Intelligent cleanliness detection device and method

Abstract

The invention discloses an intelligent cleanliness detection device and method, and belongs to the technical field of air cleanliness detection.The intelligent cleanliness detection device comprises a detector, a moving part, a data acquisition unit, a state prediction unit used for generating an air state prediction value, a state judgment unit and a control unit, the moving direction analysis unit is used for acquiring the wind speed of the space where the detector is located at present and generating the moving distance; the moving control unit is used for controlling the moving assembly to drive the detector to move according to the optimal moving direction and the moving distance; the pollutant diffusion trend can be predicted according to the real-time environmental parameters, the detection path can be automatically adjusted, the pollutant diffusion trend can be dynamically tracked, the detection path can be automatically adjusted, and the detection accuracy and efficiency are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of air cleanliness detection, and in particular relates to an intelligent cleanliness detection device and method. Background Art

[0002] Air cleanliness testing is an important technical means to evaluate and detect air quality. Its core lies in accurately detecting the concentration levels of various pollutants in the air, including particulate matter, gaseous pollutants and other components that are harmful to human health.

[0003] Current mainstream detection technologies fall into two main categories: fixed sensor networks and handheld portable detection devices. Fixed sensor networks implement continuous detection by deploying static detection nodes at specific locations. While they can provide stable local data collection, their fixed spatial location makes it difficult to effectively capture the spatial distribution characteristics and dynamic migration patterns of pollutants, resulting in significant spatial limitations in detection results. While handheld portable detection devices offer a degree of mobility, they rely entirely on manual operation, resulting in low detection efficiency and an inability to autonomously adjust detection strategies based on changes in environmental parameters. More critically, existing technologies all employ a passive response model, providing only historical or current pollution status data. They are unable to predict future pollutant diffusion trends or optimize detection paths based on real-time airflow parameters (such as wind direction and speed). These technical limitations directly hinder the timeliness and accuracy of cleanliness testing, making it difficult to meet practical requirements. This is particularly true in large spaces or complex airflow environments. Furthermore, existing devices generally lack intelligent decision-making capabilities and are unable to adaptively adjust detection locations, further limiting the representativeness and reliability of detection results. Summary of the Invention

[0004] In view of the deficiencies in the prior art, the present invention provides an intelligent cleanliness detection device and method to solve the above problems.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent cleanliness detection device, applied to the field of controlling cleanliness detection, comprising a detector, characterized in that the device further comprises: A moving component, which is arranged on one side of the detector and is used to drive the detector to move; A data acquisition unit, used to acquire air data of the space where the detector is currently located; A state prediction unit, configured to generate an air state prediction value based on air data; A state judgment unit, configured to judge the need for the moving component to drive the detector to move according to the predicted value of the air state; A moving direction analysis unit, used to obtain the wind direction of the space where the detector is currently located and generate an optimal moving direction; A moving distance analysis unit, used to obtain the wind speed in the space where the detector is currently located and generate a moving distance; A movement control unit, when the detector needs to move, the movement control unit is used to control the movement component to drive the detector to move according to the optimal movement direction and movement distance; The detection unit continues to perform air detection on the space where the detector is located after the moving component has driven the detector to move.

[0006] On the basis of the above technical solutions, the present invention also provides the following optional technical solutions: Further technical solution: The device also includes: The air collection component is used to collect the air in the space where the detector is located.

[0007] Further technical solution: The air collection component specifically includes: An air inlet duct, the air inlet duct being arranged on one side of the detector; A filter is provided at one end of the air inlet duct and is used for filtering the air entering the detector through the air inlet duct.

[0008] Further technical solution: The state prediction unit specifically includes: Data preprocessing module, used for preprocessing air data; The prediction model building module is used to build an air state prediction model based on the preprocessed air data and generate air state prediction values.

[0009] Further technical solution: The expression of the air state prediction model is specifically:

[0010] Among them, C t+Δt It represents the predicted value of air condition. represents the autoregressive coefficient, C t-j represents the measured value of the air state at the historical moment tj, n represents the number of historical moments, x t It represents the pre-processed air data, B T It represents the static influence weight vector of the preprocessed air data.

[0011] Further technical solution: The moving direction analysis unit specifically includes: A wind direction acquisition module, used to obtain the wind direction of the space where the detector is currently located; An optimal movement direction generation module is used to establish a movement direction analysis model and generate an optimal movement direction according to the wind direction and wind speed in the space where the detector is currently located; The expression of the moving direction analysis model is specifically:

[0012] Among them, d t represents the optimal moving direction, arg max represents the parameter of the maximum value of the objective function, D represents the set of candidate directions, d m It represents the mth candidate direction, d T m W t It represents the dot product of direction and wind direction, W t is the wind direction vector, I (x i loc+Δs*d m ) represents the information gain of the i-th type of air data, x i represents the i-th type of air data, loc represents the current position coordinate, Δs represents the detection distance, and γ is the information gain weight coefficient.

[0013] Further technical solution: The moving distance analysis unit specifically includes: A wind speed acquisition module, used to obtain the wind speed of the space where the detector is currently located; A moving distance generating module is used to establish a moving distance analysis model and generate a moving distance according to the wind direction of the space where the detector is currently located; The expression of the moving distance analysis model is specifically:

[0014] Among them, s t It represents the moving distance, v t It represents the wind speed in the space where the detector is currently located, ║▽C t ║ represents the gradient modulus, ▽C t It represents the cleanliness spatial gradient vector, and β represents the adaptive distance coefficient.

[0015] An intelligent cleanliness detection method specifically comprises the following steps: Place the detection device in the area to be detected; Obtaining air data of the space where the detector is currently located; Generate air status prediction values ​​based on air data; Determining the need for the mobile component to drive the detector to move according to the predicted value of the air state; Obtaining the wind direction of the space where the detector is currently located and generating an optimal moving direction; Obtaining the wind speed of the space where the detector is currently located and generating a moving distance; Controlling the moving component to drive the detector to move according to the optimal moving direction and moving distance; After the moving component has driven the detector to move, the air detection of the space where the detector is located continues.

[0016] Further technical solution: Generating the air state prediction value based on the air data specifically includes the following steps: Preprocess the air data; An air state prediction model is established based on the preprocessed air data to generate air state prediction values.

[0017] Further technical solution: The method of obtaining the wind direction of the space where the detector is currently located and generating the optimal moving direction specifically includes: Obtaining the wind direction of the space where the detector is currently located; According to the wind direction of the space where the detector is currently located, a movement direction analysis model is established to generate an optimal movement direction; The obtaining of the wind speed in the space where the detector is currently located and generating the moving distance specifically includes: Obtaining the wind speed of the space where the detector is currently located; According to the wind direction of the space where the detector is currently located, a movement distance analysis model is established to generate the movement distance.

[0018] The present invention provides an intelligent cleanliness detection device and method, which has the following beneficial effects compared with the prior art: Through the synergistic effect of the detector, mobile components, data acquisition unit, state prediction unit, movement direction analysis unit, movement distance analysis unit and movement control unit, the present invention can predict the pollutant diffusion trend according to real-time environmental parameters and autonomously adjust the detection path, thereby realizing dynamic tracking of pollutant diffusion trends, autonomous adjustment of the detection path, and improving detection accuracy and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A three-dimensional structural diagram of an intelligent cleanliness detection device provided by an embodiment of the present invention.

[0020] Figure 2 A schematic diagram of the unit structure of an intelligent cleanliness detection device provided by an embodiment of the present invention.

[0021] Figure 3 This is a module block diagram of a state prediction unit provided in an embodiment of the present invention.

[0022] Figure 4This is a module block diagram of a moving direction analysis unit provided in an embodiment of the present invention.

[0023] Figure 5 This is a module block diagram of a moving distance analysis unit provided in an embodiment of the present invention.

[0024] Figure 6 This is a flowchart of an intelligent cleanliness detection method provided by an embodiment of the present invention.

[0025] In the accompanying drawings: 1. Detector; 2. Air collection component; 3. Moving component; 201. Air inlet duct; 202. Filter; 301. Base; 302. Moving part. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0027] The specific implementation of the present invention is described in detail below with reference to specific embodiments.

[0028] See also Figure 1 and Figure 2 , an intelligent cleanliness detection device provided by an embodiment of the present invention, includes a detector 1, characterized in that the device also includes: A moving component 3, which is arranged on one side of the detector 1 and is used to drive the detector 1 to move; The data acquisition unit 10 is used to acquire the air data of the space where the detector 1 is currently located; The state prediction unit 20 is used to generate an air state prediction value based on the air data; A state judgment unit 30 is used to judge the need for the moving component 3 to drive the detector 1 to move according to the air state prediction value; A moving direction analysis unit 40 is used to obtain the wind direction of the space where the detector 1 is currently located and generate an optimal moving direction; The moving distance analysis unit 50 is used to obtain the wind speed of the space where the detector 1 is currently located and generate the moving distance; A movement control unit 60, when the detector 1 needs to move, the movement control unit 60 is used to control the movement component 3 to drive the detector 1 to move according to the optimal movement direction and movement distance; The detection unit 70, when the moving component 3 drives the detector 1 to move, continues to perform air detection on the space where the detector 1 is located; Specifically, after the detection device is started, the data acquisition unit 10 continuously collects air data at the location, including particulate matter concentration, gas composition, and temperature and humidity parameters. The state prediction unit 20 normalizes the historical data and current data, and inputs them into the air state prediction model to calculate the pollutant concentration distribution trend in the future period; when the air state prediction value is in a normal state, the state judgment unit 30 triggers the movement demand; the movement direction analysis unit 40 synchronously obtains the spatial wind direction data, and selects the direction with the most significant change in pollutant concentration as the movement path based on the calculation results; the movement distance analysis unit 50 calculates the movement distance based on the wind speed sensor data and the modulus of the current cleanliness gradient vector; the movement control unit 60 drives the movement component 3 to move the detector 1 according to the optimal movement direction and movement distance. After the movement is completed, the detection unit 70 performs accurate air cleanliness detection, forming a closed-loop control of data collection-prediction analysis-path planning-movement execution-data collection.

[0029] Preferably, see Figure 1 The present invention further proposes an air collection component 2 for collecting air in the space where the detector 1 is located.

[0030] The air collection component 2 refers to a component for actively collecting air samples; Specifically, the air collection component 2 continuously collects air samples at the location when the detector 1 moves.

[0031] Preferably, see Figure 1 The present invention further proposes that the air collection component 2 specifically includes: An air inlet duct 201, the air inlet duct 201 is provided on one side of the detector 1; A filter 202 is provided at one end of the air inlet duct 201 and is used to filter the air entering the detector 1 through the air inlet duct 201; The air inlet duct 201 is a structured channel for guiding air into the detector 1. Specifically, it can be implemented by a hollow cylindrical metal pipe. Its installation position matches the spatial layout of the detector 1 and the mobile component 3, so that the air collection direction and the movement direction of the device complement each other. The filter 202 refers to a porous medium used to intercept solid particles in the air. Specifically, it can be implemented using a stainless steel woven mesh or a polymer fiber filter membrane. The terminal installation position allows the air to complete initial flow stabilization in the pipeline before filtering, avoiding deformation of the filter material caused by airflow impact.

[0032] Specifically, the air inlet duct 201, positioned on the side of the detector 1, forms a spatially complementary relationship with the mobile assembly 3. When the detector 1 moves in a predetermined direction, the opening of the air inlet duct 201 forms an angle with the direction of movement, utilizing the relative airflow generated by the movement of the device to enhance air collection efficiency. The filter 202, positioned at the end of the duct, ensures that the air undergoes final filtration before entering the detector 1. This structural combination ensures the representativeness of the air sample while preventing contamination of the detector 1's internal precision sensors by large particulate matter.

[0033] Preferably, see Figure 1 The present invention further proposes that the mobile component 3 specifically includes: A base 301, which is provided on one side of the detector 1; A moving member 302 is provided on one side of the base 301 and is rotatably connected to the base 301 to drive the base 301 to move; Specifically, the driving motor drives the moving member 302 to rotate, and the moving member 302 rubs against the ground during rotation, causing the moving member 302 to move, thereby driving the base 301 to move, and the base 301 drives the detector 1 to move, thereby achieving position movement of the detection device; In addition, 302 can be configured as a universal wheel or other parts, so that the moving part 302 can drive the detector 1 to move in multiple directions to achieve all-round detection.

[0034] Preferably, see Figure 3 The present invention further proposes that the state prediction unit specifically includes: A data preprocessing module 21 is used to preprocess air data; The prediction model building module 22 is used to build an air state prediction model based on the pre-processed air data and generate an air state prediction value; The expression of the air state prediction model is specifically:

[0035] Among them, C t+Δt It represents the predicted value of air condition. represents the autoregressive coefficient, C t-j represents the measured value of the air state at the historical moment tj, n represents the number of historical moments, x t It represents the pre-processed air data, B T It represents the static influence weight vector of the preprocessed air data; The data preprocessing module is a processing unit that cleans and standardizes the raw air data. This can be achieved by using filtering algorithms or normalization methods to eliminate noise interference and unify the data dimensions, providing high-quality input for the model. The prediction model building module refers to a linear prediction model built based on time series analysis. Specifically, the autoregressive coefficient can be solved by the least squares method or gradient descent method. and weight vector B T , which is used to fuse historical air status with real-time air data and establish a dynamic prediction relationship.

[0036] Specifically, the data preprocessing module first removes outliers and standardizes the particle concentration, wind speed, temperature and humidity data collected by the sensor; the prediction model building module converts the preprocessed real-time data x t Compared with the historical air condition measured value C t-j Input model, through linear combination coefficients Capturing the continuous trend of pollutant diffusion, the weight matrix B T Then the influence of current environmental parameters on pollutant migration is correlated, and the cleanliness prediction value C at the future moment is finally output. t+Δt ; By integrating historical and real-time data, the model breaks through the limitations of traditional technology that only relies on current data and realizes the active prediction of the migration direction of pollutants.

[0037] Compared to existing technologies, existing fixed sensor networks can only provide local historical data and cannot predict dynamic changes in pollutants. Handheld devices rely on manual operation and lack predictive capabilities. This solution uses a time series model to combine historical trends with real-time environmental parameters to establish a continuous prediction relationship for pollutant diffusion. This allows detection devices to predict pollutant migration paths in advance, providing a scientific basis for mobile decision-making.

[0038] Through the above technical solution, the present application can predict future air state change trends based on historical data and real-time environmental parameters, so that the detection device can actively adjust the movement path, avoid detection lag problems caused by pollutant migration, reduce the number of invalid movements, and improve the efficiency and accuracy of cleanliness detection.

[0039] Preferably, see Figure 4 The present invention further proposes that the moving direction analysis unit specifically includes: A wind direction acquisition module 41 is used to obtain the wind direction of the space where the detector 1 is currently located; The optimal moving direction generating module 42 is used to establish a moving direction analysis model based on the wind direction and wind speed in the space where the detector 1 is currently located, and generate the optimal moving direction; The expression of the moving direction analysis model is specifically:

[0040] Among them, d t represents the optimal moving direction, arg max represents the parameter of the maximum value of the objective function, D represents the set of candidate directions, d m It represents the mth candidate direction, d T m W t It represents the dot product of direction and wind direction, W t is the wind direction vector, I (x i loc+Δs*d m ) represents the information gain of the i-th type of air data, x i represents the i-th type of air data, loc represents the current position coordinate, Δs represents the detection distance, and γ is the information gain weight coefficient; Among them, the wind direction vector W t It refers to the airflow direction data within the detection area of ​​the real-time detector (1), which can be realized by using vector data collected by wind speed sensors or weather stations to reflect the migration trend of pollutants in the current space; I (x i loc+Δs*d m ) refers to the information gain of the i-th category air data, which can be specifically implemented by entropy calculation and is used to quantify the amount of pollution distribution information at potential detection points in different directions; the information gain weight coefficient γ refers to the weight coefficient, which can be specifically implemented by preset empirical values ​​and is used to balance the priority relationship between airflow direction and data exploration; the candidate direction set D refers to a predefined set of movement directions, which can be specifically generated by eight-neighborhood directions or continuous angle discretization to constrain the feasibility of the movement path; Specifically, the moving direction analysis model obtains the wind direction vector W in real time t , calculate the degree of coordination between the candidate direction and the airflow direction, and at the same time combine the information gain function to evaluate the amount of pollution distribution prediction information at the corresponding position of the candidate direction; the model takes the weighted sum of the wind direction coordination degree and the information gain, and selects the direction with the highest comprehensive score from the candidate direction set D as the optimal moving direction through the arg max function. For example, when the detection device is in the pollution diffusion path, the wind direction vector W t The dominant direction is selected to avoid detection lag caused by moving against the wind; when there are unknown pollution hotspots in the detection area, the information gain function guides the detection device to actively explore high-value areas; through the information gain weight coefficient γ, it can adapt to the differences in airflow impact and data exploration needs in different scenarios.

[0041] Compared with existing technologies, existing fixed sensor networks have difficulty capturing the spatial heterogeneity of pollution due to their immobility, and handheld devices rely on manual operation and lack directional optimization capabilities. This solution integrates real-time wind direction data with information gain evaluation to enable the detection device to autonomously adjust its movement direction, avoiding detection delays caused by headwind movement while actively tracking pollution migration paths and unknown pollution sources, significantly improving spatial coverage efficiency.

[0042] Through the above technical solution, this application achieves dynamic optimization of the detection device's movement direction, solving the problem of unreasonable path planning caused by ignoring the influence of airflow in existing technologies. Through the comprehensive decision-making of wind direction coordination and information gain, the device can quickly respond to the diffusion trend of pollutants in the airflow environment, while actively exploring high-value detection areas, effectively improving the ability to identify pollution hotspots and spatial detection coverage.

[0043] Preferably, see Figure 5 The present invention further proposes the moving distance analysis unit, which specifically includes: A wind speed acquisition module 51 is used to acquire the wind speed of the space where the detector 1 is currently located; The moving distance generating module 52 is used to establish a moving distance analysis model and generate a moving distance according to the wind direction of the space where the detector 1 is currently located; The expression of the moving distance analysis model is specifically:

[0044] Among them, s t It represents the moving distance, v t It represents the wind speed in the space where the detector 1 is currently located, ║▽C t ║ represents the gradient modulus, ▽C t It represents the cleanliness spatial gradient vector, and β represents the adaptive distance coefficient; The moving distance analysis model refers to a mathematical model that dynamically calculates the moving distance based on wind speed and cleanliness gradient. Specifically, it can be implemented in the control unit using an embedded algorithm to adjust the moving distance of the detection device in real time according to environmental parameters. Wind speed refers to the airflow velocity at the current detection location, which can be achieved using an ultrasonic wind speed sensor or a hot wire anemometer to reflect the impact of airflow on pollutant diffusion; The cleanliness spatial gradient vector refers to the rate of change vector of pollutant concentration in space. It can be obtained by calculating the spatial derivative of multi-point sampling data and is used to characterize the local change trend of pollutant distribution. The coefficient β refers to the adjustment factor of the moving distance, which can be determined through experimental calibration or machine learning optimization and is used to balance detection accuracy and movement efficiency.

[0045] Specifically, the moving distance analysis model dynamically calculates the moving distance of the detection device by collecting wind speed data in real time and combining it with the modulus of the cleanliness gradient vector in the current area. When the wind speed increases, the moving distance is positively correlated with the wind speed, allowing the device to quickly expand the detection range in a strong airflow environment; when the cleanliness gradient increases, the denominator automatically reduces the moving distance, prompting the device to perform refined detection in areas where the pollutant concentration changes drastically. The coefficient β is used as a global adjustment parameter to adjust the movement sensitivity according to the needs of different scenarios. For example, in high-pollution risk areas such as industrial workshops, a smaller β value can be set to increase the detection density; in open spaces, the β value can be increased to improve coverage efficiency. The model achieves dynamic optimization of the detection path by integrating fluid mechanics parameters and pollutant distribution characteristics.

[0046] Compared to existing technologies, traditional fixed sensor networks rely on a preset fixed movement distance and are unable to adjust their detection range based on wind speed fluctuations, resulting in missed areas of contamination diffusion in strong winds. Handheld devices also rely on manual operation, making it difficult to calculate cleanliness gradients in real time and precisely control movement distance. This solution overcomes the limitations of static detection by establishing a mathematical correlation model that enables movement distance to respond simultaneously to changes in airflow dynamics and pollutant distribution.

[0047] Through the above technical solution, this application achieves adaptive movement control of the detection device in complex airflow environments. In strong winds, the movement distance is increased to avoid detection blind spots. In areas with sudden changes in pollutant concentration, the movement distance is reduced to accurately locate the pollution source. This solves the problem of low detection efficiency caused by the fixed movement distance of traditional technologies. This solution can autonomously adjust the movement strategy based on real-time environmental data, balancing detection accuracy and coverage efficiency.

[0048] Preferably, see Figure 6 The present invention also proposes an intelligent cleanliness detection method, which is applied to the above-mentioned intelligent cleanliness detection device and specifically includes the following steps: S1: Place the detection device in the area to be detected; S2: Acquire air data of the space where the detector 1 is currently located; S3: Generate air state prediction value based on air data; S4: judging the need for the moving component 3 to drive the detector 1 to move according to the predicted value of the air state; S5: Obtain the wind direction of the space where the detector 1 is currently located and generate an optimal moving direction; S6: Obtain the wind speed of the space where the detector 1 is currently located and generate the moving distance; S7: Controlling the moving component 3 to drive the detector 1 to move according to the optimal moving direction and moving distance; S8: After the moving component 3 has driven the detector 1 to move, continue to perform air detection on the space where the detector 1 is located; Among them, the air state prediction value refers to the numerical value that quantitatively predicts the diffusion trend of pollutants through a mathematical model. Specifically, it can be calculated by using a time series model combined with environmental parameters, and is used to predict changes in air states in the future. Movement demand refers to the judgment result of whether the detection device needs to adjust its position. Specifically, it can trigger a movement instruction by comparing the difference between the predicted value and the preset threshold, which is used to avoid energy loss caused by invalid movement. Among them, the optimal movement direction refers to the displacement vector that is tracked in the opposite direction of the pollutant diffusion path. Specifically, the information gain function can be combined with the wind direction vector for multi-objective optimization calculation to improve the accuracy of pollution source positioning. The moving distance refers to the distance of a single displacement of the detection device, which can be calculated by wind speed and cleanliness gradient vector to prevent positioning deviation caused by airflow disturbance.

[0049] Specifically, after the detection device is deployed in the area to be tested, it collects air data in real time through built-in sensors. The air state prediction model integrates and analyzes current and historical data to generate pollutant concentration change trends within the future time window. When the predicted value is in a normal state, the movement decision mechanism is triggered. At this time, the wind direction sensor and anemometer respectively collect ambient airflow parameters. The movement direction analysis model calculates the optimal path for reverse tracking of the pollution source based on the wind direction vector and the information gain function. The movement distance analysis model adjusts the movement distance based on the wind speed and cleanliness gradient. After the detection device autonomously moves to the target position according to the calculated optimal movement direction and movement distance, it performs high-precision air detection.

[0050] Compared to existing technologies, fixed sensor networks can only collect static data from pre-set nodes and cannot actively track pollution diffusion paths. Handheld devices rely on manual operation and lack environmental parameter feedback mechanisms. This method, by synergizing predictive models with real-time airflow parameters, enables detection devices to dynamically adjust their movement paths, achieving full coverage detection of spatially heterogeneous pollution without manual intervention.

[0051] Through the above technical solution, this application solves the problem of limited coverage of fixed sensor networks, improves the integrity of spatial pollution detection through dynamic path planning; eliminates the dependence of handheld devices on manual operation, and realizes autonomous mobile decision-making; overcomes the positioning error caused by traditional technology ignoring the influence of airflow, and uses wind direction and wind speed parameters to optimize movement parameters, so that the detection results more accurately reflect the actual pollution distribution status.

[0052] Preferably, the present invention further proposes that step S2 specifically includes: S2.1: Preprocess air data; S2.2: Establish an air state prediction model based on the preprocessed air data to generate an air state prediction value; Specifically, the air data preprocessing phase involves data cleaning and standardization, converting parameters such as particulate matter concentration and gas composition ratios collected by different sensors into a unified input sequence. This eliminates noise interference caused by device response delays or sudden environmental changes. The air state prediction model outputs predicted air state values ​​within a future time window. This predicted value reflects the spatial migration trend of pollutants and provides a basis for dynamic adjustment in subsequent movement path planning.

[0053] Preferably, the present invention further proposes that step S5 specifically includes: S5.1: Obtain the wind direction of the space where the detector 1 is currently located; S5.2: Establishing a movement direction analysis model based on the wind direction in the space where the detector 1 is currently located to generate an optimal movement direction; Specifically, this technical solution uses a wind speed sensor to obtain the wind direction of the current space in real time, converts the wind direction into vector data, and then inputs it into the movement direction analysis model. The model constructs a set of candidate directions based on the wind direction vector and calculates a comprehensive score consisting of the wind direction dominant component and the information gain component for each candidate direction. The information gain component evaluates the amount of pollution distribution information that can be obtained at the corresponding position of the candidate direction, giving priority to areas that have not been fully detected or where the pollution gradient changes significantly. Finally, the direction with the highest comprehensive score is selected as the optimal movement direction, so that the detection device can actively cover the key detector (1) points on the pollutant diffusion path while conforming to the movement law of the airflow.

[0054] Preferably, the present invention further proposes that step S6 specifically includes: S6.1: Obtain the wind speed of the space where the detector 1 is currently located; S6.2: Establishing a movement distance analysis model based on the wind direction of the space where the detector 1 is currently located to generate a movement distance; Among them, obtaining spatial wind speed refers to real-time measurement of the airflow speed in the area where the detection device is located. Specifically, this can be achieved by using a wind speed sensor. This feature provides dynamic environmental parameters for the movement distance calculation, avoiding the mismatch between the movement distance and the actual environment due to ignoring the airflow intensity; the movement distance analysis model refers to a mathematical relationship constructed based on wind speed and cleanliness gradient vector. Specifically, it can be implemented by using an embedded processor or microcontroller to perform calculations. The model dynamically adjusts the movement step size by associating wind speed with gradient changes. When the airflow interference is strong, the movement distance is shortened to maintain positioning accuracy. When the pollution spread is significant, the movement distance is increased to improve detection efficiency. Specifically, during the movement of the detection device, the sensor first collects the wind speed data of the current space in real time. The wind speed and the pre-calculated cleanliness spatial gradient vector are then input into the movement distance analysis model. The model converts the ratio of wind speed to gradient vector into actual movement distance based on the proportional coefficient and normalization processing rules set in the formula. For example, when a high wind speed is detected, the model automatically reduces the movement distance to prevent the device from being disturbed by airflow and deviating from the target area. When the cleanliness gradient changes significantly, the model appropriately increases the movement distance to quickly cover the pollution diffusion range. As a result, the movement distance generation process can dynamically adapt to environmental changes and achieve accurate planning of the detection path, thereby improving the rationality of path planning and data collection accuracy of cleanliness detection.

[0055] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent cleanliness detection device, used in the field of controlling cleanliness detection, comprising a detector (1), characterized in that: The device also includes: A moving component (3), the moving component (3) being arranged on one side of the detector (1) and being used to drive the detector (1) to move; A data acquisition unit, used to acquire air data of the space where the detector (1) is currently located; A state prediction unit, configured to generate an air state prediction value based on air data; A state judgment unit, used for judging the need for the moving component (3) to drive the detector (1) to move according to the air state prediction value; A moving direction analysis unit, used to obtain the wind direction of the space where the detector (1) is currently located and generate an optimal moving direction; A moving distance analysis unit, used to obtain the wind speed of the space where the detector (1) is currently located and generate a moving distance; A movement control unit, when the detector (1) has a movement requirement, the movement control unit is used to control the movement component (3) to drive the detector (1) to move according to the optimal movement direction and movement distance; The detection unit continues to perform air detection on the space where the detector (1) is located after the moving component (3) drives the detector (1) to move.

2. An intelligent cleanliness detection device according to claim 1, characterized in that: The device also includes: The air collection component (2) is used to collect air in the space where the detector (1) is located.

3. An intelligent cleanliness detection device according to claim 2, characterized in that: The air collection component (2) specifically includes: An air inlet duct (201), the air inlet duct (201) being arranged on one side of the detector (1); A filter (202) is provided at one end of the air inlet duct (201) and is used for filtering the air entering the detector (1) through the air inlet duct (201).

4. The intelligent cleanliness detection device according to claim 1, characterized in that: The state prediction unit specifically includes: Data preprocessing module, used for preprocessing air data; The prediction model building module is used to build an air state prediction model based on the preprocessed air data and generate air state prediction values.

5. The intelligent cleanliness detection device according to claim 4, characterized in that: The expression of the air state prediction model is specifically: ； Among them, C t+Δt It represents the predicted value of air condition. represents the autoregressive coefficient, C t-j represents the measured value of the air state at the historical moment tj, n represents the number of historical moments, x t It represents the pre-processed air data, B T It represents the static influence weight vector of the preprocessed air data.

6. The intelligent cleanliness detection device according to claim 1, characterized in that: The moving direction analysis unit specifically includes: A wind direction acquisition module, used to acquire the wind direction of the space where the detector (1) is currently located; An optimal moving direction generating module is used to establish a moving direction analysis model based on the wind direction and wind speed in the space where the detector (1) is currently located, and to generate an optimal moving direction; The expression of the moving direction analysis model is specifically: ； Among them, d t represents the optimal moving direction, arg max represents the parameter of the maximum value of the objective function, D represents the set of candidate directions, d m It represents the mth candidate direction, d T m W t It represents the dot product of direction and wind direction, W t is the wind direction vector, I (x i loc+Δs*d m ) represents the information gain of the i-th type of air data, x i represents the i-th type of air data, loc represents the current position coordinate, Δs represents the detection distance, and γ is the information gain weight coefficient.

7. An intelligent cleanliness detection device according to claim 6, characterized in that: The moving distance analysis unit specifically includes: A wind speed acquisition module, used to acquire the wind speed of the space where the detector (1) is currently located; A moving distance generating module is used to establish a moving distance analysis model and generate a moving distance according to the wind direction of the space where the detector (1) is currently located; The expression of the moving distance analysis model is specifically: ； Among them, s t It represents the moving distance, v t It represents the wind speed in the space where the detector (1) is located, ║▽C t ║ represents the gradient modulus, ▽C t It represents the cleanliness spatial gradient vector, and β represents the adaptive distance coefficient.

8. An intelligent cleanliness detection method, characterized in that: The method is applied to an intelligent cleanliness detection device according to any one of claims 1 to 7, and specifically comprises the following steps: S1: Place the detection device in the area to be detected; S2: Acquire air data of the space where the detector (1) is currently located; S3: Generate air state prediction value based on air data; S4: judging the need for the moving component (3) to drive the detector (1) to move according to the predicted value of the air state; S5: Obtain the wind direction of the space where the detector (1) is currently located, and generate an optimal moving direction; S6: Obtain the wind speed of the space where the detector (1) is currently located and generate a moving distance; S7: controlling the moving component (3) to drive the detector (1) to move according to the optimal moving direction and moving distance; S8: After the moving component (3) has driven the detector (1) to move, the air in the space where the detector (1) is located is continued to be detected.

9. An intelligent cleanliness detection method according to claim 8, characterized in that: Generating an air state prediction value based on the air data specifically includes the following steps: S2.1: Preprocess air data; S2.2: Establish an air state prediction model based on the preprocessed air data and generate an air state prediction value.

10. An intelligent cleanliness detection method according to claim 8, characterized in that: The step of obtaining the wind direction of the space where the detector (1) is currently located and generating the optimal moving direction specifically includes: S5.1: Obtain the wind direction of the space where the detector (1) is currently located; S5.2: establishing a movement direction analysis model based on the wind direction of the space where the detector (1) is currently located, and generating an optimal movement direction; The step of obtaining the wind speed in the space where the detector (1) is currently located and generating the moving distance specifically includes: S6.1: Obtain the wind speed of the space where the detector (1) is currently located; S6.2: According to the wind direction of the space where the detector (1) is currently located, a movement distance analysis model is established to generate the movement distance.

Images