Air inlet temperature processing method, device and equipment

Abstract

The embodiment of the application discloses an air inlet temperature processing method, device and equipment. The air inlet temperature sensor data obtained is dynamically corrected through a regression model, so that the determined air inlet temperature target value can still be close to the real value of the air inlet temperature at the same time under a complex external environment. The temperature change direction determined through the obtained multiple air inlet temperature target values determines the heat dissipation strategy, so that the finally determined heat dissipation strategy still has high reliability under a complex external environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of switch heat dissipation, and in particular to an air inlet temperature processing method, device and equipment. BACKGROUND

[0002] A large amount of heat is generated during high-speed operation of a switch. In order to ensure safe, stable and reliable operation of the switch, a good heat dissipation strategy is designed for the switch, which is an important guarantee for stable operation of the switch.

[0003] In a heat dissipation strategy based on an external temperature (i.e. an air inlet temperature) of the switch, a single correction or compensation method is usually used to simply correct or compensate data collected by an internal sensor of the switch to obtain an air inlet temperature target value used to determine the heat dissipation strategy. This results in a large deviation between the air inlet temperature target value and an actual external temperature of the switch in a complex external environment, thereby affecting the reliability of the heat dissipation strategy. SUMMARY

[0004] The embodiments of the present application aim to provide an air inlet temperature processing method, device and equipment, which can optimize the acquisition method of the air inlet temperature and improve the reliability of the heat dissipation strategy.

[0005] To solve the above technical problem, in a first aspect, the embodiments of the present application provide an air inlet temperature processing method, which comprises the following steps.

[0006] Collecting air inlet temperature sensor data inside a target device multiple times in a first time period to obtain air inlet temperature sensor data corresponding to each collection time;

[0007] Determining, according to each air inlet temperature sensor data, an air inlet temperature target value corresponding to each collection time by a target regression model, wherein the air inlet temperature target value is used to describe an external environment temperature in which the target device is located;

[0008] Determining, according to each air inlet temperature target value, a temperature change direction corresponding to each collection time in the first time period;

[0009] Determining a heat dissipation strategy according to multiple temperature change directions.

[0010] In a second aspect, the embodiments of the present application further provide an air inlet temperature processing device, which comprises the following modules.

[0011] A first collection module, configured to collect air inlet temperature sensor data inside a target device multiple times in a first time period to obtain air inlet temperature sensor data corresponding to each collection time;

[0012] The first processing module is configured to determine, according to the air inlet temperature sensor data, an air inlet temperature target value corresponding to each of the collection time points by a target regression model, the air inlet temperature target value being used to describe an external environment temperature in which the target device is located;

[0013] The second processing module is configured to determine, according to the air inlet temperature target values, a temperature change direction corresponding to each of the collection time points in the first time period;

[0014] The third processing module is configured to determine a heat dissipation strategy according to the temperature change directions.

[0015] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the air inlet temperature processing method according to the first aspect.

[0016] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program / instruction. The computer program / instruction is executed by a processor to implement the air inlet temperature processing method according to the first aspect.

[0017] In a fifth aspect, a computer program product is provided, which includes a computer program / instruction. The computer program / instruction is executed by a processor to implement the air inlet temperature processing method according to the first aspect.

[0018] As can be seen from the above technical solution, the air inlet temperature sensor data is dynamically corrected by the regression model, so that the determined air inlet temperature target value can still be close to the real value of the air inlet temperature at the same time under a complex external environment. The heat dissipation strategy is determined according to the temperature change directions determined by the obtained air inlet temperature target values, so that the finally determined heat dissipation strategy still has high reliability under a complex external environment. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0020] Figure 1 An implementation flowchart of an air inlet temperature processing method provided by the embodiments of the present application;

[0021] Figure 2 A schematic diagram of a system fan speed curve provided by the embodiments of the present application;

[0022] Figure 3 A flowchart of linear regression parameter calculation provided for an embodiment of the present application is shown in FIG. 1;

[0023] Figure 4 A distribution diagram of measured sample points provided for an embodiment of the present application is shown in FIG. 2;

[0024] Figure 5 A flowchart of determining a heat dissipation strategy based on an air inlet temperature provided for an embodiment of the present application is shown in FIG. 3;

[0025] Figure 6 A diagram of a sliding window provided for an embodiment of the present application is shown in FIG. 4;

[0026] Figure 7 A structural diagram of an air inlet temperature processing device provided for an embodiment of the present application is shown in FIG. 5;

[0027] Figure 8 A diagram of an electronic device provided for an embodiment of the present application is shown in FIG. 6;

[0028] Figure 9 A diagram of a computer readable storage medium provided for an embodiment of the present application is shown in FIG. 7. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0030] The terms "include" and "have" and any variations thereof in the specification and claims of the present application and the above-described drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can include steps or units not listed.

[0031] Switches, especially white-box switches, are important network devices in enterprise data centers, and are constantly transmitting and processing massive amounts of data. A large amount of heat is generated during the high-speed operation of the switches. In order to ensure the safe, stable and reliable operation of the switches, a good heat dissipation strategy for the switches becomes an important guarantee for the stable operation of the switches.

[0032] Among the current numerous heat dissipation strategies, the heat dissipation strategy based on the external temperature of the switch becomes the preferred heat dissipation design in the industry. In order to ensure the reliability, accuracy and real-time performance of the heat dissipation strategy, the accuracy of the temperature data used to determine the heat dissipation strategy must be ensured.

[0033] In the current heat dissipation strategy based on the temperature outside the switch (i.e., the temperature at the air inlet), the accuracy of the processing of the temperature sensor data collected by the internal sensor of the switch is often ignored.

[0034] In the related art, the correction or compensation of the temperature sensor data of the air inlet is relatively single. For example, a temperature sensor is arranged near the air inlet inside the switch, the temperature value of the air inlet is collected by the temperature sensor, a fixed correction value is set according to experience, and then the temperature sensor value is corrected by using the correction value to obtain the temperature data of the air inlet for determining the heat dissipation strategy. However, in actual application, the fixed correction value in the above temperature correction technology is difficult to meet the complex scene and is likely to have a correction deviation, so that the data collected by the internal sensor of the switch still deviates greatly from the actual temperature of the external environment after the simple correction or compensation, and the heat dissipation strategy is invalid, and the reliability and stability are poor.

[0035] In the related art, the correction or compensation of the temperature sensor data of the air inlet is relatively single. For example, a temperature sensor is arranged near the air inlet inside the switch, the temperature value of the air inlet is collected by the temperature sensor, a fixed correction value is set according to experience, and then the temperature sensor value is corrected by using the correction value to obtain the temperature data of the air inlet for determining the heat dissipation strategy. However, in actual application, the fixed correction value in the above temperature correction technology is difficult to meet the complex scene and is likely to have a correction deviation, so that the data collected by the internal sensor of the switch still deviates greatly from the actual temperature of the external environment after the simple correction or compensation, and the heat dissipation strategy is invalid, and the reliability and stability are poor.

[0036] To solve the above problems in the related art, the present application determines a prediction curve of the temperature of the air inlet of the switch based on a linear regression algorithm, so that the target value of the corrected temperature of the air inlet is as close as possible to the actual value of the temperature of the air inlet of the switch, and the target value of the temperature of the air inlet in the current time period (e.g., the first time period) is statistically analyzed to calculate the temperature change amplitude and the temperature change direction. According to the temperature change amplitude and the temperature change direction, the temperature hysteresis detection threshold is adaptively adjusted to determine the heat dissipation strategy, so that the stability, reliability, real-time performance and accuracy of the heat dissipation strategy can be considered at the same time when facing a complex external environment, and an effective guarantee is provided for the safe, reliable and stable operation of the switch.

[0037] The air inlet temperature processing method provided by the embodiments of the present application will be described in detail in combination with the drawings and some embodiments and application scenarios.

[0038] Firstly, see [the following] Figure 1 The diagram shown is an implementation flowchart of an air inlet temperature treatment method provided in this application embodiment. The method may include the following steps:

[0039] Step S101: Collect the air inlet temperature sensor data inside the target device multiple times within the first time period to obtain the corresponding air inlet temperature sensor data at each collection time.

[0040] The target device can be a white-box switch or other switch device that uses a heat dissipation strategy based on the external temperature of the switch.

[0041] In practical implementation, the sliding window method can be used to collect multiple air inlet temperature sensor data within the sliding window period (i.e., the first time period) corresponding to the current moment from the temperature sensor inside the target device, so as to determine the heat dissipation strategy based on the multiple air inlet temperature sensor data.

[0042] Step S102: Based on the data from each of the air inlet temperature sensors, determine the target air inlet temperature value corresponding to each acquisition time using a target regression model.

[0043] The target air inlet temperature value is used to describe the external ambient temperature of the target device.

[0044] In practical implementation, a regression algorithm can be used to pre-fit the correlation between the inlet temperature sensor data and the actual inlet temperature (i.e., the external ambient temperature) in the external environment where the target device is located, resulting in an inlet temperature prediction curve. Based on this inlet temperature prediction curve, a target regression model is determined so that the corrected inlet temperature output by the target regression model (i.e., the target inlet temperature value) can approximate the actual inlet temperature value as closely as possible. This improves the prediction accuracy of the inlet temperature under complex external environments, thereby enhancing the reliability of the heat dissipation strategy.

[0045] Step S103: Based on the target temperature values ​​of each air inlet, determine the direction of temperature change corresponding to each of the multiple sampling moments within the first time period.

[0046] In practice, by determining the direction of temperature change at each of the multiple acquisition moments within the first time period, the trend of external ambient temperature change within that first time period can be known (such as constant, continuous rise, continuous fall, or the existence of temperature abrupt change points). Based on this overall trend of external ambient temperature change, a heat dissipation strategy (such as fan speed) can be determined, which can further improve the accuracy and reliability of the heat dissipation strategy.

[0047] As one possible implementation, the temperature difference corresponding to each of the multiple sampling times can be determined by a third formula based on the target temperature values ​​of each of the air inlets.

[0048] The direction of temperature change corresponding to the temperature difference values ​​that are greater than zero among the multiple temperature difference values ​​is determined as an upward trend;

[0049] The direction of temperature change corresponding to the temperature difference values ​​that are less than zero among the multiple temperature difference values ​​is determined as a downward trend;

[0050] The third formula is characterized as follows:

[0051] Δx i =x i+1 -x i

[0052] Where, Δx i Let x represent the temperature difference at the i-th data collection time. i Let x represent the target inlet temperature value corresponding to the i-th data collection time. i+1 This represents the target inlet temperature value corresponding to the (i+1)th data collection time.

[0053] Step S104: Determine a heat dissipation strategy based on the multiple directions of temperature change.

[0054] In practical implementation, a heat dissipation strategy based on the inlet air temperature can be formulated according to, for example... Figure 2 The system fan speed control curves (temperature rise speed control curve and temperature fall speed control curve) are shown. These curves characterize the correlation between the fan speed duty cycle and the corrected inlet temperature (i.e., the inlet temperature output by the target regression model) under different external ambient temperature change trends. Optionally, when the external ambient temperature decreases, a heat dissipation strategy can be determined based on the temperature fall speed control curve and temperature hysteresis detection to further improve the reliability of the heat dissipation strategy.

[0055] Optionally, in one embodiment, when multiple temperature change directions are all upward, the fan speed can be adjusted to a set maximum speed to avoid overheating and damage to the target device; while when multiple temperature change directions show a downward trend, the fan speed can be adjusted according to the hysteresis temperature threshold corresponding to the first time period.

[0056] The hysteresis temperature threshold is determined based on the historical inlet temperature target value corresponding to each historical acquisition moment within the second time period. The second time period is the previous time period of the first time period. The two time periods and the first time period can be continuous in time to improve the real-time performance of the heat dissipation strategy.

[0057] As a possible implementation, the historical temperature difference value corresponding to each historical collection time in the second time period can be determined according to the historical inlet air temperature target value of each historical collection time. Specifically, in the second time period, the historical inlet air temperature sensor data of the target device can be collected according to a set time interval, the historical inlet air temperature target value can be determined by the target regression model according to the historical inlet air temperature sensor data, and the historical temperature difference value corresponding to each historical collection time can be determined according to the historical inlet air temperature target value corresponding to each historical collection time and the historical inlet air temperature target value corresponding to the next historical collection time of each historical collection time.

[0058] According to the plurality of historical temperature difference values, the historical temperature change direction corresponding to each of the plurality of historical collection times (which can be determined according to the positive and negative of the historical temperature difference value), the historical temperature change amplitude corresponding to each of the plurality of historical collection times (which can be determined according to the absolute value of the historical temperature difference value), and the historical temperature difference change average value can be determined.

[0059] Exemplarily, the historical temperature difference change average value can be determined by a second formula according to the plurality of historical temperature difference values.

[0060] The second formula is represented as follows:

[0061]

[0062] wherein, represents the historical temperature difference change average value, Δx i represents the historical temperature difference value corresponding to the i-th collection time, and m represents the number of historical temperature difference values.

[0063] In the case that there is a downward trend in the plurality of historical temperature change directions, and the historical temperature change amplitude corresponding to the downward trend is less than the historical hysteresis temperature threshold value corresponding to the second time period, it indicates that the decrease amplitude of the external environment temperature has changed, such as the decrease amplitude of the external environment temperature will become smaller when the central air conditioner is working, so the hysteresis temperature threshold value corresponding to the first period needs to be determined according to the historical temperature difference change average value and a set constant, to realize the adaptive adjustment of the hysteresis temperature threshold value, and avoid affecting the reliability of the heat dissipation strategy due to improper setting of the hysteresis temperature threshold value.

[0064] Exemplarily, the hysteresis temperature threshold value can be determined by a first formula according to the historical temperature difference change average value and the set constant.

[0065] The first formula is represented as follows:

[0066]

[0067] wherein, ΔT newrepresents a hysteresis temperature threshold value, represents a historical temperature difference change average value, and D represents a set constant.

[0068] In one embodiment, when all the temperature change directions are in a downward trend, the temperature difference values corresponding to each of the temperature change directions are determined, and the temperature difference change average value corresponding to the first time period is determined according to the temperature difference values.

[0069] It is detected whether the temperature difference change average value is greater than the hysteresis temperature threshold value corresponding to the first time period;

[0070] When the temperature difference change average value is not greater than the hysteresis temperature threshold value corresponding to the first time period, the fan speed is reduced to a first percentage of the current speed.

[0071] When the temperature difference change average value is greater than the hysteresis temperature threshold value corresponding to the first time period, the fan speed is reduced to a second percentage of the current speed, and the first percentage is greater than the second percentage. It can be understood that the adaptive adjustment of the hysteresis temperature threshold value can reduce the fan speed to different degrees for fans with excessively high speed, so as to reduce the power consumption of the fan while ensuring the heat dissipation effect of the target device.

[0072] In another embodiment, when there is a downward trend in a plurality of temperature change directions, and the temperature change amplitude corresponding to the downward trend meets a first set condition (i.e., there is a temperature mutation point), the target temperature change direction corresponding to the next collection time of the collection time corresponding to the downward trend is determined, that is, the temperature change direction after the temperature mutation point is determined, wherein the first set condition is that the temperature change amplitude is greater than a set multiple of the hysteresis temperature threshold value corresponding to the first time period, and the set multiple is greater than one.

[0073] When the target temperature change direction is an upward trend, the fan speed is adjusted to a set maximum speed to avoid overheating and damage to the target device.

[0074] When the target temperature change direction is a downward trend, a target time period after the collection time corresponding to the downward trend in the first time period is determined, and the fan speed is adjusted according to the target temperature difference change average value corresponding to the target time period and the hysteresis temperature threshold value corresponding to the first time period. It can be understood that if the temperature change direction after the temperature mutation point is still a downward trend, the fan can be reduced according to the hysteresis temperature threshold value.

[0075] Specifically, it can be detected whether the target temperature difference change average value is greater than the hysteresis temperature threshold corresponding to the first time period, and in the case that the target temperature difference change average value is not greater than the hysteresis temperature threshold corresponding to the first time period, the fan speed is reduced to a third percentage of the current speed; in the case that the target temperature difference change average value is greater than the hysteresis temperature threshold corresponding to the first time period, the fan speed is reduced to a fourth percentage of the current speed, and the third percentage is greater than the fourth percentage.

[0076] From the above technical solution, it can be seen that the obtained air inlet temperature sensor data is dynamically corrected by the regression model, so that the determined air inlet temperature target value can still approach the air inlet temperature true value at the same time under complex external environment, and the determined temperature change direction determined by the obtained multiple air inlet temperature target values determines the cooling strategy, so that the finally determined cooling strategy still has high reliability under complex external environment.

[0077] Optionally, in an embodiment, the target regression model can be adjusted or reconstructed in time to improve the air inlet temperature prediction accuracy under complex external environment every time the external environmental factors of the target device change, such as the device cabinet position, the device density in the cabinet, the working state of the data center machine room air conditioning control system, etc.

[0078] When adjusting or reconstructing the target regression model, as shown in Figure 3 , a plurality of groups of temperature data can be collected at different environmental temperatures first, and each group of temperature data contains the air inlet temperature sensor data inside the target device and the air inlet temperature true value outside the target device corresponding to the same collection time.

[0079] For example, the environmental temperature in the data center machine room can be adjusted, and 50 groups of internal and external air inlet temperature data of the switch are collected at different environmental temperatures at the same time, wherein the external air inlet temperature data is the ideal air inlet temperature of the switch (i.e. the air inlet temperature true value), and the internal air inlet temperature data is the measured air inlet temperature inside the switch (i.e. the air inlet temperature sensor data). In the 50 groups of temperature data, each group of data contains an ideal air inlet temperature of a switch and a corresponding measured air inlet temperature.

[0080] After a plurality of groups of temperature data are collected, a first regression model can be established, the values of the parameters in the first regression model are determined according to the plurality of groups of temperature data and the loss function corresponding to the first regression model, and then the target regression model can be determined according to the first regression model and the values of the parameters in the first regression model.

[0081] Specifically, a feature sample library can be established for the collected internal and external air inlet temperature data of the switch. Based on the data in the feature sample library, a linear regression method can be used to simulate the relationship between the measured internal air inlet temperature and the ideal air inlet temperature of the switch. At this time, the first regression model corresponding to the linear regression method can be represented as follows:

[0082] y = ax + b

[0083] where y represents the air inlet temperature target value, x represents the air inlet temperature sensor data, and a and b are the values of the parameters in the first regression model.

[0084] It can be understood that the first regression model uses a linear function to fit the relationship between the measured internal air inlet temperature and the ideal air inlet temperature of the switch. In actual application, the first regression model can also introduce multiple variables to fit the relationship between the measured internal air inlet temperature and the ideal air inlet temperature of the switch in the form of a plane function or a high-dimensional curve function.

[0085] To determine the values of the parameters in the first regression model, the mean square error can be obtained by dividing the sum of squared residuals by the sample size n. The mean square error is used as the loss function of the first regression model, and the calculation formula of the loss function J is as follows:

[0086]

[0087] As shown in FIG. 1, assuming that the best fitting straight line equation is: Figure 4

[0088] y = ax + b

[0089] then for each measured sample point (i.e., air inlet temperature sensor data) x i , the predicted value of the measured sample point determined according to the best fitting straight line equation is:

[0090]

[0091] The predicted value corresponds to the ideal true value (i.e., the true value of the air inlet temperature) y i . To make the predicted value as close as possible to the ideal true value, using the calculation formula of the loss function J, the loss function is:

[0092]

[0093] Taking the partial derivative of a and b, respectively, that is:

[0094]

[0095] ​

[0096] Available:

[0097]

[0098]

[0099] wherein, is the average value of the inlet air temperature sensor data collected at the nth time, is the average value of the true value of the inlet air temperature collected n times, n is the number of data groups in the characteristic sample library (such as 50). The obtained parameter values a and b are brought into the above-mentioned first regression model, and a target regression model is obtained.

[0100] The above-mentioned embodiments are further described below in combination with Figure 5 and Figure 6 .

[0101] After the target regression model is determined, the sliding window method is used to collect inlet air temperature sensor data in the switch at intervals of the same time Δt for n times, and the collected data is corrected by using the target regression model, so that the corrected data can approximate the true value of the inlet air temperature outside the switch as much as possible. In this example, Δt = 1 second and n = 10.

[0102] At this time, in a sliding window period (such as the first sliding window), the corrected inlet air temperature sequence is: {x1, x2, …, x 10}, and the temperature difference Δx i between adjacent two corrected inlet air temperature values is: Δx i = x i+1 - x i , (i = 1, 2, 3, … 9).

[0103] Taking the initial hysteresis temperature threshold ΔT = 3 as an example, the hysteresis detection execution steps are as follows:

[0104] (1) First, according to the temperature change direction and the temperature change amplitude in a sliding window period, the hysteresis temperature threshold ΔT of the next sliding window is adaptively adjusted. For example, in the current sliding window period, if Δx i <0 and abs(Δx i ) < ΔT, wherein abs(Δx i ) represents the absolute value of Δx i , then the new hysteresis temperature threshold ΔT new may be:

[0105]

[0106] (2) In a sliding window period, when all temperature difference Δx i = 0, it indicates that the temperature of the external environment of the switch is constant in this sliding window period, at this time, the fan speed based on the inlet temperature of the system remains unchanged (i.e. maintains the original cooling strategy), the fan body alarm light is green and no alarm log is output;

[0107] (3) In a sliding window period, when all temperature difference Δx i > 0, it indicates that the temperature of the external environment of the switch continues to rise in this sliding window period, at this time, the cooling strategy based on the inlet temperature of the switch has lagged behind the change of the temperature of the external environment of the switch (i.e. the current fan speed cannot meet the cooling demand), at this time, the system fan and the local fan of the power supply can be directly run at full speed (i.e. generate cooling strategy 1) and print alarm information to the operation and maintenance personnel at the same time.

[0108] (4) In a sliding window period, when all temperature difference Δx i < 0, it indicates that the temperature of the external environment of the switch continues to decline in this sliding window period, at this time, the cooling strategy based on the inlet temperature of the switch has led the change of the temperature of the external environment of the switch (i.e. the fan speed is too high relative to the current cooling demand), at this time, the average value of the temperature difference change in the current sliding window period is:

[0109]

[0110] If the average value of the temperature difference change is At this time, the system fan and the local fan of the power supply are adjusted to run at 80% of the current speed (i.e. generate cooling strategy 2) and print prompt information to the operation and maintenance personnel at the same time; if the average value of the temperature difference change is At this time, the system fan and the local fan of the power supply are adjusted to run at 70% of the current speed (i.e. generate cooling strategy 3) and print prompt information to the operation and maintenance personnel at the same time.

[0111] (5) In a sliding window period, if there is a temperature mutation point, i.e. there is a temperature difference Δx i < 0 and the corresponding temperature change amplitude abs(Δx i ) > 2ΔT.

[0112] If the temperature difference Δx i+1 > 0 after the temperature mutation point, i.e. the temperature change direction after the temperature mutation point is upward, at this time, the system fan and the local fan of the power supply are directly run at full speed and print alarm information to the operation and maintenance personnel at the same time.

[0113] If the temperature difference Δx i+1If the temperature change direction after the temperature mutation point is a downward trend, i.e. <0, the temperature difference change average value after the temperature mutation point is obtained by obtaining the temperature difference change average value of all temperature differences after the temperature mutation point in the sliding window period.

[0114] If the temperature difference change average value after the temperature mutation point is <0 At this time, the system fan and the power supply body fan are adjusted to run at 80% of the current speed, and a prompt information is printed to the operation and maintenance personnel; if the temperature difference change average value after the temperature mutation point is <0 At this time, the system fan and the power supply body fan are adjusted to run at 70% of the current speed, and a prompt information is printed to the operation and maintenance personnel.

[0115] Based on the above embodiment, the linear regression algorithm is used to generate a switch air inlet temperature prediction curve, the switch air inlet temperature sensor data is dynamically corrected based on the temperature prediction curve, the switch air inlet temperature sensor data is collected at the same time interval and corrected by using the sliding window method, the temperature hysteresis detection threshold is adaptively adjusted according to the temperature change amplitude and change direction in the sliding window, and the corresponding heat dissipation strategy is determined, so that when facing the complex and changeable external environment, the problems of poor reliability, low accuracy and poor real-time performance of the heat dissipation regulation and control based on the air inlet temperature in the current white box switch heat dissipation strategy can be solved. It can be understood that other regression algorithms can also be used to correct the switch air inlet temperature sensor data in specific implementation, and the hysteresis temperature threshold can also be dynamically adjusted by using other algorithms when detecting the hysteresis temperature to further improve the stability, real-time performance and accuracy of the heat dissipation strategy, which is not limited in the present application.

[0116] In a second aspect, the embodiments of the present application provide an air inlet temperature processing device, as shown in the accompanying drawings, which comprises: Figure 7

[0117] The first acquisition module 11 is configured to acquire the air inlet temperature sensor data of the target device inside at multiple times in a first time period to obtain the air inlet temperature sensor data corresponding to each acquisition time;

[0118] The first processing module 12 is configured to determine the air inlet temperature target value corresponding to each acquisition time by using a target regression model according to each air inlet temperature sensor data, and the air inlet temperature target value is used to describe the external environment temperature of the target device.

[0119] The second processing module 13 is configured to determine the temperature change direction corresponding to each acquisition time in the first time period according to each air inlet temperature target value.

[0120] ​The third processing module 14 is configured to determine a heat dissipation strategy according to the temperature change directions.

[0121] Optionally, the third processing module 14 includes:

[0122] The first adjusting module is configured to adjust the fan rotating speed to a set maximum rotating speed when all the temperature change directions are upward trends.

[0123] The second adjusting module is configured to adjust the fan rotating speed according to a hysteresis temperature threshold corresponding to the first time period when there is a downward trend in the temperature change directions, the hysteresis temperature threshold being determined according to historical inlet air temperature target values corresponding to respective historical collection time points in a second time period, the second time period being a previous time period of the first time period.

[0124] Optionally, the device further includes:

[0125] The fourth processing module is configured to determine historical temperature difference values corresponding to respective historical collection time points in the second time period according to the respective historical inlet air temperature target values.

[0126] The fifth processing module is configured to determine historical temperature change directions corresponding to the respective historical collection time points, historical temperature change amplitudes corresponding to the respective historical collection time points, and a historical temperature difference change average value according to the historical temperature difference values.

[0127] The sixth processing module is configured to determine the hysteresis temperature threshold according to the historical temperature difference change average value and a set constant when there is a downward trend in the historical temperature change directions and a historical temperature change amplitude corresponding to the downward trend is less than a historical hysteresis temperature threshold corresponding to the second time period.

[0128] Optionally, the sixth processing module includes:

[0129] The first processing submodule is configured to determine the hysteresis temperature threshold according to the historical temperature difference change average value and the set constant through a first formula.

[0130] The first formula is as follows:

[0131]

[0132] wherein ΔT new represents the hysteresis temperature threshold, represents the historical temperature difference change average value, and D represents the set constant.

[0133] Optionally, the device further includes:

[0134] The second collecting module is configured to collect each historical air inlet temperature sensor data inside the target device according to a set time interval in the second time period.

[0135] The seventh processing module is configured to determine each historical air inlet temperature target value according to each historical air inlet temperature sensor data by using the target regression model.

[0136] The fourth processing module comprises:

[0137] The second processing submodule is configured to determine a historical temperature difference value corresponding to each historical collection time according to a historical air inlet temperature target value corresponding to each historical collection time and a historical air inlet temperature target value corresponding to a next historical collection time of each historical collection time.

[0138] Optionally, the fifth processing module comprises:

[0139] The third processing submodule is configured to determine the historical temperature difference change average value by using a second formula according to a plurality of historical temperature difference values.

[0140] The second formula is represented as follows:

[0141]

[0142] wherein, represents the historical temperature difference change average value, Δx i represents an historical temperature difference value corresponding to an i-th collection time, and m represents a number of historical temperature difference values.

[0143] Optionally, the second adjusting module comprises:

[0144] The first determining module is configured to determine a temperature difference value corresponding to each temperature change direction when a plurality of temperature change directions are all in a downward trend.

[0145] The second determining module is configured to determine a temperature difference change average value corresponding to the first time period according to a plurality of temperature difference values.

[0146] The first detecting module is configured to detect whether the temperature difference change average value is greater than a hysteresis temperature threshold value corresponding to the first time period.

[0147] The first adjusting submodule is configured to reduce the fan rotating speed to a first percentage of a current rotating speed when the temperature difference change average value is not greater than the hysteresis temperature threshold value corresponding to the first time period.

[0148] The second adjusting sub-module is configured to reduce the fan rotating speed to a second percentage of the current rotating speed when the average temperature difference is greater than the hysteresis temperature threshold corresponding to the first time period, the first percentage being greater than the second percentage.

[0149] Optionally, the second adjusting module comprises:

[0150] The third determining module is configured to determine a target temperature change direction corresponding to a next collection time of a collection time corresponding to a downward trend when the downward trend exists in the multiple temperature change directions and a temperature change amplitude corresponding to the downward trend meets a first setting condition, the first setting condition being that the temperature change amplitude is greater than a set multiple of the hysteresis temperature threshold corresponding to the first time period, the set multiple being greater than one.

[0151] The fourth determining module is configured to adjust the fan rotating speed to a set maximum rotating speed when the target temperature change direction is an upward trend.

[0152] The third adjusting sub-module is configured to determine a target time period after a collection time corresponding to the downward trend within the first time period according to the target temperature change direction, and adjust the fan rotating speed according to a target average temperature difference corresponding to the target time period and the hysteresis temperature threshold corresponding to the first time period.

[0153] Optionally, the third adjusting sub-module comprises:

[0154] The second detecting module is configured to detect whether the target average temperature difference is greater than the hysteresis temperature threshold corresponding to the first time period.

[0155] The fourth adjusting sub-module is configured to reduce the fan rotating speed to a third percentage of the current rotating speed when the target average temperature difference is not greater than the hysteresis temperature threshold corresponding to the first time period.

[0156] The fifth adjusting sub-module is configured to reduce the fan rotating speed to a fourth percentage of the current rotating speed when the target average temperature difference is greater than the hysteresis temperature threshold corresponding to the first time period, the third percentage being greater than the fourth percentage.

[0157] Optionally, before determining the respective inlet air temperature target values corresponding to the respective collection times through the target regression model according to the respective inlet air temperature sensor data, the device further comprises:

[0158] The third collecting module is configured to collect a plurality of groups of temperature data under different ambient temperatures, each group of the temperature data comprising an inlet air temperature sensor data inside the target device and an inlet air temperature true value outside the target device corresponding to a same collection time point;

[0159] The first establishing module is configured to establish a first regression model, the first regression model being configured to represent a mapping relationship between the inlet air temperature sensor data and the inlet air temperature target value;

[0160] The eighth processing module is configured to determine values of parameters in the first regression model according to the plurality of groups of temperature data and a loss function corresponding to the first regression model;

[0161] The ninth processing module is configured to determine the target regression model according to the first regression model and the values of the parameters in the first regression model.

[0162] Optionally, the first regression model represents as follows:

[0163] y = ax + b

[0164] wherein y represents the inlet air temperature target value, x represents the inlet air temperature sensor data, and a and b are the values of the parameters in the first regression model.

[0165] Optionally, the second processing module 13 comprises:

[0166] The fourth processing submodule is configured to determine a temperature difference value corresponding to each of the plurality of collection time points according to each of the inlet air temperature target values by using a third formula.

[0167] The fifth processing submodule is configured to determine a temperature change direction corresponding to a temperature difference value greater than zero in the plurality of temperature difference values as an upward trend.

[0168] The sixth processing submodule is configured to determine a temperature change direction corresponding to a temperature difference value less than zero in the plurality of temperature difference values as a downward trend.

[0169] The third formula represents as follows:

[0170] Δx i = x i+1 - x i

[0171] wherein Δx i represents a temperature difference value corresponding to an i-th collection time point, x i represents an inlet air temperature target value corresponding to the i-th collection time point, and x i+1 represents an inlet air temperature target value corresponding to an (i+1)-th collection time point.

[0172] From the above technical solution can be seen, the obtained air inlet temperature sensor data is dynamically corrected by the regression model, so that the determined air inlet temperature target value can still be close to the air inlet temperature true value at the same time under the complex external environment, and the cooling strategy is determined by the temperature change direction determined by the obtained multiple air inlet temperature target values, so that the finally determined cooling strategy still has high reliability under the complex external environment.

[0173] It should be noted that the device embodiment is similar to the method embodiment, and therefore the description is relatively simple, and the relevant parts can be referred to the method embodiment.

[0174] The embodiment of the present application further provides an electronic device, which refers to Figure 8 , Figure 8 is a schematic diagram of an electronic device according to the embodiment of the present application. As shown in Figure 8 , the electronic device 100 comprises a memory 110 and a processor 120, the memory 110 and the processor 120 are connected by a bus communication, and the memory 110 stores a computer program, the computer program can run on the processor 120, and then realize the steps in the air inlet temperature processing method disclosed by the embodiment of the present application.

[0175] The embodiment of the present application further provides a computer readable storage medium, which refers to Figure 9 , Figure 9 is a schematic diagram of a computer readable storage medium according to the embodiment of the present application. As shown in Figure 9 , the computer readable storage medium 200 stores a computer program / instruction 210, and the computer program / instruction 210 is executed by the processor to realize the steps in the air inlet temperature processing method disclosed by the embodiment of the present application.

[0176] The embodiment of the present application further provides a computer program product, which comprises a computer program / instruction, and the computer program / instruction is executed by the processor to realize the steps in the air inlet temperature processing method disclosed by the embodiment of the present application.

[0177] Each embodiment in the present specification adopts a progressive manner for description, and each embodiment focuses on the different parts from other embodiments, and the same and similar parts between each embodiment can be referred to each other.

[0178] Those skilled in the art will appreciate that embodiments of the present application can be readily used as a method, apparatus or computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer readable storage media (including, but not limited to, disk memory, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0179] Embodiments of the present application are described herein with reference to the Figure 1 one or more functions specified in a flow or multiple flows and / or blocks. Figure 1 an apparatus that performs the functions specified in a flow or multiple flows and / or blocks.

[0180] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the Figure 1 one or more functions specified in a flow or multiple flows and / or blocks. Figure 1 an apparatus that performs the functions specified in a flow or multiple flows and / or blocks.

[0181] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in a flow or multiple flows and / or blocks. Figure 1 one or more functions specified in a flow or multiple flows and / or blocks. Figure 1 Figure 1 an apparatus that performs the functions specified in a flow or multiple flows and / or blocks.

[0182] Although preferred embodiments of the present application have been described, those skilled in the art will appreciate that additional modifications and alterations can be made thereto without departing from the scope of the present application. Accordingly, the appended claims are intended to cover all such modifications and alterations as fall within the scope of the present application.

[0183] Finally, it is to be understood that the phraseology or terminology such as "first" and "second" etc. used herein is merely intended to differentiate one entity or operation from another entity or operation, without necessarily requiring or implying any actual such relationship or order between such entities or operations. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0184] The above provides a kind of air inlet temperature processing method, device and equipment provided by the present application, have carried out detailed introduction, the principle and implementation mode of the present application are described in this paper by specific example, the above example is only for helping to understand the method and its core idea of the present application;For the general technical personnel in the art, according to the idea of the present application, there will be changes in specific implementation mode and application range, as described above, the content of the specification should not be understood as the limitation of the present application.

Claims

1. An air intake temperature treatment method characterized by, The method comprises: collecting air inlet temperature sensor data inside the target device multiple times in a first time period to obtain air inlet temperature sensor data corresponding to each collection time; determining, according to each air inlet temperature sensor data, a target air inlet temperature value corresponding to each collection time through a target regression model, the target air inlet temperature value being used to describe an external environment temperature in which the target device is located; determining, according to each target air inlet temperature value, a temperature change direction corresponding to each collection time in the first time period; determining a heat dissipation strategy according to multiple temperature change directions; determining a heat dissipation strategy according to multiple temperature change directions, comprising: in the case that all of the multiple temperature change directions are upward trends, adjusting a fan rotating speed to a set maximum rotating speed; in the case that there is a downward trend in the multiple temperature change directions, adjusting the fan rotating speed according to a temperature difference value corresponding to each temperature change direction and a hysteresis temperature threshold corresponding to the first time period, the hysteresis temperature threshold being determined according to a historical temperature difference value corresponding to each historical collection time in a second time period, the historical temperature difference value being determined according to each historical target air inlet temperature value, the second time period being a previous time period of the first time period.

2. The method of claim 1, wherein, The hysteresis temperature threshold is determined by the following steps: determining, according to each historical target air inlet temperature value, a historical temperature difference value corresponding to each historical collection time in the second time period; determining, according to multiple historical temperature difference values, a historical temperature change direction corresponding to each historical collection time, a historical temperature change amplitude corresponding to each historical collection time, and a historical temperature difference change average value; in the case that there is a downward trend in the multiple historical temperature change directions and a historical temperature change amplitude corresponding to the downward trend is less than a historical hysteresis temperature threshold corresponding to the second time period, determining the hysteresis temperature threshold according to the historical temperature difference change average value and a set constant.

3. The method of claim 2, wherein, Determining the hysteresis temperature threshold according to the historical temperature difference change average value and a set constant, comprising: determining the hysteresis temperature threshold through a first formula according to the historical temperature difference change average value and a set constant; The first formula is represented as follows: wherein, represents a hysteresis temperature threshold, represents a historical temperature difference change average, represents a set constant.

4. The method of claim 2, wherein, Each historical target air inlet temperature value corresponding to each historical collection time in the second time period is determined by the following steps: collecting each historical air inlet temperature sensor data inside the target device according to a set time interval in the second time period; determining each historical target air inlet temperature value through the target regression model according to each historical air inlet temperature sensor data; Each historical temperature difference value corresponding to each historical collection time is determined by the following steps: determining a historical temperature difference value corresponding to each historical collection time according to a historical target air inlet temperature value corresponding to each historical collection time and a historical target air inlet temperature value corresponding to a next historical collection time of each historical collection time.

5. The method of claim 4, wherein, The historical temperature difference change average value is determined by the following steps: The historical temperature difference change average value is determined by a second formula according to the plurality of historical temperature difference values; The second formula is characterized as follows: wherein, represents the average value of the historical temperature difference change, represents the historical temperature difference value corresponding to the i-th collection time point, represents the number of historical temperature difference values.

6. The method of claim 1, wherein, In the case that there is a downward trend in the plurality of temperature change directions, the fan speed is adjusted according to the temperature difference value corresponding to each of the plurality of temperature change directions and the hysteresis temperature threshold corresponding to the first time period, comprising: In the case that all of the plurality of temperature change directions are downward trends, the temperature difference value corresponding to each of the plurality of temperature change directions is determined; According to the plurality of temperature difference values, the average temperature difference change value corresponding to the first time period is determined; It is detected whether the average temperature difference change value is greater than the hysteresis temperature threshold corresponding to the first time period; In the case that the average temperature difference change value is not greater than the hysteresis temperature threshold corresponding to the first time period, the fan speed is reduced to a first percentage of the current speed; In the case that the average temperature difference change value is greater than the hysteresis temperature threshold corresponding to the first time period, the fan speed is reduced to a second percentage of the current speed, the first percentage being greater than the second percentage.

7. The method of claim 1, wherein, In the case that there is a downward trend in the plurality of temperature change directions, the fan speed is adjusted according to the temperature difference value corresponding to each of the plurality of temperature change directions and the hysteresis temperature threshold corresponding to the first time period, comprising: In the case that there is a downward trend in the plurality of temperature change directions, and the temperature change amplitude corresponding to the downward trend meets a first set condition, the target temperature change direction corresponding to a next collection time of the collection time corresponding to the downward trend is determined, the first set condition being that the temperature change amplitude is greater than a set multiple of the hysteresis temperature threshold corresponding to the first time period, the set multiple being greater than one; In the case that the target temperature change direction is an upward trend, the fan speed is adjusted to a set maximum speed; In the case that the target temperature change direction is a downward trend, a target time period after the collection time corresponding to the downward trend within the first time period is determined, and the fan speed is adjusted according to the target temperature difference change average value corresponding to the target time period and the hysteresis temperature threshold corresponding to the first time period.

8. The method of claim 7, wherein, Adjusting the fan speed according to the target temperature difference change average value corresponding to the target time period and the hysteresis temperature threshold corresponding to the first time period, comprising: It is detected whether the target temperature difference change average value is greater than the hysteresis temperature threshold corresponding to the first time period; In the case that the target temperature difference change average value is not greater than the hysteresis temperature threshold corresponding to the first time period, the fan speed is reduced to a third percentage of the current speed; In the case that the target temperature difference change average value is greater than the hysteresis temperature threshold corresponding to the first time period, the fan speed is reduced to a fourth percentage of the current speed, the third percentage being greater than the fourth percentage.

9. The method of claim 1, wherein, Before determining the inlet air temperature target value corresponding to each of the collection times by the target regression model according to each of the inlet air temperature sensor data, the method further comprises: Collecting multiple groups of temperature data under different ambient temperatures, each group of the temperature data containing an air inlet temperature sensor data inside the target device and an air inlet temperature true value outside the target device corresponding to the same collection time; Establishing a first regression model, the first regression model being used to represent a mapping relationship between the air inlet temperature sensor data and the air inlet temperature target value; Determining values of parameters in the first regression model according to the multiple groups of temperature data and a loss function corresponding to the first regression model; Determining the target regression model according to the first regression model and the values of the parameters in the first regression model.

10. The method of claim 9, wherein, The first regression model represents as follows: y=ax+b wherein y represents the air inlet temperature target value, x represents the air inlet temperature sensor data, and a and b are values of the parameters in the first regression model.

11. The method according to any one of claims 1 to 10, characterized in that, Determining temperature change directions corresponding to multiple collection times in the first time period according to each of the air inlet temperature target values, comprising: Determining temperature difference values corresponding to the multiple collection times according to each of the air inlet temperature target values through a third formula; Determining the temperature change direction corresponding to a temperature difference value greater than zero among the multiple temperature difference values as an upward trend; Determining the temperature change direction corresponding to a temperature difference value less than zero among the multiple temperature difference values as a downward trend; The third formula represents as follows: = wherein, represents the temperature difference value corresponding to the i-th collection time point, represents the inlet air temperature target value corresponding to the i-th collection time point, represents the inlet air temperature target value corresponding to the (i+1)-th collection time point.

12. An air intake temperature treatment device, characterized by, The device comprises: A first collection module, configured to collect air inlet temperature sensor data inside a target device multiple times in a first time period to obtain air inlet temperature sensor data corresponding to each collection time; A first processing module, configured to determine air inlet temperature target values corresponding to each of the collection times through a target regression model according to each of the air inlet temperature sensor data, the air inlet temperature target value being used to describe an external ambient temperature in which the target device is located; A second processing module, configured to determine temperature change directions corresponding to multiple collection times in the first time period according to each of the air inlet temperature target values; A third processing module, configured to determine a heat dissipation strategy according to the multiple temperature change directions; The third processing module comprises: A first adjustment module, configured to adjust a fan speed to a set maximum speed in a case where the multiple temperature change directions are all upward trends; A second adjustment module, configured to adjust the fan speed according to temperature difference values corresponding to the multiple temperature change directions and a hysteresis temperature threshold corresponding to the first time period in a case where there is a downward trend among the multiple temperature change directions, the hysteresis temperature threshold being determined according to historical temperature difference values corresponding to each of historical collection times in a second time period, the historical temperature difference value being determined according to each of historical air inlet temperature target values, and the second time period being a previous time period of the first time period.

13. An electronic device comprising a memory, a processor, and a computer program stored on the memory, wherein the computer program comprises instructions that, when executed by the processor, cause the electronic device to perform the method of any one of claims 1-12. The processor executes the computer program to implement the air inlet temperature processing method according to any one of claims 1 to 11.

Images