Multi-dimensional air purification method and system based on AI intelligent big data in-depth analysis

Abstract

The present application relates to a multi-dimensional air purification method and system based on AI intelligent big data deep analysis, which solves the problem that it is difficult for an air purifier to achieve uniform and efficient purification of the entire indoor space. The method includes: taking a region cluster that meets the conditions as a target region cluster, and calculating the centroid position of the target region cluster as the target position of the air purifier based on the coordinates and area of ​​the sub-regions contained in the target region cluster; using a path planning algorithm to plan a moving path from the current position to the target position for the air purifier, and controlling the air purifier to move along the moving path; obtaining the size, pollutant concentration and type of the polluted area of ​​the region cluster, inputting them into a preset air purifier purification parameter adjustment model, outputting the air purifier purification parameters, and adjusting the purification parameters according to the air purifier purification parameters after the air purifier moves to the target position. The present application has the following effects: improving the efficiency and quality of indoor air purification.

Description

Technical Field

[0001] The present invention relates to the field of air purification technology, and in particular to a multidimensional air purification method and system based on AI intelligent big data in-depth analysis. Background Art

[0002] As people pay more and more attention to indoor air quality, air purification technology has been widely researched and developed. Indoor air quality not only affects people's comfort, but is also closely related to health.

[0003] Traditional air purifiers are usually placed in a fixed position and run according to a preset purification mode. They mainly remove pollutants in the air, such as particulate matter (PM2.5, PM10), harmful gases such as formaldehyde and benzene, and microorganisms such as bacteria and viruses, through the principles of filter adsorption, electrostatic adsorption or chemical decomposition. Some more advanced air purifiers are also equipped with simple air quality sensors that can automatically adjust the fan speed according to changes in the concentration of surrounding air pollutants to adapt to different air quality conditions to a certain extent. For example, when the concentration of PM2.5 in the air is detected to be increased, the fan speed is increased to speed up air circulation and filtration.

[0004] Because of the fixed placement, it is difficult for air purifiers to achieve uniform and efficient purification of the entire indoor space. For example, in large rooms or environments with complex spatial layouts, areas far away from the air purifier often cannot be fully purified, resulting in large differences in air quality at different locations in the room. Moreover, traditional air purifiers are relatively simple to adjust purification parameters. Adjusting the fan speed based solely on local air quality data often fails to accurately optimize for different types of pollutants and actual pollution conditions, and cannot fully exert their purification efficiency. Summary of the invention

[0005] In order to improve the efficiency and quality of indoor air purification, the present application provides a multi-dimensional air purification method and system based on AI intelligent big data in-depth analysis.

[0006] In the first aspect, the present application provides a multi-dimensional air purification method and system based on AI intelligent big data deep analysis, which adopts the following technical solutions:

[0007] A multi-dimensional air purification method based on AI intelligent big data in-depth analysis, including:

[0008] Get the start signal of the air purifier;

[0009] The air purifier operates in response to the start signal;

[0010] Obtain the air quality data after the air purifier runs for a preset time interval and the air quality data before the air purifier runs, analyze and obtain the purification effect of the air purifier during this operation, and analyze whether the purification effect of the air purifier during this operation meets the preset purification effect standard;

[0011] If yes, it is judged that the operation is normal;

[0012] If not, the drone preset in the air purifier is started to scan each sub-area in the room according to the preset flight path. During the scanning process, the air quality of each sub-area is synchronously detected, and the room is defined to be divided into multiple sub-areas;

[0013] Based on the air quality data of each indoor sub-area, a cluster analysis algorithm is used to find sub-area groups with similar air quality, and to determine whether there is an area cluster with a pollution level that reaches the preset pollution level standard and a proportion of area effectively covered by air purifiers that is less than a preset proportion;

[0014] If yes, the region cluster that meets the conditions is used as the target region cluster, and the centroid position of the target region cluster is calculated according to the coordinates and areas of the sub-regions contained in the target region cluster as the target position of the air purifier;

[0015] According to the preset indoor map and obstacle information, a path planning algorithm is used to plan a moving path from the current position to the target position for the air purifier, and the air purifier is controlled to move along the moving path;

[0016] The size, concentration and type of pollutants in the regional cluster are obtained, and the data are input into a preset air purifier purification parameter adjustment model, and the air purifier purification parameters are output. After the air purifier is moved to the target position, the purification parameters are adjusted according to the air purifier purification parameters;

[0017] If not, the fan speed of the air purifier is analyzed and determined based on the corresponding relationship between the ratio of the purification effect of the air purifier's current operation to the preset purification standard and the fan speed of the air purifier, and the fan speed of the air purifier is updated.

[0018] Optionally, detecting the air quality of each sub-area includes:

[0019] Obtain air quality data from monitoring points in known sub-regions, and organize and pre-process the data;

[0020] The Kriging interpolation algorithm is used to infer the estimated air quality data of other unmonitored points in the room based on the air quality data of the known sub-area monitoring points.

[0021] Optionally, based on the air quality data of each indoor sub-area, a cluster analysis algorithm is used to find sub-area groups with similar air quality, including:

[0022] The air quality data of each sub-area detected by the drone is integrated with the indoor meteorological data and indoor spatial geographic information collected at the same time, and a feature vector is constructed based on the integrated data and standardized;

[0023] Run the preset clustering algorithm and use the feature vectors of each sub-region as input data for clustering analysis. The clustering algorithm divides the sub-regions into different regional clusters based on the feature similarity between them.

[0024] Optionally, judging whether there is an area cluster whose pollution level reaches a preset pollution level standard and whose effective coverage ratio of the air purifier is less than a preset ratio includes:

[0025] From the regional clusters obtained by cluster analysis, the center position coordinates and boundary range information of each regional cluster are extracted, and the air quality data of the sub-regions within the regional cluster are obtained, including the concentration distribution of various pollutants and the pollution severity assessment results;

[0026] Obtain the current location coordinates of the air purifier, and combine the preset effective coverage range of the air purifier to mark the coverage area on the indoor space map;

[0027] For each regional cluster, analyze the proportion of its area overlapping with the air purifier coverage area;

[0028] Based on the air quality data of sub-areas within the regional cluster, the preset pollution severity standards, the proportion of uncovered areas, and the preset proportion standards for areas judged to be not effectively covered by air purifiers, it is analyzed whether there is a regional cluster that meets the preset pollution severity standards and the proportion of areas effectively covered by air purifiers is less than the preset proportion.

[0029] Optionally, the method further includes steps in parallel with controlling the air purifier to move along the moving path, specifically as follows:

[0030] When the air purifier moves along the planned route, it continuously monitors the air quality change information and indoor environment dynamic information along the way;

[0031] According to the air quality change information along the route and the dynamic information of the indoor environment, analyze whether the concentration of some pollutants exceeds the preset concentration range;

[0032] If yes, the pollutants and concentration values ​​exceeding the preset concentration range are used as input objects and input into the air purifier purification parameter adjustment model, and the adapted air purifier purification parameters are output and the purification parameters are adjusted;

[0033] If not, continue moving.

[0034] Optionally, taking the region cluster that meets the conditions as the target region cluster includes:

[0035] Analyze whether there are multiple regional clusters that meet the conditions;

[0036] If not, maintain the original setting;

[0037] If yes, then obtain the indoor personnel gathering location information, and adopt the personnel gathering area cluster evaluation index quantification method to analyze and obtain the quantitative values ​​of different evaluation indicators and add them up to calculate the priority comprehensive scores of different area clusters. The evaluation indicators include personnel density, residence time, personnel activity type and overlapping area;

[0038] According to the comprehensive priority score, the order of the regional clusters to be processed first is determined, and the target regional clusters are adjusted one by one according to the order.

[0039] Optionally, the method further includes steps after outputting the purification parameters of the air purifier, which are as follows:

[0040] The quantitative values ​​of different evaluation indicators are input into the fan speed adjustment ratio model to analyze and determine the fan speed adjustment ratio;

[0041] According to the fan speed and the fan speed adjustment ratio included in the output air purifier purification parameters, the optimized fan speed is analyzed and determined, and the optimized fan speed is used as the fan speed in this air purifier.

[0042] In the second aspect, the present application provides a multi-dimensional air purification system based on AI intelligent big data deep analysis, which adopts the following technical solutions:

[0043] A multidimensional air purification system based on AI intelligent big data deep analysis, comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement the multidimensional air purification method based on AI intelligent big data deep analysis as described in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of a multi-dimensional air purification method based on AI intelligent big data in-depth analysis in an embodiment of the present application.

[0045] Figure 2 It is a schematic diagram of a flow chart of detecting the air quality of each sub-area according to another embodiment of the present application.

[0046] Figure 3This is a flowchart of another embodiment of the present application for finding sub-area groups with similar air quality using a cluster analysis algorithm based on air quality data of each indoor sub-area.

[0047] Figure 4 It is a flowchart diagram of another embodiment of the present application for determining whether there is an area cluster whose pollution level reaches a preset pollution level standard and whose effective coverage ratio by the air purifier is less than a preset ratio. DETAILED DESCRIPTION

[0048] The present application is further described in detail below in conjunction with the accompanying drawings.

[0049] Reference Figure 1 , which is a multi-dimensional air purification method based on AI intelligent big data deep analysis disclosed in this application, including:

[0050] Step S100, obtaining a start signal of the air purifier.

[0051] Among them, the start signal of the air purifier is a command signal used to trigger the air purifier to start running. It can be generated in many ways, such as manually operating the start button on the air purifier by the user, at which time the button triggers the corresponding circuit closure and other operations, thereby generating a start signal; or remotely sending instructions through the smart home system, such as the user using the smart home APP on the mobile phone, clicking the "turn on the air purifier" button on the interface, and the APP sends the start instruction to the air purifier, so that it receives the start signal; it may also be that the sensor of the air purifier itself detects that the environment meets certain conditions and automatically triggers the start signal. For example, some air purifiers are equipped with air quality sensors. When the concentration of pollutants in the air exceeds the preset threshold, it automatically generates a start signal to start purifying the air.

[0052] In step S200, the air purifier operates in response to a start signal.

[0053] The air purifier response refers to a series of actions and state changes made by the internal circuit system, control module and various purification components of the air purifier according to the pre-set program and operation mode after receiving the start signal. The operation includes the internal fan of the air purifier starting to rotate, driving the air circulation; the purification filter starts to work, adsorbing, decomposing or filtering various pollutants in the air, thereby improving the air quality.

[0054] Step S300, obtain the air quality data after the preset time interval of the air purifier and the air quality data before the operation, analyze and obtain the purification effect of the air purifier during this operation, and analyze whether the purification effect of the air purifier during this operation reaches the preset purification effect standard. If yes, execute step S400; if no, execute step S500.

[0055] Air quality data refers to a collection of data that reflects the concentration of various pollutants in the air and related environmental parameters. It usually includes information such as the concentration of inhalable particulate matter (such as PM2.5, PM10), the concentration of gaseous pollutants (such as formaldehyde, benzene, sulfur dioxide, nitrogen oxides, TVOC, etc.), oxygen content, humidity, temperature, etc. These data are collected and monitored by various sensors inside the air purifier. Different sensors use different physical or chemical principles to detect the corresponding air quality parameters.

[0056] The preset time interval is a period of time pre-set in the air purifier operation program, which is used to regularly evaluate the purification effect of the air purifier. This time interval can be set according to factors such as the use scenario of the air purifier, purification capacity and user needs, for example, it can be set to 1 hour, 2 hours or other appropriate time lengths.

[0057] Purification effect: It is an indicator to measure the degree of improvement of air quality after an air purifier has been running for a period of time. It is calculated by comparing the changes in air quality data before and after operation. It is generally expressed as the proportion of reduction in pollutant concentration or other related quantitative methods.

[0058] Preset purification effect standard: It is the target purification degree set in the air purifier based on relevant air quality standards (such as national indoor air quality standards, etc.) or the user's expectations for air quality. For example, for PM2.5 concentration, the preset purification effect standard may be set to reduce it to less than 35 micrograms per cubic meter; for formaldehyde concentration, it may be set to reduce it to less than 0.08 milligrams per cubic meter, etc.

[0059] The specific workflow is as follows:

[0060] When the air purifier starts to work, its internal air quality sensor will continue to collect air data. After the operation reaches a preset time interval, for example, set to 2 hours, the control unit of the air purifier will obtain the air quality data at this time and the air quality data recorded before the operation. For example, before the operation, the indoor PM2.5 concentration is 80 micrograms per cubic meter, and the formaldehyde concentration is 0.15 milligrams per cubic meter. After 2 hours of operation, the PM2.5 concentration becomes 40 micrograms per cubic meter, and the formaldehyde concentration becomes 0.1 milligrams per cubic meter.

[0061] Then calculate the purification effect of this operation. The purification effect for PM2.5 is calculated as: ((80-40) / 80)×100%=50%, that is, the PM2.5 concentration is reduced by 50%; the purification effect for formaldehyde is calculated as: ((0.15-0.1) / 0.15)×100%≈33.3%. Then compare these purification effects with the preset purification standards. If the preset purification effect standard for PM2.5 is to reduce to less than 35 micrograms per cubic meter, and the preset purification effect standard for formaldehyde is to reduce to less than 0.08 milligrams per cubic meter, then this operation meets the standard for PM2.5, but does not meet the standard for formaldehyde. If the purification effect reaches the preset purification effect standard, such as the PM2.5 concentration has reached the standard and the formaldehyde concentration has also reached the standard, step S400 is executed to determine that the operation is normal; if the purification effect of any pollutant does not reach the preset purification effect standard, such as the above-mentioned formaldehyde does not reach the standard, step S500 is executed to start the drone preset in the air purifier for further detection and analysis to determine the indoor air quality distribution and take corresponding measures to optimize the air purification effect.

[0062] Step S400, determining whether the operation is normal.

[0063] Step S500, start the drone preset in the air purifier, and scan each sub-area in the room according to the preset flight path. During the scanning process, the air quality of each sub-area is synchronously detected, and the room is defined to be divided into multiple sub-areas.

[0064] The drone inside the pre-set air purifier is a small drone specially designed to be placed in a specific space or structure inside the air purifier. It has the functions of flying, data collection, and communication with the air purifier control system. Its size is relatively small so that it can be stored and deployed in the limited space of the air purifier, and it is usually equipped with a high-precision air quality sensor that can detect air quality parameters at different locations during flight.

[0065] The preset flight path is the drone flight route pre-set in the intelligent control system of the air purifier. This path is planned based on the common layout of the indoor space, the air flow pattern, and the demand for air quality detection in different areas. The purpose is to ensure that the drone can comprehensively and efficiently scan and detect various sub-areas in the room. For example, the flight path planning method may be spiral, grid or partition traversal.

[0066] Sub-areas are multiple smaller area units that are divided into indoor spaces according to certain rules or algorithms. The basis for division can be the functional division of the room (such as living room, bedroom, kitchen, etc.), spatial geometry, area size and other factors. By dividing the indoor space into sub-areas, the air quality status of different local spaces can be analyzed and understood more accurately, so as to take targeted purification measures.

[0067] The workflow of step S500 is as follows: when it is determined that the purification effect of the air purifier during this operation does not meet the preset purification standard, the drone start system inside the air purifier is activated. For example, in a certain smart air purifier, once this condition is triggered, the drone in the specific cabin at the top of the purifier will automatically open the cabin door and take off smoothly with the help of the internal ejection device.

[0068] The drone starts to fly according to a preset flight path, such as a grid-shaped flight path. It starts from the location of the air purifier and flies horizontally at a set interval, just like weaving an invisible grid in the indoor space. During the flight, the air quality sensors on the drone work synchronously to detect the air quality of each sub-area. Assuming that an ordinary three-bedroom apartment is divided into sub-areas such as the living room, master bedroom, second bedroom, kitchen, and bathroom, when the drone flies over the living room, its sensor will detect the PM2.5 concentration, formaldehyde content, TVOC concentration and other air quality data in the living room, and transmit these data back to the main control system of the air purifier in real time. When the drone completes the flight scan of the entire indoor space according to the preset path, the main control system obtains detailed air quality information of each sub-area, providing a comprehensive data basis for further analysis and optimization of purification measures.

[0069] Step S600, based on the air quality data of each sub-area in the room, a cluster analysis algorithm is used to find sub-area groups with similar air quality, and it is determined whether there is a regional cluster whose pollution level reaches the preset pollution level standard and whose effective coverage ratio by the air purifier is less than the preset ratio. If yes, step S700 is executed; if no, step SA00 is executed.

[0070] Cluster analysis algorithm is a data mining technology that aims to divide data objects with similar characteristics in a data set into the same group (cluster), so that the similarity of data objects in the same cluster is high, while the similarity of data objects between different clusters is low. In this application, it is used to find sub-area groups with similar air quality based on the air quality data of each sub-area indoors, so as to better analyze and process the air quality conditions in different areas. Common clustering algorithms include K-Means algorithm, hierarchical clustering algorithm, etc.

[0071] The pollution level reaches the preset pollution level standard, which refers to a preset limit for measuring whether the concentration of pollutants in the air quality exceeds the standard. For example, for PM2.5, the preset pollution level standard may be set at 75 micrograms per cubic meter; for formaldehyde, it may be set at 0.1 milligrams per cubic meter, etc. When the concentration of one or more pollutants in the air quality data of a sub-area exceeds the standard, the sub-area is considered to have reached the preset pollution level standard.

[0072] The effective coverage ratio of an air purifier refers to the ratio of the area of ​​a certain area that can be effectively purified by an air purifier during normal operation to the total area of ​​the area. This ratio depends on the purification range of the air purifier (usually related to factors such as fan power, outlet design, air circulation mode, etc.) and factors such as the location and spatial layout of the area. For example, if the effective purification radius of an air purifier in a certain direction is 3 meters, for a square sub-area with a side length of 5 meters, if most of the sub-area is within the purification radius, its effective coverage ratio by the air purifier may be high; conversely, if the sub-area is far away from the air purifier and only a small part is within the purification range, the ratio is low.

[0073] Step S700: taking the region cluster that meets the conditions as the target region cluster, and calculating the centroid position of the target region cluster according to the coordinates and areas of the sub-regions included in the target region cluster as the target position of the air purifier.

[0074] Target area cluster: After a series of judgments, the area clusters that need to be focused on to improve air quality are determined. These area clusters are usually characterized by high pollution levels and are not fully covered by air purifiers. They are the key objects for subsequent adjustment of air purifier locations and parameters.

[0075] Coordinates of sub-areas: The position of each sub-area in the indoor space coordinate system. The coordinate system can be established based on a fixed reference point in the room (such as a corner of the room). The coordinates can accurately determine the position of the sub-area in the room, which helps to calculate the relevant geometric properties of the area cluster and its relationship with the location of the air purifier.

[0076] Centroid position: also known as the center of gravity position, is the average position of all sub-regions contained in the regional cluster in space. Calculating the centroid position is important for determining the target moving position of the air purifier, because moving the air purifier near the centroid position can theoretically purify each sub-region within the regional cluster in a more balanced manner.

[0077] When it is determined that there is a regional cluster that meets the conditions (the pollution level reaches the preset pollution level standard and the effective coverage ratio of the air purifier is less than the preset ratio), the centroid position of the target regional cluster is calculated. For example, assuming that the target regional cluster contains three sub-regions, the coordinates of sub-region A are (x1, y1) and the area is S1; the coordinates of sub-region B are (x2, y2) and the area is S2; the coordinates of sub-region C are (x3, y3) and the area is S3.

[0078] First, calculate the total area of ​​the cluster S = S1 + S2 + S3. Then calculate the horizontal coordinate of the centroid X = (x1*S1+x2*S2+x3*S3) / S.

[0079] Similarly, calculate the vertical coordinate of the centroid Y=(y1*S1+y2*S2+y3*S3) / S. In this way, the centroid position (X,Y) of the target area cluster is obtained, which is used as the target position of the air purifier. For example, in a family home, the living room and dining room are divided into an area cluster. After calculating that the centroid position is close to the center of the living room, the air purifier will take this position as the target and prepare to move to it in order to more effectively purify the air in the area cluster and improve the overall air quality of the living room and dining room.

[0080] Step S800, based on the preset indoor map and obstacle information, use the path planning algorithm to plan a moving path from the current position to the target position for the air purifier, and control the air purifier to move along the moving path.

[0081] Preset indoor map: This is a pre-built digital map of the indoor space layout. It contains the shape, size, location of each room and area in the room, as well as relevant information such as walls, furniture, etc. It can clearly present the overall structure of the indoor space and provide a basic spatial reference for subsequent path planning.

[0082] Obstacle information: refers to the objects in the room that may hinder the movement of the air purifier, including detailed data such as the location, shape, and size of furniture (such as sofas, tables, chairs, cabinets, etc.), pillars, stairs, etc. This information will be combined with the indoor map to help accurately plan a reasonable route for the air purifier to travel, so as to avoid collisions with obstacles during movement, which may damage the air purifier or affect its movement efficiency.

[0083] Path planning algorithm: It is an algorithm based on mathematical models and computational logic. Its purpose is to find a feasible and relatively optimal moving path from the starting position to the target position for a moving object (here, the air purifier) ​​in a given map environment (including obstacles and other information). Common path planning algorithms include A* algorithm, Dijkstra algorithm, etc. These algorithms will comprehensively consider multiple factors such as distance and difficulty of passage to determine the path.

[0084] The specific workflow is as follows:

[0085] After determining the target position of the air purifier (such as the centroid position of the target area cluster calculated in step S700), the control system of the air purifier will perform path planning operations based on the preset indoor map and the acquired obstacle information.

[0086] Assuming that the A* algorithm is used to plan the path, first, the current location of the air purifier is set as the starting node, and the target location is set as the target node. Then, the algorithm starts searching for possible paths from the starting node to the target node. It evaluates the "cost" of each adjacent node (representing different spatial locations on the indoor map, such as adjacent blank areas, etc.). This "cost" takes into account the estimated distance from the node to the target location (such as Euclidean distance, etc.) and whether there are obstacles at the location of the node.

[0087] For example, in a bedroom-cum-study space, the air purifier was originally placed near the door (current location), while the target location calculated in the previous steps is near the window and desk in the room. There are furniture such as beds, desks, and wardrobes in the room as obstacles. When searching for a path, the A* algorithm will prioritize avoiding nodes where these obstacles are located, and gradually build a path by selecting nodes in blank areas that are close to the target location and not blocked by obstacles.

[0088] After the algorithm calculation and path search, a moving path from the current position of the air purifier to the target position is finally determined. For example, the path may be to move along the passage between the bed and the wall for a distance, then go around the side of the desk, and finally reach the area where the target position is. After that, the control system of the air purifier will drive the moving device of the purifier (such as the rollers at the bottom) to start moving according to this planned moving path, so that the air purifier can smoothly reach the target position, so as to better perform air purification work in the corresponding area in the future.

[0089] Step S900, obtain the size, pollutant concentration and type of the polluted area of ​​the regional cluster, input them into a preset air purifier purification parameter adjustment model, output the air purifier purification parameters, and adjust the purification parameters according to the air purifier purification parameters after the air purifier moves to the target position.

[0090] The size of the polluted area of ​​the regional cluster refers to the total area of ​​the polluted and substandard air quality space within the determined target regional cluster. For example, if the target regional cluster contains three sub-regions, one of which is polluted as a whole with an area of ​​20 square meters, and the other two sub-regions are partially polluted with the polluted areas of 5 square meters and 8 square meters respectively, then the size of the polluted area of ​​the regional cluster is 20+5+8=33 square meters.

[0091] Pollutant concentration: refers to the mass or volume ratio of various pollutants (such as PM2.5, formaldehyde, benzene, TVOC, etc.) in the target area cluster. For example, the PM2.5 concentration detected in a certain area cluster is 80 micrograms per cubic meter, and the formaldehyde concentration is 0.12 milligrams per cubic meter. These data reflect the severity of air pollution and the content level of pollutants in the area.

[0092] Type: Indicates the different types of pollutants present in the air of the target area cluster. Common indoor air pollutants include inhalable particulate matter (PM10, PM2.5, etc.), gaseous pollutants (formaldehyde, benzene, toluene, xylene, TVOC, etc.), microbial pollutants (bacteria, viruses, mold, etc.), etc. Clarifying the types of pollutants helps to select purification methods and adjust purification parameters in a targeted manner.

[0093] The preset air purifier purification parameter adjustment model is a mathematical model built based on a large amount of experimental data, air purification principles and machine learning algorithms. It uses the pollution area size, pollutant concentration and type of the regional cluster as input parameters, and outputs air purifier purification parameters suitable for the current pollution situation through internal complex algorithm calculations and data processing, such as fan speed, filter replacement reminder, purification mode selection, etc., to achieve optimal adjustment of the air purifier purification effect.

[0094] When the air purifier moves to the target location, it will first conduct detailed detection and data collection on the size, concentration and type of pollutants in the target area cluster. For example, in a newly renovated office area cluster, professional air quality detection equipment detected that the polluted area covers the entire office area, with an area of ​​100 square meters, and the formaldehyde concentration is as high as 0.2 mg per cubic meter, and the TVOC concentration is 2 mg per cubic meter. The main pollutants are formaldehyde and various volatile organic compounds (TVOC).

[0095] These data will be input into the preset air purifier purification parameter adjustment model. The model may be based on a neural network algorithm and trained and learned based on a large number of purification data from different pollution scenarios in the past. For the input pollution data of the office area cluster, the model will conduct a comprehensive analysis. For example, due to the high formaldehyde concentration and the large area of ​​the polluted area, the model may output a higher fan speed recommendation, such as increasing the fan speed to 1500 revolutions per minute to speed up air circulation and formaldehyde adsorption and filtration speed; at the same time, it may be recommended to start an enhanced mode specifically for formaldehyde purification, such as activating the formaldehyde decomposition catalyst module inside the air purifier; and estimate the service life and replacement cycle of the filter element based on the pollutant concentration and area, and prompt the user to replace the filter element after a certain period of time to ensure that the air purifier can continue to effectively purify the air in the target area cluster, reduce the concentration of pollutants, and improve the air quality so that it reaches or approaches the preset air quality standards.

[0096] Step SA00, analyzing and determining the fan speed of the air purifier according to the corresponding relationship between the ratio of the purification effect of the air purifier in this operation to the preset purification standard and the fan speed of the air purifier, and updating the fan speed of the air purifier.

[0097] After determining that the purification effect of the air purifier in this operation does not meet the preset purification standard and there is no specific pollution area cluster that requires special treatment (i.e., under the condition of executing step SA00), first calculate the ratio of the purification effect of the air purifier in this operation to the preset purification standard. For example, for formaldehyde purification, the preset standard is to reduce the concentration to 0.08 mg / m3, and after actual operation, it is reduced from 0.15 mg / m3 to 0.12 mg / m3, and the purification effect is ((0.15-0.12) / 0.15)=0.2. If the ideal purification effect corresponding to the preset standard is ((0.15-0.08) / 0.15)≈0.467, then the ratio is 0.2 / 0.467≈0.429.

[0098] Then, the fan speed is determined based on the pre-established correspondence between the ratio and the fan speed of the air purifier. This correspondence is usually obtained through a large number of experimental tests and data fitting. For example, there may be such a correspondence curve or table. When the ratio is 0.4, the corresponding fan speed is 800rpm; when the ratio is 0.5, the fan speed is 1000rpm, etc. After the fan speed corresponding to the current ratio is obtained by querying or calculating, it is used as the new target fan speed, and the fan speed setting of the air purifier is updated. For example, in the above-mentioned formaldehyde purification example, the new fan speed is determined to be 900rpm according to the correspondence. The control system of the air purifier will adjust the parameters such as the fan supply voltage or frequency to increase the fan speed to 900rpm, in the hope of improving the purification efficiency in subsequent operation and making the purification effect closer to the preset standard.

[0099] Reference Figure 2 , detecting the air quality of each sub-area includes:

[0100] Step S510, obtaining air quality data of monitoring points in known sub-regions, and performing data sorting and preprocessing.

[0101] Air quality data: covers various parameters related to air quality, including but not limited to the concentration of inhalable particulate matter (such as PM2.5, PM10), the concentration of gaseous pollutants (such as formaldehyde, benzene, sulfur dioxide, nitrogen oxides, TVOC, etc.), oxygen content, humidity, temperature and other indicators. These data are collected through special air quality sensors. Different sensors use different physical or chemical principles to achieve accurate measurement of corresponding parameters.

[0102] Data collation: Preliminary processing of the raw air quality data obtained from each known sub-region monitoring point, including checking the integrity of the data (ensuring that there is no missing or erroneous data), removing outliers (such as data that significantly deviates from the normal range due to sensor failure or sudden interference), and classifying and sorting the data for subsequent more in-depth analysis and processing.

[0103] Preprocessing: Based on data organization, further normalization operations are performed on the data. For example, data collected by different sensors are unified into the same dimension or data format, and data is smoothed to reduce noise interference. It may also include feature extraction or conversion of the data to better adapt to the subsequent data analysis algorithm or model.

[0104] The specific workflow is as follows:

[0105] When the drone scans the indoor sub-areas according to the preset path, it will first reach the locations of the known sub-area monitoring points. For example, in an office environment, known sub-area monitoring points are set in the four corners and the central area of ​​the room. When the drone flies over these monitoring points, the air quality sensor it carries starts working. For PM2.5 sensors, it may use the principle of laser scattering. When the laser shines on particulate matter in the air, scattering occurs. By detecting information such as the intensity and angle of the scattered light, the concentration of PM2.5 is calculated; for formaldehyde sensors, electrochemical sensors may be used to determine the formaldehyde concentration by using the current generated by the redox reaction of formaldehyde on the electrode surface.

[0106] After collecting the original air quality data from these monitoring points, we start to organize the data. Assuming that the PM2.5 data of a certain monitoring point has a value that is significantly higher than other times and beyond the normal range, we can compare and analyze it with the historical data of the monitoring point and the data of other monitoring points during the same period, and judge it as an abnormal value and eliminate it. Then, the air quality data of different monitoring points and different types are classified according to parameter categories (such as particulate matter data, gaseous pollutant data, environmental parameter data, etc.), and sorted according to the location of the monitoring point or the order of collection time.

[0107] Then, data preprocessing is performed. For example, PM2.5 concentration data is uniformly converted to micrograms per cubic meter, and temperature data is uniformly converted to degrees Celsius. For some data with large fluctuations, smoothing techniques such as moving average method are used to reduce the impact of data fluctuations and noise. At the same time, some data features may be extracted, such as calculating the average, maximum, and minimum values ​​of PM2.5 concentrations over a period of time, or normalizing the gaseous pollutant concentration data so that its value range is between 0 and 1, so that it can be better calculated and compared in subsequent algorithms such as cluster analysis, laying the foundation for accurately estimating the air quality estimation data of other unmonitored points indoors.

[0108] Step S520, using the Kriging interpolation algorithm, based on the air quality data of the known sub-region monitoring points, the estimated air quality data of other unmonitored points in the room are calculated.

[0109] Kriging interpolation algorithm: an interpolation method based on spatial autocorrelation, which assumes that points closer in space have stronger correlation than points farther away, and uses mathematical models to infer the data values ​​of unknown points (other unmonitored points indoors) through the observation data of known finite sample points (that is, the known sub-region monitoring points here). This algorithm is widely used in many fields such as geology, meteorology, and environmental science for interpolation estimation of spatially distributed data, and can fully consider the spatial structure characteristics of data, making the inference results relatively more reasonable and accurate.

[0110] Air quality estimation data: For indoor locations without monitoring points, the estimated values ​​of various air quality parameters (such as particulate matter concentration, gaseous pollutant concentration, etc.) are calculated through the Kriging interpolation algorithm, so as to roughly understand the air quality conditions in these areas and achieve a more comprehensive portrayal of the overall indoor air quality.

[0111] After completing step S510 of collating and preprocessing the air quality data of the known sub-region monitoring points, the Kriging interpolation algorithm is used to infer the estimated air quality data of other unmonitored points in the room.

[0112] First, we need to analyze the spatial structural characteristics of the data from the known sub-region monitoring points, that is, to determine the spatial autocorrelation and variability of the data. For example, in a three-bedroom indoor environment, assuming that several known sub-region monitoring points are set up in the bedrooms, living rooms, kitchens and other areas, and by analyzing the PM2.5 concentration data collected from these monitoring points, we find that the PM2.5 concentration values ​​are closer between monitoring points that are close to each other, showing a strong spatial correlation; and as the distance increases, the difference in concentration changes will gradually appear, which reflects the spatial variability.

[0113] Next, based on this spatial structure feature, the corresponding semivariogram model is constructed. It is the core part of the Kriging interpolation algorithm and is used to describe the law of spatial autocorrelation of data with distance. For example, different types of semivariogram models such as spherical model, exponential model or Gaussian model can be selected. The specific selection depends on the fitting of the actual data. Taking the spherical model as an example, some key parameters in the model (such as nugget value, base value, range, etc.) will be determined based on the distance between known monitoring points and the corresponding PM2.5 concentration difference data, so as to accurately describe the spatial structure of PM2.5 concentration data in the indoor environment.

[0114] Then, the constructed semivariogram model is applied to the solution of the Kriging equations. For each unmonitored point in the room where the estimated air quality data needs to be estimated, the data of the known monitoring points and the spatial position relationship between them and the unmonitored point are used to solve the Kriging equations to calculate the estimated air quality data corresponding to the unmonitored point. For example, there is an unmonitored point in the middle of the living room. Based on the PM2.5 concentration data of several known monitoring points around it (such as the monitoring points in the corner of the living room, the monitoring points near the windows, etc.) and their distances from the unmonitored point, the estimated PM2.5 concentration at the unmonitored point is calculated through the calculation of the Kriging equations, assuming it is 45 micrograms per cubic meter.

[0115] In the same way, the concentrations of other gaseous pollutants (such as formaldehyde, TVOC, etc.) and other air quality parameters can also be estimated in sequence through the Kriging interpolation algorithm, and finally the complete air quality estimation data of each unmonitored point in the room can be obtained, thereby forming a relatively comprehensive and detailed understanding of the entire indoor air quality distribution, providing important data support for further analysis of the air quality similarities in different regions and formulation of targeted air purification strategies.

[0116] Reference Figure 3 ,Based on the air quality data of each indoor sub-area, a cluster analysis algorithm is used to find out the sub-area groups with similar air quality, including:

[0117] Step S610: The air quality data of each sub-area detected by the drone is integrated with the indoor meteorological data and indoor spatial geographic information collected at the same time, and a feature vector is constructed based on the integrated data, and then standardized.

[0118] Air quality data of each sub-area detected by the drone: refers to the data information about the concentration of pollutants in the air in each sub-area (such as PM2.5, PM10, formaldehyde, benzene, TVOC, etc.) and other relevant air quality indicators (such as temperature, humidity, etc.) obtained by scanning and detecting each indoor sub-area according to the preset flight path using various air quality sensors carried by the drone. These data directly reflect the air quality conditions of different sub-areas.

[0119] Indoor meteorological data mainly covers data related to meteorological elements such as temperature, humidity, air pressure, wind speed, wind direction, etc. in the indoor environment. Although the indoor meteorological conditions are relatively stable compared to those outdoors, these factors will still affect the diffusion, reaction and distribution of air pollutants. For example, high humidity may affect the adsorption and desorption process of certain gaseous pollutants, and temperature changes will also affect the convection of the air, thereby changing the distribution of pollutants.

[0120] Indoor space geographic information: including the location coordinates, shape, area size, and relative position relationship of each sub-area in the indoor space. It describes the indoor layout from a spatial perspective and plays a key role in accurately analyzing the correlation between air quality in different sub-areas and subsequently determining the coverage relationship between air purifiers and various areas.

[0121] Feature vector: a vector representation that integrates multiple related data features and arranges them in a certain order. In this step, the data of different dimensions, such as the air quality data of each sub-region detected by the drone, indoor meteorological data, and indoor spatial geographic information, are comprehensively sorted to form a vector that can fully represent the comprehensive situation of each sub-region, which is convenient for subsequent data analysis and processing, such as input into the cluster analysis algorithm.

[0122] Standardization: It is to perform a unified and standardized operation on each element (i.e., different data features) in the constructed feature vector, with the purpose of eliminating the impact of differences in dimensions, magnitudes, etc. between different features, so that each feature has equal importance and comparability in subsequent data analysis (such as clustering algorithm operations). Common standardization methods include normalization (mapping data to the range of 0 to 1), Z-score standardization (making data conform to the standard normal distribution), etc.

[0123] The specific workflow is as follows:

[0124] After the drone completes the air quality detection work of each sub-area in the room, it collects the air quality data of each sub-area. For example, in different sub-areas of a large conference room, it is detected that the PM2.5 concentration of sub-area A is 50 micrograms per cubic meter, the formaldehyde concentration is 0.08 milligrams per cubic meter, the temperature is 25 degrees Celsius, and the humidity is 50%; the PM2.5 concentration of sub-area B is 40 micrograms per cubic meter, the formaldehyde concentration is 0.06 milligrams per cubic meter, the temperature is 24 degrees Celsius, and the humidity is 48%, etc.

[0125] At the same time, indoor meteorological data is obtained through temperature and humidity sensors, air pressure sensors and other equipment installed indoors. For example, the overall indoor air pressure is 101.3 kPa, the average wind speed is 0.2 meters per second (possibly detected by a micro anemometer), and the wind direction is roughly at a certain angle.

[0126] Combined with the indoor space geographic information, assuming that the conference room is divided into several regular rectangular sub-areas, it is known that the coordinate range of sub-area A is (x1, y1) to (x2, y2), with an area of ​​20 square meters; the coordinate range of sub-area B is (x3, y3) to (x4, y4), with an area of ​​15 square meters, and so on. The relative position relationship between the sub-areas is also clear.

[0127] Next, the data from these different sources are merged to construct a feature vector. Taking sub-area A as an example, its feature vector may be [50 (PM2.5 concentration), 0.08 (formaldehyde concentration), 25 (temperature), 50% (humidity), 101.3 (air pressure), 0.2 (wind speed), specific angle (wind direction), x1, y1, x2, y2, 20 (area)], etc. (This is just a simple illustration, and a more reasonable arrangement and selection may be made according to specific needs and algorithm requirements).

[0128] Then the constructed feature vector is standardized. If the Z-score standardization method is used, for each element in the feature vector (such as the feature of PM2.5 concentration), the mean and standard deviation of the feature in all sub-region data are first calculated, and then converted into a standardized value through the formula (element value-mean) / standard deviation). After performing such processing on each element in the feature vector, a standardized feature vector is obtained, so that when the preset clustering algorithm is subsequently run, there will be no unreasonable weight deviation between different features due to issues such as data magnitude, so that each sub-region can be clustered more accurately based on feature similarity to find sub-region groups with similar air quality.

[0129] Step S620, running a preset clustering algorithm, taking the feature vectors of each sub-region as input data for clustering analysis, and the clustering algorithm divides the sub-regions into different region clusters according to the feature similarities between the sub-regions.

[0130] Feature similarity: It is an indicator to measure the similarity of feature vectors between different sub-regions after integrating air quality data, indoor meteorological data and indoor spatial geographic information. By comparing the data differences in each dimension of the feature vector, a specific similarity measurement method (such as Euclidean distance, cosine similarity, etc.) is used to determine the similarity between sub-regions. The higher the similarity, the more commonalities these sub-regions have in terms of air quality, environmental conditions and spatial location.

[0131] The specific workflow is as follows: Taking the K-Means algorithm as an example, first determine the number of clusters K. This K value can be determined based on the functional zoning of the indoor space, previous empirical data, or through some evaluation indicators (such as the silhouette coefficient, etc.). Assume that K=3 is determined, which means that the indoor sub-areas are to be divided into 3 regional clusters. Then randomly initialize K cluster centers, which are K points in the feature vector space, and each point represents the initial center position of a cluster. For example, in the feature vector space constructed by air quality data, meteorological data, and spatial geographic information, randomly select 3 points as the initial cluster centers.

[0132] Next, for each sub-region’s feature vector, calculate its similarity with the K cluster centers (e.g., using Euclidean distance calculation, the smaller the distance, the higher the similarity). Assign each sub-region to the cluster corresponding to the cluster center closest to it. For example, if the feature vector of a sub-region has the smallest Euclidean distance with the first cluster center, then this sub-region is assigned to the first cluster.

[0133] After all sub-regions are assigned, the new cluster center of each cluster is recalculated. The new cluster center is usually the average value of all sub-region feature vectors in each dimension in the cluster. Then the distance from each sub-region feature vector to the new cluster center is calculated again, and the sub-region is reallocated to the cluster corresponding to the nearest cluster center. This process is repeated until the cluster center no longer changes significantly (that is, the position change of the cluster center in the previous and subsequent iterations is less than a preset threshold) or the preset maximum number of iterations is reached. Through such an iterative process, each sub-region is finally divided into different regional clusters according to its feature similarity. For example, the living room and the open dining room connected to it are divided into a regional cluster because they are similar in air quality, temperature, humidity and spatial location; each bedroom is divided into different regional clusters because the bedroom is relatively independent from other areas in space and the air quality may be different. Such regional cluster division results provide an important basis for further analysis of the air purification needs of different regions and the formulation of accurate purification plans.

[0134] Reference Figure 4 , to determine whether there are regional clusters where the pollution level reaches the preset pollution level standard and the effective coverage ratio of the air purifier is less than the preset ratio, including:

[0135] Step S6A0, extracting the center position coordinates and boundary range information of each regional cluster from the regional clusters obtained by cluster analysis, and simultaneously obtaining the air quality data of the sub-regions within the regional cluster, including the concentration distribution of various pollutants and the pollution severity assessment results.

[0136] Boundary range information: Information about the boundaries of the area cluster in the indoor space, including the shape of the boundary, the coordinates of key points on the boundary, etc. It clarifies the spatial extension of the regional cluster, which is of great significance for accurately determining the overlap between the regional cluster and the coverage of the air purifier and evaluating the representative range of air quality data within the regional cluster.

[0137] Pollutant concentration distribution: describes the changes in the concentration of various pollutants at different locations within the regional cluster, that is, shows the uneven distribution of pollutants within the spatial range of the regional cluster. For example, in a larger regional cluster, the formaldehyde concentration may be higher near a pollution source (such as a corner under renovation), while the formaldehyde concentration in areas far away from the pollution source is relatively low. This concentration distribution information helps to gain a deeper understanding of the complexity and differences of air quality within the regional cluster.

[0138] Pollution severity assessment results: Based on the air quality data within the regional cluster and the preset pollution severity standards, the quantitative results of the overall pollution severity of the regional cluster are obtained through a specific assessment method or model. For example, the pollution severity of the regional cluster can be divided into different levels such as light pollution, moderate pollution, and severe pollution according to factors such as whether the pollutant concentration exceeds the corresponding threshold and the degree of exceeding, and the comprehensive impact of multiple pollutants, so as to more intuitively classify and compare the air quality conditions of the regional cluster.

[0139] Step S6B0, obtaining the current position coordinates of the air purifier, and combining the preset effective coverage range of the air purifier to mark the coverage area on the indoor space map.

[0140] The current position coordinates of the air purifier: the coordinate point corresponding to the current position of the air purifier in the pre-set spatial coordinate system in the room. This coordinate system usually takes a specific fixed point in the room as the origin, such as a corner or geometric center of the room. By measuring the horizontal and vertical distance of the air purifier relative to the origin, its position coordinates in the coordinate system are determined. It is generally expressed in two-dimensional coordinates (x, y) or three-dimensional coordinates (x, y, z), which are used to accurately locate the position of the air purifier in the indoor space, so as to subsequently analyze its spatial relationship with each regional cluster.

[0141] The preset effective coverage of an air purifier is the spatial area in which the air purifier can effectively purify the air, which is predetermined based on the air purifier's technical parameters, design structure, performance characteristics and other factors. This range is usually obtained by the air purifier manufacturer through experimental testing. It may be expressed in a specific geometric shape (such as a circle, ellipse, rectangle, etc.) and defined by a certain radius from the air purifier or a specific length, width, and height. For example, the preset effective coverage of an air purifier is a circular area with a radius of 5 meters with itself as the center, which means that within this circular area, the air purifier can purify the air more effectively. Beyond this range, the purification effect may gradually weaken.

[0142] When it is necessary to determine the relationship between the regional cluster and the coverage area of ​​the air purifier, first obtain the current position coordinates of the air purifier. For example, in a smart home system, the air purifier obtains its own coordinate information in the indoor space coordinate system through the built-in positioning module (such as Bluetooth positioning, Wi-Fi positioning, or a positioning system based on an indoor sensor network). Assume that its current position coordinates are (3, 4), where the coordinate unit may be meters, which means that in the coordinate system with the lower left corner of the room as the origin, the air purifier is located at a position of 3 meters in the horizontal direction and 4 meters in the vertical direction.

[0143] Then, combined with the preset effective coverage information of the air purifier. If the preset effective coverage of the air purifier is a circular area with a radius of 4 meters, draw this circular coverage area on the indoor space map with the current position (3, 4) as the center. During the drawing process, based on the data in the indoor space map, ensure that the drawing of the coverage area conforms to the actual indoor space layout, for example, it will not exceed the boundaries of the room or pass through obstacles such as walls.

[0144] Step S6C0: for each area cluster, analyze the area ratio of the overlapping part with the air purifier coverage area.

[0145] The proportion of overlapping area: refers to the proportion of the overlapping area of ​​the regional cluster and the area covered by the air purifier to the total area of ​​the regional cluster. This ratio is a key indicator to measure the degree of purification coverage of the air purifier to the regional cluster, and can intuitively reflect how much area in the regional cluster is within the effective range of the air purifier.

[0146] After obtaining relevant information of each area cluster (such as center location coordinates, boundary range information, etc.) and clarifying the marking of the air purifier's coverage area on the indoor space map, we began to analyze the area ratio of the overlapping part with the air purifier's coverage area.

[0147] First, the total area of ​​the region cluster must be determined. For example, a region cluster contains several sub-regions with different shapes. For sub-regions with regular shapes (such as rectangular sub-regions), their areas can be calculated using the corresponding geometric area calculation formula (length × width); for sub-regions with irregular shapes, it may be necessary to divide them into multiple regular shapes or use mathematical methods such as integration to calculate their areas. The total area of ​​the region cluster is obtained by adding up the areas of each sub-region. Assume that the total area of ​​the region cluster is 50 square meters.

[0148] Next, identify the overlapping parts of the area cluster and the air purifier coverage area. This needs to be determined based on their specific locations and shapes on the indoor space map. For example, the air purifier coverage area is a circle with a radius of 4 meters, and the area cluster is a polygon-like shape. By analyzing the intersection of their boundaries, the shape of the overlapping part is determined.

[0149] Then calculate the area of ​​the overlapping part. Similarly, if the shape of the overlapping part is regular, you can directly use the corresponding geometric area formula to calculate; if it is irregular, use the appropriate mathematical method to solve it. Suppose the area of ​​the overlapping part is 20 square meters.

[0150] Finally, calculate the area ratio of the overlapping part by dividing the area of ​​the overlapping part by the total area of ​​the regional cluster, that is, 20÷50=0.4 (or 40%). This 40% is the area ratio of the overlapping part between the regional cluster and the area covered by the air purifier.

[0151] Step S6D0, based on the air quality data of the sub-areas within the regional cluster, the preset pollution severity standard, the proportion of uncovered area, and the preset standard for the proportion of area judged to be not effectively covered by the air purifier, analyze whether there is a regional cluster that meets the preset pollution severity standard and the proportion of area effectively covered by the air purifier is less than the preset proportion.

[0152] Preset pollution severity standards: are indicators that are pre-set based on relevant air quality standards, health requirements, and actual application scenarios to measure whether the regional air quality has reached the pollution level limit. For example, for indoor PM2.5 concentration, the preset pollution severity standards may be set as follows: less than 35 micrograms per cubic meter is excellent, 35-75 micrograms is good, 75-115 micrograms is light pollution, 115-150 micrograms is moderate pollution, and more than 150 micrograms is severe pollution; for formaldehyde concentration, less than 0.08 milligrams per cubic meter is qualified, and higher than this value has different degrees of pollution risks, etc., in order to judge the quality of air quality within the regional cluster.

[0153] The preset area ratio standard for judging that the area is not effectively covered by the air purifier is a predetermined threshold. When the area ratio of the area cluster not covered by the air purifier exceeds this threshold, it is considered that the area cluster is not effectively covered by the air purifier and further measures need to be taken to improve its air purification condition. For example, if this standard is set to 30%, if the area ratio of a certain area cluster not covered reaches 40%, it means that it is not effectively covered and the location or operating parameters of the air purifier may need to be adjusted.

[0154] The workflow is disclosed as follows:

[0155] After obtaining the air quality data of the sub-areas within the regional cluster, clarifying the preset pollution severity standards, calculating the proportion of uncovered areas, and knowing the preset proportion standards for areas that are not effectively covered by air purifiers, the analysis and judgment of this step begins.

[0156] First, based on the air quality data of the sub-regions within the regional cluster, the pollution status of each sub-region within the regional cluster is judged against the preset pollution severity standard. For example, the PM2.5 concentration detection value of a molecular region within a regional cluster is 90 micrograms per cubic meter. According to the pollution severity standard mentioned above, this part of the sub-region is in a state of light pollution; the formaldehyde concentration of some sub-regions reaches 0.12 milligrams per cubic meter, which is higher than the qualified standard, indicating that there is formaldehyde pollution. Based on this, it is comprehensively judged that the regional cluster as a whole has a certain degree of pollution problem.

[0157] Next, check the calculated uncovered area ratio. Assuming that the total area of ​​the cluster is 80 square meters and the area not covered by the air purifier is 30 square meters, the uncovered area ratio is 30 ÷ 80 = 0.375 (37.5%).

[0158] Then, the uncovered area ratio is compared with the preset area ratio standard that is not effectively covered by the air purifier. If the preset standard is 30%, since 37.5%>30%, this means that not only does the pollution level of this regional cluster reach the preset pollution severity standard, but the area ratio that is not effectively covered by the air purifier also exceeds the preset standard, meeting the conditions for further processing. Subsequently, it may be necessary to perform related operations such as moving the air purifier and adjusting the purification parameters to improve the air quality of the regional cluster; on the contrary, if the uncovered area ratio is less than the preset standard, it means that although there may be some pollution, the air purifier is still covering it, and additional adjustment measures may not be needed for the time being.

[0159] The multi-dimensional air purification method based on AI intelligent big data deep analysis also includes steps in parallel with controlling the air purifier to move along the moving path, as follows:

[0160] Step SB00, while the air purifier moves along the planned path, continuously monitors the air quality change information and indoor environment dynamic information along the way.

[0161] Air quality change information along the way: When the air purifier moves along the planned path, the air quality sensor on board monitors the quality data changes of the surrounding air at different locations and time points in real time. These data include dynamic changes in the concentration of various pollutants (such as PM2.5, PM10, formaldehyde, benzene, TVOC, etc.), fluctuations in oxygen content, changes in humidity and temperature, etc., reflecting the real-time status of the air quality in the area where the air purifier passes during movement and the changing trend of the air purification effect.

[0162] Indoor environment dynamic information: In addition to air quality change information, other dynamic factors related to the indoor environment, such as indoor personnel activities (personnel movement, gathering, dispersion, etc.), door and window opening and closing status (whether there are new ventilation sources or pollution sources entering the room), indoor light intensity changes (may affect certain photochemical reactions and thus indirectly affect air quality), the use status of electrical equipment (such as the operation of certain electrical appliances may generate heat, electromagnetic radiation or trace gaseous pollutants, etc.), etc. This information can be combined to more comprehensively reflect the dynamic change characteristics of the indoor environment, and these changes may affect the purification effect of the air purifier and the indoor air quality.

[0163] When the air purifier starts to move along the planned path, its internal air quality sensor and other related environmental monitoring sensors start working at the same time, continuously monitoring the air quality change information and indoor environment dynamic information along the way. For example, the air purifier moves from one corner of the living room to another. During the movement, the air quality sensor will collect a sample of the surrounding air at a certain time interval (such as 1 second or 5 seconds) and detect the PM2.5 concentration, formaldehyde concentration and other data. Assume that at the initial position, the PM2.5 concentration is 60 micrograms per cubic meter. As it moves to the center of the living room, due to the high number of people activities and the possibility of some cooking fume diffusion near the kitchen, the PM2.5 concentration gradually rises to 80 micrograms per cubic meter.

[0164] At the same time, dynamic information about the indoor environment is also being monitored. For example, the movement paths and gathering areas of people in the living room are detected through the human infrared sensor installed indoors; whether the window is open is known through the door and window sensor. If the window is open, fresh air from outside may flow in, but it may also bring in pollutants from outside; through the connection with the indoor smart electrical system, the operating status of the air conditioner (cooling, heating, ventilation, etc.) and whether new electrical equipment is turned on (for example, the operation of a printer may produce pollutants such as ozone) are obtained.

[0165] The air quality change information and indoor environment dynamic information monitored along the way will be transmitted back to the air purifier's control system in real time. The control system will organize and store this data to provide data basis for subsequent analysis of whether the concentration of some pollutants exceeds the preset concentration range and to take corresponding purification measures.

[0166] Step SC00, based on the air quality change information along the way and the indoor environment dynamic information, analyze whether there are some pollutant concentrations exceeding the preset concentration range.

[0167] Preset concentration range: The concentration limit value range set for different air pollutants. These ranges are determined based on indoor air quality standards, human health impact research, and air purifier purification performance requirements. For example, the preset range for indoor PM2.5 concentration may be set to 35-75 micrograms / cubic meter for mild pollution, and more than 75 micrograms / cubic meter for more serious pollution levels; the preset range for formaldehyde concentration may be 0.08-0.12 milligrams / cubic meter for mild excess, and more than 0.12 milligrams / cubic meter for significant excess, etc. When the detected pollutant concentration exceeds the preset range, it indicates that the air quality condition deviates from the normal or expected level and needs to be handled accordingly.

[0168] As the air purifier moves along the planned path, it continuously obtains information about changes in air quality and dynamic indoor environment along the way and transmits it to its control system. For example, when the air purifier moves from room A to room B, when passing through a certain area, the sensor detects that the PM2.5 concentration instantly rises to 90 micrograms per cubic meter. At the same time, it detects that the local temperature rises by 3 degrees Celsius and the humidity drops by 10% due to the nearby electrical equipment being turned on. All of this data is recorded and transmitted.

[0169] After receiving these data, the control system will compare and analyze the concentration data of various pollutants with the corresponding preset concentration ranges one by one. For example, the PM2.5 concentration of 90 micrograms per cubic meter detected above is compared with the preset PM2.5 concentration range and is found to exceed the upper limit of the mild pollution range of 75 micrograms per cubic meter, which determines that the concentration of some pollutants exceeds the preset concentration range.

[0170] Step SD00, if yes, then some pollutants and concentration values ​​exceeding the preset concentration range are taken as input objects, input into the air purifier purification parameter adjustment model, and the adapted air purifier purification parameters are output and the purification parameters are adjusted.

[0171] Purification parameters of air purifiers: refers to various adjustable settings that affect the purification effect and operating status of air purifiers. Mainly including fan speed, which directly determines the air circulation volume of the air purifier. The higher the speed, the larger the volume of air passing through the purifier per unit time, and the adsorption and filtration efficiency of pollutants will be improved to a certain extent, but it will also bring greater noise and energy consumption; purification mode, such as normal purification mode, sleep mode, strong purification mode, etc. The working mode and intensity of the purification components inside the air purifier (such as filter combination, electrostatic adsorption device, photocatalytic device, etc.) in different modes are different; filter replacement reminder threshold, when the pollutants adsorbed by the filter reach a certain amount, its purification efficiency will decrease. This threshold is used to remind users to replace the filter in time to maintain the effective operation of the air purifier; it may also include some parameter settings related to the purification of specific pollutants, such as catalyst activity adjustment parameters for formaldehyde purification.

[0172] Preset purification parameter adjustment model: This is an intelligent model built based on a large amount of experimental data, air purification principles and mathematical algorithms. It uses the type and concentration of pollutants detected that exceed the preset concentration range, as well as the current indoor environmental parameters (such as temperature, humidity, space size, etc.) as input information, and through internal complex calculations and logical judgments, outputs the optimal purification parameter combination required for the air purifier to more efficiently purify the excessive pollutants. For example, if it is detected that the indoor formaldehyde concentration exceeds the standard and the humidity is high, the model may output suggestions to increase the fan speed, switch to the formaldehyde enhanced purification mode, and appropriately adjust the formaldehyde purification catalyst activity parameters to adapt to this specific polluted environment and improve the purification effect.

[0173] When it is determined that the concentration of some pollutants exceeds the preset concentration range, the control system of the air purifier will start the preset purification parameter adjustment model to perform parameter adjustment operations.

[0174] First, the types of pollutants that exceed the preset concentration range and their specific concentration values ​​are analyzed in detail, and relevant parameters of the current indoor environment are obtained. For example, the indoor formaldehyde concentration is detected to be 0.15 mg per cubic meter, which exceeds the preset standard of 0.1 mg per cubic meter, and the indoor temperature is 28 degrees Celsius, the humidity is 60%, and the space area is 50 square meters.

[0175] This information is then input into a preset purification parameter adjustment model, which may contain a decision tree or neural network model trained based on a machine learning algorithm, and which performs a comprehensive analysis based on input data such as formaldehyde concentration, temperature, humidity, and space area.

[0176] Based on the calculation results of the model, the purification parameters of the air purifier are adjusted. For example, the model may determine that in this case, the fan speed needs to be increased from 800 revolutions per minute to 1200 revolutions per minute to speed up air circulation and allow formaldehyde to pass through the adsorption and decomposition devices inside the purifier faster; at the same time, the purification mode is switched to the formaldehyde enhanced purification mode, which may activate chemical adsorbents or photocatalytic decomposition devices specifically for formaldehyde and increase their working power; it may also fine-tune the activity parameters of the formaldehyde purification catalyst according to the effects of temperature and humidity on formaldehyde purification, so as to optimize its decomposition efficiency of formaldehyde in the current environment.

[0177] Step SE00, if no, continue moving.

[0178] The regional clusters that meet the conditions are taken as target regional clusters, including:

[0179] Step S710: Analyze whether there are multiple region clusters that meet the conditions.

[0180] Step S720: If no, maintain the original setting.

[0181] Step S730, if yes, obtain the indoor personnel gathering location information, and adopt the personnel gathering area cluster evaluation index quantification method, analyze and obtain the quantitative values ​​of different evaluation indicators and accumulate them, and calculate the priority comprehensive scores of different area clusters. The evaluation indicators include personnel density, residence time, personnel activity type and overlapping area.

[0182] Indoor personnel gathering location information: data about the specific locations where personnel are concentrated in indoor spaces, collected through various monitoring devices installed indoors (such as cameras, human infrared sensors, Bluetooth positioning devices, etc.). For example, in a conference room, camera image analysis shows that most people gather around the conference table. This location information can be expressed in coordinate ranges or specific area descriptions, which reflects the spatial distribution of personnel in the room and plays an important role in subsequent analysis of air purification needs.

[0183] Quantification method of cluster assessment indicators for gathering areas: This is a method that converts different attributes related to gatherings of people (such as density of people, residence time, type of activity of people, overlapping area, etc.) into quantifiable values ​​through specific calculation rules and standards. For example, for density of people, it can be quantified according to the number of people per square meter; residence time can be measured and quantified in time units such as hours and minutes; the type of activity of people can be assigned corresponding values ​​according to factors such as the possibility of pollutants generated by different activities; and the overlapping area can be calculated in square meters to obtain specific values. This quantification method facilitates subsequent comprehensive analysis and comparison.

[0184] Population density: refers to the ratio of the number of people in a specific area to the area of ​​the area, usually expressed as the number of people per square meter. For example, if there are 5 people in a 10-square-meter living room, the population density is 5÷10=0.5 people / square meter, which reflects the density of people in the area. The higher the population density, the more heat, carbon dioxide and other pollutants may be generated, and the higher the demand for air purification.

[0185] Dwell time: The length of time a person stays in a particular area, which is calculated by corresponding time recording equipment (such as intelligent monitoring systems combined with timestamps, etc.), and can be measured in minutes, hours, etc. For example, people may stay in a bedroom for several hours, while they may stay in a hallway for a short time. Long dwell time means that the air quality in the area is affected by human activities for a long time, which is an important factor in air purification priority considerations.

[0186] Type of human activity: A classification method that classifies and assigns corresponding values ​​according to the different activities that people engage in indoors. Different types of activities have different degrees of impact on air quality. For example, low-intensity activities such as quiet reading have little impact on air quality and can be assigned a lower value; while activities such as cooking and exercise will generate more pollutants such as heat, water vapor and odor, which have a greater impact on air quality and can be assigned a higher value, in order to reflect the different needs for air purification in different activities.

[0187] Overlapping area: refers to the size of the overlapping area between the gathering area and the divided air purification area cluster, in square meters. For example, a gathering area and an area cluster partially overlap, and the area of ​​the overlapping part is calculated through spatial geometry analysis. The larger the overlapping area, the more closely related the area cluster is to the activities of the personnel, and the more important it needs to be considered in the arrangement of air purification.

[0188] When the "yes" condition is met (which may be based on some preset trigger conditions, such as the number of people in the room reaching a certain threshold), the information about the indoor gathering locations of people will be obtained. Assuming that in a large office space, information is collected through cameras installed on the ceiling and human infrared sensors distributed in various corners. After image recognition and data analysis, it is known that people mainly gather in the office area near the window and the conference room. The specific gathering location information of these two places is determined by coordinate ranges and other forms.

[0189] Then, a quantitative analysis is conducted using the quantitative method of cluster evaluation indicators in personnel gathering areas.

[0190] In terms of personnel density: for the office area near the window, the area is 50 square meters, and there are 20 people working there. The quantitative value of personnel density is 20÷50=0.4 people / square meter; the meeting room area is 30 square meters, and there are 15 people in a meeting. The quantitative value of personnel density is 15÷30=0.5 people / square meter.

[0191] In terms of residence time: after the system records the activity time of personnel, personnel in the window office area stay here for an average of 8 hours a day, which is converted into minutes as 8×60=480 minutes, which is used as the quantitative value of the residence time in this area; personnel in the conference room generally stay for 2 hours for meetings, which is converted into 2×60=120 minutes as its quantitative value.

[0192] In terms of personnel activity types: office work is a relatively low-intensity activity, so the window office area is assigned a value of 1; while during meetings, communication between personnel and the use of electronic devices generate relatively more pollutants, so the conference room is assigned a value of 2 as a quantitative value.

[0193] In terms of overlapping area: Assuming that after spatial analysis, the overlapping area between the divided area cluster A and the window office area is 30 square meters, and the overlapping area between the area cluster B and the conference room is 20 square meters. These are the quantitative values ​​of their respective overlapping areas.

[0194] Then, the quantitative values ​​of these different evaluation indicators are accumulated according to certain rules to calculate the priority comprehensive scores of different regional clusters. For example, for regional cluster A, the comprehensive score = quantitative value of personnel density (assuming it is 0.4) + quantitative value of residence time (assuming it is 10 after conversion according to a certain ratio) + quantitative value of personnel activity type (1) + quantitative value of overlapping area (30, assuming it is 5 after conversion) = 16.4; for regional cluster B, the comprehensive score = quantitative value of personnel density (0.5) + quantitative value of residence time (assuming it is 3 after conversion according to a certain ratio) + quantitative value of personnel activity type (2) + quantitative value of overlapping area (20, 3 after conversion) = 8.5.

[0195] Through such calculation, we can get the comprehensive priority scores of different regional clusters. By comparing the scores, we can intuitively judge that the air purification priority of regional cluster A is relatively higher. Regional cluster A will be given priority in the subsequent air purification resource allocation and air purifier location arrangement. In this way, the air purification needs of areas where people gather can be better met, ensuring that the indoor air quality meets good standards and providing people with a more comfortable and healthy indoor environment.

[0196] Step S740: Determine the order of the region clusters to be processed first according to the priority comprehensive score, and adjust the target region clusters one by one according to the order.

[0197] First, obtain the priority comprehensive score of each regional cluster. For example, suppose there are four regional clusters: regional cluster A, regional cluster B, regional cluster C, and regional cluster D. After calculation in the previous steps, their respective priority comprehensive scores are: regional cluster A scores 20 points, regional cluster B scores 15 points, regional cluster C scores 12 points, and regional cluster D scores 8 points.

[0198] Then, the priority scores are sorted. Since the higher the score, the higher the priority, the priority is sorted as follows: regional cluster A>regional cluster B>regional cluster C>regional cluster D, which forms a processing order list of regional clusters from the highest priority to the lowest priority.

[0199] Next, adjust the target area clusters one by one according to this order. The control system of the air purifier will first set area cluster A as the target area cluster to be processed with the highest priority. Based on the relevant information of area cluster A (such as its center position coordinates, boundary range, pollution level, etc.), combined with the current position of the air purifier, the preset effective coverage range and other factors, plan to move the air purifier to the best position to effectively purify area cluster A, or adjust the purification parameters of the air purifier (such as fan speed, purification mode, etc.) to enable it to better carry out purification work according to the air quality conditions of area cluster A.

[0200] When the air purification work of regional cluster A reaches a certain stage and meets some preset conditions (such as the air quality has improved after a period of purification, the pollutant concentration has dropped to a certain level, etc.) or the purification task for the regional cluster is completed, the control system will then adjust regional cluster B to the new target regional cluster according to the sorting, and repeat the above-mentioned operations such as planning the position and adjusting the parameters to ensure that the air purifier can effectively purify the air of each sorted target regional cluster in turn.

[0201] For example, area cluster A is a conference room area where people gather and activities are frequent. Prioritized purification can quickly improve the air quality where people are located. Area cluster B is a restaurant area adjacent to the kitchen. After purification, it can reduce the impact of cooking odors on the surrounding environment. And so on. By sorting and processing different target area clusters one by one, limited air purification resources can be reasonably allocated, and the air quality of the entire indoor space can be improved more efficiently to meet the needs of indoor people for a good air environment.

[0202] In short, by determining the priority area cluster ranking based on the comprehensive priority score and adjusting the target area clusters one by one according to the ranking, air purification work can be carried out in a more orderly and targeted manner, ensuring the cleanliness and health of indoor air to the greatest extent.

[0203] The multi-dimensional air purification method based on AI intelligent big data in-depth analysis also includes steps after outputting the purification parameters of the air purifier, as follows:

[0204] Step Sa00, inputting the quantified values ​​of different evaluation indicators into the fan speed adjustment ratio model, and analyzing and determining the fan speed adjustment ratio.

[0205] Quantitative values ​​of different evaluation indicators: refers to the various evaluation indicators related to the gathering of people (such as population density, residence time, types of personnel activities, and overlapping area, etc.), which are converted into specific values ​​that can be used for calculation through specific quantitative methods. For example, population density may be quantified as the specific number of people per square meter, the duration of residence time in minutes or hours, the corresponding value assigned to the type of personnel activities based on the degree of impact of the activity on air quality, and the area value of the overlapping area in square meters, etc. These quantitative values ​​can intuitively reflect the actual situation of each indicator.

[0206] Fan speed adjustment ratio model: This is a mathematical model based on the air purification principle, a large amount of past experimental data and related algorithms. The model uses the quantitative values ​​of different evaluation indicators as input parameters, and outputs the fan speed adjustment ratio after comprehensive analysis through internal preset operation logic and rules. It aims to accurately adjust the fan speed of the air purifier according to factors such as different indoor personnel gathering conditions, so as to achieve the purpose of adapting to the current air purification needs.

[0207] Fan speed adjustment ratio: indicates the ratio of the fan speed to its initial setting speed or the speed in the previous stage. For example, if the fan speed adjustment ratio is 1.2, it means that the fan speed needs to be increased to 1.2 times the original speed; if it is 0.8, it means that the fan speed needs to be reduced to 0.8 times the original speed. This ratio adjustment can be used to change the air circulation volume and purification efficiency of the air purifier, so that it can better meet the air purification needs in different scenarios.

[0208] After obtaining the quantitative values ​​of different evaluation indicators, these values ​​were input into the fan speed adjustment proportional model.

[0209] For example, for several area clusters divided from an indoor space, the quantified value of population density in one area cluster is 0.6 people / square meter (meaning there are 0.6 people per square meter), the quantified value of residence time is 300 minutes (the average length of time people stay in this area), the quantified value of personnel activity type is 2 (assuming that the activity intensity is high and has a greater impact on air quality, a higher value is assigned), and the quantified value of overlapping area is 20 square meters (the overlapping area between this area cluster and the area where people gather).

[0210] After these quantitative values ​​are input into the fan speed adjustment ratio model, the model will analyze and calculate according to the pre-set rules and algorithms. The model may comprehensively consider the impact of various indicators on air quality and air purifier load. For example, the greater the density of people, the more pollutants are generated, and faster air circulation is required for purification; the longer the stay time, the higher the intensity required for the air purifier to work continuously; the type of human activities affects the type of pollutants and the speed of generation; the larger the overlapping area, the more concentrated the key purification areas are, etc.

[0211] It is assumed that the model, based on these factors as well as past empirical data, experimental results, etc., undergoes complex calculations (may involve weighted summation, function fitting and other mathematical operations), and finally outputs a fan speed adjustment ratio of 1.3. This shows that for the situation of the cluster in this area, in order to achieve a better air purification effect, the fan speed of the air purifier needs to be increased to 1.3 times the original speed to speed up air circulation, more efficiently filter and treat various pollutants generated by human gathering activities, and ensure that the air quality in the cluster in this area can meet good standard requirements.

[0212] Step Sb00, analyzing and determining the optimized fan speed according to the fan speed and the fan speed adjustment ratio included in the output air purifier purification parameters, and using the optimized fan speed as the fan speed in this air purifier.

[0213] Based on the same inventive concept, an embodiment of the present invention provides a multi-dimensional air purification system based on AI intelligent big data deep analysis, including a memory and a processor, wherein the memory stores a program that can be executed on the processor to implement the following Figures 1 to 4 Procedure for either method.

[0214] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A multi-dimensional air purification method based on AI intelligent big data in-depth analysis, characterized in that: include: Get the start signal of the air purifier; The air purifier operates in response to the start signal; Obtain the air quality data after the air purifier runs for a preset time interval and the air quality data before the air purifier runs, analyze and obtain the purification effect of the air purifier during this operation, and analyze whether the purification effect of the air purifier during this operation meets the preset purification effect standard; If yes, it is judged that the operation is normal; If not, the drone preset in the air purifier is started to scan each sub-area in the room according to the preset flight path. During the scanning process, the air quality of each sub-area is synchronously detected, and the room is defined to be divided into multiple sub-areas; Based on the air quality data of each indoor sub-area, a cluster analysis algorithm is used to find sub-area groups with similar air quality, and to determine whether there is a regional cluster whose pollution level reaches the preset pollution level standard and whose effective coverage ratio of the air purifier is less than the preset ratio. The regional cluster is composed of sub-area groups that meet the conditions, which means that both the preset pollution level standard and the coverage ratio requirements are met; If yes, the region cluster that meets the conditions is used as the target region cluster, and the centroid position of the target region cluster is calculated according to the coordinates and areas of the sub-regions contained in the target region cluster as the target position of the air purifier; According to the preset indoor map and obstacle information, a path planning algorithm is used to plan a moving path from the current position to the target position for the air purifier, and the air purifier is controlled to move along the moving path; The size, concentration and type of pollutants in the regional cluster are obtained, and the data are input into a preset air purifier purification parameter adjustment model, and the air purifier purification parameters are output. After the air purifier is moved to the target position, the purification parameters are adjusted according to the air purifier purification parameters; If not, the fan speed of the air purifier is analyzed and determined based on the corresponding relationship between the ratio of the purification effect of the air purifier's current operation to the preset purification standard and the fan speed of the air purifier, and the fan speed of the air purifier is updated.

2. The multidimensional air purification method based on AI intelligent big data in-depth analysis according to claim 1 is characterized in that: Detecting the air quality of each sub-area includes: Obtain air quality data from monitoring points in known sub-regions, and organize and pre-process the data; The Kriging interpolation algorithm is used to infer the estimated air quality data of other unmonitored points in the room based on the air quality data of the known sub-area monitoring points.

3. The multidimensional air purification method based on AI intelligent big data in-depth analysis according to claim 1 is characterized in that: Based on the air quality data of each indoor sub-area, a cluster analysis algorithm is used to find sub-area groups with similar air quality, including: The air quality data of each sub-area detected by the drone is integrated with the indoor meteorological data and indoor spatial geographic information collected at the same time, and a feature vector is constructed based on the integrated data and standardized; Run the preset clustering algorithm and use the feature vectors of each sub-region as input data for clustering analysis. The clustering algorithm divides the sub-regions into different regional clusters based on the feature similarity between them.

4. The multidimensional air purification method based on AI intelligent big data in-depth analysis according to claim 3 is characterized in that: The area clusters for judging whether there are any areas where the pollution level reaches the preset pollution level standard and the effective coverage ratio of the air purifier is less than the preset ratio include: From the regional clusters obtained by cluster analysis, the center position coordinates and boundary range information of each regional cluster are extracted, and the air quality data of the sub-regions within the regional cluster are obtained, including the concentration distribution of various pollutants and the pollution severity assessment results; Obtain the current location coordinates of the air purifier, and combine the preset effective coverage range of the air purifier to mark the coverage area on the indoor space map; For each regional cluster, analyze the proportion of its area overlapping with the air purifier coverage area; Based on the air quality data of sub-areas within the regional cluster, the preset pollution severity standards, the proportion of uncovered areas, and the preset proportion standards for areas judged to be not effectively covered by air purifiers, it is analyzed whether there is a regional cluster that meets the preset pollution severity standards and the proportion of areas effectively covered by air purifiers is less than the preset proportion.

5. The multi-dimensional air purification method based on AI intelligent big data deep analysis according to any one of claims 1 to 4, characterized in that: The steps are also included in parallel with controlling the air purifier to move along the moving path, specifically as follows: When the air purifier moves along the planned route, it continuously monitors the air quality change information and indoor environment dynamic information along the way; According to the air quality change information along the route and the dynamic information of the indoor environment, analyze whether the concentration of some pollutants exceeds the preset concentration range; If yes, the pollutants and concentration values ​​exceeding the preset concentration range are used as input objects and input into the air purifier purification parameter adjustment model, and the adapted air purifier purification parameters are output and the purification parameters are adjusted; If not, continue moving.

6. The multi-dimensional air purification method based on AI intelligent big data deep analysis according to any one of claims 1 to 4, characterized in that: The regional clusters that meet the conditions are taken as target regional clusters, including: Analyze whether there are multiple regional clusters that meet the conditions; If not, maintain the original setting; If yes, then obtain the indoor personnel gathering location information, and adopt the personnel gathering area cluster evaluation index quantification method to analyze and obtain the quantitative values ​​of different evaluation indicators and add them up to calculate the priority comprehensive scores of different area clusters. The evaluation indicators include personnel density, residence time, personnel activity type and overlapping area; According to the comprehensive priority score, the order of the regional clusters to be processed is determined, and the target regional clusters are adjusted one by one according to the order.

7. The multi-dimensional air purification method based on AI intelligent big data in-depth analysis according to claim 6 is characterized in that: The steps after outputting the purification parameters of the air purifier are also included, which are as follows: The quantitative values ​​of different evaluation indicators are input into the fan speed adjustment ratio model to analyze and determine the fan speed adjustment ratio; According to the fan speed and the fan speed adjustment ratio included in the output air purifier purification parameters, the optimized fan speed is analyzed and determined, and the optimized fan speed is used as the fan speed in this air purifier.

8. A multi-dimensional air purification system based on AI intelligent big data in-depth analysis, characterized in that: It includes a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement a multi-dimensional air purification method based on AI intelligent big data deep analysis as described in any one of claims 1 to 7.

Images