WO2024002026A1

WO2024002026A1 - Energy-consumption optimization method, system and apparatus, and storage medium

Info

Publication number: WO2024002026A1
Application number: PCT/CN2023/102504
Authority: WO
Inventors: 刘丽霞; 孔鹏; 侯春华; 刘明; 文韬; 窦笑; 杨耀; 吴旭东
Original assignee: 中兴通讯股份有限公司
Priority date: 2022-06-30
Filing date: 2023-06-26
Publication date: 2024-01-04
Also published as: CN117391224A

Abstract

The embodiments of the present disclosure belong to the field of artificial intelligence. Provided are an energy-consumption optimization method, system and apparatus, and a storage medium. The method comprises: acquiring key index data of a data center; on the basis of a strategy fusion optimization model, determining candidate energy-consumption optimization schemes according to the key index data; on the basis of a simulation prediction model, assessing the candidate energy-consumption optimization schemes, so as to determine first assessment indexes of the candidate energy-consumption optimization schemes; on the basis of an inter-model assessment method, determining second assessment indexes of the candidate energy-consumption optimization schemes according to the first assessment indexes; on the basis of a re-assessment method, determining third assessment indexes of the candidate energy-consumption optimization schemes; according to the third assessment indexes, determining a target energy-consumption optimization scheme of the data center; and according to the target energy-consumption optimization scheme, adjusting an energy-consumption control parameter of the data center.

Description

Energy consumption optimization method, system, device and storage medium

Cross-references to related applications

This disclosure is based on the Chinese patent application CN202210760080.5 with the invention title "Energy Consumption Optimization Method, System, Device and Storage Medium" submitted on June 30, 2022, and claims the priority of this patent application, which is disclosed by reference The contents of are all incorporated into this disclosure.

Technical field

The present disclosure relates to the field of artificial intelligence technology, and in particular to a data center energy consumption optimization method, system, device and storage medium.

Background technique

Power Usage Effectiveness (PUE) is a parameter used to characterize the power utilization efficiency of a data center. Its value is the ratio of the total power consumed by all electrical equipment in the data center to the total power consumed by all electronic information equipment. In order to ensure the normal operation of electronic information equipment in the data center, such as various servers, storage devices, network equipment, etc., the temperature and humidity of the data center must be maintained within a certain range. The data center is usually equipped with HVAC equipment to adjust the ambient temperature and humidity. In order to comply with the energy consumption standards for building green data centers, in actual production it is necessary to reduce the energy consumption of HVAC equipment and thereby reduce the PUE of the data center. How to reduce the energy consumption of HVAC equipment while ensuring the safe and stable operation of the data center and determining a cooling solution that matches the actual needs of the data center has become an urgent problem that needs to be solved.

Contents of the invention

The embodiments of the present disclosure provide an energy consumption optimization method, device and storage medium, aiming to.

In a first aspect, embodiments of the present disclosure provide an energy consumption optimization method, including: obtaining key indicator data of a data center, where the key indicator data includes: data center operating status data, data center energy consumption data; based on a preset strategy Integrate the optimization model to determine at least one candidate energy consumption optimization plan based on the key indicator data; evaluate the candidate energy consumption optimization plan based on a preset simulation prediction model, and determine the corresponding energy consumption optimization plan for each candidate energy consumption optimization plan. The first evaluation index; based on the preset inter-model evaluation method, determine the second evaluation index corresponding to each of the candidate energy consumption optimization solutions according to the first evaluation index; determine the optimal energy consumption according to the second evaluation index Optimization plan: re-evaluate the optimal energy consumption optimization plan based on a preset re-evaluation method; determine the data center based on the re-evaluation results of the re-evaluation of the optimal energy consumption optimization plan. Target energy consumption optimization plan; adjust the energy consumption control parameters of the data center according to the target energy consumption optimization plan.

In a second aspect, embodiments of the present disclosure also provide an energy consumption optimization system. The energy consumption optimization system includes: an indicator data acquisition module for obtaining key indicator data of the data center. The key indicator data includes: data center operation Status data, data center energy consumption data; the strategy fusion optimization module is used to determine at least one candidate energy consumption optimization method based on the preset strategy fusion optimization model based on the key indicator data. plan; the simulation prediction and evaluation module is used to evaluate the candidate energy consumption optimization solutions based on the preset simulation prediction model, and determine the first evaluation index corresponding to each of the candidate energy consumption optimization solutions; the inter-model evaluation module is used to Based on the preset inter-model evaluation method, determine the second evaluation index corresponding to each of the candidate energy consumption optimization solutions according to the first evaluation index; a re-evaluation module is used to determine the optimal one according to the second evaluation index The energy consumption optimization plan is used to re-evaluate the optimal energy consumption optimization plan based on the preset re-evaluation method; the target solution determination module is used to re-evaluate the optimal energy consumption optimization plan according to the re-evaluation method. The evaluation results determine the target energy consumption optimization plan of the data center; the control parameter adjustment module is used to adjust the energy consumption control parameters of the data center according to the target energy consumption optimization plan.

In a third aspect, embodiments of the present disclosure also provide an energy consumption optimization device, which includes a processor, a memory, a computer program stored on the memory and executable by the processor, and a computer program for implementing A data bus is used for connection and communication between the processor and the memory. When the computer program is executed by the processor, the steps of any energy consumption optimization method provided by the embodiments of the present disclosure are implemented.

In a fourth aspect, embodiments of the present disclosure also provide a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be processed by one or more The processor is executed to implement the steps of any energy consumption optimization method provided by the embodiments of the present disclosure.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings needed to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present disclosure, which are of great significance to this field. Ordinary technicians can also obtain other drawings based on these drawings without exerting creative work.

Figure 1 is a schematic flowchart of steps of an energy consumption optimization method provided by an embodiment of the present disclosure;

Figure 2 is a schematic flowchart of sub-steps of an energy consumption optimization method provided by an embodiment of the present disclosure;

Figure 3 is a schematic flowchart of sub-steps of an energy consumption optimization method provided by an embodiment of the present disclosure;

Figure 4 is a schematic flowchart of sub-steps of an energy consumption optimization method provided by an embodiment of the present disclosure;

Figure 5 is a schematic diagram of a scenario for implementing the energy consumption optimization method provided by an embodiment of the present disclosure;

Figure 6 is a schematic block diagram of an energy consumption optimization system provided by an embodiment of the present disclosure;

FIG. 7 is a schematic structural block diagram of an energy consumption optimization device provided by an embodiment of the present disclosure.

Detailed ways

The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in this disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this disclosure.

The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all contents and operations/steps, nor are they necessarily performed in the order described. For example, some operations/steps can also be decomposed, combined or divided into Divide and merge, so the actual execution order may change based on actual conditions.

It should be understood that the terminology used in the description of the disclosure is for the purpose of describing particular embodiments only and is not intended to limit the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly dictates otherwise.

Embodiments of the present disclosure provide an energy consumption optimization method, device and storage medium. Among them, the energy consumption optimization method can be applied to mobile terminals or servers. For example, the algorithm library encapsulated by the energy consumption optimization method is deployed in a mobile terminal or server. The mobile terminal can be a mobile phone, a tablet computer, a notebook computer, Electronic devices such as desktop computers, personal digital assistants, and wearable devices; the server can be an independent server or a server cluster, or it can provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud Cloud servers for basic cloud computing services such as communications, middleware services, domain name services, security services, Content Delivery Network (CDN), and big data and artificial intelligence platforms.

Some embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. The following embodiments and features in the embodiments may be combined with each other without conflict.

Please refer to FIG. 1 , which is a schematic flow chart of an energy consumption optimization method provided by an embodiment of the present disclosure.

As shown in Figure 1, the energy consumption optimization method includes steps S101 to S107.

Step S101: Obtain key indicator data of the data center. The key indicator data includes: data center operating status data and data center energy consumption data.

For example, data centers usually include loads such as electronic information equipment, heating and ventilation equipment, and lighting equipment. Among them, the HVAC equipment is connected to the external environment. By adjusting and controlling control parameters related to natural wind, spray evaporation, chilled water, compression refrigeration, etc., it provides cooling for the electronic information equipment in the data center to ensure that the temperature and humidity of the data center are maintained. at preset levels.

For example, since the control parameters of the HVAC equipment have a certain adjustment space, the control parameters can be adjusted through a PID (Proportion Integral Differential) controller to adjust the energy consumption of the HVAC equipment, thereby improving the overall data center control energy consumption.

In an embodiment, as shown in Figure 2, step S101 includes: sub-steps S1011 to sub-step S1012.

Sub-step S1011: Obtain the indicator data of the data center through preset detection points.

Exemplarily, the indicator data is obtained by setting detection points, such as setting up sensors or software interfaces. The indicator data includes: data center operating status data used to reflect the operating status of the data center, such as the data center operating status data of each electronic information equipment in the data center. Hardware temperature, computing speed, etc.; data center energy consumption data used to reflect data center energy consumption. Of course, it is not limited to this. The indicator data may also include outdoor environmental data, such as the temperature and humidity of the external environment of the data center; and indoor environmental data, such as the temperature and humidity of the internal environment of the data center, which are not limited here.

Exemplarily, the indicator data is obtained so as to train a simulation prediction model through the indicator data with an influencing relationship in the indicator data to obtain a simulation prediction model capable of predicting relevant data.

In some implementations, step S1011 also includes performing data cleaning on the obtained indicator data.

For example, the indicator data obtained in step S1011 may include errors or outliers caused by environmental mutations. In order to ensure that the indicator data can accurately reflect the relationship between different indicators to improve the accuracy of the trained model , the data cleaning of the indicator data may include abnormality identification, and cleaning of abnormal values through statistical methods. Specifically, for example, abnormality discrimination can be performed by using the Laida criterion, the Grubbs criterion or the Dixon criterion. Of course, it is not limited to this and will not be described in detail here.

Exemplarily, the data cleaning of the indicator data may also include data conversion, such as unit conversion of the obtained indicator data, so that the cleaned indicator data can more significantly reflect the relationship between the data. .

For example, the obtained indicator data may include missing values represented by null values or placeholders, and the data cleaning of the indicator data may also include missing value processing. Specifically, missing value processing can be performed based on mean interpolation, and the missing values can be interpolated using the average value of the valid values. Of course, it is not limited to this and is not limited here.

Sub-step S1012: Analyze the indicator data based on a preset data analysis algorithm to determine the key indicator data and the independent variable data and dependent variable data in the key indicator data.

For example, the indicator data obtained in step S1011 includes a large amount of data related to the operating status of the data center and the energy consumption of HVAC equipment. In order to facilitate later model training, the indicator data needs to be analyzed in advance to determine the Key data in indicator data.

For example, the cleaned indicator data is analyzed, key data with causal relationships are determined, and independent variable data and dependent variable data in the key data are determined.

Illustratively, at least part of the index data has a mutually influencing relationship. For example, the cooling intensity of the data center HVAC equipment and the external ambient temperature of the data center will affect the internal ambient temperature of the data center and the electronic information equipment in the data center. temperature, then the cooling intensity of the HVAC equipment is determined as the independent variable data, and the temperature of the electronic information equipment is determined as the corresponding dependent variable data.

For example, the causal relationship between the cleaned indicator data can be determined based on the Bayesian causal network model. Of course, it is not limited to this. For example, it can also be determined based on the transfer entropy between the cleaned indicator data. The causal relationship will not be elaborated here.

Step S102: Based on the preset strategy fusion optimization model, determine at least one candidate energy consumption optimization plan according to the key indicator data.

For example, the strategy optimization model includes multiple sub-models, which are used to determine energy consumption optimization solutions of the data center under different circumstances. For example, the policy optimization model may include an evolutionary learning model, a statistical learning model, and a deep learning model. Of course, it is not limited to this, and there is no limit to this.

For example, the strategy optimization model is trained in advance with a certain amount of key indicator data.

In some embodiments, the energy consumption optimization method further includes: training the evolutionary learning model according to the independent variable data and the dependent variable data; according to the independent variable data and the dependent variable data, The statistical learning model is trained; the deep learning model is trained according to the independent variable data and the dependent variable data.

For example, the evolutionary learning model can be implemented by a genetic algorithm (GA), and the energy consumption control parameters of the data center are controlled when the amount of data processing is small, the data change trend is not obvious, and the external environment is unstable. Conduct strategy optimization under mutation conditions.

Illustratively, the statistical learning model may be implemented through a Bayesian optimization algorithm (Bayesian Optimization). The energy consumption control parameters of the data center are relatively sufficient when the amount of data processing is relatively sufficient, and the data changes are relatively stable or have a certain trend. Optimize strategies under certain circumstances.

Illustratively, the deep learning model may be implemented through a deterministic policy gradient (Deep Deterministic Policy Gradient, DDPG) or a deep Q-network (Deep Q-network, DQN) to control the energy consumption control parameters of the data center. Strategy optimization is performed when the amount of data processing is large and the environment changes are relatively stable.

Please refer to FIG. 3 , which is a schematic flowchart of sub-steps of an energy consumption optimization method provided by an embodiment of the present disclosure.

As shown in Figure 3, in some embodiments, step S102 includes step S1021-step S1023: Step S1021, based on the evolutionary learning model, determine the energy consumption of at least one of the candidate energy consumption optimization solutions according to the key indicator data obtained in real time. Consumption control parameters; Step S1022. Based on the statistical learning model, determine the energy consumption control parameters of at least one candidate energy consumption optimization scheme according to the key indicator data obtained in real time; Step S1023. Based on the deep learning model, determine the energy consumption control parameters of the candidate energy consumption optimization scheme according to the real-time obtained key indicator data. The key indicator data determines the energy consumption control parameters of at least one of the candidate energy consumption optimization solutions.

For example, by setting multiple sub-models of the strategy optimization model, the energy consumption optimization plan can be determined according to the actual situation of the data center, meeting the requirements for strategy optimization under diversified situations, and improving the flexibility of the energy consumption optimization method. sex and rationality.

Exemplarily, the strategy optimization model outputs at least one parameter corresponding to the current data based on the independent variable data input in real time, such as data center external environment data, based on the evolutionary learning model, the statistical learning model, and the deep learning model. The energy consumption control parameters of the center are compared with the energy consumption control parameters of candidate energy consumption optimization solutions that can reduce energy consumption.

Exemplarily, diversified candidate energy consumption optimization solutions are determined based on at least one candidate energy consumption optimization solution determined by the evolutionary learning model, the statistical learning model, and the deep learning model.

Step S103: Evaluate the candidate energy consumption optimization solutions based on a preset simulation prediction model, and determine the first evaluation index corresponding to each candidate energy consumption optimization solution.

Exemplarily, the key indicator data can also be used to train a simulation prediction model, so as to evaluate at least one candidate energy consumption optimization solution determined by the strategy optimization model through the simulation prediction model.

Illustratively, the simulation prediction model can be implemented through a deep neural network (Deep Neural Networks, DNN), and is of course not limited to this. For example, it can also be implemented through a long short-term memory network (Long Short-Term Memory, LSTM). No further details will be given here.

Exemplarily, the simulation prediction model includes a state prediction model and an energy consumption prediction model, wherein the state prediction model is used to predict data center operating status data, and the energy consumption prediction model is used to predict data center energy consumption data.

In some embodiments, the step energy consumption optimization method further includes: training a state prediction model for predicting data center operating state data based on the independent variable data and the dependent variable data; and, based on the The independent variable data and the dependent variable data are used to train an energy consumption prediction model for predicting data center energy consumption data.

Exemplarily, the state prediction model is used to predict the data center operating state data. For example, the status prediction model can determine the dependent variable data in the data center operating status data based on the independent variable data in the key data.

Exemplarily, the energy consumption prediction model is used to predict the energy consumption data of the data center. For example, the energy consumption prediction model can determine the dependent variable data in the data center energy consumption data based on the independent variable data in the key data.

Please refer to FIG. 4 , which is a schematic flowchart of sub-steps of an energy consumption optimization method provided by an embodiment of the present disclosure.

As shown in Figure 4, in some embodiments, step S103 includes step S1031-step S1032: Step S1031, based on the state prediction model, determine the operating status of the data center that executes each of the candidate energy consumption optimization solutions within a preset time period. Predict the data to determine the operating status evaluation indicators of each candidate energy consumption optimization plan; step S1032, based on the energy consumption prediction model, calculate the data center energy consumption data of each candidate energy consumption optimization plan within a preset time period. Predictions are made to determine the energy consumption evaluation indicators of each of the candidate energy consumption optimization solutions.

Exemplarily, the energy consumption control parameters of each candidate energy consumption optimization scheme are input into the state prediction model and the energy consumption prediction model, and the effects of the state prediction model and the energy consumption prediction model on each candidate energy consumption optimization scheme are obtained. Corresponding data center operating status data and prediction results of data center energy consumption data are used to determine the target energy consumption optimization plan.

Step S104: Based on the preset inter-model evaluation method and the first evaluation index, determine the second evaluation index corresponding to each of the candidate energy consumption optimization solutions.

For example, since the first evaluation index obtained in step S103 includes the operating status evaluation index determined by the state prediction model and the energy consumption evaluation index determined by the energy consumption prediction model, it is necessary to use the inter-model evaluation method to determine each of the The second evaluation index of the candidate energy consumption optimization plan is to comprehensively consider the operating status evaluation index and energy consumption evaluation index of each candidate energy consumption optimization plan to determine the optimal energy consumption optimization plan among each of the candidate energy consumption optimization plans. .

In some embodiments, step S104 includes: based on a preset inter-model evaluation method, determining an inter-model evaluation index for each of the candidate energy consumption optimization solutions according to the operating status evaluation index and the energy consumption evaluation index.

Exemplarily, the inter-model evaluation index may be, for example, Pareto efficiency (Pareto efficiency) determined based on the operating status evaluation index and energy consumption evaluation index of each candidate energy consumption optimization scheme. The calculation method of Pareto efficiency No further details will be given here.

Exemplarily, the inter-model evaluation method may also include a DM test (Diebold-Mariano Test) algorithm, which determines the confidence of the prediction results of the simulated prediction model, and filters the prediction results where the confidence is smaller than a predetermined value. Set the candidate energy consumption optimization scheme corresponding to the prediction result of reliability to ensure that the final target energy consumption optimization scheme is authentic and credible.

Step S105: Determine the optimal energy consumption optimization plan according to the second evaluation index, and re-evaluate the optimal energy consumption optimization plan based on a preset re-evaluation method.

Illustratively, the optimal energy consumption optimization solution is determined based on the second evaluation index obtained in step S104, for example, the optimal energy consumption optimization solution is determined based on the Pareto efficiency of each candidate energy consumption optimization solution.

For example, since the energy consumption control parameters of the data center have a great impact on the operation of the data center, before adjusting the energy consumption control parameters, the target energy consumption optimization plan needs to be repeatedly evaluated and tested multiple times to ensure that The feasibility of the target energy consumption optimization plan.

In some embodiments, step S105 includes: performing a constraint evaluation on the energy consumption control parameters of the optimal energy consumption optimization scheme according to the preset constraint rules for each of the energy consumption control parameters; according to the preset empirical rules, Conduct expert experience evaluation on the energy consumption control parameters of the optimal energy consumption optimization scheme; based on the simulation prediction model and the preset application benefit evaluation method, compare the current energy consumption control parameters with the optimal energy consumption optimization scheme. Data center energy consumption data of energy consumption control parameters are used to evaluate application benefits.

Illustratively, since the energy consumption control parameters have safety range constraints for normal operation, some energy consumption control parameters also have mutually exclusive or interdependent relationships. Before adopting the target energy consumption optimization plan, according to the energy consumption The safety range of the control parameters determines the safety and feasibility of the target energy consumption optimization solution.

Specifically, for example, in order to ensure the normal operation of the electronic information equipment in the data center, the spray evaporation amount of the HVAC equipment must be maintained greater than the preset amount. If the spray evaporation amount in the target energy consumption optimization plan is greater than the preset amount, Then the target energy consumption optimization plan is applied.

For example, in order to ensure the rationality of adjusting the energy consumption control parameters of the data center, empirical rules can be set in advance based on the actual situation for expert experience evaluation of the energy consumption optimization plan.

Specifically, for example, the optimal energy consumption optimization plan may have energy consumption control parameters with a large adjustment range. If the adjustments are directly made according to the optimal energy consumption optimization plan, the data center may be damaged due to sudden changes in the energy consumption control parameters. Electronic information equipment cannot operate stably.

Exemplarily, the adjustment threshold of each energy consumption control parameter in the empirical rule is obtained, if the change amplitude of the energy consumption control parameter in the optimal energy consumption optimization plan and the current energy consumption control parameter of the data center is greater than the adjustment threshold , it is possible to set the energy consumption control parameters in the optimal energy consumption optimization scheme to be adjusted step by step, that is, through multiple adjustments, the energy consumption control parameters of the data center are equal to the energy consumption control parameters of the optimal energy consumption optimization scheme. .

Exemplarily, based on the energy consumption prediction model and the preset application benefit evaluation method, the energy consumption data corresponding to the current energy consumption control parameters of the data center and the control parameters of the target energy consumption optimization plan are determined. The application benefit evaluation method For example, it can be: based on the energy consumption prediction model, determine the monthly, quarterly, and annual energy consumption data of the current energy consumption control parameters of the data center application and the control parameters of the target energy consumption optimization plan, respectively, and determine the current energy consumption control parameters based on the current energy consumption control parameters. and the monthly, quarterly, and annual energy consumption data differences between the control parameters of the target energy consumption optimization plan to determine the monthly, quarterly, and annual benefits of applying the target energy consumption optimization plan, and determine the application of the target energy consumption within the preset time. Energy consumption saved by consumption optimization solution amount, if the saved energy consumption is greater than the preset energy consumption, then the target energy consumption optimization plan is applied. Of course, it is not limited to this. The saved energy consumption can also be converted into saved costs according to the preset conversion rules. If the saved costs are greater than the preset costs, the target energy consumption optimization scheme is applied. This is not limited.

Step S106: Determine the target energy consumption optimization plan of the data center based on the re-evaluation result of the optimal energy consumption optimization plan.

For example, since the energy consumption control parameters of the data center need to ensure that the electronic information equipment in the data center can operate stably, the optimal energy consumption optimization plan among multiple candidate energy consumption optimization plans can be determined as the above after multi-level evaluation. Target energy consumption optimization plan.

Specifically, after determining the second evaluation index according to the first evaluation index, and determining the optimal energy consumption optimization scheme among the candidate energy consumption optimization schemes according to the second evaluation index, if the energy consumption in the optimal energy consumption optimization scheme is If the control parameters meet the third evaluation index, the optimal energy consumption optimization plan is determined as the target energy consumption optimization plan.

For example, the re-evaluation process may include modifying the optimal energy consumption optimization plan according to actual needs, and determining the modified energy consumption optimization plan as the target energy consumption optimization plan.

Step S107: Adjust the energy consumption control parameters of the data center according to the target energy consumption optimization plan.

Adjust the energy consumption control parameters of the data center according to the target energy consumption optimization plan.

Exemplarily, the target energy consumption optimization plan is sent to the data center, and the energy consumption control parameters of the data center are adjusted according to the target energy consumption optimization plan, such as adjusting the energy consumption of the data center HVAC equipment. Control parameters to reduce the overall energy consumption of the data center.

For example, after adjusting the energy consumption control parameters of the data center through multiple evaluations and tests, in order to prevent the target energy consumption optimization plan determined by the model from appearing unreasonable, the target energy consumption optimization plan, the constraint evaluation and the The results of the application benefit assessment are provided for users' reference so that they can confirm the target energy consumption optimization plan, which improves the safety of the energy consumption optimization method.

Please refer to Figure 5. Figure 5 is a schematic diagram of a scenario for implementing the energy consumption optimization method provided by an embodiment of the present disclosure. As shown in Figure 5, the energy consumption optimization device obtains the index data of the data center, for example, obtains the key index data of the data center. And determine the target energy consumption optimization plan based on the strategy optimization model, and deliver the target energy consumption optimization plan to the data center. Of course, it is not limited to this. For example, the energy consumption optimization device can be installed in a data center, which is not limited here.

The energy consumption optimization method provided by the above embodiments obtains key indicator data of the data center. The key indicator data includes: at least one of data center operating status data and data center energy consumption data; based on a preset strategy optimization model , determine at least one candidate energy consumption optimization plan based on the key indicator data; evaluate the candidate energy consumption optimization plan based on a preset simulation prediction model, and determine the target energy consumption optimization plan; according to the target energy consumption optimization plan , adjust the energy consumption control parameters of the data center. The technical solution of the embodiment of the present disclosure can determine the energy consumption optimization plan based on the acquired indicators of the data center, reducing the energy consumption of the data center while ensuring the safe and stable operation of the data center.

Please refer to FIG. 6 , which is a schematic structural block diagram of an energy consumption optimization system provided by an embodiment of the present disclosure.

As shown in Figure 6, the energy consumption optimization system provided by the embodiment of the present disclosure includes: an indicator data acquisition module 110, used to obtain key indicator data of the data center. The key indicator data includes: data center operating status data, data center energy consumption data; the strategy fusion optimization module 120 is used to determine at least one candidate energy consumption optimization solution based on the key indicator data based on the preset strategy fusion optimization model; the simulation prediction and evaluation module 130 is used to simulate based on the preset A prediction model is used to evaluate the candidate energy consumption optimization solutions and determine the first evaluation index corresponding to each candidate energy consumption optimization solution; the inter-model evaluation module 140 is used to evaluate the candidate energy consumption optimization solutions based on the preset inter-model evaluation method. The first evaluation index determines the second evaluation index corresponding to each of the candidate energy consumption optimization solutions; the re-evaluation module 150 is used to determine the optimal energy consumption optimization plan according to the second evaluation index, based on the preset re-evaluation method , re-evaluate the optimal energy consumption optimization plan; the target solution determination module 160 is configured to determine the target energy of the data center according to the re-evaluation result of the optimal energy consumption optimization plan. consumption optimization plan; the control parameter adjustment module 170 is used to adjust the energy consumption control parameters of the data center according to the target energy consumption optimization plan.

It can be understood that this embodiment is a system embodiment corresponding to the above method embodiment, and this embodiment can be implemented in cooperation with the above method embodiment. The relevant technical details and technical effects mentioned in the above embodiment are still valid in this embodiment, and will not be described again in order to reduce duplication. Correspondingly, the relevant technical details mentioned in this embodiment can also be applied to the above embodiments.

It should be noted that this system embodiment mainly describes the model acquisition method provided by the method embodiment at the software implementation level. Its implementation also needs to rely on hardware support. For example, the functions of related modules can be deployed on the processor. , so that the processor can implement corresponding functions. In particular, the relevant data generated by the operation can be stored in the memory for subsequent inspection and use.

It is worth mentioning that each module involved in this embodiment is a logical module. In practical applications, a logical unit can be a physical unit, or a part of a physical unit, or it can be multiple physical units. The combination of units is realized. In addition, in order to highlight the innovative part of the present disclosure, units that are not closely related to solving the technical problems raised by the present disclosure are not introduced in this embodiment, but this does not mean that other units do not exist in this embodiment.

Please refer to FIG. 7 , which is a schematic structural block diagram of an energy consumption optimization device provided by an embodiment of the present disclosure.

As shown in Figure 7, the energy consumption optimization device 300 includes a processor 301 and a memory 302. The processor 301 and the memory 302 are connected through a bus 303, which is, for example, an I2C (Inter-integrated Circuit) bus.

Specifically, the processor 301 is used to provide computing and control capabilities to support the operation of the entire energy consumption optimization device. The processor 301 can be a central processing unit (Central Processing Unit, CPU). The processor 301 can also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC). ), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gates or Transistor logic devices, discrete hardware components, etc. The general processor may be a microprocessor or the processor may be any conventional processor.

Specifically, the memory 302 may be a Flash chip, a read-only memory (ROM, Read-Only Memory) disk, an optical disk, a U disk or a mobile hard disk, etc.

Those skilled in the art can understand that the structure shown in Figure 7 is only a block diagram of a partial structure related to the embodiment of the present disclosure, and does not constitute a limitation on the energy consumption optimization device to which the embodiment of the present disclosure is applied. Specifically The energy consumption optimization device may include more or fewer components than shown in the figures, or combine certain components, or have a different arrangement of components.

Wherein, the processor is used to run a computer program stored in the memory, and implement any one of the energy consumption optimization methods provided by the embodiments of the present disclosure when executing the computer program.

In one embodiment, the processor is configured to run a computer program stored in a memory, and implement the following steps when executing the computer program:

Obtain key indicator data of the data center. The key indicator data includes: data center operating status data and data center energy consumption data;

Based on the preset strategy fusion optimization model, determine at least one candidate energy consumption optimization plan according to the key indicator data;

Evaluate the candidate energy consumption optimization solutions based on a preset simulation prediction model, and determine the first evaluation index corresponding to each candidate energy consumption optimization solution;

Based on the preset inter-model evaluation method, determine the second evaluation index corresponding to each of the candidate energy consumption optimization solutions according to the first evaluation index;

Determine the optimal energy consumption optimization plan according to the second evaluation index, and re-evaluate the optimal energy consumption optimization plan based on the preset re-evaluation method;

Determine the target energy consumption optimization plan of the data center based on the re-evaluation results of the optimal energy consumption optimization plan;

In one embodiment, when the processor implements the acquisition of key indicator data of the data center, it is used to implement:

Obtain the indicator data of the data center through preset detection points;

The indicator data is analyzed based on a preset data analysis algorithm to determine the key indicator data and the independent variable data and dependent variable data in the key indicator data.

In one embodiment, when implementing the energy consumption optimization method, the processor is used to implement:

Train the evolutionary learning model according to the independent variable data and the dependent variable data;

Train the statistical learning model according to the independent variable data and the dependent variable data;

The deep learning model is trained according to the independent variable data and the dependent variable data.

In one embodiment, when the processor implements the preset-based strategy optimization model and determines at least one candidate energy consumption optimization solution based on the key indicator data, it is used to implement:

Based on the evolutionary learning model, at least one of the candidates is determined according to the key indicator data obtained in real time. Select the energy consumption control parameters of the energy consumption optimization plan; and

Based on the statistical learning model, determine the energy consumption control parameters of at least one of the candidate energy consumption optimization solutions according to the key indicator data obtained in real time; and

Based on the deep learning model, the energy consumption control parameters of at least one of the candidate energy consumption optimization solutions are determined according to the key indicator data obtained in real time.

Based on the independent variable data and the dependent variable data, train a state prediction model for predicting data center operating state data; and,

Based on the independent variable data and the dependent variable data, an energy consumption prediction model for predicting data center energy consumption data is trained.

In one embodiment, when the processor implements the preset-based simulation prediction model, evaluates the candidate energy consumption optimization solutions, and determines the first evaluation index corresponding to each of the candidate energy consumption optimization solutions, Used to implement:

Based on the state prediction model, predict the data center operating state data for executing each of the candidate energy consumption optimization solutions within a preset time period, and determine the operating state evaluation indicators of each of the candidate energy consumption optimization solutions;

Based on the energy consumption prediction model, the energy consumption data of the data center executing each of the candidate energy consumption optimization plans within a preset time period is predicted, and the energy consumption evaluation index of each of the candidate energy consumption optimization plans is determined.

In one embodiment, when the processor implements the preset-based inter-model evaluation method and determines the second evaluation index corresponding to each of the candidate energy consumption optimization solutions according to the first evaluation index, accomplish:

Based on the preset inter-model evaluation method, the inter-model evaluation index of each candidate energy consumption optimization scheme is determined according to the operating status evaluation index and the energy consumption evaluation index.

In one embodiment, when implementing the preset-based re-evaluation method and re-evaluating the optimal energy consumption optimization plan, the processor is configured to implement:

Perform constraint evaluation on the energy consumption control parameters of the optimal energy consumption optimization scheme according to the preset constraint rules for each of the energy consumption control parameters;

According to the preset empirical rules, conduct expert experience evaluation on the energy consumption control parameters of the optimal energy consumption optimization plan;

Based on the simulation prediction model and the preset application benefit assessment method, an application benefit assessment is performed on the data center energy consumption data of the current energy consumption control parameters and the energy consumption control parameters of the optimal energy consumption optimization scheme.

It should be noted that those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the energy consumption optimization device described above can be referred to the corresponding process in the aforementioned energy consumption optimization method embodiment. This will not be described again.

Embodiments of the present disclosure also provide a storage medium for computer-readable storage. The storage medium stores one or more programs. The one or more programs can be executed by one or more processors to implement the following: The steps of any energy consumption optimization method provided by the embodiments of the present disclosure.

The storage medium may be an internal storage unit of the energy consumption optimization device described in the previous embodiment, such as a hard disk or memory of the energy consumption optimization device. The storage medium may also be an external storage device of the energy consumption optimization device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), or a secure digital (Secure Digital) equipped on the energy consumption optimization device. SD) card, Flash Card, etc.

Those of ordinary skill in the art can understand that all or some steps, systems, and functional modules/units in the devices disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may consist of several physical components. Components execute cooperatively. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit . Such software may be distributed on computer-readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). As is known to those of ordinary skill in the art, the term computer storage media includes volatile and nonvolatile media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. removable, removable and non-removable media. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Disk (DVD) or other optical disk storage, magnetic cassettes, tapes, disk storage or other magnetic storage devices, or may Any other medium used to store the desired information and that can be accessed by a computer. Additionally, it is known to those of ordinary skill in the art that communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media .

The above serial numbers of the embodiments of the present disclosure are only for description and do not represent the advantages and disadvantages of the embodiments. The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person familiar with the technical field can easily think of various equivalent methods within the technical scope disclosed in the present disclosure. Modifications or substitutions, these modifications or substitutions should be covered by the protection scope of this disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

An energy consumption optimization method, the method includes:

Obtain key indicator data of the data center. The key indicator data includes: data center operating status data and data center energy consumption data;

Based on the preset strategy fusion optimization model, determine at least one candidate energy consumption optimization plan according to the key indicator data;

Evaluate the candidate energy consumption optimization solutions based on a preset simulation prediction model, and determine the first evaluation index corresponding to each candidate energy consumption optimization solution;

Based on the preset inter-model evaluation method, determine the second evaluation index corresponding to each of the candidate energy consumption optimization solutions according to the first evaluation index;

Determine the optimal energy consumption optimization plan according to the second evaluation index, and re-evaluate the optimal energy consumption optimization plan based on the preset re-evaluation method;

Determine the target energy consumption optimization plan of the data center based on the re-evaluation results of the optimal energy consumption optimization plan;

Adjust the energy consumption control parameters of the data center according to the target energy consumption optimization plan.
The energy consumption optimization method according to claim 1, wherein said obtaining key indicator data of the data center includes:

Obtain the indicator data of the data center through preset detection points;

The indicator data is analyzed based on a preset data analysis algorithm to determine the key indicator data and the independent variable data and dependent variable data in the key indicator data.
The energy consumption optimization method according to claim 2, wherein the strategy fusion optimization model includes an evolutionary learning model, a statistical learning model, and a deep learning model, and the method further includes:

Train the evolutionary learning model according to the independent variable data and the dependent variable data;

Train the statistical learning model according to the independent variable data and the dependent variable data;

The deep learning model is trained according to the independent variable data and the dependent variable data.
The energy consumption optimization method according to claim 3, wherein the preset strategy-based fusion optimization model determines at least one candidate energy consumption optimization solution based on the key indicator data, including:

Based on the evolutionary learning model, determine the energy consumption control parameters of at least one of the candidate energy consumption optimization solutions according to the key indicator data obtained in real time; and

Based on the statistical learning model, determine the energy consumption control parameters of at least one of the candidate energy consumption optimization solutions according to the key indicator data obtained in real time; and

Based on the deep learning model, the energy consumption control parameters of at least one of the candidate energy consumption optimization solutions are determined according to the key indicator data obtained in real time.
The energy consumption optimization method according to claim 2, wherein the simulation prediction model includes a state prediction model and an energy consumption prediction model, and the method further includes:

Based on the independent variable data and the dependent variable data, train a state prediction model for predicting data center operating state data; and,

Based on the independent variable data and the dependent variable data, an energy consumption prediction model for predicting data center energy consumption data is trained.
The energy consumption optimization method according to claim 5, wherein the first evaluation index includes an operating status evaluation index and an energy consumption evaluation index; the candidate energy consumption optimization scheme is evaluated based on the preset simulation prediction model. Evaluate and determine the first evaluation indicators corresponding to each of the candidate energy consumption optimization solutions, including:

Based on the state prediction model, predict the data center operating state data for executing each of the candidate energy consumption optimization solutions within a preset time period, and determine the operating state evaluation indicators of each of the candidate energy consumption optimization solutions;

Based on the energy consumption prediction model, the energy consumption data of the data center executing each of the candidate energy consumption optimization plans within a preset time period is predicted, and the energy consumption evaluation index of each of the candidate energy consumption optimization plans is determined.
The energy consumption optimization method according to claim 6, wherein the preset-based inter-model evaluation method determines the second evaluation index corresponding to each of the candidate energy consumption optimization solutions according to the first evaluation index, including :

Based on the preset inter-model evaluation method, the inter-model evaluation index of each candidate energy consumption optimization scheme is determined according to the operating status evaluation index and the energy consumption evaluation index.
The energy consumption optimization method according to claim 7, wherein the re-evaluation method based on presets for re-evaluating the optimal energy consumption optimization plan includes:

Perform constraint evaluation on the energy consumption control parameters of the optimal energy consumption optimization scheme according to the preset constraint rules for each of the energy consumption control parameters;

According to the preset empirical rules, conduct expert experience evaluation on the energy consumption control parameters of the optimal energy consumption optimization scheme;

Based on the simulation prediction model and the preset application benefit assessment method, an application benefit assessment is performed on the data center energy consumption data of the current energy consumption control parameters and the energy consumption control parameters of the optimal energy consumption optimization scheme.
An energy consumption optimization system, the energy consumption optimization system includes:

The indicator data acquisition module is used to obtain key indicator data of the data center. The key indicator data includes: data center operating status data and data center energy consumption data;

A strategy fusion optimization module, used to determine at least one candidate energy consumption optimization plan based on the key indicator data based on the preset strategy fusion optimization model;

A simulation prediction and evaluation module, used to evaluate the candidate energy consumption optimization solutions based on a preset simulation prediction model, and determine the first evaluation index corresponding to each of the candidate energy consumption optimization solutions;

An inter-model evaluation module, configured to determine the second evaluation index corresponding to each of the candidate energy consumption optimization solutions based on the first evaluation index based on the preset inter-model evaluation method;

A re-evaluation module, configured to determine the optimal energy consumption optimization plan according to the second evaluation index, and re-evaluate the optimal energy consumption optimization plan based on a preset re-evaluation method;

A target solution determination module, configured to determine the target energy consumption optimization solution of the data center based on the re-evaluation results of the optimal energy consumption optimization solution;

A control parameter adjustment module is used to adjust the energy consumption control parameters of the data center according to the target energy consumption optimization plan.
An energy consumption optimization device, wherein the energy consumption optimization device includes a processor, a memory, a computer program stored on the memory and executable by the processor, and a computer program for implementing the processor and the memory A data bus is used for connection communication between the computer programs, and when the computer program is executed by the processor, the steps of the energy consumption optimization method according to any one of claims 1 to 8 are implemented.
A storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement claims 1 to 8 The steps of the energy consumption optimization method described in any one of the above.