Air treatment method, system and equipment with pet house linkage purification function

Abstract

The invention discloses an air treatment method, system and equipment with a pet house linkage purification function, and relates to the technical field of air purification. The method comprises the steps that a pet house sensing module is deployed on a target pet house, and pet house multi-dimensional sensing data streams are collected; fusing and analyzing the data and predicting the pollutant concentration to obtain the predicted pollutant concentration of the pet house; an air purification linkage module is tested, a graded purification parameter-pollutant concentration strategy space is established, and target air purification strategy parameters are analyzed and determined; and an air purification path is planned, and dynamic air purification treatment is conducted based on the target air purification strategy parameters and the air purification path. The technical problem that traditional air purification equipment is difficult to deal with insufficient purification or excessive energy consumption caused by dynamic change of pollutants due to fluctuation of temperature and humidity in a pet house and pet activities is solved, and accurate pollutant identification, on-demand matching of purification strategies and optimal path coverage in a pet house scene are achieved; and the technical effects of purifying effect and energy consumption are balanced.

Description

Technical Field

[0001] This invention relates to the field of air purification technology, and in particular to an air treatment method, system, and device with integrated air purification function for pet beds. Background Technology

[0002] Air purification equipment is widely used to improve indoor air quality. However, pet beds, as a frequently used space for pets, are prone to odors and pollutants that can affect pets' health and the surrounding environment. Most existing air purification devices are general-purpose for the whole house, relying on a single sensor or operating with fixed parameters. This presents significant limitations in pet bed scenarios: due to fluctuations in temperature and humidity within the pet bed and dynamic changes in pollutants caused by pet activity, traditional devices cannot accurately collect multi-dimensional environmental data, predict pollutant concentration trends, or match purification strategies to the specific pollution characteristics of the pet bed. This often results in insufficient purification, residual odors, or excessive energy consumption, failing to meet the needs for accurate air quality assessment and dynamic purification in pet bed scenarios. Summary of the Invention

[0003] This application provides an air treatment method, system, and device with a pet bed-linked purification function, which is used to solve the technical problem that traditional air purification equipment is unable to handle the temperature and humidity fluctuations in pet beds and the dynamic changes of pollutants caused by pet activities, resulting in insufficient purification or excessive energy consumption.

[0004] The first aspect of this application provides an air treatment method with a pet bed-linked air purification function. The method includes: deploying a pet bed sensing module on a target pet bed, the pet bed sensing module integrating a temperature and humidity sensor, a VOC gas sensor, and an infrared thermal imaging module; collecting multi-dimensional sensing data streams of the pet bed through the pet bed sensing module; performing fusion analysis and pollutant concentration prediction on the multi-dimensional sensing data streams of the pet bed to obtain a predicted pollutant concentration in the pet bed; acquiring an air purification linkage module; testing and recording the air purification linkage module; establishing a graded purification parameter-pollutant concentration strategy space; using the graded purification parameter-pollutant concentration strategy space to analyze the processing strategy of the predicted pollutant concentration in the pet bed to determine target air purification strategy parameters; planning and determining an air purification path for the target pet bed; and performing dynamic air purification treatment on the target pet bed based on the target air purification strategy parameters and the air purification path.

[0005] A second aspect of this application provides an air treatment system with pet bed-linked purification function. The system includes: a pet bed multi-dimensional sensing data stream acquisition module, used to deploy a pet bed sensing module on a target pet bed. The pet bed sensing module integrates a temperature and humidity sensor, a VOC gas sensor, and an infrared thermal imaging module, and acquires a pet bed multi-dimensional sensing data stream; a pet bed predicted pollutant concentration acquisition module, used to perform fusion analysis and pollutant concentration prediction on the pet bed multi-dimensional sensing data stream to obtain the predicted pollutant concentration of the pet bed; a target air purification strategy parameter acquisition module, used to acquire the air purification linkage module, test and record the air purification linkage module, establish a graded purification parameter-pollutant concentration strategy space, and use the graded purification parameter-pollutant concentration strategy space to analyze the processing strategy of the predicted pollutant concentration of the pet bed to determine the target air purification strategy parameters; and a dynamic air purification processing execution module, used to plan and determine the air purification path of the target pet bed, and perform dynamic air purification processing on the target pet bed based on the target air purification strategy parameters and the air purification path.

[0006] A third aspect of this application provides an electronic device comprising: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement an air treatment method with a pet bed-linked air purification function.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] This application utilizes a sensing module integrating temperature and humidity sensors, VOC gas sensors, and infrared thermal imaging modules deployed in the target pet bed to collect multi-dimensional sensing data streams. The data is then fused and analyzed using an LSTM network to predict the pollutant concentration in the pet bed. An air purification linkage module is tested to establish a hierarchical purification parameter-pollutant concentration strategy space. By combining a fitness function to optimize and determine the target air purification strategy parameters, an air purification path is planned. Based on these parameters and the path, the target pet bed is dynamically purified. This achieves accurate pollutant identification and on-demand purification in the pet bed scenario, making air treatment more adaptable to the characteristics of the pet bed environment, improving purification efficiency and accuracy. The application achieves the technical benefits of accurate pollutant identification, on-demand matching of purification strategies, and optimal path coverage in the pet bed scenario, balancing purification effect and energy consumption. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic flowchart of an air treatment method with pet bed linkage purification function provided in an embodiment of this application.

[0011] Figure 2 This is a schematic diagram of the air handling system with pet bed linkage purification function provided in the embodiments of this application.

[0012] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0013] Figure labeling: 1. Pet bed multi-dimensional sensing data stream acquisition module; 2. Pet bed predicted pollutant concentration acquisition module; 3. Target air purification strategy parameter acquisition module; 4. Dynamic air purification processing execution module; 5. Input device 301; 6. Processor 302; 7. Memory 303; 8. Output device 304. Detailed Implementation

[0014] This application provides an air treatment method, system, and device with a pet bed-linked purification function, which is used to solve the technical problem that traditional air purification equipment is unable to handle the temperature and humidity fluctuations in pet beds and the dynamic changes of pollutants caused by pet activities, resulting in insufficient purification or excessive energy consumption.

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0016] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0017] Example 1, as Figure 1 As shown, an air treatment method with a pet bed-linked air purification function is provided, wherein the method includes:

[0018] Step A100: Deploy a pet bed sensing module on the target pet bed. The pet bed sensing module integrates a temperature and humidity sensor, a VOC gas sensor, and an infrared thermal imaging module. The pet bed sensing module collects multi-dimensional sensing data streams of the pet bed.

[0019] In this embodiment of the application, the purpose of the VOC gas sensor is to collect VOC gas in the target pet's den, such as the concentration data of organic matter related to odors generated by pet activities. These data are a key component of the multidimensional sensing data stream of the pet's den.

[0020] Specifically, the first step is to deploy a dedicated pet-nest sensing module on the target pet's nest. This module integrates three core sensing components: a temperature and humidity sensor, a VOC gas sensor, and an infrared thermal imaging module. In practical applications, the temperature and humidity sensor can collect real-time temperature and humidity data within the pet's nest, ranging from 0-50°C and relative humidity from 20%-95%. The VOC gas sensor can accurately detect VOC gas content within the 0-1000ppm concentration range, while the infrared thermal imaging module can capture thermal imaging signals within the normal body temperature range of 30°C-40°C for pets. These sensing components work together to simultaneously acquire multi-dimensional information such as temperature and humidity, VOC gas concentration, and whether the pet is in the nest, at a sampling frequency of once every 5 minutes during the pet's daily use cycle. This data is then integrated to form a multi-dimensional sensing data stream for the pet's nest.

[0021] During the data collection process, temperature and humidity data can reflect the influence of the environment on VOC gas diffusion. For example, when the temperature exceeds 30℃, the volatilization rate of VOC gas will accelerate, and the corresponding concentration data will show an upward trend. Infrared thermal imaging data can distinguish whether the pet is in the den, avoiding unnecessary interference to the pet during subsequent purification operations. VOC gas concentration data is directly related to the odor pollution level of the pet den. The combination of these three data forms a multi-dimensional sensing data stream for the pet den, which makes up for the incompleteness of data from a single sensor and provides complete and reliable raw data support for subsequent fusion analysis, pollutant concentration prediction, and the formulation of precise purification strategies.

[0022] By deploying a pet bed sensing module integrating three major sensing components in the target pet bed and collecting multi-dimensional data periodically, comprehensive and accurate information about the pet bed environment was obtained, providing a reliable data foundation for subsequent precise air purification in the pet bed.

[0023] Step A200: Perform fusion analysis and pollutant concentration prediction on the multidimensional sensing data stream of the pet bed to obtain the predicted pollutant concentration of the pet bed.

[0024] Optionally, a multi-dimensional data preprocessing program with noise filtering and outlier cleaning is first constructed based on the data characteristics of the multi-dimensional sensing data stream of the pet nest. After preprocessing the data, feature sets of each dimension are extracted, and then multi-modal fusion and pollutant concentration prediction are performed on the feature sets of each dimension to obtain the predicted pollutant concentration of the pet nest. The specific steps are explained in detail in A210-A240.

[0025] Step A300: Obtain the air purification linkage module, test and record the air purification linkage module, establish a graded purification parameter-pollutant concentration strategy space, use the graded purification parameter-pollutant concentration strategy space to analyze the treatment strategy of the predicted pollutant concentration in the pet bed, and determine the target air purification strategy parameters.

[0026] In one embodiment of this application, firstly, according to the pet bed application scenario and the type of pet, the pollutant component type and concentration distribution threshold are preset, the air purification linkage module is tested to obtain air purification parameters and effect data, the parameters are optimized and then the correlation and hierarchical strategy is fitted to establish the hierarchical purification parameter-pollutant concentration strategy space. The specific steps are described in detail in A310-A340.

[0027] Next, the pet nest is used to predict the pollutant concentration by spatial matching of graded purification parameters and pollutant concentration strategies to obtain the threshold of the matching air purification strategy. An air purification fitness function is constructed, and then the matching air purification strategy threshold is globally optimized to determine the target air purification strategy parameters. The specific steps are explained in detail in A350-A370.

[0028] Step A400: Plan and determine the air purification path of the target pet bed, and perform dynamic air purification treatment on the target pet bed based on the target air purification strategy parameters and the air purification path.

[0029] Specifically, firstly, a 3D model of the target pet's den is performed and airflow simulation is conducted to obtain a set of simulated diffusion areas for polluted gases. Then, a set of diffusion sequence areas for polluted gases is obtained by arranging them according to concentration information. Finally, the air purification path of the target pet's den is determined through regional polling planning analysis. The specific steps are explained in detail in A410-A430.

[0030] Next, a linkage mechanism between the target air purification strategy parameters and the air purification path will be established. Based on the determined target air purification strategy parameters, such as a fan speed of 5 m / s, a HEPA + activated carbon composite filter, and a 15-minute dwell time in the core area for ammonia VOC concentrations of 0.9-1.1 ppm, a one-to-one correspondence must be established with each area of ​​the air purification path: the core area within the vent, the top diffusion area, and the frontal outer diffusion area. When the purification equipment operates according to the air purification path to the core area within the vent, the corresponding target air purification strategy parameters for that area will be automatically invoked, the composite filter will be activated, and the fan speed will be adjusted to 5 m / s to ensure that the high-concentration area is effectively purified. Sufficient purification intensity; when the air purification path switches to the top diffusion area, the target air purification strategy parameters adapted to the top area are called simultaneously. Since the concentration at the top is slightly lower, the wind speed can be finely adjusted to 4.5m / s, while the composite filter is kept running, which ensures the purification effect and avoids excessive energy consumption; when the air purification path enters the front peripheral diffusion area, the low concentration adaptation parameters are called, the wind speed is reduced to 3.5m / s, and the filter mode remains unchanged. Through the linkage between the target air purification strategy parameters and the air purification path, precise execution of parameters adapted to different areas is achieved.

[0031] During the coordinated execution process, the pet bed sensing module continuously collects real-time environmental data, forming a dynamic feedback loop. The pet bed sensing module synchronously acquires temperature, humidity, VOC concentration, and infrared thermal imaging data of the current purification area every 5 minutes: temperature and humidity data are used to determine whether the environment affects purification efficiency; for example, when the temperature rises to 32℃, VOC evaporation accelerates, requiring an assessment of whether parameters need adjustment; VOC concentration data directly reflects the current purification effect; infrared thermal imaging data monitors whether pets have entered the current purification area; if thermal imaging shows a pet lying in the core area, direct airflow should be avoided. This real-time data is transmitted to the purification control unit in real time, providing a basis for subsequent adjustments.

[0032] Then, based on real-time feedback data, the target air purification strategy parameters and air purification path are dynamically adjusted to ensure that purification is always adapted to the current state. If the concentration in the current area is not as expected, the control unit will increase the purification intensity within the allowed range of the strategy parameters based on preset parameter adjustment rules. This can extend the residence time in the core area of ​​the pet's enclosure by 5 minutes while maintaining the fan speed, until the concentration drops below the expected target before switching to the next path area. If thermal imaging data shows that the pet has entered the top diffusion area, the purification control unit will fine-tune the path angle, raising the air outlet by 5° to prevent the airflow from blowing directly onto the pet, while maintaining the current fan speed to ensure uninterrupted purification. If temperature and humidity data show a sudden increase in humidity, which will affect the VOC diffusion rate, the residence time in each area can be appropriately extended, such as by 2 minutes per area, to compensate for the impact of humidity on purification efficiency. Ultimately, a dynamic cycle of data feedback, automatic adjustment, and efficient purification is achieved.

[0033] By establishing a linkage mechanism between parameters and paths, relying on real-time data feedback from sensing modules, and dynamically adjusting operations based on feedback, the air purification process in pet beds is made to precisely adapt to dynamic environmental changes, balancing purification effectiveness, energy consumption control, and pet comfort.

[0034] Furthermore, step A200 in the method provided in this application embodiment includes:

[0035] A210: Based on the data characteristics of the multidimensional sensing data stream of the pet nest, a multidimensional data preprocessing program is constructed, which includes noise filtering and outlier cleaning.

[0036] A220: The multidimensional data preprocessing program is used to preprocess the multidimensional sensing data stream of the pet bed to obtain a usable multidimensional sensing data stream of the pet bed.

[0037] A230: Extract features from each dimension of the multidimensional sensing data stream of the available pet nest to obtain temperature and humidity feature sets, gas composition change feature sets, and thermal imaging feature sets.

[0038] A240: Perform multimodal fusion and pollutant concentration prediction on the temperature and humidity feature set, gas composition change feature set, and thermal imaging feature set to obtain the predicted pollutant concentration in the pet kennel.

[0039] Specifically, a multidimensional data preprocessing program is first constructed based on the data characteristics of the multidimensional sensing data stream from the pet's den. Since temperature and humidity data exhibit periodic fluctuations, VOC gas data contains transient pulse interference, and infrared thermal imaging data includes spatial pixel noise, the multidimensional data preprocessing program needs to integrate noise filtering and outlier cleaning functions. Noise filtering can employ a moving average algorithm to smooth the temperature, humidity, and VOC gas data collected every 5 minutes, filtering out high-frequency transient noise. Outlier cleaning involves setting reasonable threshold ranges, such as a temperature and humidity threshold of 0-50℃, a relative humidity threshold of 20%-95%, and a VOC gas concentration threshold of 0-1000ppm, to remove outlier data points that exceed normal physical ranges or deviate from the data mean by three times the standard deviation, ensuring that the multidimensional data preprocessing program can adapt to the characteristics of different dimensions of data.

[0040] Next, the multi-dimensional sensing data stream of the pet bed is input into a multi-dimensional data preprocessing program for preprocessing. For example, when the temperature and humidity sensor collects continuous data of 25℃, 28℃, 55℃, and 26℃, outlier cleaning removes abnormal data of 55℃ that exceeds the normal environmental fluctuation range of the pet bed. After noise filtering, the remaining data is then averaged to obtain smooth temperature and humidity data of approximately 26.3℃. For the VOC gas sensor collecting data of 1005ppm, 2.1ppm, 2.3ppm, and 2.2ppm, outlier cleaning removes the 1005ppm interference data, and after noise filtering, a stable VOC concentration of 2.2ppm is obtained. Infrared thermal imaging data is cleaned to remove pixels with brightness values ​​below 50 due to ambient light interference, and then noise filtering smooths the pixel grayscale, finally obtaining a usable multi-dimensional sensing data stream of the pet bed, laying the foundation for subsequent feature extraction.

[0041] Subsequently, feature extraction was performed on each dimension of the multidimensional sensing data stream of available pet beds. For available temperature and humidity data, the hourly average temperature and humidity, and the rate of temperature and humidity change (e.g., the temperature and humidity difference every 10 minutes) were extracted to form a temperature and humidity feature set, which reflects the trend of environmental influence on pollutant diffusion. For available VOC gas data, the hourly VOC concentration peak, average concentration, and concentration rise / fall rate were extracted to form a gas composition change feature set, which is directly related to the pollution level in the pet bed. For available infrared thermal imaging data, the duration of pets in the bed and the stable duration of thermal imaging signals were extracted to form a thermal imaging feature set, which is used to determine the impact of pet activities on pollutants. Through the multidimensional feature extraction steps described above, scattered sensing data can be transformed into structured features with clear physical meaning that can be used for subsequent analysis.

[0042] Finally, the temperature and humidity, gas composition changes, and thermal imaging feature sets are arranged and integrated according to time series information to obtain the corresponding sequence feature sets. Then, based on the temperature and humidity and thermal imaging sequence feature sets, the gas composition change sequence feature sets are fused with multimodal influence to obtain the gas composition change fused feature set. Finally, the LSTM network is used to identify pollution change trends and predict pollutant concentrations from the fused feature set to obtain the predicted pollutant concentration in the pet kennel. The specific steps are explained in detail in A241-A243.

[0043] By constructing a preprocessing program adapted to the characteristics of multidimensional data, preprocessing the raw data to obtain usable data, and then extracting multidimensional feature sets, the interference information in the multidimensional perception data stream of pet nests is removed and transformed into a structured feature set, providing high-quality data support for subsequent multimodal fusion and pollutant concentration prediction.

[0044] Furthermore, step A240 in the method provided in this application embodiment includes:

[0045] A241: Arrange and integrate the temperature and humidity feature set, gas composition change feature set, and thermal imaging feature set according to time sequence information to obtain the temperature and humidity sequence feature set, gas composition change sequence feature set, and thermal imaging sequence feature set.

[0046] A242: Based on the temperature and humidity sequence feature set and the thermal imaging sequence feature set, multimodal influence fusion is performed on the gas composition change sequence feature set to obtain a gas composition change fused feature set.

[0047] A243: Using an LSTM network structure, the gas composition change fusion feature set is used to identify pollution change trends and predict pollutant concentrations, thereby obtaining the predicted pollutant concentration of the pet kennel.

[0048] In this embodiment, the LSTM network structure is a special recurrent neural network used to effectively capture long-term dependencies in time-series data, that is, to identify pollution change trends and predict pollutant concentrations by fusing feature sets of gas composition changes.

[0049] Optionally, the temperature and humidity feature sets, gas composition change feature sets, and thermal imaging feature sets are first arranged and integrated according to time-series information. The time-series information can be determined based on the time nodes of data acquisition, with each acquisition cycle being 5 minutes. The temperature and humidity features, gas composition change features, and thermal imaging features corresponding to the acquisition cycles within a 24-hour period are arranged sequentially, transforming the originally independent feature sets of each dimension into a time-series feature set of temperature and humidity, gas composition change, and thermal imaging features. This allows the changing trends of each dimension to be intuitively reflected through time-series relationships, laying the foundation for subsequent fusion analysis based on temporal correlation.

[0050] After the time-series information is arranged and integrated, the pet den sensing module calls the historical multidimensional sensing dataset of the pet den. Based on this dataset, the gas composition influence analysis is performed on the temperature and humidity sequence feature set and the thermal imaging sequence feature set to determine the gas composition influence coefficient of temperature and humidity and the gas composition influence coefficient of thermal imaging. Then, these two coefficients are used to perform multimodal gain fusion on the gas composition change sequence feature set to obtain the gas composition change fused feature set. The specific steps are explained in detail in A242-1-A242-3.

[0051] Subsequently, an LSTM network structure was used to process the fusion feature set of gas composition changes. LSTM networks excel at capturing long-term dependencies in time-series data, effectively identifying periodic trends in pollutant concentrations in pet kennels, such as VOC concentration peaks caused by pets excreting at fixed times each day. In the processing, the LSTM network first identifies pollution change trends in the time-series data within the fusion feature set, distinguishing different stages such as concentration increases, stabilization, and decreases. Then, based on the identified trends and historical data patterns, pollutant concentrations for a future period are predicted, such as the VOC concentration changes over the next hour, ultimately yielding the predicted pollutant concentration in the pet kennel.

[0052] By integrating the feature sets in time sequence, fusing the gas composition sequence with multimodal features, and using an LSTM network for prediction, the system achieves the effect of accurately predicting the pollutant concentration in pet kennels, fully considering the dynamic changes in pet kennel pollution and the influence of multiple factors.

[0053] Furthermore, step A242 in the method provided in this application embodiment includes:

[0054] A242-1: Based on the pet den sensing module, call the historical multidimensional sensing dataset of the pet den.

[0055] A242-2: Based on the historical multidimensional sensing dataset of the pet den, gas composition influence analysis is performed on the temperature and humidity sequence feature set and the thermal imaging sequence feature set to determine the gas composition influence coefficient of temperature and humidity and the gas composition influence coefficient of thermal imaging.

[0056] A242-3: Using the temperature and humidity gas composition influence coefficient and the thermal imaging gas composition influence coefficient, multimodal gain fusion is performed on the gas composition change sequence feature set to obtain the gas composition change fusion feature set.

[0057] Specifically, the process begins by accessing the historical multidimensional sensing dataset of the pet kennel, which has been collected and stored by the deployed pet kennel sensing module. This historical data encompasses temperature and humidity data, VOC concentration data, and infrared thermal imaging data for the pet kennel at different time periods and under different pet activity states. For example, it includes temperature and humidity records and VOC concentration records every 5 minutes from 8:00 to 22:00 daily for the past 30 days, as well as thermal imaging records of each time a pet enters or leaves the kennel. This provides historical evidence that fits the actual scenario of the pet kennel for subsequent analysis, avoiding the mismatch between general experience values ​​and specific scenarios.

[0058] Next, based on the historical multidimensional sensing dataset of pet dens, gas composition influence analysis was conducted on the temperature and humidity sequence feature set and the thermal imaging sequence feature set, respectively. During the analysis, historical data was used to mine the correlation between temperature and humidity and gas composition. For example, the average change in VOC concentration was extracted from historical data for every 1°C change in temperature within the 20-30°C range. If the data showed that for every 1°C increase in temperature inside the pet den, the VOC concentration increased by an average of 0.15 ppm, then the temperature and humidity gas composition influence coefficient for that scenario could be calculated. At the same time, the correlation between thermal imaging sequences and gas composition was analyzed. For example, the correspondence between the duration of pet stay in the den and VOC concentration was statistically analyzed in historical data. If for every hour a pet stayed in the den, the VOC concentration increased by an average of 0.4 ppm, then the corresponding thermal imaging gas composition influence coefficient could be determined. Through this quantitative analysis based on historical data, the two influence coefficients can accurately reflect the actual environmental correlation characteristics of the pet den.

[0059] Finally, multimodal gain fusion is performed on the gas composition change sequence feature set using the determined temperature and humidity gas composition influence coefficients and thermal imaging gas composition influence coefficients. During fusion, the temperature and humidity change amplitude in the current temperature and humidity sequence feature set is multiplied by the corresponding temperature and humidity gas composition influence coefficient to obtain the gain value of temperature and humidity on the current gas composition; simultaneously, the pet's time in its den and thermal signal intensity in the current thermal imaging sequence feature set are multiplied by the corresponding thermal imaging gas composition influence coefficient to obtain the gain value of thermal imaging on the current gas composition. These two gain values ​​are then superimposed on the corresponding time period data of the gas composition change sequence feature set. For example, if the temperature and humidity gain value is 0.2 and the thermal imaging gain value is 0.3 for a certain time period, and the original VOC concentration data for that time period is 1.2 ppm, then the fused data is adjusted to 1.2 + 0.2 + 0.3 = 1.7 ppm, ultimately obtaining a gas composition change fused feature set that accurately reflects the actual impact of temperature, humidity, and thermal imaging on gas composition.

[0060] By calling up the historical multidimensional sensing dataset of pet dens, determining two influence coefficients based on historical data, and then using the coefficients to perform gain fusion on the gas composition sequence feature set, the effect of accurately matching the gas composition change fusion feature set with the actual scene correlation pattern of the pet den is achieved, thereby improving the accuracy of subsequent concentration prediction.

[0061] Furthermore, step A300 in the method provided in this application embodiment includes:

[0062] A310: Based on the application scenario of the pet kennel and the type of pet used, preset the types of pollutant components and concentration distribution thresholds.

[0063] A320: Perform purification tests and record the air purification linkage module according to the pollutant composition type and concentration distribution threshold to obtain the corresponding air purification parameters and air purification effect data.

[0064] A330: Based on the air purification effect data, the corresponding air purification parameters are optimized to obtain usable air purification parameters.

[0065] A340: Based on the pollutant component type and concentration distribution threshold, a correlation and classification strategy is fitted with the available air purification parameters to establish the classification purification parameter-pollutant concentration strategy space.

[0066] Specifically, the first step is to preset the types of pollutants and their concentration thresholds based on the application scenario and the type of pet using the pet bed. In particular, different pets' physiological activities and living environments generate different types of pollutants. For example, the main pollutant in a cat's pet bed is ammonia VOCs produced by cat litter fermentation; while in a small dog's pet bed, the main pollutants are alkane VOCs related to saliva secretions and dander. Furthermore, the scenario also affects the concentration threshold setting. In bedrooms, where people spend a lot of time, the ammonia VOC concentration threshold needs to be set lower at 0.5-0.8 ppm, while in balconies with better ventilation, the threshold can be relaxed to 0.8-1.2 ppm. This dual consideration of scenario and pet type ensures that the preset pollutant information accurately matches actual needs, providing a clear target for subsequent testing.

[0067] Next, according to the preset pollutant composition types and concentration distribution thresholds, the air purification linkage module was tested and the results were recorded. The testing process had to revolve around the preset pollutant types and concentrations. For example, for a scenario in a bedroom cat bed with an ammonia VOC concentration of 0.8 ppm, the core parameters of the air purification linkage module were gradually adjusted, including the fan speed at low speed (3 m / s), medium speed (5 m / s), and high speed (8 m / s); the filter combination was a single activated carbon filter and a HEPA + activated carbon composite filter. The purification effect under each parameter combination was recorded: for example, low fan speed + single activated carbon required 30 minutes to reduce the ammonia VOC concentration to 0.4 ppm, medium fan speed + composite filter only required 15 minutes to reduce it to 0.2 ppm, and high fan speed + composite filter reduced it to 0.1 ppm in 10 minutes, but the noise exceeded the standard, exceeding 45 decibels, which did not meet the quiet requirements of a bedroom scenario. Through multiple tests, the different parameters were matched with the corresponding purification efficiency, noise, energy consumption and other effect data and recorded one by one to obtain the corresponding air purification parameters and air purification effect data.

[0068] Subsequently, the corresponding air purification parameters were optimized based on the air purification effect data. The optimization process required setting screening criteria based on the core needs of the pet bed scenario, with the principle of achieving the required purification efficiency and adapting to the characteristics of the scenario. For example, for the bedroom cat bed scenario, high-speed parameters that cause excessive noise were excluded. Low and medium speeds were compared: although medium speed consumes more energy than low speed, the purification efficiency is improved more significantly, and the final concentration is lower, such as 0.2ppm < 0.4ppm in the example above, which better meets the high air quality requirements of the bedroom. Therefore, medium speed + HEPA + activated carbon composite filter was determined as the usable air purification parameter for the ammonia VOC concentration of 0.8ppm in the bedroom cat bed scenario. For the balcony cat bed scenario, which has no noise restrictions, high speed + composite filter can be used as the usable air purification parameter for high concentrations of ammonia VOC. Through this targeted screening, it is ensured that the usable air purification parameters meet both the purification effect and the scenario restrictions.

[0069] Finally, based on the preset pollutant component types and concentration distribution thresholds, a hierarchical correlation strategy is fitted with the screened available air purification parameters. During the correlation process, a corresponding relationship needs to be constructed according to the logic of pollutant component type → concentration range → available air purification parameter. For example, for ammonia VOCs, the concentration is divided into three ranges: 0.5-0.8 ppm for low concentration, 0.8-1.2 ppm for medium concentration, and above 1.2 ppm for high concentration. These ranges are then associated with available air purification parameters for low wind speed + single activated carbon, medium wind speed + composite filter, and high wind speed + composite filter, respectively. For alkane VOCs, the correlation between concentration range and parameters is adjusted based on their easy diffusion characteristics. For example, 0.6-1.0 ppm corresponds to medium wind speed + composite filter, and above 1.0 ppm corresponds to high wind speed + composite filter. This ultimately forms a hierarchical purification parameter-pollutant concentration strategy space covering different pet types, different scenarios, and different pollution levels, allowing subsequent purification operations to directly match the corresponding parameters according to the actual pollution situation.

[0070] By pre-setting pollutant information according to pet dwelling scenarios and pet types, conducting targeted testing of purification modules, optimizing and fitting appropriate parameters, and establishing a hierarchical purification parameter-pollutant concentration strategy space that fits the actual needs of pet dwellings, the system provides a systematic basis for subsequent precise matching of purification strategies.

[0071] Furthermore, step A300 in the method provided in this application embodiment includes:

[0072] A350: The predicted pollutant concentration in the pet kennel is matched with the treatment strategy using the graded purification parameter-pollutant concentration strategy space to determine the matching air purification strategy threshold.

[0073] A360: Construct an air purification fitness function based on the air purification goals for pet beds.

[0074] A370: Based on the air purification fitness function, perform global optimization on the threshold of the matching air purification strategy to determine the target air purification strategy parameters.

[0075] Specifically, firstly, the graded purification parameter-pollutant concentration strategy space established in the aforementioned steps is used to match treatment strategies for the predicted pollutant concentration in the pet's kennel. During execution, the core information of the predicted pollutants in the pet's kennel is first clarified, including the pollutant component type (such as ammonia VOCs, alkane VOCs) and the predicted concentration value, such as the predicted concentration change of ammonia VOCs in the next 30 minutes. Then, this information is input into the graded purification parameter-pollutant concentration strategy space. Through the preset correspondence between pollutant component type, concentration range, and available air purification parameters within the strategy space, suitable matching air purification strategy thresholds are selected. For example, if the matching air purification strategy threshold for the ammonia VOC range of 0.8-1.2 ppm in the strategy space is an air velocity of 4-6 m / s, a HEPA+activated carbon composite filter, and a running time ≥20 minutes, then this set of matching air purification strategy thresholds is directly matched. This matching method, combined with predicted concentration, can adapt to the dynamic range of pollution in advance, avoiding purification interruptions caused by frequent parameter adjustments.

[0076] Next, an air purification fitness function is constructed based on the air purification goals of the pet bed. First, the core purification goals of the pet bed are defined, which need to be determined by those skilled in the art in conjunction with the application scenario and the pet's needs. For example, for a pet bed in a bedroom where a cat is kept, the purification goals can be set as VOC removal rate ≥90%, operating noise ≤40 decibels, and energy consumption ≤45W per unit time. Then, these core purification goals are quantified into calculable indicators, and each indicator is assigned a corresponding weight, such as a removal rate weight of 0.4, noise weight of 0.3, and energy consumption weight of 0.3. These are then integrated into an air purification fitness function, for example, fitness value = (removal rate / target removal rate) × 0.4 + (1 - noise / target noise limit) × 0.3 + (1 - energy consumption / target energy consumption limit) × 0.3. This air purification fitness function converts the purification effects of different dimensions into a single fitness value. Subsequently, only the fitness value needs to be compared to determine whether the parameters meet the comprehensive purification goals of the pet bed, satisfying multi-dimensional needs.

[0077] Finally, the threshold of the matching air purification strategy is globally optimized based on the air purification fitness function. First, the threshold of the matching air purification strategy is divided into multiple purification strategy intervals, and parameters are randomly selected in each interval to obtain multiple fitness sets of strategy intervals. After comparing and determining the first purification strategy interval where the parameter with the largest fitness is located, the interval is iteratively approximated and fitness evaluated until the preset termination condition is met, and then the target air purification strategy parameters are determined. The specific steps are explained in detail in A371-A373.

[0078] By utilizing hierarchical strategy space matching to predict pollutant concentrations in pet dens to determine precise strategy thresholds, and combining this with personalized purification goals for pet dens to construct a multi-dimensional air purification fitness function, the system achieves the effect of providing precise matching directions and scientific evaluation standards for the global optimization of subsequent target air purification strategy parameters.

[0079] Furthermore, step A370 in the method provided in this application embodiment includes:

[0080] A371: Divide the matching air purification strategy threshold into multiple purification strategy intervals, and perform random parameter selection and fitness evaluation within the multiple purification strategy intervals based on the air purification fitness function to obtain multiple strategy interval fitness sets.

[0081] A372: Compare the fitness sets of the multiple strategy intervals to determine the first purification strategy interval where the parameter with the highest fitness is located.

[0082] A373: Iterate the interval approximation and fitness evaluation of the first purification strategy interval until the preset termination condition is met, and determine the target air purification strategy parameters.

[0083] In one embodiment, the matching air purification strategy threshold is first divided into multiple purification strategy intervals. The division is based on the gradient of the change of each parameter within the matching air purification strategy threshold and the sensitivity of the pet's den to purification needs. For example, if the matching air purification strategy threshold is a wind speed of 4-6 m / s, a HEPA + activated carbon composite filter, and a running time of 20-40 minutes, where wind speed is the most sensitive to the purification efficiency and noise, the wind speed can be divided into four intervals of 4-4.5 m / s, 4.5-5 m / s, 5-5.5 m / s, and 5.5-6 m / s in a 0.5 m / s gradient, and the running time can be divided into four intervals of 20-25 minutes, 25-30 minutes, 30-35 minutes, and 35-40 minutes in a 5-minute gradient. The filter type is fixed as a composite filter, forming multiple purification strategy intervals that include wind speed intervals and time intervals.

[0084] After the above steps, based on the air purification fitness function, 3-5 sets of specific parameters are randomly selected in each purification strategy interval for fitness evaluation. For example, in the purification strategy interval of 4.5-5 m / s wind speed + 25-30 minutes, three sets of parameters (4.6 m / s, 26 minutes), (4.8 m / s, 28 minutes), and (4.9 m / s, 29 minutes) are randomly selected and substituted into the air purification fitness function. Taking into account the comprehensive calculation of removal rate, noise, and energy consumption, the fitness value of each set of parameters is obtained. Finally, each purification strategy interval forms a fitness set containing 3-5 fitness values.

[0085] After evaluating the fitness of each purification strategy interval, the next step is to compare the fitness sets of multiple strategy intervals. The comparison process should focus on the maximum value and overall distribution within each set. For example, the maximum fitness set for the purification strategy interval with a wind speed of 4.5-5 m / s and a duration of 25-30 minutes is 0.92, and all three fitness values ​​in this set are above 0.85, significantly higher than the fitness values ​​of other purification strategy intervals. This indicates that the parameters within this purification strategy interval are more likely to meet the purification goals of the pet's den. Therefore, the 4.5-5 m / s wind speed + 25-30 minutes duration is determined as the first purification strategy interval. This method of screening by interval avoids misjudgment caused by the randomness of a single parameter, ensuring that subsequent optimization focuses on parameter ranges with greater potential, ultimately determining the first purification strategy interval containing the parameter with the highest fitness.

[0086] Subsequently, the first purification strategy interval is iteratively approximated and its fitness is evaluated until the preset termination condition is met. The iterative approximation adopts a step-by-step subdivision of the interval. For example, the first purification strategy interval of 4.5-5 m / s wind speed + 25-30 minutes duration is further subdivided: the wind speed is divided into five sub-intervals of 4.5-4.6 m / s, 4.6-4.7 m / s, 4.7-4.8 m / s, 4.8-4.9 m / s, and 4.9-5 m / s according to a 0.1 m / s gradient, and the duration is divided into five sub-intervals of 25-26 minutes, 26-27 minutes, 27-28 minutes, 28-29 minutes, and 29-30 minutes according to a 1 minute gradient, forming a more refined sub-purification strategy interval.

[0087] Next, within each sub-purification strategy interval, 2-3 sets of parameters are randomly selected again for fitness evaluation. After comparison, the sub-purification strategy interval containing the parameter with the highest fitness is determined. This process of subdivision, evaluation, and selection is repeated until a preset termination condition is met. The preset termination condition can be set as follows: after subdividing the purification strategy interval, the difference in the maximum fitness value between two adjacent iterations is less than 0.01, or the purification strategy interval range is narrowed to a preset precision, such as wind speed interval ≤ 0.1 m / s or duration interval ≤ 1 minute. When the iteration terminates, the parameter with the highest fitness within the current purification strategy interval is selected as the target air purification strategy parameter.

[0088] By dividing the matching strategy threshold into intervals and randomly evaluating fitness, comparing and determining the first purification strategy interval, and iteratively approximating the first interval until the termination condition, the optimal air purification strategy parameters that meet the purification goals of pet beds are determined, thus balancing optimization efficiency and accuracy, avoiding local optimum traps.

[0089] Furthermore, step A400 in the method provided in this application embodiment includes:

[0090] A410: Based on the target pet den, perform 3D modeling and airflow simulation to obtain a set of simulated diffusion areas for polluted gases.

[0091] A420: Arrange the set of simulated diffusion regions of the pollutant gas according to the concentration information to obtain the set of diffusion sequence regions of the pollutant gas.

[0092] A430: Perform regional polling planning analysis on the set of regions of the polluted gas diffusion sequence to determine the air purification path.

[0093] Optionally, a 3D model and airflow simulation are first performed based on the target pet bed. The modeling process needs to reproduce the actual structural features of the pet bed, including the length, width, and height of the bed, whether it has side rails or top cover, the material of the internal padding (e.g., cotton padding will affect gas permeability), and the environmental layout around the pet bed, such as whether it is close to walls, furniture, or other airflow obstructions. A 3D model consistent with reality is constructed through 3D scanning or parametric modeling technology.

[0094] Subsequently, based on the source location of pollutants, including VOCs, in the pet's den, which is usually the core area where the pet rests, those skilled in the art use fluid dynamics simulation tools to simulate the diffusion process of pollutants in and around the den. For example, simulating the diffusion of VOCs from the core area of ​​the den to the surrounding areas after the pet excretes, the VOCs are mainly diffused upwards and to the unobstructed front due to the obstruction of the side rails, forming different diffusion areas such as the core diffusion area inside the den, the diffusion area above the top, and the diffusion area on the outer perimeter of the front. Finally, these diffusion areas are integrated to obtain a set of simulated diffusion areas of pollutants, providing a precise basis for the spatial distribution of pollution for subsequent path planning.

[0095] Next, the simulated diffusion areas of polluted gas are arranged according to concentration information. Due to differences in the distance from the pollution source and the degree of influence from airflow, the pollutant concentrations vary significantly among different diffusion areas. For example, the core diffusion area within the confined space directly surrounds the pollution source and has the highest concentration; the diffusion area above the top has the next highest concentration due to gas rising and accumulating; and the outer diffusion area on the front is farther away and may be diluted by environmental ventilation, resulting in the lowest concentration. Arranging the diffusion areas sequentially from highest to lowest concentration forms a polluted gas diffusion sequence: core diffusion area within the confined space → diffusion area above the top → outer diffusion area on the front. This arrangement ensures that subsequent purification paths prioritize covering the most heavily polluted areas, avoiding the problem of delayed purification in high-concentration areas.

[0096] Next, a regional polling planning analysis is performed on the set of polluted gas diffusion sequence areas to determine the air purification path. The polling planning needs to consider the spatial relationship of each diffusion area and the airflow coverage logic to ensure that the purified airflow can efficiently cover each area without significant dead zones. For example, the core diffusion area within the diffused area is initially targeted, and the outlet angle of the purification equipment is adjusted to align with the core area, allowing the airflow to directly act on the high-concentration area. After the core area has been purified for a certain period, the flow shifts to the top diffusion area based on the sequence. Since the top area is directly above the core area, the outlet angle can be adjusted upwards by 15-30°, allowing the airflow to extend upwards along the core area to cover the top area, while ensuring a 5-10cm overlap between the airflow and the core area to avoid omissions at the boundary. Finally, the flow shifts to the frontal outer diffusion area, adjusting the outlet angle forward to allow the airflow to extend from the core area to the front, covering the outer low-concentration area. Throughout the polling process, the airflow simulation results can be combined to avoid structural obstruction areas, such as side rails, ensuring that the airflow can reach the target area smoothly, ultimately forming an air purification path that goes from high concentration to low concentration, covers all diffusion areas, and has no obvious dead corners.

[0097] By using 3D modeling and airflow simulation of the target pet's den to obtain the pollution diffusion area, sorting the areas by concentration to form a sequence of regions, and determining the path through polling planning analysis, the air purification path is precisely adapted to the local spatial characteristics and pollution diffusion patterns of the pet's den, thereby improving the integrity and efficiency of purification coverage.

[0098] In summary, the air treatment method with pet bed linkage purification function provided in this application embodiment has the following technical effects:

[0099] This application utilizes a pet-nest sensing module integrating temperature and humidity sensors, VOC gas sensors, and infrared thermal imaging modules to collect multi-dimensional sensing data streams from the pet-nest. Through multi-dimensional data preprocessing, multi-dimensional feature extraction, multi-modal fusion, and LSTM network prediction, the predicted pollutant concentration in the pet-nest is obtained. An air purification linkage module is tested to establish a hierarchical purification parameter-pollutant concentration strategy space. The target air purification strategy parameters are determined by optimizing the air purification fitness function. Simultaneously, air purification paths are determined through 3D modeling of the pet-nest, airflow simulation, and regional polling planning. Based on the target air purification strategy parameters and the air purification path, dynamic air purification is performed on the target pet-nest. This achieves accurate pollutant identification and on-demand purification in the pet-nest scenario, making the air treatment in the pet-nest more adaptable to its environmental characteristics and dynamic pollution changes, improving the efficiency and accuracy of air purification. The technical effect of accurately identifying pollutants, matching purification strategies on demand, and achieving optimal path coverage in the pet-nest scenario balances purification effect and energy consumption.

[0100] Example 2, as Figure 2As shown, based on the same inventive concept as in Embodiment 1 above, this application provides an air handling system with pet bed linkage purification function, the system comprising:

[0101] A multi-dimensional sensing data stream acquisition module 1 for pet beds is used to deploy a pet bed sensing module on a target pet bed. The pet bed sensing module integrates a temperature and humidity sensor, a VOC gas sensor, and an infrared thermal imaging module, and collects multi-dimensional sensing data streams of the pet bed through the pet bed sensing module.

[0102] The pet bed pollutant concentration prediction acquisition module 2 is used to perform fusion analysis and pollutant concentration prediction on the multi-dimensional sensing data stream of the pet bed to obtain the predicted pollutant concentration of the pet bed.

[0103] The target air purification strategy parameter acquisition module 3 is used to acquire the air purification linkage module, test and record the air purification linkage module, establish a graded purification parameter-pollutant concentration strategy space, and use the graded purification parameter-pollutant concentration strategy space to analyze the treatment strategy of the predicted pollutant concentration in the pet bed to determine the target air purification strategy parameters.

[0104] The dynamic air purification process execution module 4 is used to plan and determine the air purification path of the target pet bed, and to perform dynamic air purification process on the target pet bed based on the target air purification strategy parameters and the air purification path.

[0105] Furthermore, the pet kennel pollutant concentration prediction module 2 is used to perform the following steps:

[0106] Based on the data characteristics of the multidimensional sensing data stream of the pet bed, a multidimensional data preprocessing program is constructed, which includes noise filtering and outlier cleaning. The multidimensional sensing data stream of the pet bed is preprocessed using the multidimensional data preprocessing program to obtain a usable multidimensional sensing data stream of the pet bed. Features are extracted from each dimension of the usable multidimensional sensing data stream of the pet bed to obtain temperature and humidity feature sets, gas composition change feature sets, and thermal imaging feature sets. The temperature and humidity feature sets, gas composition change feature sets, and thermal imaging feature sets are then fused using multimodal methods and pollutant concentration prediction to obtain the predicted pollutant concentration of the pet bed.

[0107] Furthermore, the pet kennel pollutant concentration prediction module 2 is used to perform the following steps:

[0108] The temperature and humidity feature set, gas composition change feature set, and thermal imaging feature set are arranged and integrated according to time series information to obtain the temperature and humidity sequence feature set, the gas composition change sequence feature set, and the thermal imaging sequence feature set. Based on the temperature and humidity sequence feature set and the thermal imaging sequence feature set, multimodal influence fusion is performed on the gas composition change sequence feature set to obtain the gas composition change fused feature set. The LSTM network structure is used to identify pollution change trends and predict pollutant concentrations in the gas composition change fused feature set to obtain the predicted pollutant concentration of the pet kennel.

[0109] Furthermore, the pet kennel pollutant concentration prediction module 2 is used to perform the following steps:

[0110] According to the pet den sensing module, the historical multidimensional sensing dataset of the pet den is called; based on the historical multidimensional sensing dataset of the pet den, the gas composition influence analysis is performed on the temperature and humidity sequence feature set and the thermal imaging sequence feature set to determine the temperature and humidity gas composition influence coefficient and the thermal imaging gas composition influence coefficient; using the temperature and humidity gas composition influence coefficient and the thermal imaging gas composition influence coefficient, multimodal gain fusion is performed on the gas composition change sequence feature set to obtain the gas composition change fused feature set.

[0111] Furthermore, the target air purification strategy parameter acquisition module 3 is used to perform the following steps:

[0112] Based on the application scenario of the pet bed and the type of pet, preset pollutant component types and concentration distribution thresholds are set; the air purification linkage module is tested and recorded according to the pollutant component types and concentration distribution thresholds to obtain corresponding air purification parameters and air purification effect data; the corresponding air purification parameters are optimized according to the air purification effect data to obtain usable air purification parameters; based on the pollutant component types and concentration distribution thresholds, a correlation and hierarchical strategy is fitted with the usable air purification parameters to establish the hierarchical purification parameter-pollutant concentration strategy space.

[0113] Furthermore, the target air purification strategy parameter acquisition module 3 is used to perform the following steps:

[0114] The predicted pollutant concentration in the pet's den is matched with the treatment strategy using the graded purification parameter-pollutant concentration strategy space to determine the matching air purification strategy threshold; an air purification fitness function is constructed based on the air purification target of the pet's den; and the matching air purification strategy threshold is globally optimized based on the air purification fitness function to determine the target air purification strategy parameters.

[0115] Furthermore, the target air purification strategy parameter acquisition module 3 is used to perform the following steps:

[0116] The matching air purification strategy threshold is divided into multiple purification strategy intervals. Based on the air purification fitness function, random parameters are selected and fitness is evaluated in each of the multiple purification strategy intervals to obtain multiple strategy interval fitness sets. The multiple strategy interval fitness sets are compared to determine the first purification strategy interval where the parameter with the highest fitness is located. The first purification strategy interval is iteratively approximated and evaluated until a preset termination condition is met to determine the target air purification strategy parameters.

[0117] Furthermore, the dynamic air purification process execution module 4 is used to perform the following steps:

[0118] Based on the target pet den, a 3D model and airflow simulation are performed to obtain a set of simulated diffusion areas for polluted gases. The set of simulated diffusion areas for polluted gases is then arranged according to concentration information to obtain a set of diffusion sequence areas for polluted gases. Regional polling planning analysis is then performed on the set of diffusion sequence areas for polluted gases to determine the air purification path.

[0119] Example 3, as Figure 3 As shown, based on the same inventive concept as in Embodiment 1 above, this application provides an electronic device, the electronic device comprising:

[0120] The memory 303 is used to store executable instructions; the processor 302 is used to execute the executable instructions stored in the memory 303 to implement an air treatment method with a pet bed linkage purification function.

[0121] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present invention. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention. This electronic device is in the form of a general-purpose computing device, and its components may include, but are not limited to, an input device 301, a processor 302, a memory 303, and an output device 304. The processor 302 may be one or more; the memory 303 may include a computer-readable medium and at least one program product having a set (at least one) of program modules configured to perform the functions of the embodiments of this application.

[0122] The memory 303 shown in this embodiment of the invention can be any combination of one or more computer-readable media. The computer-readable storage medium can be, but is not limited to, an infrared, semiconductor system, device or instrument, or any combination thereof, used to store software programs, computer-executable programs and modules, such as the program instructions / modules corresponding to the air treatment method with pet bed linkage purification function in this embodiment of the invention. The processor 302 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 303, thereby realizing the above-mentioned air treatment method with pet bed linkage purification function.

[0123] The air treatment system with pet bed linkage purification function provided in the embodiments of the present invention can execute the air treatment method with pet bed linkage purification function provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0124] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0125] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. An air treatment method with pet house linkage purification function, characterized in that, The method includes: A pet bed sensing module is deployed on the target pet bed. The pet bed sensing module integrates a temperature and humidity sensor, a VOC gas sensor and an infrared thermal imaging module. Multi-dimensional sensing data streams of the pet bed are collected through the pet bed sensing module. The multidimensional sensing data stream of the pet bed is fused and analyzed, and pollutant concentration is predicted to obtain the predicted pollutant concentration of the pet bed. The air purification linkage module is obtained, the test records of the air purification linkage module are made, a graded purification parameter-pollutant concentration strategy space is established, and the treatment strategy analysis of the predicted pollutant concentration of the pet bed is performed using the graded purification parameter-pollutant concentration strategy space to determine the target air purification strategy parameters. The air purification path of the target pet bed is planned and determined, and the target pet bed is dynamically purified based on the target air purification strategy parameters and the air purification path.

2. The air treatment method with pet bed linkage purification function as described in claim 1, characterized in that, The method for obtaining the predicted pollutant concentration in the pet's kennel includes: Based on the data characteristics of the multidimensional sensing data stream from the pet den, a multidimensional data preprocessing program is constructed, which includes noise filtering and outlier cleaning. The multidimensional data preprocessing program is used to preprocess the multidimensional sensing data stream of the pet bed to obtain a usable multidimensional sensing data stream of the pet bed. Feature extraction is performed on each dimension of the multidimensional sensing data stream of the available pet nest to obtain temperature and humidity feature sets, gas composition change feature sets, and thermal imaging feature sets. The temperature and humidity feature set, gas composition change feature set, and thermal imaging feature set are fused in a multimodal manner and pollutant concentration is predicted to obtain the predicted pollutant concentration in the pet kennel.

3. The air treatment method with pet bed linkage purification function as described in claim 2, characterized in that, The method of obtaining the predicted contaminant concentration in the pet's kennel includes: The temperature and humidity feature set, gas composition change feature set, and thermal imaging feature set are arranged and integrated according to time sequence information to obtain the temperature and humidity sequence feature set, gas composition change sequence feature set, and thermal imaging sequence feature set. Based on the temperature and humidity sequence feature set and the thermal imaging sequence feature set, multimodal influence fusion is performed on the gas composition change sequence feature set to obtain a gas composition change fused feature set; The LSTM network structure is used to identify pollution change trends and predict pollutant concentrations by fusing the gas composition change feature set, thereby obtaining the predicted pollutant concentration of the pet kennel.

4. The air treatment method with pet bed linkage purification function as described in claim 3, characterized in that, The obtained gas composition change fusion feature set includes: Based on the pet den sensing module, call the historical multidimensional sensing dataset of the pet den; Based on the historical multidimensional sensing dataset of the pet den, gas composition influence analysis was performed on the temperature and humidity sequence feature set and the thermal imaging sequence feature set to determine the gas composition influence coefficient of temperature and humidity and the gas composition influence coefficient of thermal imaging. Using the temperature and humidity gas composition influence coefficient and the thermal imaging gas composition influence coefficient, multimodal gain fusion is performed on the gas composition change sequence feature set to obtain the gas composition change fused feature set.

5. The air treatment method with pet bed linkage purification function as described in claim 1, characterized in that, The establishment of the hierarchical purification parameter-pollutant concentration strategy space includes: Based on the application scenario of the pet bed and the type of pet used, preset the types of pollutant components and concentration distribution thresholds; The air purification linkage module is tested and recorded according to the pollutant composition type and concentration distribution threshold to obtain the corresponding air purification parameters and air purification effect data. Based on the air purification effect data, the corresponding air purification parameters are optimized to obtain usable air purification parameters; Based on the pollutant component type and concentration distribution threshold, a hierarchical strategy is fitted with the available air purification parameters to establish the hierarchical purification parameter-pollutant concentration strategy space.

6. The air treatment method with pet bed linkage purification function as described in claim 1, characterized in that, The parameters for determining the target air purification strategy include: The predicted pollutant concentration in the pet's nest is matched with the treatment strategy using the graded purification parameter-pollutant concentration strategy space to determine the matching air purification strategy threshold. Based on the air purification goals for pet beds, an air purification fitness function is constructed. The target air purification strategy parameters are determined by globally optimizing the matching air purification strategy threshold based on the air purification fitness function.

7. The air treatment method with pet bed linkage purification function as described in claim 6, characterized in that, The step of globally optimizing the threshold of the matching air purification strategy based on the air purification fitness function to determine the target air purification strategy parameters includes: The matching air purification strategy threshold is divided into multiple purification strategy intervals. Based on the air purification fitness function, random parameters are selected and fitness is evaluated in each of the multiple purification strategy intervals to obtain multiple strategy interval fitness sets. By comparing the fitness sets of the multiple strategy intervals, the first purification strategy interval containing the parameter with the highest fitness is determined; The first purification strategy range is iteratively approximated and its fitness is evaluated until a preset termination condition is met, thereby determining the target air purification strategy parameters.

8. The air treatment method with pet bed linkage purification function as described in claim 1, characterized in that, The plan determines the air purification path for the target pet's den, including: Based on the target pet den, a 3D model and airflow simulation were performed to obtain a set of simulated diffusion areas for polluted gases. The set of simulated diffusion regions of pollutant gas is arranged according to concentration information to obtain a set of diffusion sequence regions of pollutant gas. Regional polling planning analysis is performed on the set of regions containing the polluted gas diffusion sequence to determine the air purification path.

9. An air handling system with integrated pet bed purification function, characterized in that, The air treatment method with pet bed linkage purification function as described in any one of claims 1-8, the system comprising: A multi-dimensional sensing data stream acquisition module for pet beds is used to deploy a pet bed sensing module on a target pet bed. The pet bed sensing module integrates a temperature and humidity sensor, a VOC gas sensor, and an infrared thermal imaging module, and collects multi-dimensional sensing data streams of the pet bed through the pet bed sensing module. The pet den predicts pollutant concentration acquisition module is used to perform fusion analysis and pollutant concentration prediction on the multi-dimensional sensing data stream of the pet den to obtain the predicted pollutant concentration of the pet den. The target air purification strategy parameter acquisition module is used to acquire the air purification linkage module, test and record the air purification linkage module, establish a graded purification parameter-pollutant concentration strategy space, use the graded purification parameter-pollutant concentration strategy space to analyze the treatment strategy of the predicted pollutant concentration in the pet bed, and determine the target air purification strategy parameters. The dynamic air purification process execution module is used to plan and determine the air purification path of the target pet bed, and to perform dynamic air purification process on the target pet bed based on the target air purification strategy parameters and the air purification path.

10. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the air treatment method with pet bed linkage purification function as described in any one of claims 1 to 8.

Images