User room temperature online monitoring system and method

Abstract

The present application relates to a user room temperature online monitoring system and method. The method includes: obtaining a time series of multiple temperature data collected by temperature sensors deployed at multiple locations in the monitored room, and constructing a spatial topological matrix of the temperature sensors at multiple locations in the monitored room, and using a data processing and analysis algorithm based on deep learning to perform temporal feature encoding optimization of the temperature data, and feature extraction of the spatial topological matrix, so as to automatically obtain the recommended opening value of the electric valve according to the characterization features between the temporal features of each temperature data and the features of the spatial topological matrix, so as to realize intelligent control of heating. In this way, by deploying temperature sensors at multiple locations indoors, comprehensive temperature data can be collected, overcoming the limitations of single sensor data collection.

Description

Technical Field

[0001] The present application relates to the technical field of room temperature monitoring, and in particular, to a system and method for online monitoring of user room temperature. Background Art

[0002] In key locations such as large commercial buildings, factory workshops, and data centers, accurate room temperature monitoring and heating control can not only ensure that equipment operates in the best condition and extend its service life, but also create a comfortable working environment for staff and improve production efficiency.

[0003] However, traditional room temperature monitoring systems often fail to capture subtle changes and spatial differences in indoor temperature due to their reliance on a single sensor, limiting their ability to fully understand the temperature field. This limitation prevents the system from responding quickly and accurately to temperature fluctuations and regional temperature differences. In addition, traditional systems usually use simple threshold-based control strategies, which are easy to implement but often fail to adapt to nonlinear changes in temperature when dealing with complex temperature time series dynamics, resulting in poor control effects.

[0004] Therefore, a user room temperature online monitoring system is desired. Summary of the invention

[0005] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present application provides a user room temperature online monitoring system, the system comprising: a temperature data acquisition module, used to acquire a time series of multiple temperature data collected by temperature sensors deployed at multiple positions in a monitored room; a temperature time series feature extraction module, used to arrange the time series of the multiple temperature data into multiple temperature time series input vectors according to the time dimension, and then pass the multiple temperature time series input vectors through a time series encoder to obtain multiple temperature time series feature vectors; a temperature time series feature optimization module, used to pass the multiple temperature time series feature vectors through a feature association optimization network based on a semantic metric of sequence endogenous correlation to obtain multiple optimized temperature time series feature vectors; a temperature topology module, used to extract the temperature time series feature vectors from the temperature series feature vectors through a feature association optimization network based on a semantic metric of sequence endogenous correlation to obtain multiple optimized temperature time series feature vectors; A temperature matrix construction module is used to construct a spatial topological matrix of temperature sensors at multiple locations in the monitored room; a temperature topological feature extraction module is used to obtain a spatial topological feature matrix by passing the spatial topological matrix through a topological association feature extractor; a global temperature timing characterization module is used to pass the multiple optimized temperature timing feature vectors and the spatial topological feature matrix through a global temperature timing characterizer based on a graph attention network to obtain a global temperature timing characterization matrix guided by a spatial topological structure; an electric valve opening recommendation module is used to obtain a control result based on the global temperature timing characterization matrix guided by the spatial topological structure, and the control result is used to represent a recommended value for the opening of the electric valve.

[0007] Optionally, the temperature time series feature extraction module is used to: arrange the time series of the multiple temperature data into the multiple temperature time series input vectors according to the time dimension, and then pass the multiple temperature time series input vectors through a time encoder based on a Bi-LSTM model to obtain the multiple temperature time series feature vectors.

[0008] Optionally, the temperature time series feature optimization module includes: a temperature time series semantic association score vector calculation unit, which is used to calculate the semantic association score vector between any two temperature time series feature vectors among the multiple temperature time series feature vectors to obtain a set of temperature time series semantic association score vectors; a temperature time series sequence intrinsic correlation global representation determination unit, which is used to determine the temperature time series sequence intrinsic correlation global representation vector based on the set of temperature time series semantic association score vectors; a temperature time series semantic association optimization factor calculation unit, which is used to calculate the association optimization factor of each temperature time series feature vector among the multiple temperature time series feature vectors based on the temperature time series sequence intrinsic correlation global representation vector to obtain the temperature time series semantic association score vector. A sequence of temperature time series semantic association optimization factors; an activation unit, used to input the sequence of temperature time series semantic association optimization factors into a Softmax activation function to obtain a sequence of temperature time series semantic association optimization weight factors; a vector weighting unit, used to use each temperature time series semantic association optimization weight factor in the sequence of temperature time series semantic association optimization weight factors as a weight, and weight each temperature time series feature vector in the multiple temperature time series feature vectors to obtain the multiple optimized temperature time series feature vectors; wherein the temperature time series sequence intrinsic correlation global representation determination unit is used to: calculate the mean vector of the set of temperature time series semantic association score vectors as the global representation vector of the temperature time series sequence intrinsic correlation.

[0009] Optionally, the temperature time series semantic association score vector calculation unit is used to: cascade any two temperature time series feature vectors among the multiple temperature time series feature vectors to obtain a set of temperature time series semantic cascade feature vectors; multiply each temperature time series semantic cascade feature vector in the set of temperature time series semantic cascade feature vectors with a temperature weight matrix and then add the resultant vectors positionally with the temperature bias vector to obtain the set of temperature time series semantic association score vectors.

[0010] Optionally, the temperature time series semantic association optimization factor calculation unit is used to: multiply the global representation vector of the intrinsic correlation of the temperature time series sequence with the temperature time series global weight matrix to obtain the temperature time series global representation weight vector; multiply each temperature time series feature vector of the multiple temperature time series feature vectors with the temperature time series local weight matrix to obtain a set of temperature time series local representation weight vectors; add each temperature time series local representation weight vector in the set of temperature time series local representation weight vectors with the temperature time series global representation weight vector and the temperature time series bias vector by position to obtain a set of temperature time series local semantic association vectors; input the set of temperature time series local semantic association vectors into the Sigmoid function to obtain a sequence of temperature time series semantic association optimization factors.

[0011] Optionally, the global temperature timing characterization module includes: an optimized temperature timing change feature calculation unit, which is used to respectively calculate the change feature vector of each optimized temperature timing feature vector in the multiple optimized temperature timing feature vectors to obtain a sequence of optimized temperature timing change feature vectors; an optimized temperature timing change energy measurement coefficient calculation unit, which is used to calculate the energy measurement coefficient of each optimized temperature timing change feature vector based on the mean and standard deviation of each optimized temperature timing change feature vector in the sequence of optimized temperature timing change feature vectors to obtain a sequence of optimized temperature timing change energy measurement coefficients; a weighted feature vector calculation unit, which is used to weight the sequence of optimized temperature timing change feature vectors using each optimized temperature timing change energy measurement coefficient in the sequence of optimized temperature timing change energy measurement coefficients as a weight. Weighting to obtain a sequence of weighted eigenvectors of optimized temperature timing changes; a vector sorting unit, used to arrange the sequence of weighted eigenvectors of optimized temperature timing changes into a global temperature timing characterization matrix; a bias matrix calculation unit, used to calculate the exponential function value of each eigenvalue in the global temperature timing characterization matrix with a natural constant as the base to obtain a global temperature timing characterization bias matrix; a spatial topology matrix calculation unit, used to calculate the exponential function value of each eigenvalue in the spatial topology feature matrix with a natural constant as the base to obtain a spatial topology bias feature matrix; a fully connected encoding unit, used to perform matrix multiplication of the global temperature timing characterization bias matrix and the spatial topology bias feature matrix, and then input them into a fully connected layer for fully connected encoding processing to obtain the global temperature timing characterization matrix guided by the spatial topology structure.

[0012] Optionally, the optimized temperature timing change characteristic calculation unit is used to: calculate the difference vector between the (i+1)th optimized temperature timing characteristic vector and the i-th optimized temperature timing characteristic vector in the multiple optimized temperature timing characteristic vectors to obtain a sequence of first optimized temperature timing differential characteristic vectors; calculate the difference vector between the i-th optimized temperature timing characteristic vector and the (i-1)-th optimized temperature timing characteristic vector in the multiple optimized temperature timing characteristic vectors to obtain a sequence of second optimized temperature timing differential characteristic vectors; respectively calculate the square of the characteristic value of each position of each first optimized temperature timing differential characteristic vector in the sequence of the first optimized temperature timing differential characteristic vector to obtain a first optimized temperature timing modulation characteristic vector sequence; respectively calculate the square of the eigenvalue of each position of the second optimized temperature timing differential eigenvector in the sequence of the second optimized temperature timing differential eigenvector to obtain a sequence of the second optimized temperature timing modulation eigenvector; calculate the positional sum of each group of corresponding first optimized temperature timing modulation eigenvectors and second optimized temperature timing modulation eigenvectors in the sequence of the first optimized temperature timing modulation eigenvector and the sequence of the second optimized temperature timing modulation eigenvector to obtain a sequence of optimized temperature timing modulation eigenvectors; calculate the square root of the eigenvalue of each position of each optimized temperature timing modulation eigenvector in the sequence of the optimized temperature timing modulation eigenvector to obtain a sequence of the optimized temperature timing change eigenvectors.

[0013] Optionally, the optimized temperature timing change energy measurement coefficient calculation unit is used to: respectively calculate the mean and standard deviation of the optimized temperature timing change feature vector to obtain the optimized temperature timing change feature mean and the optimized temperature timing change feature standard deviation; calculate the positional difference between the optimized temperature timing change feature vector and the optimized temperature timing change feature mean to obtain the optimized temperature timing change difference vector; calculate the fourth power of each eigenvalue in the optimized temperature timing change difference vector to obtain the optimized temperature timing change modulation difference vector; calculate the expected value of the optimized temperature timing change modulation difference vector to obtain the optimized temperature timing change expected value; divide the optimized temperature timing change expected value by the fourth power of the optimized temperature timing change feature standard deviation to obtain the optimized temperature timing change energy measurement coefficient.

[0014] Optionally, the electric valve opening recommendation module is used to: pass the temperature time series global characterization matrix guided by the spatial topological structure through a decoder-based electric valve controller to obtain a control result, and the control result is used to represent the recommended decoding value of the electric valve opening.

[0015] In a second aspect, the present application provides a method for online monitoring of user room temperature, the method comprising: obtaining a time series of multiple temperature data collected by temperature sensors deployed at multiple locations in a monitored room; arranging the time series of the multiple temperature data into multiple temperature time series input vectors according to the time dimension, and then passing the multiple temperature time series input vectors through a time series encoder to obtain multiple temperature time series feature vectors; passing the multiple temperature time series feature vectors through a feature association optimization network based on a semantic measure of sequence intrinsic correlation to obtain multiple optimized temperature time series feature vectors; constructing a spatial topological matrix of temperature sensors at multiple locations in the monitored room; passing the spatial topological matrix through a topological association feature extractor to obtain a spatial topological feature matrix; passing the multiple optimized temperature time series feature vectors and the spatial topological feature matrix through a global temperature time series characterizer based on a graph attention network to obtain a global temperature time series characterization matrix guided by a spatial topological structure; based on the global temperature time series characterization matrix guided by the spatial topological structure, obtaining a control result, and the control result is used to represent the recommended opening value of the electric valve.

[0016] The above technical solution is adopted to obtain the time series of multiple temperature data collected by temperature sensors deployed at multiple locations in the monitored room, and to construct the spatial topological matrix of the temperature sensors at multiple locations in the monitored room, and to use the data processing and analysis algorithm based on deep learning to optimize the temporal feature encoding of the temperature data, and to extract the features of the spatial topological matrix, so as to automatically obtain the recommended opening value of the electric valve according to the characterization features between the temporal features of each temperature data and the features of the spatial topological matrix, so as to realize the intelligent control of heating. In this way, by deploying temperature sensors at multiple locations indoors, comprehensive temperature data can be collected, overcoming the limitation of single sensor data collection.

[0017] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0019] In the attached picture: Figure 1 The figure is a block diagram of a user room temperature online monitoring system according to an exemplary embodiment.

[0020] Figure 2 The figure is a flow chart of a method for online monitoring of user room temperature according to an exemplary embodiment.

[0021] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment.

[0022] Figure 4 It is an application scenario diagram of a user room temperature online monitoring system according to an exemplary embodiment. DETAILED DESCRIPTION

[0023] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0024] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0025] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0026] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0027] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0028] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0029] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0030] Figure 1 FIG. 1 is a block diagram of a user room temperature online monitoring system according to an exemplary embodiment. Figure 1 As shown, the system 100 includes: a temperature data acquisition module 101, which is used to acquire a time series of multiple temperature data collected by temperature sensors deployed at multiple positions in the monitored room; a temperature time series feature extraction module 102, which is used to arrange the time series of the multiple temperature data into multiple temperature time series input vectors according to the time dimension, and then pass the multiple temperature time series input vectors through a time series encoder to obtain multiple temperature time series feature vectors; a temperature time series feature optimization module 103, which is used to pass the multiple temperature time series feature vectors through a feature association optimization network based on a sequence endogenous correlation semantic measurement to obtain multiple optimized temperature time series feature vectors; a temperature topology matrix construction module 104, Used to construct the spatial topological matrix of temperature sensors at multiple locations in the monitored room; a temperature topological feature extraction module 105, used to obtain a spatial topological feature matrix by passing the spatial topological matrix through a topological correlation feature extractor; a global temperature timing characterization module 106, used to obtain a global temperature timing characterization matrix guided by a spatial topological structure by passing the multiple optimized temperature timing feature vectors and the spatial topological feature matrix through a global temperature timing characterizer based on a graph attention network; an electric valve opening recommendation module 107, used to obtain a control result based on the global temperature timing characterization matrix guided by the spatial topological structure, and the control result is used to represent a recommended value for the opening of the electric valve.

[0031] In response to the above technical problems, the technical concept of the present application is to obtain the time series of multiple temperature data collected by temperature sensors deployed at multiple locations in the monitored room, and to construct a spatial topological matrix of the temperature sensors at multiple locations in the monitored room, and to use a data processing and analysis algorithm based on deep learning to optimize the temporal feature encoding of the temperature data, and extract the features of the spatial topological matrix, so as to automatically obtain the recommended value of the opening of the electric valve according to the characterization features between the temporal features of each temperature data and the features of the spatial topological matrix, so as to realize intelligent control of heating. In this way, by deploying temperature sensors at multiple locations indoors, comprehensive temperature data can be collected, overcoming the limitations of single sensor data collection. At the same time, the system can identify the patterns and trends of temperature changes in real time and respond to the dynamic changes of indoor temperature to timely adjust the valve opening to adapt to temperature fluctuations and regional temperature differences, thereby improving the automation level of the user's room temperature monitoring system.

[0032] The room temperature collector based on GRPS is a remote temperature measurement tool developed for indoor temperature remote monitoring applications. It is mainly used in applications that require remote temperature monitoring, such as heating, building automation, air conditioning, environmental control, industrial temperature control, agricultural greenhouse planting, medical treatment, scientific research, and energy conservation.

[0033] In the centralized heating system, each heat exchange station can be equipped with 3-5 remote room temperature collection points as needed, which can be placed at the near end, far end, or high and bottom floors of the building in unfavorable locations such as the heating area. It can also respond to user needs and monitor the room temperature of specific users for a long time, effectively improving the heating quality.

[0034] The household-based ring heating system is recommended to use the on-off time area method, which can achieve indoor temperature control and solve the problem of uneven room temperature in the building, such as hot upstairs and cold downstairs. The on-off time of the valve can also be used to apportion the heat in front of the building and realize heat metering.

[0035] Specific implementation plan: Electric valves that can automatically open and close according to room temperature control requirements are installed on the branch pipes of each household's heating system. The measured heat at the heat settlement point is apportioned among users based on the valve's connection time and the building area of ​​each household.

[0036] The specific approach is to install a room temperature on-off control valve on the branch line of each household for the indoor heating system connected horizontally, which is used to control the circulating water of the user to achieve room temperature control of the household. At the same time, a room temperature controller is placed in the representative room of each household to measure the indoor temperature and the temperature set by the user, and transmit these two temperature values ​​to the room temperature on-off control valve. The room temperature on-off control valve determines the on-off ratio of the on-off valve within a control cycle based on the difference between the measured room temperature and the set value, and controls the on-off of the on-off regulating valve according to this on-off ratio, so as to adjust the heat sent to the room. At the same time, it records and counts the on-time of each household's on-off control valve, and apportions the heat of the entire building according to the cumulative on-time of each household combined with the heating area.

[0037] Specifically, in the technical solution of the present application, first, a time series of multiple temperature data collected by temperature sensors deployed at multiple locations in the monitored room is obtained. Next, considering that each time series of the multiple temperature data has temporal feature information in the time dimension, in order to better understand and analyze the temperature change trend between each time point in the future, in the technical solution of the present application, the time series of the multiple temperature data are arranged into multiple temperature time series input vectors according to the time dimension. Furthermore, in order to capture and mine the implicit time series features and the correlation between each other in each temperature time series input vector, in the technical solution of the present application, the multiple temperature time series input vectors are passed through a time series encoder based on a Bi-LSTM model to obtain multiple temperature time series feature vectors.

[0038] In one embodiment of the present application, the temperature time series feature extraction module is used to: arrange the time series of the multiple temperature data into the multiple temperature time series input vectors according to the time dimension, and then pass the multiple temperature time series input vectors through a time encoder based on a Bi-LSTM model to obtain the multiple temperature time series feature vectors.

[0039] Next, considering the different contributions and correlations of different temperature time series features in the multiple temperature time series feature vectors, the features between these different features have an important influence on the opening control of the electric valve. However, the use of a time series encoder based on the Bi-LSTM model for time series encoding can only capture the characteristics of the temperature time series data itself at each position, and cannot reflect the relationship between different temperature sensors. Based on this, in the technical solution of the present application, a feature association optimization network based on the semantic measurement of endogenous correlation of the sequence is further introduced to process the multiple temperature time series feature vectors to mine the implicit semantic associations of the temperature between each temperature sensor and further optimize the temperature time series feature vector. In detail, first, by calculating the cascade vector between any two features in the multiple temperature time series feature vectors to obtain a set of temperature time series semantic cascade feature vectors to reveal the mutual influence and interaction between different sensor temperatures, the model's perception of the overall dynamic characteristics of the temperature field is enhanced. Then, each cascade feature vector is semantically associated to learn and capture the implicit semantic relationship between the features to obtain a set of temperature time series semantic association score vectors. Next, the mean vector of the set of temperature time series semantic association score vectors is calculated to aggregate the semantic association features of all sensors, thereby extracting the semantic association features of the temperature field as a whole to obtain the global representation vector of the intrinsic correlation of the temperature time series sequence. Furthermore, the association optimization factor of each feature is calculated based on each temperature time series feature vector and the global representation vector of the intrinsic correlation of the temperature time series sequence, highlighting those features that are highly correlated with the temperature field dynamics, and giving these highly correlated features a greater weight to improve the model's perception of key features. Subsequently, these optimization factors are converted into weight values ​​through the Softmax activation function to weight each temperature time series feature vector to better capture the intrinsic connection between features and the dynamic characteristics of the overall temperature field to obtain the final multiple optimized temperature time series feature vectors, thereby improving the overall performance of the model.

[0040] In one embodiment of the present application, the temperature timing feature optimization module includes: a temperature timing semantic association score vector calculation unit, which is used to calculate the semantic association score vector between any two temperature timing feature vectors among the multiple temperature timing feature vectors to obtain a set of temperature timing semantic association score vectors; a temperature timing sequence intrinsic correlation global representation determination unit, which is used to determine the temperature timing sequence intrinsic correlation global representation vector based on the set of temperature timing semantic association score vectors; a temperature timing sequence semantic association optimization factor calculation unit, which is used to calculate the association optimization factor of each temperature timing feature vector among the multiple temperature timing feature vectors based on the temperature timing sequence intrinsic correlation global representation vector to obtain the temperature timing sequence intrinsic correlation global representation vector. A sequence of temperature time series semantic association optimization factors; an activation unit, used to input the sequence of temperature time series semantic association optimization factors into a Softmax activation function to obtain a sequence of temperature time series semantic association optimization weight factors; a vector weighting unit, used to use each temperature time series semantic association optimization weight factor in the sequence of temperature time series semantic association optimization weight factors as a weight, and weight each temperature time series feature vector in the multiple temperature time series feature vectors to obtain the multiple optimized temperature time series feature vectors; wherein the temperature time series sequence intrinsic correlation global representation determination unit is used to: calculate the mean vector of the set of temperature time series semantic association score vectors as the global representation vector of the temperature time series sequence intrinsic correlation.

[0041] Furthermore, in one embodiment of the present application, the temperature time series semantic association score vector calculation unit is used to: cascade any two temperature time series feature vectors among the multiple temperature time series feature vectors to obtain a set of temperature time series semantic cascade feature vectors; multiply each temperature time series semantic cascade feature vector in the set of temperature time series semantic cascade feature vectors with a temperature weight matrix and then add the resultant vectors positionally with the temperature bias vector to obtain the set of temperature time series semantic association score vectors.

[0042] Furthermore, in one embodiment of the present application, the temperature-time series semantic association optimization factor calculation unit is used to: multiply the global representation vector of the intrinsic correlation of the temperature time series sequence with the temperature time series global weight matrix to obtain the temperature time series global representation weight vector; multiply each temperature time series feature vector of the multiple temperature time series feature vectors with the temperature time series local weight matrix to obtain a set of temperature time series local representation weight vectors; add each temperature time series local representation weight vector in the set of temperature time series local representation weight vectors with the temperature time series global representation weight vector and the temperature time series bias vector by position to obtain a set of temperature time series local semantic association vectors; input the set of temperature time series local semantic association vectors into the Sigmoid function to obtain a sequence of temperature time series semantic association optimization factors.

[0043] Specifically, the multiple temperature time series feature vectors are processed by a feature association optimization network based on a semantic measurement of endogenous correlation of the sequence using the following feature association optimization formula to obtain the multiple optimized temperature time series feature vectors; wherein the feature association optimization formula is: ;in, are the multiple temperature time series feature vectors, represents the number of feature vectors in the plurality of temperature time series feature vectors, and are respectively the first and second temperature time series feature vectors of the plurality of temperature time series feature vectors. and The temperature time series characteristic vector, For cascade processing operations, is the temperature weight matrix, is the temperature bias vector, is the first temperature time series semantic association score vector in the set Temperature temporal semantic association score vector, is the number of vectors in the set of temperature time series semantic association score vectors, is the global representation vector of the endogenous correlation of the temperature time series, and They are the temperature time series global weight matrix and the temperature time series local weight matrix, is the temperature timing bias vector, is the sigmoid function, is the Softmax function, The first in the sequence of optimizing weight factors for the temperature time series semantic association Temperature time series semantic association optimization weight factor, is the first one of the plurality of optimized temperature timing feature vectors An optimized temperature timing feature vector.

[0044] Then, considering that the temperature field has certain correlation and dependence in space, that is, there is mutual influence between temperature sensors at different positions. Based on this, in the technical solution of the present application, a spatial topological matrix of temperature sensors at multiple positions in the monitored room is constructed, so that the model can learn the spatial structure information of the temperature field, thereby highlighting the correlation of the temperature space between adjacent sensors, thereby enhancing the understanding of the overall dynamics of the temperature field.

[0045] Next, considering that the spatial correlation of the temperature field often has non-local dependencies across scales, ordinary convolutional networks may be difficult to effectively model. Therefore, in order to further capture the temperature spatial structure distribution characteristic information contained in the spatial topological matrix, in the technical solution of the present application, the spatial topological matrix is ​​passed through a topological correlation feature extractor based on a dilated convolutional neural network model to obtain a spatial topological feature matrix. In other words, dilated convolution can increase the receptive field and capture a wider range of spatial dependencies without increasing the number of parameters, which can better model the non-local characteristics of the temperature field.

[0046] Further, considering that the multiple optimized temperature time series feature vectors capture the dynamic change characteristics of each temperature field in the time dimension, the spatial topological feature matrix characterizes the relevant structural characteristics of the temperature field in the spatial dimension. And different node features have different weight influences on the overall spatial feature distribution. Based on this, in order to enhance the understanding of the overall behavior of the temperature field by the characteristics of those key temperature nodes, it is necessary to aggregate the characteristics of the nodes around the node to improve the performance of temperature prediction and control. In the technical solution of the present application, the multiple optimized temperature time series feature vectors and the spatial topological feature matrix are fused and characterized by using a global temperature time series characterizer based on a graph attention network to obtain a global temperature time series characterization matrix guided by a spatial topological structure. Specifically, first, by calculating the differential features between the two adjacent time points in the optimized temperature time series feature vector at the current time point, the rate of change and the trend of change of temperature in time are reflected, so as to enhance the model's understanding of the time evolution law of the temperature field, and obtain a sequence of optimized temperature time series change feature vectors based on the differential features. Then, based on the mean and standard deviation of each optimized temperature time series change feature vector, the energy measurement coefficient is calculated to quantify the energy distribution of temperature and highlight the key change points or key areas in the temperature field. After that, each characteristic energy measurement coefficient is used as a weight to weight each optimized temperature time series change feature vector, so as to give them a higher weight, highlight and refine the key temperature time series change feature information, and remove the influence of those irrelevant or redundant temperature node features. Next, each weighted vector is arranged into a matrix and associated with the spatial topological feature matrix and input into the fully connected layer for encoding processing to associate and fuse the time series features and spatial topological features, and obtain a global temperature time series representation matrix guided by a spatial topological structure that contains both time and space information. This can fully perceive the spatiotemporal dynamic characteristics of the temperature field, thereby improving the ability to understand the overall behavior of the temperature field and obtaining a richer and more effective temperature field representation.

[0047] In one embodiment of the present application, the global temperature timing characterization module includes: an optimized temperature timing change feature calculation unit, which is used to respectively calculate the change feature vectors of each optimized temperature timing feature vector in the multiple optimized temperature timing feature vectors to obtain a sequence of optimized temperature timing change feature vectors; an optimized temperature timing change energy measurement coefficient calculation unit, which is used to calculate the energy measurement coefficient of each optimized temperature timing change feature vector based on the mean and standard deviation of each optimized temperature timing change feature vector in the sequence of optimized temperature timing change feature vectors to obtain a sequence of optimized temperature timing change energy measurement coefficients; a weighted feature vector calculation unit, which is used to use each optimized temperature timing change energy measurement coefficient in the sequence of optimized temperature timing change energy measurement coefficients as a weight to weight the optimized temperature timing change feature vector. The sequence is weighted to obtain a sequence of weighted eigenvectors of optimized temperature time series changes; a vector sorting unit is used to arrange the sequence of weighted eigenvectors of optimized temperature time series changes into a global temperature time series characterization matrix; a bias matrix calculation unit is used to calculate the exponential function value of each eigenvalue in the global temperature time series characterization matrix with a natural constant as the base to obtain a global temperature time series characterization bias matrix; a spatial topology matrix calculation unit is used to calculate the exponential function value of each eigenvalue in the spatial topology feature matrix with a natural constant as the base to obtain a spatial topology bias feature matrix; a fully connected encoding unit is used to perform matrix multiplication of the global temperature time series characterization bias matrix and the spatial topology bias feature matrix and then input them into a fully connected layer for fully connected encoding processing to obtain a global temperature time series characterization matrix guided by the spatial topology structure.

[0048] Furthermore, in one embodiment of the present application, the optimized temperature timing change characteristic calculation unit is used to: calculate the difference vector between the (i+1)th optimized temperature timing characteristic vector and the i-th optimized temperature timing characteristic vector in the multiple optimized temperature timing characteristic vectors to obtain a sequence of first optimized temperature timing differential characteristic vectors; calculate the difference vector between the i-th optimized temperature timing characteristic vector and the (i-1)-th optimized temperature timing characteristic vector in the multiple optimized temperature timing characteristic vectors to obtain a sequence of second optimized temperature timing differential characteristic vectors; respectively calculate the square of the characteristic value of each position of each first optimized temperature timing differential characteristic vector in the sequence of the first optimized temperature timing differential characteristic vector to obtain a first optimized temperature timing differential characteristic vector. A sequence of modulation feature vectors; respectively calculating the square of the eigenvalue of each position of each second optimized temperature timing differential feature vector in the sequence of the second optimized temperature timing differential feature vector to obtain a sequence of second optimized temperature timing modulation feature vectors; calculating the positional sum of each group of corresponding first optimized temperature timing modulation feature vectors and second optimized temperature timing modulation feature vectors in the sequence of the first optimized temperature timing modulation feature vectors and the sequence of the second optimized temperature timing modulation feature vectors to obtain a sequence of optimized temperature timing modulation feature vectors; calculating the square root of the eigenvalue of each position of each optimized temperature timing modulation feature vector in the sequence of the optimized temperature timing modulation feature vectors to obtain a sequence of the optimized temperature timing change feature vectors.

[0049] Furthermore, in one embodiment of the present application, the optimized temperature timing change energy measurement coefficient calculation unit is used to: respectively calculate the mean and standard deviation of the optimized temperature timing change feature vector to obtain the optimized temperature timing change feature mean and the optimized temperature timing change feature standard deviation; calculate the positional difference between the optimized temperature timing change feature vector and the optimized temperature timing change feature mean to obtain the optimized temperature timing change difference vector; calculate the fourth power of each eigenvalue in the optimized temperature timing change difference vector to obtain the optimized temperature timing change modulation difference vector; calculate the expected value of the optimized temperature timing change modulation difference vector to obtain the optimized temperature timing change expected value; divide the optimized temperature timing change expected value by the fourth power of the optimized temperature timing change feature standard deviation to obtain the optimized temperature timing change energy measurement coefficient.

[0050] Specifically, the multiple optimized temperature time series feature vectors and the spatial topology feature matrix are processed by a global temperature time series characterizer based on a graph attention network using the following characterization formula to obtain a global temperature time series characterization matrix guided by the spatial topology structure; wherein the characterization formula is: ;in, , and The first of the plurality of optimized temperature timing feature vectors is , and The optimized temperature time series feature vector of each position, For the said and The difference vector between the optimized temperature time series feature vectors, For the and The difference vector between the optimized temperature timing feature vectors, is the first in the sequence of the optimized temperature time series variation feature vector An optimized temperature time series variation feature vector, For the said The eigenvalues ​​of each position of the eigenvector of the optimized temperature time series change, and They are The mean and variance of the optimized temperature time series variation feature vector, To calculate the expected value, For the The energy metric coefficient of the optimized temperature time series variation feature vector, For each optimized temperature time series variation feature vector, The first in the sequence of weighted characteristic vectors of the optimized temperature time series variation An optimized weighted eigenvector of temperature time series variation, The first in the sequence of weighted characteristic vectors of the optimized temperature time series variation An optimized weighted eigenvector of temperature time series variation, is the matrix arrangement of the vector sequence, is the global temperature timing characterization matrix, is matrix multiplication, is the spatial topological feature matrix, Represents an exponential function with the natural constant e as the base is a fully connected encoding operation, It is the global temperature time series characterization matrix guided by the spatial topological structure.

[0051] Then, the temperature time series global characterization matrix guided by the spatial topological structure is passed through the electric valve controller based on the decoder to obtain the control result, and the control result is used to represent the recommended decoding value of the opening of the electric valve. That is, the global temperature time series characterization matrix guided by the spatial topological structure obtained by attention fusion of the multiple optimized temperature time series feature vectors and the spatial topological feature matrix is ​​decoded to automatically obtain the recommended value of the opening of the electric valve. In this way, by deploying temperature sensors at multiple locations indoors, comprehensive temperature data can be collected, overcoming the limitations of single sensor data collection. At the same time, the system can identify the pattern and trend of temperature changes in real time and respond to the dynamic changes of indoor temperature to timely adjust the valve opening to adapt to temperature fluctuations and regional temperature differences, thereby improving the automation level of the user's room temperature monitoring system.

[0052] In one embodiment of the present application, the electric valve opening recommendation module is used to: pass the temperature timing global characterization matrix guided by the spatial topological structure through a decoder-based electric valve controller to obtain a control result, and the control result is used to represent the recommended decoding value of the electric valve opening.

[0053] In particular, considering that the multiple optimized temperature time series feature vectors respectively represent the associated optimization features of the temperature at each position based on the semantic measurement of the intrinsic correlation of the short-range and long-range bidirectional time series distribution, when performing topological association based on the spatial topology of each position, the global temperature time series representation matrix guided by the spatial topological structure will also have a diversified set expression distribution of temperature time series topological association characteristics based on the expression of the graph attention network. Therefore, it is expected to improve the overall expression effect of the temperature time series global representation matrix guided by the spatial topological structure based on the diversified set distribution, thereby improving the accuracy of the control results obtained by the decoder-based electric valve controller.

[0054] In a preferred example, passing the global characterization matrix of temperature timing guided by the spatial topological structure through a decoder-based electric valve controller to obtain a control result includes: optimizing the global characterization matrix of temperature timing guided by the spatial topological structure to obtain an optimized global characterization matrix of temperature timing guided by the spatial topological structure; passing the optimized global characterization matrix of temperature timing guided by the spatial topological structure through a decoder-based electric valve controller to obtain a control result.

[0055] Based on this, the present application optimizes the temperature time series global characterization matrix under the guidance of the spatial topological structure to obtain an optimized temperature time series global characterization matrix under the guidance of the spatial topological structure, including the following steps: calculating the sum of the absolute values ​​of each eigenvalue of the temperature time series global characterization matrix under the guidance of the spatial topological structure The square root of the sum of squares ,Right now: ;in, represents the global representation matrix of temperature time series under the guidance of the spatial topological structure. The eigenvalues ​​at the positions, represents the characteristic matrix scale of the global temperature time series representation matrix guided by the spatial topological structure, represents the sum of the absolute values ​​of the eigenvalues ​​of the global characterization matrix of the temperature time series under the guidance of the spatial topological structure, It represents the square root of the sum of squares of the eigenvalues ​​of the global characterization matrix of the temperature time series under the guidance of the spatial topological structure.

[0056] Each eigenvalue of the global characterization matrix of temperature time series guided by the spatial topological structure is subtracted from After that, take the absolute value and multiply it by the characteristic matrix scale of the global characterization matrix of temperature time series guided by the spatial topological structure and The reciprocal of the product is taken to obtain the global intermediate characteristic matrix of the temperature time series under the guidance of the first spatial topological structure, and each eigenvalue of the global characterization matrix of the temperature time series under the guidance of the spatial topological structure is subtracted from After that, take the absolute value and multiply it by the feature matrix scale The square root of The reciprocal of , and take the reciprocal of the product to obtain the global intermediate characteristic matrix of the temperature time series guided by the second spatial topological structure, that is: ;in, represents the global representation matrix of temperature time series under the guidance of the spatial topological structure. The eigenvalues ​​at the positions, represents the characteristic matrix scale of the global temperature time series representation matrix guided by the spatial topological structure, represents the sum of the absolute values ​​of the eigenvalues ​​of the global characterization matrix of the temperature time series under the guidance of the spatial topological structure, represents the square root of the sum of squares of the eigenvalues ​​of the global characterization matrix of the temperature time series guided by the spatial topological structure, represents the first spatial topological structure-guided global intermediate characteristic matrix of temperature time series. represents the global intermediate characteristic matrix of temperature time series guided by the second spatial topological structure. The feature value at each position.

[0057] Finally, the global intermediate feature matrix of temperature time series guided by the first spatial topology structure is subtracted from the dot product of the global intermediate feature matrix of temperature time series guided by the second spatial topology structure and the weighted hyperparameter to obtain the optimized global representation matrix of temperature time series guided by the spatial topology structure.

[0058] Thus, by taking the difference of the eigenvalue of the global representation matrix of the temperature time series under the guidance of the spatial topological structure relative to the difference of the modulation representation of the overall feature set of the global representation matrix of the temperature time series under the guidance of the spatial topological structure as the semantic change intensity information, and performing the phase-like transformation corresponding to the position-based intensity modulation through the different modulation representation forms, so as to perform the spatial translation operation based on alternating stacking under the scale balance of the feature set of the global representation matrix of the temperature time series under the guidance of the spatial topological structure, so as to make the aggregation enhancement of the semantic change phase perception able to improve the axial aggregation receptive field along the feature aggregation direction, thereby improving the perception effect of the aggregation semantics of the global representation matrix of the temperature time series under the guidance of the spatial topological structure for the change of detailed semantics, so as to improve the overall expression effect of the global representation matrix of the temperature time series under the guidance of the spatial topological structure, and improve the accuracy of the control result obtained by the electric valve controller based on the decoder. In this way, by deploying temperature sensors at multiple locations indoors, comprehensive temperature data can be collected, overcoming the limitation of single sensor data collection. At the same time, the system can identify the pattern and trend of temperature change in real time and respond to the dynamic change of indoor temperature to adjust the valve opening in time to adapt to temperature fluctuations and regional temperature differences, thereby improving the automation level of the user room temperature monitoring system.

[0059] In summary, the above scheme is adopted to obtain the time series of multiple temperature data collected by temperature sensors deployed at multiple locations in the monitored room, and to construct the spatial topological matrix of the temperature sensors at multiple locations in the monitored room, and to use the data processing and analysis algorithm based on deep learning to optimize the temporal feature encoding of the temperature data and extract the features of the spatial topological matrix, so as to automatically obtain the recommended opening value of the electric valve according to the characterization features between the temporal features of each temperature data and the features of the spatial topological matrix, so as to realize the intelligent control of heating. In this way, by deploying temperature sensors at multiple locations indoors, comprehensive temperature data can be collected, overcoming the limitations of single sensor data collection.

[0060] Figure 2 is a flow chart of a method for online monitoring of user room temperature according to an exemplary embodiment. Figure 2As shown, the method includes: step S201, obtaining a time series of multiple temperature data collected by temperature sensors deployed at multiple positions in the monitored room; step S202, arranging the time series of the multiple temperature data into multiple temperature time series input vectors according to the time dimension, and passing the multiple temperature time series input vectors through a time series encoder to obtain multiple temperature time series feature vectors; step S203, passing the multiple temperature time series feature vectors through a feature association optimization network based on a sequence endogenous correlation semantic measurement to obtain multiple optimized temperature time series feature vectors; step S204, constructing a spatial topological matrix of the temperature sensors at multiple positions in the monitored room; step S205, passing the spatial topological matrix through a topological association feature extractor to obtain a spatial topological feature matrix; step S206, passing the multiple optimized temperature time series feature vectors and the spatial topological feature matrix through a global temperature time series characterizer based on a graph attention network to obtain a global temperature time series characterization matrix guided by a spatial topological structure; step S207, based on the global temperature time series characterization matrix guided by the spatial topological structure, obtaining a control result, and the control result is used to represent the recommended opening value of the electric valve.

[0061] Reference below Figure 3 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0062] like Figure 3 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0063] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0064] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

[0065] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0066] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0067] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0068] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0069] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0070] The modules involved in the embodiments described in this application can be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the test parameter acquisition module can also be described as a "module for acquiring device test parameters corresponding to the target device".

[0071] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0072] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0073] Figure 4 FIG. 1 is an application scenario diagram of a user room temperature online monitoring system according to an exemplary embodiment. Figure 4 As shown, in this application scenario, first, a time series of multiple temperature data collected by temperature sensors deployed at multiple locations in the monitored room is obtained (for example, Figure 4 C1 as shown in the figure); constructing a spatial topological matrix of temperature sensors at multiple locations in the monitored room (for example, Figure 4 Then, the time series and spatial topology matrix of the acquired temperature data are input to a server (for example, Figure 4 In S) shown in , the server can process the time series of the temperature data and the spatial topology matrix based on the user room temperature online monitoring algorithm to obtain a control result, and the control result is used to represent the recommended opening value of the electric valve.

[0074] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0075] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0076] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. A user room temperature online monitoring system, characterized in that: include: A temperature data acquisition module is used to acquire a time series of multiple temperature data collected by temperature sensors deployed at multiple locations in the monitored room; A temperature time series feature extraction module, configured to arrange the time series of the plurality of temperature data into a plurality of temperature time series input vectors according to the time dimension, and pass the plurality of temperature time series input vectors through a time series encoder to obtain a plurality of temperature time series feature vectors; A temperature time series feature optimization module, used for obtaining a plurality of optimized temperature time series feature vectors by subjecting the plurality of temperature time series feature vectors to a feature association optimization network based on a semantic measurement of endogenous correlation of a sequence; A temperature topology matrix building module is used to build a spatial topology matrix of temperature sensors at multiple locations in the monitored room; A temperature topological feature extraction module, used for obtaining a spatial topological feature matrix by passing the spatial topological matrix through a topological correlation feature extractor; A global temperature timing characterization module, used for passing the multiple optimized temperature timing feature vectors and the spatial topology feature matrix through a global temperature timing characterizer based on a graph attention network to obtain a global temperature timing characterization matrix guided by a spatial topology structure; An electric valve opening recommendation module, used to obtain a control result based on the global temperature time series characterization matrix guided by the spatial topological structure, wherein the control result is used to represent a recommended opening value of the electric valve; Wherein, the temperature timing characteristic optimization module includes: A temperature time series semantic association score vector calculation unit, used to calculate a semantic association score vector between any two temperature time series feature vectors among the plurality of temperature time series feature vectors to obtain a set of temperature time series semantic association score vectors; A temperature time series intrinsic correlation global representation determination unit, used to determine a temperature time series intrinsic correlation global representation vector based on a set of temperature time series semantic association score vectors; A temperature time series semantic association optimization factor calculation unit, used to calculate the association optimization factor of each temperature time series feature vector in the multiple temperature time series feature vectors based on the global representation vector of the endogenous correlation of the temperature time series sequence to obtain a sequence of temperature time series semantic association optimization factors; An activation unit, used for inputting the sequence of temperature time series semantic association optimization factors into a Softmax activation function to obtain a sequence of temperature time series semantic association optimization weight factors; A vector weighting unit, configured to weight each temperature time series feature vector in the plurality of temperature time series feature vectors by using each temperature time series semantic association optimization weight factor in the sequence of the temperature time series semantic association optimization weight factors as a weight, so as to obtain the plurality of optimized temperature time series feature vectors; The temperature time series intrinsic correlation global representation determination unit is used to: calculate the mean vector of the set of temperature time series semantic association score vectors as the temperature time series intrinsic correlation global representation vector; Wherein, the global temperature timing characterization module includes: An optimized temperature time series change characteristic calculation unit, used to respectively calculate the change characteristic vector of each optimized temperature time series characteristic vector in the plurality of optimized temperature time series characteristic vectors to obtain a sequence of optimized temperature time series change characteristic vectors; An optimized temperature time series change energy metric coefficient calculation unit, used to calculate the energy metric coefficient of each optimized temperature time series change feature vector in the sequence of optimized temperature time series change feature vectors based on the mean and standard deviation of each optimized temperature time series change feature vector to obtain a sequence of optimized temperature time series change energy metric coefficients; A weighted feature vector calculation unit, configured to weight the sequence of optimized temperature time series change feature vectors by using each optimized temperature time series change energy measurement coefficient in the sequence of optimized temperature time series change energy measurement coefficients as a weight to obtain a sequence of optimized temperature time series change weighted feature vectors; A vector sorting unit, used for arranging the sequence of the optimized temperature time series change weighted characteristic vectors into a global temperature time series characterization matrix; A bias matrix calculation unit, used for calculating an exponential function value of each eigenvalue in the global temperature time series characterization matrix with a natural constant as the base, so as to obtain a global temperature time series characterization bias matrix; A spatial topology matrix calculation unit, used for calculating an exponential function value of each eigenvalue in the spatial topology feature matrix as an exponential power with a natural constant as the base to obtain a spatial topology bias feature matrix; A fully connected encoding unit, used for performing matrix multiplication on the global temperature time series characterization bias matrix and the spatial topology bias feature matrix, and then inputting the matrix into a fully connected layer for fully connected encoding processing to obtain a global temperature time series characterization matrix guided by the spatial topology structure; Wherein, the optimization temperature time series change energy measurement coefficient calculation unit is used for: Respectively calculating the mean and standard deviation of the optimized temperature time series change feature vector to obtain the optimized temperature time series change feature mean and the optimized temperature time series change feature standard deviation; Calculate the position difference between the optimized temperature time series change feature vector and the optimized temperature time series change feature mean to obtain the optimized temperature time series change difference vector; Calculating the fourth power of each eigenvalue in the optimized temperature time series variation difference vector to obtain an optimized temperature time series variation modulation difference vector; Calculating the expected value of the optimized temperature timing change modulation difference vector to obtain the optimized temperature timing change expected value; The expected value of the optimized temperature time series change is divided by the fourth power of the standard deviation of the optimized temperature time series change characteristic to obtain the optimized temperature time series change energy measurement coefficient.

2. The user room temperature online monitoring system according to claim 1, characterized in that: The temperature time series feature extraction module is used to: after arranging the time series of the multiple temperature data into the multiple temperature time series input vectors according to the time dimension, the multiple temperature time series input vectors are passed through a time series encoder based on a Bi-LSTM model to obtain the multiple temperature time series feature vectors.

3. The user room temperature online monitoring system according to claim 2 is characterized in that: The temperature time series semantic association score vector calculation unit is used to: Cascading any two of the plurality of temperature time series feature vectors to obtain a set of temperature time series semantic cascade feature vectors; Each temperature time series semantic cascade feature vector in the set of temperature time series semantic cascade feature vectors is multiplied by a temperature weight matrix and then added to the temperature bias vector by position to obtain the set of temperature time series semantic association score vectors.

4. The user room temperature online monitoring system according to claim 3, characterized in that: The temperature time series semantic association optimization factor calculation unit is used to: Multiplying the global representation vector of the endogenous correlation of the temperature time series with the global weight matrix of the temperature time series to obtain the global representation weight vector of the temperature time series; Multiplying each temperature time series feature vector of the plurality of temperature time series feature vectors by a temperature time series local weight matrix to obtain a set of temperature time series local representation weight vectors; Adding each temperature time series local representation weight vector in the set of temperature time series local representation weight vectors to the temperature time series global representation weight vector and the temperature time series offset vector by position to obtain a set of temperature time series local semantic association vectors; The set of the temperature time series local semantic association vectors is input into a Sigmoid function to obtain a sequence of the temperature time series semantic association optimization factors.

5. The user room temperature online monitoring system according to claim 4, characterized in that: The optimized temperature time series variation characteristic calculation unit is used for: Calculating a difference vector between the (i+1)th optimized temperature timing feature vector and the i-th optimized temperature timing feature vector among the plurality of optimized temperature timing feature vectors to obtain a sequence of first optimized temperature timing difference feature vectors; Calculating a difference vector between an i-th optimized temperature timing feature vector and an (i-1)-th optimized temperature timing feature vector among the plurality of optimized temperature timing feature vectors to obtain a sequence of second optimized temperature timing difference feature vectors; Respectively calculating the square of the eigenvalue of each position of each first optimized temperature time series differential eigenvector in the sequence of the first optimized temperature time series differential eigenvectors to obtain a sequence of first optimized temperature time series modulation eigenvectors; Respectively calculating the square of the eigenvalue of each position of each second optimized temperature time series difference eigenvector in the sequence of the second optimized temperature time series difference eigenvector to obtain a sequence of second optimized temperature time series modulation eigenvectors; Calculating the positional sum of each group of corresponding first optimized temperature timing modulation feature vectors and second optimized temperature timing modulation feature vectors in the sequence of the first optimized temperature timing modulation feature vectors and the sequence of the second optimized temperature timing modulation feature vectors to obtain a sequence of optimized temperature timing modulation feature vectors; The square root of each position characteristic value of each optimized temperature timing modulation characteristic vector in the sequence of the optimized temperature timing modulation characteristic vector is calculated to obtain the sequence of the optimized temperature timing change characteristic vector.

6. The user room temperature online monitoring system according to claim 5, characterized in that: The electric valve opening recommendation module is used to: pass the temperature time series global characterization matrix guided by the spatial topological structure through a decoder-based electric valve controller to obtain a control result, and the control result is used to represent the recommended decoding value of the electric valve opening.

7. A method for online monitoring of user room temperature, using the online monitoring system for user room temperature according to claim 1, characterized in that: include: Acquire a time series of a plurality of temperature data collected by temperature sensors deployed at a plurality of positions in a monitored room; After arranging the time series of the plurality of temperature data into a plurality of temperature time series input vectors according to the time dimension, the plurality of temperature time series input vectors are passed through a time series encoder to obtain a plurality of temperature time series feature vectors; The plurality of temperature time series feature vectors are subjected to a feature association optimization network based on a semantic measurement of endogenous correlation of the sequence to obtain a plurality of optimized temperature time series feature vectors; Constructing a spatial topological matrix of temperature sensors at multiple locations in the monitored room; Passing the spatial topological matrix through a topological correlation feature extractor to obtain a spatial topological feature matrix; The multiple optimized temperature time series feature vectors and the spatial topology feature matrix are passed through a global temperature time series characterizer based on a graph attention network to obtain a global temperature time series characterization matrix guided by a spatial topology structure; Based on the global temperature time series characterization matrix guided by the spatial topological structure, a control result is obtained, and the control result is used to represent the recommended opening value of the electric valve.

Images