Dispersion type intelligent oxygen supply equipment and dynamic control system

Abstract

The invention relates to the technical field of oxygen supply control, in particular to dispersion type intelligent oxygen supply equipment and a dynamic control system. The equipment comprises a memory and a processor, and the processor executes a computer program stored in the memory so as to realize the following steps: acquiring oxygen concentrations and monitoring images of different monitoring points in an oxygen supply space; obtaining abnormal factors according to the oxygen concentration and position of each monitoring point, and determining key points and non-key points; according to the motion condition of the target person in the monitoring image of the area where each key point is located, the feature angular point and the distance between the target person and the key point, the activity frequency degree is obtained; determining a reference coefficient of the key point in combination with the staying duration, the activity frequency degree and the abnormal factor of each target person in the region where each key point is located; and determining comprehensive oxygen concentration data according to the reference coefficient of the key point, the oxygen concentration of the key point and the oxygen concentration of the non-key point, and adjusting the oxygen supply concentration of the oxygen supply equipment. The dynamic control precision of the oxygen concentration is improved.

Description

Technical Field

[0001] This invention relates to the field of oxygen supply control technology, specifically to a diffused intelligent oxygen supply device and dynamic control system. Background Technology

[0002] In high-altitude areas, air density is often lower due to altitude, causing partial hypoxia in the body's cells and leading to altitude sickness. Prolonged severe hypoxia can even cause pulmonary heart disease and mental and neurological symptoms, with serious consequences, especially for people from low-lying areas who are less tolerant of high altitudes. Therefore, some residences provide oxygen supply services, such as diffused oxygen supply systems, primarily used in hypoxic environments. Diffuse oxygen supply mainly alters the oxygen content within a specific space, providing a stable oxygen-rich environment for that space (such as residences in high-altitude areas) to improve the oxygen content in the air.

[0003] Oxygen sensor measurement accuracy is easily affected by static electricity, human activity, ventilation, etc. When monitoring oxygen concentration in a living room, the average oxygen concentration data of all monitoring points in the living room at the same time is generally obtained to reflect the overall oxygen concentration distribution in the living room. However, when multiple monitoring points experience oxygen concentration fluctuations due to local factors (such as ventilation, human activity, etc.), the oxygen concentration monitoring data will be inaccurate, causing the equipment to make incorrect oxygen adjustment decisions and reducing the accuracy of oxygen supply equipment in controlling oxygen concentration in the living room of a high-altitude residence. Summary of the Invention

[0004] To address the problem that existing methods for supplying oxygen to a space often result in inaccurate oxygen concentration monitoring, leading to incorrect oxygen regulation decisions by the equipment, this invention aims to provide a diffused intelligent oxygen supply device and dynamic control system. The specific technical solution adopted is as follows:

[0005] In a first aspect, the present invention provides a diffusion-type intelligent oxygen supply device, the device comprising a memory and a processor, the processor executing a computer program stored in the memory to achieve the following steps:

[0006] Acquire oxygen concentrations at different monitoring points within the oxygen supply space and monitoring images of the areas where these monitoring points are located;

[0007] Based on the fluctuation of oxygen concentration at each monitoring point within the current time period, the oxygen concentration fluctuation coefficient of each monitoring point is obtained; combined with the differences in the characteristics of oxygen concentration change at different monitoring points within the current time period, the relative distance, and the oxygen concentration fluctuation coefficient, the abnormal factors of oxygen concentration at each monitoring point are obtained, and key points and non-key points are determined.

[0008] Based on the movement of target personnel, characteristic corner points, and distance between target personnel and key points in the monitoring images of the areas where each key point is located within the current time period, the activity frequency of each target personnel in the area where each key point is located is obtained; combined with the dwell time, activity frequency, and abnormal factors of each target personnel in the area where each key point is located, the reference coefficient of the key point is determined; based on the reference coefficient of the key point, the oxygen concentration of the key point at the current time, and the oxygen concentration of non-key points, the comprehensive oxygen concentration data is determined.

[0009] Based on the comprehensive oxygen concentration data, adjust the oxygen supply concentration of the oxygen supply equipment; The abnormal factors of oxygen concentration at each monitoring point are obtained by combining the differences in oxygen concentration variation characteristics, relative distances, and oxygen concentration fluctuation coefficients at different monitoring points within the current time period, including: For any monitoring point: Based on the DTW distance between the second difference sequence of the oxygen concentration sequence of any monitoring point and the second difference sequence of the oxygen concentration sequence of other monitoring points within the current time period, and the difference in the oxygen concentration fluctuation coefficient of any monitoring point and other monitoring points, the oxygen concentration fluctuation consistency factor of any monitoring point and other monitoring points is obtained. The difference in the DTW distance and the oxygen concentration fluctuation coefficient are both negatively correlated with the oxygen concentration fluctuation consistency factor. Based on the consistency factor of oxygen concentration fluctuations at any given monitoring point with all other monitoring points and the distance therefrom, the abnormal factor of oxygen concentration at any given monitoring point is obtained.

[0010] Preferably, obtaining the oxygen concentration fluctuation coefficient for each monitoring point based on the fluctuation of oxygen concentration at each monitoring point within the current time period includes:

[0011] For any monitoring point:

[0012] Obtain the extreme point of the oxygen concentration curve of any of the monitoring points, wherein the oxygen concentration curve is obtained by curve fitting of the sequence of oxygen concentrations of any of the monitoring points in the current time period.

[0013] The normalized result of the difference in oxygen concentration between each extreme point and its previous extreme point is recorded as the first characteristic value of each extreme point; the normalized result of the time interval between each extreme point and its previous extreme point is recorded as the second characteristic value of each extreme point; extreme points whose first characteristic value is greater than a preset first threshold and whose second characteristic value is less than a preset second threshold are recorded as candidate points.

[0014] According to the chronological order, the time interval between the times corresponding to two adjacent candidate points constitutes a key time interval;

[0015] The oxygen concentration fluctuation coefficient of any monitoring point is obtained based on the average time interval between all adjacent key time periods of any monitoring point in the current time period, the duration of each key time period, the average rate of change of candidate points in each key time period, and the number of candidate points of any monitoring point in the current time period.

[0016] The rate of change is the ratio of the first characteristic value to the second characteristic value.

[0017] Preferably, obtaining the anomalous factor of oxygen concentration at any monitoring point based on the consistency factor and distance of oxygen concentration fluctuation at any monitoring point with all other monitoring points includes:

[0018] The ratio between the oxygen concentration fluctuation consistency factor of any monitoring point and that of each other monitoring point and the distance between any monitoring point and the same other monitoring point is recorded as the first ratio of any monitoring point to each other monitoring point.

[0019] By combining the first ratio of any monitoring point with all other monitoring points, an anomalous factor of oxygen concentration at any monitoring point is obtained, and the first ratio is negatively correlated with the anomalous factor.

[0020] Preferably, determining key points and non-key points includes: determining monitoring points with abnormal factors greater than a preset abnormal threshold as key points, and determining monitoring points with abnormal factors less than or equal to the preset abnormal threshold as non-key points.

[0021] Preferably, the step of obtaining the activity frequency of each target person in the area of ​​each key point based on the movement of the target person, feature corner points, and the distance between the target person and the key point in the monitoring images of the area where each key point is located within the current time period includes:

[0022] For any key point:

[0023] A feature matching algorithm is used to match the feature corner points of the same target person in two adjacent frames of monitoring images in the area where any key point is located, to obtain feature corner point matching pairs; based on the positions of the two feature points in the feature corner point matching pair, the motion feature value corresponding to the feature corner point matching pair is obtained;

[0024] By combining all motion feature values ​​corresponding to each target person in the area where any key point is located, the number of feature corner points of each target person in the area where any key point is located in each frame of the monitoring image, and the distance between each target person and any key point, the activity frequency of each target person in the area where any key point is located can be obtained.

[0025] Preferably, the determination of reference coefficients for key points by combining the dwell time, activity frequency, and abnormal factors of each target person in the area where each key point is located includes:

[0026] For any key point:

[0027] By combining the dwell time and activity frequency of each target person in the area where any key point is located within the current time period, the interference value of the oxygen concentration at any key point due to the activity of the person is obtained. The dwell time and the activity frequency are both positively correlated with the interference value.

[0028] Based on the abnormal factor of oxygen concentration at any monitoring point and the interference value of oxygen concentration at any key point due to human activity, a reference coefficient for any key point is obtained. The abnormal factor is negatively correlated with the reference coefficient, and the interference value is positively correlated with the reference coefficient.

[0029] Preferably, determining the comprehensive oxygen concentration data based on the reference coefficient of the key point, the oxygen concentration of the key point at the current moment, and the oxygen concentration of non-key points includes:

[0030] The ratio between the reference coefficient of each monitoring point and the sum of the reference coefficients of all monitoring points is used as the weight of each monitoring point.

[0031] The oxygen concentration data of all monitoring points at the current time is obtained by weighting and summing the oxygen concentrations using the aforementioned weights.

[0032] Among them, the reference coefficient of non-critical points is greater than the reference coefficient of reference points, and the reference coefficient of non-critical points is a preset value.

[0033] Preferably, adjusting the oxygen supply concentration of the oxygen supply equipment based on the comprehensive oxygen concentration data includes:

[0034] The comprehensive oxygen concentration data is input into the PID controller, which then sends an adjustment command to the oxygen supply equipment to adjust the oxygen concentration.

[0035] Secondly, the present invention provides a dynamic control system for diffused intelligent oxygen supply, which is used to implement the method performed by the above-mentioned device. The system includes:

[0036] The data acquisition module is used to acquire oxygen concentrations at different monitoring points within the oxygen supply space and monitoring images of the areas where different monitoring points are located;

[0037] The classification module is used to obtain the oxygen concentration fluctuation coefficient of each monitoring point based on the fluctuation of oxygen concentration at each monitoring point within the current time period; and to obtain the abnormal factors of oxygen concentration at each monitoring point by combining the differences in the change characteristics of oxygen concentration at different monitoring points within the current time period, the relative distance, and the oxygen concentration fluctuation coefficient, and to determine the key points and non-key points.

[0038] The determination module is used to determine the activity frequency of each target person in the area of ​​each key point based on the movement of the target person in the monitoring images of the area where each key point is located within the current time period, the characteristic corner points, and the distance between the target person and the key point; combined with the dwell time, activity frequency, and abnormal factors of each target person in the area where each key point is located, the reference coefficient of the key point is determined; and based on the reference coefficient of the key point, the oxygen concentration of the key point and the oxygen concentration of non-key points at the current time, the comprehensive oxygen concentration data is determined.

[0039] The control module is used to adjust the oxygen supply concentration of the oxygen supply equipment based on the comprehensive oxygen concentration data.

[0040] The present invention has at least the following beneficial effects:

[0041] This invention first analyzes the fluctuation of oxygen concentration collected from a single monitoring point within the oxygen supply space during the current time period, obtaining the oxygen concentration fluctuation coefficient. Then, combining the differences in oxygen concentration change characteristics and relative distances between different monitoring points, it determines the anomaly factor of oxygen concentration at a single monitoring point. Based on the magnitude of the anomaly factor, the monitoring points are divided into key points and non-key points. Next, by combining the movement characteristics of target personnel in the monitoring images of the areas where each key point is located, the frequency of their activities is evaluated. Furthermore, by combining the dwell time of each target personnel in the area where each key point is located with the anomaly factor, the reference coefficient of the key point is determined. This reduces the reference degree of monitoring data from locations with high anomalies and increases the reference degree of monitoring data from normal locations, thereby obtaining comprehensive oxygen concentration data that more accurately reflects the true oxygen concentration of the oxygen supply space and improving the accuracy of dynamic oxygen concentration control. Attached Figure Description

[0042] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating the method performed by a diffusion-type intelligent oxygen supply device provided in an embodiment of the present invention. Detailed Implementation

[0044] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose, the following detailed description of a diffusion-type intelligent oxygen supply device and dynamic control system proposed according to the present invention is provided in conjunction with the accompanying drawings and preferred embodiments.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0046] The following description, in conjunction with the accompanying drawings, details the specific solution of the diffusion-type intelligent oxygen supply device and dynamic control system provided by this invention.

[0047] An example of a diffusion-type intelligent oxygen supply device:

[0048] The specific scenario addressed in this embodiment is as follows: During the process of diffused oxygen supply to a space, it is necessary to adjust the oxygen supply strategy of the oxygen supply equipment in real time based on the oxygen concentration in the space. Therefore, it is necessary to ensure that the collected oxygen concentration data is as close as possible to the actual oxygen concentration in the air. Oxygen concentration is often measured using oxygen sensors. The measurement accuracy of oxygen sensors is easily affected by interference such as static electricity and ventilation, resulting in unreliable oxygen concentration data collected by oxygen sensors in areas with high interference. Therefore, this embodiment will assign different adjustment weights to sensors with different levels of interference based on the interference received by the oxygen sensors, to obtain oxygen concentration data that can reflect the actual oxygen concentration in the space, thereby achieving accurate control of the oxygen supply strategy of the oxygen supply equipment.

[0049] This embodiment of a diffusion-type intelligent oxygen supply device includes a memory and a processor. The processor executes a computer program stored in the memory to achieve, for example, Figure 1 The steps shown are as follows:

[0050] Step S1: Obtain the oxygen concentration at different monitoring points within the oxygen supply space and the monitoring images of the areas where the different monitoring points are located.

[0051] Diffuse oxygen supply includes whole-space diffuse oxygen supply and localized space diffuse oxygen supply. Whole-space diffuse oxygen supply delivers oxygen to the entire building space through the oxygen supply terminal, increasing the oxygen concentration throughout the building to achieve the desired oxygen supply effect. Since the living room is the main space for family gatherings and social interaction in high-altitude residences, it is usually affected by multiple environmental factors, such as airflow, ventilation equipment, window opening and closing, and human activity. Oxygen sensors are subject to many interference factors, resulting in significant changes in oxygen concentration. Therefore, this embodiment uses the living room area in a high-altitude residence as an example to dynamically control the oxygen supply volume of the oxygen supply equipment; that is, the living room is the oxygen supply space.

[0052] First, multiple monitoring points are evenly set up in the oxygen supply space. Optical oxygen sensors are used to collect oxygen concentration data for the current time period. In this embodiment, the oxygen concentration is collected once per second. In specific applications, the implementer can adjust the collection frequency and the number of monitoring points according to specific circumstances; this will not be elaborated further here. Simultaneously, a high-definition camera is installed at each monitoring point to continuously collect monitoring images of the area where each monitoring point is located within the current time period. Each frame of the collected monitoring image is preprocessed, including operations such as grayscale conversion and mean filtering for noise reduction. Image preprocessing is existing technology and will not be elaborated further here. It should be noted that the monitoring images mentioned later are all preprocessed monitoring images. The current time period is the set of all historical moments with a time interval less than or equal to a preset duration, plus the current moment. In this embodiment, the preset duration is 20 minutes; in specific applications, the implementer can set this according to specific circumstances.

[0053] Thus, this embodiment has collected the oxygen concentration at each monitoring point at each moment within the current time period, as well as the monitoring images of the area where each monitoring point is located.

[0054] Step S2: Based on the fluctuation of oxygen concentration at each monitoring point within the current time period, obtain the oxygen concentration fluctuation coefficient for each monitoring point; combine the differences in oxygen concentration change characteristics of different monitoring points within the current time period, the relative distance, and the oxygen concentration fluctuation coefficient to obtain the abnormal factors of oxygen concentration at each monitoring point, and determine key points and non-key points.

[0055] If the oxygen concentration at a monitoring point suddenly and drastically jumps within a short period of time, which is not possible under normal circumstances, this indicates that the sensor itself is malfunctioning or has been interfered with by the external environment.

[0056] The following embodiment uses one monitoring point as an example for explanation. The method provided in this embodiment can be used to process other monitoring points.

[0057] Specifically, for any monitoring point:

[0058] The oxygen concentration at each monitoring point is arranged in chronological order to obtain the oxygen concentration sequence for that monitoring point. The least squares method is then used to fit the data in the oxygen concentration sequence to obtain the oxygen concentration curve for that monitoring point. The horizontal axis of the oxygen concentration curve represents the collection time, and the vertical axis represents the oxygen concentration.

[0059] Obtain the extreme points of the oxygen concentration curve at the monitoring point. The normalized result of the difference in oxygen concentration between each extreme point and its preceding extreme point is recorded as the first characteristic value of each extreme point; the normalized result of the time interval between each extreme point and its preceding extreme point is recorded as the second characteristic value of each extreme point; the ratio between the first characteristic value and the second characteristic value of each extreme point is taken as the rate of change of each extreme point. The method for obtaining the difference in oxygen concentration between two extreme points is as follows: calculate the absolute value of the difference between the oxygen concentrations of these two extreme points, and take this absolute value as the difference in oxygen concentration between the two extreme points. There are many methods for normalizing the difference in oxygen concentration and the time interval, such as the maximum-minimum normalization method, or other data normalization methods, which will not be elaborated further here. Since monitoring points with a sudden increase in fluctuation amplitude are more likely to be critical points, extreme points with a first feature value greater than a preset first threshold and a second feature value less than a preset second threshold are recorded as candidate points. In this embodiment, the preset first threshold is 0.6 and the preset second threshold is 0.4. In specific applications, implementers can set them according to specific circumstances.

[0060] Based on chronological order, the time interval between any two adjacent candidate points constitutes a key time interval. Specifically: the time interval starting from the first candidate point and ending at the second candidate point is a key time interval; the time interval starting from the third candidate point and ending at the fourth candidate point is a key time interval; the time interval starting from the fifth candidate point and ending at the sixth candidate point is a key time interval; and so on, obtaining multiple key time intervals. It should be noted that if the number of candidate points is odd, the last candidate point will not be analyzed.

[0061] The time intervals between any two adjacent key time periods within the current time period are obtained for each monitoring point. Since there is a time interval between every two adjacent key time periods, the average time interval between all adjacent key time periods within the current time period is calculated. Based on the average time interval between all adjacent key time periods within the current time period, the duration of each key time period, the average rate of change of candidate points in each key time period, and the number of candidate points for the monitoring point within the current time period, the oxygen concentration fluctuation coefficient of the monitoring point is obtained.

[0062] In this embodiment, a specific formula for calculating the oxygen concentration fluctuation coefficient is given. The oxygen concentration fluctuation coefficient at the v-th monitoring point can be expressed as:

[0063]

[0064] in, This represents the oxygen concentration fluctuation coefficient at the v-th monitoring point. This represents the number of candidate points for the v-th monitoring point within the current time period. This represents the average time interval between all adjacent key time periods for the v-th monitoring point within the current time period. This represents the number of critical time periods for the v-th monitoring point within the current time period. This represents the duration of the nth critical time period within the current time period for the vth monitoring point. This represents the average rate of change of all candidate points at the v-th monitoring point during the n-th critical time period within the current time period. This represents the normalization function.

[0065] The factors of concern for the v-th monitoring point under oxygen supply conditions are characterized by the smaller the average time interval between all adjacent critical time periods within the current time period, indicating more frequent abnormal fluctuations in oxygen concentration, requiring greater attention. Conversely, longer critical time periods indicate more instances of abnormal fluctuations, necessitating more in-depth analysis. A larger oxygen concentration fluctuation coefficient indicates a greater likelihood of sudden and significant jumps in oxygen concentration at the v-th monitoring point, requiring even greater attention.

[0066] When the living room is ventilated, the significant pressure difference between the inside and outside causes air convection, which in turn affects the oxygen concentration at various monitoring points within the living room. Oxygen concentrations at each location will fluctuate; therefore, the more consistent the fluctuations across locations, the more likely the ventilation-related fluctuations are the cause.

[0067] The following explanation will use a single monitoring point as an example.

[0068] Specifically, for any monitoring point:

[0069] The oxygen concentrations at all collection times of each monitoring point within the current time period are arranged chronologically to obtain an oxygen concentration sequence for each monitoring point. Each monitoring point has a corresponding oxygen concentration sequence. The second-order difference sequence of each oxygen concentration sequence is then calculated.

[0070] Based on the DTW distance between the second-order difference sequence of the oxygen concentration sequence at this monitoring point and the second-order difference sequence of the oxygen concentration sequences at other monitoring points, and the difference in the oxygen concentration fluctuation coefficient between this monitoring point and other monitoring points, the oxygen concentration fluctuation consistency factor between this monitoring point and other monitoring points is obtained. The difference in the DTW distance and the oxygen concentration fluctuation coefficient are both negatively correlated with the oxygen concentration fluctuation consistency factor.

[0071] The ratio between the oxygen concentration fluctuation consistency factor of this monitoring point and each other monitoring point, and the ratio between the distance of this monitoring point and the distance of this monitoring point and the distance of this monitoring point to the same other monitoring point, are denoted as the first ratio of this monitoring point to each other monitoring point. A first ratio exists between this monitoring point and each other monitoring point. The anomaly factor of the oxygen concentration at this monitoring point is obtained by combining the first ratios of this monitoring point to all other monitoring points. The first ratio is negatively correlated with the anomaly factor.

[0072] In this embodiment, the calculation formula for the anomaly factor is given. The anomaly factor of the oxygen concentration at the vth monitoring point can be expressed as:

[0073]

[0074] in, This represents the abnormal factor of oxygen concentration at the v-th monitoring point. Indicates the number of monitoring points. This represents the difference in the oxygen concentration fluctuation coefficient between the v-th monitoring point and the w-th monitoring point (excluding the v-th monitoring point). Let represent the DTW distance between the second-order difference sequence of the oxygen concentration sequence at monitoring point v and the second-order difference sequence of the oxygen concentration sequence at monitoring point w (excluding monitoring point v). This represents the distance between the v-th monitoring point and the w-th monitoring point (excluding the v-th monitoring point). This represents an exponential function with the natural constant as its base.

[0075] In this embodiment, the method for obtaining the difference between the oxygen concentration fluctuation coefficient of the vth monitoring point and the wth monitoring point other than the vth monitoring point is as follows: the absolute value of the difference between the oxygen concentration fluctuation coefficient of the vth monitoring point and the oxygen concentration fluctuation coefficient of the wth monitoring point other than the vth monitoring point is taken as the difference between the oxygen concentration fluctuation coefficient of the vth monitoring point and the wth monitoring point other than the vth monitoring point.

[0076] The smaller the difference in the oxygen concentration fluctuation coefficient between the v-th monitoring point and the w-th monitoring point (excluding the v-th monitoring point), the more similar the intensity of oxygen concentration fluctuations at the two monitoring points are, and the smaller the differences in the fluctuation patterns and amplitudes, indicating higher consistency. The larger the DTW distance between the second-order difference sequence of the oxygen concentration sequence at the v-th monitoring point and the second-order difference sequence of the oxygen concentration sequence at the w-th monitoring point (excluding the v-th monitoring point), the more significant the difference in the trends of oxygen concentration changes at the two monitoring points, indicating lower consistency in oxygen concentration fluctuations. It should be noted that the method for calculating the DTW distance is existing technology and will not be elaborated further here.

[0077] The closer the v-th monitoring point is to the w-th monitoring point, the greater the likelihood that these two monitoring points are affected by similar environmental factors. The value represents the consistency factor of oxygen concentration fluctuation between the v-th and w-th monitoring points. The larger the value, the more similar the oxygen concentration fluctuation trends at the two monitoring points, the more likely the fluctuation is caused by ventilation, and the smaller the abnormal factor of oxygen concentration. This represents the first ratio between the v-th monitoring point and the w-th monitoring point.

[0078] Using the above method, the abnormal factors of oxygen concentration at each monitoring point can be obtained. Monitoring points with abnormal factors greater than a preset abnormal threshold are identified as critical points, while those with abnormal factors less than or equal to the preset abnormal threshold are identified as non-critical points. In this embodiment, the preset abnormal threshold is 0.7. In specific applications, the implementer can set it according to the specific circumstances.

[0079] Thus, this embodiment divides all monitoring points in the oxygen supply area into two categories: critical points and non-critical points.

[0080] Step S3: Based on the movement of target personnel, characteristic corner points, and distance between target personnel and key points in the monitoring images of the areas where key points are located within the current time period, obtain the activity frequency of each target personnel in the area where each key point is located; combine the dwell time, activity frequency, and abnormal factors of each target personnel in the area where each key point is located to determine the reference coefficient of the key point; based on the reference coefficient of the key point, the oxygen concentration of the key point at the current time, and the oxygen concentration of non-key points, determine the comprehensive oxygen concentration data.

[0081] In step S2 of this embodiment, several key points were selected. The oxygen concentration fluctuations at the key points were not consistent with the oxygen concentration fluctuations at other locations, indicating that the oxygen concentration fluctuations at the key points were not caused by ventilation. Therefore, it may be due to a malfunction of the oxygen sensor or interference from human activities.

[0082] High-altitude environments are typically dry, and the risk of static electricity increases in low-humidity conditions. Static electricity can interfere with the sensitivity of components in oxygen sensors. Furthermore, living rooms, as public areas, experience frequent human activity. When people move around in the living room (such as walking, resting, or talking), they directly consume the surrounding oxygen, leading to a decrease in local oxygen concentration. For example, when a person sits for a long time in the sofa area, the oxygen concentration at that location may fluctuate due to continuous oxygen consumption; if they suddenly get up and move around, their body movement will cause airflow, which may temporarily change the distribution of the surrounding oxygen concentration.

[0083] The trained YOLO object detection algorithm is used to monitor each frame of the surveillance image, obtaining several dynamic regions in each frame of video image for each keypoint. Each dynamic region is denoted as the region where a target person is located. The pixels to be segmented are divided into two categories, namely, the labels corresponding to the training set. The labeling process is as follows: single-channel semantic labels, with pixels belonging to the background region labeled as 0 and pixels belonging to the dynamic region labeled as 1. The loss function used is the cross-entropy loss function. The input is a grayscale video image, and the output is the dynamic region. The specific training process is well-known and will not be described in detail in this embodiment.

[0084] A target tracking algorithm is used to track the target person in all consecutive frames of video images in the area where each key point is located, so as to obtain the area of ​​the same target person in several frames of monitoring images.

[0085] The following explanation will use a key point as an example.

[0086] Specifically, for any key point:

[0087] A feature point detection algorithm is used to detect each frame of the monitoring image where the key point is located, and the feature corner points in each frame of the monitoring image are obtained.

[0088] The feature matching algorithm is used to match the feature corner points of the same target person in two adjacent frames of monitoring images in the area where the key point is located, and obtain feature corner point matching pairs. The two feature corner points in the same feature corner point matching pair have similar descriptors, that is, multiple feature corner point matching pairs are obtained. Each feature corner point matching pair consists of two feature corner points, and these two feature corner points come from two adjacent frames of monitoring images respectively.

[0089] For any feature corner point matching pair, the Euclidean distance between the position coordinates of the two feature corner points in the matching pair is taken as the motion feature value corresponding to the feature corner point matching pair.

[0090] By combining all motion feature values ​​of each target person in the area where the key point is located, the number of feature corner points of each target person in the area where the key point is located in each frame of the monitoring image, and the distance between each target person and the key point, the activity frequency of each target person in the area where the key point is located can be obtained.

[0091] In this embodiment, a specific formula for calculating the frequency of activity is given. The frequency of activity of the d-th target person in the area where the g-th key point is located can be expressed as:

[0092]

[0093] in, This indicates the frequency of activity of the d-th target person in the area where the g-th key point is located. This represents the number of frames in the surveillance image of the area where the g-th key point is located, showing the d-th target person. This represents the maximum value of all motion feature values ​​corresponding to the occurrence of the d-th target person in the t-th frame of the surveillance image, within the region where the g-th key point is located. This represents the number of feature corner points of the d-th target person in the t-th frame of the surveillance image where the d-th target person appears in the area where the g-th key point is located. This represents the distance between the g-th keypoint and the d-th target person in the t-th frame of the surveillance image where the d-th target person appears in its area. This represents the normalization function.

[0094] In determining the distance between the key point and the target person, this embodiment uses the distance between the key point and the center point of the target person's body area as the distance between the key point and the target person.

[0095] It should be noted that since the motion feature value is obtained based on two adjacent monitoring images, the motion feature value obtained based on the monitoring image of frame t and the monitoring image of frame t+1 is recorded as the motion feature value corresponding to the monitoring image of frame t. In addition, the motion feature value corresponding to the monitoring image of the second to last frame is taken as the motion feature value corresponding to the monitoring image of the last frame. That is, each monitoring image has its corresponding motion feature value.

[0096] The larger the maximum value of all motion feature values ​​corresponding to the t-th frame monitoring image of the d-th target person in the area where the g-th key point is located, the greater the activity level and oxygen consumption of the target person. This is used to characterize the activity level of the d-th target person in the area where the g-th keypoint is located in the z-th video image. The closer the g-th keypoint is to the d-th target person in the t-th monitoring image where the d-th target person appears in the same area, the more directly the target person's activity will affect the air around the sensor, and the more significant the impact on the oxygen concentration at the g-th keypoint location will be, the more likely it will cause fluctuations in the collected oxygen concentration data.

[0097] Furthermore, for any key point: by combining the dwell time and activity frequency of each target person in the area where the key point is located within the current time period, the interference value of the oxygen concentration at the key point due to the activities of the people is obtained, and the dwell time and the activity frequency are both positively correlated with the interference value.

[0098] In this embodiment, a specific formula for calculating the interference value of oxygen concentration at key points due to personnel activity is given. The interference value of oxygen concentration at the g-th key point due to personnel activity can be expressed as:

[0099]

[0100] in, Let represent the disturbance value of oxygen concentration at the g-th key point caused by human activity, and D represent the number of target personnel in the area where the g-th key point is located during the current time period. This indicates the frequency of activity of the d-th target person in the area where the g-th key point is located within the current time period. This indicates the duration of time that the d-th target person stays in the area where the g-th key point is located within the current time period. This represents the normalization function.

[0101] The longer the target personnel stay in the area of ​​the g-th key point and the more frequent their activities, the more likely the fluctuation in oxygen concentration at that location is caused by personnel activities. In other words, the greater the interference value of personnel activities on the oxygen concentration of the g-th key point.

[0102] The greater the likelihood of human activity affecting the oxygen concentration at the g-th key point, the more likely the abnormal fluctuations in oxygen concentration at this point are due to frequent human activity, and the more accurately the oxygen concentration data collected at this point reflects the actual changes in oxygen concentration in the living room. Conversely, the less likely the oxygen concentration at the g-th key point is affected by human activity, the more likely the abnormal fluctuations are due to sensor malfunctions such as electrostatic discharge, and the less reliable the collected oxygen concentration data. When predicting future oxygen concentration data, the representativeness of this point should be reduced, and the accuracy of dynamically adjusting the oxygen supply of the oxygen supply equipment should be improved to ensure the comfort of people in the space.

[0103] Based on the above characteristics, this embodiment obtains the reference coefficient of the key point by using the abnormal factor of the oxygen concentration at the monitoring point and the interference value of the oxygen concentration at the key point due to human activities. The abnormal factor is negatively correlated with the reference coefficient, and the interference value is positively correlated with the reference coefficient.

[0104] In this embodiment, the specific calculation formula for the reference coefficient of the key point is given. The reference coefficient of the g-th key point can be expressed as:

[0105]

[0106] in, This represents the reference coefficient for the g-th key point. The abnormal factor representing the oxygen concentration at the g-th key point. This represents the value of oxygen concentration at the g-th critical point affected by human activity. This represents an exponential function with the natural constant as its base.

[0107] The greater the interference value of human activity on the oxygen concentration at the g-th key point, the more likely the abnormal fluctuation of the oxygen concentration at the g-th key point is caused by human activity. The degree of abnormality at this location should be smaller. In other words, the oxygen concentration at the g-th key point can better reflect the actual oxygen concentration change in the oxygen supply space, and the more representative it is. That is, the larger the reference coefficient of the g-th key point is.

[0108] The above method can be used to obtain the reference coefficient for each key point. For non-key points: the reference coefficient for non-key points is set to a preset value; wherein, the reference coefficient for non-key points is greater than the reference coefficient for reference points. In this embodiment, the preset value is 1, that is, the reference coefficient for non-key points is 1. Using the above method, the reference coefficients for each monitoring point within the oxygen supply space are obtained.

[0109] The larger the reference coefficient of any monitoring point, the greater the weight should be given to the oxygen concentration data at that location at the current moment, so as to ensure that the final comprehensive oxygen concentration data can better reflect the overall trend of oxygen concentration in the living room, rather than relying solely on the oxygen concentration data of a specific location.

[0110] Specifically, the cumulative sum of reference coefficients for all monitoring points within the oxygen supply space is calculated; then, the ratio between the reference coefficient of each monitoring point and the cumulative sum is used as the weight of each monitoring point; next, the oxygen concentration of all monitoring points at the current moment is weighted and summed using the weights of each monitoring point to obtain comprehensive oxygen concentration data.

[0111] In this embodiment, a formula for calculating the comprehensive oxygen concentration data is given, and the comprehensive oxygen concentration data can be expressed as:

[0112]

[0113] in, This represents the combined oxygen concentration data, where V represents the number of monitoring points within the oxygen supply area. This represents the reference coefficient for the v-th monitoring point. This represents the sum of the reference coefficients for all monitoring points within the oxygen supply space. This represents the oxygen concentration at the v-th monitoring point at the current time.

[0114] The weight of the v-th monitoring point is used to weight the oxygen concentration. This reduces the large fluctuations in oxygen concentration caused by local factors at individual monitoring points, as well as the impact of measurement errors or equipment failures at single monitoring points. As a result, a smoother and more accurate comprehensive oxygen concentration data that reflects the true oxygen concentration in the oxygen supply space is obtained, thus improving the accuracy of oxygen concentration control in the living room.

[0115] Step S4: Based on the comprehensive oxygen concentration data, adjust the oxygen supply concentration of the oxygen supply equipment.

[0116] After obtaining comprehensive oxygen concentration data, a PID controller is used to regulate the oxygen supply of the oxygen supply equipment.

[0117] Specifically, the comprehensive oxygen concentration data is input into the PID controller. The PID controller first calculates the deviation between the comprehensive oxygen concentration data and the target oxygen concentration value, i.e.: ,in, Indicates the target oxygen concentration value. This represents the combined oxygen concentration data. This indicates the deviation between the current overall oxygen concentration data and the target oxygen concentration value. The target oxygen concentration value is set by the implementer according to the specific circumstances.

[0118] Then, the three components of the PID controller are calculated, specifically:

[0119]

[0120] in, It is a proportional term, directly proportional to the deviation, designed to respond quickly to deviations; The proportional gain determines the strength of the proportional control.

[0121] in, This represents the integral term corresponding to the current moment. This is the integral term corresponding to the previous time step; the integral term accumulates past deviations, aiming to eliminate long-term deviations. This is the integral gain.

[0122]

[0123] in, The differential term predicts the trend of the deviation, aiming to correct the impending deviation in advance. For differential gain, It is a time interval. This represents the deviation between the overall oxygen concentration data from the previous moment and the target oxygen concentration value.

[0124] Based on the three components of the PID algorithm, the controller calculates the control quantity, i.e., the output signal. According to the output signal, the PID controller sends adjustment commands to the oxygen supply equipment to adjust the oxygen concentration. This process is existing technology and will not be elaborated further here.

[0125] Thus, the method provided in this embodiment has enabled intelligent control of the diffused oxygen supply equipment.

[0126] This embodiment first analyzes the fluctuation of oxygen concentration collected from a single monitoring point within the oxygen supply space during the current time period, obtaining the oxygen concentration fluctuation coefficient. Then, combining the differences in oxygen concentration change characteristics and relative distances between different monitoring points, the abnormal factors of oxygen concentration at a single monitoring point are determined. Based on the magnitude of the abnormal factors, the monitoring points are divided into key points and non-key points. Next, combining the movement characteristics of target personnel in the monitoring images of the areas where each key point is located, the frequency of their activities is evaluated. Combining the dwell time of each target person in the area where each key point is located with the abnormal factors, the reference coefficient of the key point is determined. This reduces the reference degree of monitoring data from locations with high abnormality and increases the reference degree of monitoring data from normal locations, thereby obtaining comprehensive oxygen concentration data that more accurately reflects the true oxygen concentration of the oxygen supply space and improving the accuracy of dynamic oxygen concentration control.

[0127] An embodiment of a dynamic control system for diffused intelligent oxygen supply:

[0128] An embodiment of the present invention provides a dynamic control system for diffused intelligent oxygen supply that may include a data acquisition module, a classification module, a determination module, and a control module.

[0129] The data acquisition module is used to acquire oxygen concentrations at different monitoring points within the oxygen supply space and monitoring images of the areas where the different monitoring points are located.

[0130] The classification module is used to obtain the oxygen concentration fluctuation coefficient of each monitoring point based on the fluctuation of oxygen concentration at each monitoring point within the current time period; and to obtain the abnormal factors of oxygen concentration at each monitoring point by combining the differences in the change characteristics of oxygen concentration at different monitoring points within the current time period, the relative distance, and the oxygen concentration fluctuation coefficient, and to determine the key points and non-key points.

[0131] The determination module is used to determine the activity frequency of each target person in the area of ​​each key point based on the movement of the target person in the monitoring images of the area where each key point is located within the current time period, the characteristic corner points, and the distance between the target person and the key point; combined with the dwell time, activity frequency, and abnormal factors of each target person in the area where each key point is located, the reference coefficient of the key point is determined; and based on the reference coefficient of the key point, the oxygen concentration of the key point and the oxygen concentration of non-key points at the current time, the comprehensive oxygen concentration data is determined.

[0132] The control module is used to adjust the oxygen supply concentration of the oxygen supply equipment based on the comprehensive oxygen concentration data.

[0133] It should be understood that the modules of the dynamic control system for diffused intelligent oxygen supply can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated hardware. Those skilled in the art will understand that the above-described methods and systems can be implemented using computer-executable instructions and / or included in processor control code, for example, on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The systems and modules of this specification can be implemented not only by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., but also by software executed by various types of processors, or by a combination of the above-described hardware circuits and software (e.g., firmware).

[0134] For more details about the above modules, please refer to other parts of this manual; they will not be repeated here.

[0135] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A diffusion-type intelligent oxygen supply device, characterized in that, The device includes a memory and a processor, the processor executing a computer program stored in the memory to perform the following steps: Acquire oxygen concentrations at different monitoring points within the oxygen supply space and monitoring images of the areas where these monitoring points are located; Based on the fluctuation of oxygen concentration at each monitoring point within the current time period, the oxygen concentration fluctuation coefficient of each monitoring point is obtained; combined with the differences in the characteristics of oxygen concentration change at different monitoring points within the current time period, the relative distance, and the oxygen concentration fluctuation coefficient, the abnormal factors of oxygen concentration at each monitoring point are obtained, and key points and non-key points are determined. Based on the movement of target personnel, characteristic corner points, and distance between target personnel and key points in the monitoring images of the areas where each key point is located within the current time period, the activity frequency of each target personnel in the area where each key point is located is obtained; combined with the dwell time, activity frequency, and abnormal factors of each target personnel in the area where each key point is located, the reference coefficient of the key point is determined; based on the reference coefficient of the key point, the oxygen concentration of the key point at the current time, and the oxygen concentration of non-key points, the comprehensive oxygen concentration data is determined. Based on the comprehensive oxygen concentration data, adjust the oxygen supply concentration of the oxygen supply equipment; The abnormal factors of oxygen concentration at each monitoring point are obtained by combining the differences in oxygen concentration variation characteristics, relative distances, and oxygen concentration fluctuation coefficients at different monitoring points within the current time period, including: For any monitoring point: Based on the DTW distance between the second difference sequence of the oxygen concentration sequence of any monitoring point and the second difference sequence of the oxygen concentration sequence of other monitoring points within the current time period, and the difference in the oxygen concentration fluctuation coefficient of any monitoring point and other monitoring points, the oxygen concentration fluctuation consistency factor of any monitoring point and other monitoring points is obtained. The difference in the DTW distance and the oxygen concentration fluctuation coefficient are both negatively correlated with the oxygen concentration fluctuation consistency factor. Based on the consistency factor of oxygen concentration fluctuations at any given monitoring point with all other monitoring points and the distance therefrom, the abnormal factor of oxygen concentration at any given monitoring point is obtained.

2. The diffusion-type intelligent oxygen supply device according to claim 1, characterized in that, The process of obtaining the oxygen concentration fluctuation coefficient for each monitoring point based on the fluctuation of oxygen concentration at each monitoring point within the current time period includes: For any monitoring point: Obtain the extreme point of the oxygen concentration curve of any of the monitoring points, wherein the oxygen concentration curve is obtained by curve fitting of the sequence of oxygen concentrations of any of the monitoring points in the current time period. The normalized result of the difference in oxygen concentration between each extreme point and its previous extreme point is recorded as the first characteristic value of each extreme point; the normalized result of the time interval between each extreme point and its previous extreme point is recorded as the second characteristic value of each extreme point; extreme points whose first characteristic value is greater than a preset first threshold and whose second characteristic value is less than a preset second threshold are recorded as candidate points. According to the chronological order, the time interval between the times corresponding to two adjacent candidate points constitutes a key time interval; The oxygen concentration fluctuation coefficient of any monitoring point is obtained based on the average time interval between all adjacent key time periods of any monitoring point in the current time period, the duration of each key time period, the average rate of change of candidate points in each key time period, and the number of candidate points of any monitoring point in the current time period. The rate of change is the ratio of the first characteristic value to the second characteristic value.

3. The diffusion-type intelligent oxygen supply device according to claim 1, characterized in that, The step of obtaining the anomalous factor of oxygen concentration at any monitoring point based on the consistency factor and distance of oxygen concentration fluctuation at any monitoring point with all other monitoring points includes: The ratio between the oxygen concentration fluctuation consistency factor of any monitoring point and that of each other monitoring point and the distance between any monitoring point and the same other monitoring point is recorded as the first ratio of any monitoring point to each other monitoring point. By combining the first ratio of any monitoring point with all other monitoring points, an anomalous factor of oxygen concentration at any monitoring point is obtained, and the first ratio is negatively correlated with the anomalous factor.

4. The diffusion-type intelligent oxygen supply device according to claim 1, characterized in that, Determining critical and non-critical points includes: identifying monitoring points with abnormal factors greater than a preset abnormal threshold as critical points, and identifying monitoring points with abnormal factors less than or equal to a preset abnormal threshold as non-critical points.

5. The diffusion-type intelligent oxygen supply device according to claim 1, characterized in that, The process of determining the activity frequency of each target person in the area of ​​each key point based on the movement of the target person in the monitoring images of the area where each key point is located within the current time period, the characteristic corner points, and the distance between the target person and the key point includes: For any key point: A feature matching algorithm is used to match the feature corner points of the same target person in two adjacent frames of monitoring images in the area where any key point is located, to obtain feature corner point matching pairs; based on the positions of the two feature points in the feature corner point matching pair, the motion feature value corresponding to the feature corner point matching pair is obtained; By combining all motion feature values ​​corresponding to each target person in the area where any key point is located, the number of feature corner points of each target person in the area where any key point is located in each frame of the monitoring image, and the distance between each target person and any key point, the activity frequency of each target person in the area where any key point is located can be obtained.

6. The diffusion-type intelligent oxygen supply device according to claim 1, characterized in that, The reference coefficients for each key point are determined by combining the dwell time, activity frequency, and abnormal factors of each target person in the area where each key point is located. For any key point: By combining the dwell time and activity frequency of each target person in the area where any key point is located within the current time period, the interference value of the oxygen concentration at any key point due to the activity of the person is obtained. The dwell time and the activity frequency are both positively correlated with the interference value. Based on the abnormal factor of oxygen concentration at any monitoring point and the interference value of oxygen concentration at any key point due to human activity, a reference coefficient for any key point is obtained. The abnormal factor is negatively correlated with the reference coefficient, and the interference value is positively correlated with the reference coefficient.

7. A diffusion-type intelligent oxygen supply device according to claim 4, characterized in that, The process of determining comprehensive oxygen concentration data based on the reference coefficients of key points, the current oxygen concentration at key points, and the oxygen concentration at non-key points includes: The ratio between the reference coefficient of each monitoring point and the sum of the reference coefficients of all monitoring points is used as the weight of each monitoring point. The oxygen concentration data of all monitoring points at the current time is obtained by weighting and summing the oxygen concentrations using the aforementioned weights. Among them, the reference coefficient of non-critical points is greater than the reference coefficient of reference points, and the reference coefficient of non-critical points is a preset value.

8. A diffusion-type intelligent oxygen supply device according to claim 1, characterized in that, The adjustment of the oxygen supply concentration of the oxygen supply equipment based on the comprehensive oxygen concentration data includes: The comprehensive oxygen concentration data is input into the PID controller, which then sends an adjustment command to the oxygen supply equipment to adjust the oxygen concentration.

9. A dynamic control system for diffused intelligent oxygen supply, the system being used to implement the steps performed by the device according to claim 1, characterized in that, The system includes: The data acquisition module is used to acquire oxygen concentrations at different monitoring points within the oxygen supply space and monitoring images of the areas where different monitoring points are located; The classification module is used to obtain the oxygen concentration fluctuation coefficient of each monitoring point based on the fluctuation of oxygen concentration at each monitoring point within the current time period; and to obtain the abnormal factors of oxygen concentration at each monitoring point by combining the differences in the characteristics of oxygen concentration change at different monitoring points within the current time period, the relative distance, and the oxygen concentration fluctuation coefficient, and to determine the key points and non-key points. The determination module is used to determine the activity frequency of each target person in the area of ​​each key point based on the movement of the target person in the monitoring images of the area where each key point is located within the current time period, the characteristic corner points, and the distance between the target person and the key point; combined with the dwell time, activity frequency, and abnormal factors of each target person in the area where each key point is located, the reference coefficient of the key point is determined; and based on the reference coefficient of the key point, the oxygen concentration of the key point and the oxygen concentration of non-key points at the current time, the comprehensive oxygen concentration data is determined. The control module is used to adjust the oxygen supply concentration of the oxygen supply equipment based on the comprehensive oxygen concentration data.

Images