Method and system for dynamically and automatically adjusting rotating speed of fan, electronic equipment and medium

Abstract

The invention provides a fan rotating speed dynamic automatic adjusting method and system, electronic equipment and a storage medium, and aims to solve the problems that air volume adjustment lacks accuracy and is low in energy utilization rate. Inputting the data into a trained preset prediction model to generate pre-judgment data of the wind speed and the local temperature of the equipment in a period of time in the future; the expected rotating speed and the air volume of the fan are determined according to the data, and the optimized pre-adjusting rotating speed value and the optimized air volume value are obtained so as to meet the large-air-volume and small-enthalpy-difference cooling requirements of the equipment; comparing the pre-regulation rotating speed value with the current fan rotating speed, comparing the pre-regulation air volume value with the current air volume, and respectively calculating difference values; whether the difference value exceeds an adjusting threshold value or not is judged, if yes, the rotating speed and the air volume are adjusted to a pre-adjusting rotating speed value and an air volume value, and the air volume is accurately sent to a local high-temperature area of the equipment; and if not, maintaining the current rotating speed and the current air volume of the fan. Air supply of the system can be optimized, and energy conservation and precise cooling are achieved.

Description

Technical Field

[0001] The present disclosure relates to the field of communication and energy-saving technology, and in particular to a method for dynamically and automatically adjusting a fan speed, a system for dynamically and automatically adjusting a fan speed, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the increase in high-temperature equipment scenarios in data centers and industrial computer rooms, the role of air-conditioning systems in equipment operation has become increasingly important. The stability and life of equipment in high-temperature environments depend to a large extent on the cooling effect of the air-conditioning system. However, traditional air-conditioning systems often use fixed wind speeds or extensive full-area air supply modes, which cannot meet the dual needs of high-temperature equipment for precise cooling and energy saving.

[0003] At present, most air-conditioning systems disperse cold air to the equipment area through preset fixed wind speed and air volume. Although this traditional mode can alleviate the heat dissipation demand of the equipment to a certain extent, it has obvious defects. On the one hand, the air-conditioning fan cannot adjust the air volume according to the real-time temperature changes of the equipment and local needs in the fixed speed mode, resulting in some high-temperature equipment not getting enough cooling, while low-temperature equipment may be over-cooled, causing energy waste. On the other hand, the air volume distribution of the full-area air supply mode is not accurate enough to provide directional air supply to high-temperature areas, and the problem of excessive air volume dispersion often occurs, affecting the cooling efficiency.

[0004] In recent years, some studies have attempted to improve the energy efficiency of air-conditioning systems by adjusting air volume and speed control. A simple PID controller is used to adjust the fan speed or a temperature sensor is used to monitor real-time temperature changes and then set a predetermined air volume adjustment plan. The following problems have been exposed in actual applications: First, the response speed and adjustment accuracy of the PID controller are limited, making it difficult to cope with the rapid heat dissipation requirements of high-temperature equipment; second, the preset adjustment plan lacks intelligence and real-time performance, and cannot accurately distribute air volume according to the dynamic changes in equipment temperature; finally, the cooling effect on local areas of high-temperature equipment is poor and cannot meet the current data center scenario's demand for precise cooling.

[0005] In summary, the existing technology has the following shortcomings in energy saving and precise air supply and cooling of air-conditioning fans: air volume adjustment lacks real-time and precision; the cooling effect is poor, and local cooling of high-temperature equipment is insufficient; low energy utilization efficiency causes a large amount of energy waste. Therefore, there is an urgent need for an air-conditioning fan adjustment method that combines intelligent prediction and dynamic optimization to achieve energy saving and efficient precise cooling, thereby meeting the higher requirements of modern high-temperature equipment scenarios for air-conditioning systems. Summary of the invention

[0006] In order to at least solve the problems of lack of real-time and precision in air volume regulation, poor cooling effect and low energy efficiency in the prior art, the present invention provides a fan speed dynamic automatic regulation method, a fan speed dynamic automatic regulation system, an electronic device and a computer-readable storage medium, which can optimize the air supply of an air conditioning system to achieve energy saving and precise air supply and cooling.

[0007] In a first aspect, the present disclosure provides a method for dynamically and automatically adjusting the speed of a fan, the method comprising:

[0008] Obtain wind speed, wind direction, temperature and real-time temperature data of the fan's surrounding environment, and input them into the trained preset prediction model to generate wind speed prediction data and prediction data of the local temperature of the equipment for a period of time in the future;

[0009] Determine the expected speed and air volume of the fan according to the wind speed prediction data and the local temperature prediction data of the equipment, and obtain the optimized pre-adjusted speed value and pre-adjusted air volume value to meet the cooling demand of the equipment with large air volume and small enthalpy difference;

[0010] Comparing the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, and calculating the difference between the two respectively;

[0011] Determine whether the difference exceeds the preset adjustment threshold. If the difference exceeds the preset adjustment threshold, control the fan to adjust the speed and air volume to the pre-adjusted speed value and pre-adjusted air volume value, and accurately deliver the air volume to the local high-temperature area of ​​the equipment to reduce the temperature; if the difference does not exceed the adjustment threshold, maintain the current fan speed and air volume.

[0012] Furthermore, the method further comprises:

[0013] Analyze the fluctuation pattern of real-time wind speed, extract the self-similar features in wind speed fluctuation, and identify the periodic fluctuation and random fluctuation components of wind speed;

[0014] When comparing the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, the difference between the two is calculated based on the identified periodic fluctuation and random fluctuation components of the wind speed.

[0015] Furthermore, the method further comprises:

[0016] Collect historical wind speed, wind direction, temperature and local temperature data of the fan's surrounding environment to build a training data set;

[0017] The training data set is used to train the preset prediction model so that the preset prediction model can predict the wind speed and local temperature changes of the equipment within a period of time in the future, thereby obtaining a trained preset prediction model.

[0018] Furthermore, the method further comprises:

[0019] The adjusted fan speed, air volume, environmental real-time data and equipment temperature data are stored together, and the database is updated to optimize the preset prediction model.

[0020] Furthermore,

[0021] The historical wind speed, wind direction, temperature and local temperature data of the equipment surrounding the fan are collected to construct a training data set, including:

[0022] S11. Collect historical data of the wind turbine’s surrounding environment, including wind speed V h 、wind directionθ h , Temperature T h And the local temperature of the equipment T eq :

[0023] S12. Preprocess the collected historical data, including data cleaning, denoising, and normalization, to obtain standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n ;

[0024] S13. The standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n Combined into training data set D train :

[0025] D train ={(V h,n ,θ h,n ,T h,n ,T eq,n )};

[0026] The method of using the training data set to train the preset prediction model so that the preset prediction model can predict the wind speed and local temperature changes of the equipment within a period of time in the future, and obtaining the trained preset prediction model, includes:

[0027] S21. The training data set D train Organized into input and output pairs suitable for time series forecasting:

[0028] S22. Define the neural network prediction model f θ To preset the prediction model, the neural network parameter is θ, which is used to capture the time dependence and nonlinear relationship between environmental data and device temperature;

[0029] S23. Using the loss function L(θ) to train the neural network prediction model;

[0030] S24. Update the neural network parameters θ by iteratively minimizing the loss function L(θ), and continue iterative training until the loss function L(θ) converges to the preset threshold ∈ L , and obtain the optimized neural network prediction model as the trained preset prediction model.

[0031] Further, the expected speed and air volume of the fan are determined according to the wind speed prediction data and the equipment local temperature prediction data to obtain the optimized pre-adjusted speed value and pre-adjusted air volume value, including:

[0032] S51: Receive wind speed forecast data And equipment local temperature prediction data Combined with real-time ambient temperature and real-time local device temperature Construct fan speed and air volume optimization function:

[0033]

[0034] Where n is the fan speed, Q is the fan air volume, and Represent the predicted local temperature of the device and the real-time local temperature of the device, respectively. and Represent the predicted wind speed and real-time wind speed respectively. Where Δh represents the enthalpy difference, H(Q) represents the air volume per unit enthalpy difference, α1, β1 and γ1 are the speed adjustment weight coefficient, air volume adjustment weight coefficient and enthalpy difference air volume optimization weight coefficient respectively;

[0035] S52: Solve the optimization function F opt (n,Q), determine the pre-adjusted speed n that satisfies the minimization target opt and air volume Q opt :

[0036]

[0037] S53: Calculate the pre-adjusted air volume value Q that meets the requirements of large air volume and small enthalpy difference through the enthalpy difference optimization formula opt :

[0038]

[0039] in, represents the difference in local temperature of the equipment, and R represents the ambient heat transfer constant.

[0040] Furthermore,

[0041] The analysis of the fluctuation law of real-time wind speed, extraction of self-similar features in wind speed fluctuation, and identification of periodic fluctuation and random fluctuation components of wind speed include:

[0042] S61: Get real-time wind speed data Time series:

[0043]

[0044] in, represents the real-time wind speed at time step t, t w is the analysis window length;

[0045] S62: Construct a fractal dimension model of real-time wind speed based on spatial fractal algorithm and calculate the fractal dimension D of time series f :

[0046]

[0047] Among them, ∈ is the segmentation scale, N(∈) represents the minimum number of segments covering the wind speed time series at scale ∈;

[0048] S63: Through fractal dimension D f Extract the self-similar features of wind speed time series. If D f ≈1.5, it indicates that the wind speed change has a significant random fluctuation component; if 1.0 <D f <1.5, it indicates that the wind speed change has periodic fluctuation characteristics;

[0049] S64: The wind speed time series is decomposed using the characteristic separation model of periodic fluctuations and random fluctuations. series It is expressed as:

[0050] V series =V periodic +V random ;

[0051] Among them, V periodic Represents the periodic fluctuation component of wind speed, V random Represents the random fluctuation component of wind speed, V series The remainder after deducting the periodic fluctuations;

[0052] The method of comparing the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, respectively calculating the difference between the two based on the identified periodic fluctuation and random fluctuation components of the wind speed, includes:

[0053] S71: Obtaining the periodic fluctuation component V of the wind speed periodic and the random fluctuation component V random , combined with the current fan speed ncurrent and air volume Q current , calculate the predicted periodic speed n periodic and randomly adjust the speed n random :

[0054]

[0055] Among them, n opt is the pre-adjusted speed value, is the predicted total wind speed, n periodic Indicates the speed adjustment to periodic fluctuations, n random Indicates the compensation of random fluctuations in speed;

[0056] S72: Compare the current fan speed n current and pre-adjusted speed n opt , calculate the total speed difference Δn and the total air volume difference ΔQ:

[0057] Δn=|n current -(n periodic +n random )|;

[0058] ΔQ=|Q current -Q opt |;

[0059] Among them, Δn represents the difference between the current speed and the total adjusted speed, ΔQ represents the difference between the current air volume and the pre-adjusted air volume, Q current is the current air volume, Q opt To pre-adjust the air volume value.

[0060] Further, the determination of whether the difference exceeds a preset adjustment threshold value, if the difference exceeds the preset adjustment threshold value, controlling the fan to adjust the speed and air volume to the pre-adjusted speed value and pre-adjusted air volume value, and accurately delivering the air volume to the local high temperature area of ​​the equipment to reduce the temperature; if the difference does not exceed the adjustment threshold value, maintaining the current fan speed and air volume, including:

[0061] S81: Determine whether the speed difference Δn and the air volume difference ΔQ exceed the preset adjustment threshold ∈ n and ∈ Q , the judgment conditions are:

[0062]

[0063] Where Δn and ΔQ are the root mean square difference between the current speed and the total adjusted speed, and the integral difference between the current air volume and the pre-adjusted air volume, respectively. w is the observation time window, n periodic (k) and n random (k) are the periodic and random speed adjustment components, ∈n and ∈ Q They are the adjustment thresholds for the rotation speed and air volume respectively;

[0064] S82: If the condition is met, adjust the fan speed n in real time adjust and air volume Q adjust :

[0065]

[0066] Among them, η n and η Q is the speed and air volume adjustment rate, σ T Adjust the sensitivity parameter for the temperature difference, f t Adjust frequency for air volume fluctuations;

[0067] S83: Adjust the air volume Q adjust Accurately guide the heat to the local high temperature area of ​​the equipment, through the heat transfer rate R heat To verify:

[0068]

[0069] Where ρ is the air density, c p is the specific heat capacity of air, L, W and H are the length, width and height of the local area of ​​the equipment, Indicates the actual equivalent temperature of the local high temperature area of ​​the equipment, T target Indicates the target temperature;

[0070] S84: If Δn≤∈ n And ΔQ≤∈ Q , the current fan status remains unchanged, and the status data S is recorded current :

[0071]

[0072] And stored in database D state :

[0073] D state =D state ∪S current .

[0074] In a second aspect, the present disclosure provides a system for dynamically and automatically adjusting the speed of a fan, the system comprising:

[0075] A prediction module is configured to obtain wind speed, wind direction, temperature and real-time temperature data of the fan surrounding environment, and input the data into a trained preset prediction model to generate wind speed prediction data and prediction data of the local temperature of the equipment for a period of time in the future;

[0076] A determination module, which is configured to determine the expected speed and air volume of the fan according to the wind speed prediction data and the equipment local temperature prediction data, and obtain an optimized pre-adjusted speed value and pre-adjusted air volume value to meet the cooling demand of the equipment with large air volume and small enthalpy difference;

[0077] a difference calculation module, configured to compare the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, and respectively calculate the difference between the two;

[0078] A judgment and adjustment module is configured to judge whether the difference exceeds a preset adjustment threshold. If the difference exceeds the preset adjustment threshold, the fan is controlled to adjust the speed and air volume to the pre-adjusted speed value and pre-adjusted air volume value, and the air volume is accurately delivered to the local high-temperature area of ​​the equipment to reduce the temperature; if the difference does not exceed the adjustment threshold, the current fan speed and air volume are maintained.

[0079] Furthermore, the system also includes an identification module;

[0080] The identification module is configured to analyze the fluctuation pattern of the real-time wind speed, extract the self-similar features in the wind speed fluctuation, and identify the periodic fluctuation and random fluctuation components of the wind speed;

[0081] The difference calculation module is specifically configured to calculate the difference between the pre-adjusted speed value and the current fan speed, and the pre-adjusted air volume value and the current air volume, based on the identified periodic fluctuation and random fluctuation components of the wind speed.

[0082] In a third aspect, the present disclosure provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes a method for dynamic and automatic adjustment of fan speed as described in any one of the first aspects.

[0083] In a fourth aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for dynamically and automatically adjusting the fan speed as described in any one of the above-mentioned first aspects is implemented.

[0084] Beneficial effects:

[0085] The present invention provides a method for dynamically and automatically adjusting the fan speed, a system for dynamically and automatically adjusting the fan speed, an electronic device and a storage medium; through a trained prediction model, it is possible to dynamically predict the future wind speed and equipment temperature, thereby realizing advance pre-adjustment of the speed to achieve adaptive adjustment of the air volume; and by optimizing the demand formula for the air volume enthalpy difference, the goal of large air volume and small enthalpy difference is achieved, thereby improving the accuracy and practicality of automatic adjustment; achieving energy saving and precise air supply and cooling, and enhancing the adaptability of the fan in a dynamic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 A schematic diagram of a flow chart of a method for dynamically and automatically adjusting the speed of a fan provided in the first embodiment of the present disclosure;

[0087] Figure 2 A schematic diagram of the overall process of a method for dynamically and automatically adjusting the speed of a fan provided in the second embodiment of the present disclosure;

[0088] Figure 3 This is an architecture diagram of a fan speed dynamic automatic adjustment system provided in the fourth embodiment of the present disclosure;

[0089] Figure 4 This is an architecture diagram of an electronic device provided in Embodiment 5 of the present disclosure. DETAILED DESCRIPTION

[0090] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is further described in detail below in conjunction with the drawings and embodiments. It should be understood that the specific embodiments and drawings described herein are only used to explain the present invention, rather than to limit the present invention.

[0091] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence; and, in the absence of conflict, the embodiments in the present disclosure and the features in the embodiments can be arbitrarily combined with each other.

[0092] The terms used in the embodiments of the present disclosure are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure. The singular forms of "a", "said" and "the" used in the embodiments of the present disclosure and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0093] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present disclosure, and have no specific meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.

[0094] The technical solution of the present invention and how the technical solution of the present invention solves the technical problems existing in the prior art are described in detail below with specific embodiments. It is understandable that in the embodiments of the present application, the execution subject can perform some or all of the steps in the embodiments of the present application, and these steps or operations are only examples. The embodiments of the present application can also perform other operations or deformations of various operations. In addition, each step can be performed in a different order presented in the embodiments of the present application, and it is possible not to perform all the operations in the embodiments of the present application. And, the following several specific embodiments can be combined with each other, and may not be repeated in some embodiments for the same or similar concepts or processes.

[0095] Figure 1 A schematic diagram of a flow chart of a method for dynamically and automatically adjusting the fan speed provided in the first embodiment of the present disclosure, such as Figure 1 As shown, the method includes:

[0096] Step S101: Obtain wind speed, wind direction, temperature and real-time temperature data of the wind turbine surrounding environment, and input them into a trained preset prediction model to generate wind speed prediction data and prediction data of local temperature of the equipment within a period of time in the future;

[0097] Step S102: determining the expected speed and air volume of the fan according to the wind speed prediction data and the local temperature prediction data of the equipment, and obtaining the optimized pre-adjusted speed value and pre-adjusted air volume value to meet the cooling demand of the equipment with large air volume and small enthalpy difference;

[0098] Step S103: comparing the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, and calculating the difference between the two respectively;

[0099] Step S104: Determine whether the difference exceeds a preset adjustment threshold. If the difference exceeds the preset adjustment threshold, control the fan to adjust the speed and air volume to the pre-adjusted speed value and pre-adjusted air volume value, and accurately deliver the air volume to the local high temperature area of ​​the equipment to reduce the temperature; if the difference does not exceed the adjustment threshold, maintain the current fan speed and air volume.

[0100] The existing machine room fan air volume adjustment and speed control are difficult to meet the rapid heat dissipation requirements of high-temperature equipment; and the preset adjustment scheme lacks intelligence and real-time performance, and cannot accurately distribute air volume according to the dynamic changes in equipment temperature. To solve this problem, the embodiment of the present disclosure combines intelligent prediction and dynamic optimization to adjust the air-conditioning fan, achieve energy-saving and efficient precise cooling. First, a prediction model is pre-trained to predict the wind speed data and the local temperature of the equipment for a subsequent period of time based on the data of the current fan environment. The prediction model can be a linear regression or support vector machine based on a statistical model, or a neural network model, etc. The preset prediction model of the embodiment of the present disclosure is preferably a neural network model. The wind speed, wind direction, temperature and real-time temperature data of the fan environment are obtained, and input into the trained preset prediction model to generate wind speed prediction data and prediction data of the local temperature of the equipment for a future period of time. Compared with the prior art that usually uses fixed rules or simple statistical models to adjust wind speed and temperature, which cannot fully capture high-dimensional nonlinear characteristics, the embodiment of the present disclosure uses a prediction model to combine real-time input data and historical data to perform high-precision wind speed and equipment temperature prediction, which can significantly improve prediction accuracy and adjustment efficiency.

[0101] After predicting the future wind speed and local temperature of the equipment, the expected speed and air volume of the fan in the future are determined by optimizing the demand formula of air volume enthalpy difference, so as to achieve the goal of large air volume and small enthalpy difference. The optimal air volume demand of the local high-temperature area of ​​the equipment is calculated by the enthalpy difference optimization model, which can ensure the accurate distribution and efficient use of air volume.

[0102] Then, the difference between the two is calculated based on the actual condition of the current fan, including the fan speed and air volume value. Whether the difference exceeds the preset adjustment threshold is used to determine whether the current fan speed and air volume need to be adjusted. The preset adjustment threshold is set based on historical experience and actual conditions to meet the equipment cooling requirements. If the difference exceeds the adjustment threshold, the fan speed and air volume are controlled to be adjusted to the pre-adjusted speed value and pre-adjusted air volume value, and the air volume is accurately delivered to the local high-temperature area of ​​the equipment to reduce the temperature. If the difference does not exceed the adjustment threshold, the current fan speed and air volume are maintained.

[0103] The disclosed embodiment can dynamically predict future wind speed and equipment temperature through a trained prediction model, thereby realizing advance pre-adjustment of the rotation speed to achieve adaptive adjustment of the air volume; and by optimizing the demand formula for the air volume enthalpy difference, the goal of large air volume and small enthalpy difference is achieved, thereby improving the accuracy and practicality of automatic adjustment; achieving energy saving and precise air supply and cooling, and enhancing the adaptability of the fan in a dynamic environment.

[0104] Furthermore, the method further comprises:

[0105] Analyze the fluctuation pattern of real-time wind speed, extract the self-similar features in wind speed fluctuation, and identify the periodic fluctuation and random fluctuation components of wind speed;

[0106] When comparing the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, the difference between the two is calculated based on the identified periodic fluctuation and random fluctuation components of the wind speed.

[0107] Since the wind speed of the fan has a fluctuation pattern, in order to better adjust the wind speed of the fan, the disclosed embodiment also provides an accurate theoretical basis for optimizing the fan speed and air volume by separating the periodic fluctuation and random fluctuation components of the wind speed. The wind speed fluctuation characteristics are analyzed by spatial fractal algorithm, Fourier transform, wavelet analysis or other frequency domain analysis techniques, thereby separating the fluctuation characteristics, which can reduce unnecessary adjustment frequency, extend the service life of the fan, and improve the accuracy of air volume distribution.

[0108] Furthermore, the method further comprises:

[0109] Collect historical wind speed, wind direction, temperature and local temperature data of the fan's surrounding environment to build a training data set;

[0110] The training data set is used to train the preset prediction model so that the preset prediction model can predict the wind speed and local temperature changes of the equipment within a period of time in the future, thereby obtaining a trained preset prediction model.

[0111] In order to achieve accurate prediction of wind speed and equipment temperature, the embodiment of the present disclosure collects historical data of the environment around the fan, constructs a training data set, trains a preset prediction model, namely a neural network model, and captures the time dependency and nonlinear relationship between environmental data and equipment temperature; after the model training is completed, an optimized neural network prediction model is obtained, thereby combining real-time input data and historical data to perform high-precision wind speed and equipment temperature prediction, significantly improving prediction accuracy and regulation efficiency. The response speed and adjustment accuracy of the fan are improved, and the energy efficiency loss caused by high temperature of the equipment is significantly reduced.

[0112] Furthermore, the method further comprises:

[0113] The adjusted fan speed, air volume, environmental real-time data and equipment temperature data are stored together, and the database is updated to optimize the preset prediction model.

[0114] By continuously enriching the database and updating and optimizing the prediction model, the model can be enabled to perform high-precision wind speed and equipment temperature predictions, significantly improving prediction accuracy and regulation efficiency.

[0115] Furthermore,

[0116] The historical wind speed, wind direction, temperature and local temperature data of the equipment surrounding the fan are collected to construct a training data set, including:

[0117] S11. Collect historical data of the wind turbine’s surrounding environment, including wind speed V h 、wind directionθ h , Temperature T h And the local temperature of the equipment T eq :

[0118] S12. Preprocess the collected historical data, including data cleaning, denoising, and normalization, to obtain standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n ;

[0119] S13. The standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n Combined into training data set D train :

[0120] D train ={(V h,n ,θ h,n ,T h,n ,T eq,n )};

[0121] The method of using the training data set to train the preset prediction model so that the preset prediction model can predict the wind speed and local temperature changes of the equipment within a period of time in the future, and obtaining the trained preset prediction model, includes:

[0122] S21. The training data set D train Organized into input and output pairs suitable for time series forecasting:

[0123] S22. Define the neural network prediction model f θ To preset the prediction model, the neural network parameter is θ, which is used to capture the time dependence and nonlinear relationship between environmental data and device temperature;

[0124] S23. Using the loss function L(θ) to train the neural network prediction model;

[0125] S24. Update the neural network parameters θ by iteratively minimizing the loss function L(θ), and continue iterative training until the loss function L(θ) converges to the preset threshold ∈ L , and obtain the optimized neural network prediction model as the trained preset prediction model.

[0126] Wind speed V h Indicates the change in wind speed within the historical time interval; wind direction θ h Indicates the direction data of wind direction in the historical time interval; temperature T h Indicates the ambient temperature within the historical time interval; the local temperature of the equipment T eq Represents the temperature change data of the fan equipment in the local area; the collected historical data is preprocessed, and the standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n ; Combined into training data set: D train ={(V h,n ,θ h,n ,T h,n ,T eq,n )}.

[0127] After the training data set is constructed, the neural network prediction model is trained using the training data set, including:

[0128] S21. The training data set D train Organized into input and output pairs suitable for time series forecasting:

[0129] Input sequence Represents the time step it l The historical data of standardized wind speed, wind direction, ambient temperature and local temperature of the equipment from i to i, where t l is the lookback window length;

[0130] Target Output represents the normalized wind speed and local temperature of the equipment at the future time step i+Δt, where Δt is the prediction time interval;

[0131] S22. Define the neural network prediction model f θ To preset the prediction model, the neural network parameter is θ, which is used to capture the time dependence and nonlinear relationship between environmental data and equipment temperature. The neural network prediction model structure includes:

[0132] Input layer, accepts input sequence X (i) ;

[0133] Multiple hidden layers, using long short-term memory units to process time series data;

[0134] Output layer, generating predicted future wind speeds and local temperature of the equipment satisfy:

[0135]

[0136] S23. Use the loss function L(θ) to train the neural network prediction model. The loss function is defined as:

[0137]

[0138] Where N is the number of training samples, and is the true normalized wind speed and local temperature of the equipment at time step i+Δt, α2 and β2 are weight coefficients, γ2 is the regularization coefficient, K is the total number of model parameters, θ k is the kth model parameter;

[0139] S24. Update the neural network parameters θ by iteratively minimizing the loss function L(θ):

[0140]

[0141] Among them, θ (t) is the parameter of the tth iteration, η is the learning rate, is the gradient of the loss function with respect to the parameter θ, and the training is continued iteratively until the loss function L(θ) converges to the preset threshold ∈ L , and obtain the optimized neural network prediction model as the trained preset prediction model.

[0142] Preset threshold ∈ L It can be determined by analyzing the characteristics of the training data (noise level and number of samples) and the prediction accuracy requirements of the model. For example, when using mean square error as the loss function, if the data has been standardized, the threshold range can be set to 0.01, etc.

[0143] Furthermore, the input to the trained preset prediction model to generate wind speed prediction data and prediction data of the local temperature of the equipment in the future period includes:

[0144] The neural network prediction model is used to process the input real-time wind speed, wind direction and temperature data to generate wind speed prediction data for a period of time in the future and prediction data of the local temperature of the equipment, including:

[0145] S41. Obtain real-time input data sequence around the wind turbine:

[0146]

[0147] in, represents the real-time wind speed at time step i, represents the real-time wind direction at time step i, represents the real-time ambient temperature at time step i, represents the real-time local temperature of the device at time step i;

[0148] S42. Input real-time data sequence Input to the optimized neural network prediction model f θ , the wind speed forecast data for future time steps is generated through the nonlinear mapping function of the neural network prediction model And the prediction data of the local temperature of the equipment

[0149] Further, the expected speed and air volume of the fan are determined according to the wind speed prediction data and the equipment local temperature prediction data to obtain the optimized pre-adjusted speed value and pre-adjusted air volume value, including:

[0150] S51: Receive wind speed forecast data And equipment local temperature prediction data Combined with real-time ambient temperature and real-time local device temperature Construct fan speed and air volume optimization function:

[0151]

[0152] Where n is the fan speed, Q is the fan air volume, and Represent the predicted local temperature of the device and the real-time local temperature of the device, respectively. and Represent the predicted wind speed and real-time wind speed respectively. Where Δh represents the enthalpy difference, H(Q) represents the air volume per unit enthalpy difference, α1, β1 and γ1 are the speed adjustment weight coefficient, air volume adjustment weight coefficient and enthalpy difference air volume optimization weight coefficient respectively;

[0153] S52: Solve the optimization function F opt (n,Q), determine the pre-adjusted speed n that satisfies the minimization target opt and air volume Q opt :

[0154]

[0155] S53: Calculate the pre-adjusted air volume value Q that meets the requirements of large air volume and small enthalpy difference through the enthalpy difference optimization formula opt :

[0156]

[0157] in, represents the difference in local temperature of the equipment, and R represents the ambient heat transfer constant.

[0158] Enthalpy difference refers to the magnitude of the change in the enthalpy value of a substance during a chemical reaction or physical process. The size of the enthalpy difference can reflect the thermal effect of a reaction or process. A small enthalpy difference means that the difference in unit energy before and after the air is treated is small, that is, the proportion of sensible heat in the air heat is high. This situation is called a high sensible heat ratio. A high sensible heat ratio means that when the fan system processes the air, it mainly achieves temperature regulation through sensible heat exchange rather than through changes in humidity. Reactions or processes with small enthalpy differences have weak exothermic or endothermic properties; the optimal air volume requirements of local high-temperature areas of the equipment are calculated through the enthalpy difference optimization model to ensure accurate distribution and efficient utilization of the air volume. Precise air supply significantly improves the cooling efficiency of the equipment and reduces overall energy consumption.

[0159] S52 optimization function F opt (n,Q) calculates the speed n when the overall objective function is minimized opt and air volume Q opt ,The purpose is to find the wind turbine operating parameters suitable for the current scenario from the perspective of global optimization, taking ,into account of various trade-off factors.

[0160] The S53 formula is to recalculate Q opt In order to meet specific conditions (meet the needs of large air volume and small enthalpy difference), based on the global optimization of S52, an independent thermodynamic constraint or local optimization strategy is used to further correct and confirm Q opt The specific value of .

[0161] Introducing temperature difference ΔT into the calculation of S53 eq The heat transfer constant R is due to the need to further clarify the relationship between air volume and local temperature regulation due to thermodynamic requirements. Unlike the global optimization of S52, S53 focuses more on specific physical requirements rather than optimizing overall system performance. The air volume calculation of S53 is a thermodynamic supplement to the air volume in S52, ensuring the final air volume Q opt It not only meets the goal of global optimization of the system, but also effectively meets the local heat dissipation needs of the equipment. eq The specific values ​​of and R are dynamic and depend on the current operating conditions.

[0162] Furthermore,

[0163] The analysis of the fluctuation law of real-time wind speed, extraction of self-similar features in wind speed fluctuation, and identification of periodic fluctuation and random fluctuation components of wind speed include:

[0164] S61: Get real-time wind speed data Time series:

[0165]

[0166] in, represents the real-time wind speed at time step t, t w is the analysis window length;

[0167] S62: Construct a fractal dimension model of real-time wind speed based on spatial fractal algorithm and calculate the fractal dimension D of time series f :

[0168]

[0169] Among them, ∈ is the segmentation scale, N(∈) represents the minimum number of segments covering the wind speed time series at scale ∈;

[0170] S63: Through fractal dimension D f Extract the self-similar features of wind speed time series. If D f ≈1.5, it indicates that the wind speed change has a significant random fluctuation component; if 1.0 <D f <1.5, it indicates that the wind speed change has periodic fluctuation characteristics;

[0171] S64: The wind speed time series is decomposed using the characteristic separation model of periodic fluctuations and random fluctuations. series It is expressed as:

[0172] V series =V periodic +V random ;

[0173] Among them, V periodic Represents the periodic fluctuation component of wind speed, V random Represents the random fluctuation component of wind speed, V series The remainder after deducting the periodic fluctuations;

[0174] The method of comparing the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, respectively calculating the difference between the two based on the identified periodic fluctuation and random fluctuation components of the wind speed, includes:

[0175] S71: Obtaining the periodic fluctuation component V of the wind speed periodic and the random fluctuation component V random , combined with the current fan speed n current and air volume Q current , calculate the predicted periodic speed n periodic and randomly adjust the speed n random :

[0176]

[0177] Among them, n opt is the pre-adjusted speed value, is the predicted total wind speed, n periodic Indicates the speed adjustment to periodic fluctuations, n random Indicates the compensation of random fluctuations in speed;

[0178] S72: Compare the current fan speed n current and pre-adjusted speed n opt , calculate the total speed difference Δn and the total air volume difference ΔQ:

[0179] Δn=|n current -(n periodic +n random )|;

[0180] ΔQ=|Q current -Q opt |;

[0181] Among them, Δn represents the difference between the current speed and the total adjusted speed, ΔQ represents the difference between the current air volume and the pre-adjusted air volume, Q current is the current air volume, Q opt To pre-adjust the air volume value.

[0182] Fan speed (n current ) and air volume (Q current ) is the actual measured parameter, which is usually collected in real time through the fan's own sensors and monitoring system. Specifically, the fan's speed is measured by a speed sensor, and the air volume is calculated by a flow sensor.

[0183] S101 obtains real-time wind speed and direction data of the external environment, which provide basic input for the subsequent neural network prediction model to predict future wind speed and local temperature changes of the equipment. current and Q current It reflects the current operating status of the fan and will be used together with the predicted value as a reference input for dynamic adjustment of the fan.

[0184] Traditional methods do not conduct in-depth analysis of the periodic and random fluctuation characteristics of wind speed, and the adjustment strategy is simple and rough. The present invention separates the periodic fluctuation and random fluctuation components of wind speed through a spatial fractal algorithm to provide an accurate theoretical basis for the optimal adjustment of fan speed and air volume; separating the fluctuation characteristics can reduce unnecessary adjustment frequency, extend the service life of the fan, and improve the accuracy of air volume distribution.

[0185] Further, the determination of whether the difference exceeds a preset adjustment threshold value, if the difference exceeds the preset adjustment threshold value, controlling the fan to adjust the speed and air volume to the pre-adjusted speed value and pre-adjusted air volume value, and accurately delivering the air volume to the local high temperature area of ​​the equipment to reduce the temperature; if the difference does not exceed the adjustment threshold value, maintaining the current fan speed and air volume, including:

[0186] S81: Determine whether the speed difference Δn and the air volume difference ΔQ exceed the preset adjustment threshold ∈ n and ∈ Q , the judgment conditions are:

[0187]

[0188] Among them, Δn2 and ΔQ2 are the root mean square difference between the current speed and the total adjusted speed, and the integral difference between the current air volume and the pre-adjusted air volume, respectively. w is the observation time window, n periodic (k) and n random (k) are the periodic and random speed adjustment components, ∈ n and ∈ Q They are the adjustment thresholds for the rotation speed and air volume respectively;

[0189] S82: If the condition is met, adjust the fan speed n in real time adjust and air volume Q adjust :

[0190]

[0191] Among them, η n and η Q is the speed and air volume adjustment rate, σ T Adjust the sensitivity parameter for the temperature difference, f t Adjust frequency for air volume fluctuations;

[0192] S83: Adjust the air volume Q adjust Accurately guide the heat to the local high temperature area of ​​the equipment, through the heat transfer rate R heat To verify:

[0193]

[0194] Where ρ is the air density, c p is the specific heat capacity of air, L, W and H are the length, width and height of the local area of ​​the equipment, Indicates the actual equivalent temperature of the local high temperature area of ​​the equipment, T target Indicates the target temperature;

[0195] S84: If Δn≤∈ n And ΔQ≤∈ Q , the current fan status remains unchanged, and the status data S is recorded current :

[0196]

[0197] And stored in database D state :

[0198] D state =D state ∪S current .

[0199] Indicates the actual equivalent temperature of a local high temperature area of ​​the equipment. The superscript (i) indicates that the temperature value corresponds to a specific area and the temperature is calculated by region. The definition of equivalent temperature is based on the thermodynamic state of the equipment. T target Indicates the target temperature, that is, the set temperature that the local high temperature area is expected to eventually reach. The relationship between the two is: Indicates the difference between the actual temperature of the equipment and the target temperature. It is used to quantify the temperature deviation in the local area and guide the adjustment of the air volume Q adjust The difference is balanced by heat transfer.

[0200] ∈ n (Speed ​​regulation threshold) setting: ∈ n Indicates the RMS difference threshold between the current fan speed and the total regulated speed. It can be set according to the system's sensitivity to speed adjustment and the requirements for operating stability. Setting basis: Allowable speed fluctuation range: For example, the fan may allow a speed fluctuation range of 1%-5% of the total speed. System tolerance to load or air volume fluctuations: A lower threshold can be selected to ensure regulation accuracy when the load is high, and the threshold can be appropriately increased in light load scenarios.

[0201] Example: If the typical speed of the fan is 1500RPM and the allowable fluctuation range is 2%, then ∈ n =0.02×1500=30RPM.

[0202] ∈ Q (Air volume adjustment threshold) setting: ∈ Q Indicates the integral difference threshold between the current air volume and the pre-adjusted air volume, reflecting the adjustment accuracy requirement of the actual fan output air volume and the expected air volume. Basis: Fan output air volume requirement: For example, for cooling equipment, the required air volume adjustment needs to ensure that the temperature can be controlled within a small range. Dynamics of air volume adjustment: If the air volume change has a greater impact on system performance, a smaller threshold is selected.

[0203] Example: If the target air volume of the fan is 5000m 3 / h, the allowable error is 5%, then ∈ Q =0.05×5000=250m 3 / h.

[0204] Each periodic and random speed adjustment component is accumulated in the observation time window, and the speed difference Δn and air volume difference ΔQ are judged whether they exceed the preset adjustment threshold ∈ by the root mean square difference between the current speed and the total adjustment speed, and the integral difference between the current air volume and the pre-adjusted air volume. n and ∈ Q , can make the judgment more accurate, and adjust the fan speed n in real time adjust and air volume Q adjust The formula can improve the adjustment accuracy, thereby reducing energy efficiency loss while meeting the equipment cooling requirements as much as possible, and through the heat transfer rate R heat Verification can better ensure the effectiveness of adjustments and achieve energy saving and precise air supply and cooling.

[0205] The disclosed embodiment uses an optimized neural network prediction model, combined with real-time input data and historical data, to perform high-precision wind speed and equipment temperature prediction, significantly improving prediction accuracy and regulation efficiency. Accurate prediction directly improves the response speed and adjustment accuracy of the fan, thereby significantly reducing the energy efficiency loss caused by high temperature of the equipment. The separation of the periodic fluctuation and random fluctuation components of the wind speed through the spatial fractal algorithm provides an accurate theoretical basis for the optimization and regulation of the fan speed and air volume, which can reduce unnecessary adjustment frequency, extend the service life of the fan, and improve the accuracy of air volume distribution. The optimal air volume requirement of the local high temperature area of ​​the equipment is calculated through the enthalpy difference optimization model to ensure accurate distribution and efficient use of the air volume. Accurate air supply significantly improves the cooling efficiency of the equipment and reduces overall energy consumption.

[0206] like Figure 2 As shown, the second embodiment of the present disclosure also provides a method for dynamically and automatically adjusting the speed of a fan, comprising the following steps:

[0207] S1. Collect historical wind speed, wind direction, temperature and local temperature data of the fan surrounding environment to build a training data set;

[0208] S2. Using the training data set to train the neural network prediction model, so that the neural network prediction model can predict the wind speed and local temperature changes of the equipment in the future, and obtain an optimized neural network prediction model;

[0209] S3. Obtain wind speed, wind direction, temperature and real-time temperature data of the fan surrounding environment, and input the real-time temperature data into the trained neural network prediction model;

[0210] S4. Processing the input real-time temperature data using the neural network prediction model to generate wind speed prediction data for a period of time in the future and prediction data of the local temperature of the equipment;

[0211] S5. Determine the expected speed and air volume of the fan according to the wind speed prediction data and the local temperature prediction data of the equipment, obtain the optimized pre-adjusted speed value and pre-adjusted air volume value, and meet the cooling requirements of the equipment with large air volume and small enthalpy difference;

[0212] S6. Analyze the fluctuation law of real-time wind speed based on spatial fractal algorithm, extract the self-similar features in wind speed fluctuation, and identify the periodic fluctuation and random fluctuation components of wind speed;

[0213] S7. After identifying the periodic fluctuation and random fluctuation components of the wind speed, compare the pre-adjusted speed value, the pre-adjusted air volume value with the current fan speed and air volume, calculate the difference between the two, and determine whether the difference exceeds the preset adjustment threshold;

[0214] S8. If the difference exceeds the adjustment threshold, the fan speed and air volume are controlled to the pre-adjusted speed value and pre-adjusted air volume value, and the air volume is accurately delivered to the local high temperature area of ​​the equipment to reduce the temperature; if the difference does not exceed the adjustment threshold, the current fan speed and air volume are maintained;

[0215] S9. Store the adjusted fan speed, air volume, environmental real-time data and equipment temperature data together, update the database, and optimize the neural network prediction model.

[0216] In this implementation, the S1 includes:

[0217] S11. Collect historical data of the wind turbine’s surrounding environment, including wind speed V h 、wind directionθ h , Temperature T h And the local temperature of the equipment T eq :

[0218] Wind speed V h Indicates the change in wind speed within the historical time interval;

[0219] Wind direction θ h Direction data indicating wind direction within a historical time interval;

[0220] Temperature T h Indicates the ambient temperature in the historical time interval;

[0221] Equipment local temperature T eq Indicates the temperature change data of the fan equipment in the local area;

[0222] S12. Preprocess the collected historical data by data cleaning, denoising and normalization to obtain standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n ;

[0223] S13. The standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n Combined into a training data set:

[0224] D train ={(V h,n ,θ h,n ,T h,n ,T eq,n )};

[0225] In this implementation, S2 includes:

[0226] S21. The training data set D train Organized into input and output pairs suitable for time series forecasting:

[0227] Input sequence Represents the time step it l The historical data of standardized wind speed, wind direction, ambient temperature and local temperature of the equipment from i to i, where t l is the lookback window length;

[0228] Target Output represents the normalized wind speed and local temperature of the equipment at the future time step i+Δt, where Δt is the prediction time interval;

[0229] S22. Define the neural network prediction model f θ , the neural network parameter is θ, which is used to capture the time dependence and nonlinear relationship between environmental data and device temperature. The neural network prediction model structure includes:

[0230] Input layer, accepts input sequence X (i) ;

[0231] Several hidden layers, using long short-term memory units to process time series data;

[0232] Output layer, generating predicted future wind speeds and local temperature of the equipment satisfy:

[0233]

[0234] S23. Use the loss function L(θ) to train the neural network prediction model. The loss function is defined as:

[0235]

[0236] Where N is the number of training samples, and is the true normalized wind speed and local temperature of the equipment at time step i+Δt, α2 and β2 are weight coefficients, γ2 is the regularization coefficient, K is the total number of model parameters, θ k is the kth model parameter;

[0237] S24. Update the neural network parameters θ by iteratively minimizing the loss function L(θ):

[0238]

[0239] Among them, θ (t) is the parameter of the tth iteration, η is the learning rate, is the gradient of the loss function with respect to the parameter θ, and the training is continued iteratively until the loss function L(θ) converges to the preset threshold ∈ L , and obtain the optimized neural network prediction model.

[0240] In this implementation, S4 includes:

[0241] S41. Obtain real-time input data sequence around the wind turbine:

[0242]

[0243] in, represents the real-time wind speed at time step i, represents the real-time wind direction at time step i, represents the real-time ambient temperature at time step i, represents the real-time local temperature of the device at time step i;

[0244] S42. Input real-time data sequence Input to the optimized neural network prediction model f θ , the wind speed forecast data for future time steps is generated through the nonlinear mapping function of the neural network prediction model And the prediction data of the local temperature of the equipment

[0245] In this implementation, S5 includes:

[0246] S51. Receive wind speed forecast data And equipment local temperature prediction data Combined with real-time ambient temperature and real-time local device temperature Construct fan speed and air volume optimization function:

[0247]

[0248] Where n is the fan speed, Q is the fan air volume, and Represent the predicted local temperature of the device and the real-time local temperature of the device, respectively. and Represent the predicted wind speed and real-time wind speed respectively. Where Δh represents the enthalpy difference, H(Q) represents the air volume per unit enthalpy difference, α1, β1 and γ1 are the speed adjustment weight coefficient, air volume adjustment weight coefficient and enthalpy difference air volume optimization weight coefficient respectively;

[0249] S52. Solve the optimization function F opt (n,Q), determine the pre-adjusted speed n that satisfies the minimization target opt and air volume Q opt :

[0250]

[0251] S53. Calculate the pre-adjusted air volume value Q that meets the requirements of large air volume and small enthalpy difference through the enthalpy difference optimization formula opt :

[0252]

[0253] in, represents the difference in local temperature of the equipment, and R represents the ambient heat transfer constant.

[0254] In this implementation, S6 includes:

[0255] S61. Get real-time wind speed data Time series:

[0256]

[0257] in, represents the real-time wind speed at time step t, t w is the analysis window length;

[0258] S62. Construct a fractal dimension model of real-time wind speed based on spatial fractal algorithm and calculate the fractal dimension D of time series f :

[0259]

[0260] Among them, ∈ is the segmentation scale, N(∈) represents the minimum number of segments covering the wind speed time series at scale ∈;

[0261] S63. Through fractal dimension D f Extract the self-similar features of wind speed time series. If D f ≈1.5, it indicates that the wind speed change has a significant random fluctuation component; if 1.0 <D f<1.5, it indicates that the wind speed change has periodic fluctuation characteristics;

[0262] S64. The wind speed time series is decomposed using the feature separation model of periodic fluctuations and random fluctuations. series It is expressed as:

[0263] V series =V periodic +V random ;

[0264] Among them, V periodic Represents the periodic fluctuation component of wind speed, V random Represents the random fluctuation component of wind speed, V series The remainder after subtracting periodic fluctuations.

[0265] In this implementation, S7 includes:

[0266] S71. Obtain the periodic fluctuation component V of the wind speed periodic and the random fluctuation component V random , combined with the current fan speed n current and air volume Q current , calculate the predicted periodic speed n periodic and randomly adjust the speed n random :

[0267]

[0268] Among them, n opt is the pre-adjusted speed value, is the predicted total wind speed, n periodic Indicates the speed adjustment to periodic fluctuations, n random Indicates the compensation of random fluctuations in speed;

[0269] S72. Compare the current fan speed n current and pre-adjusted speed n opt , calculate the total speed difference Δn and the total air volume difference ΔQ:

[0270] Δn=|n current -(n periodic +n random )|;

[0271] ΔQ=|Q current -Q opt |;

[0272] Among them, Δn represents the difference between the current speed and the total adjusted speed, ΔQ represents the difference between the current air volume and the pre-adjusted air volume, Q current is the current air volume, Q opt To pre-adjust the air volume value.

[0273] In this implementation, S8 includes:

[0274] S81. Determine whether the speed difference Δn and the air volume difference ΔQ exceed the preset adjustment threshold ∈ n and ∈ Q , the judgment conditions are:

[0275]

[0276] Among them, Δn2 and ΔQ2 are the root mean square difference between the current speed and the total adjusted speed, and the integral difference between the current air volume and the pre-adjusted air volume, respectively. w is the observation time window, n periodic (k) and n random (k) are the periodic and random speed adjustment components, ∈ n and ∈ Q They are the adjustment thresholds for the rotation speed and air volume respectively;

[0277] S82. If the adjustment condition is met, adjust the fan speed n in real time adjust and air volume Q adjust :

[0278]

[0279] Among them, η n and η Q is the speed and air volume adjustment rate, σ T Adjust the sensitivity parameter for the temperature difference, f t Adjust frequency for air volume fluctuations;

[0280] S83. Adjust the air volume Q adjust Accurately guide the heat to the local high temperature area of ​​the equipment, through the heat transfer rate R heat To verify:

[0281]

[0282] Where ρ is the air density, c p is the specific heat capacity of air, L, W and H are the length, width and height of the local area of ​​the equipment, Indicates the actual equivalent temperature of the local high temperature area of ​​the equipment, T target Indicates the target temperature;

[0283] S84. If Δn≤∈ n And ΔQ≤∈ Q , the current fan status remains unchanged, and the status data S is recorded current :

[0284]

[0285] And stored in database D state :

[0286] D state =D state ∪S current .

[0287] The disclosed embodiment can dynamically predict future wind speed and equipment temperature changes through time series analysis of a pre-trained neural network model based on historical data sets, so as to achieve adaptive adjustment of air volume and enhance the adaptability of the fan in a dynamic environment; and achieve the goal of large air volume and small enthalpy difference by optimizing the demand formula of air volume enthalpy difference. In addition, the disclosed embodiment realizes the separation of wind speed components through fractal algorithm when processing wind speed fluctuations, and then makes more detailed speed adjustment. This real-time prediction and fine adjustment function improves the accuracy and practicality of automatic adjustment.

[0288] The third embodiment of the present disclosure also provides a method for dynamically and automatically adjusting the fan speed. In a large data center computer room, the heat generated by the operation of the equipment is extremely high, especially in summer, the local temperature of some high-performance servers will quickly rise to above 70°C. If the temperature cannot be effectively reduced in time, the equipment performance will be seriously damaged, and even the system will be paralyzed. The traditional air conditioning system uses a constant wind speed or a rough zoned air supply method for cooling, which leads to large differences in the local temperature of the equipment, serious waste of air volume, and low overall energy efficiency.

[0289] This embodiment optimizes the air supply of the air conditioning system by applying a fan dynamic adjustment method based on neural network prediction and spatial fractal algorithm, thereby achieving energy saving and precise air supply and cooling.

[0290] A dynamically adjustable fan is installed in the computer room of the data center. The fan is connected to the temperature monitoring system and the central air-conditioning control system. The monitoring system collects the following data every minute: surface temperature of the server equipment, ambient temperature of the computer room, wind speed and air volume inside the equipment, and transmits it to the fan control module in real time for analysis and adjustment.

[0291] Example On a summer afternoon (August 3, 2024, 14:00-15:00), the fan system conducted a 1-hour precise air supply and cooling test on the high-temperature equipment in the computer room. The specific process is as follows:

[0292] The data center temperature monitoring system recorded the surface temperature of 10 high-performance servers in the computer room, ranging from [65°C to 72°C], and the room ambient temperature was 35°C. The wind speed data collected in real time was [0.8m / s, 1.2m / s], and the air volume data was 200m / min.

[0293] Using the neural network prediction model trained with historical data, the fan control system predicts the temperature change trend of the equipment in the next 10 minutes as follows:

[0294] Equipment maximum temperature: [72.0, 73.5, 75.2, 76.1, 77.0]°C;

[0295] Average device temperature: [67.5, 68.8, 70.2, 71.0, 72.0]°C;

[0296] Through the forecast data, it is preliminarily determined that the fan speed needs to be adjusted from the current 600rpm to 850rpm, and the air volume needs to be increased to 350m / min to meet the cooling requirements.

[0297] The spatial fractal algorithm is used to analyze the real-time wind speed data and extract the periodic and random components. The results show that:

[0298] Periodic fluctuation: [0.9m / s, 1.1m / s, 1.0m / s, 1.2m / s];

[0299] Random fluctuation: [-0.05m / s, +0.03m / s, -0.02m / s, +0.04m / s];

[0300] The fan system separates and compensates the wind speed according to the fluctuation characteristics, and finally determines the adjusted speed to be 870rpm, and dynamically adjusts the air volume to 360m3 / min in real time.

[0301] According to the surface temperature distribution of the equipment, the fan sends most of the air volume to the server area with the highest temperature. The target air volume is calculated to be 400m

[0302] / min, ultimately achieving optimized cooling rate. The air supply direction is precisely controlled to the surface of the equipment through the air guide system to avoid ineffective air volume dispersion.

[0303] To verify the effect, a comparison was made with the traditional fixed speed air supply method. The core data during the test are as follows:

[0304]

[0305] Description of training samples:

[0306] 1. Sample size: 500,000 groups, covering operating data in different seasons.

[0307] 2. Wind speed range: [0.5m / s, 3.0m / s]; temperature range: [30℃, 80℃].

[0308] 3. Equipment characteristics: Contains heat distribution data of various server models, covering high-temperature and medium-temperature equipment operation scenarios.

[0309] The present disclosure successfully achieved efficient cooling of the equipment in this test scenario, with the surface temperature of the equipment dropping from 72°C to below 68°C, and the cooling time shortened by 35%. At the same time, the precise air supply strategy effectively avoided the waste of air volume, improved the overall energy efficiency by about 18%, and significantly reduced the system energy consumption, proving the energy-saving and precise cooling effect of the present disclosure in the application of air conditioning systems.

[0310] The fourth embodiment of the present disclosure also provides a fan speed dynamic automatic adjustment system, such as Figure 3 As shown, the system comprises:

[0311] The prediction module 11 is configured to obtain the wind speed, wind direction, temperature and real-time temperature data of the wind turbine surrounding environment and input them into the trained preset prediction model to generate wind speed prediction data and prediction data of the local temperature of the equipment in the future;

[0312] A determination module 12 is configured to determine the expected speed and air volume of the fan according to the wind speed prediction data and the equipment local temperature prediction data, and obtain an optimized pre-adjusted speed value and pre-adjusted air volume value to meet the cooling demand of the equipment with large air volume and small enthalpy difference;

[0313] A difference calculation module 13 is configured to compare the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, and calculate the difference between the two respectively;

[0314] The judgment and adjustment module 14 is configured to judge whether the difference exceeds a preset adjustment threshold. If the difference exceeds the preset adjustment threshold, the fan is controlled to adjust the speed and air volume to the pre-adjusted speed value and pre-adjusted air volume value, and the air volume is accurately delivered to the local high-temperature area of ​​the equipment to reduce the temperature; if the difference does not exceed the adjustment threshold, the current fan speed and air volume are maintained.

[0315] Furthermore, the system also includes an identification module 15;

[0316] The identification module 15 is configured to analyze the fluctuation pattern of the real-time wind speed, extract the self-similar features in the wind speed fluctuation, and identify the periodic fluctuation and random fluctuation components of the wind speed;

[0317] The difference calculation module 13 is specifically configured to calculate the difference between the pre-adjusted speed value and the current fan speed, and the pre-adjusted air volume value and the current air volume based on the identified periodic fluctuation and random fluctuation components of the wind speed.

[0318] Furthermore, the system also includes a model training module 16;

[0319] The model training module 16 is configured to collect historical wind speed, wind direction, temperature and local temperature data of the wind turbine surrounding environment to construct a training data set; and

[0320] The training data set is used to train the preset prediction model so that the preset prediction model can predict the wind speed and local temperature changes of the equipment within a period of time in the future, thereby obtaining a trained preset prediction model.

[0321] Furthermore, the system also includes an updating module 17;

[0322] The updating module 17 is configured to store the adjusted fan speed, air volume, environmental real-time data and equipment temperature data together, and update the database to optimize the preset prediction model.

[0323] Furthermore,

[0324] The historical wind speed, wind direction, temperature and local temperature data of the equipment surrounding the fan are collected to construct a training data set, including:

[0325] S11. Collect historical data of the wind turbine’s surrounding environment, including wind speed V h 、wind directionθ h , Temperature T h And the local temperature of the equipment T eq :

[0326] S12. Preprocess the collected historical data, including data cleaning, denoising, and normalization, to obtain standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n ;

[0327] S13. The standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n Combined into training data set D train :

[0328] D train ={(V h,n ,θ h,n ,T h,n ,T eq,n )};

[0329] The method of using the training data set to train the preset prediction model so that the preset prediction model can predict the wind speed and local temperature changes of the equipment within a period of time in the future, and obtaining the trained preset prediction model, includes:

[0330] S21. The training data set D train Organized into input and output pairs suitable for time series forecasting:

[0331] S22. Define the neural network prediction model f θ To preset the prediction model, the neural network parameter is θ, which is used to capture the time dependence and nonlinear relationship between environmental data and device temperature;

[0332] S23. Using the loss function L(θ) to train the neural network prediction model;

[0333] S24. Update the neural network parameters θ by iteratively minimizing the loss function L(θ), and continue iterative training until the loss function L(θ) converges to the preset threshold ∈ L , and obtain the optimized neural network prediction model as the trained preset prediction model.

[0334] Further, the expected speed and air volume of the fan are determined according to the wind speed prediction data and the equipment local temperature prediction data to obtain the optimized pre-adjusted speed value and pre-adjusted air volume value, including:

[0335] S51: Receive wind speed forecast data And equipment local temperature prediction data Combined with real-time ambient temperature and real-time local device temperature Construct fan speed and air volume optimization function:

[0336]

[0337] Where n is the fan speed, Q is the fan air volume, and Represent the predicted local temperature of the device and the real-time local temperature of the device, respectively. and Represent the predicted wind speed and real-time wind speed respectively. Where Δh represents the enthalpy difference, H(Q) represents the air volume per unit enthalpy difference, α1, β1 and γ1 are the speed adjustment weight coefficient, air volume adjustment weight coefficient and enthalpy difference air volume optimization weight coefficient respectively;

[0338] S52: Solve the optimization function F opt (n,Q), determine the pre-adjusted speed n that satisfies the minimization target opt and air volume Q opt :

[0339]

[0340] S53: Calculate the pre-adjusted air volume value Q that meets the requirements of large air volume and small enthalpy difference through the enthalpy difference optimization formula opt:

[0341]

[0342] in, represents the difference in local temperature of the equipment, and R represents the ambient heat transfer constant.

[0343] Furthermore,

[0344] The identification module 15 is specifically configured to identify the periodic fluctuation and random fluctuation components of the wind speed through the following steps, including:

[0345] S61: Get real-time wind speed data Time series:

[0346]

[0347] in, represents the real-time wind speed at time step t, t w is the analysis window length;

[0348] S62: Construct a fractal dimension model of real-time wind speed based on spatial fractal algorithm and calculate the fractal dimension D of time series f :

[0349]

[0350] Among them, ∈ is the segmentation scale, N(∈) represents the minimum number of segments covering the wind speed time series at scale ∈;

[0351] S63: Through fractal dimension D f Extract the self-similar features of wind speed time series. If D f ≈1.5, it indicates that the wind speed change has a significant random fluctuation component; if 1.0 <D f <1.5, it indicates that the wind speed change has periodic fluctuation characteristics;

[0352] S64: The wind speed time series is decomposed using the characteristic separation model of periodic fluctuations and random fluctuations. series It is expressed as:

[0353] V series =V periodic +V random ;

[0354] Among them, V periodic Represents the periodic fluctuation component of wind speed, V random Represents the random fluctuation component of wind speed, V series The remainder after deducting the periodic fluctuations;

[0355] The difference calculation module 13 is specifically configured to calculate the difference between the pre-adjusted speed value and the current fan speed, and the pre-adjusted air volume value and the current air volume respectively when comparing the two in the following manner, including:

[0356] S71: Obtaining the periodic fluctuation component V of the wind speed periodic and the random fluctuation component V random , combined with the current fan speed n current and air volume Q current , calculate the predicted periodic speed n periodic and randomly adjust the speed n random :

[0357]

[0358] Among them, n opt is the pre-adjusted speed value, is the predicted total wind speed, n periodic Indicates the speed adjustment to periodic fluctuations, n random Indicates the compensation of random fluctuations in speed;

[0359] S72: Compare the current fan speed n current and pre-adjusted speed n opt , calculate the total speed difference Δn and the total air volume difference ΔQ:

[0360] Δn=|n current -(n periodic +n random )|

[0361] ΔQ=|Q current -Q opt |;

[0362] Among them, Δn represents the difference between the current speed and the total adjusted speed, ΔQ represents the difference between the current air volume and the pre-adjusted air volume, Q current is the current air volume, Q opt To pre-adjust the air volume value.

[0363] Furthermore, the judgment and adjustment module 14 is specifically configured as follows:

[0364] S81: Determine whether the speed difference Δn and the air volume difference ΔQ exceed the preset adjustment threshold ∈ n and ∈ Q , the judgment conditions are:

[0365]

[0366] Among them, Δn2 and ΔQ2 are the root mean square difference between the current speed and the total adjusted speed, and the integral difference between the current air volume and the pre-adjusted air volume, respectively.w is the observation time window, n periodic (k) and n random (k) are the periodic and random speed adjustment components, ∈ n and ∈ Q They are the adjustment thresholds for the rotation speed and air volume respectively;

[0367] S82: If the condition is met, adjust the fan speed n in real time adjust and air volume Q adjust :

[0368]

[0369] Among them, η n and η Q is the speed and air volume adjustment rate, σ T Adjust the sensitivity parameter for the temperature difference, f t Adjust frequency for air volume fluctuations;

[0370] S83: Adjust the air volume Q adjust Accurately guide the heat to the local high temperature area of ​​the equipment, through the heat transfer rate R heat To verify:

[0371]

[0372] Where ρ is the air density, c p is the specific heat capacity of air, L, W and H are the length, width and height of the local area of ​​the equipment;

[0373] S84: If Δn≤∈ n And ΔQ≤∈ Q , the current fan status remains unchanged, and the status data S is recorded current :

[0374]

[0375] And stored in database D state :

[0376] D state =D state ∪S current .

[0377] The fan speed dynamic automatic regulation system of the disclosed embodiment is used to implement the fan speed dynamic automatic regulation method in method embodiment 1 and embodiment 2, so the description is relatively simple. For details, please refer to the relevant description in the previous method embodiment, which will not be repeated here.

[0378] In addition, if Figure 4As shown, the fifth embodiment of the present disclosure further provides an electronic device, including a memory 100 and a processor 200, wherein the memory 100 stores a computer program, and when the processor 200 runs the computer program stored in the memory 100, the processor 200 executes the above-mentioned various possible methods.

[0379] The memory 100 is connected to the processor 200. The memory 100 may be a flash memory, a read-only memory or other memory, and the processor 200 may be a central processing unit or a single-chip microcomputer.

[0380] In addition, an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to perform the above-mentioned various possible methods.

[0381] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program modules or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), Digital Video Disc (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0382] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and substance of the present disclosure, and these modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. A method for dynamically and automatically adjusting the speed of a fan, characterized in that: The method comprises: Obtain wind speed, wind direction, temperature and real-time temperature data of the fan's surrounding environment, and input them into the trained preset prediction model to generate wind speed prediction data and prediction data of the local temperature of the equipment for a period of time in the future; Determine the expected speed and air volume of the fan according to the wind speed prediction data and the local temperature prediction data of the equipment, and obtain the optimized pre-adjusted speed value and pre-adjusted air volume value to meet the cooling demand of the equipment with large air volume and small enthalpy difference; Comparing the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, and calculating the difference between the two respectively; Determine whether the difference exceeds the preset adjustment threshold. If the difference exceeds the preset adjustment threshold, control the fan to adjust the speed and air volume to the pre-adjusted speed value and pre-adjusted air volume value, and accurately deliver the air volume to the local high-temperature area of ​​the equipment to reduce the temperature; if the difference does not exceed the adjustment threshold, maintain the current fan speed and air volume.

2. The method according to claim 1, characterized in that The method further comprises: Analyze the fluctuation pattern of real-time wind speed, extract the self-similar features in wind speed fluctuation, and identify the periodic fluctuation and random fluctuation components of wind speed; When comparing the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, the difference between the two is calculated based on the identified periodic fluctuation and random fluctuation components of the wind speed.

3. The method according to claim 2, characterized in that The method further comprises: Collect historical wind speed, wind direction, temperature and local temperature data of the fan's surrounding environment to build a training data set; The training data set is used to train the preset prediction model so that the preset prediction model can predict the wind speed and local temperature changes of the equipment within a period of time in the future, thereby obtaining a trained preset prediction model.

4. The method according to claim 3, characterized in that The method further comprises: The adjusted fan speed, air volume, environmental real-time data and equipment temperature data are stored together, and the database is updated to optimize the preset prediction model.

5. The method according to claim 3, characterized in that: The historical wind speed, wind direction, temperature and local temperature data of the equipment surrounding the fan are collected to construct a training data set, including: S11. Collect historical data of the wind turbine’s surrounding environment, including wind speed V h 、wind directionθ h , Temperature T h And the local temperature of the equipment T eq : S12. Preprocess the collected historical data, including data cleaning, denoising, and normalization, to obtain standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n ; S13. The standardized wind speed data V h,n 、Wind direction dataθ h,n , Temperature data T h,n And local temperature data T eq,n Combined into training data set D train : D train ={(V h,n ,θ h,n ,T h,n ,T eq,n )}; The method of using the training data set to train the preset prediction model so that the preset prediction model can predict the wind speed and local temperature changes of the equipment within a period of time in the future, and obtaining the trained preset prediction model, includes: S21. The training data set D train Organized into input and output pairs suitable for time series forecasting: S22. Define the neural network prediction model f θ To preset the prediction model, the neural network parameter is θ, which is used to capture the time dependence and nonlinear relationship between environmental data and device temperature; S23. Using the loss function L(θ) to train the neural network prediction model; S24. Update the neural network parameters θ by iteratively minimizing the loss function L(θ), and continue iterative training until the loss function L(θ) converges to the preset threshold ∈ L , and obtain the optimized neural network prediction model as the trained preset prediction model.

6. The method according to claim 3, characterized in that Determining the expected speed and air volume of the fan according to the wind speed prediction data and the equipment local temperature prediction data to obtain the optimized pre-adjusted speed value and pre-adjusted air volume value includes: S51: Receive wind speed forecast data And equipment local temperature prediction data Combined with real-time ambient temperature and real-time local device temperature Construct fan speed and air volume optimization function: Where n is the fan speed, Q is the fan air volume, and Represent the predicted local temperature of the device and the real-time local temperature of the device, respectively. and Represent the predicted wind speed and real-time wind speed respectively. Where Δh represents the enthalpy difference, H(Q) represents the air volume per unit enthalpy difference, α1, β1 and γ1 are the speed adjustment weight coefficient, air volume adjustment weight coefficient and enthalpy difference air volume optimization weight coefficient respectively; S52: Solve the optimization function F opt (n,Q), determine the pre-adjusted speed n that satisfies the minimization target opt and pre-adjusted air volume Q opt : S53: Calculate the pre-adjusted air volume value Q that meets the requirements of large air volume and small enthalpy difference through the enthalpy difference optimization formula opt : in, represents the difference in local temperature of the equipment, and R represents the ambient heat transfer constant.

7. The method according to claim 3, characterized in that The analysis of the fluctuation law of real-time wind speed, extraction of self-similar features in wind speed fluctuation, and identification of periodic fluctuation and random fluctuation components of wind speed include: S61: Get real-time wind speed data Time series: in, represents the real-time wind speed at time step t, t w is the analysis window length; S62: Construct a fractal dimension model of real-time wind speed based on spatial fractal algorithm and calculate the fractal dimension D of time series f : Among them, ∈ is the segmentation scale, N(∈) represents the minimum number of segments covering the wind speed time series at scale ∈; S63: Through fractal dimension D f Extract the self-similar features of wind speed time series. If D f ≈1.5, it indicates that the wind speed change has a significant random fluctuation component; if 1.0 <D f <1.5, it indicates that the wind speed change has periodic fluctuation characteristics; S64: The wind speed time series is decomposed using the characteristic separation model of periodic fluctuations and random fluctuations. series It is expressed as: V series =V periodic +V random ; Among them, V periodic Represents the periodic fluctuation component of wind speed, V random Represents the random fluctuation component of wind speed, V series The remainder after deducting the periodic fluctuations; The method of comparing the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, respectively calculating the difference between the two based on the identified periodic fluctuation and random fluctuation components of the wind speed, includes: S71: Obtaining the periodic fluctuation component V of the wind speed periodic and the random fluctuation component V random , combined with the current fan speed n current and air volume Q current , calculate the predicted periodic speed n periodic and randomly adjust the speed n random : Among them, n opt is the pre-adjusted speed value, is the predicted total wind speed, n periodic Indicates the speed adjustment to periodic fluctuations, n random Indicates the compensation of random fluctuations in speed; S72: Compare the current fan speed n current and pre-adjusted speed n opt , calculate the total speed difference Δn and the total air volume difference ΔQ: Δn=|n current -(n periodic +n random )| ΔQ=|Q current -Q opt |; Among them, Δn represents the difference between the current speed and the total adjusted speed, ΔQ represents the difference between the current air volume and the pre-adjusted air volume, Q current is the current air volume, Q opt It is the pre-adjusted air volume value.

8. The method according to claim 3, characterized in that The determination as to whether the difference exceeds a preset adjustment threshold value is made. If the difference exceeds the preset adjustment threshold value, the fan is controlled to adjust the speed and air volume to the pre-adjusted speed value and the pre-adjusted air volume value, so as to accurately deliver the air volume to the local high temperature area of ​​the equipment to reduce the temperature; If the difference does not exceed the adjustment threshold, the current fan speed and air volume are maintained, including: S81: Determine whether the speed difference Δn and the air volume difference ΔQ exceed the preset adjustment threshold ∈ n and ∈ Q , the judgment conditions are: Among them, Δn2 and ΔQ2 are the root mean square difference between the current speed and the total adjusted speed, and the integral difference between the current air volume and the pre-adjusted air volume, respectively. w is the observation time window, n periodic (k) and n random (k) are the periodic and random speed adjustment components, ∈ n and ∈ Q They are the adjustment thresholds for the rotation speed and air volume respectively; S82: If the condition is met, adjust the fan speed n in real time adjust and air volume Q adjust : Among them, η n and η Q is the speed and air volume adjustment rate, σ T Adjust the sensitivity parameter for the temperature difference, f t Adjust frequency for air volume fluctuations; S83: Adjust the air volume Q adjust Accurately guide the heat to the local high temperature area of ​​the equipment, through the heat transfer rate R heat To verify: Where ρ is the air density, c p is the specific heat capacity of air, L, W and H are the length, width and height of the local area of ​​the equipment, Indicates the actual equivalent temperature of the local high temperature area of ​​the equipment, T target Indicates the target temperature; S84: If Δn≤∈ n And ΔQ≤∈ Q , the current fan status remains unchanged, and the status data S is recorded current : And stored in database D state : D state =D state ∪S current 。 9. A fan speed dynamic automatic adjustment system, characterized in that: The system comprises: A prediction module is configured to obtain wind speed, wind direction, temperature and real-time temperature data of the fan surrounding environment, and input the data into a trained preset prediction model to generate wind speed prediction data and prediction data of the local temperature of the equipment for a period of time in the future; A determination module, which is configured to determine the expected speed and air volume of the fan according to the wind speed prediction data and the equipment local temperature prediction data, and obtain an optimized pre-adjusted speed value and pre-adjusted air volume value to meet the cooling demand of the equipment with large air volume and small enthalpy difference; a difference calculation module, configured to compare the pre-adjusted speed value with the current fan speed, and the pre-adjusted air volume value with the current air volume, and respectively calculate the difference between the two; A judgment and adjustment module is configured to judge whether the difference exceeds a preset adjustment threshold. If the difference exceeds the preset adjustment threshold, the fan is controlled to adjust the speed and air volume to the pre-adjusted speed value and pre-adjusted air volume value, and the air volume is accurately delivered to the local high-temperature area of ​​the equipment to reduce the temperature; if the difference does not exceed the adjustment threshold, the current fan speed and air volume are maintained.

10. The system according to claim 9, characterized in that The system also includes an identification module; The identification module is configured to analyze the fluctuation pattern of the real-time wind speed, extract the self-similar features in the wind speed fluctuation, and identify the periodic fluctuation and random fluctuation components of the wind speed; The difference calculation module is specifically configured to calculate the difference between the pre-adjusted speed value and the current fan speed, and the pre-adjusted air volume value and the current air volume based on the identified periodic fluctuation and random fluctuation components of the wind speed.

11. An electronic device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the method for dynamic automatic adjustment of fan speed as described in any one of claims 1-8.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for dynamically and automatically adjusting the fan speed according to any one of claims 1 to 8 is implemented.

Images