Data center environment intelligent regulation and control system and method based on artificial intelligence technology

Abstract

The invention provides a data center environment intelligent regulation and control system and method based on an artificial intelligence technology, and the system comprises a perception fusion module, a self-adaptive data preprocessing module, a feature mining module, an intelligent prediction module, a decision module, a dynamic optimization module, and an energy management module. And the sensing fusion module is used for collecting environmental parameters and equipment information in the data center. According to the method, the future environmental parameters and the equipment state of the data center are predicted, and the optimal regulation and control strategy is made, so that dynamic and accurate prediction and decision making are realized, and the system can dynamically adjust energy distribution according to real-time requirements. Compared with a traditional system, the change trend of the data center can be predicted in advance, the optimal regulation and control strategy is formulated according to the change trend, and therefore energy waste is avoided.

Description

Technical Field

[0001] The present invention relates to an intelligent control system and method, specifically an intelligent control system and method for a data center environment based on artificial intelligence technology, and belongs to the field of intelligent control technology. Background Art

[0002] Data centers house a large number of servers and network equipment, generating significant heat and requiring effective cooling systems to maintain operating temperatures. The data center intelligent environmental control system integrates advanced computer networks, sensor technology, automated control, and data processing capabilities to enable comprehensive monitoring and intelligent management of the data center environment and equipment operating status. This system not only provides real-time monitoring of power supply and environmental control but also includes intelligent management and fault prevention for various equipment within the computer room.

[0003] Intelligent data center environmental control systems are widely used in key facilities such as data centers and communication base stations to ensure stable operation under optimal conditions. For large and medium-sized data centers, this system provides a centralized, visual management interface, allowing operations and maintenance personnel to monitor equipment operating status and environmental conditions at all times, enabling rapid response to emergencies and improving management efficiency.

[0004] Traditional data center environmental intelligent control systems often rely on pre-set rules and thresholds to perform monitoring and control tasks in their design and operation. While this static parameter-based operating mode can achieve a certain degree of environmental parameter monitoring and equipment control, its efficiency and accuracy are significantly limited. Because the rules and thresholds are pre-set, the system often appears to be unable to cope with the complex and changing conditions that arise during actual data center operations, making it difficult to make timely and accurate adjustments. More critically, traditional systems lack the ability to dynamically adjust energy allocation based on real-time demand. In actual data center operations, energy demand fluctuates constantly with changes in a variety of factors, including business load, equipment status, and external environmental conditions. However, traditional control systems are unable to effectively perceive these real-time changes and flexibly adjust energy allocation strategies accordingly. This results in the system still operating in a preset high-energy consumption mode when energy demand is low, resulting in unnecessary energy waste. Furthermore, during peak energy demand periods, the system is unable to allocate sufficient energy in a timely manner, affecting the normal operation of the data center. Therefore, the shortcomings of traditional data center environmental intelligent control systems in terms of efficiency and accuracy, as well as their inability to dynamically adjust energy distribution according to real-time demand, not only increase data center operating costs but also pose a serious challenge to their sustainable development. To this end, an intelligent control system and method for data center environment based on artificial intelligence technology is proposed. Summary of the Invention

[0005] In view of this, the present invention provides an intelligent control system and method for a data center environment based on artificial intelligence technology to solve or alleviate the technical problems existing in the prior art and at least provide a beneficial option.

[0006] The technical solution of the embodiment of the present invention is implemented as follows: an intelligent control system for data center environment based on artificial intelligence technology, including a perception fusion module, an adaptive data preprocessing module, a feature mining module, an intelligent prediction module, a decision module, a dynamic optimization module and an energy management module: The perception fusion module is used to collect environmental parameters and equipment information in the data center, including temperature, humidity, light, air quality, server load and power consumption information; The adaptive data preprocessing module is used to process the collected raw data and automatically adjust the processing strategy according to the data characteristics; The feature mining module is used to mine deep features from preprocessed data and capture complex relationships between data; The intelligent prediction module is used to predict the future environmental parameters and equipment status of the data center through a prediction algorithm; The prediction basis of the intelligent prediction module includes: Historical data: Historical data accumulated over the long term operation of the data center, including environmental parameters and equipment status; Real-time data: environmental parameters and equipment status data collected in real time by the data center; External factors: external weather changes, power supply conditions and changes in business needs; The decision-making module is used to analyze the future environmental parameters and equipment status output by the intelligent prediction module, formulate the optimal control strategy based on the prediction results and the actual needs of the data center, evaluate the formulated control strategy, predict the effect after its implementation, and adjust and optimize it according to the evaluation results; The dynamic optimization module is used to dynamically adjust the control strategy based on the real-time collected data and prediction results to achieve real-time optimization of energy distribution and equipment start-up and shutdown; The energy management module is used to optimize energy distribution by combining the energy demand of the data center with real-time electricity price information and utilizing intelligent management algorithms.

[0007] Further preferably, the prediction algorithm comprises the following steps: Data acquisition: Acquire environmental parameters and equipment status data mined by the feature mining module; Model construction: An autoregressive integrated moving average model is used to capture periodic changes and trends in environmental parameters and equipment status. A prediction submodule based on a gradient boosting tree is constructed to handle nonlinear and complex relationships. The outputs of the time series model and machine learning model are used as input to construct a convolutional neural network for fusion prediction. Joint prediction mechanism: Analyze the mutual influence between environmental parameters and equipment status, and construct an association matrix or graph structure to transmit and integrate information during the prediction process; Multi-step prediction: A multi-step prediction strategy is used to gradually predict future environmental parameters and equipment status. The prediction results of each step are used as input for the next step, forming a recursive prediction process. Dynamic weight adjustment: Dynamically adjust the weights of different prediction submodules based on prediction error and model performance.

[0008] Further preferably, when the dynamic optimization module dynamically adjusts the control strategy, the following steps are included: Strategy formulation and optimization: Establish a strategy library containing multiple control strategies, and select control strategies from the strategy library based on model prediction results and real-time data; Parameter optimization: Optimize the parameters of the selected control strategy; Dynamic adjustment: Adjust the parameters and execution methods of the control strategy in real time based on the dynamic changes of real-time data and model prediction results; Execute control: Send the optimized control strategy to the device control module to achieve real-time optimization.

[0009] Further preferably, the control strategies in the strategy library include environmental parameter control, equipment status control, energy management control and safety emergency control: Among them, environmental parameter control includes: adjusting air conditioning temperature, humidity control and light intensity; Equipment status control includes adjusting server load distribution, optimizing power supply, and performing equipment maintenance; Energy management and control include adjusting energy distribution based on real-time electricity price information and adopting energy-saving technologies to reduce energy consumption; Safety emergency control: including formulating emergency plans, conducting safety drills and setting alarm thresholds.

[0010] Further preferably, the feature mining module mines environmental parameter features, device status features, and data correlation features: The environmental parameter characteristics include: Temperature characteristics: Exploring temperature change trends in different areas and time periods within the data center, as well as correlation characteristics between temperature, humidity, and light environment parameters; Humidity characteristics: Analyze the changing patterns of humidity and the impact of humidity on equipment performance and energy consumption; Lighting characteristics: Extracting light intensity and distribution characteristics, as well as the impact of light on data center energy consumption and equipment heat dissipation; Air quality characteristics: explore the characteristics of pollutant concentration and oxygen content in the air; The device status characteristics include: Server load characteristics: Analyze the server's CPU usage, memory usage, disk I / O load characteristics, and load change trends over time; Power consumption characteristics: Explore the power consumption patterns of devices and the correlation between power consumption, environmental parameters, and device load; Equipment failure characteristics: By monitoring the equipment's operation logs and error code information, we can extract precursor characteristics of equipment failures, including abnormal vibration, temperature rise, and current fluctuations. Equipment life characteristics: Based on the historical operation data of the equipment, predict the remaining life of the equipment and explore the characteristics of key factors affecting the equipment life.

[0011] Further preferably, the adaptive data processing module performs dynamic cleaning, denoising, normalization and adaptive strategy adjustment on the collected raw data.

[0012] Further preferably, it also includes a device control module, which is used to connect various devices in the data center through Internet of Things technology, realize remote monitoring and intelligent control of the equipment, and automatically adjust the equipment operation status according to the control strategy.

[0013] More preferably, it further includes a user interaction module, which is used to provide a visual interface to display the environmental parameters, equipment status and energy consumption information of the data center.

[0014] Further preferably, it also includes an early warning module, which is used to generate intelligent alarm information when an abnormal situation is found or a preset threshold is reached, and automatically trigger an emergency response process.

[0015] An artificial intelligence-based intelligent data center environment control method includes the following steps: Step 1: Obtain environmental parameters and equipment information in the data center, and dynamically clean, denoise, normalize, and adjust adaptive strategies for the raw data; Step 2: Mining deep features from the preprocessed data, including environmental parameter features, equipment status features, and data correlation features; Step 3: Predict the future environmental parameters and equipment status of the data center; Step 4: Analyze future environmental parameters and equipment status, formulate the optimal control strategy based on the predicted results and the actual needs of the data center, evaluate the formulated control strategy, predict its effect after implementation, and adjust and optimize it based on the evaluation results; Step 5: Dynamically adjust the control strategy based on real-time collected data and prediction results to achieve real-time optimization of energy distribution and equipment start-up and shutdown; Step 6: Optimize energy distribution using intelligent management algorithms, combining data center energy demand and real-time electricity price information. Step 7: Connect various devices in the data center through IoT technology to achieve remote monitoring and intelligent control of the equipment, and automatically adjust the equipment operating status according to the control strategy.

[0016] The embodiment of the present invention adopts the above technical solution, which has the following advantages: 1. The present invention mines deep features from preprocessed data and captures the complex relationships between data, thereby extracting more representative features and improving the accuracy and generalization ability of model training. Through deep feature mining, the system can more deeply understand the intrinsic connection between data center environmental parameters and equipment status, thereby formulating more accurate control strategies. By predicting the future environmental parameters and equipment status of the data center and formulating the optimal control strategy, dynamic and accurate prediction and decision-making are achieved, enabling the system to dynamically adjust energy allocation according to real-time demand. Compared with traditional systems, the present invention can predict the changing trends of the data center in advance and formulate the optimal control strategy accordingly, thereby avoiding energy waste.

[0017] 2. The present invention realizes real-time optimization of parameters such as energy distribution and equipment start and stop by dynamically adjusting the control strategy, ensuring that the data center operates in the best state, improving the response speed and control accuracy. Through real-time dynamic optimization, the system can quickly adapt to changes in the data center environment and adjust the control strategy in time, thereby further reducing energy waste. Through energy management, the system can reduce energy costs and improve energy utilization efficiency while meeting the energy needs of the data center.

[0018] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present invention will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0020] Figure 1 This is a structural diagram of an intelligent control system for data center environment based on artificial intelligence technology of the present invention; Figure 2 A flowchart of the steps of the prediction algorithm of the present invention; Figure 3 A flow chart of the steps for dynamically adjusting the control strategy of the present invention; Figure 4 This is a flowchart of the steps of an intelligent control method for data center environment based on artificial intelligence technology of the present invention. DETAILED DESCRIPTION

[0021] Hereinafter, only certain exemplary embodiments are briefly described. As will be appreciated by those skilled in the art, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and description are to be considered as illustrative in nature and not restrictive.

[0022] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0023] like Figures 1-4 As shown, an embodiment of the present invention provides an intelligent control system for a data center environment based on artificial intelligence technology, including a perception fusion module, an adaptive data preprocessing module, a feature mining module, an intelligent prediction module, a decision module, a dynamic optimization module, and an energy management module: The perception fusion module is used to collect environmental parameters and equipment information in the data center, including temperature, humidity, light, air quality, server load, and power consumption information; The feature mining module is used to mine deep features from preprocessed data and capture complex relationships between data; The intelligent prediction module is used to predict the future environmental parameters and equipment status of the data center through a prediction algorithm. The prediction algorithm includes the following steps: Data acquisition: Acquire environmental parameters and equipment status data mined by the feature mining module; Model construction: An autoregressive integrated moving average model is used to capture periodic changes and trends in environmental parameters and equipment status. A prediction submodule based on a gradient boosting tree is constructed to handle nonlinear and complex relationships. The outputs of the time series model and machine learning model are used as input to construct a convolutional neural network for fusion prediction. Joint prediction mechanism: Analyze the mutual influence between environmental parameters and equipment status, and construct an association matrix or graph structure to transmit and integrate information during the prediction process; Multi-step prediction: A multi-step prediction strategy is used to gradually predict future environmental parameters and equipment status. The prediction results of each step are used as input for the next step, forming a recursive prediction process. Dynamic weight adjustment: Dynamically adjust the weights of different prediction submodules based on prediction error and model performance; By accurately predicting future environmental parameters and equipment status, the prediction algorithm allows data centers to formulate control strategies in advance, optimize energy distribution, improve equipment utilization, reduce operating costs, and ensure the safe and stable operation of the data center.

[0024] The prediction basis of the intelligent prediction module includes: Historical data: Historical data accumulated over the long term operation of the data center, including environmental parameters and equipment status; Real-time data: environmental parameters and equipment status data collected in real time by the data center; External factors: external weather changes, power supply conditions and changes in business needs; The decision-making module is used to analyze the future environmental parameters and equipment status output by the intelligent prediction module, formulate the optimal control strategy based on the prediction results and the actual needs of the data center, evaluate the formulated control strategy, predict the effect after its implementation, and adjust and optimize it according to the evaluation results; The main functions of the decision-making module include: Analyze prediction results: Analyze the future environmental parameters and equipment status output by the intelligent prediction module to understand the meaning and potential impact of the prediction results; Formulate control strategies: Based on the forecast results and the actual needs of the data center, formulate the optimal control strategy. The control strategy aims to ensure that the data center operates in the best state, improve operational efficiency and energy utilization, and reduce operating costs. Evaluate strategy effectiveness: Evaluate the formulated regulatory strategy, predict its effect after implementation, and adjust and optimize it based on the evaluation results.

[0025] Through the effective operation of the intelligent prediction module and decision-making module, the data center can achieve more efficient and accurate intelligent environmental control, improve operational efficiency and reliability, reduce operating costs, and provide strong support for the sustainable development of the data center.

[0026] The dynamic optimization module is used to dynamically adjust the control strategy based on real-time collected data and prediction results to achieve real-time optimization of energy distribution and equipment start-up and shutdown; The energy management module is used to combine the energy demand of the data center with real-time electricity price information and optimize energy distribution using intelligent management algorithms; The intelligent management algorithm includes the following steps: Data collection and preprocessing: Collect energy demand data (such as server load and cooling system energy consumption) and real-time electricity price information from the data center, and perform preprocessing operations such as cleaning, denoising, and normalization on the collected data to ensure data quality; Energy demand forecasting: Build an energy demand forecasting model using a long-short-term memory artificial neural network. Use historical energy demand data to train the forecasting model and optimize model parameters. Real-time energy demand data is then input into the forecasting model to predict energy demand over a period of time. Electricity price analysis: Obtain real-time electricity price information through API interfaces or data crawling, perform time series analysis on real-time electricity prices, and identify peak and low electricity price periods; Optimize energy allocation strategy: With the goal of minimizing energy costs, we establish an energy allocation optimization model. This model considers energy demand forecasts, real-time electricity price information, and the data center's energy supply capacity. A linear programming algorithm is used to solve the optimization model and determine the optimal energy allocation strategy. Based on real-time electricity price fluctuations and energy demand changes, the energy allocation strategy is dynamically adjusted to ensure efficient energy utilization and cost savings. Execution and monitoring: The optimized energy allocation strategy is sent to the data center's energy management system to execute energy allocation operations. The data center's energy usage and electricity price changes are monitored in real time through the sensor network to ensure the effective execution of the energy allocation strategy. Based on the real-time monitoring results, the energy allocation strategy is fed back and optimized to improve the adaptability and accuracy of the algorithm.

[0027] Intelligent management algorithms can be widely used in various types of data centers, especially those that are sensitive to energy costs and have large fluctuations in energy demand. By implementing this algorithm, data centers can achieve efficient energy utilization and cost savings, and improve operational efficiency and economic benefits.

[0028] In one embodiment, when the dynamic optimization module dynamically adjusts the control strategy, the following steps are included: Strategy formulation and optimization: Establish a strategy library containing multiple control strategies, and select control strategies from the strategy library based on model prediction results and real-time data; Parameter optimization: Optimize the parameters of the selected control strategy; Dynamic adjustment: Adjust the parameters and execution methods of the control strategy in real time based on the dynamic changes of real-time data and model prediction results; Execution control: Send the optimized control strategy to the equipment control module to achieve real-time optimization; For example, in the temperature control of a data center, the specific implementation process of the real-time dynamic optimization module is as follows: Data collection: Temperature and humidity sensors collect temperature data from the data center in real time; Data processing and analysis: The central monitoring host receives temperature data, performs pre-processing and model prediction, and derives the temperature change trend over the next period of time; Strategy formulation and optimization: Based on the prediction results and real-time temperature data, the most appropriate temperature control strategy (such as adjusting the air conditioning temperature, turning on or off the humidifier, etc.) is selected from the strategy library and the strategy parameters are optimized; Dynamic adjustment and execution: Real-time adjustment of parameters such as air conditioning temperature and humidifier operating status to ensure that the data center temperature remains within the optimal range; Feedback and correction: Use temperature and humidity sensors to monitor the temperature control effect in real time and correct and optimize the control strategy; Visualization and alarm: Provides a visual interface for temperature control and triggers an alarm mechanism when the temperature exceeds the preset threshold.

[0029] In one embodiment, the control strategies in the policy library include environmental parameter control, device status control, energy management control, and safety emergency control: Among them, environmental parameter control includes: adjusting air conditioning temperature, humidity control and light intensity; Equipment status control includes adjusting server load distribution, optimizing power supply, and performing equipment maintenance; Energy management and control include adjusting energy distribution based on real-time electricity price information and adopting energy-saving technologies to reduce energy consumption; Safety emergency control: including formulating emergency plans, conducting safety drills and setting alarm thresholds.

[0030] In one embodiment, the feature mining module mines environmental parameter features, device status features, and data correlation features; Environmental parameter characteristics include: Temperature characteristics: Exploring temperature change trends in different areas and time periods within the data center, as well as correlation characteristics between temperature, humidity, and light environment parameters; Humidity characteristics: Analyze the changing patterns of humidity and the impact of humidity on equipment performance and energy consumption; Lighting characteristics: Extracting light intensity and distribution characteristics, as well as the impact of light on data center energy consumption and equipment heat dissipation; Air quality characteristics: explore the characteristics of pollutant concentration and oxygen content in the air; Device status characteristics include: Server load characteristics: Analyze the server's CPU usage, memory usage, disk I / O load characteristics, and load change trends over time; Power consumption characteristics: Explore the power consumption patterns of devices and the correlation between power consumption, environmental parameters, and device load; Equipment failure characteristics: By monitoring the equipment's operation logs and error code information, we can extract precursor characteristics of equipment failures, including abnormal vibration, temperature rise, and current fluctuations. Equipment life characteristics: Based on the equipment's historical operating data, predict the equipment's remaining life and explore the key factors affecting the equipment's life; The characteristics of data association include: Correlation characteristics between environmental parameters and device status: Analyze the impact of environmental parameter changes on device status, such as increased temperature leading to decreased server performance and increased humidity leading to increased device failure rates; Inter-device correlation features: Mining the correlation features of the operating status of different devices, such as excessive load on one server causing increased load on other servers, or network device failure affecting communication across the entire data center; Spatiotemporal correlation features: Considering the temporal and spatial distribution characteristics of data, we can mine data correlation features between different regions and time periods, such as temperature differences between different floors in a data center and changes in power consumption at different times of the day. It also includes deep abstract features, which include: Nonlinear features: Leverage the nonlinear mapping capabilities of deep learning models to mine nonlinear relationship features in data, such as the complex nonlinear mapping relationship between environmental parameters and device status; High-dimensional features: Deep learning models automatically learn high-dimensional representations of data, extract more representative high-dimensional features, and improve the accuracy and generalization ability of model training; Sequence features: For time series data, mine the sequence pattern features in the data, such as the changing trend of environmental parameters over time and the periodic changes in equipment load.

[0031] In one embodiment, the adaptive data pre-processing module is used to process the collected raw data and automatically adjust the processing strategy according to the data characteristics; The adaptive data processing module dynamically cleans, denoises, normalizes and adjusts the adaptive strategy for the collected raw data, including: Dynamic cleaning: Noise removal: Due to sensor errors, signal interference, and other reasons, the collected data contains noise. Adaptive algorithms can identify and remove this noise, retaining the valid part of the data. Handling missing values: During the data collection process, some data points may be missing. The adaptive data preprocessing module will use interpolation and mean filling methods to handle missing values according to the data characteristics to ensure data integrity; Outlier detection and processing: Using statistical methods or machine learning models, we can identify and process outliers in the data to prevent them from interfering with subsequent analysis. Denoising: Filtering technology: Apply low-pass filtering algorithm to remove high-frequency noise in the data and retain low-frequency valid signals; Smoothing: Use moving average, exponential smoothing and other methods to smooth the data to reduce data fluctuations and improve data stability.

[0032] Normalization processing: Data scaling: scaling the data to the same scale, such as normalizing data in different ranges to the interval [0, 1] or [-1, 1], to facilitate subsequent model training; Standardization: By calculating the mean and standard deviation of the data, the data is converted into a standard normal distribution, eliminating the dimensionality effect and improving the convergence speed and accuracy of the model; Adaptive strategy adjustment: Adjust processing strategies based on data characteristics: The adaptive data preprocessing module can automatically adjust processing strategies based on data characteristics (such as distribution, trend, periodicity, etc.) to ensure optimal preprocessing results; Real-time feedback and adjustment: During the preprocessing process, the module will monitor the processing effect in real time and dynamically adjust the processing parameters and strategies based on the feedback results to adapt to data changes.

[0033] In one embodiment, it also includes a device control module, which is used to connect various devices in the data center through the Internet of Things technology to achieve remote monitoring and intelligent control of the equipment, and automatically adjust the equipment operation status according to the control strategy.

[0034] In one embodiment, a user interaction module is further included, which is used to provide a visual interface to display the environmental parameters, equipment status and energy consumption information of the data center.

[0035] In one embodiment, an early warning module is also included, which is used to generate intelligent alarm information when an abnormal situation is found or a preset threshold is reached, and automatically trigger an emergency response process.

[0036] In one embodiment, a system security module is further included to control access, ensure system security and data confidentiality, and prevent unauthorized access and operation; This includes data encryption, access control, audit monitoring, and data minimization principles, including: Data encryption technologies include: Encrypted storage: AES encryption algorithm is used to encrypt sensitive data stored in the data center, ensuring that even if the data is illegally obtained, it cannot be decrypted and read; Encrypted transmission: During data transmission, encryption protocols such as SSL / TLS are used to encrypt data to prevent data from being stolen or tampered with during transmission; Access controls include: Role-based access control: Grant different access rights based on the user's role and responsibilities. For example, ordinary users can only view some data, while administrators have higher permissions. Fine-grained permission management: Fine-grained management of data access rights ensures that only authorized users can access specific data; Multi-factor authentication: Combining multiple authentication methods such as passwords, biometrics, and tokens to improve identity authentication security and prevent unauthorized users from accessing the system; Data minimization principles include: Collect only necessary data: During the data collection process, follow the principle of data minimization and only collect the minimum amount of data required to achieve a specific purpose, thereby reducing the risk of data leakage; Data desensitization: For data that needs to be shared or made public, data desensitization techniques are used to process it, such as replacing, deleting, or generalizing sensitive information, to ensure that the data can still be used for analysis and research without leaking privacy; Security audit monitoring includes: Real-time auditing: Real-time monitoring and auditing of system access and operations, recording user access behavior and operation log information; Anomaly detection: Through security audit tools and analysis algorithms, abnormal behaviors such as illegal access and data leakage can be discovered and identified in a timely manner, and corresponding measures can be taken to deal with them.

[0037] An artificial intelligence-based intelligent data center environment control method includes the following steps: Step 1: Obtain environmental parameters and equipment information in the data center, and dynamically clean, denoise, normalize, and adjust adaptive strategies for the raw data; Step 2: Mining deep features from the preprocessed data, including environmental parameter features, equipment status features, and data correlation features; Step 3: Predict the future environmental parameters and equipment status of the data center; Step 4: Analyze future environmental parameters and equipment status, formulate the optimal control strategy based on the predicted results and the actual needs of the data center, evaluate the formulated control strategy, predict its effect after implementation, and adjust and optimize it based on the evaluation results; Step 5: Dynamically adjust the control strategy based on real-time collected data and prediction results to achieve real-time optimization of energy distribution and equipment start-up and shutdown; Step 6: Optimize energy distribution using intelligent management algorithms, combining data center energy demand and real-time electricity price information. Step 7: Connect various devices in the data center through IoT technology to achieve remote monitoring and intelligent control of the equipment, and automatically adjust the equipment operating status according to the control strategy.

[0038] When the present invention is working: the environmental parameters and equipment information in the data center are collected through the perception fusion module and sent to the adaptive data preprocessing module. The adaptive data preprocessing module processes the collected raw data and automatically adjusts the processing strategy according to the data characteristics. The feature mining module mines deep-level features from the preprocessed data and captures the complex relationship between the data. The intelligent prediction module predicts the future environmental parameters and equipment status of the data center through a prediction algorithm. The decision module analyzes the future environmental parameters and equipment status output by the intelligent prediction module, formulates the optimal control strategy based on the prediction results and the actual needs of the data center, and evaluates the formulated control strategy to predict the effect after its implementation, and adjusts and optimizes it according to the evaluation results. The dynamic optimization module dynamically adjusts the control strategy based on the real-time collected data and prediction results to achieve real-time optimization of energy distribution and equipment start and stop. The energy management module combines the energy demand and real-time electricity price information of the data center and uses the intelligent management algorithm to optimize energy distribution. The equipment control module connects various devices in the data center through the Internet of Things technology to achieve remote monitoring and intelligent control of the equipment, and automatically adjusts the equipment operation status according to the control strategy.

[0039] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various modifications and substitutions within the technical scope disclosed in the present invention, and such modifications and substitutions are intended to be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. An artificial intelligence-based data center environment intelligent control system, comprising a perception fusion module, an adaptive data preprocessing module, a feature mining module, an intelligent prediction module, a decision-making module, a dynamic optimization module, and an energy management module, characterized by: The perception fusion module is used to collect environmental parameters and equipment information in the data center, including temperature, humidity, light, air quality, server load and power consumption information; The adaptive data preprocessing module is used to process the collected raw data and automatically adjust the processing strategy according to the data characteristics; The feature mining module is used to mine deep features from preprocessed data and capture complex relationships between data; The intelligent prediction module is used to predict the future environmental parameters and equipment status of the data center through a prediction algorithm; The prediction basis of the intelligent prediction module includes: Historical data: Historical data accumulated over the long term operation of the data center, including environmental parameters and equipment status; Real-time data: environmental parameters and equipment status data collected in real time by the data center; External factors: external weather changes, power supply conditions and changes in business needs; The decision-making module is used to analyze the future environmental parameters and equipment status output by the intelligent prediction module, formulate the optimal control strategy based on the prediction results and the actual needs of the data center, evaluate the formulated control strategy, predict the effect after its implementation, and adjust and optimize it according to the evaluation results; The dynamic optimization module is used to dynamically adjust the control strategy based on the real-time collected data and prediction results to achieve real-time optimization of energy distribution and equipment start-up and shutdown; The energy management module is used to optimize energy distribution by combining the energy demand of the data center with real-time electricity price information and utilizing intelligent management algorithms.

2. The data center environment intelligent control system based on artificial intelligence technology according to claim 1 is characterized by: The prediction algorithm includes the following steps: Data acquisition: Acquire environmental parameters and equipment status data mined by the feature mining module; Model construction: An autoregressive integrated moving average model is used to capture periodic changes and trends in environmental parameters and equipment status. A prediction submodule based on a gradient boosting tree is constructed to handle nonlinear and complex relationships. The outputs of the time series model and machine learning model are used as input to construct a convolutional neural network for fusion prediction. Joint prediction mechanism: Analyze the mutual influence between environmental parameters and equipment status, and construct an association matrix or graph structure to transmit and integrate information during the prediction process; Multi-step prediction: A multi-step prediction strategy is used to gradually predict future environmental parameters and equipment status. The prediction results of each step are used as input for the next step, forming a recursive prediction process. Dynamic weight adjustment: Dynamically adjust the weights of different prediction submodules based on prediction error and model performance.

3. The data center environment intelligent control system based on artificial intelligence technology according to claim 1 is characterized by: When the dynamic optimization module dynamically adjusts the control strategy, the following steps are included: Strategy formulation and optimization: Establish a strategy library containing multiple control strategies, and select control strategies from the strategy library based on model prediction results and real-time data; Parameter optimization: Optimize the parameters of the selected control strategy; Dynamic adjustment: Adjust the parameters and execution methods of the control strategy in real time based on the dynamic changes of real-time data and model prediction results; Execute control: Send the optimized control strategy to the device control module to achieve real-time optimization.

4. The data center environment intelligent control system based on artificial intelligence technology according to claim 3 is characterized by: The control strategies in the strategy library include environmental parameter control, equipment status control, energy management control and safety emergency control: Among them, environmental parameter control includes: adjusting air conditioning temperature, humidity control and light intensity; Equipment status control includes adjusting server load distribution, optimizing power supply, and performing equipment maintenance; Energy management and control include adjusting energy distribution based on real-time electricity price information and adopting energy-saving technologies to reduce energy consumption; Safety emergency control: including formulating emergency plans, conducting safety drills and setting alarm thresholds.

5. The data center environment intelligent control system based on artificial intelligence technology according to claim 1 is characterized by: The feature mining module mines environmental parameter features, device status features, and data association features: The environmental parameter characteristics include: Temperature characteristics: Exploring temperature change trends in different areas and time periods within the data center, as well as correlation characteristics between temperature, humidity, and light environment parameters; Humidity characteristics: Analyze the changing patterns of humidity and the impact of humidity on equipment performance and energy consumption; Lighting characteristics: Extracting light intensity and distribution characteristics, as well as the impact of light on data center energy consumption and equipment heat dissipation; Air quality characteristics: explore the characteristics of pollutant concentration and oxygen content in the air; The device status characteristics include: Server load characteristics: Analyze the server's CPU usage, memory usage, disk I / O load characteristics, and load change trends over time; Power consumption characteristics: Explore the power consumption patterns of devices and the correlation between power consumption, environmental parameters, and device load; Equipment failure characteristics: By monitoring the equipment's operation logs and error code information, we can extract precursor characteristics of equipment failures, including abnormal vibration, temperature rise, and current fluctuations. Equipment life characteristics: Based on the historical operation data of the equipment, predict the remaining life of the equipment and explore the characteristics of key factors affecting the equipment life.

6. The data center environment intelligent control system based on artificial intelligence technology according to claim 1 is characterized by: The adaptive data processing module performs dynamic cleaning, denoising, normalization and adaptive strategy adjustment on the collected raw data.

7. The data center environment intelligent control system based on artificial intelligence technology according to claim 1 is characterized by: It also includes a device control module, which is used to connect various devices in the data center through Internet of Things technology, realize remote monitoring and intelligent control of the equipment, and automatically adjust the equipment operation status according to the control strategy.

8. The data center environment intelligent control system based on artificial intelligence technology according to claim 1 is characterized by: It also includes a user interaction module, which is used to provide a visual interface to display the environmental parameters, equipment status and energy consumption information of the data center.

9. The data center environment intelligent control system based on artificial intelligence technology according to claim 1 is characterized by: It also includes an early warning module, which is used to generate intelligent alarm information when an abnormal situation is found or a preset threshold is reached, and automatically trigger an emergency response process.

10. A data center environment intelligent control method based on artificial intelligence technology, applied to a data center environment intelligent control system based on artificial intelligence technology according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: Obtain environmental parameters and equipment information in the data center, and dynamically clean, denoise, normalize, and adjust adaptive strategies for the raw data; Step 2: Mining deep features from the preprocessed data, including environmental parameter features, equipment status features, and data correlation features; Step 3: Predict the future environmental parameters and equipment status of the data center; Step 4: Analyze future environmental parameters and equipment status, formulate the optimal control strategy based on the predicted results and the actual needs of the data center, evaluate the formulated control strategy, predict its effect after implementation, and adjust and optimize it based on the evaluation results; Step 5: Dynamically adjust the control strategy based on real-time collected data and prediction results to achieve real-time optimization of energy distribution and equipment start-up and shutdown; Step 6: Combine the data center's energy demand and real-time electricity price information to optimize energy distribution using intelligent management algorithms; Step 7: Connect various devices in the data center through IoT technology to achieve remote monitoring and intelligent control of the equipment, and automatically adjust the equipment operating status according to the control strategy.

Images