Air conditioner self-adaptive temperature adjusting system based on environment perception

Abstract

The invention provides an air conditioner self-adaptive temperature adjusting system based on environment perception, and relates to the technical field of air conditioning, the system comprises a design drawing acquisition module, a sensor array generation module, a temperature sensor array generation module, a state identification acquisition module, a state identification module, a state identification module and a state identification module, m partition device state identifiers are obtained, and M fusion subspaces are obtained by a space fusion module; the temperature data extraction module is used for obtaining M fusion temperature value sets; the parameter configuration module is used for performing parameter configuration; and the temperature adjusting module is used for adjusting the temperature of the target area. According to the invention, the problems of excessive consumption and waste of energy resources, lack of flexibility and incapability of performing real-time temperature adjustment according to dynamic changes of spatial layout due to insufficient control precision in the prior art can be solved, real-time monitoring and temperature control of the spatial structure of the target area are realized, and the comfort level of the space and the energy use efficiency are improved.

Description

Technical Field

[0001] This application relates to the technical field of air conditioning, and particularly to an air conditioner adaptive temperature regulation system based on environmental perception. Background Art

[0002] With the development of social economy and the improvement of people's living standards, air conditioners, as important devices for providing a comfortable indoor environment, are increasingly widely used in modern life. However, traditional air conditioner control methods often cannot meet the needs of different indoor environments and may also have the problem of excessive energy consumption. This has prompted people to seek more intelligent and efficient air conditioner control methods. People's demands for the comfort and energy efficiency of the indoor environment are also constantly increasing.

[0003] In summary, the control accuracy of the prior art is insufficient, resulting in excessive consumption and waste of energy resources, lacking flexibility and unable to perform real-time temperature adjustment according to the dynamic changes in the spatial layout. Summary of the Invention

[0004] The purpose of this application is to provide an air conditioner adaptive temperature regulation system based on environmental perception to solve the problems of insufficient control accuracy in the prior art, resulting in excessive consumption and waste of energy resources, lacking flexibility and unable to perform real-time temperature adjustment according to the dynamic changes in the spatial layout.

[0005] In view of the above problems, this application provides an air conditioner adaptive temperature regulation system based on environmental perception.

[0006] The present application provides an air conditioner adaptive temperature regulation system based on environmental perception. Among them, the system includes: a design drawing acquisition module, which is used to collect the spatial structure design drawing of the target area. Among them, the spatial structure design drawing includes K sub-spaces, and there are partition devices between several of the K sub-spaces. The target area has M partition devices, and the partition devices are used to isolate or connect two adjacent sub-spaces; a sensor array generation module, which is used to deploy the panoramic camera array at the M partition devices in all target areas, and determine the number of temperature sensors to be deployed according to the space size in the K sub-spaces, and deploy temperature sensors according to the number of temperature sensors deployed in K, and generate the temperature sensor array. Among them, the panoramic camera array includes M panoramic cameras, and the temperature sensor array includes K temperature sensor sets; a status identification acquisition module, which is used to extract M panoramic videos of the M partition devices by the M panoramic cameras in a preset monitoring window before the current moment, and perform status identification on the M partition devices according to the M panoramic videos to obtain M partition device status identifications. Among them, the partition device status identification includes a passage identification and a break identification; a space fusion module, which is used to identify the corresponding positions of the spatial structure design drawing by using the M partition device status identifications, and perform space fusion on the K sub-spaces according to the real-time status map of the marked space structure to obtain M fused sub-spaces; a temperature data extraction module, which is used to target the spatial positions corresponding to the M fused sub-spaces, extract the real-time temperature data of the K temperature sensor sets, and obtain M sets of fused temperature values; a parameter configuration module, which is used to configure the parameters of the M air conditioner unit sets corresponding to the M fused sub-spaces respectively according to the degree of deviation between the M sets of fused temperature values and the preset reference temperature value and the spatial area of the M fused sub-spaces; a temperature regulation module, which is used to use the configured M air conditioner unit sets to regulate the temperature of the target area.

[0007] One or more technical solutions provided in the present application have at least the following technical effects or advantages: Through a sensor array generation module, the sensor array generation module is configured to deploy the panoramic camera array at M partition devices in all target areas, and determine the number of temperature sensors to be deployed according to the space size within the K subspaces, deploy the temperature sensors according to the number of temperature sensors to be deployed in the K subspaces, and generate the temperature sensor array, where the panoramic camera array includes M panoramic cameras, and the temperature sensor array includes K temperature sensor sets; a status identifier acquisition module, the status identifier acquisition module is configured to extract M panoramic videos of the M partition devices by the M panoramic cameras in a preset monitoring window before the current moment, and perform status identification on the M partition devices according to the M panoramic videos to obtain M partition device status identifiers, where the partition device status identifier includes a passage identifier and a break identifier; a space fusion module, the space fusion module is configured to use the M partition device status identifiers to identify corresponding positions of the space structure design diagram, and perform space fusion on the K subspaces according to the real-time status diagram of the marked space structure to obtain M fused subspaces; a temperature data extraction module, the temperature data extraction module is configured to target the space positions corresponding to the M fused subspaces, extract the real-time temperature data of the K temperature sensor sets, and obtain M sets of fused temperature values; a parameter configuration module, the parameter configuration module is configured to respectively configure parameters of the M air conditioner unit sets corresponding to the M fused subspaces according to the degree of deviation between the M sets of fused temperature values and a preset reference temperature value and the space areas of the M fused subspaces; a temperature adjustment module, the temperature adjustment module is configured to use the M configured air conditioner unit sets to adjust the temperature of the target area, effectively solving the problems of insufficient control accuracy in the prior art, resulting in excessive consumption and waste of energy resources, lack of flexibility, and inability to perform real-time temperature adjustment according to dynamic changes in the space layout, realizing real-time monitoring of the space structure and temperature control of the target area, and improving the comfort of the space and the energy use efficiency.

[0008] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically illustrates the embodiments of the present application. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understood through the following description. Brief Description of the Drawings

[0009] To more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only exemplary. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on the provided drawings.

[0010] Figure 1 It is a schematic structural diagram of the air conditioner adaptive temperature regulation system based on environmental perception of the present application.

[0011] Explanation of the reference numerals: Design drawing acquisition module 11, sensor array generation module 12, status flag acquisition module 13, spatial fusion module 14, temperature data extraction module 15, parameter configuration module 16, temperature regulation module 17. Detailed implementation manners

[0012] By providing the air conditioner adaptive temperature regulation system based on environmental perception, the present application solves the problems of insufficient control accuracy in the prior art, resulting in excessive consumption and waste of energy resources, lack of flexibility, and inability to perform real-time temperature adjustment according to the dynamic changes in the spatial layout, realizing real-time monitoring and temperature control of the spatial structure of the target area, and improving the comfort of the space and the energy use efficiency.

[0013] Next, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described here. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application. Additionally, it should be noted that for the sake of convenience of description, only the parts related to the present application are shown in the accompanying drawings rather than all of them.

[0014] Embodiment 1 The present application provides an air conditioner adaptive temperature regulation system based on environmental perception. Please refer to the attached Figure 1 , the system is applied to a temperature regulation platform, and the temperature regulation platform is communicatively connected to a panoramic camera array and a temperature sensor array. The system includes, the system includes: A design drawing acquisition module 11, which is used to collect the spatial structure design drawing of the target area. Among them, the spatial structure design drawing includes K sub-spaces, and there are partition devices between some of the K sub-spaces. The target area has M partition devices, and the partition devices are used to isolate or connect two adjacent sub-spaces; Specifically, a spatial structure design diagram of the target area is collected. This design diagram describes the situation where the target area is divided into K sub-spaces and the relationships between these sub-spaces. Each sub-space may represent a room, an area, or a functional block. The spatial structure design diagram is a digital model that shows the overall layout and structure of the target area. It includes the positions, sizes, and mutual relationships of the K sub-spaces, as well as the positions and functions of the M partition devices. The K sub-spaces are the different parts into which the target area is divided. They can be actual rooms, offices, corridors, etc. The partition devices are designed to isolate or connect adjacent sub-spaces as needed. For example, they can be movable walls, folding screens, sliding doors, etc. Isolation and connection are the main functions of the partition devices. When privacy or noise reduction is required, the partition devices can be closed or moved to the isolation position; when openness and fluidity are needed, the partition devices can be opened or moved to connect adjacent sub-spaces.

[0015] The sensor array generation module 12 is configured to deploy the panoramic camera array at the M partition devices in all target areas, and determine the number of temperature sensors to be deployed according to the space size within the K sub-spaces, deploy the temperature sensors according to the number of temperature sensors deployed in the K sub-spaces, and generate the temperature sensor array, where the panoramic camera array includes M panoramic cameras, and the temperature sensor array includes K temperature sensor sets; Specifically, M panoramic cameras are respectively deployed at the M partition devices in the target area. The purpose of this is to monitor the status of the partition devices in real time, such as whether they are open or closed, and to monitor the flow of people or goods passing through these partition devices. The panoramic camera can provide a 360-degree field of view. Through the panoramic camera array, the usage situation of the partition devices can be obtained in real time. According to the sizes of the K sub-spaces, the number of temperature sensors to be deployed in each sub-space is determined. Larger sub-spaces may require more temperature sensors to ensure uniform temperature monitoring. In each sub-space, the temperature sensors are deployed according to the determined number to form K temperature sensor sets, and these sets together constitute the temperature sensor array. The temperature sensors are used to monitor the temperature of each sub-space in real time, and the temperature data is used for the decision-making of the air-conditioning adaptive temperature regulation system to ensure that the temperature of each sub-space is controlled within the set comfortable range.

[0016] The status identification acquisition module 13 is configured to extract M panoramic videos of the M partition devices from the preset monitoring window of the M panoramic cameras before the current moment, and perform status identification on the M partition devices according to the M panoramic videos to obtain M partition device status identifications, where the partition device status identifications include a passage identification and a break identification; Specifically, video data within a preset monitoring window before the current moment is extracted from the M panoramic cameras. This preset monitoring window may be a fixed time period, such as the past few minutes or hours. Using image processing and computer vision techniques, each panoramic video is analyzed to identify the state of the partition device. This includes detecting features such as the position, shape, and color of the partition device, as well as tracking its dynamic changes. Preferably, deep learning models, image segmentation techniques, or motion detection algorithms can be used for video analysis. Based on the results of the state identification, a state identifier is assigned to each partition device. This identifier reflects the state of the partition device at the end of the preset monitoring window. It includes a passage identifier and a break identifier. The passage identifier indicates that the partition device is in an open state and allows passage; while the break identifier indicates that the partition device is in a closed state and does not allow passage.

[0017] The spatial fusion module 14 is configured to use the M partition device state identifiers to identify corresponding positions on the spatial structure design diagram, and perform spatial fusion on the K sub-spaces based on the identified real-time state diagram of the spatial structure, to obtain M fused sub-spaces; Specifically, according to the M partition device state identifiers, the passage identifier or the break identifier, corresponding positions on the spatial structure design diagram are identified. On the digital model, the state of each partition device will be clearly marked. It converts the static spatial design diagram into a dynamic spatial state diagram, reflecting the real-time spatial configuration. With the identified real-time state diagram of the spatial structure, it is possible to determine which sub-spaces are connected based on the open or closed state of the partition device. When the partition devices between two or more sub-spaces are in an open state, these sub-spaces substantially form a continuous space, and thus spatial fusion can be performed. These connected sub-spaces are treated as a whole. After spatial fusion processing, M fused sub-spaces are obtained. These fused sub-spaces are dynamically combined from the original sub-spaces according to the state of the partition device. Each fused sub-space may contain one or more of the original sub-spaces, depending on which partition devices are in an open state.

[0018] The temperature data extraction module 15 is configured to target the spatial positions corresponding to the M fused sub-spaces, and extract the real-time temperature data of the K temperature sensor sets, to obtain M sets of fused temperature values; Specifically, according to the location of the fusion subspace, temperature data is extracted from the previously arranged set of K temperature sensors. Each fusion subspace may contain multiple temperature sensors. For each fusion subspace, a representative temperature value is calculated. The calculation of the fusion temperature value can be based on various methods, such as taking the average value, median, weighted average, etc. If a fusion subspace contains a relatively large number of temperature sensors and is unevenly distributed, a more complex algorithm is adopted to ensure the accuracy and representativeness of the fusion temperature value. After the above steps, a fusion temperature value is calculated for each fusion subspace. These fusion temperature values together constitute a set of M fusion temperature values, with each set corresponding to a fusion subspace.

[0019] The parameter configuration module 16 is used to configure the parameters of the set of M air-conditioning units corresponding to the M fusion subspaces respectively according to the degree of deviation between the set of M fusion temperature values and the preset reference temperature value and the spatial area of the M fusion subspaces. Specifically, compare the temperature data in each set of fusion temperature values with the preset reference temperature value, and calculate the degree of deviation between them. This degree of deviation can be measured by means such as difference value, absolute difference value, temperature difference percentage, etc. The preset reference temperature value is set according to the requirements of indoor environmental comfort and energy conservation, representing the ideal indoor temperature. For a larger space, it may require stronger cooling or heating capacity to maintain the ideal temperature. According to the temperature deviation degree and the spatial area, the parameters such as the air-conditioning power and air supply volume required for each fusion subspace can be determined more precisely. Configure the parameters of the set of air-conditioning units corresponding to each fusion subspace. This includes adjusting the operating mode of the air conditioner, such as cooling, heating, air supply, etc., setting the temperature target value, adjusting the wind speed, controlling the direction and opening degree of the air supply outlet, etc. The parameter configuration is to make the temperature of each fusion subspace reach and stabilize near the preset reference temperature value as soon as possible, while considering the balance between energy efficiency and comfort.

[0020] The temperature adjustment module 17 is used to adjust the temperature of the target area by using the configured set of M air-conditioning units.

[0021] Specifically, according to the parameter configuration, commands are sent to each air conditioner unit to start running according to the set parameters. The air conditioner units will start working according to the configured operating modes, such as cooling, heating, air supply, etc., and adjust the air supply temperature, wind speed and direction to change the temperature of the fusion subspace. Continuously monitor the real-time temperature of each fusion subspace and compare it with the preset reference temperature value to ensure that the temperature gradually approaches the preset value. During the operation of the air conditioner units, collect the real-time data of each set of temperature sensors and evaluate the effect of temperature adjustment. If it is found that the temperature of a certain fusion subspace deviates too much from the preset value or the adjustment speed is too slow, the parameters of the corresponding air conditioner unit are adjusted in real time to optimize the temperature adjustment effect. When adjusting the temperature, energy conservation is considered. For example, after reaching the preset temperature, the air conditioner unit will automatically switch to a lower power mode to reduce energy consumption.

[0022] Furthermore, the system further includes: A status recognition module, configured to construct a status recognizer to recognize the on / off status of the partition devices in M panoramic videos, and obtain M on / off status time series sequences; A sequence retrieval module, configured to retrieve the M on / off status time series sequences with the current moment as the index to obtain M current partition device statuses; A status judgment module, configured to judge whether the M current partition device statuses are in a conducting state. If so, use the conducting state flag as the M partition device status flags; A changed status extraction module, if not, extract the status change frequency of the partition devices in the M on / off status time series sequences. When the status change frequency is greater than the preset change frequency, use the conducting state flag as the M partition device status flags; A status flag generation module, configured to use the open circuit flag as the M partition device status flags when the status change frequency is less than or equal to the preset change frequency.

[0023] Specifically, a state recognizer is constructed, which can analyze panoramic videos and accurately identify the on / off state of the partition device. By processing M panoramic videos with the state recognizer, the on / off state of the partition device in each video can be identified, thereby obtaining M on / off state time series. These time series record the change of the partition device state over time. Taking the current moment as the index, retrieving the M on / off state time series, the state of each partition device at the current moment can be found, that is, obtaining the M current partition device states. Check whether these M current partition device states are in a conductive state. If it is in a conductive state, directly use the conductive identifier as the state identifier of these partition devices. If the current partition device state is not in a conductive state, that is, in an open or closed state, it is necessary to further analyze the on / off state time series. Extract the state change frequency of each partition device from the M on / off state time series, that is, count the number of times the state of each partition device changes from the start of the series to the current moment. When the state change frequency of a certain partition device is greater than a preset change frequency, this partition device frequently changes its state in the near future and may be in use or about to be used, so the conductive identifier is used as the state identifier of this partition device. On the contrary, if the state change frequency is less than or equal to the preset change frequency, it indicates that the state of this partition device is relatively stable in the near future and is currently in an open state, so the open identifier is used as the state identifier of this partition device.

[0024] Furthermore, the system further includes: A window period state extraction module. If not, based on the M on / off state time series, extract M window periods from the current moment to the previous state change node. When the M window periods are lower than a preset window period, use the conductive identifier as the M partition device state identifiers.

[0025] Specifically, extract the time period between the current moment and the previous state change node, that is, the time point when the state of the partition device last changed. This time period is called the window period. For each partition device, a corresponding window period will be obtained, so there are a total of M window periods. Compare these M window periods with a preset window period. This preset window period is a threshold used to judge the recency of the partition device state change. The window period can be set by those skilled in the art. If the window period of a certain partition device is lower than, that is, shorter than the preset window period, the time from the last state change to now is very short, indicating that this partition device has a state change in the near future. When the window period of a certain partition device is lower than the preset window period, even if it is currently in an open state, it is considered that it may change its state again, either open or closed, soon, so the conductive identifier is used as the state identifier of this partition device.

[0026] Furthermore, the system further includes: The state recognizer composition module, where the state recognizer includes a first state extraction path, a second state extraction path, and a state recognition network layer; The first image frame set acquisition module is used to extract image frames from the M panoramic videos respectively according to the first extraction step length by using the first state extraction path, and obtain M first image frame sets; The second image frame set acquisition module is used to extract image frames from the M panoramic videos respectively according to the second extraction step length by using the second state extraction path, and obtain M second image frame sets, where the second extraction step length is greater than the second extraction step length; The on-off state time series acquisition module is used to use the state recognition network layer to identify the on-off state of the partition device for the M first image frame sets and the M second image frame sets, and obtain M on-off state time series.

[0027] Specifically, the first state extraction path is used to extract image frames from the panoramic video according to a shorter step length, the first extraction step length. The second state extraction path is used to extract image frames from the panoramic video according to a longer step length, the second extraction step length, and greater than the first extraction step length. The state recognition network layer is responsible for processing the extracted image frames and identifying the on-off state of the partition device therein. Image frames are extracted from the M panoramic videos with the first extraction step length. Frames will be captured from the video relatively frequently, so as to be able to capture the rapid changes in the state of the partition device. This process will generate M first image frame sets. Image frames are extracted from the same M panoramic videos with the second extraction step length greater than the first extraction step length. Due to the longer step length, this path will capture relatively sparse frames in the video, which may be used to capture the long-term changes or overall trends in the state of the partition device. This process will generate M second image frame sets. The state recognition network layer receives the M first image frame sets and the M second image frame sets of the image frame sets from the two extraction paths as inputs. By analyzing these image frames, the state recognition network layer can identify the on-off state of the partition device in each video and generate M on-off state time series. These time series reflect the change of the partition device state over time. The state recognition network layer can analyze information extracted from various data sources, such as video streams and sensor data, to identify and determine the current state of the target object, such as the partition device. Through a deep learning model, such as a convolutional neural network CNN, features helpful for state recognition are extracted from the raw data. According to the extracted features, the state recognition network layer can be trained to recognize different states of the target object. For example, in the scenario of the partition device, the model will recognize whether the device is in a conducting state or an open state.

[0028] Furthermore, the system further includes: A change event acquisition module, configured to extract the status change times of the partition devices within the target area in a historical window, so as to obtain a plurality of historical status change times; A first extraction compensation acquisition module, configured to use the minimum value among the plurality of historical status change times as a first extraction step length, where the first extraction step length is the minimum time length between adjacent two image frames when extracting image frames from M panoramic videos; A second extraction compensation acquisition module, configured to use the maximum value among the plurality of historical status change times as a second extraction step length, where the second extraction step length is the maximum time length between adjacent two image frames when extracting image frames from M panoramic videos.

[0029] Specifically, view the status change records of the partition devices in the past period of time. These records will show when the partition device changes from the open state to the closed state, or from the closed state to the open state. From these historical records, the specific time of each status change can be extracted. For example, if the partition device changes from closed to open at a certain time point, then this time point will be recorded. Among all the extracted status change times, find the shortest time interval between two status changes. This time interval represents the shortest time for the status of the partition device to change rapidly. To capture such rapid changes, set this time interval as the first extraction step length. When extracting image frames from the panoramic video, extract according to this shortest time interval or a slightly longer time to ensure that no changes are missed, so as to ensure that all status changes can be captured. Similarly, find the longest time interval among all the status change times. This time interval represents the longest time required for the status change of the partition device. To optimize the data processing efficiency, set the second extraction step length according to this longest time interval. When the status changes are not frequent, extract image frames according to this longer step length, so as to reduce the data processing volume.

[0030] Furthermore, the system further includes: A first subspace acquisition module, configured to randomly select a first subspace from K subspaces in the real-time status graph of the identified spatial structure; A first fusion subspace acquisition module, configured to determine whether a plurality of partition devices within the first subspace have path identifiers. If so, extract a plurality of first partition devices with path identifiers in the first subspace, and fuse the plurality of subspaces adjacent to the first subspace through the plurality of first partition devices with the first subspace to obtain a first fusion subspace; A space update module, configured to determine again whether multiple partition devices in the first fusion subspace have a path identifier. If so, extract multiple first fusion partition devices with path identifiers in the first fusion subspace, and perform space fusion on multiple subspaces adjacent to the first fusion subspace through the multiple first fusion partition devices, and update the first fusion subspace according to the fusion result; A status elimination module, if not, eliminate the first fusion subspace from the real-time status diagram of the identified space structure, and then randomly select a second subspace from the remaining multiple subspaces; A second fusion subspace acquisition module, configured to perform space fusion analysis on the second subspace to obtain a second fusion subspace; A fusion analysis module, configured to perform multiple space fusion analyses until all of the K subspaces are analyzed, to obtain M fusion subspaces.

[0031] Specifically, randomly select one from the K subspaces in the real-time status diagram of the identified space structure, which is called the first subspace. Check whether the partition devices in the first subspace have a path identifier. If a certain partition device has a path identifier, it can be regarded as a passable channel rather than a partition. If there are partition devices with path identifiers in the first subspace, extract these partition devices with path identifiers, which are called the first partition devices. Determine which subspaces are adjacent to the first subspace through these first partition devices with path identifiers. Perform fusion on these adjacent subspaces and the first subspace to form a larger space, which is called the first fusion subspace. Repeat the above judgment and extraction process for the first fusion subspace. If there are still first fusion partition devices with path identifiers in this larger fusion space, continue to fuse the subspaces adjacent to the first fusion subspace through these partition devices. Continuously update the range of the first fusion subspace according to the new fusion result. If at a certain fusion stage, there are no more partition devices with path identifiers in the fusion subspace, eliminate the current fusion subspace such as the first fusion subspace from the real-time status diagram of the identified space structure. Randomly select a new subspace from the remaining unanalyzed subspaces, such as the second subspace, and perform the same space fusion analysis as above on it. Repeat the above process to perform space fusion analysis on each new subspace until all K subspaces are analyzed. After multiple space fusion analyses, M fusion subspaces will be obtained, and these fusion subspaces represent the space regions connected to each other through the partition devices with path identifiers.

[0032] Further, the system further includes: A first fluctuation coefficient acquisition module, configured to calculate the variances of the M sets of fusion temperature values respectively to obtain M first fluctuation coefficients; The first fusion temperature value acquisition module is configured to calculate the M means of the M fusion temperature value sets when the M first fluctuation coefficients are greater than a preset fluctuation coefficient, and use the M means as starting points to retrieve in the M fusion temperature value sets according to a preset search step size, so as to obtain M first fusion temperature values; The temperature value judgment module is configured to judge whether the temperature value concentration of the M means is greater than the temperature value concentration of the M first fusion temperature values. If not, the M first fusion temperature values are used as the M stage fusion temperature values; The target fusion temperature value acquisition module is configured to update the M stage fusion temperature values as starting points, and continue to retrieve in the M fusion temperature value sets according to the preset search step size until the difference between the temperature value concentrations of two adjacent retrievals is less than a preset increment, stop the retrieval, and use the M stage fusion temperature values corresponding to the maximum temperature value and weight during the retrieval process as the M target fusion temperature values.

[0033] Specifically, calculate the variance for each of the M fusion temperature value sets. Variance is a statistic that measures the degree of dispersion of numerical values in a dataset. By calculating the variance, M first fluctuation coefficients can be obtained, and these coefficients reflect the degree of dispersion of the temperature values in each set. When the first fluctuation coefficient of a certain set is greater than a preset fluctuation coefficient, the temperature values in this set are relatively dispersed and may require further optimization. For the sets with fluctuation coefficients greater than the preset value, calculate their means, and use this mean as the starting point to search in the set according to a preset search step size. Search for a more concentrated or representative temperature value, that is, the first fusion temperature value. After finding the M first fusion temperature values, judge the temperature value concentration of these means, which refers to whether the aggregation degree or frequency of the temperature values near these means is greater than the temperature value concentration of the first fusion temperature values. If the temperature value concentration near the means is not greater than the concentration of the first fusion temperature values, then these M first fusion temperature values are used as the stagewise optimal solutions, that is, the M stage fusion temperature values. Next, update these M stage fusion temperature values as the new search starting points, and continue to search in the fusion temperature value set according to the preset search step size. Each iteration will update the stage fusion temperature values until the difference between the temperature value concentrations of two adjacent retrievals is less than a preset increment. When the difference between the temperature value concentrations of two adjacent retrievals during the search process is less than the preset increment, the search stops. Finally, use the temperature value concentration and the corresponding weight during the search process, which refers to the M stage fusion temperature values corresponding to the maximum importance or frequency of the temperature value, as the M target fusion temperature values.

[0034] Furthermore, the system further includes: The mean calculation module is used to calculate the mean of multiple fusion temperature values in M first regions respectively to obtain M fusion temperature values, where the M first regions are regions constructed with the M target fusion temperature values as the centers and the preset search step length as the radius; The temperature deviation value generation module is used to calculate the degree of deviation between the M fusion temperature values and a preset reference temperature value respectively, and generate M temperature deviation values; The unit parameter configuration module is used to configure the parameters of the M air conditioner unit sets corresponding to the M fusion subspaces according to the M temperature deviation values and the spatial areas of the M fusion subspaces.

[0035] Specifically, M target fusion temperature values are determined. With each target fusion temperature value as the center and the preset search step length as the radius, M first regions are constructed. In each first region, there are multiple fusion temperature values. The mean of these temperature values is calculated to obtain M fusion temperature values. These fusion temperature values represent the average temperature conditions in their respective regions. Then, the degree of deviation between these M fusion temperature values and a preset reference temperature value is calculated. This reference temperature value can be the ideal indoor temperature, the energy-saving standard temperature, or the comfortable temperature set by the user, etc. By comparing the fusion temperature values with the reference temperature value, M temperature deviation values can be generated. These deviation values reflect the difference between the actual temperature and the ideal state. Combining with the spatial areas of the M fusion subspaces, personalized parameter configuration can be performed for the M air conditioner unit sets corresponding to each fusion subspace. For example, for the subspace with a large temperature deviation, it is necessary to increase the cooling or heating power of the air conditioner unit to adjust the indoor temperature to the reference value faster; while for the subspace with a small temperature deviation, the power of the air conditioner unit can be appropriately reduced to save energy. According to the above analysis, the parameters of the M air conditioner unit sets corresponding to the M fusion subspaces are adjusted. This includes changing the working mode, wind speed, temperature set point, etc. of the air conditioner unit. The goal of the adjustment is to make the temperature of each subspace as close as possible to the preset reference temperature value, while considering energy conservation and comfort.

[0036] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0037] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of this application and its equivalent technologies, this application is also intended to include these changes and modifications.

Claims

1. The air conditioning adaptive temperature control system based on environmental perception is characterized by: The system is applied to a temperature regulating platform, and the temperature regulating platform is communicatively connected with a panoramic camera array and a temperature sensor array. The system includes: A design acquisition module, the design acquisition module is used to collect a spatial structure design of a target area, wherein the spatial structure design includes K subspaces, several of the K subspaces are provided with partition devices, and the target area has M partition devices, and the partition devices are used to isolate or connect two adjacent subspaces; A sensor array generation module, the sensor array generation module is used to deploy the panoramic camera array at the M partition devices of all target areas, and determine the number of deployed temperature sensors in the K subspaces according to the space size, deploy the temperature sensors according to the K number of deployed temperature sensors, and generate the temperature sensor array, wherein the panoramic camera array includes M panoramic cameras, and the temperature sensor array includes K temperature sensor sets; A status identification acquisition module, the status identification acquisition module is used to extract M panoramic videos of the M partition devices in a preset monitoring window before the current moment by the M panoramic cameras, and perform status identification on the M partition devices according to the M panoramic videos to obtain M partition device status identifications, wherein the partition device status identifications include a passage identification and a disconnection identification; A spatial fusion module, the spatial fusion module is used to mark the corresponding positions of the spatial structure design diagram using the M partition device status identifiers, and to perform spatial fusion on the K subspaces according to the marked real-time status diagram of the spatial structure to obtain M fused subspaces; A temperature data extraction module, the temperature data extraction module is used to extract the real-time temperature data of the K temperature sensor sets with the spatial positions corresponding to the M fusion subspaces as the target, and obtain M fusion temperature value sets; A parameter configuration module, the parameter configuration module is used to perform parameter configuration on the M air-conditioning unit sets corresponding to the M fusion subspaces according to the degree of deviation between the M fusion temperature value sets and the preset reference temperature value and the spatial area of ​​the M fusion subspaces; A temperature adjustment module is used to adjust the temperature of the target area using the configured set of M air-conditioning units.

2. The system according to claim 1, characterized in that Extracting M panoramic videos of the M partition devices in a preset monitoring window of the M panoramic cameras before the current moment, and identifying the states of the partition devices according to the M panoramic videos to obtain M partition device state identifiers, the system includes: A state recognition module is used to construct a state recognizer to recognize the on and off states of the partition device in the M panoramic videos, and obtain M on and off state time series; A sequence retrieval module, used to retrieve the M on-off state time sequence sequences with the current time as the index, and obtain M current isolation device states; A state judgment module, used to judge whether the states of the M current partition devices are passages, and if so, use the passage identifier as the state identifier of the M partition devices; A change state extraction module, if not, extracts the state change frequency of the isolation device in the M on-off state timing sequences, and when the state change frequency is greater than the preset change frequency, uses the passage identifier as the state identifier of the M isolation devices; The status identification generating module is used to use the circuit breaking identification as the status identification of the M isolation devices when the status change frequency is less than or equal to the preset change frequency.

3. The system according to claim 2, characterized in that Determining whether the M current partition device states are passages, the system further includes: The window period state extraction module, if not, extracts M window periods from the current moment to the last state change node based on the M on-off state timing sequences, and when the M window periods are lower than the preset window periods, uses the passage identifier as the M isolation device state identifier.

4. The system according to claim 2, characterized in that A state identifier is constructed to identify the on-off state of the partition device in M ​​panoramic videos to obtain M on-off state time series. The system includes: A state recognizer component module, wherein the state recognizer includes a first state extraction path, a second state extraction path, and a state recognition network layer; A first image frame set acquisition module, configured to extract image frames from the M panoramic videos respectively according to a first extraction step length by using the first state extraction path to obtain M first image frame sets; A second image frame set acquisition module, configured to extract image frames from the M panoramic videos respectively according to a second extraction step length by using the second state extraction path to obtain M second image frame sets, wherein the second extraction step length is greater than the second extraction step length; The on-off state timing sequence acquisition module is used to use the state recognition network layer to identify the on-off state of the partition device on the M first image frame sets and the M second image frame sets to obtain M on-off state timing sequences.

5. The system according to claim 4, characterized in that The system comprises: A change event acquisition module, used to extract the state change time of the partition device in the target area within the historical window to obtain multiple historical state change times; A first extraction compensation acquisition module, configured to use a minimum value among the multiple historical state change times as a first extraction step length, wherein the first extraction step length is a minimum length of an interval between two adjacent image frames when image frames are extracted from M panoramic videos; The second extraction compensation acquisition module is used to take the maximum value of the multiple historical state change times as the second extraction step length, wherein the second extraction step length is the maximum duration of the interval between two adjacent image frames when extracting image frames from M panoramic videos.

6. The system according to claim 1, characterized in that The corresponding positions of the spatial structure design diagram are marked by using the M partition device status marks, and the K subspaces are spatially fused according to the marked real-time status diagram of the spatial structure to obtain M fused subspaces. The system includes: A first subspace acquisition module is used to randomly select a first subspace from the K subspaces in the marked spatial structure real-time state diagram; A first fused subspace acquisition module is used to determine whether the multiple partitioning devices in the first subspace have a passage mark, and if so, extract the multiple first partitioning devices with the passage mark in the first subspace, and spatially fuse the multiple subspaces adjacent to the first subspace through the multiple first partitioning devices with the first subspace to obtain a first fused subspace; A spatial updating module, used for judging again whether the multiple partitioning devices in the first fused subspace have a passage identifier, and if so, extracting the multiple first fused partitioning devices with the passage identifier in the first fused subspace, and spatially fusing the multiple subspaces adjacent to the first fused subspace through the multiple first fused partitioning devices with the first fused subspace, and updating the first fused subspace according to the fusion result; A state elimination module, if not, eliminates the first fused subspace from the real-time state diagram of the identified spatial structure, and then randomly selects a second subspace from the remaining multiple subspaces; A second fused subspace acquisition module, used for performing spatial fusion analysis on the second subspace to obtain a second fused subspace; The fusion analysis module is used to obtain M fused subspaces after multiple spatial fusion analyses until all the K subspaces are analyzed.

7. The system according to claim 1, characterized in that According to the degree of deviation between the M fusion temperature value sets and the preset reference temperature value and the spatial area of ​​the M fusion subspaces, parameter configuration is performed on the M air-conditioning unit sets corresponding to the M fusion subspaces, and the system includes: A first fluctuation coefficient acquisition module, used to respectively calculate the variances of the M fusion temperature value sets to obtain M first fluctuation coefficients; A first fusion temperature value acquisition module, configured to calculate M means of the M fusion temperature value sets when the M first fluctuation coefficients are greater than a preset fluctuation coefficient, and use the M means as a starting point to search in the M fusion temperature value sets according to a preset search step length to obtain M first fusion temperature values; a temperature value judgment module, used to judge whether the temperature value concentration of the M average values ​​is greater than the temperature value concentration of the M first fusion temperature values; if not, taking the M first fusion temperature values ​​as the M stage fusion temperature values; The target fusion temperature value acquisition module is used to update the M stage fusion temperature values ​​as the starting point, continue to search in the M fusion temperature value set according to the preset search step, until the difference in the concentration of temperature values ​​between two adjacent searches is less than the preset increment, stop the search, and use the M stage fusion temperature values ​​corresponding to the maximum temperature values ​​and weights during the search process as the M target fusion temperature values.

8. The system according to claim 7, characterized in that After obtaining M target fusion temperature values, the system further comprises: a mean value calculation module, used to respectively calculate the mean values ​​of a plurality of fusion temperature values ​​in the M first regions to obtain M fusion temperature values, wherein the M first regions are regions constructed with the M target fusion temperature values ​​as the center and the preset search step as the radius; A temperature deviation value generating module, used to respectively calculate the degree of deviation between the M fusion temperature values ​​and a preset reference temperature value, and generate M temperature deviation values; The unit parameter configuration module is used to perform parameter configuration on the M air-conditioning unit sets corresponding to the M fusion subspaces according to the M temperature deviation values ​​and the spatial areas of the M fusion subspaces.

Images