Intelligent Electric Fan Control System and Method Based on Internet of Things Technology

Abstract

The present application discloses an intelligent electric fan control system and method based on Internet of Things technology, which obtains the time series of ambient temperature and the time series of wind speed; performs data regularization on the time series of ambient temperature and the time series of wind speed to obtain a sequence of ambient temperature local time series input vectors and a sequence of wind speed local time series input vectors; calculates the sample covariance correlation matrix between each pair of corresponding ambient temperature local time series input vectors and wind speed local time series input vectors in the sequence of ambient temperature local time series input vectors and the sequence of wind speed local time series input vectors to obtain a sequence of temperature-wind speed correlation matrices in the local time domain; and determines a fan control instruction based on the implicit correlation features of the sequence of temperature-wind speed correlation matrices in the local time domain. In this way, the power of the electric fan can be automatically adjusted according to the changes in ambient temperature and wind speed, so as to provide an appropriate wind speed under different environmental conditions and improve the comfort of users.

Description

Technical Field

[0001] This application relates to the technical field of intelligent electric fan control, and particularly to an intelligent electric fan control system and method based on Internet of Things technology. Background Art

[0002] An electric fan is a common household item that can lower the body temperature and perceived temperature by blowing air, thereby improving people's comfort. However, traditional electric fans usually require users to manually adjust parameters such as switches and wind speeds, which not only brings inconvenience to users but also causes energy waste. For example, when the ambient temperature drops, the electric fan may still operate at a high power, resulting in power loss. Therefore, an intelligent electric fan control system and method are expected. Summary of the Invention

[0003] This application provides an intelligent electric fan control system and method based on Internet of Things technology, which acquires the time series of ambient temperature and the time series of wind speed; performs data regularization on the time series of ambient temperature and the time series of wind speed to obtain a sequence of ambient temperature local time series input vectors and a sequence of wind speed local time series input vectors; calculates the sample covariance correlation matrix between each pair of corresponding ambient temperature local time series input vectors and wind speed local time series input vectors in the sequence of ambient temperature local time series input vectors and the sequence of wind speed local time series input vectors to obtain a sequence of temperature-wind speed correlation matrices in the local time domain; and determines the fan control instruction based on the implicit correlation features of the sequence of temperature-wind speed correlation matrices in the local time domain. In this way, the power of the electric fan can be automatically adjusted according to the changes in ambient temperature and wind speed, so as to provide an appropriate wind speed under different environmental conditions and improve the comfort of users.

[0004] This application also provides an intelligent electric fan control method based on Internet of Things technology, which includes:

[0005] Acquiring the time series of ambient temperature collected by a temperature sensor;

[0006] Acquiring the time series of wind speed collected by a wind speed sensor;

[0007] Performing data regularization on the time series of ambient temperature and the time series of wind speed to obtain a sequence of ambient temperature local time series input vectors and a sequence of wind speed local time series input vectors;

[0008] Calculating the sample covariance correlation matrix between each pair of corresponding ambient temperature local time series input vectors and wind speed local time series input vectors in the sequence of ambient temperature local time series input vectors and the sequence of wind speed local time series input vectors to obtain a sequence of temperature-wind speed correlation matrices in the local time domain;

[0009] Determine the fan control instruction based on the implicit correlation features of the sequence of the temperature-wind speed correlation matrix within the local time domain.

[0010] In the above intelligent electric fan control method based on the Internet of Things technology, data regularization is performed on the time series of the environmental temperature and the time series of the wind speed to obtain a sequence of local time series input vectors of the environmental temperature and a sequence of local time series input vectors of the wind speed, including: sequence segmentation is performed on the time series of the environmental temperature and the time series of the wind speed based on a predetermined time scale to obtain a sequence of local time series of the environmental temperature and a sequence of local time series of the wind speed; the sequence of local time series of the environmental temperature and the sequence of local time series of the wind speed are respectively regularized in the time dimension to obtain the sequence of local time series input vectors of the environmental temperature and the sequence of local time series input vectors of the wind speed.

[0011] In the above intelligent electric fan control method based on the Internet of Things technology, calculate the sample covariance correlation matrix between each pair of corresponding local time series input vectors of the environmental temperature and local time series input vectors of the wind speed in the sequence of local time series input vectors of the environmental temperature and the sequence of local time series input vectors of the wind speed to obtain a sequence of temperature-wind speed correlation matrices within the local time domain, including: calculate the sample covariance matrix of the local time series input vector of the environmental temperature with respect to the local time series input vector of the wind speed using the following sample covariance correlation formula to obtain the temperature-wind speed correlation matrix within the local time domain; wherein, the sample covariance correlation formula is: ; wherein, is the local time series input vector of the environmental temperature, is the local time series input vector of the wind speed, is the temperature-wind speed correlation matrix within the local time domain.

[0012] In the above intelligent electric fan control method based on the Internet of Things technology, determining the fan control instruction based on the implicit association features of the sequence of temperature-wind speed association matrices in the local time domain includes: using a deep learning network model to perform implicit feature extraction on each temperature-wind speed association matrix in the sequence of temperature-wind speed association matrices in the local time domain to obtain a sequence of temperature-wind speed association feature vectors in the local time domain; passing the sequence of temperature-wind speed association feature vectors in the local time domain through an importance assignment network based on an adaptive attention module to obtain a weighted sequence of temperature-wind speed association feature vectors in the local time domain; performing fusion optimization processing on the sequence of temperature-wind speed association feature vectors in the local time domain and the weighted sequence of temperature-wind speed association feature vectors in the local time domain to obtain an optimized full-time domain temperature-wind speed association feature vector; passing the optimized full-time domain temperature-wind speed association feature vector through a classifier to obtain the fan control instruction, and the fan control instruction is used to indicate whether the power of the electric fan at the current time point can be reduced.

[0013] In the above intelligent electric fan control method based on the Internet of Things technology, the deep learning network model is a temperature-wind speed association feature extractor based on a convolutional neural network model.

[0014] In the above intelligent electric fan control method based on the Internet of Things technology, using a deep learning network model to perform implicit feature extraction on each temperature-wind speed association matrix in the sequence of temperature-wind speed association matrices in the local time domain to obtain a sequence of temperature-wind speed association feature vectors in the local time domain includes: passing each temperature-wind speed association matrix in the sequence of temperature-wind speed association matrices in the local time domain through the temperature-wind speed association feature extractor based on the convolutional neural network model to obtain the sequence of temperature-wind speed association feature vectors in the local time domain.

[0015] In the above intelligent electric fan control method based on the Internet of Things technology, passing the sequence of temperature-wind speed association feature vectors in the local time domain through an importance assignment network based on an adaptive attention module to obtain a weighted sequence of temperature-wind speed association feature vectors in the local time domain includes: processing the sequence of temperature-wind speed association feature vectors in the local time domain according to the following importance assignment formula to obtain the weighted sequence of temperature-wind speed association feature vectors in the local time domain; where the importance assignment formula is: ; where is the th temperature-wind speed association feature vector in the sequence of temperature-wind speed association feature vectors in the local time domain, is pooling processing, is a pooling vector, is a weight matrix, is the offset vector, is the activation process, is the initial meta-weight feature vector, is the th eigenvalue at the position in the initial meta-weight feature vector, is the corrected meta-weight feature vector, is the th eigenvalue at the position in the corrected meta-weight feature vector, is the th weighted local time-domain temperature-wind speed correlation feature vector in the sequence of weighted local time-domain temperature-wind speed correlation feature vectors, represents the dot product process.

[0016] In the above intelligent electric fan control method based on the Internet of Things technology, the optimized full-time-domain temperature-wind speed correlation feature vector is passed through a classifier to obtain the fan control instruction, and the fan control instruction is used to indicate whether the power of the electric fan at the current time point can be reduced, including: performing fully connected encoding on the optimized full-time-domain temperature-wind speed correlation feature vector using multiple fully connected layers of the classifier to obtain an encoded classification feature vector; and passing the encoded classification feature vector through the Softmax classification function of the classifier to obtain the classification result.

[0017] This application also provides an intelligent electric fan control system based on the Internet of Things technology, which includes:

[0018] A time series acquisition module for ambient temperature, configured to acquire the time series of ambient temperature collected by a temperature sensor;

[0019] A time series acquisition module for wind speed, configured to acquire the time series of wind speed collected by a wind speed sensor;

[0020] A data regularization module, configured to regularize the time series of ambient temperature and the time series of wind speed to obtain a sequence of ambient temperature local time series input vectors and a sequence of wind speed local time series input vectors;

[0021] A sample covariance correlation matrix calculation module, configured to calculate the sample covariance correlation matrix between each pair of corresponding ambient temperature local time series input vectors and wind speed local time series input vectors in the sequence of ambient temperature local time series input vectors and the sequence of wind speed local time series input vectors to obtain a sequence of local time-domain temperature-wind speed correlation matrices;

[0022] A fan control instruction determination module, configured to determine a fan control instruction based on the implicit correlation features of the sequence of local time-domain temperature-wind speed correlation matrices.

[0023] In the above intelligent electric fan control system based on Internet of Things technology, the data regularization module includes: a sequence segmentation unit, configured to segment the time series of the ambient temperature and the time series of the wind speed based on a predetermined time scale to obtain a sequence of local time series of the ambient temperature and a sequence of local time series of the wind speed; and a sequence regularization unit, configured to regularize the sequence of local time series of the ambient temperature and the sequence of local time series of the wind speed along the time dimension respectively to obtain a sequence of local time series input vectors of the ambient temperature and a sequence of local time series input vectors of the wind speed.

[0024] Compared with the prior art, the intelligent electric fan control system and method based on Internet of Things technology provided by this application obtain the time series of the ambient temperature and the time series of the wind speed; perform data regularization on the time series of the ambient temperature and the time series of the wind speed to obtain a sequence of local time series input vectors of the ambient temperature and a sequence of local time series input vectors of the wind speed; calculate the sample covariance correlation matrix between each pair of corresponding local time series input vectors of the ambient temperature and the local time series input vectors of the wind speed in the sequence of local time series input vectors of the ambient temperature and the sequence of local time series input vectors of the wind speed to obtain a sequence of temperature-wind speed correlation matrices in the local time domain; and determine the fan control instruction based on the implicit correlation features of the sequence of temperature-wind speed correlation matrices in the local time domain. In this way, the power of the electric fan can be automatically adjusted according to the changes in the ambient temperature and the wind speed, so as to provide an appropriate wind speed under different environmental conditions and improve the comfort of the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. In the drawings:

[0026] Figure 1 It is a flowchart of an intelligent electric fan control method based on Internet of Things technology provided in an embodiment of the present application.

[0027] Figure 2 It is a schematic diagram of the system architecture of an intelligent electric fan control method based on Internet of Things technology provided in an embodiment of the present application.

[0028] Figure 3 It is a block diagram of an intelligent electric fan control system based on Internet of Things technology provided in an embodiment of the present application.

[0029] Figure 4This is an application scenario diagram of an intelligent electric fan control method provided in an embodiment of the present application based on Internet of Things technology. Detailed implementation manners

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer and more understandable, the following further elaborates on the embodiments of the present application with reference to the accompanying drawings. Here, the illustrative embodiments of the present application and their descriptions are used to explain the present application, but do not limit the present application.

[0031] Unless otherwise specified, all the technical and scientific terms used in the embodiments of the present application have the same meanings as those commonly understood by those skilled in the technical field of the present application. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the scope of the present application.

[0032] In the description of the embodiments of the present application, it should be noted that unless otherwise specified and defined, the term "connection" should be understood in a broad sense. For example, it can be an electrical connection or the connection inside two components. It can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific situations.

[0033] It should be noted that the terms "first", "second", and "third" involved in the embodiments of the present application are only used to distinguish similar objects and do not represent a specific order for the objects. It can be understood that "first", "second", and "third" can be interchanged in a specific order or sequence when permitted. It should be understood that the objects distinguished by "first", "second", and "third" can be interchanged appropriately so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here.

[0034] An electric fan is a common household appliance, usually used to provide air flow and cooling in an indoor space. By rotating the blades, air flow is generated, enabling the indoor air to circulate, reducing the indoor temperature, improving air quality, and enhancing comfort. An electric fan is usually driven by an electric motor, and the blades rotate through a rotating shaft connected to the blades. The rotation of the blades causes the flow of the surrounding air, thus forming wind. An electric fan usually has different speed settings and can adjust the wind speed according to needs.

[0035] In addition to cooling, an electric fan can also provide a certain degree of comfort when an air conditioner is unavailable or unnecessary. Some electric fans also have a shaking function and can swing left and right or up and down to cover a wider area. In some regions, due to energy conservation and environmental protection considerations, people tend to use electric fans rather than air conditioners. Electric fans are usually more energy-efficient than air conditioners and can provide sufficient comfort in some cases where the weather is not extremely hot.

[0036] Traditional electric fan control systems have some drawbacks. Traditional electric fans usually require users to manually adjust parameters such as switches and wind speeds, which brings inconvenience to users, especially when frequent adjustments of wind speed or switches are needed. Due to the lack of intelligent control in traditional electric fans, when the ambient temperature drops, the electric fan may still operate at a high power, resulting in a waste of electric energy. In such cases, the electric fan fails to make intelligent adjustments according to actual needs, causing a waste of energy. Traditional electric fans usually lack automation functions and cannot automatically adjust wind speed and switches according to ambient temperature, humidity, or user needs, which limits their energy efficiency and user experience.

[0037] To solve these problems, an intelligent electric fan control system and method should be able to achieve automatic adjustment, intelligently adjust wind speed and switches according to environmental conditions and user needs, so as to improve energy utilization efficiency and enhance user experience. This may include using sensors to monitor environmental conditions, adopting intelligent algorithms to automatically adjust the operating state of the fan, and even integrating with a smart home system to achieve remote control and intelligent management.

[0038] In one embodiment of the present application, Figure 1 It is a flowchart of an intelligent electric fan control method based on Internet of Things technology provided in an embodiment of the present application. Figure 2 It is a schematic diagram of the system architecture of an intelligent electric fan control method based on Internet of Things technology provided in an embodiment of the present application. As Figure 1 and Figure 2 shown, the intelligent electric fan control method based on Internet of Things technology according to an embodiment of the present application includes: 110, obtaining a time series of ambient temperature collected by a temperature sensor; 120, obtaining a time series of wind speed collected by a wind speed sensor; 130, performing data regularization on the time series of the ambient temperature and the time series of the wind speed to obtain a sequence of ambient temperature local time series input vectors and a sequence of wind speed local time series input vectors; 140, calculating the sample covariance correlation matrix between each group of corresponding ambient temperature local time series input vectors and wind speed local time series input vectors in the sequence of the ambient temperature local time series input vectors and the sequence of the wind speed local time series input vectors to obtain a sequence of temperature-wind speed correlation matrices in the local time domain; 150, determining a fan control instruction based on the implicit correlation features of the sequence of the temperature-wind speed correlation matrices in the local time domain.

[0039] In step 110, obtain the time series of the ambient temperature collected by the temperature sensor, ensure the accuracy and stability of the temperature sensor to obtain reliable ambient temperature data, and provide basic data for subsequent steps by obtaining an accurate ambient temperature time series. In step 120, obtain the time series of the wind speed collected by the wind speed sensor, ensure the accuracy and stability of the wind speed sensor to obtain reliable wind speed data, and provide basic data for subsequent steps by obtaining an accurate wind speed time series. In step 130, perform data regularization on the time series of the ambient temperature and the time series of the wind speed to obtain a sequence of local time series input vectors of the ambient temperature and a sequence of local time series input vectors of the wind speed, and ensure appropriate time series regularization for the ambient temperature and wind speed data, such as time alignment, unified sampling rate, etc. Provide preparation for data processing and analysis in subsequent steps by obtaining regularized local time series input vector sequences of the ambient temperature and the wind speed. In step 140, calculate the sample covariance correlation matrix between each pair of corresponding local time series input vectors of the ambient temperature and the local time series input vectors of the wind speed in the sequence of local time series input vectors of the ambient temperature and the sequence of local time series input vectors of the wind speed to obtain a sequence of temperature-wind speed correlation matrices in the local time domain, and ensure that the time relationship between samples is considered during the calculation process to obtain an accurate temperature-wind speed correlation matrix in the local time domain. Provide a sequence of correlation matrices between the ambient temperature and the wind speed for subsequent analysis of the correlation characteristics between the temperature and the wind speed. In step 150, determine the fan control instruction based on the implicit correlation characteristics of the sequence of temperature-wind speed correlation matrices in the local time domain, perform effective feature extraction and analysis using the implicit features of the correlation matrix to determine an appropriate fan control instruction, and realize intelligent fan control by intelligently determining the fan control instruction according to the correlation characteristics between the ambient temperature and the wind speed, improving energy utilization efficiency and user experience.

[0040] Traditional electric fans usually require users to manually adjust parameters such as switches and wind speeds, which not only brings inconvenience to users but also causes energy waste. For example, when the ambient temperature drops, the electric fan may still operate at a high power, resulting in power loss. To implement an intelligent electric fan control system, multiple factors need to be considered, among which the most important ones are the ambient temperature and the wind speed. The ambient temperature is one of the important factors affecting the perceived temperature of the human body, while the wind speed is one of the important factors affecting the power and effect of the electric fan. Therefore, if the correlation relationship between the ambient temperature and the wind speed can be established and appropriate fan control instructions can be generated based on this correlation relationship, an intelligent electric fan control system can be realized.

[0041] However, the correlation between environmental temperature and wind speed is not a simple linear relationship, but a complex time-varying dynamic relationship. In this regard, the technical concept of this application is: to obtain the time series of environmental temperature collected by a temperature sensor and the time series of wind speed collected by a wind speed sensor, and use a deep learning model to extract the implicit correlation between environmental temperature and wind speed, and based on this implicit correlation to achieve intelligent adjustment of the power of the electric fan.

[0042] Among them, the deep learning model has powerful non-linear modeling capabilities and can learn and capture complex non-linear relationships. The correlation between environmental temperature and wind speed is often non-linear, and traditional linear models or simple machine learning algorithms may not be able to accurately model this complex relationship. The deep learning model can learn more complex non-linear mapping relationships through the combination of multiple hidden layers and activation functions, so as to better capture the implicit correlation between environmental temperature and wind speed. In this way, the power of the electric fan can be automatically adjusted according to the changes in environmental temperature and wind speed, so as to provide a suitable wind speed under different environmental conditions and improve the comfort of users.

[0043] Based on this, in the technical solution of this application, first, obtain the time series of environmental temperature collected by a temperature sensor; and obtain the time series of wind speed collected by a wind speed sensor. It should be understood that the change of environmental temperature has a great impact on the comfort of the human body. In hot weather, a lower environmental temperature can bring a sense of comfort, while in cold weather, a higher environmental temperature is more comfortable. Therefore, obtaining the time series of environmental temperature can help analyze the real-time environmental situation. In addition, wind speed is also an important factor affecting the perceived temperature of the human body. Different environmental conditions may require different wind speeds to achieve a comfortable effect. Therefore, obtaining the time series of environmental temperature and the time series of wind speed can monitor environmental changes in real time, automatically adjust the power of the electric fan according to the actual situation, provide a more comfortable and personalized wind speed experience, and reduce energy waste at the same time.

[0044] Then, based on a predetermined time scale, sequence segmentation is performed on the time series of the ambient temperature and the time series of the wind speed to obtain a sequence of local time series of the ambient temperature and a sequence of local time series of the wind speed; and the sequence of local time series of the ambient temperature and the sequence of local time series of the wind speed are respectively regularized according to the time dimension to obtain a sequence of local time series input vectors of the ambient temperature and a sequence of local time series input vectors of the wind speed. Among them, the time series of the ambient temperature and the time series of the wind speed often contain a large amount of information. If the time series of the ambient temperature and the time series of the wind speed are directly used for data analysis and feature extraction, it may increase the computational complexity. Here, by performing sequence segmentation on the time series of the ambient temperature and the time series of the wind speed based on a predetermined time scale, the entire time series can be segmented into smaller local sequences, so as to guide the model to better capture the local changes of the ambient temperature and the wind speed while reducing the computational amount of the subsequent model. Then, regularizing the sequence of local time series of the ambient temperature and the sequence of local time series of the wind speed according to the time dimension respectively can integrate the data structure into a unified vector representation, providing a more convenient and reliable data form for further data analysis and processing.

[0045] In a specific embodiment obtained in the present application, data regularization is performed on the time series of the ambient temperature and the time series of the wind speed to obtain a sequence of local time series input vectors of the ambient temperature and a sequence of local time series input vectors of the wind speed, including: performing sequence segmentation on the time series of the ambient temperature and the time series of the wind speed based on a predetermined time scale to obtain a sequence of local time series of the ambient temperature and a sequence of local time series of the wind speed; and respectively regularizing the sequence of local time series of the ambient temperature and the sequence of local time series of the wind speed according to the time dimension to obtain the sequence of local time series input vectors of the ambient temperature and the sequence of local time series input vectors of the wind speed.

[0046] Next, calculate the sample covariance correlation matrix between each pair of corresponding local time series input vectors of the ambient temperature and local time series input vectors of the wind speed in the sequence of local time series input vectors of the ambient temperature and the sequence of local time series input vectors of the wind speed to obtain a sequence of temperature-wind speed correlation matrices in the local time domain. That is, the correlation between each pair of corresponding local time series input vectors of the ambient temperature and local time series input vectors of the wind speed is constructed by calculating the sample covariance correlation matrix. Specifically, the sample covariance correlation matrix reflects the strength and direction of the linear relationship between the local time series input vectors of the ambient temperature and the local time series input vectors of the wind speed. By analyzing the elements of the covariance matrix, the correlation between the ambient temperature and the wind speed can be understood.

[0047] In a specific embodiment of the present application, calculating the sample covariance correlation matrix between each pair of corresponding ambient temperature local time-series input vectors and wind speed local time-series input vectors in the sequence of the ambient temperature local time-series input vectors and the sequence of the wind speed local time-series input vectors to obtain a sequence of temperature-wind speed correlation matrices in the local time domain, including: calculating the sample covariance matrix of the ambient temperature local time-series input vectors relative to the wind speed local time-series input vectors according to the following sample covariance correlation formula to obtain the temperature-wind speed correlation matrix in the local time domain; wherein, the sample covariance correlation formula is: ; wherein, is the ambient temperature local time-series input vector, is the wind speed local time-series input vector, is the temperature-wind speed correlation matrix in the local time domain.

[0048] Further, in an embodiment of the present application, based on the implicit correlation features of the sequence of the temperature-wind speed correlation matrices in the local time domain, determining a fan control instruction, including: using a deep learning network model to perform implicit feature extraction on each of the temperature-wind speed correlation matrices in the sequence of the temperature-wind speed correlation matrices in the local time domain to obtain a sequence of temperature-wind speed correlation feature vectors in the local time domain; passing the sequence of the temperature-wind speed correlation feature vectors in the local time domain through an importance assignment network based on an adaptive attention module to obtain a weighted sequence of temperature-wind speed correlation feature vectors in the local time domain; performing fusion optimization processing on the sequence of the temperature-wind speed correlation feature vectors in the local time domain and the weighted sequence of the temperature-wind speed correlation feature vectors in the local time domain to obtain an optimized full-time domain temperature-wind speed correlation feature vector; passing the optimized full-time domain temperature-wind speed correlation feature vector through a classifier to obtain the fan control instruction, and the fan control instruction is used to indicate whether the power of the electric fan at the current time point can be reduced.

[0049] Subsequently, each temperature-wind speed correlation matrix within the local time domain in the sequence of temperature-wind speed correlation matrices within the local time domain is passed through a temperature-wind speed correlation feature extractor based on a convolutional neural network model to obtain a sequence of temperature-wind speed correlation feature vectors within the local time domain. Among them, the convolutional neural network (CNN) model is good at capturing complex local spatial and temporal patterns. By taking each temperature-wind speed correlation matrix within the local time domain as input, the CNN can automatically learn local patterns, spatial relationships, and temporal features, so as to better identify the implicit correlation features between temperature and wind speed within the local neighborhood. More specifically, the CNN model can extract the local neighborhood correlation features between temperature and wind speed by stacking convolutional layers and pooling layers. Moreover, the non-linear characteristics of the CNN model enable it to model complex temperature-wind speed correlation features. By using appropriate activation functions (such as ReLU) and non-linear layers (such as Batch Normalization), the CNN can better adapt to various correlation patterns between temperature and wind speed.

[0050] Among them, the deep learning network model is a temperature-wind speed correlation feature extractor based on a convolutional neural network model.

[0051] Furthermore, in a specific embodiment of the present application, using a deep learning network model to perform implicit feature extraction on each temperature-wind speed correlation matrix in the sequence of temperature-wind speed correlation matrices within the local time domain to obtain a sequence of temperature-wind speed correlation feature vectors within the local time domain includes: passing each temperature-wind speed correlation matrix in the sequence of temperature-wind speed correlation matrices within the local time domain through the temperature-wind speed correlation feature extractor based on the convolutional neural network model to obtain the sequence of temperature-wind speed correlation feature vectors within the local time domain.

[0052] Furthermore, the sequence of temperature-wind speed correlation feature vectors within the local time domain is passed through an importance assignment network based on an adaptive attention module to obtain a sequence of weighted temperature-wind speed correlation feature vectors within the local time domain. Here, by passing the sequence of temperature-wind speed correlation feature vectors within the local time domain through the importance assignment network based on the adaptive attention module, dynamic assignment based on the importance degree of each temperature-wind speed correlation feature vector can be performed, so that each of the weighted temperature-wind speed correlation feature vectors within the local time domain can highlight important feature regions, thereby specifically focusing on the features related to fan control.

[0053] In a specific embodiment of the present application, passing the sequence of temperature-wind speed correlation feature vectors in the local time domain through an importance assignment network based on an adaptive attention module to obtain a sequence of weighted temperature-wind speed correlation feature vectors in the local time domain includes: processing the sequence of temperature-wind speed correlation feature vectors in the local time domain with the following importance assignment formula to obtain the sequence of weighted temperature-wind speed correlation feature vectors in the local time domain; wherein, the importance assignment formula is: ; wherein, is the -th temperature-wind speed correlation feature vector in the sequence of temperature-wind speed correlation feature vectors in the local time domain, is pooling processing, is a pooling vector, is a weight matrix, is a bias vector, is activation processing, is an initial meta-weight feature vector, is the eigenvalue at the -th position in the initial meta-weight feature vector, is a corrected meta-weight feature vector, is the eigenvalue at the -th position in the corrected meta-weight feature vector, is the -th weighted temperature-wind speed correlation feature vector in the sequence of weighted temperature-wind speed correlation feature vectors in the local time domain, represents dot product processing.

[0054] In the technical solution of the present application, the sequence of temperature-wind speed correlation feature vectors in the local time domain expresses the high-order local time series correlation features of the local time series full covariance correlation between the ambient temperature and the wind speed in the global time domain via sequence segmentation determined in the local time domain. In this way, after passing the sequence of temperature-wind speed correlation feature vectors in the local time domain through the importance assignment network based on the adaptive attention module, the global distribution based on the local time domain of the sequence of temperature-wind speed correlation feature vectors in the local time domain can be strengthened, thereby strengthening the expression effect of the sequence of weighted temperature-wind speed correlation feature vectors in the local time domain.

[0055] However, this will also cause the temporal correlation feature representation of the sequence of the weighted local time-domain temperature-wind speed correlation feature vectors to deviate from the temporal correlation feature representation based on the local time-domain distribution of the sequence of the local time-domain temperature-wind speed correlation feature vectors. Therefore, the applicant of the present application considers improving the semantic feature representation of the sequence of the weighted local time-domain temperature-wind speed correlation feature vectors by further fusing the sequence of the local time-domain temperature-wind speed correlation feature vectors with the sequence of the weighted local time-domain temperature-wind speed correlation feature vectors.

[0056] Moreover, considering the difference in the feature distribution information representation between the sequence of the weighted local time-domain temperature-wind speed correlation feature vectors and the sequence of the weighted local time-domain temperature-wind speed correlation feature vectors, in order to improve the consistency of the distribution information representation during fusion, the applicant of the present application performs fusion optimization on the sequence of the local time-domain temperature-wind speed correlation feature vectors and the sequence of the weighted local time-domain temperature-wind speed correlation feature vectors, which is specifically expressed as: fusing and optimizing the sequence of the local time-domain temperature-wind speed correlation feature vectors and the sequence of the weighted local time-domain temperature-wind speed correlation feature vectors with the following optimization formula to obtain the optimized full-time-domain temperature-wind speed correlation feature vectors; wherein, the optimization formula is: ; wherein, is the first feature vector obtained by concatenating the sequence of the local time-domain temperature-wind speed correlation feature vectors, and is the second feature vector obtained by concatenating the sequence of the weighted local time-domain temperature-wind speed correlation feature vectors, and respectively represent the squares of the 1-norm and 2-norm of the feature vectors. The first feature vector and the second feature vector have the same feature vector length , and is the weight hyperparameter, and are the eigenvalues of the first feature vector and the second feature vector respectively, is the eigenvalue of the optimized full-time-domain temperature-wind speed correlation feature vector, represents the logarithmic function with base 2.

[0057] Here, in order to improve the consistency of the distribution information representation between the sequence of the temperature-wind speed correlation feature vectors in the local time domain and the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain in the feature fusion scenario, the absolute coordinates of the distribution regression are predefined through the feature scale and structural representation of the feature vectors to be fused as the benchmark for the feature value cross-geometric registration. In this way, the rigid grid consistency of the information distribution can be maintained, and the idea of the probability chamfer loss is used to penalize the distance-based misalignment and incomplete overlap between the feature distribution information representations, so as to achieve the feature fusion with consistent distribution information representation between the sequence of the temperature-wind speed correlation feature vectors in the local time domain and the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain. In this way, the expression effect of the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain is improved, and thus the expression effect of the full-time domain temperature-wind speed correlation feature vector obtained by cascading the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain is improved, and the accuracy of the classification result obtained by the classifier is improved.

[0058] Subsequently, the optimized full-time domain temperature-wind speed correlation feature vector is passed through a classifier to obtain a fan control instruction, and the fan control instruction is used to indicate whether the power of the electric fan at the current time point can be reduced. That is, the information of the local time series is integrated into the full-time domain temperature-wind speed correlation feature distribution through cascading to comprehensively represent the temperature-wind speed correlation pattern and dynamic changes within the entire time domain range. Then, the classifier is used to learn the mapping relationship from the optimized full-time domain temperature-wind speed correlation feature vector to the fan control instruction, so as to intelligently judge whether the power of the electric fan can be reduced.

[0059] In a specific embodiment of the present application, passing the optimized full-time domain temperature-wind speed correlation feature vector through a classifier to obtain the fan control instruction, where the fan control instruction is used to indicate whether the power of the electric fan at the current time point can be reduced, includes: performing fully connected encoding on the optimized full-time domain temperature-wind speed correlation feature vector using multiple fully connected layers of the classifier to obtain an encoded classification feature vector; and passing the encoded classification feature vector through the Softmax classification function of the classifier to obtain the classification result.

[0060] In summary, the intelligent electric fan control method based on the Internet of Things technology according to the embodiments of the present application is clarified. It obtains the time series of the ambient temperature collected by the temperature sensor and the time series of the wind speed collected by the wind speed sensor, uses a deep learning model to extract the implicit correlation relationship between the ambient temperature and the wind speed, and realizes the intelligent adjustment of the power of the electric fan according to this implicit correlation relationship.

[0061] Figure 3It is a block diagram of an intelligent electric fan control system based on Internet of Things technology provided in an embodiment of this application. As Figure 3 shown, the intelligent electric fan control system 200 based on Internet of Things technology includes: a time series acquisition module 210 of ambient temperature, configured to acquire the time series of ambient temperature collected by a temperature sensor; a time series acquisition module 220 of wind speed, configured to acquire the time series of wind speed collected by a wind speed sensor; a data regularization module 230, configured to regularize the time series of ambient temperature and the time series of wind speed to obtain a sequence of local time series input vectors of ambient temperature and a sequence of local time series input vectors of wind speed; a sample covariance correlation matrix calculation module 240, configured to calculate the sample covariance correlation matrix between each pair of corresponding local time series input vectors of ambient temperature and local time series input vectors of wind speed in the sequence of local time series input vectors of ambient temperature and the sequence of local time series input vectors of wind speed to obtain a sequence of temperature-wind speed correlation matrices in the local time domain; and a fan control instruction determination module 250, configured to determine a fan control instruction based on the implicit correlation features of the sequence of temperature-wind speed correlation matrices in the local time domain.

[0062] In the intelligent electric fan control system based on Internet of Things technology, the data regularization module includes: a sequence segmentation unit, configured to segment the time series of ambient temperature and the time series of wind speed based on a predetermined time scale to obtain a sequence of local time series of ambient temperature and a sequence of local time series of wind speed; and a sequence regularization unit, configured to regularize the sequence of local time series of ambient temperature and the sequence of local time series of wind speed respectively according to the time dimension to obtain the sequence of local time series input vectors of ambient temperature and the sequence of local time series input vectors of wind speed.

[0063] Those skilled in the art can understand that the specific operations of each step in the above intelligent electric fan control system based on Internet of Things technology have been introduced in detail in the description of the Figures 1 to 2 intelligent electric fan control method based on Internet of Things technology, and therefore, the repeated description thereof will be omitted.

[0064] As described above, the intelligent electric fan control system 200 based on the Internet of Things technology according to the embodiments of the present application can be implemented in various terminal devices, such as a server for intelligent electric fan control based on the Internet of Things technology. In one example, the intelligent electric fan control system 200 based on the Internet of Things technology according to the embodiments of the present application can be integrated into the terminal device as a software module and / or a hardware module. For example, the intelligent electric fan control system 200 based on the Internet of Things technology can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the intelligent electric fan control system 200 based on the Internet of Things technology can also be one of the many hardware modules of the terminal device.

[0065] Alternatively, in another example, the intelligent electric fan control system 200 based on the Internet of Things technology and the terminal device can also be separate devices, and the intelligent electric fan control system 200 based on the Internet of Things technology can be connected to the terminal device through a wired and / or wireless network and transmit interaction information in accordance with a predefined data format.

[0066] Figure 4 This is an application scenario diagram of an intelligent electric fan control method provided in the embodiments of the present application. As Figure 4 shown, in this application scenario, first, obtain the time series of the ambient temperature collected by the temperature sensor (for example, C1 as illustrated in Figure 4 ); and obtain the time series of the wind speed collected by the wind speed sensor (for example, C2 as illustrated in Figure 4 ); then, input the obtained time series of the ambient temperature and the time series of the wind speed into a server (for example, S as illustrated in Figure 4 ) where an intelligent electric fan control algorithm based on the Internet of Things technology is deployed, and the server can process the time series of the ambient temperature and the time series of the wind speed based on the intelligent electric fan control algorithm based on the Internet of Things technology to determine a fan control instruction.

[0067] The specific embodiments described above further elaborate on the purpose, technical solutions, and beneficial effects of the present application. It should be understood that the above are only specific embodiments of the present application and are not used to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent electric fan control method based on Internet of Things technology, characterized in that, Including: Obtain the time series of the ambient temperature collected by the temperature sensor; Obtain the time series of the wind speed collected by the wind speed sensor; Perform data regularization on the time series of the ambient temperature and the time series of the wind speed to obtain a sequence of ambient temperature local time series input vectors and a sequence of wind speed local time series input vectors; Calculate the sample covariance correlation matrix between each pair of corresponding ambient temperature local time series input vectors and wind speed local time series input vectors in the sequence of ambient temperature local time series input vectors and the sequence of wind speed local time series input vectors to obtain a sequence of temperature-wind speed correlation matrices in the local time domain; Determine the fan control instruction based on the implicit correlation features of the sequence of temperature-wind speed correlation matrices in the local time domain; Among them, determining the fan control instruction based on the implicit correlation features of the sequence of temperature-wind speed correlation matrices in the local time domain includes: Use a deep learning network model to perform implicit feature extraction on each temperature-wind speed correlation matrix in the sequence of temperature-wind speed correlation matrices in the local time domain to obtain a sequence of temperature-wind speed correlation feature vectors in the local time domain; Pass the sequence of temperature-wind speed correlation feature vectors in the local time domain through an importance assignment network based on an adaptive attention module to obtain a sequence of weighted temperature-wind speed correlation feature vectors in the local time domain; Perform fusion optimization processing on the sequence of temperature-wind speed correlation feature vectors in the local time domain and the sequence of weighted temperature-wind speed correlation feature vectors in the local time domain to obtain an optimized full-time domain temperature-wind speed correlation feature vector; Pass the optimized full-time domain temperature-wind speed correlation feature vector through a classifier to obtain the fan control instruction, and the fan control instruction is used to indicate whether the power of the electric fan at the current time point can be reduced; Among them, passing the sequence of temperature-wind speed correlation feature vectors in the local time domain through an importance assignment network based on an adaptive attention module to obtain a sequence of weighted temperature-wind speed correlation feature vectors in the local time domain includes: Process the sequence of the temperature-wind speed correlation feature vectors in the local time domain with the following importance assignment formula to obtain the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain; wherein, the importance assignment formula is: ; wherein, is the -th temperature-wind speed correlation feature vector in the sequence of the temperature-wind speed correlation feature vectors in the local time domain, is the pooling process, is the pooling vector, is the weight matrix, is the bias vector, is the activation process, is the initial meta-weight feature vector, is the eigenvalue at the -th position in the initial meta-weight feature vector, is the corrected meta-weight feature vector, is the eigenvalue at the -th position in the corrected meta-weight feature vector, is the -th weighted temperature-wind speed correlation feature vector in the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain, represents the dot product process; Among them, the sequence of the temperature-wind speed correlation feature vectors in the local time domain and the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain are fused and optimized, which is specifically expressed as: the sequence of the temperature-wind speed correlation feature vectors in the local time domain and the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain are fused and optimized with the following optimization formula to obtain the optimized full-time domain temperature-wind speed correlation feature vector; among them, the optimization formula is: ; where is the first feature vector obtained by concatenating the sequence of the temperature-wind speed correlation feature vectors in the local time domain, and is the second feature vector obtained by concatenating the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain, and respectively represent the squares of the 1-norm and 2-norm of the feature vector. The first feature vector and the second feature vector have the same feature vector length , and is the weight hyperparameter, and are the eigenvalues of the first feature vector and the second feature vector respectively, is the eigenvalue of the optimized full-time domain temperature-wind speed correlation feature vector, represents the logarithmic function with base 2.

2. The intelligent electric fan control method based on Internet of Things technology according to claim 1, characterized in that, Performing data regularization on the time series of the ambient temperature and the time series of the wind speed to obtain a sequence of ambient temperature local time series input vectors and a sequence of wind speed local time series input vectors includes: Perform sequence segmentation on the time series of the ambient temperature and the time series of the wind speed based on a predetermined time scale to obtain a sequence of local time series of the ambient temperature and a sequence of local time series of the wind speed; Regularize the sequence of local time series of the ambient temperature and the sequence of local time series of the wind speed respectively according to the time dimension to obtain the sequence of ambient temperature local time series input vectors and the sequence of wind speed local time series input vectors.

3. The intelligent electric fan control method based on Internet of Things technology according to claim 2, wherein, Calculating the sample covariance correlation matrix between each pair of corresponding ambient temperature local time series input vectors and wind speed local time series input vectors in the sequence of ambient temperature local time series input vectors and the sequence of wind speed local time series input vectors to obtain a sequence of temperature-wind speed correlation matrices in the local time domain includes: Calculate the sample covariance matrix of the local time series input vector of the environmental temperature with respect to the local time series input vector of the wind speed using the following sample covariance correlation formula to obtain the temperature-wind speed correlation matrix within the local time domain; wherein, the sample covariance correlation formula is: ; wherein, is the local time series input vector of the environmental temperature, is the local time series input vector of the wind speed, is the temperature-wind speed correlation matrix within the local time domain.

4. The intelligent electric fan control method based on Internet of Things technology according to claim 3, wherein, The deep learning network model is a temperature-wind speed correlation feature extractor based on a convolutional neural network model.

5. The intelligent electric fan control method based on Internet of Things technology according to claim 4, characterized in that Performing implicit feature extraction on each local time-domain temperature-wind speed correlation matrix in the sequence of local time-domain temperature-wind speed correlation matrices using a deep learning network model to obtain a sequence of local time-domain temperature-wind speed correlation feature vectors, including: Passing each local time-domain temperature-wind speed correlation matrix in the sequence of local time-domain temperature-wind speed correlation matrices through the temperature-wind speed correlation feature extractor based on the convolutional neural network model to obtain the sequence of local time-domain temperature-wind speed correlation feature vectors.

6. The intelligent electric fan control method based on Internet of Things technology according to claim 5, characterized in that, Passing the optimized full-time-domain temperature-wind speed correlation feature vector through a classifier to obtain the fan control instruction, where the fan control instruction is used to indicate whether the power of the electric fan at the current time point can be reduced, including: Performing fully connected encoding on the optimized full-time-domain temperature-wind speed correlation feature vector using multiple fully connected layers of the classifier to obtain an encoded classification feature vector; and Passing the encoded classification feature vector through the Softmax classification function of the classifier to obtain a classification result.

7. An intelligent electric fan control system based on Internet of Things technology, characterized in that, Including: An environmental temperature time series acquisition module for acquiring the time series of environmental temperature collected by a temperature sensor; A wind speed time series acquisition module for acquiring the time series of wind speed collected by a wind speed sensor; A data regularization module for regularizing the time series of environmental temperature and the time series of wind speed to obtain a sequence of environmental temperature local time series input vectors and a sequence of wind speed local time series input vectors; A sample covariance correlation matrix calculation module for calculating the sample covariance correlation matrix between each pair of corresponding environmental temperature local time series input vectors and wind speed local time series input vectors in the sequence of environmental temperature local time series input vectors and the sequence of wind speed local time series input vectors to obtain a sequence of local time-domain temperature-wind speed correlation matrices; A fan control instruction determination module for determining a fan control instruction based on the implicit correlation features of the sequence of local time-domain temperature-wind speed correlation matrices; Wherein, determining the fan control instruction based on the implicit correlation features of the sequence of local time-domain temperature-wind speed correlation matrices includes: Performing implicit feature extraction on each local time-domain temperature-wind speed correlation matrix in the sequence of local time-domain temperature-wind speed correlation matrices using a deep learning network model to obtain a sequence of local time-domain temperature-wind speed correlation feature vectors; Passing the sequence of local time-domain temperature-wind speed correlation feature vectors through an importance assignment network based on an adaptive attention module to obtain a sequence of weighted local time-domain temperature-wind speed correlation feature vectors; Performing fusion optimization processing on the sequence of local time-domain temperature-wind speed correlation feature vectors and the sequence of weighted local time-domain temperature-wind speed correlation feature vectors to obtain an optimized full-time-domain temperature-wind speed correlation feature vector; Passing the optimized full-time-domain temperature-wind speed correlation feature vector through a classifier to obtain the fan control instruction, where the fan control instruction is used to indicate whether the power of the electric fan at the current time point can be reduced; Among them, obtaining a sequence of weighted temperature-wind speed correlation feature vectors in the local time domain by passing the sequence of temperature-wind speed correlation feature vectors in the local time domain through an importance assignment network based on an adaptive attention module includes: Process the sequence of temperature-wind speed correlation feature vectors in the local time domain with the following importance assignment formula to obtain the sequence of weighted temperature-wind speed correlation feature vectors in the local time domain; wherein, the importance assignment formula is: ; wherein, is the -th temperature-wind speed correlation feature vector in the sequence of temperature-wind speed correlation feature vectors in the local time domain, is pooling processing, is a pooling vector, is a weight matrix, is a bias vector, is activation processing, is an initial meta-weight feature vector, is the eigenvalue at the -th position in the initial meta-weight feature vector, is a corrected meta-weight feature vector, is the eigenvalue at the -th position in the corrected meta-weight feature vector, is the -th weighted temperature-wind speed correlation feature vector in the sequence of weighted temperature-wind speed correlation feature vectors in the local time domain, represents dot product processing; Among them, the sequence of the temperature-wind speed correlation feature vectors in the local time domain and the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain are fused and optimized, which is specifically expressed as: the following optimization formula is used to fuse and optimize the sequence of the temperature-wind speed correlation feature vectors in the local time domain and the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain to obtain the optimized full-time domain temperature-wind speed correlation feature vector; among them, the optimization formula is: ; where is the first feature vector obtained by concatenating the sequence of the temperature-wind speed correlation feature vectors in the local time domain, and is the second feature vector obtained by concatenating the sequence of the weighted temperature-wind speed correlation feature vectors in the local time domain, and respectively represent the squares of the 1-norm and 2-norm of the feature vector. The first feature vector and the second feature vector have the same feature vector length , and is the weight hyperparameter, and are the eigenvalues of the first feature vector and the second feature vector respectively, is the eigenvalue of the optimized full-time domain temperature-wind speed correlation feature vector, represents the logarithmic function with base 2.

8. The intelligent electric fan control system based on the Internet of Things technology according to claim 7, characterized in that, The data regularization module includes: A sequence segmentation unit, configured to perform sequence segmentation on the time series of the ambient temperature and the time series of the wind speed based on a predetermined time scale to obtain a sequence of local time series of the ambient temperature and a sequence of local time series of the wind speed; A sequence regularization unit, configured to regularize the sequence of local time series of the ambient temperature and the sequence of local time series of the wind speed respectively according to the time dimension to obtain a sequence of local time series input vectors of the ambient temperature and a sequence of local time series input vectors of the wind speed.

Images