CN114489307A

CN114489307A - Energy efficiency optimization method and device for internet data center

Info

Publication number: CN114489307A
Application number: CN202210134044.8A
Authority: CN
Inventors: 方昕; 徐永田; 朱昊; 姜伟明; 陈昱; 孟晓林; 吴轩; 王加龙; 王东; 杨浩林; 理栈; 陆增义
Original assignee: Alibaba China Co Ltd
Current assignee: Alibaba China Co Ltd
Priority date: 2022-02-14
Filing date: 2022-02-14
Publication date: 2022-05-13

Abstract

The embodiment of the specification provides an energy efficiency optimization method and device for an internet data center. The method can comprise the following steps: receiving a working condition recommendation instruction aiming at a target machine room of an internet data center; according to the working condition recommendation instruction, obtaining current actual power consumption and non-power consumption type target operation working condition parameters of a target machine room, and executing a plurality of times of power consumption prediction operations according to a preset parameter adjustment strategy to obtain first predicted power consumption corresponding to the plurality of times of power consumption prediction operations respectively; in response to that the minimum value of the first predicted power consumptions respectively corresponding to the plurality of times of power consumption prediction operations is smaller than the actual power consumption, determining a plurality of adjusted controllable parameters adopted in predicting the minimum value as a plurality of optimized parameters; and generating and outputting working condition recommendation information comprising the plurality of items of optimization parameters.

Description

Energy efficiency optimization method and device for internet data center

Technical Field

The embodiment of the specification relates to the technical field of computers, in particular to an energy efficiency optimization method and device for an internet data center.

Background

With the explosive increase of Data traffic and the general decrease of computational cost, global computational resources are greatly increased, and the power consumption of Internet Data Centers (IDCs) is also sharply increased. The proportion of the power consumption of the internet data center to the power consumption of the whole society rises year by year, and the power consumption of the internet data center exceeds 2.3% of the total domestic power consumption in 2020. Currently, China is continuously strengthening policy guidance, the requirement on energy consumption of an internet data center is increasingly strict, and energy conservation of the internet data center is not slow along with popularization of stricter energy consumption limitation to the whole country.

Therefore, a reasonable and reliable scheme is urgently needed, so that the energy consumption of the internet data center can be reduced, and the energy efficiency is improved.

Disclosure of Invention

The embodiment of the specification provides an energy efficiency optimization method for an internet data center, which can reduce energy consumption of the internet data center and improve energy efficiency.

In a first aspect, an embodiment of the present specification provides an energy efficiency optimization method for an internet data center, including: receiving a working condition recommendation instruction aiming at a target machine room of an internet data center; acquiring current actual power consumption and non-power consumption type target operation condition parameters of the target machine room according to the condition recommendation instruction; according to a preset parameter adjustment strategy, executing a plurality of power consumption prediction operations, wherein any power consumption prediction operation comprises adjusting values of a plurality of controllable parameters in the target operation condition parameters, and performing power consumption prediction on the target machine room according to the adjusted controllable parameters and other parameters in the target operation condition parameters by using a target power consumption prediction model to obtain first predicted power consumption; in response to that the minimum value of the first predicted power consumptions respectively corresponding to the plurality of times of power consumption prediction operations is smaller than the actual power consumption, determining a plurality of adjusted controllable parameters adopted in predicting the minimum value as a plurality of optimized parameters; and generating and outputting working condition recommendation information comprising the plurality of optimization parameters.

In some embodiments, the target machine room comprises a cooling system and a plurality of servers, and the target operating condition parameters comprise a plurality of parameters of the cooling system and the plurality of servers respectively.

In some embodiments, the method further comprises: determining the minimum value as an optimized power consumption; the working condition recommendation information further comprises the optimized power consumption.

In some embodiments, the receiving of the operating condition recommendation instruction for the target machine room of the internet data center includes: and receiving the working condition recommendation instruction triggered by the timing task.

In some embodiments, the target communication platform creates a target group chat related to the operation and maintenance of the target machine room; and the output comprises the working condition recommendation information of the plurality of optimization parameters, and comprises the following steps: and sending the working condition recommendation information to the target group chat through the target communication platform.

In some embodiments, the internet data center is locally deployed with a building dynamic ring system for equipment control; and the output comprises the working condition recommendation information of the plurality of optimization parameters, and comprises the following steps: and sending the working condition recommendation information to the building dynamic loop system.

In some embodiments, the receiving of the operating condition recommendation instruction for the target machine room of the internet data center includes: receiving the working condition recommendation instruction triggered by a user; the output comprises working condition recommendation information of the plurality of optimization parameters, and the working condition recommendation information comprises: and providing the working condition recommendation information for the user.

In some embodiments, the internet data center is locally deployed with a facility-side system for periodically collecting data from the cooling system and a server-side system for periodically collecting data from the plurality of servers; and before the receiving of the working condition recommendation instruction for the target machine room of the internet data center, the method further comprises the following steps: receiving power consumption data of the cooling system reported by the facility side system and a first operation condition parameter of a non-power consumption class; and receiving the power consumption data of the plurality of servers and the second operation working condition parameters of the non-power consumption class reported by the server side system.

In some embodiments, after the receiving the power consumption data of the cooling system and the first operating condition parameter of the non-power consumption class reported by the facility-side system, the method further includes: determining whether a plurality of parameters in the first operation condition parameters meet corresponding early warning conditions; and if at least one parameter in the plurality of parameters meets the corresponding early warning condition, generating and outputting early warning information related to the at least one parameter.

In some embodiments, the method further comprises: obtaining second predicted power consumption of the target machine room at present, wherein the second predicted power consumption is predicted based on historical operating condition parameters of the target machine room; determining a difference between the second predicted power consumption and the actual power consumption; and if the difference exceeds a difference threshold, taking the current actual power consumption and the target operation condition parameter as training data to train the target power consumption prediction model.

In some embodiments, the plurality of parameters of the cooling system include a plurality of: the temperature of the chilled water return, the flow of a chilled water main line, the flow of a bypass, the pressure difference between a water separator and a water collector, the pressure difference at the tail end of a chilled water branch, the opening of a chilled water terminal device cold water valve, the opening of a bypass valve, the outlet water temperature of a cooling tower, the outdoor dry bulb temperature, the outdoor wet bulb temperature, the outdoor relative humidity, the frequency of the cooling tower, the frequency of a cooling circulating water pump, the frequency of a freezing circulating water pump, the frequency of a tail end air conditioner, the frequency of a refrigerator, the operation percentage of the refrigerator, the temperature of the chilled water return, the output power of the refrigerator, the operation mode and the air supply temperature of the air conditioner; the multiple parameters of the plurality of servers comprise server type, IT load rate, server inlet air temperature and server fan rotating speed.

In some embodiments, a single said controllable parameter comprises any one of: cooling tower frequency, cooling circulating water pump frequency, freezing circulating water pump frequency, tail end air conditioner frequency, cold machine frequency, server air inlet temperature, server fan rotating speed and air conditioner air supply temperature.

In a second aspect, an embodiment of the present specification provides an energy efficiency optimization apparatus for an internet data center, including: the receiving unit is configured to receive a working condition recommendation instruction aiming at a target machine room of the Internet data center; the obtaining unit is configured to obtain current actual power consumption and non-power consumption type target operation condition parameters of the target machine room according to the condition recommendation instruction; the prediction unit is configured to execute a plurality of power consumption prediction operations according to a preset parameter adjustment strategy, wherein any power consumption prediction operation comprises the steps of adjusting the values of a plurality of controllable parameters in the target operation condition parameters, and performing power consumption prediction on the target machine room according to the adjusted plurality of controllable parameters and other parameters in the target operation condition parameters by using a target power consumption prediction model to obtain first predicted power consumption; a determining unit configured to determine, as a plurality of optimized parameters, a plurality of adjusted controllable parameters to be used in predicting a minimum value in response to the minimum value being smaller than the actual power consumption in the first predicted power consumptions respectively corresponding to the plurality of power consumption prediction operations; and the recommending unit is configured to generate and output the working condition recommending information comprising the plurality of items of optimization parameters.

In a third aspect, the present specification provides a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed in a computer, the computer is caused to execute the method described in any implementation manner in the first aspect.

In a fourth aspect, the present specification provides a computing device, including a memory and a processor, where the memory stores executable code, and the processor executes the executable code to implement the method described in any implementation manner of the first aspect.

In a fifth aspect, the present specification provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the method described in any implementation manner of the first aspect.

According to the scheme provided by the embodiment of the specification, the current actual power consumption and non-power consumption type target operation condition parameters of the target machine room can be obtained according to the received working condition recommendation instruction for the target machine room of the internet data center, then, the power consumption prediction operations can be executed for a plurality of times according to a preset parameter adjustment strategy so as to obtain the first predicted power consumption corresponding to the power consumption prediction operations respectively, then, the minimum value of the first predicted power consumption corresponding to the power consumption prediction operations respectively can be responded to and is smaller than the actual power consumption, a plurality of controlled parameters which are adopted when the minimum value is predicted are determined to be a plurality of optimized parameters, and then, the working condition recommendation information comprising the optimized parameters can be generated and output. Therefore, the working condition recommendation information can be used for equipment control of the target machine room, and power consumption of the target machine room can be effectively reduced, so that energy consumption of the internet data center can be reduced, and energy efficiency is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments disclosed in the present specification, the drawings needed to be used in the description of the embodiments will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments disclosed in the present specification, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.

FIG. 1 is an exemplary system architecture diagram to which some embodiments of the present description may be applied;

FIG. 2 is a flow diagram of one embodiment of a method for energy efficiency optimization for an Internet data center;

FIG. 3 is a schematic diagram of a condition recommendation information transmission process;

FIG. 4 is a diagram illustrating the effect of behavior recommendation information in target group chat;

FIG. 5 is a flow diagram of one embodiment of a method for energy efficiency optimization for an Internet data center;

FIG. 6 is a diagram illustrating the effect of inter-package temperature-adjusting recommended content in a target group chat;

fig. 7 is a schematic configuration diagram of an energy efficiency optimization apparatus for an internet data center.

Detailed Description

The present specification will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. The described embodiments are only a subset of the embodiments described herein and not all embodiments described herein. All other embodiments obtained by a person skilled in the art based on the embodiments in the present specification without any inventive step are within the scope of the present application.

It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings. The embodiments and features of the embodiments in the present description may be combined with each other without conflict.

As mentioned above, under the continuous policy guidance of our country, the requirement for energy consumption of the internet data center is increasingly strict, and with the more strict energy consumption limitation spreading to the whole country, the energy saving of the internet data center is still unbearable.

Based on this, some embodiments of the present disclosure provide an energy efficiency optimization method for an internet data center, which may reduce energy consumption of the internet data center and improve energy efficiency. In particular, FIG. 1 illustrates an exemplary system architecture diagram suitable for use with these embodiments.

As shown in fig. 1, the system architecture may include an internet data center and an energy efficiency optimization platform for energy efficiency optimization for the internet data center. Wherein, the energy efficiency optimization platform can be a cloud platform.

The internet data center may include a plurality of rooms, such as room 1, room 2, …, room N shown in fig. 1, where N is the total number of rooms in the internet data center. Further, any of the plurality of rooms may include a cooling system and IT equipment. The cooling system may include, for example, a chilled water system and an AHU (Air Handling Unit) system. The IT device may include, for example, a plurality of servers.

In practice, the energy efficiency optimization platform can receive working condition recommendation instructions for a target machine room of the internet data center in real time or in a timing mode. The target machine room may be one or more of the plurality of machine rooms, and is not particularly limited herein. A condition may refer to an operating state of a device under conditions directly related to its action. And then, the energy efficiency optimization platform can acquire current actual power consumption and non-power consumption target operation condition parameters of the target machine room according to the condition recommendation instruction. The target operating condition parameters may include, for example, various parameters of the cooling system. Further, the target operating condition parameters may also include a plurality of parameters for a plurality of servers. Then, the energy efficiency optimization platform can determine working condition recommendation information aiming at the target machine room according to a preset parameter adjustment strategy, the actual power consumption and the target operation working condition parameters. Then, the energy efficiency optimization platform can output the working condition recommendation information.

The parameter adjustment policy may include, for example, the number of times of adjustment, a controllable parameter type, an adjustment rule corresponding to the controllable parameter type, and the like. Wherein the controllable parameter type may be used to define the controllable parameter to be adjusted. A controllable parameter is a parameter that can be controlled. For example, the cooling tower frequency, the cooling cycle water pump frequency, the chilled cycle water pump frequency, the terminal air conditioning frequency, and the chiller frequency are among the controllable parameters. The adjustment rules may include, for example, the manner of adjustment (e.g., increase or decrease), the step size of the adjustment (the value of each increase or decrease), and the safety range of the parameter (the range within which the value of the adjusted parameter should fall).

It should be understood that the parameter adjustment strategy may be configured according to actual requirements, and is not specifically limited herein.

In the following, taking the target machine room as the machine room 1 and the receiving mode of the working condition recommendation instruction as the timing receiving example, the energy efficiency optimization process executed by the energy efficiency optimization platform is further described.

Specifically, the energy efficiency optimization platform may be configured with a timing task for triggering the working condition recommendation instruction, the timing task may trigger the working condition recommendation instruction once every set time (for example, 30 minutes or 1 hour, and the like), and the energy efficiency optimization platform may receive the working condition recommendation instruction for the machine room 1 triggered by the timing task.

Then, the energy efficiency optimization platform can obtain the current actual power consumption and non-power consumption type target operation condition parameters of the machine room 1. Then, the energy efficiency optimization platform may execute a plurality of power consumption prediction operations according to a preset parameter adjustment strategy to obtain first predicted power consumption corresponding to the plurality of power consumption prediction operations respectively. As an example, the power consumption prediction operation at any time may include adjusting values of a plurality of controllable parameters in the target operating condition parameters, and performing power consumption prediction on the target machine room according to the adjusted plurality of controllable parameters and other parameters in the target operating condition parameters by using the target power consumption prediction model to obtain the first predicted power consumption.

Then, the energy efficiency optimization platform may respond that a minimum value of the first predicted power consumptions respectively corresponding to the plurality of power consumption prediction operations is smaller than the actual power consumption, and determine the working condition recommendation information according to the minimum value. For example, the adjusted controllable parameters used in predicting the minimum value may be determined as optimized parameters, and the condition recommendation information including the optimized parameters may be generated.

Then, the energy efficiency optimization platform can output the working condition recommendation information. As an example, as shown in fig. 1, the system architecture may further include a target communication platform, the energy efficiency optimization platform may be communicatively connected to the target communication platform, and a target group chat related to operation and maintenance of the computer room 1 may be created in the target communication platform. It should be understood that the group members of the target group chat may include operation and maintenance personnel of the machine room 1. Based on this, the energy efficiency optimization platform can send the working condition recommendation information to the target group chat through the target communication platform, so that the operation and maintenance personnel can obtain the working condition recommendation information, and further, based on the working condition recommendation information, the equipment control is performed on the machine room 1.

By adopting the energy efficiency optimization process described above, the energy consumption of the internet data center can be reduced and the energy efficiency can be improved by reducing the power consumption of the machine room.

The following describes specific implementation steps of the above method with reference to specific examples.

Referring to fig. 2, a flow 200 of one embodiment of a method for energy efficiency optimization for an internet data center is shown. The execution subject of the method can be an energy efficiency optimization platform as shown in fig. 1. The method comprises the following steps:

step 202, receiving a working condition recommendation instruction aiming at a target machine room of an internet data center;

step 204, obtaining current actual power consumption and non-power consumption target operation condition parameters of a target machine room according to the condition recommendation instruction;

step 206, according to a preset parameter adjustment strategy, executing a plurality of power consumption prediction operations, wherein any power consumption prediction operation comprises adjusting values of a plurality of controllable parameters in the target operation condition parameters, and performing power consumption prediction on the target machine room according to the adjusted plurality of controllable parameters and other parameters in the target operation condition parameters by using a target power consumption prediction model to obtain first predicted power consumption;

step 208, in response to that the minimum value of the first predicted power consumptions respectively corresponding to the plurality of times of power consumption prediction operations is smaller than the actual power consumption, determining a plurality of adjusted controllable parameters adopted in predicting the minimum value as a plurality of optimized parameters;

step 210, generating and outputting working condition recommendation information including a plurality of items of optimization parameters.

The above steps are further explained below.

In step 202, the energy efficiency optimization platform may receive a condition recommendation command for a target machine room of the internet data center in a timed or real-time manner. The target machine room may include a cooling system and a plurality of servers, among other things. The cooling system may include, for example, a chilled water system and an AHU system.

Specifically, in step 202, the energy efficiency optimization platform may receive a condition recommendation instruction triggered by the timed task, or receive a condition recommendation instruction triggered by a user. The condition recommendation command may include a machine room name of the target machine room, a building where the target machine room is located, and the like.

Next, in step 204, the energy efficiency optimization platform may obtain current actual power consumption and non-power consumption target operation condition parameters of the target machine room according to the condition recommendation instruction.

In practice, the internet data center can report power consumption data and non-power consumption type operation condition parameters of the target machine room to the energy efficiency optimization platform in a timing mode. The current actual power consumption of the target machine room can be determined according to the latest reported power consumption data of the target machine room by the internet data center. The target operation condition parameters can be determined according to the operation condition parameters of the target machine room non-power consumption type reported by the internet data center.

The target operating condition parameters may include, for example, but are not limited to, various parameters of the cooling system. The plurality of parameters may include, for example, a plurality of: chilled water return temperature, chilled water main flow, bypass flow, pressure difference between a water separator and a water collector, chilled water branch end pressure difference, chilled water terminal equipment cold water valve opening, bypass valve opening, cooling tower outlet water temperature, outdoor dry bulb temperature, outdoor wet bulb temperature, outdoor relative humidity, cooling tower frequency, cooling circulating water pump frequency, chilled circulating water pump frequency, end air conditioning frequency, chiller operation percentage, chilled water return temperature, chiller output power, operation mode, air conditioner supply temperature, and the like.

It should be understood that the above listed parameters of the cooling system are merely exemplary parameters. In practice, the parameter types of the parameters to be included in the target operating condition parameters can be configured in advance according to actual requirements.

Next, in step 206, the energy efficiency optimization platform may perform several power consumption prediction operations according to a preset parameter adjustment policy. The power consumption prediction operation of any time can include, for example, adjusting values of a plurality of controllable parameters in the target operation condition parameters, and performing power consumption prediction on the target machine room by using the target power consumption prediction model according to the adjusted plurality of controllable parameters and other parameters in the target operation condition parameters to obtain first predicted power consumption.

The parameter adjustment policy may include, for example, the number of times of adjustment, a controllable parameter type, an adjustment rule corresponding to the controllable parameter type, and the like. Here, for the specific explanation of the parameter adjustment strategy, reference may be made to the related description in the foregoing, and details are not repeated here.

When the target operating condition parameters include multiple parameters of the cooling system, the single controllable parameter may include, for example, any one of: cooling tower frequency, cooling circulating water pump frequency, chilled circulating water pump frequency, terminal air conditioning frequency, chiller frequency, air conditioning supply air temperature, and the like.

Additionally, the target power consumption prediction model may include a facility-side power consumption model related to the cooling system. Further, when the cooling system includes a chilled water system and an AHU system, the facility-side power consumption model may include a chilled water system power consumption model and an AHU system power consumption model. Still further, the chilled water system may relate to a plurality of refrigeration equipment types, and the chilled water system power consumption model may include equipment power consumption models corresponding to the plurality of refrigeration equipment types, respectively. In addition, the AHU system may also relate to a plurality of refrigeration device types, and the AHU system power consumption model may include device power consumption models corresponding to the plurality of refrigeration device types, respectively. It should be appreciated that the device power consumption model may be used to predict the device power consumption of a refrigeration device for its corresponding refrigeration device type.

By way of example, the cooling system may include a plurality of refrigeration units, wherein any refrigeration unit may include a plurality of refrigeration devices. It is assumed that the plurality of refrigeration units include a 4# refrigeration unit, a 5# refrigeration unit, and a 6# refrigeration unit, and the 3 refrigeration units each include one cooling circulation water pump and two cooling towers. Wherein, "4 #", "5 #" and "6 #" may be refrigeration unit numbers. Among the target operating condition parameters, the plurality of parameters of the cooling system may include, but are not limited to, the frequencies of the cooling circulation water pumps and the cooling towers of the 4# refrigeration unit, the 5# refrigeration unit, and the 6# refrigeration unit, respectively.

When the controllable parameter types in the parameter adjustment strategy comprise cooling circulating water pump frequency and cooling tower frequency, in the execution process of the power consumption prediction operation for any time, the energy efficiency optimization platform can adjust the frequencies of the cooling circulating water pumps and the cooling towers of the 4# refrigeration unit, the 5# refrigeration unit and the 6# refrigeration unit respectively according to the parameter adjustment strategy, and predict the power consumption of the target machine room by using the target power consumption prediction model according to the adjusted frequencies and other parameters in the target operation condition parameters to obtain first predicted power consumption.

Further, when power consumption prediction is performed on a target machine room, the target power consumption prediction model can be used for predicting the equipment power consumption of the cooling circulating water pumps and the cooling towers of the 4# refrigeration unit, the 5# refrigeration unit and the 6# refrigeration unit, and then the first predicted power consumption can be determined according to the predicted equipment power consumption.

After step 206 is executed, first predicted power consumptions respectively corresponding to the plurality of power consumption prediction operations may be obtained. Then, the energy efficiency optimization platform may determine whether a minimum value of the obtained first predicted power consumptions is smaller than a current actual power consumption of the target machine room. If the minimum value is less than the actual power consumption, step 208 may be performed; otherwise, the execution of the process 200 at this time may be ended.

In step 208, the energy efficiency optimization platform may determine, as a plurality of optimized parameters, the plurality of adjusted controllable parameters used in predicting the minimum value in response to the minimum value being smaller than the current actual power consumption of the target equipment room.

Next, in step 210, the energy efficiency optimization platform may generate condition recommendation information including the optimization parameters, and output the condition recommendation information.

As described above, the energy efficiency optimization platform may receive a condition recommendation instruction triggered by a timing task, or receive a condition recommendation instruction triggered by a user.

Under the condition that the energy efficiency optimization platform receives a working condition recommendation instruction triggered by a timing task, the energy efficiency optimization platform can adopt various working condition recommendation modes.

In one example, the energy efficiency optimization platform may be communicatively coupled to a target communication platform, and the target communication platform may create a target group chat related to operation and maintenance of a target machine room. Based on this, in step 210, the energy efficiency optimization platform may send the working condition recommendation information to the target group chat via the target communication platform as shown in fig. 3. It should be noted that the target communication platform may include a robot (for example, the second recommended condition shown in fig. 4) responsible for condition recommendation, and the target communication platform may send the condition recommendation information to the target group chat by using the robot.

Continuing with the example of the 4# refrigeration unit, the 5# refrigeration unit and the 6# refrigeration unit as described above, it is assumed that the optimization parameters in the condition recommendation information include: the optimized frequency of a cooling circulating water pump of the 4# refrigeration unit is 30.2HZ (hertz), and the optimized frequency of a cooling tower is 44.0 HZ; the optimized frequency of a cooling circulating water pump of the 5# refrigeration unit is 29.99HZ, and the optimized frequency of a cooling tower is 45.0 HZ; the optimized frequency of the cooling circulating water pump of the 6# refrigeration unit is 0.0HZ, and the optimized frequency of the cooling tower is 0.0 HZ. The effect of these optimized frequencies may be shown as reference numeral 401 in fig. 4.

Optionally, the condition recommendation information may include other information items besides the optimization parameter. For example, the minimum value may be determined as the optimized power consumption, and the condition recommendation information may further include the optimized power consumption (e.g., the optimized power consumption 601.38KW (kilowatt) shown in fig. 4).

For another example, the condition recommendation information may further include a machine room name of the target machine room (e.g., EA118 shown in fig. 4), a building where the target machine room is located, a recommended time, a current condition of the target machine room (e.g., current frequencies of the cooling circulation water pump and the cooling tower of the 4# refrigeration unit, the 5# refrigeration unit, and the 6# refrigeration unit shown in fig. 4), a current actual power consumption of the refrigeration equipment in the cooling system, a current equipment optimized power consumption of the refrigeration equipment, a current actual power consumption of the target machine room, and/or a current second predicted power consumption of the target machine room, and so on.

It should be noted that the current optimal power consumption of the refrigeration equipment is the predicted power consumption of the refrigeration equipment in the prediction process of the current optimal power consumption of the target machine room. The current second predicted power consumption of the target machine room can be obtained by predicting according to the historical operating condition parameters of the target machine room in advance. In practice, the energy efficiency optimization platform can predict the power consumption of the target machine room in a future period of time according to the acquired operating condition parameters of the target machine room at regular time (every 2.5, 10 or 20 minutes, etc.).

In addition, when the target operating condition parameters include only parameters of the cooling system, the optimized power consumption determined according to the minimum value may be regarded as the facility-side optimized power consumption, the current actual power consumption of the target machine room may be regarded as the facility-side actual power consumption, and the current second predicted power consumption of the target machine room may be regarded as the facility-side predicted power consumption.

It should be understood that the group members of the target group chat may include operation and maintenance personnel of the target machine room, and the operation and maintenance personnel may obtain the condition recommendation information by sending the condition recommendation information to the target group chat, so as to perform device control on the target machine room according to the condition recommendation information. Therefore, equipment control can be carried out on the target machine room in an artificial control mode, and the power consumption of the target machine room is reduced, so that the energy consumption of the internet data center is reduced, and the energy efficiency is improved.

In another example, an internet data center may be locally deployed with a building dynamic ring system for device control. Based on the condition, the energy efficiency optimization platform can send the working condition recommendation information to the building dynamic loop system, so that the building dynamic loop system automatically controls equipment of the target machine room according to the working condition recommendation information. Therefore, equipment control can be carried out on the target machine room in an automatic control mode, and the power consumption of the target machine room is reduced, so that the energy consumption of the internet data center is reduced, and the energy efficiency is improved.

Under the condition that the energy efficiency optimization platform receives a condition recommendation instruction triggered by a user, in step 210, the energy efficiency optimization platform may provide condition recommendation information to the user. By way of example, the energy efficiency optimization platform may provide a target interface for viewing the condition recommendation information to the user, and the user may access the target interface through the user equipment and trigger the condition recommendation instruction on the target interface. Based on this, the energy efficiency optimization platform can send the working condition recommendation information to the user equipment, so that the user equipment displays the working condition recommendation information to the user.

According to the scheme provided by the embodiment corresponding to fig. 2, the current actual power consumption and non-power consumption type target operation condition parameters of the target machine room can be obtained according to the received working condition recommendation instruction for the target machine room of the internet data center, then, the power consumption prediction operations can be executed for a plurality of times according to the preset parameter adjustment strategy so as to obtain the first prediction power consumptions respectively corresponding to the power consumption prediction operations for the plurality of times, then, the minimum value of the first prediction power consumptions respectively corresponding to the power consumption prediction operations for the plurality of times can be responded to, the adjusted controllable parameters adopted when the minimum value is predicted are determined as the optimized parameters, and then, the working condition recommendation information comprising the optimized parameters can be generated and output. Therefore, the working condition recommendation information can be used for equipment control of the target machine room, and power consumption of the target machine room can be effectively reduced, so that energy consumption of the internet data center is reduced, and energy efficiency is improved.

In an embodiment, a facility-side system may be locally deployed in the internet data center, the facility-side system may be configured to perform data acquisition on the cooling system in a timing manner, and before step 202, the energy efficiency optimization platform may further receive power consumption data of the cooling system and a first operation condition parameter of a non-power consumption class, which are reported by the facility-side system.

In practice, variations in IT load rates and power consumption characteristics of different types of servers also affect the power consumption of internet data centers. The IT load rate is the actual power consumption of the server/the designed power consumption of the server. In order to further reduce the energy consumption of the internet data center and improve the energy efficiency, the operation condition parameters of the cooling system and the server can be comprehensively considered.

Next, taking an example that the current non-power consumption type target operation condition parameters of the target machine room include multiple parameters of the cooling system and the servers, a method for optimizing the energy efficiency of the internet data center is further described.

Referring to fig. 5, a flow 500 of one embodiment of a method for energy efficiency optimization for an internet data center is shown. The execution subject of the method can be an energy efficiency optimization platform as shown in fig. 1. The method comprises the following steps:

step 502, receiving a working condition recommendation instruction for a target machine room of an internet data center, wherein the target machine room comprises a cooling system and a plurality of servers;

step 504, obtaining current actual power consumption and non-power consumption type target operation condition parameters of a target machine room according to the condition recommendation instruction, wherein the target operation condition parameters comprise respective multiple parameters of a cooling system and a plurality of servers;

step 506, according to a preset parameter adjustment strategy, executing a plurality of power consumption prediction operations, wherein any power consumption prediction operation comprises adjusting values of a plurality of controllable parameters in the target operation condition parameters, and performing power consumption prediction on the target machine room according to the adjusted plurality of controllable parameters and other parameters in the target operation condition parameters by using a target power consumption prediction model to obtain first predicted power consumption;

step 508, in response to that the minimum value of the first predicted power consumptions respectively corresponding to the plurality of times of power consumption prediction operations is smaller than the actual power consumption, determining a plurality of adjusted controllable parameters adopted in predicting the minimum value as a plurality of optimized parameters;

step 510, generating and outputting condition recommendation information including a plurality of optimization parameters.

The specific implementation manner of the steps 502-510 is similar to that of the steps 202-210 in the corresponding embodiment of fig. 2, and reference may be made to the relevant description in the corresponding embodiment of fig. 2, which will not be described in detail herein.

IT should be noted that the target operating condition parameters include, in addition to the parameters of the cooling system described above, parameters of the servers, which may include, for example, server type, IT load rate, server intake air temperature, and server fan speed, etc.

The single controllable parameter may comprise any one of: cooling tower frequency, cooling circulating water pump frequency, refrigeration circulating water pump frequency, terminal air conditioning frequency, chiller frequency, server inlet air temperature, server fan speed, air conditioner supply air temperature, and the like.

The target power consumption prediction model includes a server-side model in addition to the facility-side power consumption model described above. Further, the server-side model may include, for example, server power consumption models corresponding to a plurality of server types, respectively.

The condition recommendation information may include condition recommendation content on the server side in addition to condition recommendation content on the facility side.

In practice, the target room may include a plurality of bays, each of which may include multiple ones of the plurality of servers described above. Based on this, the air-conditioning air supply temperature, the server air inlet temperature and the outdoor relative humidity in the target operation condition parameters may be the device-level temperature or the inter-package-level temperature, and are not specifically limited herein.

In one embodiment, the energy efficiency optimization platform may support inter-packet level temperature adjustment policy delivery. As an example, when the target operating condition parameter includes the server intake air temperature of each of the plurality of bays, and the controllable parameter type in the parameter adjustment strategy includes the server intake air temperature, the server intake air temperature of each of the plurality of bays in the target operating condition parameter may be adjusted during the execution of the power consumption prediction operation for any number of times. Based on the above, the condition recommendation information may include inter-package temperature adjustment recommendation content on the server side in addition to the condition recommendation content on the facility side.

Taking the server intake air temperature as an example, the recommended inter-pack temperature adjustment content on the server side may include the current server optimized intake air temperature (may be simply referred to as recommended temperature) of the plurality of inter-packs. Further, the temperature recommendation content may further include a machine room name of the target machine room, a building where the target machine room is located, a current average temperature of the server intake air, a preferred server intake air temperature, a current expected energy saving, names of the plurality of bays, a current server intake air temperature of the plurality of bays (which may be simply referred to as a current temperature), a current outdoor relative humidity (which may be simply referred to as a relative humidity) of the plurality of bays, and/or a temperature rise and fall value, and the like.

The average temperature of the current server intake air may be an average value of the current temperatures among the plurality of packets. The preferred server inlet air temperature may be an average of the recommended temperatures among the plurality of packets. The current expected energy saving can be calculated according to the current actual power consumption and the optimized power consumption of the target machine room. The temperature rise and fall value can be calculated according to the current temperature and the recommended temperature of the plurality of the packages.

In one embodiment, the condition recommendation contents on the facility side and the server side can be pushed respectively. Taking the example of pushing the working condition recommendation information to the target group chat as described above, the target communication platform may use the same or different robots to push the working condition recommendation contents on the facility side and the server side.

Assuming that the name of the robot responsible for pushing the working condition recommended content at the server side is "recommended little two", the working condition recommended content at the server side is specifically inter-package temperature adjustment recommended content, and the inter-package temperature adjustment recommendation relates to the intake air temperature of the server, and the display effect of the inter-package temperature adjustment recommended content in the target group chat can be as shown in fig. 6.

In one embodiment, the internet data center may be locally deployed with a facility-side system that may be used to periodically collect data from the cooling system and a server-side system that may be used to periodically collect data from multiple servers. Based on this, before step 502, the energy efficiency optimization platform may receive the power consumption data of the cooling system and the first operating condition parameters of the non-power consumption class reported by the facility-side system, and receive the power consumption data of the servers and the second operating condition parameters of the non-power consumption class reported by the server-side system.

The scheme provided by the embodiment corresponding to fig. 5 may comprehensively consider the operating condition parameters of the cooling system and the servers in the target machine room, and implement full link optimization of the cooling system and the servers, so that not only the Power Usage Efficiency (PUE) of the internet data center can be reduced, but also the energy consumption of the internet data center can be reduced, and the total energy consumption of the internet data center is lower.

It should be noted that, for the measurement of the energy consumption of the internet data center, PUE is a measurement index widely used by the internet data center industry at home and abroad at present. PUE can be defined as the ratio of the total equipment energy consumption of the internet data center to the energy consumption of IT equipment. PUE is a ratio, with a benchmark of generally 2, with closer to 1 indicating better energy efficiency levels. The lower the PUE value is, the lower the energy consumption of the internet data center for use outside IT equipment is, and the more energy is saved.

In one embodiment, the target power consumption prediction model has a possibility of low prediction accuracy. Based on the method, after the energy efficiency optimization platform obtains the current actual power consumption and the target operation condition parameters of the target machine room, the second predicted power consumption of the target machine room can be obtained through prediction in advance, and the second predicted power consumption can be obtained through prediction based on the historical operation condition parameters of the target machine room. The energy efficiency optimization platform may then determine a difference between the second predicted power consumption and the actual power consumption. If the difference exceeds the difference threshold, the energy efficiency optimization platform can use the actual power consumption and the target operation condition parameters as training data to train a target power consumption prediction model. Therefore, the rapid training and the optimized result output of the target power consumption prediction model can be realized under the condition of not excessively depending on the type, the number and the training period of the measuring points.

In one embodiment, in order to maximize an extended AI (Artificial Intelligence) control validity time, etc., an intelligent early warning may be performed for the cooling system.

Specifically, after power consumption data of the cooling system and first operating condition parameters reported by the facility-side system are received, it may be determined whether a plurality of parameters in the first operating condition parameters satisfy corresponding early warning conditions, and if at least one parameter in the plurality of parameters satisfies the corresponding early warning conditions, early warning information related to the at least one parameter may be generated and output.

Specifically, for any one of the several parameters, the early warning condition corresponding to the parameter may include, for example, a first threshold and a second threshold, where the second threshold is greater than the first threshold. If the value of the parameter is greater than or equal to the first threshold and smaller than the second threshold, the parameter can be determined to meet the corresponding early warning condition; otherwise, the parameter can be determined not to satisfy the corresponding early warning condition.

When the early warning information related to the at least one parameter is output, in an example, the early warning information may be sent to the target group chat via the target communication platform, so that the operation and maintenance personnel of the target machine room can know the device abnormality of the target machine room in time, and quickly complete the abnormality relief.

Further, in order to enable operation and maintenance personnel of the target machine room to complete exception resolution more quickly, for the parameter in the at least one parameter, the energy efficiency optimization platform can adjust the value of the parameter according to a preset control point optimization strategy. Based on this, the warning information may further include the adjusted at least one parameter.

In practice, the energy efficiency optimization platform may also support users to view other information related to the internet data center.

In one embodiment, the energy efficiency optimization platform may receive a viewing request of a user for the management and control data of the target machine room, and provide the management and control data of the target machine room to the user according to the viewing request. The management and control data may include detail data, AI benefits, and/or real-time status of the target machine room.

The detailed data may include, for example, a current AI management and control status of the target machine room (e.g., in-take or not-take), a cooling structure of the cooling system, a region to which the target machine room belongs, a current actual power consumption, a current design power consumption, and/or a current IT load rate of the IT equipment, a current actual power consumption, a current design power consumption, and/or a current load rate of the utility power, and/or an average wet bulb temperature, and so on.

The AI benefits may include, current AI cumulative intervention times of the target machine room, AI cumulative intervention duration, average relative error of the facility-side power consumption model over a recent period of time (e.g., a recent day), average relative error of PUE prediction over a recent period of time, AI management and control acceptance unit count, cumulative energy savings, prediction of optimization over a recent period of time to reduce PUE, maximum duration of continuous AI hosting, and/or prediction of optimization over a recent period of time to conserve energy, and so on.

The real-time status may include, for example, CLF (Cooling Load Factor), PLF (Power Load Factor), and/or PUE of the target room for a recent period of time (e.g., a recent hour), and/or the like. The CLF may be defined as the ratio of refrigeration equipment power consumption to server power consumption in an internet data center. The PLF may be defined as the ratio of power consumption of the power supply and distribution system to power consumption of the server in the internet data center.

In one embodiment, the energy efficiency optimization platform may receive a user's view request for an actual PUE, an IT load rate, a predicted PUE, and/or an optimized PUE for a target machine room over a target time period. The view request may include the target time period. Based on this, the energy efficiency optimization platform can obtain the PUE data of the target machine room in the target time period according to the viewing request, and provide the PUE data to the user. Wherein the PUE data comprises an actual PUE, an IT load rate, a predicted PUE, and/or an optimized PUE. In addition, the PUE data may be data in the form of text, graph, or the like.

It should be understood that in performing the PUE calculation, the reference factor for the actual PUE includes a corresponding actual power consumption, the reference factor for the predicted PUE includes a corresponding second predicted power consumption (which may be referred to simply as predicted power consumption), and the reference factor for the optimized PUE includes a corresponding optimized power consumption.

Optionally, the target time period may include a plurality of sub-time periods, and the PUE data may specifically include an actual PUE, an IT load rate, a predicted PUE, and/or an optimized PUE of the target equipment room in each of the plurality of sub-time periods.

In one embodiment, the energy efficiency optimization platform may receive a user's viewing request for actual power consumption, predicted power consumption, and/or optimized power consumption of a target object of a target computer room within a target time period. Further, the energy efficiency optimization platform may receive the viewing request after providing the PUE data to the user. The view request may include the target object. The target object may be, among other things, a building, a system (e.g., a cooling system, etc.), a unit (e.g., a refrigeration unit, etc.), or a device (e.g., a cooling tower, etc.).

Based on this, the energy efficiency optimization platform can obtain power consumption data of the target machine room in the target time period according to the viewing request, and provide the power consumption data for the user. Wherein the power consumption data comprises actual power consumption, predicted power consumption and/or optimized power consumption. It should be understood that, when the target time period includes a plurality of sub-time periods, the power consumption data may specifically include actual power consumption, predicted power consumption, and/or optimized power consumption of the target computer room in each of the plurality of sub-time periods. In addition, the power consumption data may be data in the form of text or a graph or the like.

In one embodiment, the target computer room may have several predicted records related to server power consumption, which may correspond to timestamps. The energy efficiency optimization platform can receive a viewing request of a user for the prediction record in the target time period, acquire the prediction record of the corresponding timestamp in the target time period according to the viewing request, and provide the acquired prediction record for the user.

Specifically, the user may be provided with a timestamp option corresponding to each acquired predicted record, and the record content of the predicted record with the latest predicted time in each predicted record. The user can check the prediction record corresponding to the time stamp option by triggering the time stamp option.

Any of the prediction records may include the total power consumption data of the server in the target equipment room. The server total power consumption data can comprise the server total power consumption of the target machine room at the air supply temperatures of the plurality of air conditioners. The total power consumption of the server of the target machine room at the air supply temperature of the plurality of air conditioners can be data in a chart form. Optionally, the server total power consumption data may further include a temperature range of the computer room, and/or a coverage rate of the server-side power consumption model, and the like.

Further, the arbitrary prediction record may further include temperature options corresponding to the air supply temperatures of the plurality of air conditioners, and a list of high temperature servers corresponding to a certain temperature option. The user can check the high-temperature server list corresponding to the temperature option by triggering the temperature option.

The list of high temperature servers may include, for example, the server identification, model, parent, server inlet temperature, power consumption, etc. of each server in the target room for which the server inlet temperature reaches a temperature threshold (e.g., 30℃.) at its corresponding air conditioner supply temperature.

In one embodiment, the energy efficiency optimization platform may support a user to modify the plurality of prediction records. Specifically, the energy efficiency optimization platform may receive a modification request of a user for the target prediction record, and modify the target prediction record according to the modification request. It should be understood that the target prediction record is one of the prediction records described above.

In one embodiment, the energy efficiency optimization platform may support a user to view a list of models for any room, such as a cooling side model list and/or a server side model list.

Taking the model list of the cooling side as an example, the energy efficiency optimization platform may receive a viewing request of a user for the model list of the cooling side of the target machine room, and provide the model list to the user according to the viewing request. The viewing request may include, for example, a machine room name of the target machine room. Further, the view request may also include, among other things, the model type (e.g., chilled water or AHU), building, and/or model name.

With further reference to fig. 7, the present specification provides one embodiment of an energy efficiency optimization apparatus for an internet data center, which may be applied to the energy efficiency optimization platform shown in fig. 1.

As shown in fig. 7, the energy efficiency optimization apparatus 700 for an internet data center according to the present embodiment includes: a receiving unit 701, an obtaining unit 702, a predicting unit 703, a determining unit 704, and a recommending unit 705. The receiving unit 701 is configured to receive a working condition recommendation instruction for a target machine room of the internet data center; the obtaining unit 702 is configured to obtain target operation condition parameters of current actual power consumption and non-power consumption of the target machine room according to the condition recommendation instruction; the prediction unit 703 is configured to perform a plurality of power consumption prediction operations according to a preset parameter adjustment strategy, where any power consumption prediction operation includes adjusting values of a plurality of controllable parameters in the target operating condition parameters, and performing power consumption prediction on the target machine room according to the adjusted plurality of controllable parameters and other parameters in the target operating condition parameters by using a target power consumption prediction model to obtain a first predicted power consumption; the determining unit 704 is configured to determine, as the plurality of optimized parameters, the adjusted plurality of controllable parameters employed in predicting the minimum value in response to the minimum value of the first predicted power consumptions respectively corresponding to the plurality of power consumption prediction operations being smaller than the actual power consumption; recommendation unit 705 is configured to generate and output condition recommendation information including the plurality of optimization parameters.

In some embodiments, the target machine room may include a cooling system and a plurality of servers, and the target operating condition parameters may include a plurality of parameters for each of the cooling system and the plurality of servers.

In some embodiments, the determining unit 704 may be further configured to: determining the minimum value as the optimized power consumption; the condition recommendation information may further include the optimized power consumption.

In some embodiments, the receiving unit 701 may be further configured to: and receiving a working condition recommendation instruction triggered by the timing task.

In some embodiments, the energy efficiency optimization platform may be in communication connection with a target communication platform, and the target communication platform may create a target group chat related to operation and maintenance of a target machine room; and the recommending unit 705 may be further configured to: and sending the working condition recommendation information to the target group chat through the target communication platform.

In some embodiments, the internet data center may be locally deployed with a building dynamic ring system for device control; and the recommending unit 705 may be further configured to: and sending the working condition recommendation information to a building dynamic loop system.

In some embodiments, the receiving unit 701 may be further configured to: receiving a working condition recommendation instruction triggered by a user; the recommending unit 705 may be further configured to: and providing the working condition recommendation information for a user.

In some embodiments, the internet data center may be locally deployed with a facility-side system and a server-side system, the facility-side system may be configured to perform data acquisition on the cooling system at regular time, and the server-side system may be configured to perform data acquisition on the plurality of servers at regular time; and the receiving unit 701 may be further configured to: before a working condition recommendation instruction for a target machine room of an internet data center is received, power consumption data of a cooling system and a first non-power-consumption operation working condition parameter reported by a facility side system are received; and receiving the power consumption data of the plurality of servers and the second operation working condition parameters of the non-power consumption class reported by the server side system.

In some embodiments, the determining unit 704 may be further configured to: after the receiving unit 701 receives power consumption data of the cooling system and first operation condition parameters of non-power consumption types reported by the facility side system, whether a plurality of parameters in the first operation condition parameters meet corresponding early warning conditions is determined; the apparatus 700 may further include: and an early warning unit (not shown in the figure) configured to generate and output early warning information related to at least one of the parameters if the at least one of the parameters meets a corresponding early warning condition.

In some embodiments, the obtaining unit 702 may be further configured to: acquiring second predicted power consumption of the target machine room which is predicted in advance, wherein the second predicted power consumption is predicted based on the historical operating condition parameters of the target machine room; the determining unit 704 may be further configured to: determining a difference between the second predicted power consumption and the actual power consumption; and if the difference exceeds the difference threshold, taking the current actual power consumption and the target operation condition parameter as training data to train a target power consumption prediction model.

In some embodiments, the plurality of parameters of the cooling system may include a plurality of: the temperature of the chilled water return, the flow of a chilled water main line, the flow of a bypass, the pressure difference between a water separator and a water collector, the pressure difference at the tail end of a chilled water branch, the opening of a chilled water terminal device cold water valve, the opening of a bypass valve, the outlet water temperature of a cooling tower, the outdoor dry bulb temperature, the outdoor wet bulb temperature, the outdoor relative humidity, the frequency of the cooling tower, the frequency of a cooling circulating water pump, the frequency of a freezing circulating water pump, the frequency of a tail end air conditioner, the frequency of a refrigerator, the operation percentage of the refrigerator, the temperature of the chilled water return, the output power of the refrigerator, the operation mode, the air supply temperature of the air conditioner, and the like; the plurality of parameters of the plurality of servers may include a server type, an IT load rate, a server intake air temperature, a server fan speed, and the like.

In some embodiments, the single controllable parameter may include any one of: cooling tower frequency, cooling circulating water pump frequency, refrigeration circulating water pump frequency, terminal air conditioning frequency, chiller frequency, server inlet air temperature, server fan speed, air conditioner supply air temperature, and the like.

In the embodiment corresponding to fig. 7, the detailed processing of each unit and the technical effect thereof can refer to the related description of the method embodiment in the foregoing, and are not repeated herein.

The present specification also provides a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed in a computer, the computer program causes the computer to execute the energy efficiency optimization method for an internet data center, which is respectively described in the above method embodiments.

The embodiment of the present specification further provides a computing device, which includes a memory and a processor, where the memory stores executable code, and the processor executes the executable code to implement the energy efficiency optimization method for an internet data center, which is respectively described in the above method embodiments.

The present specification further provides a computer program, wherein when the computer program is executed in a computer, the computer program causes the computer to execute the energy efficiency optimization method for an internet data center, which is respectively described in the above method embodiments.

Those skilled in the art will recognize that, in one or more of the examples described above, the functions described in the embodiments disclosed herein may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium.

In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The above-mentioned embodiments, objects, technical solutions and advantages of the embodiments disclosed in the present specification are further described in detail, it should be understood that the above-mentioned embodiments are only specific embodiments of the embodiments disclosed in the present specification, and are not intended to limit the scope of the embodiments disclosed in the present specification, and any modifications, equivalent substitutions, improvements and the like made on the basis of the technical solutions of the embodiments disclosed in the present specification should be included in the scope of the embodiments disclosed in the present specification.

Claims

1. An energy efficiency optimization method for an internet data center comprises the following steps:

receiving a working condition recommendation instruction aiming at a target machine room of an internet data center;

acquiring current actual power consumption and non-power consumption type target operation condition parameters of the target machine room according to the condition recommendation instruction;

according to a preset parameter adjustment strategy, executing a plurality of power consumption prediction operations, wherein any power consumption prediction operation comprises adjusting values of a plurality of controllable parameters in the target operation condition parameters, and performing power consumption prediction on the target machine room according to the adjusted controllable parameters and other parameters in the target operation condition parameters by using a target power consumption prediction model to obtain first predicted power consumption;

in response to that the minimum value of the first predicted power consumptions respectively corresponding to the plurality of times of power consumption prediction operations is smaller than the actual power consumption, determining a plurality of adjusted controllable parameters adopted in predicting the minimum value as a plurality of optimized parameters;

and generating and outputting working condition recommendation information comprising the plurality of optimization parameters.

2. The method of claim 1, wherein the target machine room comprises a cooling system and a plurality of servers, and the target operating condition parameters comprise a plurality of parameters for each of the cooling system and the plurality of servers.

3. The method of claim 1, further comprising:

determining the minimum value as an optimized power consumption;

the working condition recommendation information further comprises the optimized power consumption.

4. The method of claim 1, wherein the receiving of the condition recommendation instruction for the target machine room of the internet data center comprises:

and receiving the working condition recommendation instruction triggered by the timing task.

5. The method of claim 4, wherein the target communication platform creates a target group chat related to the operation and maintenance of the target machine room; and

the output comprises working condition recommendation information of the plurality of optimization parameters, and the working condition recommendation information comprises:

and sending the working condition recommendation information to the target group chat through the target communication platform.

6. The method of claim 4, wherein the internet data center is locally deployed with a building dynamic ring system for device control; and

and sending the working condition recommendation information to the building dynamic loop system.

7. The method of claim 1, wherein the receiving of the condition recommendation instruction for the target machine room of the internet data center comprises:

receiving the working condition recommendation instruction triggered by a user;

and providing the working condition recommendation information for the user.

8. The method of claim 2, wherein the internet data center is locally deployed with a facility-side system for periodically collecting data from the cooling system and a server-side system for periodically collecting data from the plurality of servers; and

before the receiving of the operating condition recommendation instruction for the target machine room of the internet data center, the method further includes:

receiving power consumption data of the cooling system reported by the facility side system and a first operation condition parameter of a non-power consumption class;

and receiving the power consumption data of the plurality of servers and the second operation working condition parameters of the non-power consumption class reported by the server side system.

9. The method of claim 8, wherein after receiving the power consumption data of the cooling system and the first operating condition parameter of the non-power consumption class reported by the facility-side system, further comprising:

determining whether a plurality of parameters in the first operation condition parameters meet corresponding early warning conditions;

and if at least one parameter in the plurality of parameters meets the corresponding early warning condition, generating and outputting early warning information related to the at least one parameter.

10. The method of claim 1, further comprising:

obtaining second predicted power consumption of the target machine room at present, wherein the second predicted power consumption is predicted based on historical operating condition parameters of the target machine room;

determining a difference between the second predicted power consumption and the actual power consumption;

and if the difference exceeds a difference threshold, taking the current actual power consumption and the target operation condition parameter as training data to train the target power consumption prediction model.

11. The method of claim 2, wherein,

the plurality of parameters of the cooling system include a plurality of: the temperature of the chilled water return, the flow of a chilled water main line, the flow of a bypass, the pressure difference between a water separator and a water collector, the pressure difference at the tail end of a chilled water branch, the opening of a chilled water terminal device cold water valve, the opening of a bypass valve, the outlet water temperature of a cooling tower, the outdoor dry bulb temperature, the outdoor wet bulb temperature, the outdoor relative humidity, the frequency of the cooling tower, the frequency of a cooling circulating water pump, the frequency of a freezing circulating water pump, the frequency of a tail end air conditioner, the frequency of a refrigerator, the operation percentage of the refrigerator, the temperature of the chilled water return, the output power of the refrigerator, the operation mode and the air supply temperature of the air conditioner;

the multiple parameters of the plurality of servers comprise server type, IT load rate, server inlet air temperature and server fan rotating speed.

12. An energy efficiency optimization device for an internet data center, comprising:

the receiving unit is configured to receive a working condition recommendation instruction aiming at a target machine room of the Internet data center;

the obtaining unit is configured to obtain current actual power consumption and non-power consumption type target operation condition parameters of the target machine room according to the condition recommendation instruction;

the prediction unit is configured to execute a plurality of power consumption prediction operations according to a preset parameter adjustment strategy, wherein any power consumption prediction operation comprises the steps of adjusting the values of a plurality of controllable parameters in the target operation condition parameters, and performing power consumption prediction on the target machine room according to the adjusted plurality of controllable parameters and other parameters in the target operation condition parameters by using a target power consumption prediction model to obtain first predicted power consumption;

a determining unit configured to determine, as a plurality of optimized parameters, a plurality of adjusted controllable parameters to be used in predicting a minimum value in response to the minimum value being smaller than the actual power consumption in the first predicted power consumptions respectively corresponding to the plurality of power consumption prediction operations;

and the recommending unit is configured to generate and output the working condition recommending information comprising the plurality of items of optimization parameters.

13. A computer-readable storage medium, on which a computer program is stored, wherein the computer program causes a computer to carry out the method of any one of claims 1-11, when the computer program is carried out in the computer.

14. A computing device comprising a memory and a processor, wherein the memory has stored therein executable code that when executed by the processor implements the method of any of claims 1-11.