Infrared Air Conditioning Energy-Saving Control Method and System Based on Crowd Characteristics and Spatial Recognition

Abstract

This invention discloses an infrared air conditioning energy-saving control method and system based on crowd characteristics and spatial recognition, belonging to the field of image processing technology. It acquires video images and infrared thermal images in real time, performs face recognition, limb detection, and region segmentation and spatial alignment processing to obtain infrared temperature data for each image region. Thermal comfort is determined based on the infrared reference temperature of each image region, and air supply execution parameters are obtained by combining the spatial distribution of people, changes in spatial reference temperature, and temperature change slope. After air supply processing, heat load change analysis is performed. When the heat load change result is unqualified, a first or second air supply adjustment process is executed based on the cause of the heat load change, dynamically adjusting the air supply temperature, air volume, and swing frequency. This invention can improve the regional cooling effect in scenarios with uneven crowd distribution, reduce air conditioning energy consumption, and improve air conditioning energy utilization efficiency.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology, specifically relating to an infrared air conditioning energy-saving control method and system based on crowd characteristics and spatial recognition. Background Technology

[0002] With the development of smart home and image recognition technology, existing technologies have developed intelligent air conditioner control solutions based on facial recognition. Existing air conditioner control methods based on facial recognition acquire the facial image of the target user, determine the corresponding control model based on the facial image, and then combine the target user's body surface temperature information to determine the air conditioner control parameters, thereby realizing intelligent adjustment of the air conditioner's operating status.

[0003] For example, Chinese invention patent CN110929671B discloses an air conditioner, an air conditioner control method based on face recognition, and a storage medium. The method includes: determining the face image of a target user; determining the corresponding control model based on the face image; determining the target user's body surface temperature information; and determining the corresponding air conditioner control parameters based on the control model and the body surface temperature information; and controlling the operation of the air conditioner according to the control parameters. Specifically, the method determines the control model corresponding to the target user based on the target user's face image, and determines the air conditioner control parameters by combining the target user's body surface temperature information.

[0004] However, existing technologies primarily analyze the temperature of a single user or the entire human body, lacking comprehensive analysis considering the spatial distribution of people, localized heat accumulation, and differences in thermal perception across different areas of the body. For example, in scenarios with uneven population density and large spatial spans, such as university classrooms with open lectures, the actual cooling effect can vary across different areas due to localized clustering and dispersion of people, leading to significant temperature deviations. If a fixed cooling mode is maintained, localized areas may experience poor cooling; conversely, blindly increasing the overall cooling capacity can waste cooling resources and increase energy consumption. Therefore, a new energy-saving air conditioning control method is needed to address the low energy utilization rate of existing technologies. Summary of the Invention

[0005] To address the aforementioned issues, this invention discloses an infrared air conditioning energy-saving control method and system based on crowd characteristics and spatial recognition. By combining human thermal state information such as the temperature of the face area and the temperature of the limbs area, as well as area division and spatial processing technology, the air conditioning air supply parameters are dynamically corrected, thereby improving the cooling effect while reducing energy consumption and improving the energy utilization efficiency of the air conditioning.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows:

[0007] In a first aspect, this invention discloses an infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition, comprising the following steps: real-time acquisition of video images and infrared thermal imaging images of people in the initial operation mode of the air conditioner; performing face recognition and limb detection based on the video images, and performing region segmentation and alignment processing in conjunction with the infrared thermal imaging images to obtain the infrared temperature of each image region, each image region including each face region and each limb region; acquiring the infrared temperature data of each image region within a first preset retrospective time period, and performing temperature suitability condition detection to obtain a thermal comfort judgment result; if the thermal comfort judgment result is that the temperature is suitable, then maintaining the current steady-state operation state, otherwise performing temperature adjustment; performing temperature adjustment, specifically by: extracting the regional spatial location of the video images, and performing regional heat analysis to obtain the air supply execution parameters of the air conditioner, thereby performing air supply adjustment processing; acquiring the personnel heat load change parameters within a second preset time period, and analyzing the heat load change result; if the heat load change result is qualified, then maintaining the current air supply execution parameters, otherwise performing secondary air supply adjustment until the heat load change result is qualified.

[0008] Secondly, this invention discloses an infrared air conditioning energy-saving control system based on crowd characteristics and spatial recognition, comprising the following modules:

[0009] The data acquisition module is used to acquire video images and infrared thermal images of personnel in the initial operation mode of the air conditioner in real time;

[0010] The region segmentation module is used to perform face recognition and limb detection based on video images, and to perform region segmentation and alignment processing in conjunction with infrared thermal imaging images to obtain the infrared temperature of each image region. Each image region includes each face region and each limb region.

[0011] The thermal comfort determination module is used to acquire infrared temperature data of each image area within the first preset retrospective period, and to detect the temperature suitability conditions to obtain the thermal comfort determination result. If the thermal comfort determination result is that the temperature is suitable, the current steady-state operation is maintained; otherwise, temperature adjustment is performed.

[0012] The air supply adjustment module is used to perform temperature adjustment. The specific method is as follows: extract the regional spatial location of the video image, perform regional heat analysis, obtain the air supply execution parameters of the air conditioner, and then perform air supply adjustment processing.

[0013] The secondary air supply adjustment module is used to obtain the personnel heat load change parameters and infrared reference temperature within the second preset time period, and analyze the heat load change results. If the heat load change results are qualified, the current air supply execution parameters are maintained; otherwise, secondary air supply adjustment is performed until the heat load change results are qualified.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0015] 1. This invention acquires video images and infrared thermal images in real time, performs face recognition and limb detection based on the video images, and performs region segmentation and alignment processing in conjunction with the infrared thermal images to obtain the face region temperature and limb region temperature corresponding to each image region. This enables regionalized air supply control based on the spatial distribution of people and the thermal state of the human body, effectively solving the problem of excessive cooling and low energy utilization caused by large differences in cooling effect in local areas in the existing technology when people are unevenly distributed.

[0016] 2. This invention obtains the spatial reference temperature of each spatial area within a first preset retrospective period and performs temperature change analysis. Based on the heat change status of different spatial areas, it performs air volume adjustment or air supply temperature adjustment respectively, thereby achieving dynamic matching between cooling capacity and air circulation capacity, improving the cooling efficiency of local high-heat areas, and reducing energy waste caused by continuous high-power operation of the overall air conditioner.

[0017] 3. This invention obtains the personnel heat load change parameters and infrared reference temperature within a second preset time period, and combines the personnel temperature change value, infrared temperature change rate and the change ratio of the exposed area of ​​the limbs to perform heat load change analysis, thereby providing feedback on the actual cooling effect after the current air supply adjustment, and thus realizing dynamic air supply correction based on changes in human body thermal state, improving the accuracy of air conditioning air supply control and environmental adaptability.

[0018] 4. This invention performs source analysis based on heat load change results and performs corresponding secondary air supply adjustment processes for the influence of spatial reference temperature and heat load change, thereby achieving fine dynamic control of air supply temperature, air volume and swing frequency. This improves the uniformity of cold air coverage in areas where people gather, avoids continuous overheating or overcooling in local areas, and improves overall comfort and air conditioning energy saving effect. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the overall process of infrared air conditioning energy-saving control according to the present invention.

[0020] Figure 2 This is a flowchart illustrating the generation of air supply execution parameters for the present invention.

[0021] Figure 3 This is a system structure diagram of the present invention. Detailed Implementation

[0022] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention.

[0023] like Figure 1 As shown in this embodiment, an infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition includes the following steps: real-time acquisition of video images and infrared thermal imaging images of people in the initial operation mode of the air conditioner; performing face recognition and limb detection based on the video images, and performing region segmentation and alignment processing in conjunction with the infrared thermal imaging images to obtain the infrared temperature of each image region, where each image region includes each face region and each limb region; acquiring the infrared temperature data of each image region within a first preset retrospective time period, and performing temperature suitability condition detection to obtain a thermal comfort judgment result; if the thermal comfort judgment result indicates that the temperature is suitable, then maintaining the current steady-state operation state, otherwise performing temperature adjustment; performing temperature adjustment, specifically by: extracting the regional spatial location of the video images and performing regional heat analysis to obtain the air supply execution parameters of the air conditioner, thereby performing air supply adjustment processing; acquiring the personnel heat load change parameters within a second preset time period, and analyzing the heat load change result; if the heat load change result is qualified, then maintaining the current air supply execution parameters, otherwise performing secondary air supply adjustment until the heat load change result is qualified.

[0024] The method for obtaining infrared temperature data of each image region described in this embodiment is as follows: performing face recognition and limb detection based on video images to identify the face region and limb region of each person; spatially aligning the infrared thermal imaging image with the video image to obtain the infrared temperature data of each face region and the infrared temperature data of each limb region; jointly labeling the face region and limb region as each image region; the infrared temperature data is specifically the infrared temperature value of each pixel in each image region.

[0025] In this embodiment, spatial alignment refers to mapping the face and limb regions in the video image to the temperature regions in the infrared thermal imaging image, so that the human body region in the video image can correspond to the temperature pixel region in the infrared thermal imaging image.

[0026] Face recognition and limb detection are performed based on video images to identify the face and limb regions of each person. This can be achieved by using the RetinaFace face recognition model to identify the face bounding box in the video image, then using the OpenPose model (open pose estimation model) to extract the human skeleton key points, and constructing the limb regions based on the shoulder key points, elbow key points, wrist key points, and knee key points. Finally, the corresponding regions are mapped onto the infrared thermal imaging image to form the image region, thereby identifying the face and limb regions of the person.

[0027] Infrared temperature value refers to the temperature value corresponding to each pixel in an infrared thermal imaging image. Thermal imaging data can be directly obtained through an infrared thermal imaging camera.

[0028] By spatially aligning video images with infrared thermal images, a precise correspondence between human body regions and temperature regions is achieved. This allows for the separate acquisition of infrared temperature data for the face and limb regions, improving the accuracy of human body thermal state analysis. Furthermore, by using the OpenPose model to detect the limb regions, the thermal differences in different areas of the body can be further analyzed, avoiding the problem of inaccurate local thermal state identification caused by relying solely on overall body temperature analysis. Statistical processing of pixel temperatures within the image region reduces errors caused by single-point temperature fluctuations, improving the stability of airflow regulation.

[0029] The specific method for obtaining the thermal comfort determination result in this embodiment is as follows: The infrared temperature data of each image region in the current video image are averaged to obtain the average infrared temperature of each image region, which serves as the infrared reference temperature for each image region; the number of face regions whose infrared reference temperature does not belong to the preset face infrared temperature range is counted and recorded as the number of people with overheated faces; the number of limb regions whose infrared reference temperature does not belong to the preset limb infrared temperature range is counted and recorded as the number of people with overheated limbs; the number of people with overheated faces and limbs are compared with their corresponding thresholds. If both the number of people with overheated faces and limbs are less than their corresponding thresholds, the thermal comfort determination result is that the temperature is suitable; otherwise, the thermal comfort determination result is that the temperature is unsuitable.

[0030] In this embodiment, it should be noted that the infrared reference temperature refers to the regional representative temperature obtained by averaging the infrared temperature values ​​of all pixels within the corresponding image area (including the face area and the limb area).

[0031] By constructing separate infrared reference temperatures for the face and limbs, differentiated analysis of the thermal state of different areas of the human body is achieved, which can more accurately reflect the true thermal sensation of people. By jointly determining the thermal comfort state by considering the number of people with overheated faces and limbs, the problem of misjudgment caused by relying solely on the temperature of a single human body area can be avoided, thus improving the reliability of thermal comfort assessment results. The face area mainly reflects the core thermal sensation state of the human body, while the limb area can reflect local environmental temperature changes. The joint analysis of the two can provide more refined data support for subsequent air supply parameter adjustment, thereby improving overall cooling comfort and reducing ineffective cooling energy consumption.

[0032] The specific method for obtaining the air supply execution parameters of the air conditioner in this embodiment is as follows: Each video image within a first preset retrospective time period is divided into spatial regions based on the spatial distribution of people, resulting in spatial regions for each video image; each spatial region is spatially aligned with each image region to obtain a spatial reference temperature for each spatial region; the spatial reference temperature of each spatial region within the first preset retrospective time period is analyzed for trend changes to obtain the spatial reference temperature change value and the corresponding temperature change slope for each spatial region; based on this temperature change slope and compared with a preset temperature change slope threshold, if the absolute value of the temperature change slope is greater than the temperature change threshold threshold... If the slope threshold is set, an airflow adjustment value is obtained based on the space reference temperature change value, thereby increasing the airflow (because the scenario in this embodiment is summer, the air conditioner operates in cooling mode, and will automatically enter sleep mode after the cooling temperature reaches the target space cooling temperature, so the absolute value of the temperature change slope here is only used to represent the speed of cooling change, without considering the existence of heating); if the absolute value of the temperature change slope is less than the temperature change slope threshold, an airflow temperature adjustment value is obtained based on the space reference temperature change value, thereby reducing the airflow temperature; the airflow and airflow temperature are jointly marked as air supply execution parameters.

[0033] In this embodiment, as Figure 2 As shown, Figure 2 The flowchart for generating air supply execution parameters of this invention is as follows: spatial areas are divided based on the spatial distribution of personnel, and the spatial reference temperature of each spatial area is analyzed to obtain the spatial reference temperature change value and temperature change slope; when the absolute value of the temperature change slope is greater than the threshold, the air volume is adjusted, and when the absolute value of the temperature change slope is less than the threshold, the air supply temperature is adjusted, thereby obtaining the corresponding air supply execution parameters.

[0034] A spatial region refers to a local indoor area defined based on the spatial distribution of people. The spatial region division based on the spatial distribution of people yields various spatial regions for each video image. Specifically, this involves: extracting the center coordinates of people in the video image; using the K-means clustering algorithm to cluster the center coordinates of people; and forming multiple spatial regions based on the clustering results.

[0035] The spatial reference temperature of a space area refers to the infrared temperature reference value of people in that space area, that is, the average infrared temperature in that space area.

[0036] The first preset backtracking period refers to the historical backtracking starting from the current time (real-time time), and backtracking according to the preset backtracking period length. This backtracking period is used as the first preset backtracking period, such as 10 minutes.

[0037] The second preset time period refers to the continuous monitoring of future moments starting from the moment after the air supply adjustment is performed. The monitoring duration is the preset second monitoring duration, and the second preset time period is obtained from this. For example, 5 minutes.

[0038] The change value of the spatial reference temperature in the spatial region is specifically obtained by subtracting the spatial reference temperature at the end of the first preset backtracking period from the current spatial reference temperature, and using this spatial reference temperature backtracking difference as the change value of the spatial reference temperature.

[0039] The temperature change slope is specifically calculated by dividing the spatial reference temperature change value by the length of the backtracking period. It should be noted that this embodiment considers summer; therefore, the air supply adjustment and air conditioning mode in this embodiment are both in cooling mode. If the indoor ambient temperature reaches the set sleep mode, the air conditioner automatically enters sleep mode. Therefore, the temperature adjustment in this embodiment will only be a gradual cooling adjustment, and thus the temperature change slope is always negative, representing the slope of the cooling process.

[0040] The airflow adjustment value is obtained based on the spatial reference temperature change value. Specifically, the method is as follows: Obtain the spatial reference temperature change value for each spatial area within the first preset retrospective period; then call the airflow adjustment mapping table in the database. This table is established based on historical classroom cooling experiment data, recording different spatial reference temperature change ranges and their corresponding optimal airflow adjustment values; normalize the spatial reference temperature change value of the current spatial area (using a unified unit) and map it to the corresponding change range; determine the target range location of the current spatial reference temperature change value; if the current spatial reference temperature... If the degree of temperature change falls between two historical intervals, a linear interpolation method is used to calculate the corresponding airflow adjustment value. Specifically, the optimal airflow adjustment value is obtained from the nodes of the two historical intervals adjacent to the current spatial reference temperature change value, and the position ratio of the current spatial reference temperature change value between the two intervals is calculated. Based on this position ratio, the two historical airflow adjustment values ​​are weighted and calculated to obtain the optimal airflow adjustment value as the airflow adjustment value. Among them, the larger the spatial reference temperature change value, the more obvious the temperature fluctuation in the current area, so a larger airflow is needed to enhance the air circulation capacity of the area.

[0041] The air outlet temperature adjustment value is obtained based on the spatial reference temperature change value. Specifically, the method is as follows: First, obtain the spatial reference temperature change value for each spatial area within the first preset retrospective time period. Then, call the air supply temperature adjustment mapping table in the database. This table is established based on historical air conditioning cooling operation data and records different spatial reference temperature change ranges and their corresponding optimal air supply temperature adjustment values. Next, normalize the spatial reference temperature change value of the current spatial area (using a unified unit) and map it to the corresponding change range. Then, determine the target range location where the current spatial reference temperature change value is located. If the current spatial reference temperature change... If the temperature change value falls between two historical intervals, the corresponding optimal supply air temperature adjustment value is calculated using linear interpolation. Specifically, the optimal supply air temperature adjustment value is obtained from the nodes of the two historical intervals adjacent to the current spatial reference temperature change value, and the position ratio of the current spatial reference temperature change value between the two intervals is calculated. Then, based on this position ratio, the two historical supply air temperature adjustment values ​​are weighted and calculated to obtain the optimal supply air temperature adjustment value as the final outlet air temperature adjustment value. Among them, when the spatial reference temperature change value is larger, it indicates that the overall thermal state of the current area is more obvious, so the supply air temperature needs to be reduced more to improve the area's cooling capacity.

[0042] When the absolute value of the temperature change slope is greater than the temperature change slope threshold, it indicates that the temperature in the corresponding space is dropping rapidly. This suggests that the air conditioning cooling capacity is effectively covering the area, but the airflow velocity within the area is insufficient, leading to uneven heat and cold distribution in some areas. This phenomenon is usually caused by high population density, insufficient air circulation velocity, or limited air supply coverage. Therefore, when the absolute value of the temperature change slope is large, it is necessary to increase the airflow. Increasing the airflow velocity improves the air circulation velocity, allowing cold air to quickly diffuse to the surrounding areas, thereby reducing local heat accumulation and improving the overall temperature uniformity. Since the area is already experiencing a significant temperature drop, there is no need to further reduce the supply air temperature to avoid causing excessive cold in some areas due to excessively low supply air temperatures.

[0043] When the absolute value of the temperature change slope is less than the temperature change slope threshold, it indicates that the temperature in the corresponding space is decreasing slowly. This suggests that the air circulation in the area is relatively stable, but the overall cooling capacity is insufficient, resulting in low cooling efficiency despite the temperature decreasing. This phenomenon is usually caused by high indoor heat load, continuous heat dissipation from people, or insufficient supply air cooling capacity. When the absolute value of the temperature change slope is small, it is necessary to reduce the supply air temperature to enhance the cooling effect of the area by increasing the cooling capacity of the cold air itself, thereby improving the overall cooling efficiency of the area. Since the current air circulation state already meets the area coverage requirements, there is no need to further increase the air volume to avoid energy waste and discomfort for people due to excessive air volume.

[0044] This embodiment dynamically divides indoor areas based on the spatial distribution of people, enabling independent analysis of different heat accumulation areas. This avoids the problem of uneven cooling in local areas caused by traditional overall air supply methods. At the same time, by analyzing the rate of heat change in different spatial areas through temperature change slope analysis, the air volume is increased first when the temperature change rate of an area is fast, thereby enhancing air circulation capacity. When the temperature change of an area is slow but remains consistently hot, the air volume is reduced to improve cooling capacity. This allows for differentiated adjustment of air supply parameters under different scenarios, improving overall air conditioning operating efficiency and regional cooling comfort.

[0045] The specific method for obtaining the heat load change results described in this embodiment is as follows: Based on the infrared reference temperature of each image region within the first preset retrospective time period and combined with the infrared thermal imaging image, the change amplitude is analyzed to obtain the personnel heat load change parameters and personnel distribution density of each image region; the personnel heat load change parameters include personnel temperature change value, infrared temperature change rate, and change ratio of exposed limb area; a preset heat load comparison set is obtained from the database and compared with the personnel heat load change parameters and personnel distribution density of each image region to obtain the heat load comparison analysis value, and then weighted and coupled to obtain the heat load change value of the video image; the video within the first preset retrospective time period is extracted. Infrared temperature data for each spatial region in the image are collected and averaged to obtain a spatial reference temperature for each region. The spatial reference temperature is the spatial reference temperature for the face and / or the spatial reference temperature for the limbs. A preset spatial reference temperature threshold is obtained and compared with the spatial reference temperature of each spatial region. If the spatial reference temperature of any spatial region is greater than the spatial reference temperature threshold, the heat load change result is unqualified; otherwise, a heat load change judgment is performed. A preset heat load change value threshold is obtained and compared with the heat load change value. If the heat load change value is above the heat load change value threshold, the heat load change result is qualified; otherwise, the heat load change result is unqualified.

[0046] In this embodiment, the personnel distribution density refers to the personnel density within the video image area (i.e., the target area), which is obtained by dividing the actual spatial volume corresponding to the video image by the number of people that the space can accommodate.

[0047] The change in human body temperature can be obtained by subtracting the infrared reference temperature at the end of the first preset backtracking period from the current infrared reference temperature. Since the image area includes the face and limbs, the change in human body temperature can be either the face or the limbs. The rate of change in infrared temperature can be obtained by dividing the change in human body temperature by the corresponding duration. The change ratio of the exposed limb area can be obtained by subtracting the exposed limb area at the end of the first preset backtracking period from the current infrared reference temperature, obtaining the difference in exposed limb area, and then dividing this difference by the exposed limb area at the end of the first preset backtracking period to obtain the change ratio of the exposed limb area.

[0048] The heat load control set includes human temperature change control value, infrared temperature change rate control value, change ratio of exposed limb area control value, and human distribution density control value.

[0049] The specific method for obtaining the heat load change value of the video image is as follows:

[0050] ;

[0051] In the formula, This represents the change in heat load in the video image. Indicates the first Temperature variation values ​​of people in each image region Indicates the number of the image region. , This indicates the total number of image regions. This represents the control value for changes in human body temperature. Indicates the first The rate of infrared temperature change in each image region Indicates the infrared temperature change rate reference value. Indicates the first The proportion of changes in the exposed limb area in each image region This represents the control value for the percentage change in the exposed area of ​​the limbs. Indicates the first Personnel distribution density in each image region This represents the control value for population distribution density. Indicates the weight of personnel temperature change values, Indicates the weight of the rate of change of infrared temperature. This indicates the weighted proportion of changes in the exposed area of ​​the limbs. This indicates the weight of personnel distribution density.

[0052] The weights for changes in personnel temperature, infrared temperature change rate, changes in the proportion of exposed limb area, and personnel distribution density can be obtained through analysis of historical data in a database. For example, historical classroom cooling data can be extracted from the database, including historical changes in personnel temperature, historical infrared temperature change rate, historical changes in the proportion of exposed limb area, historical personnel distribution density, and corresponding historical thermal comfort assessment results. Data with historical thermal comfort assessment results indicating suitable temperature are then recorded as valid sample data. The Pearson correlation coefficients between historical personnel temperature changes, historical infrared temperature change rate, historical changes in the proportion of exposed limb area, and historical personnel distribution density and historical thermal comfort assessment results are calculated. The absolute values ​​of each Pearson correlation coefficient are normalized to obtain the corresponding weights for each parameter, namely, the weights for personnel temperature changes, infrared temperature change rate, changes in the proportion of exposed limb area, and personnel distribution density.

[0053] The specific method for performing secondary air supply adjustment described in this embodiment is as follows: based on the heat load change results, the source is traced to obtain the cause of the heat load change. The cause of the heat load change includes the influence of the space reference temperature and the influence of the heat load change. If the cause of the heat load change is the influence of the space reference temperature, then the first air supply adjustment process is performed based on the space reference temperature of each space area. If the cause of the heat load change is the influence of the heat load change, then the second air supply adjustment process is performed based on the heat load change value.

[0054] In this embodiment, the source is traced based on the heat load change results to find the cause of the heat load change. The specific method is as follows: when the heat load change result is unqualified, it is determined whether there is a situation where the space reference temperature in any space area is greater than the space reference temperature threshold. If so, the cause of the heat load change is the influence of the space reference temperature. Otherwise, the cause of the heat load change is the influence of the heat load change, and the source tracing is completed.

[0055] The causes of heat load variation refer to the types of reasons why the target cooling effect is not achieved after current air supply adjustment. Specifically, the cause of heat load variation affecting the space reference temperature means that the failure to achieve the target cooling effect is mainly due to excessively high temperatures in localized areas. The cause of heat load variation affecting the overall cooling effect means that the failure to achieve the target cooling effect is mainly due to insufficient overall heat load variation. By tracing the source of heat load variation results, different causes of air supply anomalies can be differentiated and addressed, improving the targeting of air conditioning parameter adjustments and reducing ineffective energy consumption.

[0056] The specific method for performing the first air supply adjustment process described in this embodiment is as follows: The spatial reference temperature of each space area is processed by difference with the preset target spatial reference temperature to obtain the personnel target temperature difference. An air supply temperature adjustment value is obtained based on the personnel target temperature difference. The current air supply temperature is reduced based on the air supply temperature adjustment value to obtain the air supply temperature adjustment execution value. Air supply processing is performed based on the air supply temperature adjustment execution value. The spatial reference temperature of each space area at the end of the third preset time period is obtained and recorded as the personnel adjustment temperature of each space area. If the personnel adjustment temperature of each space area is below the preset spatial reference temperature threshold, the first air supply adjustment process is completed; otherwise, the difference between the spatial reference temperature at the end of the third preset time period and the target spatial reference temperature is obtained and recorded as the personnel adjustment temperature difference. An air supply speed adjustment value is obtained based on the personnel adjustment temperature difference. The current air supply speed is increased based on the air supply speed adjustment value.

[0057] In this embodiment, the target temperature difference of the personnel is obtained by subtracting the infrared reference temperature from the target infrared reference temperature.

[0058] The method for obtaining the air supply temperature adjustment value based on the target temperature difference of personnel is as follows: First, obtain the target temperature difference of personnel in each spatial area. Then, call the air supply temperature adjustment mapping table in the database. This table is established based on historical air supply adjustment experimental data, recording different target temperature difference ranges for personnel and their corresponding optimal air supply temperature adjustment values. Next, normalize the current target temperature difference of personnel and map it to the corresponding target temperature difference range (normalization is performed to unify units). If the current target temperature difference of personnel is located between the boundaries of two ranges, a linear interpolation method is used to calculate the air supply temperature adjustment value. Specifically, obtain the air supply temperature adjustment values ​​in the two historical range nodes adjacent to the current difference, and calculate the position ratio of the current difference in the two ranges. Based on the position ratio, perform a weighted calculation on the two historical air supply temperature adjustment values ​​to obtain the final optimal air supply temperature adjustment value, which is then used as the air supply temperature adjustment value. The larger the target temperature difference of personnel, the more significant the deviation between the actual human thermal state and the target thermal state in the current area, and therefore, the larger the corresponding air supply temperature adjustment value.

[0059] The specific method for obtaining the supply air temperature adjustment execution value is as follows: subtract the supply air temperature adjustment value from the current supply air temperature to obtain the supply air temperature adjustment execution value.

[0060] The third preset time period refers to the continuous monitoring of future moments starting from the moment when the current air supply temperature is reduced. The monitoring duration is the preset third monitoring duration, and the third preset time period is obtained by statistical analysis.

[0061] The air supply speed adjustment value is obtained by matching the temperature difference adjusted by personnel. The specific method is as follows:

[0062] The process involves: obtaining the temperature difference for personnel at the end of the third preset time period; then calling the air supply speed adjustment mapping table in the database, which is established based on historical air supply speed adjustment experimental data and records different temperature difference intervals for personnel and their corresponding optimal air supply speed adjustment values; normalizing the current temperature difference for personnel and mapping it to the corresponding interval; determining the target interval position of the current temperature difference for personnel; if the current temperature difference for personnel is located between two historical intervals, calculating the corresponding air supply speed adjustment value using linear interpolation; specifically, obtaining the air supply speed adjustment values ​​in the two historical interval nodes adjacent to the current temperature difference for personnel, and calculating the position ratio of the current temperature difference for personnel between the two intervals; weighting the two historical air supply speed adjustment values ​​based on this position ratio to obtain the final optimal air supply speed adjustment value as the air supply speed adjustment value; where, the larger the temperature difference for personnel, the more significant the residual heat in the current area after air supply temperature adjustment, thus requiring further increase in air supply speed.

[0063] The current air supply speed is increased based on the air supply speed adjustment value. Specifically, the air supply speed is increased by adding the air supply speed adjustment value to the current air supply speed.

[0064] By prioritizing the reduction of supply air temperature to improve the cooling capacity of local areas, and determining whether the temperature adjustment of personnel in each space area is below the space reference temperature threshold, i.e., whether the temperature of each space area has been reduced to the ideal temperature after adjustment, it is determined whether the cooling adjustment is complete. If the adjustment is complete, the supply air speed will not be increased to avoid over-adjustment. If the adjustment is not complete, the supply air speed will be increased to form a progressive supply air adjustment, so as to improve the cooling efficiency of the area and avoid over-adjustment of local areas by only adjusting the supply air temperature in order to achieve the ideal cooling effect, thus causing waste of resources.

[0065] The specific method for performing the second air supply adjustment process described in this embodiment is as follows: Areas with a reference temperature above the average reference temperature of the target space are designated as high-heat areas. The average reference temperature of each high-heat area is calculated and recorded as the average temperature of personnel in the high-heat area. The proportion of the area of ​​the high-heat area to the total area of ​​the space is calculated and recorded as the high-heat area area ratio. Based on the high-heat area area ratio and the average temperature of personnel in the high-heat area, a comprehensive calorific value of the high-heat area is obtained. An airflow adjustment reference value is obtained based on the comprehensive calorific value of the high-heat area. The variance of the reference temperature of each high-heat area is calculated and compared with a variance threshold. If the variance is greater than the variance threshold, a preset airflow adjustment base value is used as the airflow adjustment execution value to increase the current airflow, and a secondary adjustment of the swing frequency is performed. If the variance is less than the variance threshold, the airflow adjustment reference value is used as the airflow adjustment execution value to increase the current airflow, and no secondary adjustment of the swing frequency is performed.

[0066] In this embodiment, the comprehensive calorific value of the high-heat area is obtained by comparing the area ratio of the high-heat area and the average temperature of people in the high-heat area with preset comparison values ​​for the area ratio and temperature of people in the high-heat area in the database. The comparison results for each high-heat area are then multiplied by their corresponding weights and summed to obtain the comprehensive calorific value of the high-heat area. The specific formula is as follows:

[0067] ;

[0068] In the formula, This indicates the overall calorific value of the high-heat region. This indicates the percentage of the area in high-heat regions. This represents the percentage of area in high-heat regions. This represents the average temperature of people in high-temperature areas. This represents the temperature control value for people in high-temperature areas. This indicates the weight of the area proportion of high-heat regions. This represents the weight of the average temperature of people in high-temperature areas.

[0069] The weights for the area proportion of high-heat regions and the weights for the average temperature of people in high-heat regions can be obtained through database matching. The specific method is as follows:

[0070] Historical high-temperature area operation data were extracted from the database, including the area proportion of historical high-temperature areas, the average temperature of people in historical high-temperature areas, and the corresponding historical cooling effect evaluation results. These evaluation results were then divided into different cooling levels. The distribution frequencies of different high-temperature area proportions and average temperatures of people in different high-temperature areas were then statistically analyzed within each cooling level. Based on these distribution frequencies, the probability of the corresponding parameters' influence on the cooling effect was calculated. The probability of influence was then normalized to obtain the weights of the high-temperature area proportion and the average temperature of people in high-temperature areas.

[0071] The reference value for adjusting the air volume is obtained based on the comprehensive calorific value matching of the high-heat zone. The specific method is as follows:

[0072] Obtain the comprehensive calorific value of each high-heat zone; then call the airflow adjustment reference mapping table in the database. This table is established based on historical high-heat zone cooling experiment data and records the comprehensive calorific value ranges of different high-heat zones and the corresponding optimal airflow adjustment reference values; normalize the current comprehensive calorific value of the high-heat zone and map it to the corresponding calorific value range; determine the target range position of the current comprehensive calorific value of the high-heat zone; if the current comprehensive calorific value of the high-heat zone is between two historical calorific value ranges, then use linear interpolation to calculate the corresponding airflow adjustment reference value. Specifically, the air volume adjustment reference values ​​are obtained from the nodes of two historical calorific value intervals adjacent to the current high-heat area's comprehensive calorific value, and the position ratio of the current high-heat area's comprehensive calorific value between the two intervals is calculated. Then, based on this position ratio, the two historical air volume adjustment reference values ​​are weighted and calculated to obtain the final optimal air volume adjustment reference value as the air volume adjustment reference value. Among them, when the comprehensive calorific value of the high-heat area is larger, it indicates that the heat accumulation degree of the current area is more obvious, so a larger air volume needs to be increased to enhance the cold air coverage capacity of the high-heat area and improve the area's cooling efficiency.

[0073] When the variance of the spatial reference temperature is greater than the variance threshold, it indicates that there is a large temperature difference between different high-heat areas. Therefore, the swing frequency is increased to enhance the coverage of cold air. This embodiment achieves dynamic air supply adjustment for different heat accumulation states by jointly analyzing the comprehensive calorific value of high-heat areas and the variance of the spatial reference temperature, thereby improving the regional cooling uniformity.

[0074] The specific method for secondary adjustment of the swing frequency described in this embodiment is as follows: the absolute difference between the air volume adjustment reference value and the air volume adjustment execution value is processed to obtain the air volume adjustment difference; the swing frequency adjustment value is obtained by matching the air volume adjustment difference; and the current swing frequency is increased based on the swing frequency adjustment value.

[0075] In this embodiment, the swing frequency adjustment value is obtained based on the air volume adjustment difference matching. Specifically, the method is as follows: Obtain the air volume adjustment difference between the air volume adjustment reference value and the air volume adjustment execution value; then call the swing frequency adjustment mapping table in the database. This mapping table is established based on historical swing control experimental data, recording different air volume adjustment difference intervals and their corresponding optimal swing frequency adjustment values; normalize the current air volume adjustment difference and map it to the corresponding interval; determine the target interval location where the current air volume adjustment difference is located; if the current air volume adjustment difference is located in two historical intervals... Between intervals, a linear interpolation method is used to calculate the corresponding swing frequency adjustment value. Specifically, the swing frequency adjustment values ​​of the two historical interval nodes adjacent to the current air volume adjustment difference are obtained, and the position ratio of the current air volume adjustment difference between the two intervals is calculated. Based on this position ratio, the two historical swing frequency adjustment values ​​are weighted and calculated to obtain the final optimal swing frequency adjustment value as the swing frequency adjustment value. Among them, when the air volume adjustment difference is larger, it indicates that the deviation between the current air supply capacity and the target air supply capacity is more obvious. Therefore, a larger swing frequency needs to be increased to enhance the cold air coverage capacity.

[0076] The current swing frequency is increased based on the swing frequency adjustment value. Specifically, the swing frequency is increased by adding the swing frequency adjustment value to the current swing frequency to obtain the adjusted swing frequency.

[0077] like Figure 3 As shown, this invention also introduces a system comprising the following modules: a data acquisition module, used to acquire in real time video images and infrared thermal imaging images of personnel in the initial operation mode of the air conditioner; a region division module, used to perform face recognition and limb detection based on the video images, and to perform region division and alignment processing in conjunction with the infrared thermal imaging images to obtain the infrared temperature of each image region, each image region including each face region and each limb region; a thermal comfort determination module, used to acquire the infrared temperature data of each image region within a first preset retrospective time period, and to perform temperature suitability condition detection to obtain a thermal comfort determination result. If the thermal comfort determination result is that the temperature is suitable, the current steady-state operation is maintained; otherwise, temperature adjustment is performed; an air supply adjustment module, used to perform temperature adjustment, specifically by extracting the regional spatial location of the video images and performing regional heat analysis to obtain the air supply execution parameters of the air conditioner, thereby performing air supply adjustment processing; and a secondary air supply adjustment module, used to acquire personnel heat load change parameters and infrared reference temperature within a second preset time period, and to analyze and obtain the heat load change result. If the heat load change result is qualified, the current air supply execution parameters are maintained; otherwise, secondary air supply adjustment is performed until the heat load change result is qualified.

[0078] It should be noted that the above content merely illustrates the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. For those skilled in the art, various improvements and modifications can be made without departing from the principle of the present invention, and all such improvements and modifications fall within the scope of protection of the claims of the present invention.

Claims

1. An infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition, characterized in that: Includes the following steps: S1 acquires real-time video and infrared thermal images of personnel during the initial operation mode of the air conditioner; S2 performs face recognition and limb detection based on video images, and performs region segmentation and alignment processing in conjunction with infrared thermal imaging images to obtain the infrared temperature of each image region, wherein each image region includes each face region and each limb region; S3 acquires infrared temperature data of each image region within the first preset retrospective time period, performs temperature suitability detection, and obtains thermal comfort judgment results. If the thermal comfort judgment results indicate that the temperature is suitable, the current steady-state operation is maintained; otherwise, temperature adjustment is performed. The specific method for performing temperature adjustment is as follows: extract the regional spatial location of the video image, perform regional heat analysis, obtain the air supply execution parameters of the air conditioner, and then perform air supply adjustment processing. S4 acquires the personnel heat load change parameters within the second preset time period and analyzes the heat load change results. If the heat load change results are qualified, the current air supply execution parameters are maintained; otherwise, a secondary air supply adjustment is performed until the heat load change results are qualified.

2. The infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition according to claim 1, characterized in that: Step S2 describes obtaining the infrared temperature data for each image region. The specific method is as follows: Perform face recognition and limb detection based on video images to identify the face and limb regions of each person; The infrared thermal imaging images and video images were spatially aligned to obtain the infrared temperature data of each face region and each limb region. The face region and the limb region are jointly labeled as each image region; The infrared temperature data specifically refers to the infrared temperature value of each pixel in each image region.

3. The infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition according to claim 1, characterized in that: Step S3 yields the thermal comfort assessment result, specifically through the following method: The infrared temperature data of each image region in the current video image are averaged to obtain the average infrared temperature of each image region, which is used as the infrared reference temperature of each image region. The number of facial regions whose infrared reference temperature does not fall within the preset facial infrared temperature range is counted and recorded as the number of people with overheated faces. The number of limb regions whose infrared reference temperature does not fall within the preset limb infrared temperature range is counted and recorded as the number of limbs with overheating. The number of people with facial and limb overheating is compared with the corresponding thresholds. If the number of people with facial and limb overheating is less than the corresponding threshold, the thermal comfort judgment result is that the temperature is suitable; otherwise, the thermal comfort judgment result is that the temperature is unsuitable.

4. The infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition according to claim 1, characterized in that: Step S3 obtains the air supply execution parameters of the air conditioner, specifically through the following method: The video images within the first preset retrospective time period are divided into spatial regions based on the spatial distribution of people, resulting in each spatial region of each video image; Spatial alignment is performed between each spatial region and each image region to obtain the spatial reference temperature of each spatial region. The spatial reference temperature of each spatial region during the first preset retrospective period is analyzed to obtain the spatial reference temperature change value of each spatial region and the corresponding temperature change slope. Based on the temperature change slope and compared with a preset temperature change slope threshold, if the absolute value of the temperature change slope is greater than the temperature change slope threshold, then the air volume adjustment value is obtained by matching the spatial reference temperature change value, thereby increasing the air volume. If the absolute value of the temperature change slope is less than the temperature change slope threshold, the outlet air temperature adjustment value is obtained based on the spatial reference temperature change value, thereby reducing the outlet air temperature. The air volume and air temperature are jointly marked as the air supply execution parameters.

5. The infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition according to claim 1, characterized in that: The heat load change results mentioned in step S4 are specifically obtained through the following method: Based on the infrared reference temperature of each image region within the first preset retrospective period and combined with the infrared thermal imaging image, the change amplitude analysis is performed to obtain the personnel heat load change parameters and personnel distribution density of each image region. The parameters for changes in human thermal load include changes in human temperature, rate of change in infrared temperature, and percentage change in the exposed area of ​​limbs. Obtain a preset heat load comparison set from the database and compare it with the heat load change parameters and population distribution density of each image area to obtain the heat load comparison analysis value. Then, perform weighted coupling processing to obtain the heat load change value of the video image. Infrared temperature data of each spatial region in the video images within the first preset retrospective time period are extracted and averaged to obtain the spatial reference temperature of each spatial region. The spatial reference temperature is the face spatial reference temperature and / or the limb spatial reference temperature; Obtain the preset space reference temperature threshold and compare it with the space reference temperature of each space region. If the space reference temperature of any space region is greater than the space reference temperature threshold, the heat load change result is unqualified; otherwise, perform heat load change judgment. Obtain the preset heat load change value threshold and compare it with the heat load change value. If the heat load change value is above the heat load change value threshold, the heat load change result is qualified; otherwise, the heat load change result is unqualified.

6. The infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition according to claim 1, characterized in that: Step S4 involves performing secondary air supply adjustment, specifically as follows: The causes of heat load changes are traced based on the results of heat load changes, including the influence of space reference temperature and the influence of heat load changes. If the cause of the heat load change is the influence of the space reference temperature, then the first air supply adjustment process shall be performed based on the space reference temperature of each space area. If the cause of the heat load change is the heat load change itself, then the second air supply adjustment process will be performed based on the heat load change value.

7. The infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition according to claim 6, characterized in that: The specific method for performing the first air supply adjustment process is as follows: The spatial reference temperature of each space area is compared with the preset target spatial reference temperature to obtain the target temperature difference for personnel. The supply air temperature adjustment value is obtained based on the target temperature difference for personnel. The current supply air temperature is reduced based on the supply air temperature adjustment value to obtain the supply air temperature adjustment execution value. Based on the air supply temperature adjustment execution value, air supply processing is performed, the spatial reference temperature of each spatial area at the end of the third preset time period is obtained, and recorded as the personnel adjustment temperature of each spatial area; If the temperature adjustment of personnel in each space area is below the preset space reference temperature threshold, the first air supply adjustment process is completed; otherwise, the difference between the space reference temperature and the target space reference temperature at the end of the third preset time period is obtained and recorded as the personnel adjustment temperature difference. The air supply speed adjustment value is obtained by matching the temperature difference adjusted by personnel, and the current air supply speed is increased based on the air supply speed adjustment value.

8. The infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition according to claim 6, characterized in that: The specific method for performing the second air supply adjustment process is as follows: Spatial areas with a spatial reference temperature above the average target spatial reference temperature are designated as high-heat areas. The average spatial reference temperature of each high-heat area is calculated and recorded as the average temperature of people in the high-heat area. The proportion of the area of ​​the high-heat region to the total area of ​​the space region is recorded as the high-heat region area ratio. Based on the analysis of the area ratio of high-heat regions and the average temperature of people in high-heat regions, the comprehensive calorific value of high-heat regions is obtained. The reference value for adjusting the air volume is obtained based on the comprehensive calorific value matching of the high-heat zone; The variance of the spatial reference temperature in each high-heat zone is calculated and compared with the variance threshold. If the variance value is greater than the variance threshold, the preset air volume adjustment base value is used as the air volume adjustment execution value to increase the current air volume, and the swing frequency is adjusted a second time. If the variance value is below the variance threshold, the air volume adjustment reference value will be used as the air volume adjustment execution value to increase the current air volume, and the swing frequency will not be adjusted again.

9. The infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition according to claim 8, characterized in that: The specific method for performing a secondary adjustment of the swing frequency is as follows: The absolute difference between the air volume adjustment reference value and the air volume adjustment execution value is processed to obtain the air volume adjustment difference value. The swing frequency adjustment value is obtained by matching the air volume adjustment difference, and the current swing frequency is increased based on the swing frequency adjustment value.

10. A system applying the infrared air conditioning energy-saving control method based on crowd characteristics and spatial recognition as described in any one of claims 1-9, characterized in that: Includes the following modules: The data acquisition module is used to acquire video images and infrared thermal images of personnel in the initial operation mode of the air conditioner in real time; The region segmentation module is used to perform face recognition and limb detection based on video images, and to perform region segmentation and alignment processing in conjunction with infrared thermal imaging images to obtain the infrared temperature of each image region, wherein each image region includes each face region and each limb region; The thermal comfort determination module is used to acquire infrared temperature data of each image area within the first preset retrospective period, and to detect the temperature suitability conditions to obtain the thermal comfort determination result. If the thermal comfort determination result is that the temperature is suitable, the current steady-state operation is maintained; otherwise, temperature adjustment is performed. The air supply adjustment module is used to perform temperature adjustment. The specific method is as follows: extract the regional spatial location of the video image, perform regional heat analysis, obtain the air supply execution parameters of the air conditioner, and then perform air supply adjustment processing. The secondary air supply adjustment module is used to obtain the personnel heat load change parameters and infrared reference temperature within the second preset time period, and analyze the heat load change results. If the heat load change results are qualified, the current air supply execution parameters are maintained; otherwise, secondary air supply adjustment is performed until the heat load change results are qualified.

Images