceed a preset temperature threshold are extracted and marked as thermal regions to be intervened in. Based on the geometric center location, volume parameters, and spatial relationship with the cooling units inside the cabinet of the heat-affected area to be intervened, a preliminary cooling resource allocation scheme is generated for each heat-affected area to be intervened; the preliminary cooling resource allocation schemes of all heat-affected areas to be intervened are integrated, and combined with the upper limit of the total available cooling power of the low-voltage security cabinet, a global cooling strategy instruction is generated; the global cooling strategy instruction is decomposed into specific control instructions for different cooling units inside the cabinet; the specific control instructions are executed to adjust the power of the designated cooling units inside the low-voltage security cabinet; the construction of a three-dimensional heat distribution map within the current time window based on the temperature data sequence and the pre-stored cabinet spatial layout coordinates includes: performing time-series filtering on the temperature data sequence to eliminate random temperature measurement noise; binding each temperature value in the time-series filtered temperature data sequence with the corresponding monitoring point spatial coordinates in the pre-stored cabinet spatial layout coordinates to form temperature point data with spatial coordinates; based on all temperature point data with spatial coordinates, a spatial interpolation algorithm is used to estimate the temperature of the spatial locations inside the cabinet where no monitoring points are set, generating continuous temperature field data covering the entire internal space of the cabinet; the... The continuous temperature field data is mapped onto a three-dimensional mesh model established based on the spatial layout coordinates of the cabinet, thereby forming a three-dimensional thermal distribution map within the current time window. The three-dimensional thermal distribution map is then input into a heat conduction simulation unit, which performs calculations based on preset material thermal property parameters and fluid dynamic parameters. This includes: obtaining the power parameters and spatial location information of the main heat-generating components inside the low-voltage security cabinet; inputting the power parameters of the main heat-generating components as an internal steady-state heat source into the heat conduction simulation unit; using the thermal conductivity coefficient of the structural shell material of the low-voltage security cabinet and the convective heat transfer coefficient of the air inside the cabinet as boundary condition parameters of the heat conduction simulation unit; using the three-dimensional thermal distribution map within the current time window as the initial temperature field condition of the heat conduction simulation unit; calling the heat conduction simulation unit to perform iterative calculations based on the internal steady-state heat source, the boundary condition parameters, and the initial temperature field condition, according to a set future time step; updating the temperature value of each grid node in the three-dimensional mesh model in each iteration until a preset future time is reached, and outputting the final predictive three-dimensional thermal distribution map for the future time.

2. The method for cooling a temperature-controlled cabinet in a low-voltage security system according to claim 1, characterized in that, The process involves generating a preliminary cooling resource allocation scheme for each heat-affected area based on its geometric center location, volume parameters, and spatial relationship with the cooling units inside the cabinet. This includes: projecting a spatial image from the geometric center of the heat-affected area onto the plane containing the cooling units inside the cabinet, identifying the nearest cooling unit as an associated cooling unit; calculating the ratio between the volume parameters of the heat-affected area and the rated heat dissipation capacity of the associated cooling unit as a basic cooling demand intensity index; attenuating the basic cooling demand intensity index based on the spatial distance between the associated cooling unit and the geometric center of the heat-affected area to obtain a corrected cooling demand intensity index; and determining the initial power allocation ratio for the associated cooling unit based on the corrected cooling demand intensity index, thereby forming the preliminary cooling resource allocation scheme.

3. The method for cooling a temperature-controlled cabinet for low-voltage security as described in claim 2, characterized in that, The process integrates the preliminary cooling resource allocation schemes for all areas to be intervened in, and combines this with the total available cooling power limit of the low-voltage security cabinet to generate a global cooling strategy instruction. This includes: summarizing the initial power ratio allocated to each associated cooling unit in all preliminary cooling resource allocation schemes; for the same associated cooling unit involved in multiple preliminary cooling resource allocation schemes, superimposing all the initial power ratios allocated to it to obtain the total requested power ratio of the associated cooling unit; comparing the total requested power ratio of each associated cooling unit with the maximum allowable operating power ratio of the associated cooling unit; if the total requested power ratio exceeds the maximum allowable operating power ratio, setting the final power ratio of the associated cooling unit to its maximum allowable operating power ratio, and recording the overflow power difference; normalizing and adjusting the final power ratio of all associated cooling units according to the total available cooling power limit of the low-voltage security cabinet to ensure that the sum of the actual operating power of all associated cooling units does not exceed the total available cooling power limit; and generating a global cooling strategy instruction containing the target power value of each cooling unit based on the adjusted final power ratio of each associated cooling unit.

4. A temperature-controlled cabinet cooling method for low-voltage security as described in claim 3, characterized in that, The processing of the overflow power difference includes: when an associated cooling unit's requested power exceeds its maximum allowable operating power ratio, resulting in an overflow power difference, identifying the heat region to be intervened for the associated cooling unit; finding other cooling units spatially adjacent to the heat region to be intervened for which the power request has not reached the upper limit as auxiliary cooling units; allocating part or all of the overflow power difference to the auxiliary cooling unit based on the spatial distance between the auxiliary cooling unit and the heat region to be intervened for and the remaining available power capacity of the auxiliary cooling unit, and updating the total requested power ratio of the auxiliary cooling unit.

5. A temperature-controlled cabinet cooling method for low-voltage security as described in claim 1, characterized in that, After generating the global cooling strategy command and before executing the specific control command, a cooling strategy verification step is also included: constructing a simplified real-time thermal feedback verification model; inputting the global cooling strategy command into the simplified real-time thermal feedback verification model to simulate the cooling effect within a shortened future time window; obtaining the verification three-dimensional thermal distribution map output by the simplified model after the simulation is completed; and determining whether there are still spatial regions in the verification three-dimensional thermal distribution map that exceed the preset temperature threshold. If it exists, the power value of the corresponding cooling unit in the global cooling strategy instruction is fine-tuned according to the verification result to generate a corrected global cooling strategy instruction.

6. A temperature-controlled cabinet cooling method for low-voltage security as described in claim 5, characterized in that, The construction of the simplified real-time thermal feedback verification model includes: establishing a local three-dimensional mesh sub-model focusing on the thermal region to be intervened in and its adjacent space based on the three-dimensional thermal distribution map within the current time window and the global cooling strategy command; ignoring temperature changes outside the boundary of the local three-dimensional mesh sub-model and treating the boundary temperature as constant; converting the power value of the cooling unit located within or within the effective range of the local three-dimensional mesh sub-model in the global cooling strategy command into the equivalent heat dissipation rate in the local three-dimensional mesh sub-model; and performing rapid iterative calculations on the local three-dimensional mesh sub-model using a time step larger than that of the thermal conduction simulation unit to obtain the verification three-dimensional thermal distribution map at the end of the shortened future time window.

7. A temperature-controlled cabinet cooling method for low-voltage security as described in claim 1, characterized in that, The process of collecting real-time temperature information at designated monitoring points inside the low-voltage security cabinet includes: simultaneously activating multiple temperature sensors arranged at different depths and heights inside the low-voltage security cabinet; driving the multiple temperature sensors with synchronous sampling pulses to obtain temperature readings at each monitoring point at the same physical moment; packaging the obtained temperature readings with the corresponding temperature sensor identifiers and transmitting them to the data processing node through a redundant data link to form a set of raw temperature information.

8. A temperature-controlled cabinet cooling system for low-voltage security, characterized in that, It includes a main controller and a distributed temperature sensing array and a controllable cooling unit array connected in communication with it, wherein the main controller is configured to perform a temperature-controlled cabinet cooling method for low-voltage security as described in any one of claims 1 to 7.

Images