Fan Sleep Control Method and System Based on Load Prediction

Abstract

The present invention relates to a fan dormancy control method and system based on load prediction. The method includes: inputting the historical operation data of the fans into a gradient neural network for feature extraction and state space modeling to obtain temperature prediction error data and heat dissipation load prediction error data; inputting the temperature prediction error data and the heat dissipation load prediction error data into a load prototype library for pattern analysis to obtain a fan operating condition vector and an operating condition state transition matrix; inputting the fan operating condition vector and the operating condition state transition matrix into a transferability evaluation model for multi-dimensional similarity calculation to obtain heat dissipation load prediction data, and generating a dormancy control instruction according to the heat dissipation load prediction data; and performing collaborative optimization control on the fan group according to the dormancy control instruction to generate a fan group dormancy execution plan. The present invention comprehensively considers the airflow influence and spatial layout among the fans, optimizes the overall heat dissipation efficiency of the group, and realizes coordinated control from a single machine to a group.

Description

Technical Field

[0001] The present invention relates to the technical field of load prediction, and particularly to a fan dormancy control method and system based on load prediction. Background Art

[0002] The fan system faces a complex and changeable thermal load environment and different heat dissipation requirements, which makes fan load prediction and dormancy control extremely challenging. Traditional control methods based on fixed thresholds are difficult to adapt to the dynamic changes of the thermal environment and cannot effectively handle the collaborative work among fan groups.

[0003] Currently, fan dormancy control mainly relies on simple temperature thresholds and fixed switching strategies, lacking the ability to accurately predict real-time load status and dynamically adjust. Due to the volatility of the environmental temperature and the non-linear characteristics of the heat dissipation load, a single control strategy is difficult to achieve the optimal operation of fan groups. In addition, existing control methods often ignore the air flow interference and spatial layout relationship among fan groups, resulting in reduced overall heat dissipation efficiency and energy waste. In terms of fan load prediction, traditional methods mainly rely on real-time data from temperature sensors, making it difficult to predict in advance the load change trend and the influence law of environmental factors. At the same time, the differences in heat dissipation requirements and the diversity of working conditions among different equipment areas pose serious challenges to the applicability of the prediction model. These problems severely restrict the intelligent control level of fan groups and the improvement of energy-saving efficiency. Summary of the Invention

[0004] The main object of the present invention is to provide a fan dormancy control method and system based on load prediction. The present invention comprehensively considers the air flow influence and spatial layout among fans, optimizes the overall heat dissipation efficiency of the group, and realizes coordinated control from single machines to groups.

[0005] To achieve the above object, the present invention provides a fan dormancy control method based on load prediction, including the following steps:

[0006] Input the fan operation historical data into a gradient neural network for feature extraction and state space modeling to obtain temperature prediction error data and heat dissipation load prediction error data;

[0007] Input the temperature prediction error data and the heat dissipation load prediction error data into a load prototype library for pattern analysis to obtain a fan working condition vector and a working condition state transition matrix;

[0008] Input the fan working condition vector and the working condition state transition matrix into a transferability evaluation model for multi-dimensional similarity calculation to obtain heat dissipation load prediction data, and generate a dormancy control instruction according to the heat dissipation load prediction data;

[0009] Perform collaborative optimization control on the fan group according to the sleep control instruction to generate a fan group sleep execution plan.

[0010] The present invention also provides a fan sleep control system based on load prediction, including:

[0011] A modeling module for inputting the fan operation historical data into a gradient neural network for feature extraction and state space modeling to obtain temperature prediction error data and heat dissipation load prediction error data;

[0012] A mode analysis module for inputting the temperature prediction error data and the heat dissipation load prediction error data into a load prototype library for mode analysis to obtain a fan working condition vector and a working condition state transition matrix;

[0013] A calculation module for inputting the fan working condition vector and the working condition state transition matrix into a transferability evaluation model for multi-dimensional similarity calculation to obtain heat dissipation load prediction data, and generating a sleep control instruction according to the heat dissipation load prediction data;

[0014] A generation module for performing collaborative optimization control on the fan group according to the sleep control instruction to generate a fan group sleep execution plan.

[0015] The present invention also provides a computer device, including a memory and a processor, where a computer program is stored in the memory, and when the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0016] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0017] In summary, the technical solution provided by the present invention combines a gradient neural network and a state space model to achieve accurate extraction of heat dissipation load characteristics and accurate calculation of multi-dimensional coupling errors, improving the accuracy of load prediction. Based on the design of a load prototype library and a working condition state transition matrix, it realizes adaptive classification and dynamic prediction of the operating conditions of the fan, enhancing the adaptability of the control strategy to temperature environment changes. Using a transferability evaluation model for multi-dimensional similarity calculation, it solves the problem of dynamic adjustment of heat dissipation requirements in different regions and improves the practicality of the prediction model. Introducing an event trigger mechanism and a sliding mode control strategy, it realizes real-time optimization of fan sleep control and reduces the energy consumption of the system. Through the design of a group cooperative control model, it comprehensively considers the airflow influence and spatial layout among fans and optimizes the overall heat dissipation efficiency of the group. Establishing a complete abnormal handling and emergency response mechanism improves the stability of the fan group sleep control system. Using a multi-objective optimization method, taking into account both heat dissipation effect and energy consumption, it improves the energy-saving level of the system. Through a hierarchical control architecture design, it realizes coordinated control from a single machine to a group, ensuring the balanced performance of the heat dissipation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a schematic diagram of the steps of a fan sleep control method based on load prediction in an embodiment of the present invention;

[0019] Figure 2 is a block diagram of the structure of a fan sleep control system based on load prediction in an embodiment of the present invention;

[0020] Figure 3 is a schematic block diagram of the structure of a computer device in an embodiment of the present invention.

[0021] The implementation, functional features, and advantages of the objectives of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0023] Referring to Figure 1 , this embodiment provides a fan sleep control method based on load prediction, including the following steps:

[0024] S1, input the historical operation data of the fan into a gradient neural network for feature extraction and state space modeling to obtain temperature prediction error data and heat dissipation load prediction error data;

[0025] Among them, the historical operation data of the fan is divided into an environmental parameter sequence and a heat dissipation parameter sequence. Among them, the environmental parameter sequence includes operating conditions related to the external environment such as environmental temperature parameters, humidity parameters, and air flow parameters, while the heat dissipation parameter sequence includes equipment temperature parameters and heat dissipation load parameters, and these parameters directly reflect the heat exchange and load characteristics of the fan operation. The environmental parameter sequence is input into the input mapping unit of the gradient neural network for feature extraction. In the input mapping unit, a fully connected parameter layer and a feature extraction layer are designed. The fully connected parameter layer contains 4 neuron nodes, corresponding to environmental temperature, humidity, air flow, and time parameters respectively. This structure extracts the basic characteristics of each environmental parameter through preliminary weighting and bias adjustment. The feature extraction layer processes the output of the fully connected layer. This layer consists of 8 neuron nodes and uses the ReLU activation function, so as to effectively capture the non-linear characteristics between environmental parameters and generate an environmental mapping vector. The environmental mapping vector is merged with the heat dissipation parameter sequence and input into the double-layer hidden unit of the gradient neural network to complete feature fusion. In the double-layer hidden unit, a deep residual layer and an attention calculation layer are designed. The deep residual layer contains 16 neuron nodes, which are used to retain the key information of environmental parameters during the feature extraction process and at the same time alleviate the problem of gradient disappearance introduced by the deep network. At the same time, the attention calculation layer, through its 12 neuron nodes, can dynamically focus on the important associations between heat dissipation parameters and environmental characteristics, improving the effectiveness of feature fusion. The heat dissipation mapping vector is input into the feature fusion unit of the gradient neural network for multi-layer feature fusion. Through the collaborative action of multi-layer neurons, the network can comprehensively model the input data from a higher dimension and a wider perspective, extract the deep feature relationships in the heat dissipation mapping vector, and generate a fan heat dissipation feature vector. The fan heat dissipation feature vector is input into the state space model to complete multi-dimensional coupling calculation. In the state space model, using the time series and multi-dimensional features contained in the feature vector, the model can establish a dynamic expression of the fan operation state and obtain temperature prediction error data and heat dissipation load prediction error data through iterative calculations.

[0026] In this embodiment, the fan heat dissipation feature vector is decomposed in the state space to obtain a temperature feature sequence and a heat dissipation load feature sequence. Using the temperature feature sequence and the heat dissipation load feature sequence, a state space model is constructed and a state equation is generated. During the construction of the state equation, a temperature prediction error variable and a heat dissipation load prediction error variable are introduced to describe the deviation between the system prediction and the actual operating state. These variables are embedded in the state equation in a specific mathematical form to ensure that the model can capture the coupled dynamics of temperature and load during the operation of the fan. The state equation is input into the state transition calculation unit to construct the coupling features and generate a state transition matrix. The state transition matrix is an important parameter in the state space model that describes the dynamic changes of the system, and its determinant value is constrained between 0 and 1 to ensure the stability and invertibility of the matrix. The eigenvalues of this matrix are all set to positive real numbers, indicating that the system described by the model has stable dynamic evolution characteristics. Through this construction method, the state transition matrix can not only effectively capture the interaction relationship between temperature and heat dissipation load, but also avoid unstable phenomena in numerical calculations to a certain extent. By performing quantization calculations on the real-time collected temperature data and heat dissipation load data, an observation equation is constructed. The observation equation is another key component in the state space model, which is used to describe the relationship between the real-time observation data of the system and the predicted state of the model. By combining the observation equation with the state equation, a covariance matrix is established. The covariance matrix is used to quantify the error between the predicted state and the observation data and provide a reference basis for subsequent error correction. The construction of the covariance matrix needs to fully consider the dynamic characteristics of temperature and heat dissipation load and the correlation between the two to ensure the accuracy of error analysis. Perform Kalman filtering operation on the error quantization result to obtain a state update vector. Kalman filtering is a recursive optimal estimation method that reduces the random error in system prediction by updating the state estimate value in real time. The Kalman filtering operation combines the state equation, the observation equation, and the covariance matrix to optimize the error quantization result and obtain a state update vector. The state update vector reflects the state adjustment information of the system within the current time step. The state update vector is subjected to a coordinate mapping operation to obtain temperature prediction error data and heat dissipation load prediction error data.

[0027] S2. Input the temperature prediction error data and the heat dissipation load prediction error data into the load prototype library for pattern analysis to obtain a fan operating condition vector and an operating condition state transition matrix;

[0028] Specifically, preprocess the temperature prediction error data and the heat dissipation load prediction error data. Decompose the original data into multiple time series segments in a time series segmentation manner to capture the operating characteristics of the system in different time periods, while retaining the time dependence and change trend of the data. Construct a correlation coefficient matrix for the time series segments. The elements of this matrix represent the correlation between each time series segment, and the diagonal elements of the matrix are fixed at 1 to reflect the perfect correlation of the sequence with itself. By analyzing the correlation coefficient matrix, generate a time series correlation matrix and extract the potential correlation characteristics between the data. Based on the time series correlation matrix, conduct heat dissipation feature analysis and transform the time series segments into heat dissipation feature sequences with more physical significance. The heat dissipation feature sequences include temperature change sequences, heat dissipation efficiency sequences, and heat dissipation trend sequences, which respectively describe the temperature change pattern of the fan during operation, the efficiency fluctuation of the heat dissipation system, and the long-term trend of the heat dissipation capacity. Input the heat dissipation feature sequences into the load prototype library for feature matching to generate a prototype feature set. The load prototype library is a database containing various typical heat dissipation characteristics, and each feature prototype represents a common heat dissipation mode or operating condition characteristic. During the matching process, compare the similarity between the heat dissipation feature sequences and the prototypes in the library one by one, and finally determine the most similar prototype feature set. The prototype feature set contains M heat dissipation feature prototypes, where M is an integer greater than 1, indicating that the current heat dissipation characteristics of the system are described by the combination of M different prototype features. Perform K-means clustering on the prototype feature set to find the commonalities between the feature prototypes and classify them into several clustering centers. The clustering center set represents the core distribution points of the heat dissipation characteristics in the data space. Input the clustering center set into the Markov state transition unit for state calculation to generate an M×M-dimensional operating condition state transition matrix. Each element of this matrix represents the transition probability from one heat dissipation operating condition state to another. Through the calculation of the Markov model, capture the dynamic change law of the fan between different operating conditions and generate a complete state transition picture. Classify and calculate the current heat dissipation operating condition data according to the operating condition state transition matrix to generate a fan operating condition vector. Utilize the dynamic characteristics of the transition matrix to match the current operating condition data with possible operating condition states and determine the operating state of the fan at the current time point. The fan operating condition vector contains a comprehensive description of information such as temperature, heat dissipation efficiency, and trend.

[0029] S3, input the fan operating condition vector and the operating condition state transition matrix into the transferability evaluation model for multi-dimensional similarity calculation to obtain the heat dissipation load prediction data, and generate a sleep control instruction according to the heat dissipation load prediction data;

[0030] It should be noted that the fan operating condition vector and the condition state transition matrix are input into the eigen - decomposition unit of the transferability evaluation model for data splitting. By analyzing the structures of the operating condition vector and the state transition matrix, the eigen - decomposition unit extracts the environmental feature sequence and the heat - dissipation feature sequence respectively. The environmental feature sequence describes the external environmental conditions during the operation of the fan, including parameters such as temperature, humidity, and air flow. The heat - dissipation feature sequence focuses on heat - dissipation performance and load characteristics, such as key data like heat distribution and system efficiency. The environmental feature sequence is input into the multi - dimensional similarity calculation unit of the transferability evaluation model for feature mapping. Through the embedding learning method, the multi - dimensional similarity calculation unit maps the high - dimensional environmental features into a similarity space to generate environmental similarity parameters. The environmental similarity parameters quantify the similarity between the current environmental conditions and the typical environments in historical data. At the same time, the heat - dissipation feature sequence is input into the load distribution calculation unit of the transferability evaluation model. By probability calculation, the load distribution calculation unit analyzes the distribution characteristics of the heat - dissipation load. The load distribution calculation unit uses a statistical model to evaluate the probability distribution of the heat - dissipation features in different states and generates heat - dissipation distribution features to capture the heat - dissipation behavior rules of the current operating state. The heat - dissipation distribution features are input into the distribution consistency calculation unit for similarity calculation to obtain distribution similarity parameters. By comparing the differences between the current heat - dissipation distribution features and the historical distribution features, the distribution consistency calculation unit calculates their similarity and reveals the degree of consistency between the current heat - dissipation state and the historical typical patterns. At the same time, the environmental similarity parameters and the distribution similarity parameters are input into the weight fusion unit of the transferability evaluation model. These parameters are combined through weight adjustment according to their importance to obtain a transferability weight coefficient that comprehensively reflects the similarity of environmental and heat - dissipation features, indicating the applicability of the current fan operating condition in the prediction model. The transferability weight coefficient is input into the parameter optimization unit of the transferability evaluation model for coefficient update. The parameter optimization unit adopts dynamic optimization algorithms, such as gradient descent or adaptive learning rate algorithms, to iteratively update and optimize the transferability weight coefficient to obtain a more accurate sequence of prediction parameters. These prediction parameter sequences contain the key variables required for fan heat - dissipation load prediction. The optimized prediction parameter sequence is input into the load prediction unit of the transferability evaluation model for calculation to generate heat - dissipation load prediction data, which describes the change of the heat - dissipation load of the fan in the future time period. The heat - dissipation load prediction data is input into the event - triggered controller, and a sleep control instruction is generated through sliding - mode control optimization. Based on the heat - dissipation load prediction data, the event - triggered controller determines whether the fan needs to remain running or enter the sleep state within a specific time period. The sliding - mode control optimization method generates precise control instructions through the dynamic adjustment of the prediction data and the current system state to ensure the efficiency and reliability of the sleep strategy.

[0031] In this embodiment, the heat dissipation load prediction data is subjected to a sliding interval operation to dynamically capture the change trend of the heat dissipation load in different time periods, and a trigger condition function is constructed. The trigger condition function consists of a temperature error threshold variable and a heat dissipation state change rate variable. The former quantifies the deviation between the current system temperature and the set target, and the latter reflects the dynamic change rate of the heat dissipation system. The sliding interval operation smooths the data fluctuations through a gradually moving time window to ensure that the trigger condition function can accurately describe the operating state of the system. The trigger condition function is input into an event-triggered controller for error calculation to generate an error product matrix. During the error calculation process, the temperature error variable and the heat dissipation deviation variable are respectively extracted and a two-dimensional matrix is constructed through a product operation to describe the coupling relationship between the system temperature and the heat dissipation deviation. A Lyapunov-Krasovskii function is constructed based on the error product matrix to analyze the stability of the system. The Lyapunov-Krasovskii function is a stability analysis tool for delay systems, and its purpose is to evaluate whether the current system can maintain stable operation under input disturbances. By constructing a stability function within the Lyapunov-Krasovskii framework, the dynamic behavior of the system is transformed into a mathematically solvable problem. The stability function is input into a linear matrix inequality (LMI) solving unit for parameter calculation to generate a controller parameter matrix. The linear matrix inequality solving unit uses optimization algorithms such as semidefinite programming (SDP) to extract the optimal control parameters from the stability conditions to ensure that the system is robust while achieving the control objective. An adaptive law is constructed for the controller parameter matrix to generate an adaptive control function. The adaptive control function can dynamically respond to changes in the system operating environment by adjusting the control parameters in real time to improve the accuracy and adaptability of the control. The adaptive control function is input into a state feedback control unit for parameter adjustment to generate a sleep control parameter sequence. The sleep control parameter sequence includes a sleep start time variable, a sleep duration variable, and a wake-up trigger variable, which respectively correspond to when the fan enters the sleep state, the duration of the sleep, and when to wake up from the sleep state. Through the parameter adjustment of the state feedback control unit, these variables are optimized to the optimal values that meet the system performance requirements, thereby ensuring that the fan group minimizes energy consumption while maintaining the heat dissipation demand. The sleep control parameter sequence is input into a stability constraint analysis unit to calculate the stable control interval to ensure that the execution of the control instruction does not cause system instability or excessive shock. Through constraint analysis, it is judged whether the current parameter sequence falls within a safe and feasible control range, and the parameters that do not meet the conditions are adjusted. After the stable control interval analysis is completed, the optimized sleep control parameter sequence is converted into specific sleep control instructions. The instruction conversion process maps the parameter sequence into the operation instructions of the fan group, including which fans enter the sleep state, the specific sleep duration, and the wake-up time point.

[0032] S4. According to the sleep control instruction, perform collaborative optimization control on the fan group to generate a fan group sleep execution plan.

[0033] Specifically, the dormancy control instruction is input into the decision - making processing unit of the group collaborative control model to extract and calculate the fan influence data. In the decision - making processing unit, the spatial layout calculation module and the airflow interference calculation module undertake different calculation tasks. The spatial layout calculation module, based on the layout relationship of the fans in the physical space, uses the radial basis function (RBF) to calculate the influence coefficient between adjacent fans, thereby quantifying the interaction between the fans. At the same time, the airflow interference calculation module simulates the airflow distribution during the operation of the fans and its impact on the heat dissipation efficiency through a fluid dynamics model, and calculates the heat dissipation efficiency loss caused by airflow interference. These calculation results are integrated into the fan influence data, which describes the spatial coupling and airflow interaction characteristics of each fan in the group. The fan influence data is input into the objective function construction unit of the group collaborative control model to construct a multi - objective optimization objective function. In the objective function construction unit, the heat dissipation efficiency calculation module and the energy consumption calculation module respectively model and quantify the system performance from different perspectives. The heat dissipation efficiency calculation module, based on the heat exchange model, evaluates the overall heat dissipation of each fan in the group to ensure that the heat dissipation requirements of the system can be met under different dormancy strategies. The energy consumption calculation module, on the other hand, uses a load model to quantify the energy consumption level of the system under different strategies, thus providing support for energy efficiency optimization. By integrating the results of these two modules, the objective function construction unit generates a multi - objective optimization function to balance the relationship between heat dissipation efficiency and energy consumption. According to the optimization objective function, multi - objective genetic algorithm optimization is carried out to obtain the Pareto optimal solution set. The genetic algorithm generates a set of solutions that balance between different optimization objectives by simulating the natural selection process. These solutions form a solution set on the Pareto front, and each solution represents a possible dormancy execution strategy. The Pareto optimal solution set is input into the timing planning unit of the group collaborative control model to generate an execution timing plan according to the time - series requirements of the fan operation. The execution timing plan specifies when each fan in the fan group should dormancy, when to start, and the operation sequence in different time periods, providing a specific plan in the time dimension for collaborative control. The execution timing plan is input into the heat dissipation optimization unit for air volume distribution to obtain a heat dissipation control plan. The heat dissipation optimization unit optimizes the overall heat dissipation capacity of the group by distributing the air volume of each fan according to the operation characteristics and timing requirements of the fans, ensuring that the system can maintain stable heat dissipation performance during the execution of the dormancy strategy. The heat dissipation optimization unit constructs the state monitoring rules for the heat dissipation control plan to generate a monitoring parameter sequence. The monitoring parameter sequence includes temperature deviation variables, heat dissipation efficiency variables, and energy consumption intensity variables, which are used to monitor the system state in real - time to ensure the safety and stability of the control strategy during execution. The monitoring parameter sequence is input into the exception handling unit, and through the analysis of potential abnormal situations, an emergency control plan is generated. The emergency control plan includes specific countermeasures for fan failures or sudden changes in heat dissipation requirements to ensure that the system can still operate normally in case of accidents.Integrate and optimize the execution timing scheme, heat dissipation control scheme, and emergency control scheme to generate the final fan group dormancy execution scheme.

[0034] In one example, the steps of inputting the fan operation historical data into the gradient neural network for feature extraction and state space modeling include:

[0035] Divide the fan operation historical data into an environmental parameter sequence and a heat dissipation parameter sequence, where the environmental parameter sequence includes environmental temperature parameters, humidity parameters, and air flow parameters, and the heat dissipation parameter sequence includes equipment temperature parameters and heat dissipation load parameters;

[0036] Input the environmental parameter sequence into the input mapping unit of the gradient neural network for feature extraction to obtain an environmental mapping vector, where the input mapping unit includes a fully connected parameter layer and a feature extraction layer. The fully connected parameter layer contains 4 neuron nodes corresponding to environmental temperature, humidity, air flow, and time parameters respectively, and the feature extraction layer contains 8 neuron nodes and uses the ReLU activation function;

[0037] Input the environmental mapping vector and the heat dissipation parameter sequence into the double-layer hidden unit of the gradient neural network for feature fusion to obtain a heat dissipation mapping vector, where the double-layer hidden unit includes a deep residual layer and an attention calculation layer. The deep residual layer contains 16 neuron nodes for retaining environmental feature information, and the attention calculation layer contains 12 neuron nodes;

[0038] Input the heat dissipation mapping vector into the feature fusion unit of the gradient neural network for multi-layer feature fusion to obtain the fan heat dissipation feature vector;

[0039] Input the fan heat dissipation feature vector into the state space model for multi-dimensional coupling calculation to obtain temperature prediction error data and heat dissipation load prediction error data.

[0040] In this example, the fan operation historical data is divided into an environmental parameter sequence and a heat dissipation parameter sequence. The environmental parameter sequence includes environmental temperature , humidity , air flow velocity , and time feature , and these variables reflect the external conditions during fan operation. The heat dissipation parameter sequence includes equipment temperature and heat dissipation load , which are the core performance indicators of fan operation. Through division, the input data is modularized, facilitating the neural network to process environmental characteristics and heat dissipation characteristics separately. Input the environmental parameter sequence into the input mapping unit of the gradient neural network for feature extraction. The input mapping unit contains two main parts: a fully connected parameter layer and a feature extraction layer. In the fully connected parameter layer, each input parameter passes through the weight of a neuron node And bias Perform a linear transformation, and the calculation formula is:

[0041] ;

[0042] Where Is the output after linear transformation, Is the input parameter (such as ), Is the weight, Is the bias. The output of the fully connected layer is input to the feature extraction layer, which contains 8 neuron nodes, and each node uses the ReLU activation function. The calculation formula is:

[0043] ;

[0044] Through ReLU activation, the feature extraction layer captures the non-linear characteristics of the input data and generates the environment mapping vector , where each Is the extracted high-dimensional feature. The environment mapping vector And the heat dissipation parameter sequence Are input into the double-layer hidden unit of the gradient neural network for feature fusion. The double-layer hidden unit consists of a deep residual layer and an attention calculation layer. The deep residual layer maintains the environmental feature information through 16 neuron nodes, and the residual calculation formula is:

[0045] ;

[0046] Where Is the non-linear transformation of the feature, Is the residual output. In this way, the information flow of the environmental features is retained in the deep network to avoid information loss. The attention calculation layer performs weighted processing on the heat dissipation characteristics through 12 neuron nodes, and the attention weight The calculation formula is:

[0047] ;

[0048] Where And Are the query and key features in the attention mechanism, Indicates the Attention weight of the individual feature. The final output is the heat dissipation mapping vector , which is used to comprehensively describe the environmental and heat dissipation characteristics. The heat dissipation mapping vector Is input into the feature fusion unit of the gradient neural network for multi-layer feature fusion operations to generate the fan heat dissipation feature vector 。The fusion process consists of multiple fully connected layers and activation functions. The goal is to further compress and refine high-dimensional features so that they can represent the overall heat dissipation behavior of the fan. The heat dissipation feature vector of the fan is input into the state space model for multi-dimensional coupling calculation to generate temperature prediction error data and heat dissipation load prediction error data. In the state space model, the system state is represented by the state variable , and its dynamic equation is:

[0049] ;

[0050] where is the state transition matrix, is the control matrix, is the control matrix, is the process noise. The output equation of the system is:

[0051] ;

[0052] where is the output matrix, is the observation noise. By performing Kalman filter calculation on the dynamic equation and output equation of the system, the predicted temperature error and the heat dissipation load error are predicted, and their formula is:

[0053] ;

[0054] where and are the predicted temperature and load values of the model.

[0055] In one example, the steps of inputting the heat dissipation feature vector of the fan into the state space model for multi-dimensional coupling calculation to obtain temperature prediction error data and heat dissipation load prediction error data include:

[0056] Performing state space decomposition on the heat dissipation feature vector of the fan to obtain a temperature feature sequence and a heat dissipation load feature sequence;

[0057] Constructing a state space model based on the temperature feature sequence and the heat dissipation load feature sequence to obtain a state equation, where the state equation includes a temperature prediction error variable and a heat dissipation load prediction error variable;

[0058] Inputting the state equation into the state transition calculation unit for coupling feature construction to obtain a state transition matrix, where the determinant value of the state transition matrix is between 0 and 1 and the eigenvalues are all positive real numbers;

[0059] Quantify and calculate the temperature data and heat dissipation load data collected in real time to obtain an observation equation, and construct a covariance matrix based on the state equation and the observation equation to obtain an error quantization result;

[0060] Perform Kalman filtering operation on the error quantization result to obtain a state update vector, and perform coordinate mapping on the state update vector to obtain temperature prediction error data and heat dissipation load prediction error data.

[0061] In this example, the fan heat dissipation feature vector is decomposed in the state space, and the high-dimensional feature vector is divided into two main parts: the temperature feature sequence

[0062] and the heat dissipation load feature sequence . The temperature feature sequence describes the thermal characteristics of the system, including variables such as heat distribution and temperature gradient, while the heat dissipation load feature sequence reflects the load characteristics of the system, such as energy consumption and heat dissipation efficiency. The decomposition process is achieved through the linear mapping matrices and :

[0063] ;

[0064] where and are the mapping matrices respectively used to project the original feature vector into the corresponding feature space. Based on the temperature feature sequence and the heat dissipation load feature sequence a state space model is constructed to generate a state equation. The core of the state space model is to describe the dynamic evolution of the system, and its state equation is defined as:

[0065] ;

[0066] where is the state variable, including the temperature feature state and the load feature state ; is the state transition matrix, describing the dynamic relationship of the system from time step to ; is the control input matrix; is the control input; is the process noise, assumed to be a zero-mean Gaussian distribution. The determinant value of the state transition matrix is constrained within the range of [0, 1], and its eigenvalues are required to be positive real numbers to ensure the stability and reversibility of the system. To complete state estimation, the temperature data and heat dissipation load data collected in real time are input into the observation equation, and the form of the observation equation is:

[0067] ;

[0068] Among them, is the observed variable, including the temperature and heat dissipation load data at the current moment; is the observation matrix, connecting the state variable and the observed variable; is the observation noise, assumed to be a zero-mean Gaussian distribution. Through the state equation and the observation equation, the dynamic characteristics of the system are combined with the observed data. After the state and observation models are established, a covariance matrix is constructed based on the two. The covariance matrix is used to quantify the prediction error and observation error of the system, and its definition is:

[0069] ;

[0070] Among them, is the estimated value of the state. The covariance matrix is iteratively updated through the Kalman filter algorithm. The prediction step of the Kalman filter is:

[0071] ;

[0072] ;

[0073] Among them, is the time step of the predicted state, is the predicted covariance matrix, is the process noise covariance matrix. After observing new data, the update step of the Kalman filter is:

[0074] ;

[0075] ;

[0076] ;

[0077] Among them, is the Kalman gain matrix, is the observation noise covariance matrix. Through the prediction and update steps, the Kalman filter generates the state update vector . Coordinate mapping is performed on the state update vector to convert it into temperature prediction error data and heat dissipation load prediction error data. The formula is:

[0078] ;

[0079] Among them, and respectively represent the prediction errors of temperature and heat dissipation load.

[0080] In one example, the steps of inputting temperature prediction error data and heat dissipation load prediction error data into a load prototype library for pattern analysis to obtain a fan operating condition vector and an operating condition state transition matrix include:

[0081] Segment the temperature prediction error data and the heat dissipation load prediction error data by time series to obtain time series segments, and construct a correlation coefficient matrix for the time series segments to obtain a time series correlation matrix, where the diagonal elements of the correlation coefficient matrix are 1;

[0082] Conduct heat dissipation feature analysis on the time series correlation matrix to obtain a heat dissipation feature sequence, where the heat dissipation feature sequence includes a temperature change sequence, a heat dissipation efficiency sequence, and a heat dissipation trend sequence;

[0083] Input the heat dissipation feature sequence into the load prototype library for feature matching to obtain a prototype feature set, where the prototype feature set includes M heat dissipation feature prototypes, and M is an integer greater than 1;

[0084] Perform K-means clustering operation on the prototype feature set to obtain a clustering center set, and input the clustering center set into a Markov state transition unit for state calculation to obtain an M×M-dimensional operating condition state transition matrix, where each element of the operating condition state transition matrix represents the transition probability between heat dissipation operating conditions;

[0085] Classify and calculate the current heat dissipation operating condition data according to the operating condition state transition matrix to obtain a fan operating condition vector.

[0086] In this example, the temperature prediction error data and the heat dissipation load prediction error data are organized into a time series and segmented according to a fixed time window to form time series segments where each segment contains the error data within that window. The segmentation method can effectively capture the dynamic changes in heat dissipation characteristics within the time window. Construct a correlation coefficient matrix for the time series segments, define the correlation coefficient matrix whose element represents the correlation between time series segments and , and the calculation formula is:

[0087] ;

[0088] where, is the covariance of sequence segments and , and and are the standard deviations of the sequence segments respectively. The diagonal elements of the matrix , indicating that the sequence segment is completely correlated with itself. Based on the correlation coefficient matrix , the heat dissipation characteristics of the time series are analyzed. By the method of eigenvalue decomposition, the main eigenvalues and eigenvectors of the matrix are extracted to generate the heat dissipation characteristic sequence. The heat dissipation characteristic sequence includes three main parts: the temperature change sequence , the heat dissipation efficiency sequence and the heat dissipation trend sequence . The temperature change sequence describes the fluctuation characteristics of the temperature error in each time period, and its calculation formula is:

[0089] ;

[0090] where is the mean value of the temperature error. The heat dissipation efficiency sequence is calculated by the ratio of the heat dissipation load to the temperature change, and the formula is:

[0091] ;

[0092] The heat dissipation trend sequence uses regression analysis to calculate the change trend of the temperature error. The heat dissipation characteristic sequence is input into the load prototype library for feature matching. The load prototype library stores a variety of typical heat dissipation characteristic prototypes, and each prototype represents a common heat dissipation working condition mode. Feature matching is achieved by calculating the similarity between the heat dissipation characteristic sequence and the prototype features, and the similarity calculation formula is:

[0093] ;

[0094] where is the input heat dissipation characteristic sequence, is the th prototype feature, is the scale parameter that controls the similarity distribution. Through matching, a prototype feature set is generated, where represents the number of matched prototypes. K-means clustering operation is performed on the prototype feature set. The K-means algorithm generates a clustering center set by minimizing the sum of the squared distances between each clustering center and its members. Each clustering center represents the average characteristics of a group of similar heat dissipation features, and the optimization objective of the clustering process is:

[0095] ;

[0096] where is the th cluster. The clustering center set is input into the Markov state transition unit for state calculation to generate Multi - condition state transition matrix The matrix elements represent the probability of transitioning from the condition state to the state and satisfy:

[0097] ;

[0098] The state transition probability is calculated through frequency statistics of the time series. According to the multi - condition state transition matrix, the current heat dissipation condition data is classified and calculated to generate the fan condition vector Each component of the condition vector represents the characteristic weight of the current fan state.

[0099] In one example, the steps of inputting the fan condition vector and the multi - condition state transition matrix into the transferability evaluation model for multi - dimensional similarity calculation, obtaining the heat dissipation load prediction data, and generating the sleep control instruction based on the heat dissipation load prediction data include:

[0100] Input the fan condition vector and the multi - condition state transition matrix into the feature decomposition unit of the transferability evaluation model for data splitting to obtain the environmental feature sequence and the heat dissipation feature sequence;

[0101] Input the environmental feature sequence into the multi - dimensional similarity calculation unit of the transferability evaluation model for feature mapping to obtain the environmental similarity parameter, and input the heat dissipation feature sequence into the load distribution calculation unit of the transferability evaluation model for probability calculation to obtain the heat dissipation distribution feature;

[0102] Input the heat dissipation distribution feature into the distribution consistency calculation unit of the transferability evaluation model for similarity operation to obtain the distribution similarity parameter, and input the environmental similarity parameter and the distribution similarity parameter into the weight fusion unit of the transferability evaluation model for feature combination to obtain the transferability weight coefficient;

[0103] Input the transferability weight coefficient into the parameter optimization unit of the transferability evaluation model for coefficient update to obtain the prediction parameter sequence, and input the prediction parameter sequence into the load prediction unit of the transferability evaluation model for heat dissipation load calculation to obtain the heat dissipation load prediction data;

[0104] Input the heat dissipation load prediction data into the event - triggered controller for sliding - mode control optimization to obtain the sleep control instruction.

[0105] In this example, the fan condition vector and the multi - condition state transition matrix are input into the feature decomposition unit of the transferability evaluation model for data splitting. Through the matrix decomposition method, these two inputs are respectively decomposed into the environmental feature sequence

[0106] and the heat dissipation feature sequence . The core formula for feature decomposition is:

[0107] ;

[0108] where and are mapping matrices respectively used to extract environmental features and heat dissipation features. The environmental feature sequence reflects the influence of the external environment on the operation of the fan, while the heat dissipation feature sequence describes the internal dynamic characteristics of the heat dissipation system. The environmental feature sequence is input into the multi-dimensional similarity calculation unit of the transferability evaluation model for feature mapping. Through the kernel function method, the environmental features are mapped to a high-dimensional similarity space to generate the environmental similarity parameter . The mapping formula is:

[0109] ;

[0110] where is the reference environmental feature sequence, is the scale parameter controlling the similarity distribution. The environmental similarity parameter represents the similarity between the current environment and the historical environmental data. At the same time, the heat dissipation feature sequence is input into the load distribution calculation unit, and the heat dissipation distribution feature is calculated through the probability modeling method. The calculation formula for the distribution feature is:

[0111] ;

[0112] where is the parameter controlling the weight decay. The heat dissipation distribution feature describes the position of the current heat dissipation state in the overall load distribution. The heat dissipation distribution feature is input into the distribution consistency calculation unit and compared with the historical distribution feature to obtain the distribution similarity parameter . The calculation formula for the distribution similarity is:

[0113] ;

[0114] where is the historical heat dissipation distribution feature, and the closer the value of is to 1, the more consistent the current distribution is with the historical distribution. The environmental similarity parameter and the distribution similarity parameter are input into the weight fusion unit for feature combination to generate the transferability weight coefficient . The fusion process uses the weighted average method, and the formula is:

[0115] ;

[0116] Among them, and are weight parameters corresponding to environmental and distribution similarities respectively. The transferability weight coefficient is input into the parameter optimization unit for dynamic update to generate a predicted parameter sequence . The parameter optimization is iteratively updated by the gradient descent method:

[0117] ;

[0118] Among them, is the learning rate, is the objective function, usually the sum of squares of load prediction errors. The predicted parameter sequence is input into the load prediction unit to calculate the heat dissipation load prediction data . The prediction formula is:

[0119] ;

[0120] Among them, is the feature mapping function, which is used to generate load predictions by combining environmental and heat dissipation characteristics. The heat dissipation load prediction data is input into the event-triggered controller for sliding mode control optimization to generate a sleep control instruction. The core of the sliding mode control is to design a switching surface , and adjust the control strategy according to its dynamic behavior, defined as:

[0121] ;

[0122] Among them, and are sliding mode control parameters, is the reference load value. When , the control strategy is adjusted to reduce the load; when , the strategy is adjusted to increase the load. The finally generated sleep control instruction includes specific fan sleep time, duration, and trigger conditions.

[0123] In one example, the steps of inputting the heat dissipation load prediction data into the event-triggered controller for sliding mode control optimization to obtain the sleep control instruction include:

[0124] Performing a sliding interval operation on the heat dissipation load prediction data to obtain a trigger condition function, where the trigger condition function includes a temperature error threshold variable and a heat dissipation state change rate variable;

[0125] Inputting the trigger condition function into the event-triggered controller for error calculation to obtain an error product matrix, where the error product matrix includes a temperature error variable and a heat dissipation deviation variable;

[0126] Construct a Lyapunov-Krasovskii function based on the error product matrix to obtain a stability function, and input the stability function into a linear matrix inequality solving unit for parameter calculation to obtain the controller parameter matrix;

[0127] Construct an adaptive law for the controller parameter matrix to obtain an adaptive control function, and input the adaptive control function into a state feedback control unit for parameter adjustment to obtain a sleep control parameter sequence, where the sleep control parameter sequence includes a sleep start time variable, a sleep duration variable, and a wake-up trigger variable;

[0128] Conduct a stability constraint analysis on the sleep control parameter sequence to obtain a stable control interval, and perform command conversion on the sleep control parameter sequence according to the stable control interval to obtain a sleep control command.

[0129] In this example, the heat dissipation load prediction data is operated in a sliding interval according to a time window of fixed length to capture the short-term dynamic characteristics of the load change. The sliding interval operation is achieved by calculating the sliding average and the change rate, and the formula is:

[0130] ;

[0131] where, is the sliding average load, is the load change rate. Based on these operation results, a trigger condition function is constructed, which includes a temperature error threshold variable and a heat dissipation state change rate variable , and is defined as:

[0132] ;

[0133] where, represents the deviation between the measured temperature and the reference temperature . Input the trigger condition function into an event-triggered controller for error calculation to generate an error product matrix . The elements of the error product matrix are calculated by the following formula:

[0134] ;

[0135] where, and are the -th and -th components of the trigger condition function respectively. The error product matrix is a symmetric matrix used to describe the interaction effect of temperature error and heat dissipation deviation. According to the error product matrix Construct the Lyapunov-Krasovskii function to analyze the stability of the system. The Lyapunov-Krasovskii function has the general form of:

[0136] ;

[0137] where is the system state vector, and are positive definite matrices, is the system delay. By incorporating the error product matrix into the Lyapunov-Krasovskii framework, the stability function of the system is obtained. Input the stability function into the linear matrix inequality (LMI) solving unit for parameter calculation to generate the controller parameter matrix . The basic form of the LMI is:

[0138] ;

[0139] Through numerical optimization algorithms such as semidefinite programming (SDP), solve the controller parameter matrix that satisfies the stability conditions. Use the controller parameter matrix to construct the adaptive law to generate the adaptive control function . The adaptive control function is defined as:

[0140] ;

[0141] where is the control bias used to dynamically adjust the control input. Input the adaptive control function into the state feedback control unit for parameter adjustment to obtain the sleep control parameter sequence . The sleep control parameter sequence includes the sleep start time variable , the sleep duration variable and the wake-up trigger variable . Conduct a stability constraint analysis on the sleep control parameter sequence to ensure that the control instructions will not cause the system to become unstable during execution. By analyzing the control constraint region of the system, determine the stable control interval, and the constraint conditions are:

[0142] ;

[0143] where It is the stable control interval. According to the analysis results, the sleep control parameter sequence is optimized and adjusted. The optimized sleep control parameter sequence is converted into actual executable sleep control instructions. The instruction format includes the specific fan group number, sleep start time, duration, and wake-up condition.

[0144] In one example, the steps of performing collaborative optimization control on a fan group according to the sleep control instruction to generate a fan group sleep execution plan include:

[0145] Input the sleep control instruction into the decision processing unit of the group collaborative control model for data preprocessing to obtain fan influence data. The decision processing unit includes a spatial layout calculation module and an air flow interference calculation module. The spatial layout calculation module calculates the adjacent fan influence coefficient using a radial basis function, and the air flow interference calculation module calculates the heat dissipation efficiency loss based on a fluid dynamics model;

[0146] Input the fan influence data into the objective function construction unit of the group collaborative control model for multi-objective construction to obtain an optimized objective function. The objective function construction unit includes a heat dissipation efficiency calculation module and an energy consumption calculation module. The heat dissipation efficiency calculation module calculates the heat dissipation of the group based on a heat exchange model, and the energy consumption calculation module calculates the system energy consumption using a load model;

[0147] Perform multi-objective genetic algorithm optimization according to the optimized objective function to obtain a Pareto optimal solution set, and input the Pareto optimal solution set into the timing planning unit of the group collaborative control model for sequence generation to obtain an execution timing plan;

[0148] Input the execution timing plan into the heat dissipation optimization unit of the group collaborative control model for air volume distribution to obtain a heat dissipation control plan, and construct a state monitoring rule for the heat dissipation control plan to obtain a monitoring parameter sequence. The monitoring parameter sequence includes temperature deviation variables, heat dissipation efficiency variables, and energy consumption intensity variables;

[0149] Input the monitoring parameter sequence into the exception handling unit for emergency strategy generation to obtain an emergency control plan, and integrate and optimize the execution timing plan, heat dissipation control plan, and emergency control plan to obtain a fan group sleep execution plan.

[0150] In this example, input the sleep control instruction into the decision processing unit of the group collaborative control model for data preprocessing. In the decision processing unit, use the spatial layout calculation module and the air flow interference calculation module to evaluate the mutual influence between fans and the overall heat dissipation performance respectively. The spatial layout calculation module uses a radial basis function (RBF) to evaluate the influence coefficient of adjacent fans, defined as:

[0151] ;

[0152] Where, Denote the fan For the fan influence coefficient and are respectively the and position vectors of the fan is the influence range parameter. The airflow interference calculation module evaluates the loss of heat dissipation efficiency caused by airflow interference based on the fluid mechanics model, and its formula is:

[0153] ;

[0154] where is the loss of heat dissipation efficiency of the fan is the proportionality coefficient is the fan air volume. Through the above calculations, fan influence data and airflow interference data

[0155] are generated. The fan influence data is input into the objective function construction unit of the group cooperative control model to construct a multi-objective optimization objective function. The objective function construction unit includes a heat dissipation efficiency calculation module and an energy consumption calculation module. The heat dissipation efficiency calculation module evaluates the total heat dissipation of the group based on the heat exchange model, which is defined as:

[0156] ;

[0157] where is the total heat dissipation is the heat dissipation efficiency of the fan is the specific heat capacity of air is the fan air volume is the fan temperature difference. The energy consumption calculation module evaluates the total energy consumption of the system using the load model, which is defined as:

[0158] ;

[0159] where is the total energy consumption denotes the power consumption of the fan and are the power consumption model parameters. Combining these two modules, the multi-objective optimization function is obtained:

[0160] ;

[0161] ​​​The multi-objective genetic algorithm is used to solve the optimization objective function, generating a Pareto optimal solution set. The multi-objective genetic algorithm optimizes the objective function through an evolutionary process (selection, crossover, mutation), and finally obtains multiple solution sets that balance heat dissipation efficiency and energy consumption. The Pareto optimal solution set contains multiple optimal solutions, and each solution corresponds to a sleep strategy. Input the Pareto optimal solution set into the timing planning unit of the group collaborative control model, and generate an execution timing plan through time series analysis. According to the operating characteristics of the fan group, the timing planning unit assigns the operating and sleeping time periods for each fan, and the timing planning formula is:

[0162] ;

[0163] where, represents the operating time period of the fan , and and are the start and end times of operation respectively. Input the execution timing plan into the heat dissipation optimization unit, perform air volume distribution, and generate a heat dissipation control plan. Heat dissipation optimization optimizes the air volume distribution of each fan through linear programming methods to maximize the total heat dissipation of the group. The optimization objective is:

[0164] ;

[0165] where, is the upper limit of the total air volume of the fan group. At the same time, construct a state monitoring rule for the heat dissipation control plan, generate a monitoring parameter sequence, including the temperature deviation variable , the heat dissipation efficiency variable and the energy consumption intensity variable . The monitoring rule evaluates the operating state of the fan group by detecting these variables in real time. Input the monitoring parameter sequence into the exception handling unit to generate an emergency control plan. The emergency control plan responds to emergencies (such as fan failures or temperature anomalies) through a rule engine and threshold analysis. For example, when the temperature deviation of a certain fan exceeds the set threshold, the emergency control plan adjusts the air volume of other fans or starts the standby fan. Integrate and optimize the execution timing plan, the heat dissipation control plan, and the emergency control plan to generate the final fan group sleep execution plan. This plan includes the sleep time, operating air volume, and exception handling strategy of each fan.

[0166] Referring to Figure 2 , this embodiment provides a fan sleep control system based on load prediction, including:

[0167] A modeling module 1, which is used to input the fan operation historical data into a gradient neural network for feature extraction and state space modeling, obtaining temperature prediction error data and heat dissipation load prediction error data;

[0168] A pattern analysis module 2, configured to input temperature prediction error data and heat dissipation load prediction error data into a load prototype library for pattern analysis, so as to obtain a fan operating condition vector and an operating condition state transition matrix;

[0169] A calculation module 3, configured to input the fan operating condition vector and the operating condition state transition matrix into a transferability evaluation model for multi-dimensional similarity calculation, so as to obtain heat dissipation load prediction data, and generate a sleep control instruction according to the heat dissipation load prediction data;

[0170] A generation module 4, configured to perform cooperative optimization control on a fan group according to the sleep control instruction, and generate a fan group sleep execution plan.

[0171] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the description in the above method embodiment, and details are not described herein again.

[0172] Refer to Figure 3 , in the embodiment of the present invention, a computer device is further provided. The computer device may be a server, and its internal structure may be as Figure 3 shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0173] Those skilled in the art can understand that Figure 3 the structure shown in

[0174] is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0175] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0176] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, apparatus, article or method including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, apparatus, article or method. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, apparatus, article or method including the element.

[0177] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall equally be included in the patent protection scope of the present invention.

Claims

1. A fan sleep control method based on load prediction, characterized in that: The following steps are involved: Input the fan operation history data into the gradient neural network for feature extraction and state space modeling to obtain temperature prediction error data and heat dissipation load prediction error data; specifically including: dividing the fan operation history data into an environmental parameter sequence and a heat dissipation parameter sequence, wherein the environmental parameter sequence includes environmental temperature parameters, humidity parameters and airflow parameters, and the heat dissipation parameter sequence includes equipment temperature parameters and heat dissipation load parameters; input the environmental parameter sequence into the input mapping unit of the gradient neural network for feature extraction to obtain an environmental mapping vector, wherein the input mapping unit includes a fully connected parameter layer and a feature extraction layer, the fully connected parameter layer includes 4 neuron nodes corresponding to environmental temperature, humidity, airflow and time parameters respectively, and the feature extraction layer includes 4 neuron nodes corresponding to environmental temperature, humidity, airflow and time parameters respectively. The feature extraction layer includes 8 neuron nodes and adopts the ReLU activation function; the environment mapping vector and the heat dissipation parameter sequence are input into the double-layer hidden unit of the gradient neural network for feature fusion to obtain the heat dissipation mapping vector, wherein the double-layer hidden unit includes a deep residual layer and an attention calculation layer, the deep residual layer includes 16 neuron nodes and is used to retain the environmental feature information, and the attention calculation layer includes 12 neuron nodes; the heat dissipation mapping vector is input into the feature fusion unit of the gradient neural network for multi-layer feature fusion to obtain the fan heat dissipation feature vector; the fan heat dissipation feature vector is input into the state space model for multi-dimensional coupling calculation to obtain the temperature prediction error data and the heat dissipation load prediction error data; Inputting the temperature prediction error data and the heat dissipation load prediction error data into a load prototype library for pattern analysis to obtain a fan operating state vector and an operating state transfer matrix; Inputting the fan operating condition vector and the operating condition state transfer matrix into a transferability evaluation model to perform multi-dimensional similarity calculation to obtain heat dissipation load prediction data, and generating a sleep control instruction according to the heat dissipation load prediction data; The fan group is collaboratively optimized and controlled according to the sleep control instruction to generate a sleep execution plan for the fan group. The sleep execution plan for the fan group includes the sleep time, operating air volume and abnormality handling strategy of each fan.

2. The method according to claim 1, characterized in that The step of inputting the fan heat dissipation characteristic vector into a state space model for multi-dimensional coupling calculation to obtain the temperature prediction error data and the heat dissipation load prediction error data comprises: Decomposing the fan heat dissipation feature vector in state space to obtain a temperature feature sequence and a heat dissipation load feature sequence; Constructing a state space model according to the temperature characteristic sequence and the heat dissipation load characteristic sequence to obtain a state equation, wherein the state equation includes a temperature prediction error variable and a heat dissipation load prediction error variable; Inputting the state equation into a state transfer calculation unit for coupling feature construction to obtain a state transfer matrix, wherein the determinant value of the state transfer matrix is ​​between 0 and 1 and the eigenvalues ​​are all positive real numbers; Quantitatively calculate the temperature data and heat dissipation load data collected in real time to obtain an observation equation, and construct a covariance matrix based on the state equation and the observation equation to obtain an error quantification result; A Kalman filter operation is performed on the error quantization result to obtain a state update vector, and coordinate mapping is performed on the state update vector to obtain the temperature prediction error data and the heat dissipation load prediction error data.

3. The method according to claim 2, characterized in that The step of inputting the temperature prediction error data and the heat dissipation load prediction error data into a load prototype library for pattern analysis to obtain a fan operating state vector and an operating state transfer matrix comprises: The temperature prediction error data and the heat dissipation load prediction error data are segmented into time series to obtain time series segments, and a correlation coefficient matrix is ​​constructed for the time series segments to obtain a time series correlation matrix, wherein the diagonal elements of the correlation coefficient matrix are 1; Performing heat dissipation feature analysis on the time series correlation matrix to obtain a heat dissipation feature sequence, wherein the heat dissipation feature sequence includes a temperature change sequence, a heat dissipation efficiency sequence, and a heat dissipation trend sequence; Inputting the heat dissipation feature sequence into a load prototype library for feature matching to obtain a prototype feature set, wherein the prototype feature set includes M heat dissipation feature prototypes, where M is an integer greater than 1; Performing a K-means clustering operation on the prototype feature set to obtain a cluster center set, and inputting the cluster center set into a Markov state transfer unit for state calculation to obtain an M×M-dimensional operating state transfer matrix, wherein each element of the operating state transfer matrix represents a transition probability between heat dissipation operating conditions; The current heat dissipation working condition data is classified and calculated according to the working condition state transfer matrix to obtain a fan working condition vector.

4. The method according to claim 3, characterized in that The step of inputting the fan operating condition vector and the operating condition state transfer matrix into a transferability evaluation model to perform multi-dimensional similarity calculation to obtain heat dissipation load prediction data, and generating a sleep control instruction according to the heat dissipation load prediction data includes: Inputting the fan operating condition vector and the operating condition state transfer matrix into the feature decomposition unit of the transferability evaluation model for data splitting to obtain an environmental feature sequence and a heat dissipation feature sequence; Input the environmental feature sequence into the multidimensional similarity calculation unit of the transferability evaluation model for feature mapping to obtain environmental similarity parameters, and input the heat dissipation feature sequence into the load distribution calculation unit of the transferability evaluation model for probability calculation to obtain heat dissipation distribution features; Input the heat dissipation distribution feature into the distribution consistency calculation unit of the transferability evaluation model to perform similarity calculation to obtain a distribution similarity parameter, and input the environment similarity parameter and the distribution similarity parameter into the weight fusion unit of the transferability evaluation model to perform feature combination to obtain a transferability weight coefficient; Inputting the transferability weight coefficient into the parameter optimization unit of the transferability evaluation model to update the coefficient to obtain a prediction parameter sequence, and inputting the prediction parameter sequence into the load prediction unit of the transferability evaluation model to calculate the heat dissipation load to obtain heat dissipation load prediction data; The heat dissipation load prediction data is input into an event trigger controller to perform sliding mode control optimization to obtain a sleep control instruction.

5. The method according to claim 4, characterized in that The step of inputting the heat dissipation load prediction data into an event triggering controller to perform sliding mode control optimization to obtain a sleep control instruction comprises: Performing a sliding interval operation on the heat dissipation load prediction data to obtain a trigger condition function, wherein the trigger condition function includes a temperature error threshold variable and a heat dissipation state change rate variable; Inputting the trigger condition function into an event trigger controller to perform error calculation to obtain an error product matrix, wherein the error product matrix includes a temperature error variable and a heat dissipation deviation variable; Constructing a Lyapunov-Krasovskii function according to the error product matrix to obtain a stability function, and inputting the stability function into a linear matrix inequality solving unit for parameter operation to obtain a controller parameter matrix; An adaptive rule is constructed for the controller parameter matrix to obtain an adaptive control function, and the adaptive control function is input into a state feedback control unit for parameter adjustment to obtain a sleep control parameter sequence, wherein the sleep control parameter sequence includes a sleep start time variable, a sleep duration variable, and a wake-up trigger variable; A stability constraint analysis is performed on the sleep control parameter sequence to obtain a stable control interval, and an instruction conversion is performed on the sleep control parameter sequence according to the stable control interval to obtain a sleep control instruction.

6. The method according to claim 5, characterized in that The step of performing collaborative optimization control on the wind turbine group according to the sleep control instruction and generating a sleep execution plan for the wind turbine group comprises: Input the sleep control instruction into the decision processing unit of the group collaborative control model for data preprocessing to obtain fan influence data, wherein the decision processing unit includes a space layout calculation module and an airflow interference calculation module, the space layout calculation module uses a radial basis function to calculate the influence coefficient of adjacent fans, and the airflow interference calculation module calculates the heat dissipation efficiency loss based on a fluid mechanics model; Inputting the fan impact data into the objective function construction unit of the group collaborative control model for multi-objective construction to obtain an optimized objective function, wherein the objective function construction unit includes a heat dissipation efficiency calculation module and an energy consumption calculation module, the heat dissipation efficiency calculation module calculates the group heat dissipation based on the heat exchange model, and the energy consumption calculation module uses the load model to calculate the system energy consumption; Perform multi-objective genetic algorithm optimization according to the optimization objective function to obtain a Pareto optimal solution set, and input the Pareto optimal solution set into the timing planning unit of the group collaborative control model for sequence generation to obtain an execution timing plan; The execution timing scheme is input into the heat dissipation optimization unit of the group collaborative control model to distribute the air volume, so as to obtain a heat dissipation control scheme, and a state monitoring rule is constructed for the heat dissipation control scheme to obtain a monitoring parameter sequence, wherein the monitoring parameter sequence includes a temperature deviation variable, a heat dissipation efficiency variable and an energy consumption intensity variable; The monitoring parameter sequence is input into the abnormality processing unit to generate an emergency strategy to obtain an emergency control plan, and the execution timing plan, the heat dissipation control plan and the emergency control plan are integrated and optimized to obtain a fan group sleep execution plan.

7. A fan dormancy control system based on load prediction, characterized in that: For implementing the steps of the method according to any one of claims 1 to 6, the system comprises: A modeling module, used to input the fan operation history data into the gradient neural network for feature extraction and state space modeling, and obtain temperature prediction error data and heat dissipation load prediction error data; A mode analysis module, used for inputting the temperature prediction error data and the heat dissipation load prediction error data into a load prototype library for mode analysis to obtain a fan operating state vector and an operating state transfer matrix; A calculation module, used for inputting the fan operating condition vector and the operating condition state transfer matrix into a transferability evaluation model to perform multi-dimensional similarity calculation, obtain heat dissipation load prediction data, and generate a sleep control instruction according to the heat dissipation load prediction data; A generation module is used to perform collaborative optimization control on the fan group according to the sleep control instruction and generate a sleep execution plan for the fan group.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images