Energy management method and system for immersion liquid cooling server
By collecting and analyzing the operating load data of the server, combined with the various flow rate adjustments of the liquid cooling system, the precise temperature control of the immersed liquid cooling server is achieved, solving the problem of unstable temperature adjustment in the existing technology, and improving the performance of the server and service life.
Patent Information
- Application Number
- CN202411225221.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-09-03
AI Technical Summary
The existing immersion liquid-cooled server lacks intelligent control functions and cannot achieve precise temperature control, resulting in unstable server working environment and affecting performance and service life.
By collecting the operating load data of the server, establishing a relationship diagram between the time period and the operating load, deploying a liquid cooling system and establishing a connection between the liquid cooling unit and the server, recording the temperature change data at different flow rates, and establishing the flow rate and temperature data of the liquid cooling unit. The relationship between the server realizes initial and fine temperature adjustment to the server, ensuring that the temperature is always within the preset range.
It realizes precise temperature control of immersed liquid-cooled servers, improves the stability and performance of the server's working environment, and extends the service life.
Smart Images

Figure CN118733397B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of server energy management, and in particular relates to an energy management method and system for an immersion liquid cooling server. Background Art
[0002] With the rapid development of technologies such as cloud computing, big data, and artificial intelligence, the demand for server computing power continues to increase, and the power consumption of servers has also increased accordingly. In order to ensure the stable operation of the server, the server needs to be effectively cooled and temperature controlled. Traditional air cooling technology has been unable to meet the heat dissipation needs of high-power servers, so liquid cooling technology has become one of the solutions. Immersion liquid cooling technology is a technology that directly immerses the server in liquid coolant for heat dissipation. Compared with traditional air cooling technology, it has the advantages of high heat dissipation efficiency, low noise, and energy saving. However, immersion liquid cooling technology also faces challenges in energy management and temperature control. Existing server immersion liquid cooling servers use fixed temperature refrigeration for temperature adjustment, often lack intelligent control functions, and cannot achieve precise temperature control of the server, resulting in an unstable server working environment, affecting the performance and service life of the server.
[0003] Therefore, how to improve the energy management of the existing immersion liquid cooling server, achieve precise temperature control of the immersion liquid cooling server, stabilize the server working environment, improve the performance of the immersion liquid cooling server, and extend its service life. Summary of the invention
[0004] The purpose of the present invention is to provide an energy management method and system for an immersion liquid cooling server, so as to improve the energy management of the existing immersion liquid cooling server, realize precise temperature control of the immersion liquid cooling server, stabilize the server working environment, improve the performance of the immersion liquid cooling server, and extend its service life.
[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is as follows:
[0006] In a first aspect, a method for energy management of an immersion liquid cooling server is provided, comprising the following steps:
[0007] S1: collecting the operating loads of several servers within a specified period, obtaining the temperature data of each server under different operating loads within the specified period, and establishing a relationship diagram between each server and the operating load in different time periods, and a relationship diagram between different operating loads and temperatures;
[0008] S2: deploying a liquid cooling system, establishing connections between each liquid cooling unit of the liquid cooling system and the plurality of servers, starting each liquid cooling unit of the liquid cooling system and operating them at different flow rates, recording temperature change data of the servers at different flow rates of each liquid cooling unit, and establishing a data set of corresponding relationships between the flow rate of each liquid cooling unit of each server and the operating load and temperature data of the server in combination with the relationship between the server at different operating loads and temperature data;
[0009] S3: determining the current time period, and based on the operation load data of the server in different time periods in the data set, calling the data set, obtaining the flow rate of the liquid cooling unit corresponding to the current time period, and making each liquid cooling unit operate at the corresponding flow rate to achieve initial temperature adjustment of the server;
[0010] S4: Collect the real-time temperature data of each server, and determine whether the real-time temperature of each server is within the preset temperature range of the current time period. If so, continue to run at the flow rate under the initial temperature adjustment in step S3. If the real-time temperature is higher than the maximum value of the preset temperature range or lower than the minimum value of the preset temperature range, execute step S5;
[0011] S5: When the real-time temperature of a server is higher than the maximum value of the preset temperature range, the flow rate of the corresponding liquid cooling unit is increased by a fixed value; when the real-time temperature of a server is lower than the minimum value of the preset temperature range, the flow rate of the corresponding liquid cooling unit is reduced by a fixed value;
[0012] S6: cyclically executing steps S4 and S5, cyclically adjusting each server so that the temperature of each server is always within a preset temperature range.
[0013] Preferably, step S1 includes the following specific process:
[0014] S11: Obtaining operation load data and corresponding temperature data of each server for at least 30 days, wherein the operation load data is counted in units of 24 hours, and a relationship diagram between different time periods and operation load of each server within 24 hours is constructed based on the average load data of 30 days;
[0015] S12: constructing a relationship diagram between the operating load and temperature of each server in different time periods within 24 hours based on the 30-day average temperature data.
[0016] Preferably, step S2 includes the following specific process:
[0017] S21: deploying a liquid cooling system, wherein the liquid cooling system includes a plurality of liquid cooling units, the number of the liquid cooling units is the same as the number of the servers, each liquid cooling unit is an independently controlled unit, and establishing a connection between each liquid cooling unit and a corresponding server;
[0018] S22: Start the liquid cooling system, make each liquid cooling unit run at a different flow rate, respectively, make each server run at the same first load for a specified time, record the temperature change data of the liquid cooling unit when the server runs at the first load at different flow rates, make each server run at the same second load for a specified time, record the temperature change data of the liquid cooling unit when the server runs at the second load at different flow rates, ..., make each server run at the same Nth load for a specified time, record the temperature change data of the liquid cooling unit when the server runs at the Nth load at different flow rates, the first load, the second load to the Nth load are the average loads of each server in different time periods;
[0019] S23: Based on the relationship between the server's operating loads and temperature data at different times, a data set is established that shows the corresponding relationship between the flow rate of each server's liquid cooling unit and the server's operating load and temperature data, and the lowest flow rate that meets the preset temperature range is extracted under different operating loads.
[0020] Preferably, step S3 includes the following specific process:
[0021] S31: obtaining the current time period, and based on the operation load data of each server in the data set in the time period, calling the minimum flow rate of each server in the data set that meets the preset temperature range;
[0022] S32: Based on the minimum flow rate of each server that meets the preset temperature range in the current time period, each liquid cooling unit is controlled to operate at the corresponding minimum flow rate to achieve initial temperature adjustment of each server.
[0023] Preferably, during step S5, when the real-time temperature of a server is higher than the maximum value of a preset temperature range, the initial temperature of the liquid-cooling medium in the corresponding liquid-cooling unit is reduced by a preset value, and the temperature of the server is continuously measured within a fixed time. If the temperature of the server is still higher than the maximum value of the preset temperature range within the fixed time, the initial temperature of the liquid-cooling medium in the corresponding liquid-cooling unit is further reduced by a preset value, and the temperature of the server is cyclically detected and the temperature of the liquid-cooling medium is reduced until the real-time temperature of the server is within the preset temperature range.
[0024] Preferably, during the temperature measurement of the server, a thermocouple temperature sensor is used to monitor the temperature of the server entering the liquid cooling medium.
[0025] In a second aspect, an energy management system for an immersion liquid cooling server is provided, which is used to implement any one of the energy management methods for an immersion liquid cooling server, including a liquid cooling system, a temperature acquisition module, a data set creation module, a pressure sensor, and a control module, wherein the liquid cooling system, the temperature acquisition module, the data set creation module, and the pressure sensor are all connected to the control module;
[0026] The liquid cooling system is used to circulate the liquid cooling medium into the immersion liquid cooling server to achieve temperature regulation of the server;
[0027] The temperature acquisition module is used to immerse in the cooling liquid and monitor the temperature data of the server in real time;
[0028] The data set creation module is used to establish a data set of the corresponding relationship between the flow rate of each server liquid cooling unit and the server operating load and temperature data based on the temperature change data of the server at different flow rates of each liquid cooling unit and in combination with the relationship between the server at different operating loads and temperature data;
[0029] The pressure sensor is installed on the liquid cooling system pipeline and is used to monitor the pressure change of the cooling liquid to infer the quality and flow rate of the liquid cooling medium;
[0030] The control module is used to control the liquid cooling system, the temperature acquisition module, the data set creation module and the pressure sensor.
[0031] Preferably, a water quality sensor and a vibration sensor are also included, both of which are connected to the control module. The water quality sensor is used to monitor the concentration of various chemical substances in the liquid cooling medium, including dissolved solids, ion concentration and pH value, and the vibration sensor is used to monitor bubbles and suspended matter in the cooling liquid, as well as the quality and state of the liquid.
[0032] Preferably, it further comprises an infrared sensor, which is connected to the control module and is used to detect the surface temperature of the liquid cooling medium and the height of the liquid level.
[0033] The beneficial effects of the present invention include:
[0034] The energy management method and system of the immersion liquid cooling server provided by the present invention collect the operation load of the server within a specified period, establish a relationship diagram between the server's time period and the operation load, and a relationship diagram between the operation load and the temperature; start each liquid cooling unit of the liquid cooling system and operate at different flow rates, and establish a data set of the corresponding relationship between the flow rate of each server liquid cooling unit and the server's operation load and temperature data; call the data set, and make each liquid cooling unit operate at a corresponding flow rate to achieve initial temperature adjustment of the server; collect the real-time temperature data of each server, if the real-time temperature is within a preset temperature range, operate at the flow rate under the initial temperature adjustment, if the real-time temperature is higher than the maximum value of the preset temperature range or lower than the minimum value, when the real-time temperature of a server is higher than the maximum value of the preset temperature range or lower than the minimum value, the flow rate of the corresponding liquid cooling unit is increased or decreased by a fixed value.
[0035] First, by collecting the operating loads of several servers within a specified period and obtaining the temperature data of each server under different operating loads during the specified period, a relationship diagram between the server's time period and the operating load is established to understand the operating load pattern of the server in each time period, and the relationship between the operating load and temperature based on the load pattern, providing a preliminary data basis for subsequent temperature adjustment through the liquid cooling system.
[0036] Secondly, by having each liquid cooling unit of the liquid cooling system operate at different flow rates, a data set is established that shows the correspondence between the flow rate of each server liquid cooling unit and the server operating load and temperature data, further providing a data basis for subsequent temperature adjustment through the liquid cooling system.
[0037] Thirdly, according to the current time period, and based on the relationship between the server's time period and operating load, the relationship between the operating load and temperature, and the data set of the correspondence between the flow rate of each server's liquid cooling unit and the server's operating load and temperature data, initial temperature adjustment is performed, and then further fine temperature adjustment is achieved based on the comparison results between the real-time temperature and the preset temperature range. Through the process of cyclic adjustment, the server's temperature is always within the preset temperature range, making its working environment stable, improving the performance of the immersion liquid cooling server, and extending its service life. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 It is a schematic flow chart of the energy management method of the immersion liquid cooling server of the present invention.
[0039] Figure 2 The present invention is a schematic diagram of the energy management system of an immersion liquid cooling server in the prior art.
[0040] Figure 3The figure is a schematic diagram of the energy management system of the immersion liquid cooling server of the present invention. DETAILED DESCRIPTION
[0041] The following is combined with Figure 1~Figure 3 The present invention is further described in detail:
[0042] Example 1
[0043] See attached Figure 1 As shown, an energy management method for an immersion liquid cooling server includes the following steps:
[0044] S1: Collect the operating loads of several servers within a specified period, obtain the temperature data of each server under different operating loads within the specified period, and establish a relationship diagram between each server and the operating load in different time periods, as well as a relationship diagram between different operating loads and temperatures. Since there is an obvious regularity between the operating load of the server and the time period, for example, the operating load during the day in 24 hours a day will be significantly higher than the operating load at night, the operating load data of several servers within a specified period, such as 30 days, is collected to obtain the regularity between the time period and the operating load. For the server, there is also an obvious relationship between the operating load of the server and its temperature change. The higher the operating load, the higher the temperature of the server. Therefore, the relationship between the operating load and the temperature is established through the temperature data of the server, providing a preliminary data basis for the subsequent temperature adjustment of the liquid cooling system.
[0045] S2: Deploy the liquid cooling system and establish connections between each liquid cooling unit of the liquid cooling system and several servers. Each server has an independently controllable liquid cooling unit to achieve temperature adjustment for the corresponding server. Start each liquid cooling unit of the liquid cooling system and operate it at different flow rates. Record the temperature change data of the server at different flow rates for each liquid cooling unit. Combined with the relationship between the server's different operating loads and temperature data, establish a data set of the corresponding relationship between the flow rate of each server's liquid cooling unit and the server's operating load and temperature data. Since the greater the flow rate of the liquid cooling unit, the faster the server's temperature drops, the corresponding relationship between the flow rate of the liquid cooling unit and the server's operating load and temperature data can be obtained, thereby providing a further data basis for subsequent temperature adjustment of the liquid cooling system.
[0046] S3: Determine the current time period, based on the operating load data of the server in different time periods in the data set, call the data set, obtain the flow rate of the liquid cooling unit corresponding to the current time period, and make each liquid cooling unit run at the corresponding flow rate to achieve initial temperature adjustment of the server. According to the time period, its operating load can be predicted, and the corresponding liquid cooling unit flow rate is extracted based on the predicted operating load, and the liquid cooling unit is run at the flow rate to achieve initial temperature adjustment of the server. The initial temperature adjustment here is to pre-adjust the server operating load predicted according to the time period.
[0047] S4: Collect the real-time temperature data of each server, and determine whether the real-time temperature of each server is within the preset temperature range of the current time period. If so, continue to run at the flow rate under the initial temperature adjustment in step S3. If the real-time temperature is higher than the maximum value of the preset temperature range or lower than the minimum value of the preset temperature range, execute step S5;
[0048] S5: When the real-time temperature of a server is higher than the maximum value of the preset temperature range, the flow rate of the corresponding liquid cooling unit is increased by a fixed value; when the real-time temperature of a server is lower than the minimum value of the preset temperature range, the flow rate of the corresponding liquid cooling unit is reduced by a fixed value. The reason for collecting the real-time temperature here is that there may be a certain gap between the predicted operating load and the actual operating load, so the flow rate of the liquid cooling unit after the initial temperature adjustment may not be completely appropriate, so the flow rate of the liquid cooling unit is further adjusted based on the real-time temperature.
[0049] S6: Execute steps S4 and S5 in a loop, and adjust each server in a loop so that the temperature of each server is always within the preset temperature range. Since increasing or decreasing the fixed value once may not necessarily adjust the flow rate of the liquid cooling unit to a suitable temperature, it may be necessary to adjust the flow rate in a loop for multiple times to reach a suitable flow rate. In addition, the load of the server is not fixed, but may change over time, so it is necessary to continuously monitor the temperature of the server so that the temperature of the server is always within the preset temperature range.
[0050] In the energy management process of the entire immersion liquid cooling server, by collecting the operating loads of several servers within a specified period, and obtaining the temperature data of each server under different operating loads within the specified period, a relationship diagram between the server time period and the operating load is established to obtain the operating load law of the server in each time period, and the relationship between the operating load and the temperature is obtained on the basis of the load law, providing a preliminary data basis for subsequent temperature adjustment through the liquid cooling system. Through the liquid cooling system, each liquid cooling unit works at different flow rates, and a data set of the corresponding relationship between the flow rate of each server liquid cooling unit and the server operating load and temperature data is established, further providing a data basis for subsequent temperature adjustment through the liquid cooling system. According to the current time period, and based on the relationship between the server time period and the operating load, the relationship between the operating load and the temperature, and the data set of the corresponding relationship between the flow rate of each server liquid cooling unit and the server operating load and temperature data, initial temperature adjustment is performed, and then further fine temperature adjustment is achieved based on the comparison result between the real-time temperature and the preset temperature range, and the temperature of the server is always within the preset temperature range through the cyclic adjustment process, so that its working environment is stable, the performance of the immersion liquid cooling server is improved, and its service life is extended.
[0051] Example 2
[0052] Step S1 includes the following specific processes:
[0053] S11: Obtaining at least 30 days of operation load data and corresponding temperature data of each server, wherein the operation load data is counted in units of 24 hours, and a relationship diagram between different time periods and operation load of each server within 24 hours is constructed based on the average load data of 30 days, because the load data of the server generally changes in a cycle of 24 hours, and the reason for collecting data for 30 days is that there is a certain degree of randomness if only a few days of data are collected, so at least 30 days of load data and temperature data are collected;
[0054] S12: constructing a relationship diagram between the operating load and temperature of each server in different time periods within 24 hours based on the 30-day average temperature data.
[0055] Step S2 includes the following specific processes:
[0056] S21: deploying a liquid cooling system, wherein the liquid cooling system includes a plurality of liquid cooling units. The number of the liquid cooling units is the same as the number of the servers. Each server corresponds to a liquid cooling unit, so that one server is independently controlled by one liquid cooling unit. Each liquid cooling unit is an independently controlled unit, and a connection is established between each liquid cooling unit and the corresponding server.
[0057] S22: Start the liquid cooling system, make each liquid cooling unit run at a different flow rate, respectively, make each server run at the same first load for a specified time, record the temperature change data of the liquid cooling unit when the server runs at the first load at different flow rates, make each server run at the same second load for a specified time, record the temperature change data of the liquid cooling unit when the server runs at the second load at different flow rates, ..., make each server run at the same Nth load for a specified time, record the temperature change data of the liquid cooling unit when the server runs at the Nth load at different flow rates, the first load, the second load to the Nth load are the average loads of each server in different time periods;
[0058] S23: In combination with the relationship between the server's different operating loads and temperature data, a data set of the corresponding relationship between the flow rate of the liquid cooling unit of each server and the server's operating load and temperature data is established, and under different operating loads, the minimum flow rate that meets the preset temperature range is extracted. The reason for extracting the minimum flow rate that meets the preset temperature range is that in the daily operation of the server, as long as the temperature of the server continues to be within the preset temperature range, the server can work normally without the need for the server to be at a very low temperature. The server being at a very low temperature requires the liquid cooling unit to consume more energy, and the minimum flow rate that meets the preset temperature range can ensure that the server is within the normal temperature range on the one hand, and on the other hand, it can also ensure the normal operation of the server.
[0059] Example 3
[0060] Based on Example 1 or Example 2, step S3 includes the following specific process:
[0061] S31: Obtain the current time period, and based on the operating load data of each server in the data set in the time period, call the minimum flow rate of each server in the data set that meets the preset temperature range. The minimum flow rate is preset according to the load data of each server, so the operating conditions of each server in each time period are different. Based on the operating load data and temperature data of each server in each time period obtained in steps S1 and S2, the flow rate of the liquid cooling medium of the corresponding liquid cooling unit is obtained;
[0062] S32: Based on the minimum flow rate of each server that meets the preset temperature range in the current time period, each liquid cooling unit is controlled to operate at the corresponding minimum flow rate to achieve initial temperature adjustment of each server. The reason for performing initial temperature adjustment on each server according to the current time period is that the operating load of each server is different in each time period of the server. If the temperature is adjusted at the same flow rate in each time period, on the one hand, it may cause the temperature of the server to not be continuously within the preset temperature range. On the other hand, when the two time periods alternate, the temperature can be adjusted at a preset flow rate so that the server can be roughly within the preset temperature range. Even if it exceeds the temperature range, it is unlikely to change too much. Combined with subsequent fine adjustment, the temperature of the server can be continuously within the preset temperature range.
[0063] During step S5, when the real-time temperature of a server is higher than the maximum value of a preset temperature range, the initial temperature of the liquid-cooling medium in the corresponding liquid-cooling unit is reduced by a preset value, and the temperature of the server is continuously measured within a fixed time. If the temperature of the server is still higher than the maximum value of the preset temperature range within the fixed time, the initial temperature of the liquid-cooling medium in the corresponding liquid-cooling unit is further reduced by a preset value, and the temperature of the server is cyclically detected and the temperature of the liquid-cooling medium is reduced until the real-time temperature of the server is within the preset temperature range.
[0064] According to the current time period, and based on the relationship between the server's time period and operating load, the relationship between the operating load and temperature, and the data set of the correspondence between the flow rate of each server liquid cooling unit and the server's operating load and temperature data, initial temperature adjustment is performed, and then further fine temperature adjustment is achieved based on the comparison result between the real-time temperature and the preset temperature range. Through the process of cyclic adjustment, the server's temperature is always within the preset temperature range, so that its working environment is stable, the performance of the immersion liquid cooling server is improved, and its service life is extended.
[0065] In the process of measuring the temperature of the server, a thermocouple temperature sensor is used to monitor the temperature of the server entering the liquid cooling medium, which can accurately monitor the real-time temperature data of the server during operation.
[0066] See attached Figure 2 As shown, the energy management system of the immersion liquid cooling server in the prior art uniformly controls all servers through the liquid cooling system, has no independent control and adjustment process, and cannot achieve precise control of the temperature of the server.
[0067] See attached Figure 3As shown, an energy management system for an immersion liquid cooling server is used to implement any one of the energy management methods for an immersion liquid cooling server, including a liquid cooling system, a temperature acquisition module, a data set creation module, a pressure sensor, and a control module, wherein the liquid cooling system, temperature acquisition module, data set creation module, and pressure sensor are all connected to the control module.
[0068] The liquid cooling system is used to circulate the liquid cooling medium to the immersed liquid cooling server to achieve temperature regulation of the server; the temperature acquisition module is used to immerse in the cooling liquid to monitor the temperature data of the server in real time; the data set creation module is used to establish a data set of the corresponding relationship between the flow rate of each server liquid cooling unit and the server operating load and temperature data based on the temperature change data of the server at different flow rates of each liquid cooling unit and in combination with the relationship between the server at different operating loads and temperature data; the pressure sensor is installed on the liquid cooling system pipeline to monitor the pressure change of the cooling liquid to infer the quality and flow rate of the liquid cooling medium; the control module is used to control the liquid cooling system, the temperature acquisition module, the data set creation module and the pressure sensor.
[0069] The energy management system of the immersion liquid cooling server also includes a water quality sensor and a vibration sensor, both of which are connected to the control module. The water quality sensor is used to monitor the concentration of various chemical substances in the liquid cooling medium, including dissolved solids, ion concentration and pH value. The data results obtained by the water quality sensor can be used to know the state of the liquid cooling medium, and to know that when the water quality no longer meets the preset requirements, for example, the amount of dissolved solids exceeds the preset threshold, or the ion concentration exceeds its preset threshold or the pH value is not within the preset range, the cooling medium needs to be replaced. The vibration sensor is used to monitor bubbles and suspended matter in the cooling liquid, as well as the quality and state of the liquid. It also includes an infrared sensor, which is connected to the control module and is used to detect the surface temperature and liquid level of the liquid cooling medium. The infrared sensor detects the surface temperature and liquid level of the liquid cooling medium to know whether the temperature adjustment of the server, that is, the surface temperature and liquid level of the liquid cooling medium during the cooling process, meet the corresponding requirements. When the liquid level temperature is higher than the preset value, it is necessary to increase its flow rate or reduce its initial temperature. When the liquid level is not within the preset height range, the liquid level needs to be adjusted.
[0070] In summary, the energy management method and system for an immersion liquid-cooled server provided by the present invention collects the operating load of the server within a specified period, establishes a relationship diagram between the server's time period and the operating load, and a relationship diagram between the operating load and the temperature; starts each liquid cooling unit of the liquid cooling system and operates at different flow rates, and establishes a data set of the corresponding relationship between the flow rate of each server liquid cooling unit and the server's operating load and temperature data; calls the data set, and makes each liquid cooling unit operate at a corresponding flow rate to achieve initial temperature adjustment of the server; collects real-time temperature data of each server, and if the real-time temperature is within the preset temperature range, operates at the flow rate under the initial temperature adjustment; if the real-time temperature is higher than the maximum value of the preset temperature range or lower than the minimum value, when the real-time temperature of a server is higher than the maximum value of the preset temperature range or lower than the minimum value, the flow rate of the corresponding liquid cooling unit is increased or decreased by a fixed value.
[0071] By collecting the operating loads of several servers within a specified period, and obtaining the temperature data of each server under different operating loads within the specified period, a relationship diagram between the time period and the operating load of the server is established to obtain the operating load law of the server in each time period, and the relationship between the operating load and the temperature is obtained based on the load law, providing a preliminary data basis for subsequent temperature adjustment through the liquid cooling system. Through the liquid cooling system, each liquid cooling unit works at different flow rates, and a data set of the corresponding relationship between the flow rate of each server liquid cooling unit and the server operating load and temperature data is established, further providing a data basis for subsequent temperature adjustment through the liquid cooling system. According to the current time period, and based on the relationship between the server time period and the operating load, the relationship between the operating load and the temperature, and the data set of the corresponding relationship between the flow rate of each server liquid cooling unit and the server operating load and temperature data, initial temperature adjustment is performed, and then further fine temperature adjustment is achieved based on the comparison result between the real-time temperature and the preset temperature range, and the temperature of the server is always within the preset temperature range through the cyclic adjustment process, so that its working environment is stable, the performance of the immersion liquid cooling server is improved, and its service life is extended.
Claims
1. An energy management method for an immersion liquid cooling server, characterized in that: The following steps are involved: S1: collecting the operating loads of several servers within a specified period, obtaining the temperature data of each server under different operating loads within the specified period, and establishing a relationship diagram between each server and the operating load in different time periods, and a relationship diagram between different operating loads and temperatures; S2: deploying a liquid cooling system, establishing connections between each liquid cooling unit of the liquid cooling system and the plurality of servers, starting each liquid cooling unit of the liquid cooling system and operating them at different flow rates, recording temperature change data of the servers at different flow rates of each liquid cooling unit, and establishing a data set of corresponding relationships between the flow rate of each liquid cooling unit of each server and the operating load and temperature data of the server in combination with the relationship between the server at different operating loads and temperature data; S3: determining the current time period, and based on the operation load data of the server in different time periods in the data set, calling the data set, obtaining the flow rate of the liquid cooling unit corresponding to the current time period, and making each liquid cooling unit operate at the corresponding flow rate to achieve initial temperature adjustment of the server; S4: Collect the real-time temperature data of each server, and determine whether the real-time temperature of each server is within the preset temperature range of the current time period. If so, continue to run at the flow rate under the initial temperature adjustment in step S3. If the real-time temperature is higher than the maximum value of the preset temperature range or lower than the minimum value of the preset temperature range, execute step S5; S5: When the real-time temperature of a server is higher than the maximum value of the preset temperature range, the flow rate of the corresponding liquid cooling unit is increased by a fixed value; when the real-time temperature of a server is lower than the minimum value of the preset temperature range, the flow rate of the corresponding liquid cooling unit is reduced by a fixed value; S6: cyclically executing steps S4 and S5, cyclically adjusting each server so that the temperature of each server is always within a preset temperature range; Wherein, step S1 includes the following specific process: S11: Obtaining operation load data and corresponding temperature data of each server for at least 30 days, wherein the operation load data is counted in units of 24 hours, and a relationship diagram between different time periods and operation load of each server within 24 hours is constructed based on the average load data of 30 days; S12: constructing a relationship diagram between the operating load and temperature of each server in different time periods within 24 hours based on the average temperature data of 30 days; Step S2 includes the following specific processes: S21: deploying a liquid cooling system, wherein the liquid cooling system includes a plurality of liquid cooling units, the number of the liquid cooling units is the same as the number of the servers, each liquid cooling unit is an independently controlled unit, and establishing a connection between each liquid cooling unit and a corresponding server; S22: Start the liquid cooling system, make each liquid cooling unit run at a different flow rate, respectively, make each server run at the same first load for a specified time, record the temperature change data of the liquid cooling unit when the server runs at the first load at different flow rates, make each server run at the same second load for a specified time, record the temperature change data of the liquid cooling unit when the server runs at the second load at different flow rates, ..., make each server run at the same Nth load for a specified time, record the temperature change data of the liquid cooling unit when the server runs at the Nth load at different flow rates, the first load, the second load to the Nth load are the average loads of each server in different time periods; S23: combining the relationship between different operating loads and temperature data of the server, establishing a data set of the corresponding relationship between the flow rate of the liquid cooling unit of each server and the operating load and temperature data of the server, and extracting the lowest flow rate that meets the preset temperature range under different operating loads; Step S3 includes the following specific processes: S31: obtaining the current time period, and based on the operation load data of each server in the data set in the time period, calling the minimum flow rate of each server in the data set that meets the preset temperature range; S32: Based on the minimum flow rate of each server that meets the preset temperature range in the current time period, each liquid cooling unit is controlled to operate at the corresponding minimum flow rate to achieve initial temperature adjustment of each server.
2. The energy management method of an immersion liquid cooling server according to claim 1, characterized in that: During step S5, when the real-time temperature of a server is higher than the maximum value of a preset temperature range, the initial temperature of the liquid-cooling medium in the corresponding liquid-cooling unit is reduced by a preset value, and the temperature of the server is continuously measured within a fixed time. If the temperature of the server is still higher than the maximum value of the preset temperature range within the fixed time, the initial temperature of the liquid-cooling medium in the corresponding liquid-cooling unit is further reduced by a preset value, and the temperature of the server is cyclically detected and the temperature of the liquid-cooling medium is reduced until the real-time temperature of the server is within the preset temperature range.
3. The energy management method for an immersion liquid cooling server according to claim 1, characterized in that: During the temperature measurement of the server, a thermocouple temperature sensor is used to monitor the temperature of the server entering the liquid cooling medium.
4. An energy management system for an immersion liquid cooling server, used to implement an energy management method for an immersion liquid cooling server according to any one of claims 1 to 3, characterized in that: It includes a liquid cooling system, a temperature acquisition module, a data set creation module, a pressure sensor, and a control module, wherein the liquid cooling system, the temperature acquisition module, the data set creation module, and the pressure sensor are all connected to the control module; The liquid cooling system is used to circulate the liquid cooling medium into the immersion liquid cooling server to achieve temperature regulation of the server; The temperature acquisition module is used to immerse in the cooling liquid and monitor the temperature data of the server in real time; The data set creation module is used to establish a data set of the corresponding relationship between the flow rate of each server liquid cooling unit and the server operating load and temperature data based on the temperature change data of the server at different flow rates of each liquid cooling unit and in combination with the relationship between the server at different operating loads and temperature data; The pressure sensor is installed on the liquid cooling system pipeline and is used to monitor the pressure change of the cooling liquid to infer the quality and flow rate of the liquid cooling medium; The control module is used to control the liquid cooling system, the temperature acquisition module, the data set creation module and the pressure sensor.
5. The energy management system for an immersion liquid cooling server according to claim 4, characterized in that: It also includes a water quality sensor and a vibration sensor, both of which are connected to the control module. The water quality sensor is used to monitor the concentration of various chemical substances in the liquid cooling medium, including dissolved solids, ion concentration and pH value, and the vibration sensor is used to monitor bubbles and suspended matter in the liquid cooling medium and the quality and state of the liquid cooling medium.
6. The energy management system for an immersion liquid cooling server according to claim 4, characterized in that: It also includes an infrared sensor, which is connected to the control module and is used to detect the surface temperature of the liquid cooling medium and the height of the liquid level.
