Warm water circulation energy-saving heat dissipation device and heat dissipation system of high-performance server

Abstract

The invention relates to the technical field of non-electrical variable control, in particular to a warm water circulation energy-saving heat dissipation device and system of a high-performance server, and the device comprises a heat dissipation data collection module which is used for collecting temperature data and power consumption data of the server in real time, and collecting various parameter data of the heat dissipation device in real time; the heat dissipation data analysis module is used for comprehensively analyzing all data in a preset time period before the current adjustment moment and judging whether to adjust the PID control parameters at the current adjustment moment, and comprises the following steps: acquiring a change index, power consumption mutability, a temperature change factor, a temperature change coefficient and a temperature adjustment index at the current adjustment moment; whether the PID control parameters are adjusted at the current adjusting moment or not is judged; and the control parameter adjusting module is used for adjusting the PID control parameters at the current adjusting moment. The invention aims to adjust the PID control parameters according to the actual situation and improve the adjustment precision.

Description

Technical Field

[0001] The present application relates to the technical field of control of non-electrical variables, and in particular to a warm water circulation energy-saving heat dissipation device and a heat dissipation system for a high-performance server. Background Art

[0002] With the development of information technology, the performance and density of data center servers have been continuously improved, resulting in higher heat density. Liquid cooling technology has gradually attracted widespread attention due to its energy-saving, high-efficiency and quiet characteristics. In the process of using liquid cooling technology to dissipate heat from the server, a PID (Proportional Integral Derivative) controller is used to maintain the operating temperature of server components (such as CPU, GPU) within a preset safety range. According to the actual temperature feedback from the sensor, the PID controller dynamically adjusts the pump speed, fan speed and valve opening to ensure that the water flow rate and temperature are suitable for the current load conditions.

[0003] In the process of adjusting the temperature of the server by the warm water circulation energy-saving cooling system, considering the highly dynamic nature of the server load, the traditional fixed PID control parameters are not sufficient to cope with all situations, resulting in poor temperature regulation effect. Therefore, the PID control parameters are dynamically adjusted according to the actual situation during the temperature regulation process. Summary of the invention

[0004] In view of the above, it is necessary to provide a warm water circulation energy-saving heat dissipation device and heat dissipation system for high-performance servers. Compared with the traditional heat dissipation device of high-performance servers, while ensuring the heat dissipation effect, the parameters of the heat dissipation device are dynamically adjusted according to the actual heat dissipation requirements to avoid unnecessary energy waste: In a first aspect, an embodiment of the present application provides a warm water circulation energy-saving heat dissipation device for a high-performance server, the device comprising: The heat dissipation data collection module is used to collect the temperature data and power consumption data of the server in real time, and collect various parameter data of the heat dissipation device in real time, including the water inlet temperature and water outlet temperature of the heat dissipation device, and the water flow rate in the water pipe of the heat dissipation device; The heat dissipation data analysis module is used to preset the adjustment cycle of the PID control parameters, obtain the current adjustment time of the PID control parameters, analyze all data within the preset time period, and determine whether to adjust the PID control parameters at the current adjustment time. The process is: Based on the similarity of the change between all parameter data except the water inlet temperature and the water outlet temperature of the heat sink and all power consumption data, the change index at the current adjustment moment is obtained; Based on the mutation of all power consumption data, obtain the power consumption mutation at the current adjustment moment; Based on the difference between the water inlet temperature and the water outlet temperature at each collection moment, combined with the water flow rate at each collection moment, the temperature change factor at each collection moment is obtained; based on the discrete degree of all temperature data and the correlation between all power consumption data and all temperature change factors, the temperature change coefficient at the current adjustment moment is obtained; Based on the power consumption mutation, the change index and the temperature change coefficient, obtaining the temperature adjustment index at the current adjustment moment, and determining whether to adjust the PID control parameters at the current adjustment moment; The control parameter adjustment module is used to adjust the PID control parameters. Based on the similarity of all data except temperature data between each historical adjustment moment and the current adjustment moment, and the similarity of the PID control parameters, the reference adjustment moments of the current adjustment moment are obtained, and the PID control parameters of the next moment of each reference adjustment moment are adjusted in combination with the temperature adjustment index.

[0005] In one embodiment, the various parameter data of the heat dissipation device also include the pumping speed of a water pump in the heat dissipation device, the opening of a valve for adjusting the water flow rate, and the rotation speed of a heat dissipation fan.

[0006] In one embodiment, the expression of the variation index is: ; Wherein, Fr represents the change index at the current adjustment moment; S represents the number of all power consumption data within the preset time period; Indicates the difference in power consumption data between the qth collection moment and the q-1th collection moment within the preset time period; represents the difference in power consumption data between the qth collection moment and the q+1th collection moment within the preset time period; the pump speed, rotation speed, opening degree and water flow speed at each collection moment form a heat dissipation parameter vector, It represents the distance of the heat dissipation parameter vector between the qth collection moment and the q+1th collection moment within the preset time period; ε represents a preset value greater than 0.

[0007] In one embodiment, the process of obtaining the sudden change in power consumption is as follows: The difference in power consumption data between each acquisition moment and its adjacent acquisition moment is recorded as the power consumption difference at each acquisition moment, a threshold segmentation method is used to obtain a segmentation threshold of the power consumption difference at all acquisition moments, and the power consumption difference greater than the segmentation threshold is recorded as high difference data; The expression of power consumption mutation at the current adjustment moment is: ; Wherein, Tn represents the power consumption mutation at the current adjustment moment; A represents the mean of all high difference data; M represents the number of high difference data; S represents the number of all power consumption data within the preset time period; Indicates the time interval between the i-th high difference data and the i+1-th high difference data; Represents the average value of the time interval between any two adjacent high-difference data; wherein the high-difference data are arranged in time sequence.

[0008] In one embodiment, the process of obtaining the temperature change factor is: The difference between the water inlet temperature and the water outlet temperature at each sampling moment is recorded as the temperature difference; The temperature variation factor is the product of the temperature difference at each collection moment and the water flow velocity.

[0009] In one embodiment, the process of obtaining the temperature variation coefficient is as follows: Calculate the reciprocal of the sum of the discrete degree and a preset value greater than 0; Calculating the correlation coefficients between all power consumption data and all temperature change factors within the preset time period; The temperature variation coefficient is the product of the reciprocal and the correlation coefficient.

[0010] In one embodiment, the expression of the temperature adjustment index is: ; In the formula, Bc represents the temperature adjustment index at the current adjustment moment; Tn represents the power consumption mutation at the current adjustment moment; Fr represents the change index at the current adjustment moment; H represents the temperature change coefficient at the current adjustment moment; norm( ) represents the normalization operation; β represents a preset constant greater than 0.

[0011] In one of the embodiments, based on the temperature adjustment index, it is determined whether the PID control parameters need to be adjusted at the current adjustment moment. The specific method is: when the temperature adjustment index is less than or equal to a preset adjustment threshold, the PID control parameters at the current adjustment moment are not adjusted; otherwise, the PID control parameters at the current adjustment moment are adjusted.

[0012] In one embodiment, the process of adjusting the PID control parameters at the current adjustment time is: The PID control parameters, pump speed, rotation speed, opening, water flow rate, water inlet temperature, water outlet temperature and power consumption data at each adjustment moment are combined into a control vector, and all adjustment moments are clustered according to the distance between the control vectors; The remaining adjustment moments in the same cluster as the current adjustment moment are recorded as reference adjustment moments of the current adjustment moment; ; In the formula, represents the adjusted PID control parameter at the current adjustment time; Bc represents the temperature adjustment index at the current adjustment time; Pi represents the PID control parameter at the current adjustment time; N represents the total number of reference adjustment times at the current adjustment time; Indicates the PID control parameters at the next moment of the u-th reference adjustment moment of the current adjustment moment.

[0013] In the second aspect, an embodiment of the present application also provides a warm water circulation energy-saving heat dissipation system for a high-performance server, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, it is used to implement the above-mentioned warm water circulation energy-saving heat dissipation device for a high-performance server.

[0014] This application has at least the following beneficial effects: This application analyzes whether various parameter data of the heat dissipation device have changed accordingly when the power consumption data of the server changes, and obtains a change index to reflect the regulation hysteresis of the PID control parameters; Furthermore, by obtaining the sudden change of power consumption at the current adjustment moment, abnormal changes in power consumption can be discovered in time; by obtaining the temperature change factor through the difference between the water inlet temperature and the water outlet temperature, combined with the water flow rate, the heat dissipation capacity of the heat dissipation device is analyzed to provide a basis for whether the PID control parameters are adjusted; based on the discrete degree of temperature data and combined with the correlation between power consumption data and temperature change factor, the temperature change coefficient is obtained to analyze the adjustment ability of the PID control parameters, which helps to understand the effectiveness of the current PID control parameters in regulating the parameters of the heat dissipation device; and then comprehensively considering the change index, power consumption sudden change and temperature change coefficient to determine whether the current PID control parameters need to be adjusted, so that the adjustment can be more in line with actual needs. If the parameters of the heat dissipation device are adjusted according to the current PID control parameters, resulting in poor heat dissipation effect, the PID control parameters are adjusted, and then the fan speed, pump speed and other parameters are adjusted to avoid unnecessary energy waste and achieve more efficient and energy-saving heat dissipation control. However, if the parameters of the heat dissipation device are adjusted according to the current PID control parameters, the parameters of the heat dissipation device can adapt to the power consumption changes of the server, and then the PID control parameters are not adjusted, thereby avoiding unnecessary adjustment operations and saving time and resources. Furthermore, when adjusting the PID control parameters, not only the temperature adjustment index at the current adjustment moment is considered, but also the adjustment moments with similar PID control parameters and heat dissipation device parameters as the current adjustment moment are combined, so that the PID control parameters can be adjusted more finely and the adjustment accuracy can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A block diagram of a warm water circulation energy-saving heat dissipation device for a high-performance server provided in one embodiment of the present application; Figure 2 It is a schematic diagram of the process of obtaining the temperature regulation index; Figure 3 A flow chart for determining whether to adjust PID control parameters. DETAILED DESCRIPTION

[0017] In the description of the embodiments of the present application, words such as "exemplary", "or", "for example" and the like are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "or", "for example" and the like is intended to present related concepts in a concrete manner.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the present application. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. It should be understood that, unless otherwise specified, " / " means or.

[0019] It should also be noted that the terms "first" and "second" in the present application are used to distinguish similar objects rather than to describe a specific order or sequence.

[0020] The specific scheme of a warm water circulation energy-saving heat dissipation device for a high-performance server provided by the present application is described in detail below with reference to the accompanying drawings.

[0021] See also Figure 1 , which shows a block diagram of a warm water circulation energy-saving heat dissipation device for a high-performance server provided by an embodiment of the present application. The device includes: a heat dissipation data acquisition module 101, a heat dissipation data analysis module 102, and a control parameter adjustment module 103.

[0022] The heat dissipation data collection module 101 is used to collect the temperature data and power consumption data of the server in real time, and collect various parameter data of the heat dissipation device in real time, including the water inlet temperature and outlet temperature of the heat dissipation device, and the water flow rate in the water pipe of the heat dissipation device.

[0023] The heat dissipation device mainly includes a water pump, a cooling fan, a radiator, a water pipe and a water tank. A speed sensor is installed on the motor of the water pump to collect the pump speed in real time; a flow meter is installed in the water pipe to collect the water flow speed in real time; a temperature sensor is installed at the inlet of the pipe where water enters the heat dissipation device to collect the water inlet temperature in real time; a temperature sensor is installed at the outlet of the pipe where water flows out of the heat dissipation device to collect the water outlet temperature in real time; for the valve used to adjust the water flow speed in the heat dissipation device, the valve opening is collected in real time through a position sensor; and a photoelectric sensor is used to collect the speed of the cooling fan in real time. The temperature and power consumption of the server are obtained through the server management interface.

[0024] Real-time collection of server temperature data, server power consumption data, water inlet and outlet temperatures of the heat sink, and multiple parameter data of the heat sink, including the water flow rate in the water pipe, the pump speed of the water pump, the opening of the valve that adjusts the water flow rate, and the speed of the cooling fan.

[0025] In this embodiment, the collection frequency of temperature data, power consumption data, water inlet temperature, water outlet temperature, water flow velocity, pump speed, opening degree and rotation speed is once per second. The value of the collection frequency is preset manually and can be set by the implementer. This application does not impose any special restrictions.

[0026] In this embodiment, the adjustment period of the control parameters of the PID control algorithm is W1 minutes, and the adjustment periods of the pump speed, rotation speed, opening and water flow rate of the heat dissipation device are all W2 minutes, wherein the values ​​of W1 and W2 are preset manually and can be set by the implementer. In this embodiment, the values ​​of W1 and W2 are 10 and 1 respectively.

[0027] The heat dissipation data analysis module 102 is used to preset the adjustment cycle of the PID control parameters, obtain the current adjustment time of the PID control parameters, comprehensively analyze all data in the preset time period before the current adjustment time, and determine whether to adjust the PID control parameters at the current adjustment time.

[0028] Since the control parameters of the PID control algorithm are mainly adjusted through residuals, there is a certain hysteresis in the adjustment process. In the process of adjusting the parameters of the heat sink by the PID control parameters, if the difference between the server power consumption at the previous moment and the server power consumption at the current moment is large, and the difference between the server power consumption at the current moment and the server power consumption at the next moment is small, if the adjustment result of the PID control parameters is reliable, the difference between the various heat sink parameters of the heat sink between the current moment and the next moment is small. Therefore, this application analyzes the hysteresis of PID regulation.

[0029] The time when the control parameters of the PID control algorithm are adjusted is recorded as the adjustment time. All data in the preset time period before the current adjustment time are comprehensively analyzed to determine whether to adjust the PID control parameters at the current adjustment time. The specific process is as follows: Based on the similarity of the changes between all parameter data except the water inlet temperature and water outlet temperature of the heat sink and all power consumption data, the change index at the current adjustment moment is obtained, and the expression is: ; Wherein, Fr represents the change index at the current adjustment moment; S represents the number of all power consumption data within the preset time period; Indicates the difference in power consumption data between the qth collection moment and the q-1th collection moment within the preset time period; represents the difference in power consumption data between the qth collection moment and the q+1th collection moment within the preset time period; the pump speed, rotation speed, opening degree and water flow speed at each collection moment form a heat dissipation parameter vector, It represents the distance of the heat dissipation parameter vector between the qth collection moment and the q+1th collection moment within the preset time period; ε represents a preset value greater than 0, which is used to avoid the denominator being 0. The value of ε is preset manually and can be set by the implementer. In this embodiment, the value of ε is 0.01.

[0030] In this embodiment, the preset time period is the time period between the current adjustment moment and its adjacent previous adjustment moment.

[0031] In this embodiment, the distance between the heat dissipation parameter vectors is the Euclidean distance. As other implementation methods, on the basis of being able to measure the distance between the heat dissipation parameter vectors, the implementer may adopt other existing technologies for measurement, such as Manhattan distance, cosine distance, etc., and this application does not impose any special restrictions.

[0032] In this embodiment, in the process of calculating the change index, the differences involved are all absolute values ​​of the difference. As other implementation methods, on the basis of being able to measure the difference between the two power consumption data, the implementer can adopt other calculation methods, such as ratio, square of the difference, etc., and this application does not impose any special restrictions.

[0033] It should be noted that: if the difference between the power consumption data at each collection moment and its previous collection moment is greater, and the difference between the power consumption data at each collection moment and its subsequent collection moment is smaller, and at the same time, the difference between the various heat dissipation device parameters at each collection moment and its subsequent collection moment is smaller, then the changes in the power consumption data and the various heat dissipation device parameters are more similar, indicating that the regulation lag of the PID control parameters at the current adjustment moment is weaker.

[0034] The stronger the adjustment hysteresis of the PID control parameters, the more it is necessary to adjust the PID control parameters. Therefore, it is necessary to analyze the PID control parameters to obtain the necessity of their adjustment. In the PID control process, when the stability of the server power consumption change is stronger, the impact of the adjustment hysteresis of the PID control parameters is smaller. Therefore, it is necessary to analyze the power change characteristics of the server to determine the impact of the adjustment hysteresis of the PID control parameters.

[0035] Based on the above analysis, the difference in power consumption data between each acquisition moment and its adjacent acquisition moment is recorded as the power consumption difference at each acquisition moment. The threshold segmentation method is used to obtain the segmentation threshold of the power consumption difference at all acquisition moments, and the power consumption difference greater than the segmentation threshold is recorded as high difference data.

[0036] In this embodiment, the difference between the power consumption data of each acquisition moment and its adjacent previous acquisition moment is recorded as the power consumption difference of each acquisition moment. The difference between the power consumption data is the absolute value of the difference. As other implementation methods, on the basis of being able to measure the difference between the power consumption data, the implementer can use other calculation methods, such as the square of the difference, etc., and this application does not make any special restrictions.

[0037] In this embodiment, the Otsu threshold segmentation algorithm is used to obtain the segmentation threshold of the power consumption difference at all acquisition moments. The Otsu threshold segmentation algorithm is a well-known technology and will not be described in detail in this application. As other implementation methods, the implementer may use other existing algorithms to obtain the segmentation threshold of the power consumption difference at all acquisition moments, such as global threshold segmentation, iterative threshold segmentation, etc. This application does not impose any special restrictions.

[0038] Based on the mutation of all power consumption data, the power consumption mutation at the current adjustment moment is obtained. The expression is: ; Wherein, Tn represents the power consumption mutation at the current adjustment moment; A represents the mean of all high difference data; M represents the number of high difference data; S represents the number of all power consumption data within the preset time period; Indicates the time interval between the i-th high difference data and the i+1-th high difference data; Represents the average value of the time interval between any two adjacent high-difference data; wherein the high-difference data are arranged in time sequence.

[0039] It should be noted that: when the difference in power consumption of the server at adjacent collection moments is smaller, and the proportion of large changes in the power consumption of the server is smaller, the regularity of the change in the power consumption of the server is stronger, which means that the mutation of the power consumption change of the server is weaker, the influence of the regulation lag of the PID control parameters on the adjustment lag of the heat dissipation device parameters is smaller, and there is less need to adjust the PID control parameters.

[0040] All temperature data within the preset time period are arranged in time sequence to form a temperature sequence; all power consumption data within the preset time period are arranged in time sequence to form a power consumption sequence. Then, the temperature variation coefficient is analyzed according to the temperature sequence and the power consumption sequence.

[0041] First, the temperature stability of the server is calculated. The specific calculation method is: calculate the discreteness of all data in the temperature sequence, and calculate the inverse of the sum of the discreteness and the value α preset to be greater than 0; the larger the inverse, the stronger the temperature stability of the server. The purpose of adding α is to avoid the denominator being 0. The value of α is preset by humans and can be set by the implementer. In this embodiment, the value of α is 0.01.

[0042] In this embodiment, the degree of discreteness of all data in the temperature sequence is the variance. As other implementation methods, on the basis of being able to measure the uneven distribution of all data in the temperature sequence, the implementer may adopt other existing technologies for measurement, such as standard deviation, coefficient of variation, etc., and this application does not impose any special restrictions.

[0043] The difference between the water inlet temperature and the water outlet temperature at each collection moment is recorded as the temperature difference, and the product of the temperature difference at each collection moment and the water flow velocity is used as the temperature change factor at each collection moment. The greater the difference between the water inlet temperature and the water outlet temperature and the greater the water flow velocity, the greater the heat dissipation capacity of the heat dissipation device.

[0044] In this embodiment, the difference between the water inlet temperature and the water outlet temperature is the absolute value of the difference. On the basis of being able to measure the difference between the water inlet temperature and the water outlet temperature, the implementer may adopt other calculation methods, such as the square of the difference, the ratio, etc., and this application does not impose any special restrictions.

[0045] The temperature change factors of all acquisition moments within the preset time period are arranged in time sequence to form a temperature change sequence, and the correlation coefficient between the power consumption sequence and the temperature change sequence is calculated. The larger the correlation coefficient, the better the regulation effect of the PID control parameters.

[0046] In this embodiment, the correlation coefficient between the power consumption sequence and the temperature change sequence is the Pearson correlation coefficient. As other implementation methods, on the basis of being able to measure the correlation between the power consumption sequence and the temperature change sequence, the implementer may adopt other existing technologies to measure the correlation between the power consumption sequence and the temperature change sequence, such as the Spearman correlation coefficient, the Kendall rank correlation coefficient, etc., and this application does not impose any special restrictions.

[0047] Furthermore, the product of the reciprocal and the correlation coefficient is used as the temperature variation coefficient at the current adjustment moment; when the temperature stability of the server is stronger and the correlation between the heat dissipation capacity and the server power consumption during the PID control process is stronger, the value of the temperature variation coefficient is larger, indicating that the adjustment ability of the PID control parameters is stronger.

[0048] Furthermore, based on the power consumption mutation at the current adjustment time, the change index at the current adjustment time, and the temperature change coefficient at the current adjustment time, the temperature adjustment index at the current adjustment time is obtained, and the expression is: ; Wherein, Bc represents the temperature adjustment index at the current adjustment moment; Tn represents the power consumption mutation at the current adjustment moment; Fr represents the change index at the current adjustment moment; H represents the temperature change coefficient at the current adjustment moment; norm ( ) represents the normalization operation; β represents a constant preset greater than 0, which is used to avoid the denominator being 0. The value of β is preset by humans and can be set by the implementer. In this embodiment, the value of β is 0.01. In this embodiment, the hyperbolic tangent function is used for normalization operation.

[0049] It should be noted that: after analyzing all the data within the preset time period, the stronger the adjustment ability of the PID control parameters, the weaker the sudden change of the power consumption of the server, and the smaller the regulation lag of the PID control parameters, the less necessary it is to adjust the PID control parameters at the current adjustment time. The schematic diagram of the temperature adjustment index acquisition process is as follows Figure 2 shown.

[0050] Furthermore, the time period between the current adjustment moment and the next adjustment moment is used as the adjustment interval of the next adjustment moment, and an adjustment threshold is set. If the temperature adjustment index of the current adjustment moment is less than or equal to the preset adjustment threshold, the PID control parameters of the current adjustment moment are still used to adjust the heat dissipation device parameters within the adjustment interval of the next adjustment moment after the current adjustment moment; otherwise, the PID control parameters of the current adjustment moment need to be adjusted. The flowchart for determining whether to adjust the PID control parameters is as follows Figure 3 shown.

[0051] In this embodiment, the value of the preset adjustment threshold is 0.5. The value of the preset adjustment threshold is preset manually and can be set by the implementer. This application does not impose any special restrictions.

[0052] The control parameter adjustment module 103 is used to adjust the PID control parameters. Based on the similarity of all data except temperature data between each historical adjustment moment and the current adjustment moment, and the similarity of the PID control parameters, each reference adjustment moment of the current adjustment moment is obtained, and based on the PID control parameters of the next moment of each reference adjustment moment, the PID control parameters of the current adjustment moment are adjusted in combination with the temperature adjustment index.

[0053] The specific method for adjusting the PID control parameters at the current adjustment time is as follows: The PID control parameters, pump speed, rotation speed, opening, water flow rate, water inlet temperature, water outlet temperature and power consumption data at each adjustment moment are used to form the control vector of each adjustment moment. All adjustment moments are clustered according to the distance between the control vectors. The remaining adjustment moments in the same cluster as the current adjustment moment are recorded as the reference adjustment moments of the current adjustment moment.

[0054] In this embodiment, the K-Means algorithm is used to cluster all adjustment moments according to the distance between the control vectors, where the distance between the control vectors is the Euclidean distance. The K-Means algorithm is a well-known technology and will not be described in detail in this application. The implementer can select other feasible clustering algorithms to cluster all adjustment moments, and this application does not impose any special restrictions.

[0055] Furthermore, based on the temperature adjustment index at the current adjustment moment and the PID control parameters at the next moment of each reference adjustment moment at the current adjustment moment, the PID control parameters at the current adjustment moment are adjusted, and the expression is: ; In the formula, represents the adjusted PID control parameter at the current adjustment time; Bc represents the temperature adjustment index at the current adjustment time; Pi represents the PID control parameter at the current adjustment time; N represents the total number of reference adjustment times at the current adjustment time; Indicates the PID control parameter at the next moment of the u-th reference adjustment moment of the current adjustment moment. The initial value of the control parameter of the PID control algorithm is determined by the attenuation curve method, which is a well-known technology and will not be described in detail in this application.

[0056] In an adjustment interval at a subsequent adjustment moment of the current adjustment moment, the heat dissipation device parameters are adjusted using the adjusted PID control parameters at the current adjustment moment.

[0057] Based on the same inventive concept as the above method, an embodiment of the present application also provides a warm water circulation energy-saving heat dissipation system for a high-performance server, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, it is used to implement the above-mentioned warm water circulation energy-saving heat dissipation device for a high-performance server.

[0058] In summary, this application analyzes whether various parameter data of the heat dissipation device have corresponding changes when the power consumption data of the server changes, and obtains a change index to reflect the regulation hysteresis of the PID control parameters; Furthermore, by obtaining the sudden change of power consumption at the current adjustment moment, abnormal changes in power consumption can be discovered in time; by obtaining the temperature change factor through the difference between the water inlet temperature and the water outlet temperature, combined with the water flow rate, the heat dissipation capacity of the heat dissipation device is analyzed to provide a basis for whether the PID control parameters are adjusted; based on the discrete degree of temperature data and combined with the correlation between power consumption data and temperature change factor, the temperature change coefficient is obtained to analyze the adjustment ability of the PID control parameters, which helps to understand the effectiveness of the current PID control parameters in regulating the parameters of the heat dissipation device; and then comprehensively considering the change index, power consumption sudden change and temperature change coefficient to determine whether the current PID control parameters need to be adjusted, so that the adjustment can be more in line with actual needs. If the parameters of the heat dissipation device are adjusted according to the current PID control parameters, resulting in poor heat dissipation effect, the PID control parameters are adjusted, and then the fan speed, pump speed and other parameters are adjusted to avoid unnecessary energy waste and achieve more efficient and energy-saving heat dissipation control. However, if the parameters of the heat dissipation device are adjusted according to the current PID control parameters, the parameters of the heat dissipation device can adapt to the power consumption changes of the server, and then the PID control parameters are not adjusted, thereby avoiding unnecessary adjustment operations and saving time and resources; Furthermore, when adjusting the PID control parameters, not only the temperature adjustment index at the current adjustment moment is considered, but also the adjustment moments with similar PID control parameters and heat dissipation device parameters as the current adjustment moment are combined, so that the PID control parameters can be adjusted more finely and the adjustment accuracy can be improved.

[0059] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

[0060] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the basic features of the present application. Therefore, no matter from which point of view, the above embodiments of the present application should be regarded as exemplary and non-restrictive.

Claims

1. A warm water circulation energy-saving heat dissipation device for a high-performance server, characterized in that: The device includes: The heat dissipation data collection module is used to collect the temperature data and power consumption data of the server in real time, and collect various parameter data of the heat dissipation device in real time, including the water inlet temperature and water outlet temperature of the heat dissipation device, and the water flow rate in the water pipe of the heat dissipation device; The heat dissipation data analysis module is used to preset the adjustment cycle of the PID control parameters, obtain the current adjustment time of the PID control parameters, analyze all data within the preset time period, and determine whether to adjust the PID control parameters at the current adjustment time. The process is: Based on the similarity of the change between all parameter data except the water inlet temperature and the water outlet temperature of the heat sink and all power consumption data, the change index at the current adjustment moment is obtained; Based on the mutation of all power consumption data, obtain the power consumption mutation at the current adjustment moment; Based on the difference between the water inlet temperature and the water outlet temperature at each collection moment, combined with the water flow rate at each collection moment, the temperature change factor at each collection moment is obtained; based on the discrete degree of all temperature data and the correlation between all power consumption data and all temperature change factors, the temperature change coefficient at the current adjustment moment is obtained; Based on the power consumption mutation, the change index and the temperature change coefficient, obtaining the temperature adjustment index at the current adjustment moment, and determining whether to adjust the PID control parameters at the current adjustment moment; The control parameter adjustment module is used to adjust the PID control parameters. Based on the similarity of all data except temperature data between each historical adjustment moment and the current adjustment moment, and the similarity of the PID control parameters, the reference adjustment moments of the current adjustment moment are obtained, and the PID control parameters of the next moment of each reference adjustment moment are adjusted in combination with the temperature adjustment index.

2. A warm water circulation energy-saving heat dissipation device for a high-performance server as claimed in claim 1, characterized in that: The various parameter data of the heat dissipation device also include the pumping speed of the water pump in the heat dissipation device, the opening of the valve for adjusting the water flow speed, and the rotation speed of the heat dissipation fan.

3. A warm water circulation energy-saving heat dissipation device for a high-performance server as claimed in claim 2, characterized in that: The expression of the variation index is: ; Wherein, Fr represents the change index at the current adjustment moment; S represents the number of all power consumption data within the preset time period; Indicates the difference in power consumption data between the qth collection moment and the q-1th collection moment within the preset time period; represents the difference in power consumption data between the qth collection moment and the q+1th collection moment within the preset time period; the pump speed, rotation speed, opening degree and water flow speed at each collection moment form a heat dissipation parameter vector, represents the distance of the heat dissipation parameter vector between the qth collection moment and the q+1th collection moment within the preset time period; ε indicates a preset value greater than 0.

4. A warm water circulation energy-saving heat dissipation device for a high-performance server as claimed in claim 1, characterized in that: The process of obtaining the power consumption mutation is as follows: The difference in power consumption data between each acquisition moment and its adjacent acquisition moment is recorded as the power consumption difference at each acquisition moment, a threshold segmentation method is used to obtain a segmentation threshold of the power consumption difference at all acquisition moments, and the power consumption difference greater than the segmentation threshold is recorded as high difference data; The expression of power consumption mutation at the current adjustment moment is: ; Wherein, Tn represents the power consumption mutation at the current adjustment moment; A represents the mean of all high difference data; M represents the number of high difference data; S represents the number of all power consumption data within the preset time period; Indicates the time interval between the i-th high difference data and the i+1-th high difference data; Represents the average value of the time interval between any two adjacent high-difference data; wherein the high-difference data are arranged in time sequence.

5. A warm water circulation energy-saving heat dissipation device for a high-performance server as claimed in claim 1, characterized in that: The process of obtaining the temperature change factor is as follows: The difference between the water inlet temperature and the water outlet temperature at each sampling moment is recorded as the temperature difference; The temperature variation factor is the product of the temperature difference at each collection moment and the water flow velocity.

6. A warm water circulation energy-saving heat dissipation device for a high-performance server as claimed in claim 1, characterized in that: The process of obtaining the temperature variation coefficient is as follows: Calculate the reciprocal of the sum of the discrete degree and a preset value greater than 0; Calculating the correlation coefficients between all power consumption data and all temperature change factors within the preset time period; The temperature variation coefficient is the product of the reciprocal and the correlation coefficient.

7. A warm water circulation energy-saving heat dissipation device for a high-performance server as claimed in claim 1, characterized in that: The expression of the temperature regulation index is: ; In the formula, Bc represents the temperature adjustment index at the current adjustment moment; Tn represents the power consumption mutation at the current adjustment moment; Fr represents the change index at the current adjustment moment; H represents the temperature change coefficient at the current adjustment moment; norm( ) represents the normalization operation; β represents a preset constant greater than 0.

8. A warm water circulation energy-saving heat dissipation device for a high-performance server as claimed in claim 1, characterized in that: The specific method for judging whether to adjust the PID control parameters at the current adjustment moment is as follows: when the temperature adjustment index is less than or equal to a preset adjustment threshold, the PID control parameters at the current adjustment moment are not adjusted; otherwise, the PID control parameters at the current adjustment moment are adjusted.

9. A warm water circulation energy-saving heat dissipation device for a high-performance server as claimed in claim 2, characterized in that: The process of adjusting the PID control parameters at the current adjustment time is as follows: The PID control parameters, pump speed, rotation speed, opening, water flow rate, water inlet temperature, water outlet temperature and power consumption data at each adjustment moment are combined into a control vector, and all adjustment moments are clustered according to the distance between the control vectors; The remaining adjustment moments in the same cluster as the current adjustment moment are recorded as reference adjustment moments of the current adjustment moment; ; In the formula, represents the adjusted PID control parameter at the current adjustment time; Bc represents the temperature adjustment index at the current adjustment time; Pi represents the PID control parameter at the current adjustment time; N represents the total number of reference adjustment times at the current adjustment time; Indicates the PID control parameters at the next moment of the u-th reference adjustment moment of the current adjustment moment.

10. A warm water circulation energy-saving cooling system for a high-performance server, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, it is used to implement the warm water circulation energy-saving heat dissipation device of the high-performance server as described in any one of claims 1-9.

Images