Air conditioner self-cleaning method and device, air conditioner and storage medium

Abstract

The invention relates to an air conditioner self-cleaning method and device, an air conditioner and a storage medium, and the air conditioner self-cleaning method comprises the steps that a composite coating image of a composite coating on the surface of a to-be-cleaned assembly in the air conditioner is obtained; inputting the composite coating image into a preset coating state detection model to obtain coating state information; according to the coating state information, whether the smudginess degree of the to-be-cleaned assembly meets the self-cleaning condition of the air conditioner or not is determined; and if the smudginess degree of the to-be-cleaned assembly meets the self-cleaning condition of the air conditioner, a composite coating on the surface of the to-be-cleaned assembly is subjected to self-cleaning. The self-cleaning opportunity of the air conditioner can be automatically judged according to the coating state information of the composite coating, the composite coating on the surface of the to-be-cleaned assembly is automatically self-cleaned, subjective judgment of a user is avoided, the user can conveniently use the air conditioner, and the situation that excessive dirt is accumulated in the air conditioner and influences the health of the user is avoided.

Description

Technical Field

[0001] This application relates to the field of air conditioning, and more particularly to a self-cleaning method, apparatus, air conditioner, and storage medium for air conditioning. Background Technology

[0002] Air conditioner UV self-cleaning is a technology that uses the bactericidal and disinfecting properties of ultraviolet (UV) light to clean the components inside the air conditioner that are prone to microbial growth. Its core is to destroy the DNA or RNA structure of microorganisms such as bacteria, viruses, and mold through the radiation of ultraviolet light, inhibit their reproduction or kill them directly, thereby reducing the accumulation of pollutants inside the air conditioner and ensuring hygienic air supply.

[0003] However, when using ultraviolet light for air conditioner self-cleaning, it is difficult for users to determine when self-cleaning is needed. Failure to perform self-cleaning in a timely manner will lead to the accumulation of pollutants inside the air conditioner, and the pollutants carried by the air conditioner during airflow will affect the user's health. Summary of the Invention

[0004] In order to solve the above-mentioned technical problems, or at least partially solve the above-mentioned technical problems, this application provides an air conditioner self-cleaning method, apparatus, air conditioner and storage medium.

[0005] In a first aspect, this application provides a self-cleaning method for an air conditioner, the method comprising:

[0006] Obtain composite coating images of the composite coating on the surface of the components to be cleaned in an air conditioner;

[0007] The composite coating image is input into a preset coating state detection model to obtain coating state information;

[0008] Based on the coating status information, determine whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner;

[0009] If the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner, the composite coating on the surface of the component to be cleaned will be self-cleaned.

[0010] Optionally, the preset coating state detection model includes: a shared encoder and a dirt detection branch decoder;

[0011] The shared encoder is used to extract local dirt features from the composite coating image;

[0012] The dirt detection branch decoder is used to determine the dirt coverage area and dirt area information of the component to be cleaned based on the local dirt characteristics, so as to serve as the coating status information.

[0013] Optionally, determining whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner based on the coating status information includes:

[0014] Determine whether the dirty area information in the coating status information is greater than a preset dirty area threshold;

[0015] If the dirty area information is greater than the preset dirty area threshold, it is determined that the degree of dirtiness of the component to be cleaned meets the self-cleaning conditions of the air conditioner.

[0016] Optionally, the composite coating on the surface of the component to be cleaned is self-cleaned, including:

[0017] The irradiation path of the ultraviolet light spot is determined based on the dirt coverage area in the coating status information;

[0018] Adjust the angle of the ultraviolet light emitted by the ultraviolet irradiation component according to the irradiation path until the composite coating is self-cleaned.

[0019] Optionally, the self-cleaning of the composite coating on the surface of the component to be cleaned further includes:

[0020] Obtain data on the characteristic absorption peak intensity and microbial inactivation efficiency of pollutants on the surface of the composite coating;

[0021] The irradiation period length, irradiation intensity, and phase difference between different ultraviolet light sources in the ultraviolet irradiation component are determined based on the intensity of the characteristic absorption peak.

[0022] The combination of ultraviolet light wavelengths for the ultraviolet irradiation component is determined based on the microbial inactivation efficiency data.

[0023] The ultraviolet irradiation component emits ultraviolet light according to the combination of the irradiation period length, the irradiation intensity, the phase difference, and the ultraviolet light wavelength.

[0024] Optionally, the preset coating state detection model further includes: an aging detection branch decoder;

[0025] The shared encoder is also used to extract global aging features from the composite coating image;

[0026] The aging detection branch decoder is used to determine the aging degree information of the component to be cleaned based on the global aging characteristics, so as to serve as the coating state information.

[0027] Optionally, the method further includes:

[0028] Determine whether the aging degree information in the coating status information is greater than a preset aging degree threshold;

[0029] If the aging information exceeds a preset aging threshold, an aging warning for the composite coating will be issued.

[0030] Secondly, this application provides an air conditioner self-cleaning device, the device comprising:

[0031] The acquisition module is used to acquire composite coating images of the composite coating on the surface of the components to be cleaned in the air conditioner;

[0032] The input module is used to input the composite coating image into a preset coating state detection model to obtain coating state information;

[0033] The first determining module is used to determine whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner based on the coating status information.

[0034] The self-cleaning module is used to self-clean the composite coating on the surface of the component to be cleaned if the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner.

[0035] Optionally, the preset coating state detection model includes: a shared encoder and a dirt detection branch decoder;

[0036] The shared encoder is used to extract local dirt features from the composite coating image;

[0037] The dirt detection branch decoder is used to determine the dirt coverage area and dirt area information of the component to be cleaned based on the local dirt characteristics, so as to serve as the coating status information.

[0038] Optionally, determining whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner based on the coating status information includes:

[0039] Determine whether the dirty area information in the coating status information is greater than a preset dirty area threshold;

[0040] If the dirty area information is greater than the preset dirty area threshold, it is determined that the degree of dirtiness of the component to be cleaned meets the self-cleaning conditions of the air conditioner.

[0041] Optional self-cleaning module, including:

[0042] The first determining unit is used to determine the irradiation path of the ultraviolet light spot based on the dirt coverage area in the coating status information;

[0043] An adjustment unit is used to adjust the angle of ultraviolet light emitted by the ultraviolet irradiation component according to the irradiation path until the composite coating is self-cleaned.

[0044] Optional, the self-cleaning module also includes:

[0045] The acquisition unit is used to acquire the characteristic absorption peak intensity and microbial inactivation efficiency data of pollutants on the surface of the composite coating;

[0046] The second determining unit is used to determine the irradiation period length, irradiation intensity, and phase difference between different ultraviolet light sources in the ultraviolet irradiation component based on the intensity of the characteristic absorption peak.

[0047] The third determining unit is used to determine the ultraviolet light wavelength combination of the ultraviolet irradiation component based on the microbial inactivation efficiency data.

[0048] The control unit is used to control the ultraviolet irradiation component to emit ultraviolet light according to the combination of the irradiation period length, the irradiation intensity, the phase difference and the ultraviolet light wavelength.

[0049] Optionally, the preset coating state detection model further includes: an aging detection branch decoder;

[0050] The shared encoder is also used to extract global aging features from the composite coating image;

[0051] The aging detection branch decoder is used to determine the aging degree information of the component to be cleaned based on the global aging characteristics, so as to serve as the coating state information.

[0052] Optionally, the device further includes:

[0053] The second determining module is used to determine whether the aging degree information in the coating state information is greater than a preset aging degree threshold.

[0054] The reminder module is used to issue a composite coating aging reminder if the aging information is greater than a preset aging threshold.

[0055] Thirdly, this application provides an air conditioner, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0056] Memory, used to store computer programs;

[0057] When the processor executes a program stored in the memory, it implements the air conditioner self-cleaning method described in any of the first aspects.

[0058] Fourthly, this application provides a computer-readable storage medium storing a program for an air conditioner self-cleaning method, wherein when the program for the air conditioner self-cleaning method is executed by a processor, it implements the steps of the air conditioner self-cleaning method described in any of the first aspects.

[0059] The technical solutions provided in this application have the following advantages compared with the prior art:

[0060] This application embodiment uses a preset coating state detection model to determine the coating state information of the composite coating based on the composite coating image. If the degree of dirt on the component to be cleaned is determined to meet the self-cleaning conditions of the air conditioner based on the coating state information, then the composite coating on the surface of the component to be cleaned will be self-cleaned. This realizes the automatic determination of the timing of air conditioner self-cleaning based on the coating state information of the composite coating, and the automatic self-cleaning of the composite coating on the surface of the component to be cleaned. This avoids subjective judgment by the user, facilitates user use, and prevents excessive dirt from accumulating inside the air conditioner, which could affect the user's health. Attached Figure Description

[0061] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0062] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0063] Figure 1 A flowchart of an air conditioner self-cleaning method provided in this application embodiment;

[0064] Figure 2 A flowchart of an air conditioner self-cleaning method provided in this application embodiment;

[0065] Figure 3 A structural diagram of an air conditioner self-cleaning device provided in an embodiment of this application;

[0066] Figure 4 This is a schematic diagram of the structure of an air conditioner provided in an embodiment of this application. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, 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 some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0068] Currently, when using ultraviolet light for air conditioner self-cleaning, users often find it difficult to determine when self-cleaning is needed. Failure to clean the air conditioner promptly leads to the accumulation of pollutants inside, which, when carried by the air conditioner's airflow, can negatively impact user health. Therefore, this application provides an air conditioner self-cleaning method, apparatus, air conditioner, and storage medium. By combining the spectral characteristics of photocatalytic materials, the light field distribution is adaptively adjusted to achieve comprehensive dynamic sterilization. Furthermore, this application employs deep learning technology to analyze the material wear and the distribution of bacteria and dust in real time using the spectral characteristics of the photocatalyst and filter, promptly reminding users to activate the self-cleaning mode and replace materials.

[0069] This application provides an air conditioner self-cleaning system, which includes: a spectral imaging device (such as a hyperspectral imaging, infrared imaging, terahertz imaging, etc.) and an ultraviolet irradiation component. The ultraviolet irradiation component may include multiple micro ultraviolet light-emitting diode (UV-LED) arrays arranged in a biomimetic compound eye pattern, and each micro ultraviolet light-emitting diode array is equipped with an independently driven pulse width modulation circuit.

[0070] This application provides an air conditioner self-cleaning method, such as... Figure 1 As shown, the method includes:

[0071] Step S101: Obtain an image of the composite coating on the surface of the component to be cleaned in the air conditioner.

[0072] In this embodiment of the application, the components to be cleaned include: heat exchange components, airflow driving components, airflow guiding components, and auxiliary contact components, etc. For example, the heat exchange components include evaporators (indoor side), condensers (outdoor side), etc., the air filtration components include various types of filters (pre-filter, high-efficiency filter, activated carbon, etc.), etc., the airflow driving components include cross-flow fans (fan wheel), axial flow fans (outdoor unit), etc., the airflow guiding components include air guide plates, air outlet grilles, etc., and the auxiliary contact components include water collection trays (places where condensate is easily left behind), etc.

[0073] The composite coating in this application refers to a photocatalytic material coating, which can be excited by light (usually ultraviolet or visible light) to produce a photocatalytic reaction. It has the characteristics of decomposing organic matter and sterilizing and disinfecting, and has wide applications in air purification, self-cleaning and other fields. Specifically, it can refer to a TiO2-graphene composite coating.

[0074] In practical applications, spectral imaging equipment (such as hyperspectral imaging, infrared imaging, terahertz (THz) imaging, etc.) can be used to acquire hyperspectral images, infrared images, and terahertz images of the composite coating surface as images of the composite coating.

[0075] Step S102: Input the composite coating image into a preset coating state detection model to obtain coating state information;

[0076] In this embodiment, a preset coating state detection model can be trained in advance. Specifically, composite coating images of components to be cleaned under different degrees of dirt are collected. To improve the accuracy of the model's output coating state information, thermal maps and pseudo-color images (such as HSV, Hot, and parula) of the composite coating of the components to be cleaned can also be collected based on the different spectral absorption characteristics of different substances. The collected composite coating images, thermal maps, and pseudo-color images are labeled and used as input to the deep learning model. The preset coating state detection model can combine composite coating images, thermal maps, and pseudo-color images to analyze the distribution of bacteria and dust, the degree of material aging, etc., to obtain coating state information.

[0077] In one embodiment of this application, the preset coating state detection model includes: a shared encoder and a dirt detection branch decoder;

[0078] The shared encoder is used to extract local dirt features from the composite coating image;

[0079] The dirt detection branch decoder is used to determine the dirt coverage area and dirt area information of the component to be cleaned based on the local dirt characteristics, so as to serve as the coating status information.

[0080] In this step, the composite coating image can be input into a preset coating state detection model to obtain coating state information. The coating state information includes at least: dirt area information, and in some embodiments, it may also include dirt coverage area information and aging degree information.

[0081] Step S103: Determine whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner based on the coating status information;

[0082] In this step, the dirt level of the component to be cleaned can be determined based on the coating status information, such as the dirt area and dirt density. The numerical range corresponding to the dirt level is the degree of dirt. For example, if the dirt area is 60%-70%, the degree of dirt is 6; if the dirt area is 80%-90%, the degree of dirt is 9; if the dirt density is 100 colonies per square centimeter, the degree of dirt is 6; if the dirt density is 500 colonies per square centimeter, the degree of dirt is 8, and so on. The degree of dirt is then compared with a dirt degree threshold. For example, if the dirt degree threshold is 6, then if the degree of dirt is greater than 6, it is determined that the degree of dirt meets the self-cleaning conditions of the air conditioner.

[0083] To save computation, the coating status information can be directly compared with the corresponding threshold to determine whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner. In one embodiment of this application, it can be determined whether the dirt area information in the coating status information is greater than a preset dirt area threshold; if the dirt area information is greater than the preset dirt area threshold, it is determined that the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner.

[0084] Step S104: If the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner, perform self-cleaning on the composite coating on the surface of the component to be cleaned.

[0085] In this step, self-cleaning of the composite coating on the surface of the component to be cleaned can refer to activating the ultraviolet irradiation component to emit ultraviolet light in order to self-clean the composite coating on the surface of the component to be cleaned.

[0086] In one embodiment of this application, self-cleaning of the composite coating on the surface of the component to be cleaned includes: determining the irradiation path of the ultraviolet light spot based on the dirt-covered area in the coating status information; adjusting the angle of the ultraviolet light emitted by the ultraviolet irradiation component according to the irradiation path until the composite coating is self-cleaned.

[0087] When determining the irradiation path, the principle of uniform and effective irradiation of ultraviolet light to all areas that need to be cleaned can be followed to automatically generate the irradiation path. After the irradiation path is determined, the angle of ultraviolet light emitted by the ultraviolet irradiation component can be continuously adjusted along the irradiation path to enable the ultraviolet light to self-clean the composite coating.

[0088] In another embodiment of this application, during the process of adjusting the angle of ultraviolet light emitted by the ultraviolet irradiation component according to the irradiation path until the composite coating is self-cleaned, the characteristic absorption peak intensity and microbial inactivation efficiency data of the pollutants on the surface of the composite coating can be obtained; the irradiation cycle length, irradiation intensity, and phase difference between different ultraviolet light sources in the ultraviolet irradiation component are determined based on the characteristic absorption peak intensity; the ultraviolet light wavelength combination of the ultraviolet irradiation component is determined based on the microbial inactivation efficiency data; and the ultraviolet light emitted by the ultraviolet irradiation component is controlled according to the irradiation cycle length, the irradiation intensity, the phase difference, and the ultraviolet light wavelength combination.

[0089] Among them, the intensity of characteristic absorption peaks can be collected by multispectral imaging equipment, and the microbial inactivation efficiency data can be obtained by photodetector on the fin surface.

[0090] For example, the combination of ultraviolet light wavelengths may include 275nm (main band) and 310nm (auxiliary band), and the light field distribution and sterilization effect can be optimized by changing the relative proportion of different wavelengths.

[0091] This application utilizes the surface plasmon waveguide effect of the photocatalytic material coating to directionally transmit ultraviolet light along the fin surface to the physically shielded area. It also couples a reinforcement learning algorithm to analyze the light field distribution in real time and dynamically adjust the phase and power of the UV-LED array to achieve autonomous enhancement and path optimization of the photocatalytic reaction in the shaded area. This enables the simultaneous cleaning of dust and bacteria, and allows for autonomous cleaning of hidden corners, significantly improving the efficiency and intelligence of the air conditioner's self-cleaning function.

[0092] This application embodiment uses a preset coating state detection model to determine the coating state information of the composite coating based on the composite coating image. If the degree of dirt on the component to be cleaned is determined to meet the self-cleaning conditions of the air conditioner based on the coating state information, then the composite coating on the surface of the component to be cleaned will be self-cleaned. This realizes the automatic determination of the timing of air conditioner self-cleaning based on the coating state information of the composite coating, and the automatic self-cleaning of the composite coating on the surface of the component to be cleaned. This avoids subjective judgment by the user, facilitates user use, and prevents excessive dirt from accumulating inside the air conditioner, which could affect the user's health.

[0093] Based on the foregoing embodiments, in another embodiment of this application, the preset coating state detection model includes, in addition to the shared encoder and the dirt detection branch decoder, an aging detection branch decoder.

[0094] The shared encoder is also used to extract global aging features from the composite coating image;

[0095] The aging detection branch decoder is used to determine the aging degree information of the component to be cleaned based on the global aging characteristics, so as to serve as the coating state information.

[0096] Based on this, such as Figure 2 As shown, the air conditioner self-cleaning method further includes:

[0097] Step S201: Determine whether the aging degree information in the coating state information is greater than a preset aging degree threshold;

[0098] Step S202: If the aging degree information is greater than the preset aging degree threshold, a composite coating aging reminder is issued.

[0099] It can automatically analyze the aging degree of composite materials based on the coating condition detection model and the spectral characteristics of self-cleaning materials. If the aging degree of the composite coating exceeds the preset aging degree threshold, it will issue a composite coating aging reminder to prompt the user to update the composite coating. This eliminates the need to rely on experience when replacing composite materials and reduces the possibility that users will forget to replace the materials.

[0100] For ease of understanding, this application also provides an embodiment in practical application, as follows:

[0101] Step 1: Data Collection

[0102] This invention first collects spectral images of photocatalytic materials (a composite coating of TiO2-graphene), filters, and other materials. The main steps are as follows: First, install spectral imaging equipment (such as hyperspectral imaging, infrared imaging, terahertz imaging, etc.). Second, acquire the spectral images of the materials using the imaging equipment. Third, based on the different spectral absorption characteristics of different substances, analyze the distribution of bacterial dust and the degree of material aging by combining their thermograms, pseudo-color maps, etc. (such as HSV, Hot, and parula maps). Finally, label the acquired dataset as input to a deep learning model.

[0103] Step 2: Identification of material aging and dust / bacteria

[0104] This model is based on a multi-task learning framework and adopts a shared encoder-task branch decoder structure.

[0105] First, ResNet or EfficientNet networks are used as shared encoders to extract global aging features and local contamination details on the material surface. The core of this approach is to learn multi-scale features from the input image through convolutional layers or Transformer modules. For example, low-level networks capture texture details (such as bacterial spots and dust particles), while high-level networks extract global semantics (such as color gradations caused by aging and crack distribution).

[0106] ResNet is a deep neural network architecture that solves the vanishing gradient problem during deep network training by introducing residual blocks, thus enabling more efficient training of very deep networks. Input: Input image, such as a spectral image of a material surface. Output: Scale feature map, including texture details captured by lower-level networks (such as bacterial spots and dust particles) and global semantic information extracted by higher-level networks (such as color gradations caused by aging and crack distribution).

[0107] EfficientNet is an efficient and lightweight deep learning network architecture that balances model size, resolution, and depth through a compound scaling method, thereby reducing computational resource consumption while maintaining high performance. Input: Input image, such as a spectral image of a material surface. Output: Multi-scale feature maps, including texture details captured by lower-level networks and global semantic information extracted by higher-level networks.

[0108] The shared encoder can include either ResNet or EfficientNet, but not both. The choice depends on the specific application scenario and performance requirements: ResNet is chosen if a deeper network is needed to extract complex features. EfficientNet is chosen if more efficient computation and lower resource consumption are required.

[0109] Secondly, the output features of the shared encoder are then distributed to three task branches: 1. The aging assessment branch compresses the feature map into a one-dimensional vector using global average pooling, connects it to a fully connected layer to regress an aging score (0-1 value), directly reflecting the overall degradation degree of the material; 2. The bacteria and dust segmentation branches each adopt a decoding structure similar to U-Net, outputting pixel-level binary masks to mark bacterial colonies and dust-covered areas respectively. In this way, during training, each branch reuses parameters by sharing the low-level features of the shared encoder (containing basic texture and edge information), improving model efficiency. During training, the loss weights of different task branches are dynamically adjusted (in the early stages of training, more attention may be needed for certain tasks to ensure the model converges quickly; as training progresses, the loss weights can be dynamically adjusted according to the progress of the tasks, using adaptive strategies such as weight adjustment based on task performance or simple linear scheduling strategies), balancing the optimization speed of different tasks, so that the model improves the learning efficiency of the synergistic effect of aging and contamination while reducing redundant computation.

[0110] Step 3: Material replacement and self-cleaning assessment

[0111] The system uses threshold settings to determine whether the air conditioner should perform self-cleaning and whether materials should be replaced. For example, if the aging score exceeds 0.7, the user is reminded to replace materials promptly; if the area covered by bacteria and dust exceeds a certain threshold, the user is reminded to activate the air conditioner's self-cleaning function.

[0112] Step 4: Sterilization by dynamic distribution of ultraviolet light field

[0113] When step 3 determines that the area covered by bacteria and dust exceeds a preset threshold, the system triggers a photocatalytic synergistic sterilization mechanism. In the composite coating pre-applied on the surface of the evaporator fins, ultraviolet light is transmitted along the fin geometry to the physically shielded area through the surface plasmon waveguide effect, breaking through the physical limitation of traditional straight-line propagation.

[0114] During the dynamic distribution sterilization process of ultraviolet light field, it is necessary to control the angle of ultraviolet light to ensure that the light can evenly and effectively irradiate all areas that need to be cleaned, especially those shadowy areas that are difficult to irradiate directly.

[0115] The micro UV-LED array integrated on the back of the evaporator adopts a biomimetic compound eye arrangement. Each LED unit is equipped with an independently driven pulse width modulation (PWM) circuit, which can independently control the switching state and duty cycle of each LED. By adjusting the phase difference of the PWM signal, the phase shift between LEDs can be achieved.

[0116] The pollution heat map generated in step 2 is analyzed in real time using a reinforcement learning model (a dynamic control strategy based on the PPO algorithm), the optimal coverage path of the light spot is calculated, and the phase difference of each LED is dynamically adjusted according to the calculation results to optimize the light field distribution.

[0117] The system dynamically adjusts the irradiance and phase difference of each LED with a period of 10ms, allowing the overlap rate of adjacent light spots to adaptively fluctuate within the range of 15%-25%. Simultaneously, based on the microbial inactivation efficiency data fed back by the photodetector on the fin surface (which reflects the sterilization effect under the current ultraviolet wavelength combination), it determines whether the microbial inactivation efficiency under the current wavelength combination has reached the expected target, identifies which areas have low inactivation efficiency, and adjusts the phase difference and irradiance of the LEDs in real time based on the identification results. It also optimizes the ultraviolet wavelength combination in reverse (adjusting the ratio of the 275nm main band to the 310nm auxiliary band; for example, if the inactivation efficiency of a certain area is low, the proportion of the 275nm wavelength of LEDs near that area can be increased to enhance the sterilization effect). By changing the relative proportion of different wavelengths, it optimizes the light field distribution and sterilization effect, and achieves active compensation for the photocatalytic reaction rate in the shaded area (the purpose is to avoid blind spots in ultraviolet irradiation and enhance the sterilization effect).

[0118] During the dynamic distribution sterilization process using ultraviolet light, a multispectral imaging device continuously monitors the intensity of characteristic absorption peaks of organic pollutant decomposition products (e.g., the characteristic absorption peak intensity of VOCs at a specific wavelength, such as 9.6 μm, reflecting the degree of pollutant decomposition and sterilization effect). A closed-loop control algorithm dynamically adjusts the wavelength combination based on feedback data to ensure optimal sterilization within the energy consumption threshold. Furthermore, the algorithm can dynamically adjust the length of the irradiation cycle (high absorption peak intensity indicates incomplete pollutant decomposition, requiring a longer irradiation cycle; low absorption peak intensity indicates near-complete decomposition, allowing for a shorter or terminated irradiation cycle). This ensures that the total bacterial count is <15 CFU / cm³ within the energy consumption threshold. 2 The purification target is determined, and the light field distribution is dynamically optimized to ensure that the shadow areas are effectively illuminated.

[0119] In practice, intensity, phase difference, ultraviolet light, and irradiation period are usually adjusted simultaneously to achieve optimal sterilization results. The specific steps are as follows:

[0120] Initialization settings:

[0121] Set the initial irradiance, phase difference, ultraviolet wavelength combination, and irradiation period.

[0122] Real-time monitoring:

[0123] The efficiency of microbial inactivation and the intensity of characteristic absorption peaks of organic pollutants are monitored in real time using photodetectors and multispectral imaging equipment.

[0124] Dynamic adjustment:

[0125] Based on the monitoring data, the following parameters will be dynamically adjusted:

[0126] Irradiation intensity: controlled by adjusting the PWM duty cycle of the LED.

[0127] Phase difference: Controlled by adjusting the phase difference of the PWM signal.

[0128] Ultraviolet wavelength combination: Adjust the relative proportions of different wavelengths based on inactivation efficiency data.

[0129] Irradiation cycle: The irradiation cycle is dynamically extended or terminated based on the characteristic absorption peak intensity of organic pollutants.

[0130] Closed-loop control:

[0131] By using a closed-loop control algorithm to comprehensively consider the above parameters, the best sterilization effect can be achieved within the energy consumption threshold.

[0132] This application utilizes a dynamic optical adjustment algorithm and the optical properties of photocatalytic materials to achieve omnidirectional ultraviolet radiation distribution, significantly improving the efficiency and intelligence of air conditioner self-cleaning. Simultaneously, a deep learning-based optical imaging fusion method dynamically identifies the distribution of bacteria and dust, as well as the aging degree of the self-cleaning materials, promptly reminding users to activate the air conditioner's self-cleaning function and replace materials, thus enhancing the user experience.

[0133] In another embodiment of this application, an air conditioner self-cleaning device is also provided, such as... Figure 3 As shown, the device includes:

[0134] The acquisition module 11 is used to acquire composite coating images of the composite coating on the surface of the component to be cleaned in the air conditioner;

[0135] Input module 12 is used to input the composite coating image into a preset coating state detection model to obtain coating state information;

[0136] The first determining module 13 is used to determine whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner based on the coating status information.

[0137] The self-cleaning module 14 is used to self-clean the composite coating on the surface of the component to be cleaned if the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner.

[0138] Optionally, the preset coating state detection model includes: a shared encoder and a dirt detection branch decoder;

[0139] The shared encoder is used to extract local dirt features from the composite coating image;

[0140] The dirt detection branch decoder is used to determine the dirt coverage area and dirt area information of the component to be cleaned based on the local dirt characteristics, so as to serve as the coating status information.

[0141] Optionally, determining whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner based on the coating status information includes:

[0142] Determine whether the dirty area information in the coating status information is greater than a preset dirty area threshold;

[0143] If the dirty area information is greater than the preset dirty area threshold, it is determined that the degree of dirtiness of the component to be cleaned meets the self-cleaning conditions of the air conditioner.

[0144] Optional self-cleaning module, including:

[0145] The first determining unit is used to determine the irradiation path of the ultraviolet light spot based on the dirt coverage area in the coating status information;

[0146] An adjustment unit is used to adjust the angle of ultraviolet light emitted by the ultraviolet irradiation component according to the irradiation path until the composite coating is self-cleaned.

[0147] Optional, the self-cleaning module also includes:

[0148] The acquisition unit is used to acquire the characteristic absorption peak intensity and microbial inactivation efficiency data of pollutants on the surface of the composite coating;

[0149] The second determining unit is used to determine the irradiation period length, irradiation intensity, and phase difference between different ultraviolet light sources in the ultraviolet irradiation component based on the intensity of the characteristic absorption peak.

[0150] The third determining unit is used to determine the ultraviolet light wavelength combination of the ultraviolet irradiation component based on the microbial inactivation efficiency data.

[0151] The control unit is used to control the ultraviolet irradiation component to emit ultraviolet light according to the combination of the irradiation period length, the irradiation intensity, the phase difference and the ultraviolet light wavelength.

[0152] Optionally, the preset coating state detection model further includes: an aging detection branch decoder;

[0153] The shared encoder is also used to extract global aging features from the composite coating image;

[0154] The aging detection branch decoder is used to determine the aging degree information of the component to be cleaned based on the global aging characteristics, so as to serve as the coating state information.

[0155] Optionally, the device further includes:

[0156] The second determining module is used to determine whether the aging degree information in the coating state information is greater than a preset aging degree threshold.

[0157] The reminder module is used to issue a composite coating aging reminder if the aging information is greater than a preset aging threshold.

[0158] In another embodiment of this application, an air conditioner is also provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus.

[0159] Memory, used to store computer programs;

[0160] The processor, when executing a program stored in memory, implements the air conditioner self-cleaning method described in any of the foregoing method embodiments.

[0161] The air conditioner provided in this embodiment of the invention uses a processor to execute a program stored in a memory to determine the coating state information of the composite coating based on a preset coating state detection model and a composite coating image. If the degree of dirt on the component to be cleaned is determined to meet the self-cleaning conditions of the air conditioner based on the coating state information, the processor performs self-cleaning on the composite coating on the surface of the component to be cleaned. This achieves automatic determination of the timing of self-cleaning of the air conditioner based on the coating state information of the composite coating, and automatically performs self-cleaning on the composite coating on the surface of the component to be cleaned. This avoids subjective judgment by the user, facilitates user use, and prevents excessive dirt accumulation inside the air conditioner, which could affect the user's health.

[0162] The communication bus 1140 mentioned above in the air conditioner can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus 1140 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0163] Communication interface 1120 is used for communication between the air conditioner and other devices.

[0164] The memory 1130 may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0165] The processor 1110 mentioned above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0166] In another embodiment of this application, a computer-readable storage medium is also provided, on which a program for an air conditioner self-cleaning method is stored. When the program for the air conditioner self-cleaning method is executed by a processor, it implements the steps of the air conditioner self-cleaning method described in any of the foregoing method embodiments.

[0167] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0168] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A self-cleaning method for air conditioners, characterized in that, The method includes: Obtain composite coating images of the composite coating on the surface of the components to be cleaned in an air conditioner; The composite coating image is input into a preset coating state detection model to obtain coating state information; Based on the coating status information, determine whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner; If the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner, the composite coating on the surface of the component to be cleaned will be self-cleaned.

2. The air conditioner self-cleaning method according to claim 1, characterized in that, The preset coating state detection model includes: a shared encoder and a dirt detection branch decoder; The shared encoder is used to extract local dirt features from the composite coating image; The dirt detection branch decoder is used to determine the dirt coverage area and dirt area information of the component to be cleaned based on the local dirt characteristics, so as to serve as the coating status information.

3. The air conditioner self-cleaning method according to claim 1, characterized in that, Determining whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner based on the coating status information includes: Determine whether the dirty area information in the coating status information is greater than a preset dirty area threshold; If the dirty area information is greater than the preset dirty area threshold, it is determined that the degree of dirtiness of the component to be cleaned meets the self-cleaning conditions of the air conditioner.

4. The air conditioner self-cleaning method according to claim 1, characterized in that, Self-cleaning of the composite coating on the surface of the component to be cleaned includes: The irradiation path of the ultraviolet light spot is determined based on the dirt coverage area in the coating status information; Adjust the angle of the ultraviolet light emitted by the ultraviolet irradiation component according to the irradiation path until the composite coating is self-cleaned.

5. The air conditioner self-cleaning method according to claim 4, characterized in that, The self-cleaning process for the composite coating on the surface of the component to be cleaned further includes: Obtain data on the characteristic absorption peak intensity and microbial inactivation efficiency of pollutants on the surface of the composite coating; The irradiation period length, irradiation intensity, and phase difference between different ultraviolet light sources in the ultraviolet irradiation component are determined based on the intensity of the characteristic absorption peak. The combination of ultraviolet light wavelengths for the ultraviolet irradiation component is determined based on the microbial inactivation efficiency data. The ultraviolet irradiation component emits ultraviolet light according to the combination of the irradiation period length, the irradiation intensity, the phase difference, and the ultraviolet light wavelength.

6. The air conditioner self-cleaning method according to claim 2, characterized in that, The preset coating state detection model also includes: an aging detection branch decoder; The shared encoder is also used to extract global aging features from the composite coating image; The aging detection branch decoder is used to determine the aging degree information of the component to be cleaned based on the global aging characteristics, so as to serve as the coating state information.

7. The air conditioner self-cleaning method according to claim 1, characterized in that, The method further includes: Determine whether the aging degree information in the coating status information is greater than a preset aging degree threshold; If the aging information exceeds a preset aging threshold, an aging warning for the composite coating will be issued.

8. An air conditioner self-cleaning device, characterized in that, The device includes: The acquisition module is used to acquire composite coating images of the composite coating on the surface of the components to be cleaned in the air conditioner; The input module is used to input the composite coating image into a preset coating state detection model to obtain coating state information; The first determining module is used to determine whether the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner based on the coating status information. The self-cleaning module is used to self-clean the composite coating on the surface of the component to be cleaned if the degree of dirt on the component to be cleaned meets the self-cleaning conditions of the air conditioner.

9. An air conditioner, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor, when executing a program stored in the memory, implements the air conditioner self-cleaning method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program for an air conditioner self-cleaning method, which, when executed by a processor, implements the steps of the air conditioner self-cleaning method according to any one of claims 1-7.

Images