Indoor pollutant detection method and device driven by air quality data

Abstract

The invention provides an indoor pollutant detection method and device driven by air quality data, and relates to the technical field of pollutant detection, and the method comprises the steps: building an indoor reference model of a target space, and carrying out scene configuration based on user configuration; obtaining the attention height of the scene, and carrying out two-dimensional simplification; activating a multi-source sensor, and collecting air quality data; constructing a pollutant diffusion model with the simplified indoor model, and performing indoor pollutant diffusion simulation; and performing point source information extraction on the pollutant distribution information according to a preset pollutant control point set, and outputting multiple pieces of point source information as indoor pollutant detection results. Through the method and the device, the technical problem of relatively low indoor pollutant detection precision caused by environmental complexity and spatial difference in the prior art can be solved, accurate monitoring on the indoor pollutant concentration is realized by performing scene configuration on the indoor reference model and combining indoor and outdoor air data, and the indoor pollutant detection precision is improved. And the precision and the efficiency of indoor pollutant detection are improved.

Description

Technical Field

[0001] This application relates to the technical field of pollutant detection, and particularly to an indoor pollutant detection method and device driven by air quality data. Background Art

[0002] Indoor air pollution not only affects human health, but is also closely related to energy consumption, building maintenance, and air circulation. The quality of indoor air directly affects the physical health, work efficiency, and quality of life of occupants. Long-term exposure to harmful pollutants can cause a series of health problems. Currently, indoor pollutant detection methods are carried out through fixed-point positions and model-based prediction methods, which can help evaluate indoor air quality to a certain extent, but there are some technical problems. Different indoor spaces have different structures, layouts, ventilation systems, and object distributions, etc., and these factors will significantly affect the diffusion pattern of pollutants in the indoor environment. In addition, the indoor environment is not only affected by the building structure, but is also closely related to the external environment. Outdoor air quality, weather conditions (such as wind speed, temperature, humidity), etc. will all affect how pollutants enter the indoor environment and the diffusion process indoors.

[0003] In summary, in the prior art, there is a technical problem that due to the lack of full consideration of environmental complexity and spatial differences, the diffusion behavior of pollutants in different spaces cannot be effectively simulated, resulting in a low accuracy of indoor pollutant detection. Summary of the Invention

[0004] The purpose of this application is to provide an indoor pollutant detection method and device driven by air quality data, so as to solve the technical problem in the prior art that due to the lack of full consideration of environmental complexity and spatial differences, the diffusion behavior of pollutants in different spaces cannot be effectively simulated, resulting in a low accuracy of indoor pollutant detection.

[0005] In view of the above problems, this application provides an indoor pollutant detection method and device driven by air quality data.

[0006] In a first aspect, the present application provides an indoor pollutant detection method driven by air quality data. The indoor pollutant detection method driven by air quality data is implemented through an indoor pollutant detection device driven by air quality data. Among them, the indoor pollutant detection method driven by air quality data includes: establishing an indoor benchmark model for the target space, and performing scenario configuration on the indoor benchmark model based on user configuration to generate a target indoor model; obtaining the scene attention height, and performing two-dimensional simplification of the target indoor model according to the scene attention height to obtain a simplified indoor model; activating multi-source sensors to collect air quality data of the target space, where the air quality data includes local air quality data and external environment air quality data; constructing a pollutant diffusion model based on the air quality data and the simplified indoor model, and performing indoor pollutant diffusion simulation according to the pollutant diffusion model to obtain pollutant distribution information; extracting point source information from the pollutant distribution information according to a preset set of pollutant control points, and outputting multiple point source information as the indoor pollutant detection result.

[0007] In a second aspect, the present application further provides an indoor pollutant detection device driven by air quality data, which is used to execute the indoor pollutant detection method driven by air quality data as described in the first aspect. Among them, the indoor pollutant detection device driven by air quality data includes: a scenario configuration module, which is used to establish an indoor benchmark model for the target space and perform scenario configuration on the indoor benchmark model based on user configuration to generate a target indoor model; a two-dimensional simplification module, which is used to obtain the scene attention height and perform two-dimensional simplification of the target indoor model according to the scene attention height to obtain a simplified indoor model; a data collection module, which is used to activate multi-source sensors to collect air quality data of the target space, where the air quality data includes local air quality data and external environment air quality data; a model construction module, which is used to construct a pollutant diffusion model based on the air quality data and the simplified indoor model, and perform indoor pollutant diffusion simulation according to the pollutant diffusion model to obtain pollutant distribution information; an information extraction module, which is used to extract point source information from the pollutant distribution information according to a preset set of pollutant control points, and output multiple point source information as the indoor pollutant detection result.

[0008] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0009] By establishing an indoor reference model of the target space and performing scenario configuration on the indoor reference model based on user configuration, a target indoor model is generated; the scene attention height is obtained, and two-dimensional simplification of the target indoor model is performed according to the scene attention height to obtain a simplified indoor model; multi-source sensors are activated to collect air quality data of the target space, where the air quality data includes local air quality data and external environment air quality data; based on the air quality data and the simplified indoor model, a pollutant diffusion model is constructed, and indoor pollutant diffusion simulation is performed according to the pollutant diffusion model to obtain pollutant distribution information; according to a preset set of pollutant control points, point source information extraction is performed on the pollutant distribution information, and multiple point source information is output as the indoor pollutant detection result. That is to say, by establishing a configurable indoor reference model, making scenario adjustments according to user needs, introducing the concept of attention height and simplifying the indoor model, combining indoor and outdoor air quality data, constructing a pollutant diffusion model, analyzing the pollutant distribution information, accurately identifying the location of the pollution source, and outputting it in the form of point source information, an effective indoor pollutant detection result is formed, realizing precise monitoring of the indoor pollutant concentration, and improving the accuracy and efficiency of indoor pollutant detection.

[0010] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically illustrates the specific embodiments of the present application. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only exemplary, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0012] Figure 1 It is a schematic flowchart of the indoor pollutant detection method driven by air quality data of the present application;

[0013] Figure 2 It is a schematic structural diagram of the indoor pollutant detection device driven by air quality data of the present application.

[0014] Description of the reference numerals: Scenario configuration module 11, two-dimensional simplification module 12, data acquisition module 13, model construction module 14, information extraction module 15. Detailed implementation manners

[0015] By providing an indoor pollutant detection method and device driven by air quality data, this application solves the technical problem in the prior art that due to the insufficient consideration of environmental complexity and spatial differences, the diffusion behavior of pollutants in different spaces cannot be effectively simulated, resulting in low accuracy of indoor pollutant detection. By establishing a configurable indoor benchmark model, making scenario-based adjustments according to user requirements, introducing the concept of attention height and simplifying the indoor model, combining indoor and outdoor air quality data, constructing a pollutant diffusion model, analyzing the pollutant distribution information, accurately identifying the location of the pollution source, and outputting it in the form of point source information to form an effective indoor pollutant detection result, realizing precise monitoring of the indoor pollutant concentration, and improving the accuracy and efficiency of indoor pollutant detection.

[0016] Next, the technical solutions in this application will be described clearly and completely with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited by the example embodiments described here. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application. In addition, it should be noted that for the sake of description, only parts related to this application are shown in the accompanying drawings rather than all.

[0017] Embodiment 1. Please refer to the attached Figure 1 , this application provides an indoor pollutant detection method driven by air quality data. Among them, the indoor pollutant detection method driven by air quality data is applied to an indoor pollutant detection device driven by air quality data. The indoor pollutant detection method driven by air quality data specifically includes the following steps:

[0018] Step 1: Establish an indoor benchmark model for the target space, and perform scenario-based configuration on the indoor benchmark model based on user configuration to generate a target indoor model.

[0019] Specifically, the spatial characteristic information of the target space includes spatial dimension information (such as the length, width, and height of a room) and spatial layout information (such as the positions of walls, partitions, floors, ceilings, doors, and windows, etc.). According to the spatial characteristic information, the spatial dimension information is transformed into a standard geometric model, usually a simple cuboid or rectangle. Then, based on the spatial layout information, room partitioning, door and window positions, etc. are determined to form an indoor reference model. The specific configurations of items in the space and the ventilation settings in the space are collected from the user. The item configuration information will describe in detail the quantity, size, and layout of various items in the room (such as tables, chairs, computers, etc.). The ventilation configuration information includes information such as ventilation form, ventilation obstacles (such as window screens, dust-proof nets, the opening and closing states of doors and windows, etc.).

[0020] According to the user configuration information, elements such as furniture and decorations are added to the reference model. The configuration information adjusted by the user through the visualization interface is imported into 3D modeling software. The user configuration includes furniture positions, sizes, styles, etc. The user configuration information is parsed and converted into a format that can be recognized by the 3D model. The positions, sizes, and angles of these elements are adjusted to correctly reflect the actual scene of the target space. All elements (including the elements of the reference model and the user configuration) are integrated into a complete 3D model to obtain the target indoor model. According to the item configuration information provided by the user, simple geometric bodies (such as rectangles, cubes, etc.) corresponding to each item (such as tables, chairs, bookshelves, etc.) are constructed, and the simplified models of the items are mapped into the indoor reference model so that each item occupies its correct position in the actual space.

[0021] Using the ventilation configuration information provided by the user, the indoor reference model is adjusted and optimized to generate the target indoor model. The air flow and pollutant diffusion paths are configured according to the ventilation conditions in the room (such as the opening and closing of windows, the positions of air conditioner vents, etc.). That is to say, according to the actual usage scenarios of the room (such as window ventilation, air conditioner working status, etc.), the air flow paths are simulated and these factors are introduced into the target indoor model. By mapping the specific structure and layout of the target space to the indoor reference model and correcting the indoor reference model in combination with specific scenarios, the real scene is accurately reflected, making the indoor pollutant diffusion model more suitable for the actual application scenario.

[0022] Step 2: Obtain the scene attention height, and perform two-dimensional simplification of the target indoor model according to the scene attention height to obtain a simplified indoor model.

[0023] Specifically, according to the spatial attribute characteristics of the target space (such as height, area, volume, etc.) and personnel configuration characteristics (such as the number of people, activity habits, activity areas, etc.), the scene attention height is defined. The target indoor model is two-dimensionally simplified according to the scene attention height, and the three-dimensional space information is compressed into one or more two-dimensional plane models, reflecting the model situation near the scene attention height. For example, in a simplified model of an office space, assuming the scene attention height is from 1.0 meter to 1.5 meters, then the simplified two-dimensional model may only retain the parts related to desks, chairs, and walls, while ignoring the ceiling above or the floor information below, making the simplified model more in line with the actual activity area.

[0024] After two-dimensional simplification, the obtained simplified indoor model retains the core elements related to the scene attention height in the target space, including the planar layout of each object, without complex three-dimensional geometric details, and is suitable for efficient simulation calculations. The key layers of all simplified two-dimensional plane models are superimposed and fused according to their corresponding weights, that is, weighted combination is carried out according to the importance or function of different layers, to obtain the indoor simplified model. By obtaining the scene attention height and two-dimensionally simplifying the target indoor model according to this height, a refined and highly relevant simplified indoor model can be obtained, improving the efficiency of analysis and simulation, while ensuring that the model retains the spatial information most relevant to a specific scene.

[0025] Step 3: Activate multi-source sensors to collect air quality data of the target space, where the air quality data includes local air quality data and external environment air quality data.

[0026] Specifically, determine the air quality indicators to be monitored, such as PM2.5, CO 2 , VOC (volatile organic compounds), etc. Select appropriate multi-source sensors according to the monitoring objectives, including but not limited to particulate matter sensors, gas sensors, temperature and humidity sensors, etc. Design the sensor layout to ensure coverage of the key areas of the target space, and consider possible influencing factors, such as ventilation, human flow, etc. Install the sensors at predetermined positions within the target space, install local air quality sensors inside the target space to monitor the indoor air quality. For external environment air quality monitoring, the sensors should be installed outdoors to be able to reflect the external air quality of the target space. Set the working parameters of the sensors, such as sampling frequency, data transmission method, etc., to ensure the stable operation of the sensors, and calibrate regularly to ensure the accuracy of the data. Multi-source sensors refer to using multiple sensors or devices to collect data simultaneously in the same target space. For example, in air quality monitoring, a PM2.5 sensor (measuring fine particulate matter in the air), CO 2Sensors (for measuring carbon dioxide concentration), temperature and humidity sensors (for measuring temperature and humidity in the air), etc., work simultaneously or periodically to provide comprehensive air quality data.

[0027] Activate multi-source sensors and collect air quality data of the target space according to the preset sampling frequency, including local air quality data and external environment air quality data. Local air quality data refers to the data collected by sensors arranged inside the target space (such as indoors), which reflects the state of indoor air quality. The parameters involved include temperature, humidity, CO 2 concentration, concentration of pollutants in the air, etc. External environment air quality data refers to the data collected by external monitoring systems or sensors, which usually represents the air quality status of the outside of the target space, provides a reference for the external air pollution situation, and helps to evaluate the impact of external pollution sources on indoor air quality. By activating multi-source sensors, the air quality of the target space can be comprehensively monitored, including both local (indoor) air quality data and external environment air quality data. The comprehensive monitoring makes the evaluation of indoor air quality more accurate.

[0028] Step 4: Based on the air quality data and the simplified indoor model, construct a pollutant diffusion model, and perform indoor pollutant diffusion simulation according to the pollutant diffusion model to obtain pollutant distribution information.

[0029] Specifically, obtain the historical air quality data of the target space, including historical indoor air quality data and historical outdoor air quality data. According to the air quality data and the simplified indoor model, construct a pollutant diffusion model for simulating the propagation of pollutants in the indoor space. The pollutant diffusion model takes into account factors such as air flow, temperature, humidity, ventilation conditions, etc. to calculate the concentration change of pollutants from the source to each location. Collect air quality data inside and outside the room, including information such as pollutant types, concentrations, time series, etc., and determine the boundary conditions and initial conditions of the indoor model, such as the air flow rate at the ventilation opening, the outdoor pollutant concentration, etc. Select a suitable mathematical model, such as a turbulence model, a multiphase flow model, a pollutant transport model, etc., set the simulation parameters in CAD software, such as time step, grid division, solver type, etc., run the simulation, and monitor the calculation process. Analyze the simulation results, obtain the concentration distribution map, streamline map, etc. of pollutants in the room, identify the high-concentration areas of pollutants, and evaluate the health risks.

[0030] When conducting the simulation, it is assumed that the air flowing into the room from the window is simulated, and the air conditioning system causes the air to circulate in the room. The pollutant diffusion model will consider these air flow factors and predict the pollutant concentrations in different areas. After the simulation is completed, the distribution information of pollutants in the indoor space is generated, which describes the concentration changes of pollutants at various positions in the room during the simulation time. The simulation based on the pollutant diffusion model provides accurate prediction of indoor pollutant diffusion. By predicting the results of pollutant diffusion through simulation, air quality problems can be identified in advance without relying on a large number of actual detections, reducing the costs of hardware equipment and manual detection.

[0031] Step Five: According to the preset pollutant control point set, extract point source information from the pollutant distribution information and output multiple pieces of point source information as the indoor pollutant detection results.

[0032] Specifically, the pollutant control point set refers to several key points preset in the indoor space, usually the positions that are focused on in indoor air quality monitoring, located in different areas of the indoor space, and used to centrally monitor the concentration changes of pollutants. For example, in an office, the pollutant control points include positions such as the air conditioner intake, near the window, and around the conference table. The pollutant distribution information refers to the concentration distribution of pollutants in the indoor space, usually obtained through the simulation of the pollutant diffusion model, which describes the pollutant concentration in each area or position and its change over time.

[0033] According to the preset pollutant control point set, identify the key positions or known pollution source areas in the indoor space. Extract possible pollution source information from the pollutant distribution information, which is usually related to the areas with higher pollutant concentrations in the room and represents the source of pollutants. Extract detailed data of multiple key points from the point source information, and use the extracted point source data as the indoor pollutant detection results to evaluate the air quality of the entire space and ensure that each pollution source is accurately monitored and identified. Point source information extraction refers to extracting information such as the position and intensity of the pollution source from the pollutant distribution information, which is often the source of pollutants released by specific areas or equipment (such as air conditioners, heaters, windows, etc.). By extracting point source information, the source and diffusion path of pollutants can be further understood. Through point source information extraction, accurately locate the position of indoor pollution sources and estimate the pollutant concentration, helping to discover those pollution sources that are easily overlooked, such as areas like air conditioners, equipment, or windows. By extracting point source information and analyzing its intensity and diffusion range, users can take targeted measures to improve air quality.

[0034] Furthermore, Step One of this application includes:

[0035] Based on the spatial characteristic information of the target space, an indoor reference model is established, where the spatial characteristic information includes spatial dimension information and spatial layout information; obtain the user configuration information of the target space, including object configuration information and ventilation configuration information; construct multiple object simplified models of the target space based on the object configuration information, and map the multiple object simplified models to the indoor reference model; based on the ventilation configuration information, perform scenario configuration on the indoor reference model to obtain the target indoor model.

[0036] Specifically, according to the spatial characteristic information of the target space, including spatial dimension information (such as the length, width, and height of a room) and spatial layout information (such as room partitioning, door and window positions, etc.). The spatial characteristic information refers to the data describing the basic characteristics of the indoor space. The spatial dimension information refers to the physical dimension data of the target space (such as a room or an office), and the spatial layout information refers to the physical layout inside the space, such as the arrangement of walls, doors, windows, furniture, and the layout of equipment such as air conditioners and ventilation openings. The spatial characteristic information of the target space can be obtained by a 3D scanner or a laser measurement tool to obtain the specific dimensions and layout information of the indoor space. Use CAD software (such as AutoCAD) or 3D modeling software (such as SketchUp) to establish an indoor reference model, which is usually a simple cuboid or rectangle, including the basic structures of walls, floors, ceilings, doors, and windows.

[0037] Obtain the user configuration information of the target space, including object configuration information and ventilation configuration information, that is, the specific configuration of items and the ventilation settings in the space. The object configuration information will detail the quantity, dimensions, and layout of various furniture objects (such as tables, chairs, computers, etc.) in the room. The ventilation configuration information includes information such as ventilation form and ventilation obstacles (screen windows, dust-proof nets, the opening and closing states of doors and windows, etc.), which will affect the simulation of air flow and pollutant diffusion. The user configuration information refers to the specific requirements or preferences of the user, the specific requirements for the indoor environment, including the placement of furniture and other objects (object configuration information) and the settings of the ventilation system (ventilation configuration information). The object configuration information is obtained by collecting real-time photos or videos of the target space and detecting the position information and dimension information of specific objects in the real-time photos or videos through an object detection algorithm, and displaying it on the user side. The user can modify the position or dimensions of the furniture according to the actual situation, and re-upload and update the adjusted information by the user and store it as the final user configuration information. The ventilation configuration information is composed of understanding the indoor ventilation system (such as fresh air system, exhaust system, air circulation path, etc.), using sensors to monitor the indoor air quality, and understanding the user's requirements for indoor ventilation, etc.

[0038] According to the object configuration information, use a parametric design tool (such as CAD software) to construct simple geometric bodies (such as rectangles, cubes, etc.) for each object, obtaining multiple simplified models of the objects. Map them to the corresponding positions in the indoor reference model to ensure consistency with the actual space layout. Simplified design refers to geometric simplification of these objects, removing unnecessary details and retaining the basic shapes and dimensions. Integrate all the simplified models of the objects into the indoor reference model to form a complete indoor space layout. According to the space layout and functional requirements, adjust the size, orientation, and position of the objects. For example, construct a table as a rectangular simplified model with a length of 1.5 meters, a width of 0.75 meters, and a height of 0.75 meters; construct a bookshelf as a rectangular simplified model with a length of 2 meters, a width of 0.3 meters, and a height of 2 meters.

[0039] Use the ventilation configuration information to adjust and optimize the indoor reference model according to the actual usage of the target space (such as personnel activities, ventilation conditions, etc.). Configure the air flow and pollutant diffusion paths according to the ventilation conditions of the room (such as the opening and closing of windows, the position of air conditioner vents, etc.). Scenario-based configuration means simulating the air flow path according to the actual usage scenarios of the room (such as window ventilation, air conditioner working status, etc.) and introducing these factors into the target indoor model. For example, if the window is open, the air flow path will change, resulting in a different pattern of pollutant diffusion. By combining the space characteristic information, user configuration information, simplified object models, and ventilation configuration information, construct a more realistic indoor environment model. Especially considering the furniture layout and ventilation conditions, make the indoor pollutant diffusion model more suitable for the actual application scenario.

[0040] Furthermore, the present application further includes the following steps:

[0041] The interactive user terminal obtains the live image information and ventilation configuration information of the target space collected by the user; through the target detection algorithm, performs object detection on the live image information to obtain the object position information and object size information, generating the original user configuration information; feeds back the object configuration information to the user terminal for visual display, updates and transmits the original user configuration information adjusted by the user, and stores it as the user configuration information.

[0042] Specifically, interacting with the client refers to the interface or platform for interacting with users, which can be a mobile application, a web application, or desktop software. The real-time photos or videos (live image information) of the room are taken through a terminal device (camera or other image acquisition device). Through the client, the user manually or automatically inputs the ventilation configuration information of the room, including the ventilation form (such as natural ventilation, mechanical ventilation, or hybrid ventilation), ventilation obstacles (such as window screens, dust-proof nets, the opening and closing states of doors and windows, etc.). The object detection algorithm is a computer vision algorithm used to identify and locate specific objects in an image, which can identify the positions and sizes of objects (such as tables, chairs), including YOLO, Faster R-CNN, etc.

[0043] Use the object detection algorithm (such as YOLOv5 or Faster R-CNN) to analyze the live image information, identify specific objects, and extract the position information and size information of the objects. The object detection algorithm will mark the bounding boxes of the objects and convert the position information (such as the center point coordinates) and sizes (such as length, width, and height) of these objects into data. Object detection refers to identifying specific objects (such as tables, chairs, windows, etc.) in an image and determining their positions. The specific object detection process is as follows: Preprocess the live image information, such as resizing, normalizing, denoising, etc., to meet the requirements of the object detection algorithm. Select a suitable pre-trained model or train a new model according to the type of objects to be detected. Use the selected model to perform object detection on the preprocessed image, and output the category, position (usually represented in the form of a bounding box), and confidence of each detected object. Extract the position (coordinates of the bounding box) and size (width and height of the bounding box) of each object from the output of the algorithm. Integrate the extracted position and size information into a structured data format, such as a JSON or XML file. Summarize the information of the detected objects to form a preliminary room configuration dataset, including the position information and size information of each piece of furniture, as well as the states of doors and windows (such as open / closed, equipped with window screens).

[0044] The generated original user configuration information is presented to the user in a visual form, that is, on the user-side interface, the detected objects are superimposed on the original image in the form of layers, showing the positions and sizes of the objects. For example, a top view or a 3D model of a room is displayed on the APP interface, and the user can see the positions and sizes of the furniture. The user-side tools (such as dragging, zooming, rotating, etc.) are interactive, allowing the user to click, drag, or adjust the furniture in the model for correction. The user can also modify the category or other attributes of the object through the input field. The user can modify the position or size of the furniture according to the actual situation, and the adjustments on the user side should be reflected on the interface in real time, allowing the user to see the adjusted effect. After the user confirms that the adjustment is correct, the adjusted information of the user will be transmitted back to the server and stored as the final user configuration information. Through the live image information and the object detection algorithm, the objects in the room can be automatically identified, reducing the error of manual input. Through the visual display and user adjustment functions, the user can calibrate and optimize the data generated by the system according to the actual situation, improving the fitting degree of the model.

[0045] Further, step two of this application includes:

[0046] According to the spatial attribute features and personnel configuration features of the target space, define the scene attention height; perform height slicing of the target indoor model based on the scene attention height to obtain an attention slice model, where the attention slice model has slice thickness information; extract key layers from the attention slice model and fuse multiple key layers of the key layer extraction results to generate the simplified indoor model.

[0047] Specifically, according to the spatial attribute features of the target space (such as height, area, volume) and personnel configuration features (such as the number of people, activity habits, activity areas), determine the height range that needs to be focused on (i.e., the scene attention height). The spatial attribute features include parameters related to the physical structure of the space, such as height, area, volume, etc., which determine the basic geometric form of the space. The personnel configuration features refer to information such as the number of people, activity range, and activity habits in the target space. For example, if the target space is an office with a space height of 3 meters, the number of people may be 10, the activity range is concentrated around the desks, and the activity habit is to sit and discuss, then the attention height may be in the seat height range (about 0.7 meters to 1.2 meters).

[0048] In CAD software (such as AutoCAD) or 3D modeling software (such as SketchUp), slice the target indoor model according to the attention height to generate a sliced model that reflects the attention area. Use 3D modeling software or a dedicated slicing tool to perform the slicing operation on the indoor model along the determined attention height. After slicing, a sliced model that reflects the situation near the attention height is obtained, that is, the attention sliced model, which has slice thickness information, that is, the height range covered by this slice. For example, a slice thickness may be 0.4 meters, representing the area from 1 meter to 1.4 meters.

[0049] According to the preset key layer interval, take the central height of the attention sliced model as the initial plane, and symmetrically extract two-dimensional plane models of different interval floor heights upward and downward to obtain multiple key layers. Each key layer contains the main structural features in the height range of this space, such as furniture, doors and windows, walls, etc. According to the height difference between each key layer and the initial plane, calculate the fusion weight corresponding to each key layer through normalization processing. Superimpose and fuse multiple key layers according to the corresponding fusion weights to obtain a simplified indoor model, which represents the key features of the entire space but removes relatively subtle details to reduce the computational complexity. By defining the scene attention height and extracting key layers, the model can focus on the most important areas in the target space (such as seat height, activity area), improve the pertinence of analysis, remove the geometric information of non-key areas, and greatly reduce the amount of calculation.

[0050] Furthermore, the present application further includes the following steps:

[0051] Based on the preset key layer interval, take the central height of the attention sliced model as the initial plane, symmetrically extract two-dimensional plane models of multiple interval floor heights to obtain multiple key layers; according to the height differences between multiple key layers and the initial plane, calculate and obtain the fusion weights of multiple key layers through normalization processing; according to the fusion weights, perform the superimposed fusion of multiple key layers to obtain the simplified indoor model.

[0052] Specifically, the preset key layer interval refers to the height difference between each layer set according to the scene attention height in the indoor model. The central height of the attention slice model refers to the height selected as the initial plane in the attention slice model, which is usually a reference height defined according to the scene attention height (for example, the height of a desktop or a chair), serving as the reference plane in the middle of the model. According to the preset key layer interval, starting from the initial plane, multiple two-dimensional plane models with the height of the interval layer are symmetrically extracted upward and downward. Since the space is three-dimensional, it is necessary to symmetrically expand upward and downward during extraction to ensure that there are corresponding levels both upward and downward from the initial plane. Each extracted layer will generate a two-dimensional plane model, which is retained as a key layer, representing the spatial information within different height ranges of the target space and enabling a more refined analysis of each region of the space.

[0053] The height of the interval layer refers to the vertical height difference between each layer, which determines the vertical height of each layer. The two-dimensional plane model refers to the planarization process of each layer of the space, simplifying the originally three-dimensional spatial information into two-dimensional data, which helps to improve the analysis efficiency. Each two-dimensional plane model contains the key spatial elements of that layer, such as walls, doors, windows, furniture, etc. The key layer refers to the important spatial regions among the extracted layers. Each layer contains the geometric information related to that height range. The key layer helps to focus on a specific height range, thereby understanding the characteristics such as the structure and layout of the space.

[0054] For each extracted key layer, calculate the vertical height difference from it to the initial plane. Normalize the height differences of each key layer to eliminate the weight deviation caused by the height differences between different layers. The purpose of the normalization process is to make the numerical values of the height differences within the range of [0, 1] for subsequent weighted superposition. The normalization calculation is usually (height difference - minimum height difference) / (maximum height difference - minimum height difference), where the minimum height difference and the maximum height difference are respectively the minimum and maximum height difference values among all key layers. The fusion weight is the weight value of each key layer calculated after normalizing the height differences, reflecting the importance of different height layers to the final simplified model.

[0055] Using the calculated fusion weights, multiple key layers are stacked. The two-dimensional plane models of each key layer are weighted according to their fusion weights, and the weighted key layer models are stacked. Through the stacked fusion, a simplified indoor model is finally obtained, which synthesizes the information of each key layer, simplifies the details, and retains the necessary structural information near the attention height. The process of weighted stacking is actually information fusion between key layers, enabling the final simplified indoor model to better reflect the spatial structure in different height ranges. Through refined key layer extraction and weighted fusion, the simplified indoor model can more accurately reflect the important spatial information in different height ranges, avoid over-simplification, and ensure the accuracy of the indoor model in key areas. By using weighted stacking instead of fully detailed 3D modeling, the unnecessary computational burden is reduced. The details of each key layer are weighted according to their importance, and the final simplified model can effectively reduce the consumption of computing resources and improve the efficiency of subsequent analysis and simulation.

[0056] Furthermore, the present application further includes the following steps:

[0057] The simplified indoor model includes multiple layers of simplified indoor sub-models corresponding to the scene attention heights, and the multiple layers of simplified indoor sub-models correspond to multiple attention targets.

[0058] Specifically, for each scene, a simplified indoor sub-model is constructed according to the defined attention height, and each sub-model only contains elements related to a specific attention height. Each level of the simplified indoor sub-model represents a specific attention target, that is, each layer of the model focuses on a specific function or usage scenario. For example, the desktop activity area will be extracted at a lower slice height, and the air conditioners and lights on the ceiling will appear at a higher slice height. The concerns of the multi-layer sub-models are different, which helps to conduct more targeted analysis and optimization. Combining these sub-models at different heights generates a simplified indoor model with multiple levels. Each level represents a specific area and spatial function, and the corresponding geometric details will be streamlined according to different attention targets.

[0059] Furthermore, step four of the present application includes:

[0060] Obtain the historical air quality data of the target space, including historical indoor air quality data and historical outdoor air quality data; initialize the pollutant concentration field of the simplified indoor model according to the outdoor air quality data, and obtain a diffusion verification model; activate the diffusion verification model to perform simulation verification on the simplified indoor model until the state of the pollutant concentration field is balanced, and obtain the verification diffusion result; compare the verification diffusion result with the historical indoor air quality data, calculate the simulation verification deviation, and if the simulation verification deviation meets the detection deviation constraint, perform indoor pollutant diffusion simulation according to the simplified indoor model.

[0061] Specifically, collect the historical air quality data of the target space, including indoor and outdoor air quality data. The indoor air quality data can be directly obtained through sensors, while the outdoor air quality data can be obtained through external monitoring stations or data interfaces. Historical air quality data refers to the data information on air quality collected over a past period of time, usually including indoor and outdoor air quality data, which reflects the changes in parameters such as the concentration, temperature, and humidity of various pollutants in the air within a certain time range. According to the outdoor air quality data, initialize the pollutant concentration field of the simplified indoor model, that is, use the outdoor air quality data as the initial concentration distribution of the simplified indoor model, as the diffusion verification model. The pollutant concentration field is a data distribution in a three-dimensional space, representing the concentration values of pollutants (such as PM2.5, CO 2 etc.) at each point, reflecting the diffusion state of pollutants in space. The diffusion verification model is a mathematical or physical model used to simulate the diffusion process of pollutants in an indoor space, predicting the change of pollutant concentration by simulating air flow, the diffusion of pollutants, and the influence of other environmental factors.

[0062] Activate the diffusion verification model to simulate the indoor air flow and the diffusion process of pollutants. According to the set initial conditions, calculate the change of pollutant concentration at each point in the room over time. Continuously simulate until the pollutant concentration field reaches a stable state, that is, the concentration no longer changes significantly over time, indicating that the indoor and outdoor pollutant exchange and the indoor pollutant diffusion have reached a dynamic balance. Compare the simulation results with the historically collected indoor air quality data, calculate the simulation verification deviation, which is the difference between the pollutant concentration distribution obtained through simulation and the actual historical data collected, including indicators such as the average deviation and root mean square error (RMSE).

[0063] The detection deviation constraint is a tolerance range used to determine whether the results of the simulation model conform to the actual situation. If the deviation between the simulation results and the historical data is less than the preset constraint value, the simulation results are considered valid. After passing the verification, a simplified indoor model is used to perform actual indoor pollutant diffusion simulations and predict the distribution of pollutants under different environmental conditions. The simulation results are applied to indoor environmental assessment, ventilation design optimization, etc. Verification is carried out through historical air quality data to ensure the accuracy of the diffusion simulation model. By comparing historical data with simulation results, it is possible to determine whether the simulation model is reliable, ensure that the simulation results can truly reflect the actual situation, accurately simulate the diffusion behavior of pollutants, and thus improve the accuracy and efficiency of indoor pollutant detection.

[0064] In summary, the indoor pollutant detection method driven by air quality data provided by the present application has the following technical effects:

[0065] By establishing an indoor reference model for the target space and performing scenario configuration on the indoor reference model based on user configuration to generate a target indoor model; obtaining the scene attention height and performing two-dimensional simplification of the target indoor model according to the scene attention height to obtain a simplified indoor model; activating multi-source sensors to collect air quality data of the target space, where the air quality data includes local air quality data and external environment air quality data; constructing a pollutant diffusion model based on the air quality data and the simplified indoor model, and performing indoor pollutant diffusion simulation according to the pollutant diffusion model to obtain pollutant distribution information; extracting point source information from the pollutant distribution information according to a preset set of pollutant control points, and outputting multiple point source information as the indoor pollutant detection result. That is to say, by establishing a configurable indoor reference model, making scenario adjustments according to user needs, introducing the concept of attention height and simplifying the indoor model, combining indoor and outdoor air quality data, constructing a pollutant diffusion model, analyzing the pollutant distribution information, accurately identifying the location of the pollution source, and outputting it in the form of point source information to form an effective indoor pollutant detection result, realizing precise monitoring of indoor pollutant concentration, and improving the accuracy and efficiency of indoor pollutant detection.

[0066] Embodiment 2, based on the same inventive concept as the indoor pollutant detection method driven by air quality data in the first embodiment above, the present application also provides an indoor pollutant detection device driven by air quality data. Please refer to the appendix Figure 2 , the indoor pollutant detection device driven by air quality data includes:

[0067] A scenario configuration module 11, which is used to establish an indoor reference model for the target space and perform scenario configuration on the indoor reference model based on user configuration to generate a target indoor model.

[0068] A two-dimensional simplification module 12, which is used to obtain the scene attention height and perform two-dimensional simplification of the target indoor model according to the scene attention height to obtain a simplified indoor model.

[0069] A data acquisition module 13, which is used to activate multi-source sensors to collect air quality data of the target space. Among them, the air quality data includes local air quality data and external environment air quality data.

[0070] A model construction module 14, which is used to construct a pollutant diffusion model based on the air quality data and the simplified indoor model, and perform indoor pollutant diffusion simulation according to the pollutant diffusion model to obtain pollutant distribution information.

[0071] An information extraction module 15, which is used to perform point source information extraction on the pollutant distribution information according to a preset set of pollutant control points, and output multiple point source information as indoor pollutant detection results.

[0072] Furthermore, the scenario configuration module 11 in the indoor pollutant detection device driven by air quality data is further used for:

[0073] Establish an indoor reference model according to the spatial characteristic information of the target space, where the spatial characteristic information includes spatial dimension information and spatial layout information; obtain the user configuration information of the target space, including object configuration information and ventilation configuration information; construct multiple object simplified models of the target space based on the object configuration information, and map the multiple object simplified models to the indoor reference model; perform scenario configuration on the indoor reference model based on the ventilation configuration information to obtain the target indoor model.

[0074] Furthermore, the scenario configuration module 11 in the indoor pollutant detection device driven by air quality data is further used for:

[0075] An interactive user terminal, which obtains the live image information and ventilation configuration information of the target space collected by the user; performs object detection on the live image information through a target detection algorithm to obtain object position information and object size information, generates original user configuration information; feeds back the object configuration information to the user terminal for visual display, updates and transmits the original user configuration information adjusted by the user, and stores it as the user configuration information.

[0076] Furthermore, the two-dimensional simplification module 12 in the indoor pollutant detection device driven by air quality data is further used for:

[0077] Define the scene attention height according to the spatial attribute characteristics and personnel configuration characteristics of the target space; perform height slicing of the target indoor model based on the scene attention height to obtain an attention slice model, where the attention slice model has slice thickness information; extract key layers from the attention slice model and fuse multiple key layers of the key layer extraction results to generate the simplified indoor model.

[0078] Further, the two-dimensional simplification module 12 in the indoor pollutant detection device driven by air quality data is further configured to:

[0079] Based on a preset key layer interval, symmetrically extract two-dimensional plane models with multiple interval layer heights with the central height of the attention slice model as the initial plane to obtain multiple key layers; calculate and obtain the fusion weights of multiple key layers through normalization processing according to the height differences between multiple key layers and the initial plane; according to the fusion weights, perform superposition fusion of multiple key layers to obtain the simplified indoor model.

[0080] Further, the two-dimensional simplification module 12 in the indoor pollutant detection device driven by air quality data is further configured to:

[0081] The simplified indoor model includes multiple layers of simplified indoor sub-models corresponding to multiple scene attention heights, and the multiple layers of simplified indoor sub-models correspond to multiple attention targets.

[0082] Further, the model construction module 14 in the indoor pollutant detection device driven by air quality data is further configured to:

[0083] Obtain the historical air quality data of the target space, including historical indoor air quality data and historical outdoor air quality data; initialize the pollutant concentration field of the simplified indoor model according to the outdoor air quality data to obtain a diffusion verification model; activate the diffusion verification model to perform simulation verification of the simplified indoor model until the state of the pollutant concentration field is balanced to obtain a verified diffusion result; compare the verified diffusion result with the historical indoor air quality data, calculate the simulation verification deviation, and if the simulation verification deviation meets the detection deviation constraint, perform indoor pollutant diffusion simulation according to the simplified indoor model.

[0084] The various embodiments in this specification are described in a progressive manner, and the key points of each embodiment are the differences from other embodiments. The foregoing Figure 1The indoor pollutant detection method and specific examples in Embodiment 1 are equally applicable to the indoor pollutant detection device driven by air quality data in this embodiment. Through the detailed description of the indoor pollutant detection method driven by air quality data above, those skilled in the art can clearly know the indoor pollutant detection device driven by air quality data in this embodiment. Therefore, for the sake of simplicity of the specification, it will not be elaborated here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0085] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0086] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is also intended to include these changes and variations.

Claims

1. An indoor pollutant detection method driven by air quality data, characterized in that: include: Establishing an indoor benchmark model of the target space, and performing scenario-based configuration on the indoor benchmark model based on user configuration to generate a target indoor model; Acquire a scene attention height, and perform two-dimensional simplification of the target indoor model according to the scene attention height to acquire a simplified indoor model; Activate multi-source sensors to collect air quality data of the target space, wherein the air quality data includes local air quality data and external environment air quality data; Based on the air quality data and the simplified indoor model, a pollutant diffusion model is constructed, and indoor pollutant diffusion simulation is performed according to the pollutant diffusion model to obtain pollutant distribution information; According to a preset pollutant control point set, point source information is extracted from the pollutant distribution information, and a plurality of point source information is output as indoor pollutant detection results.

2. The air quality data driven indoor pollutant detection method according to claim 1, characterized in that: Establishing an indoor benchmark model of the target space, and performing scenario-based configuration on the indoor benchmark model based on user configuration to generate a target indoor model, including: Establishing an indoor reference model according to the spatial characteristic information of the target space, wherein the spatial characteristic information includes spatial dimension information and spatial layout information; Obtain user configuration information of the target space, including object configuration information and ventilation configuration information; constructing a plurality of simplified object models of the target space based on the object configuration information, and mapping the plurality of simplified object models to the indoor reference model; Based on the ventilation configuration information, the indoor reference model is configured in a scenario-based manner to obtain the target indoor model.

3. The air quality data driven indoor pollutant detection method according to claim 2, characterized in that: Obtain the user configuration information of the target space, including object configuration information and ventilation configuration information, including: The interactive user terminal obtains the real-time image information and ventilation configuration information of the target space collected by the user; Perform object detection on the live image information through a target detection algorithm, obtain object position information and object size information, and generate original user configuration information; The object configuration information is fed back to the user end for visual display, and the original user configuration information adjusted by the user is updated and transmitted and stored as the user configuration information.

4. The air quality data driven indoor pollutant detection method according to claim 3, characterized in that: Acquiring a scene attention height, and performing two-dimensional simplification of the target indoor model according to the scene attention height to obtain a simplified indoor model, including: Defining the scene attention height according to the spatial attribute characteristics and personnel configuration characteristics of the target space; Performing fixed-height slicing of the target indoor model based on the scene attention height to obtain an attention slicing model, wherein the attention slicing model has slicing thickness information; Key layers are extracted from the attention slice model, and multiple key layers of the key layer extraction results are fused to generate the simplified indoor model.

5. The air quality data driven indoor pollutant detection method according to claim 4, characterized in that: Extracting key layers from the attention slice model and fusing multiple key layers of the key layer extraction results to generate the simplified indoor model, including: Based on the preset key layer interval, taking the center height of the attention slice model as the initial plane, symmetrically extracting a plurality of two-dimensional plane models with interval layer heights to obtain a plurality of the key layers; According to the height differences between the multiple key layers and the initial plane, a fusion weight of the multiple key layers is obtained by normalization calculation; According to the fusion weight, multiple key layers are superimposed and fused to obtain the simplified indoor model.

6. The air quality data driven indoor pollutant detection method according to claim 5, characterized in that: Based on the air quality data and the simplified indoor model, a pollutant diffusion model is constructed, and indoor pollutant diffusion simulation is performed according to the pollutant diffusion model to obtain pollutant distribution information, including: Obtain historical air quality data of the target space, including historical indoor air quality data and historical outdoor air quality data; Initializing the pollutant concentration field of the simplified indoor model according to the outdoor air quality data to obtain a diffusion verification model; Activating the diffusion verification model to perform simulation verification of the simplified indoor model until the pollutant concentration field state is balanced, and obtaining a verification diffusion result; The verification diffusion result is compared with the historical indoor air quality data to calculate the simulation verification deviation. If the simulation verification deviation satisfies the detection deviation constraint, indoor pollutant diffusion simulation is performed according to the simplified indoor model.

7. The air quality data driven indoor pollutant detection method according to claim 1, characterized in that: The simplified indoor model includes multiple layers of simplified indoor sub-models corresponding to multiple scene attention heights, and the multiple layers of simplified indoor sub-models correspond to multiple attention targets.

8. An indoor pollutant detection device driven by air quality data, characterized in that: The method for detecting indoor pollutants driven by air quality data according to any one of claims 1 to 7 is implemented, wherein the indoor pollutant detection device driven by air quality data comprises: A scenario configuration module, the scenario configuration module is used to establish an indoor benchmark model of the target space, and perform scenario configuration on the indoor benchmark model based on user configuration to generate a target indoor model; A two-dimensional simplification module, the two-dimensional simplification module is used to obtain a scene attention height, and perform two-dimensional simplification of the target indoor model according to the scene attention height to obtain a simplified indoor model; A data acquisition module, the data acquisition module is used to activate the multi-source sensor to collect air quality data of the target space, wherein the air quality data includes local air quality data and external environment air quality data; A model building module, the model building module is used to build a pollutant diffusion model based on the air quality data and the simplified indoor model, and perform indoor pollutant diffusion simulation according to the pollutant diffusion model to obtain pollutant distribution information; The information extraction module is used to extract point source information from the pollutant distribution information according to a preset pollutant control point set, and output a plurality of point source information as indoor pollutant detection results.

Images