A method and system for optimizing energy efficiency in data centers with wind and liquid homogeneity

Abstract

This invention relates to the field of data processing technology, and particularly to a method and system for optimizing energy efficiency in data centers using a combined air-liquid and air-source approach. The method includes: acquiring acoustic spectrum signals from the cooling system during operation via sensors to obtain a raw acoustic spectrum dataset; performing frequency domain transformation on the raw acoustic spectrum dataset to obtain a frequency band distribution feature set; extracting acoustic features from the frequency band distribution feature set and determining the system energy consumption status based on a pre-established mapping model; determining the cooling system mode switching requirement; predicting the acoustic feature change trend based on the mode switching requirement to determine the mode switching time point; adjusting the cooling system operating parameters according to the mode switching time point, executing the mode switch, and obtaining the updated system operating status; and updating the mapping model based on the updated system operating status to obtain an optimized energy consumption status prediction model. This invention solves the problem of low accuracy in data center energy efficiency optimization and improves the accuracy of energy efficiency optimization.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for optimizing energy efficiency in a data center with wind and liquid homogeneity. Background Technology

[0002] With the rapid development of information technology, the scale of facilities such as data centers and server clusters has gradually expanded, generating a large number of computing tasks, which places higher demands on cooling systems. Cooling systems play a crucial role in data centers, not only ensuring the normal operation of equipment but also directly affecting the system's energy efficiency and energy-saving performance. However, with fluctuations in equipment load and changes in environmental conditions, traditional cooling methods cannot perceive the system's operating status in real time, making precise energy efficiency optimization difficult and resulting in significant energy waste.

[0003] Traditional cooling methods primarily rely on preset cooling modes, such as air cooling and liquid cooling. However, these methods typically cannot adjust in a timely manner based on real-time load and system temperature. While air cooling and liquid cooling each have their advantages, in complex environments like data centers with drastically changing loads, relying on a single cooling mode often fails to achieve optimal energy efficiency. Furthermore, most existing cooling systems fail to effectively utilize signals generated during system operation, such as acoustic signals, for real-time monitoring and adjustment. These systems often cannot accurately predict and determine the optimal time for mode switching, leading to unstable system energy efficiency and energy waste.

[0004] Meanwhile, acoustic signals, as a byproduct of system operation, contain a wealth of useful operational information. By analyzing the acoustic signals generated by changes in fan speed and fluid flow, rich characteristics of the system's operating status can be obtained. However, extracting effective features from this complex acoustic data and establishing a precise mapping relationship between them and energy consumption status still faces significant technical challenges. Current research and application in this field are still in their early stages and cannot effectively solve the problems of intelligent regulation and energy efficiency optimization of cooling systems.

[0005] Therefore, how to accurately monitor the cooling system in real time and adjust its operating status using intelligent algorithms to achieve intelligent switching between air-cooling and liquid-cooling modes, avoid energy waste, and improve the overall energy efficiency of the system has become a pressing technical challenge. This invention addresses this problem by proposing an optimization method for a co-source air-liquid cooling system based on acoustic signal analysis. This method uses sensors to collect acoustic data in real time and combines this data with intelligent algorithms to dynamically adjust the operating status of the cooling system, thereby achieving intelligent control and high-efficiency energy management of the cooling system. Summary of the Invention

[0006] This invention provides a method and system for optimizing the energy efficiency of a data center with integrated air and liquid cooling systems, enabling precise monitoring, flexible adjustment, and continuous optimization of the cooling system. In a first aspect, this invention provides a method for optimizing the energy efficiency of a data center with integrated air and liquid cooling systems, comprising:

[0007] Step S1: Acquire the acoustic spectrum signal of the cooling system during operation through sensors to obtain the original acoustic spectrum dataset; perform frequency domain transformation on the original acoustic spectrum dataset to obtain the frequency band distribution feature set;

[0008] Step S2: Extract acoustic features based on the frequency band distribution feature set, determine the system energy consumption status by combining with the pre-established mapping model, and determine the cooling system mode switching requirements based on the relationship between the system energy consumption status and the preset threshold.

[0009] Step S3: Predict the acoustic feature change trend based on the mode switching requirements and determine the mode switching time point;

[0010] Step S4: Adjust the cooling system operating parameters according to the mode switching time point, perform mode switching, and obtain the updated system operating status;

[0011] Step S5: Update the mapping model according to the updated system operating status to obtain the optimized energy consumption state prediction model.

[0012] As a preferred embodiment of the present invention, the step of acquiring the acoustic spectrum signal of the cooling system during operation through a sensor to obtain the original acoustic spectrum dataset includes:

[0013] Acoustic sensors are used to collect audio signals generated by changes in fan speed and sound waves generated by liquid flow; the audio signals and sound waves are continuously sampled to generate acoustic data containing multiple time points; the acoustic data is preprocessed to filter out environmental noise and obtain the original acoustic spectrum dataset.

[0014] As a preferred embodiment of the present invention, the step of performing frequency domain transformation on the original acoustic spectrum dataset to obtain a frequency band distribution feature set includes:

[0015] The original acoustic spectrum dataset is converted into a frequency domain signal using a fast Fourier transform algorithm; the frequency domain signal is discretized into multiple frequency bands, and feature vectors for each frequency band are generated; a frequency band distribution feature set containing low-frequency and high-frequency bands is generated based on the feature vectors.

[0016] As a preferred embodiment of the present invention, the step of extracting acoustic features based on the frequency band distribution feature set and determining the system energy consumption state by combining it with a pre-established mapping model includes:

[0017] Extract low-frequency features corresponding to fan speed changes and high-frequency features corresponding to liquid flow from the frequency band distribution feature set; input the low-frequency features and the high-frequency features into the pre-established acoustic feature-energy consumption state mapping model; determine the current system energy consumption state based on the output of the mapping model.

[0018] As a preferred embodiment of the present invention, determining the cooling system mode switching requirement based on the relationship between the system energy consumption status and a preset threshold includes:

[0019] If the system energy consumption exceeds a preset energy consumption threshold, the frequency band distribution feature set is classified using a support vector machine algorithm; based on the classification result, it is determined whether the cooling system is in a critical state of air cooling or liquid cooling mode; and based on the critical state, the mode switching requirement is generated.

[0020] As a preferred embodiment of the present invention, the step of predicting the acoustic feature change trend based on the mode switching requirement and determining the mode switching time point includes:

[0021] Based on the mode switching requirements, analyze the dynamic changes of acoustic features in the frequency band distribution feature set; use a long short-term memory network algorithm to model the dynamic changes of the acoustic features; predict the trend of acoustic feature changes in the future period based on the modeling results, and generate the mode switching time point.

[0022] As a preferred embodiment of the present invention, the step of adjusting the cooling system operating parameters according to the mode switching time point, performing mode switching, and obtaining the updated system operating state includes:

[0023] Based on the mode switching time point, combined with load monitoring information and temperature sensor data, the optimal switching parameters between air cooling and liquid cooling modes are determined; the operating parameters of the cooling system are adjusted through preset control logic; the switching between air cooling and liquid cooling modes is executed, and the updated system operating status is generated.

[0024] As a preferred embodiment of the present invention, the step of updating the mapping model according to the updated system operating state to obtain an optimized energy consumption state prediction model includes:

[0025] Acquire the acoustic spectrum signal under the updated system operating state to generate a new frequency band distribution feature set; extract new acoustic features based on the new frequency band distribution feature set; update the mapping model between the acoustic features and energy consumption state based on the new acoustic features to generate the optimized energy consumption state prediction model.

[0026] Secondly, the present invention also provides a wind-liquid co-source data center energy efficiency optimization system for implementing the above-mentioned method, the system comprising:

[0027] The conversion unit is used to acquire the acoustic spectrum signal of the cooling system during operation through sensors to obtain the original acoustic spectrum dataset; and to perform frequency domain conversion on the original acoustic spectrum dataset to obtain the frequency band distribution feature set.

[0028] The judgment unit is used to extract acoustic features based on the frequency band distribution feature set, determine the system energy consumption status by combining the pre-established mapping model, and determine the cooling system mode switching requirements based on the relationship between the system energy consumption status and the preset threshold.

[0029] The determining unit is used to predict the acoustic feature change trend based on the mode switching requirements and determine the mode switching time point.

[0030] The update unit is used to adjust the operating parameters of the cooling system according to the mode switching time point, perform mode switching, and obtain the updated system operating status.

[0031] An optimization unit is used to update the mapping model based on the updated system operating status to obtain an optimized energy consumption state prediction model.

[0032] Thirdly, the present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0033] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0034] This invention discloses a method and system for optimizing energy efficiency in data centers using a combined air-cooling and liquid-cooling approach. By collecting acoustic signals from the cooling system during operation, performing frequency domain conversion and feature extraction, and combining this with a pre-established mapping model, the system's energy consumption status is determined. Based on the relationship between the energy consumption status and preset thresholds, mode switching requirements are assessed, and the optimal switching time is determined by predicting the trend of acoustic feature changes. This invention adjusts operating parameters according to the switching time, achieving intelligent switching between air-cooling and liquid-cooling modes, and optimizes the energy consumption prediction model based on the updated operating status. This method achieves real-time monitoring and intelligent adjustment of the cooling system through acoustic analysis, effectively improving system energy efficiency and reducing energy consumption, providing a new technical solution for optimizing cooling systems in data centers and other similar scenarios. Attached Figure Description

[0035] Figure 1 This is a flowchart of a wind-liquid homogeneous data center energy efficiency optimization method according to an embodiment of the present invention;

[0036] Figure 2 This is a structural diagram of a data center energy efficiency optimization system based on the principle of wind and liquid homology, according to an embodiment of the present invention. Detailed Implementation

[0037] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] like Figure 1 This embodiment of a method for optimizing energy efficiency in a data center based on the co-source of air and liquid can specifically include:

[0039] Step S1: Acquire the acoustic spectrum signal of the cooling system during operation through sensors to obtain the original acoustic spectrum dataset; perform frequency domain transformation on the original acoustic spectrum dataset to obtain the frequency band distribution feature set;

[0040] The process involves acquiring acoustic spectrum signals from the cooling system during operation via sensors to obtain a raw acoustic spectrum dataset. This includes: acquiring audio signals generated by changes in fan speed and sound waves generated by liquid flow using acoustic sensors; continuously sampling the audio signals and sound waves to generate acoustic data containing multiple time points; and preprocessing the acoustic data to filter out environmental noise to obtain the raw acoustic spectrum dataset.

[0041] Specifically, acoustic sensors capture low-frequency sound wave signals generated by changes in fan speed and high-frequency sound wave signals generated by liquid flow. Changes in these signals reflect the operating status of the cooling system. By continuously sampling these acoustic signals and recording data at multiple time points, an acoustic dataset containing information from multiple moments is generated. To improve the accuracy and reliability of the signals, preprocessing is performed to filter noise from the collected acoustic data, removing environmental noise and other interference to ensure a high-quality raw acoustic spectrum dataset. The above technical solution guarantees the accuracy of data acquisition and provides a basis for subsequent energy consumption status determination and mode switching.

[0042] Furthermore, the original acoustic spectrum dataset is frequency domain transformed to obtain a frequency band distribution feature set, including:

[0043] The original acoustic spectrum dataset is converted into a frequency domain signal using a fast Fourier transform; the frequency domain signal is discretized into multiple frequency bands, and feature vectors for each frequency band are generated; a frequency band distribution feature set containing low-frequency and high-frequency bands is generated based on the feature vectors.

[0044] Specifically, by performing frequency domain transformation on the original acoustic spectrum dataset, the characteristics of the cooling system's operating state are extracted. That is, through Fast Fourier Transform (FFT), the original time-domain acoustic signal is converted into a frequency-domain signal, thereby obtaining the energy distribution of the signal at different frequencies. The frequency-domain transformed signal is discretized into multiple frequency bands, which represent the energy characteristics within different frequency ranges. By calculating the feature vector of each frequency band, a frequency band distribution feature set is generated. The feature vector includes the amplitude spectrum, energy distribution, frequency distribution, spectral entropy, and root mean square value of the corresponding frequency band. The amplitude spectrum represents the signal intensity within a certain frequency band, typically calculated by measuring the amplitude of the signal in that band. The frequency distribution, obtained from the above, reflects the acoustic energy level of the frequency band. The energy distribution represents the total energy distribution of the corresponding frequency band, indicating the overall energy quantization value of the signal within that band. The frequency distribution represents the frequency distribution of the acoustic signal within a specific frequency band, including at least the peak frequency and the center value of the frequency. The spectral entropy reflects the complexity and information content of the frequency band signal; the higher the spectral entropy, the higher the signal complexity of the frequency band. The root mean square value describes the energy intensity of the signal within the frequency band, usually calculated by the square root average of the signal within that band. The frequency band distribution feature set contains feature information from both low-frequency and high-frequency bands, reflecting different acoustic characteristics of fan speed changes and liquid flow, respectively. The low-frequency band mainly corresponds to the fan's operating status, while the high-frequency band corresponds to the state of liquid flow. These feature vectors provide a basis for subsequent data center energy consumption prediction.

[0045] Step S2: Extract acoustic features based on the frequency band distribution feature set, and determine the system energy consumption status by combining it with a pre-established mapping model; determine the cooling system mode switching requirements based on the relationship between the system energy consumption status and a preset threshold.

[0046] The process of extracting acoustic features based on the frequency band distribution feature set and determining the system energy consumption status by combining them with a pre-established mapping model includes:

[0047] Extract low-frequency features corresponding to fan speed changes and high-frequency features corresponding to liquid flow from the frequency band distribution feature set; input the low-frequency features and the high-frequency features into the pre-established acoustic feature-energy consumption state mapping model; determine the current energy consumption state based on the output of the mapping model.

[0048] Specifically, by extracting acoustic features from a frequency band distribution feature set and combining them with a pre-established mapping model, the energy consumption status of the cooling system is determined, solving the problem that traditional cooling systems cannot monitor and optimize energy efficiency in real time. This method collects acoustic signals from the cooling system during operation, performs frequency domain conversion, and obtains a frequency band distribution feature set. The low-frequency band corresponds to changes in fan speed, while the high-frequency band corresponds to acoustic signals generated by liquid flow. After feature extraction, these acoustic signals yield low-frequency and high-frequency feature vectors reflecting the fan operation and liquid flow states, respectively. The low-frequency features caused by fan speed changes reflect the air-cooling mode of the cooling system, while the high-frequency features generated by liquid flow reflect the system state under liquid-cooling mode. The extracted low-frequency and high-frequency features are input into a pre-established acoustic feature-energy consumption status mapping model. The model outputs the current system energy consumption status based on this input information. Regarding power consumption, this mapping model, based on historical data learning, establishes a mapping relationship between acoustic features and energy consumption status. It can accurately determine the current energy consumption level of the system based on the input acoustic features. For example, in data center applications, when the fan speed increases, the low-frequency features will be enhanced, indicating that the system is operating in air-cooled mode; if the liquid flow rate increases, the high-frequency features will be enhanced, indicating that the system is operating in liquid-cooled mode. By monitoring these changes in real time and combining them with the pre-established mapping model, the cooling mode can be dynamically adjusted to optimize energy efficiency. The mapping model is trained using the historical frequency band features of the cooling system and the corresponding energy consumption status as training data. This technical solution solves the problem of insufficient energy efficiency of the cooling system under different loads and environments, ensuring that the cooling system always operates in the most suitable mode, reducing energy waste and improving the overall efficiency of the system.

[0049] Furthermore, based on the relationship between the system energy consumption status and a preset threshold, the cooling system mode switching requirement is determined, including:

[0050] If the system energy consumption exceeds a preset energy consumption threshold, the frequency band distribution feature set and frequency band feature vector are input into a vector machine model to classify the frequency band distribution feature set; based on the classification result, it is determined whether the cooling system is in a critical state of air cooling or liquid cooling mode; and based on the critical state, the mode switching requirement is generated.

[0051] Specifically, by comparing the system's energy consumption status with a preset threshold, it is determined whether the cooling system needs to switch modes to optimize cooling efficiency. Here, the energy consumption status refers to power consumption. In the implementation of the above technical solution, it is determined whether the current energy consumption status exceeds the set energy consumption threshold. If the system's energy consumption status exceeds this threshold, it is further determined whether the cooling system is in a critical state between air cooling and liquid cooling modes. A critical state typically refers to a system energy consumption approaching or exceeding its optimal operating state, which may lead to decreased efficiency or excessive energy consumption. By inputting the acoustic feature vector corresponding to the current data center cooling system into the above vector machine model to obtain the energy consumption status classification result, it is possible to determine whether there is a mode switching requirement. If the classification indicates that the system is in a critical state, a mode switching requirement is generated. The signal indicates a switch from one mode to another, such as from air cooling to liquid cooling, thereby improving the overall energy efficiency of the cooling system. The technical solution, through a close integration of algorithm optimization and data analysis, achieves intelligent control of the cooling system, reducing unnecessary energy waste and ensuring stable system operation. For example, when the data center load increases, air cooling may not meet the heat load demand. The system predicts this through changes in acoustic characteristics and generates a switching demand for liquid cooling mode, thus avoiding overheating, reducing energy consumption, and maintaining effective system cooling. This technical solution solves the problems of low cooling system energy efficiency and untimely mode switching by dynamically judging and optimizing the timing of mode switching, improving overall energy efficiency and reducing unnecessary energy waste.

[0052] Step S3: Predict the acoustic characteristic change trend based on the mode switching requirements, and determine the mode switching time point; specifically including:

[0053] Based on the mode switching requirements, analyze the dynamic changes of acoustic features in the frequency band distribution feature set; use a long short-term memory network model to model the dynamic changes of acoustic features; predict the trend of acoustic feature changes in the future period based on the modeling results, and generate the mode switching time point.

[0054] Specifically, the above technical solution determines the mode switching time of the cooling system by predicting the changing trends of acoustic characteristics, thereby achieving intelligent cooling system optimization. This step relies on a long short-term memory network model, which models the dynamic changes of acoustic characteristics to predict the changing trends of acoustic characteristics over a future period and generates mode switching time points based on these trends. During implementation, according to mode switching requirements, the system analyzes the frequency band distribution characteristics of the collected acoustic data. These characteristics, after Fourier transform, form feature vectors representing the dynamic changes in low-frequency and high-frequency bands. The low-frequency band typically corresponds to fan speed, and the high-frequency band typically corresponds to liquid flow. By analyzing these changes in acoustic characteristics, energy consumption fluctuations or system state changes can be identified. For example, in air-cooled mode, an increase in fan speed leads to an enhancement of low-frequency characteristics, while in liquid-cooled mode, the liquid flow characteristics are enhanced. Changes can indicate whether a system is about to reach a critical energy efficiency state. Employing a Long Short-Term Memory (LSTM) network model, which is suitable for processing time-series data and can capture long-term dependencies in feature changes over time, LSTM, through its unique network structure, can learn and remember the patterns of acoustic feature changes over time, predict the trend of acoustic feature changes in the future, and thus predict when a mode switching requirement will occur. Specifically, the LSTM model is trained on historical acoustic feature data, learning the change patterns of acoustic features under different loads and temperatures, predicting future acoustic signal changes, and determining whether a mode switching is necessary. For example, in a data center environment, when the system load increases, the increase in fan speed will gradually enhance the low-frequency acoustic features. The LSTM model can predict this trend and determine when to switch to liquid cooling mode to avoid excessive energy consumption and system overheating. During prediction, the LSTM model can comprehensively consider factors such as fan speed, liquid flow rate, and temperature to analyze system state changes in the future. In this way, the system can predict changes in cooling demand in advance, thereby determining the optimal switching time to achieve energy-saving optimization.

[0055] Step S4: Adjust the cooling system operating parameters according to the mode switching time point, perform mode switching, and obtain the updated system operating status, including:

[0056] Based on the mode switching time point, combined with load monitoring information and temperature sensor data, the optimal switching parameters between air cooling and liquid cooling modes are determined; the operating parameters of the cooling system are adjusted through preset control logic; the switching between air cooling and liquid cooling modes is executed, and the updated system operating status is generated.

[0057] Specifically, the technical solution in step S4 combines load monitoring information and temperature sensor data to adjust the cooling system operating parameters at the mode switching time point and execute mode switching to generate an updated system operating status. In the implementation process, load monitoring information and temperature sensor data are used as input data sources to provide real-time feedback on the operating status of the cooling system. Load monitoring information provides real-time data on changes in system load, reflecting indicators such as CPU utilization and memory usage of servers in the data center, thereby indirectly inferring changes in cooling demand. Temperature sensor data provides temperature changes in the cooling system and the data center environment, directly affecting the judgment of cooling effect. When abnormal fluctuations occur in load monitoring information and temperature sensor data, the system needs to dynamically adjust the cooling mode to ensure optimal energy efficiency and temperature control.

[0058] After obtaining the mode switching time, the preset control logic begins execution. The control logic determines whether to switch from air cooling mode to liquid cooling mode, or vice versa, based on set parameter ranges such as upper temperature limits and load thresholds. The execution of the control logic is not solely based on static rules, but rather adjusts according to the analysis results of the previous step and real-time data changes. For example, when the server load continues to increase and the ambient temperature exceeds the effective operating range of air cooling mode, the control logic automatically calculates the optimal time to switch to liquid cooling mode, thereby activating the liquid cooling system to prevent overheating.

[0059] During this process, the optimal switching parameters between air-cooling and liquid-cooling modes are determined through comprehensive calculations, taking into account both real-time temperature data from temperature sensors and load monitoring information to optimize the balance between the air-cooling and liquid-cooling systems. For example, when the temperature in air-cooling mode approaches a set threshold, the system adjusts the operating parameters of liquid-cooling mode, such as liquid flow rate and pump speed, to ensure that switching does not cause excessive energy consumption. Simultaneously, the liquid-cooling mode may also optimize its parameters based on changes in load information, enabling it to effectively leverage its advantages under high-load conditions and reduce energy waste.

[0060] By executing these mode-switching control logics, the system can ultimately automatically switch between air-cooling and liquid-cooling modes at precise times, optimizing cooling performance and avoiding energy inefficiencies caused by excessively high or low temperatures. This optimization method, through precise mode-switching timing and parameter adjustments, not only improves the overall energy efficiency of the data center cooling system but also effectively reduces energy consumption while maintaining system stability. For example, during peak data center load periods, assuming load monitoring information indicates a sharp increase in server processing capacity demand, and temperature sensors show that the ambient temperature has reached a preset critical value, the system will promptly determine and automatically adjust the cooling mode switching time. Through optimized control logic, the air-cooling system will switch to liquid-cooling mode in advance when the temperature approaches the critical point, and precisely adjust the liquid flow rate and pump speed of the liquid-cooling system to avoid overheating, ensuring the stability and efficiency of the cooling effect.

[0061] The above technical solution solves the problem that traditional cooling modes in data center cooling systems cannot be flexibly adjusted and respond to load changes in a timely manner. By comprehensively utilizing load monitoring information and temperature sensor data, the system can ensure the best cooling effect under different load and temperature conditions without increasing additional energy consumption, thereby improving the system's energy efficiency and reducing energy waste.

[0062] Step S5: Update the mapping model according to the updated system operating status to obtain the optimized energy consumption state prediction model, including:

[0063] Acquire the acoustic spectrum signal under the updated system operating state to generate a new frequency band distribution feature set; extract new acoustic features based on the new frequency band distribution feature set; update the mapping model between the acoustic features and energy consumption state based on the new acoustic features to generate the optimized energy consumption state prediction model.

[0064] Specifically, step S5 involves updating the mapping model between acoustic features and energy consumption status based on the updated system operating status to obtain an optimized energy consumption status prediction model. This process, through a series of data acquisition, processing, and algorithm optimization operations, aims to further improve system energy efficiency and solve the technical problem of traditional cooling systems' inflexible adjustment. During the operation of the cooling system, the acoustic spectrum signals collected in real time by sensors reflect the dynamic changes in the system's operating status. The frequency band distribution feature set generated from these signals includes acoustic features within different frequency ranges. The low-frequency band mainly corresponds to the sound wave signals generated by changes in fan speed, while the high-frequency band reflects the noise generated by liquid flow. These features provide important basis for judging system energy efficiency and mode switching.

[0065] When the system's operating state changes, new acoustic spectrum signals are collected and processed to generate a new frequency band distribution feature set. This feature set not only includes the acoustic characteristics of the fan and liquid flow, but also converts these time-domain signals into frequency-domain signals through fast Fourier transform, thus providing richer data input for subsequent energy efficiency state prediction. The new frequency band distribution feature set provides more accurate characteristics that reflect the operating state of the cooling system, and can better match the actual energy consumption performance of the system under different load and environmental conditions.

[0066] By utilizing the new frequency band distribution feature set, new acoustic features are further extracted, representing the energy consumption characteristics of the cooling system under the current operating mode. These acoustic features are combined with the previously established energy consumption state mapping model, and calculated and predicted through input model, to obtain the current energy consumption state of the system. To adapt to the dynamic changes of the cooling system, the mapping model needs to be updated based on the new data to ensure that it can more accurately predict the system's energy efficiency level.

[0067] Through the above process, the mapping model continuously adjusts its parameters based on the latest acoustic characteristics, making the predicted energy consumption state more accurate and able to respond promptly to changes in system operation. This optimization process effectively solves the problems of inflexible and untimely energy efficiency optimization in existing technologies. The optimized energy consumption state prediction model can achieve more intelligent and accurate switching between air-cooled and liquid-cooled modes. When the load increases, it can predict in advance and switch to liquid-cooled mode, thereby avoiding excessive energy consumption and improving cooling efficiency and system stability. For example, in a practical application scenario, when the fan speed in the cooling system changes due to increased load, the low-frequency band features with concentrated frequency band distribution characteristics will show an increasing trend. Through the updated mapping model, the system energy efficiency state at this time can be accurately determined, and it can be predicted whether to switch to liquid-cooled mode to adapt to higher cooling demands. If the mapping model detects that the energy efficiency threshold is close, the system can switch modes at the appropriate time, thereby achieving energy saving and optimization effects.

[0068] This invention also provides a wind-liquid co-source data center energy efficiency optimization system for implementing the above-mentioned method, such as... Figure 2 As shown, the system includes:

[0069] The conversion unit is used to acquire the acoustic spectrum signal of the cooling system during operation through sensors to obtain the original acoustic spectrum dataset; and to perform frequency domain conversion on the original acoustic spectrum dataset to obtain the frequency band distribution feature set.

[0070] The judgment unit is used to extract acoustic features based on the frequency band distribution feature set, determine the system energy consumption status by combining the pre-established mapping model, and determine the cooling system mode switching requirements based on the relationship between the system energy consumption status and the preset threshold.

[0071] The determining unit is used to predict the acoustic feature change trend based on the mode switching requirements and determine the mode switching time point.

[0072] The update unit is used to adjust the operating parameters of the cooling system according to the mode switching time point, perform mode switching, and obtain the updated system operating status.

[0073] An optimization unit is used to update the mapping model based on the updated system operating status to obtain an optimized energy consumption state prediction model.

[0074] The present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0075] In summary, this invention provides accurate feedback on the real-time status of the cooling system through acoustic spectrum signals collected by sensors. By performing frequency domain transformation on these signals, a frequency band distribution feature set is obtained, which can clearly distinguish different frequency signals generated by changes in fan speed and liquid flow. These signals not only represent the operating status of the cooling system but also reflect energy efficiency fluctuations caused by changes in system load. This step provides a precise real-time data foundation for subsequent optimization, avoiding the problem of traditional cooling systems being unable to flexibly adapt to load changes. By combining acoustic features extracted from the frequency band distribution feature set with a pre-established energy consumption state mapping model, the current energy consumption state of the system can be effectively assessed. Through the mapping of these features by the algorithm, the system can determine in real time whether to switch operating modes, avoiding excessive energy waste. For example, when the air-cooling mode cannot meet the cooling demand, the system will automatically determine and switch to liquid-cooling mode, thereby improving cooling efficiency and reducing energy consumption. Furthermore, by introducing a mode switching demand prediction algorithm, a long short-term memory network is used to model the trend of acoustic feature changes, thereby predicting changes in the operating state of the cooling system in advance, optimizing the timing of mode switching, ensuring accurate switching of cooling modes, avoiding over-reliance on fixed rules and manual intervention, and improving the system's adaptive adjustment capability. By continuously optimizing the energy consumption state prediction model, the system can continuously update the mapping model according to the real-time operating status, making energy efficiency prediction more accurate and adaptable. This optimization cycle ensures that the cooling system always operates in the most suitable mode, avoiding unnecessary energy waste, and improving the long-term stability and efficiency of the system through the self-updating mapping model.

[0076] The above are only some preferred embodiments of the present invention, but the present invention is not limited thereto, and many improvements and modifications can be made. Any improvements and modifications made based on the basic principles of the present invention should be considered to fall within the protection scope of the present invention.

Claims

1. A method for optimizing energy efficiency in a data center with wind and liquid homology, characterized in that, include: Step S1: Acquire the acoustic spectrum signal of the cooling system during operation through sensors to obtain the raw acoustic spectrum dataset; The original acoustic spectrum dataset is transformed in the frequency domain to obtain a frequency band distribution feature set; Step S2: Extract acoustic features based on the frequency band distribution feature set, determine the system energy consumption status by combining with the pre-established mapping model, and determine the cooling system mode switching requirements based on the relationship between the system energy consumption status and the preset threshold. Step S3: Predict the acoustic feature change trend based on the mode switching requirements and determine the mode switching time point; including: analyzing the dynamic changes of acoustic features in the frequency band distribution feature set based on the mode switching requirements; modeling the dynamic changes of the acoustic features using a long short-term memory network algorithm; predicting the acoustic feature change trend in the future period based on the modeling results, and generating the mode switching time point; Step S4: Adjust the cooling system operating parameters according to the mode switching time point, perform mode switching, and obtain the updated system operating status; Step S5: Update the mapping model according to the updated system operating status to obtain the optimized energy consumption state prediction model; The determination of cooling system mode switching requirements based on the relationship between the system energy consumption status and a preset threshold includes: If the system energy consumption exceeds a preset energy consumption threshold, the frequency band distribution feature set is classified using a support vector machine algorithm; based on the classification result, it is determined whether the cooling system is in a critical state of air cooling or liquid cooling mode; and based on the critical state, the mode switching requirement is generated.

2. The method as described in claim 1, characterized in that, The process of acquiring acoustic spectrum signals from the cooling system during operation via sensors to obtain a raw acoustic spectrum dataset includes: Acoustic sensors are used to collect audio signals generated by changes in fan speed and sound waves generated by liquid flow; the audio signals and sound waves are continuously sampled to generate acoustic data containing multiple time points; the acoustic data is preprocessed to filter out environmental noise and obtain the original acoustic spectrum dataset.

3. The method as described in claim 1, characterized in that, The step of performing frequency domain transformation on the original acoustic spectrum dataset to obtain a frequency band distribution feature set includes: The original acoustic spectrum dataset is converted into a frequency domain signal using a fast Fourier transform algorithm; the frequency domain signal is discretized into multiple frequency bands, and feature vectors for each frequency band are generated; a frequency band distribution feature set containing low-frequency and high-frequency bands is generated based on the feature vectors.

4. The method as described in claim 1, characterized in that, The step of extracting acoustic features based on the frequency band distribution feature set and determining the system energy consumption status by combining it with a pre-established mapping model includes: Extract low-frequency features corresponding to fan speed changes and high-frequency features corresponding to liquid flow from the frequency band distribution feature set; input the low-frequency features and the high-frequency features into the pre-established acoustic feature-energy consumption state mapping model; determine the current system energy consumption state based on the output of the mapping model.

5. The method as described in claim 1, characterized in that, The step of adjusting the cooling system operating parameters according to the mode switching time point, performing mode switching, and obtaining the updated system operating status includes: Based on the mode switching time point, combined with load monitoring information and temperature sensor data, the optimal switching parameters between air cooling and liquid cooling modes are determined; the operating parameters of the cooling system are adjusted through preset control logic; the switching between air cooling and liquid cooling modes is executed, and the updated system operating status is generated.

6. The method as described in claim 1, characterized in that, The step of updating the mapping model based on the updated system operating state to obtain the optimized energy consumption state prediction model includes: Acquire the acoustic spectrum signal under the updated system operating state to generate a new frequency band distribution feature set; extract new acoustic features based on the new frequency band distribution feature set; update the mapping model between the acoustic features and energy consumption state based on the new acoustic features to generate the optimized energy consumption state prediction model.

7. A wind-liquid co-source data center energy efficiency optimization system, used to implement the method as described in any one of claims 1-6, characterized in that, The system includes: The conversion unit is used to acquire the acoustic spectrum signal of the cooling system during operation through sensors to obtain the original acoustic spectrum dataset; and to perform frequency domain conversion on the original acoustic spectrum dataset to obtain the frequency band distribution feature set. The judgment unit is used to extract acoustic features based on the frequency band distribution feature set, determine the system energy consumption status by combining the pre-established mapping model, and determine the cooling system mode switching requirements based on the relationship between the system energy consumption status and the preset threshold. The determining unit is used to predict the acoustic feature change trend based on the mode switching requirements and determine the mode switching time point. The update unit is used to adjust the operating parameters of the cooling system according to the mode switching time point, perform mode switching, and obtain the updated system operating status. An optimization unit is used to update the mapping model based on the updated system operating status to obtain an optimized energy consumption state prediction model.

8. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the method as described in any one of claims 1-6.

Images