Method and device for identifying frost layer of air conditioner, air conditioner

Abstract

The application relates to the technical field of intelligent household appliances, and discloses a method and device for identifying frost layers of an air conditioner and the air conditioner. The method comprises the following steps: dividing samples in an initial sample set into a first frost layer type and a second frost layer type according to original frost layer labels; determining a similarity score of each sample in the initial sample set and the second frost layer type; wherein the frost thickness of the second frost layer type is greater than that of the first frost layer type; updating the original frost layer labels of the samples in the initial sample set according to the similarity score to form a new sample set; and inputting the new sample set into a neural network for training to obtain a frost layer type identification model. The frost layer identification model can output the probability that the frost layer belongs to thick frost, thereby identifying whether the degree of frost removal is required, and more accurately judging the frost layer state.

Description

Technical Field

[0001] The present application relates to the technical field of smart home appliances, for example, to a method and device for identifying frost layer of an air conditioner, and an air conditioner. Background Art

[0002] When an air conditioner operates in heating mode, frost frequently forms on the surface of the outdoor unit's heat exchanger. This frost significantly increases the heat transfer resistance, significantly reducing heating efficiency and even causing system shutdown in severe cases. Therefore, precise defrost control is crucial to ensuring optimal heating performance.

[0003] Related technology discloses an air conditioning control method, device, air conditioner and storage medium, wherein the control method includes: respectively calculating a first temperature difference between the user-set temperature and the indoor ambient temperature and a second temperature difference between the outdoor ambient temperature and the external disk temperature; if the first temperature difference is within the first preset temperature range and the second temperature difference is within the second preset temperature range, then determining that the frosting state of the outdoor unit is the thin frost state; respectively determining whether the first temperature difference is within the third preset temperature range and whether the second temperature difference is within the fourth preset temperature range; if the first temperature difference is within the third preset temperature range and the second temperature difference is within the fifth preset temperature range, then determining that the frosting state of the outdoor unit is the thick frost state.

[0004] During the implementation of the embodiments of the present disclosure, it was found that at least the following problems exist in the related art:

[0005] Related technologies use the difference between the first and second temperatures to determine frost formation. These rely on preset thresholds and simple logic, making them ineffective in addressing the nonlinear nature of frost growth. This can lead to significant errors in determining the critical states between light and heavy frost, causing defrost decisions to deviate significantly from actual demand, resulting in energy waste and significant fluctuations in indoor temperature.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute prior art known to ordinary technicians in this field. Summary of the Invention

[0007] In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not an extensive review, nor is it intended to identify key / critical elements or delineate the scope of protection of these embodiments, but rather serves as a prelude to the detailed description that follows.

[0008] The embodiments of the present disclosure provide a method and device for identifying the frost layer of an air conditioner, and an air conditioner, so as to more accurately determine the frost layer status.

[0009] In some embodiments, the method for identifying frost layer of an air conditioner includes: dividing the samples in the initial sample set into a first frost layer type and a second frost layer type according to the original frost layer labels; determining a similarity score between each sample in the initial sample set and the second frost layer type; wherein the frost thickness of the second frost layer type is greater than the frost thickness of the first frost layer type; updating the original frost layer labels of the samples in the initial sample set according to the similarity scores to form a new sample set; and inputting the new sample set into a neural network for training to obtain a frost layer type recognition model.

[0010] In some embodiments, the apparatus for identifying frost layer of an air conditioner includes: a processor and a memory storing program instructions, and the processor is configured to execute the aforementioned method for identifying frost layer of an air conditioner when running the program instructions.

[0011] In some embodiments, the air conditioner includes: an air conditioner body; and the aforementioned device for identifying frost layer of the air conditioner, which is installed on the air conditioner body.

[0012] In some embodiments, the computer-readable storage medium stores program instructions, and when the program instructions are executed, the aforementioned method for identifying frost layer of an air conditioner is executed.

[0013] The method and device for identifying frost layer of an air conditioner, and the air conditioner provided by the embodiments of the present disclosure can achieve the following technical effects:

[0014] First, the samples in the initial sample set are divided into the first and second frost types, and the frost thickness of the samples is preliminarily distinguished using the existing method. A similarity score is then determined between each sample in the initial sample set and the second frost type to determine the sample's proximity to thick frost. The original frost labels of the samples in the initial sample set are updated based on the similarity scores, and the similarity scores are used to more finely distinguish the frost thickness of the samples, thus forming a new sample set. This new sample set is then fed into a neural network for training, ultimately resulting in a frost recognition model. The frost recognition model outputs the probability that the frost layer is thick, thereby identifying whether defrosting is necessary and more accurately determining the frost condition.

[0015] The above general description and the following description are exemplary and explanatory only and are not intended to limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] One or more embodiments are exemplarily described by corresponding drawings. These exemplary descriptions and drawings do not limit the embodiments. Elements with the same reference numerals in the drawings are shown as similar elements. The drawings do not constitute a scale limitation. In addition,

[0017] Figure 1 is a schematic diagram of a first method for identifying a frost state of an air conditioner provided by an embodiment of the present disclosure;

[0018] Figure 2 is a schematic diagram of a second method for identifying a frost state of an air conditioner provided by an embodiment of the present disclosure;

[0019] Figure 3 is a schematic diagram of a method for determining sample centers of a first frost layer type and a second frost layer type provided by an embodiment of the present disclosure;

[0020] Figure 4 is a schematic diagram of a third method for identifying the frost state of an air conditioner provided by an embodiment of the present disclosure;

[0021] Figure 5 is a schematic diagram of a fourth method for identifying a frost state of an air conditioner provided by an embodiment of the present disclosure;

[0022] Figure 6 is a schematic diagram of a fifth method for identifying a frost state of an air conditioner provided by an embodiment of the present disclosure;

[0023] Figure 7 is a schematic diagram of a method for obtaining a real-time frost state of an air conditioner provided by an embodiment of the present disclosure;

[0024] Figure 8 is a schematic diagram of a first device for identifying the frost state of an air conditioner provided by an embodiment of the present disclosure;

[0025] Figure 9 is a schematic diagram of a second device for identifying the frost state of an air conditioner provided by an embodiment of the present disclosure;

[0026] Figure 10 Schematic diagram of an air conditioner provided by an embodiment of the present disclosure.

[0027] Reference numerals:

[0028] 80. Device for identifying frost state of air conditioner; 81. Classification module; 82. Determination module; 83. Update module; 84. Training module;

[0029] 90. Device for identifying frost status of an air conditioner; 91. Processor; 92. Memory; 93. Communication interface; 94. Bus;

[0030] 100. Air conditioner. DETAILED DESCRIPTION

[0031] In order to be able to understand the features and technical content of the embodiments of the present disclosure in more detail, the implementation of the embodiments of the present disclosure is described in detail below in conjunction with the accompanying drawings. The accompanying drawings are for reference only and are not used to limit the embodiments of the present disclosure. In the following technical description, for the sake of convenience of explanation, a full understanding of the disclosed embodiments is provided through multiple details. However, one or more embodiments can still be implemented without these details. In other cases, to simplify the drawings, well-known structures and devices can be simplified for display.

[0032] In the description and claims of the embodiments of the present disclosure, as well as in the accompanying drawings, the terms "first," "second," and the like are used to distinguish similar items and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate to describe the embodiments of the present disclosure herein. In addition, the terms "including," "having," and any variations thereof are intended to cover non-exclusive inclusions.

[0033] Unless otherwise stated, the term "plurality" means two or more.

[0034] In the embodiment of the present disclosure, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B means: A or B.

[0035] The term "and / or" describes an association between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or A and B.

[0036] The term "correspondence" may refer to an association relationship or a binding relationship. The correspondence between A and B means that there is an association relationship or a binding relationship between A and B.

[0037] Combine Figure 1 As shown, an embodiment of the present disclosure provides a method for identifying a frost state of an air conditioner, comprising:

[0038] S101: The processor divides samples in the initial sample set into a first frost layer type and a second frost layer type according to the original frost layer labels of the samples.

[0039] S102, the processor determines a similarity score between each sample in the initial sample set and a second frost layer type; wherein the frost thickness of the second frost layer type is greater than the frost thickness of the first frost layer type.

[0040] S103: The processor updates the original frost layer labels of the samples in the initial sample set according to the similarity scores to form a new sample set.

[0041] S104: The processor inputs the new sample set into the neural network for training to obtain a frost layer type recognition model.

[0042] Collect sample data in advance. Using existing sensors, collect air conditioner operating parameters, indoor environmental parameters, and outdoor environmental parameters under different environmental conditions, such as temperature, humidity, and time. Optionally, operating parameters include outdoor coil temperature, indoor coil temperature, compressor frequency, electronic expansion valve opening, and compressor exhaust temperature. This sample data forms the initial sample set. The temperature, frequency, and opening variation characteristics presented in this sample data can reflect changes in the degree of frost on the air conditioner's outdoor unit.

[0043] Existing frost classifications generally use thick frost and thin frost, with thick frost labeled 1 and thin frost labeled 0. Based on these existing frost labels, the samples in the initial sample set are divided into the first and second frost types. These samples contain feature data from multiple dimensions, such as outdoor coil temperature, indoor coil temperature, outdoor ambient temperature, indoor ambient temperature, compressor frequency, electronic expansion valve opening, and compressor exhaust temperature. The frost thickness of the second frost type is greater than that of the first frost type, meaning the first frost type is thin frost and the second frost type is thick frost. Existing frost labels indicate that thick frost has a higher label than thin frost. Therefore, a similarity score is determined for each sample in the initial sample set with the second frost type. A higher similarity score indicates a closer similarity to thick frost. The similarity scores of the samples in the initial sample set follow the same trend as the original frost labels. Based on the obtained similarity scores, the original frost labels of the samples in the initial sample set are updated to form a new sample set. This new sample set is then fed into a neural network for training, resulting in a frost type recognition model.

[0044] According to the method for identifying frost layer of air conditioners provided by the embodiment of the present disclosure, the samples in the initial sample set are first divided into a first frost layer type and a second frost layer type, and the thickness of the frost layer of the samples is preliminarily distinguished in the original way. Then, the similarity score between each sample in the initial sample set and the second frost layer type is determined to judge the degree of proximity of the sample to thick frost. The original frost layer labels of the samples in the initial sample set are updated according to the similarity scores, and the frost layer thickness of the samples is distinguished more finely using the similarity scores, thereby forming a new sample set. The new sample set is input into a neural network for training, and finally a frost layer recognition model is obtained. The frost layer recognition model can output the probability that the frost layer is thick frost, and then identify whether it has reached the level that requires defrosting, and more accurately judge the frost layer status.

[0045] Combine Figure 2 As shown, the embodiment of the present disclosure provides another method for identifying the frost state of an air conditioner, including:

[0046] S101: The processor divides samples in the initial sample set into a first frost layer type and a second frost layer type according to the original frost layer labels of the samples.

[0047] S112 , the processor determines sample centers of the first frost layer type and the second frost layer type.

[0048] S122, the processor uses one of the methods selected from Euclidean distance, principal component analysis, autoencoding, and K-means clustering to analyze each sample and sample center in the initial sample set to obtain a similarity score between each sample in the initial sample set and the second frost layer type.

[0049] S103: The processor updates the original frost layer labels of the samples in the initial sample set according to the similarity scores to form a new sample set.

[0050] S104: The processor inputs the new sample set into the neural network for training to obtain a frost layer type recognition model.

[0051] To determine the similarity score between each sample in the initial sample set and the second frost type, we first determine the sample center of each frost type. As mentioned previously, there are two frost types. The sample centers for the first frost type (thin frost) and the second frost type (thick frost) are determined separately. The sample centers reflect the average frost thickness of samples of the same frost type. Using one of the following methods, such as Euclidean distance, principal component analysis, autoencoding, or K-means clustering, we comprehensively analyze each sample and sample center in the initial sample set to determine a similarity score between each sample in the initial sample set and the second frost type. This improves the accuracy of identifying the frost state represented by the sample.

[0052] Combine Figure 3 As shown, at S112, the processor determines the sample centers of the first frost layer type and the second frost layer type, including:

[0053] S1112: The processor calculates the average data value of samples of the same frost layer type in multiple characteristic dimensions reflecting frost formation.

[0054] S1122: The processor uses each obtained average value as the sample center of the corresponding frost layer type.

[0055] For the first frost layer type, sum the data for each frost-reflecting dimension for all samples in the first frost layer type and then divide by the number of samples in the first frost layer type to obtain the average data for each dimension. For example, if there are n samples in the first frost layer type, each with m dimensions, the average data for the first frost layer type in the jth dimension is: the sum of the data for all samples in the first frost layer type in the jth dimension divided by n. In this way, the average data for the first frost layer type in each of the m dimensions is obtained. Combining these averages forms the sample center of the first frost layer type.

[0056] Similarly, the same operation is performed on the second frost layer type to calculate the average value of the data of the thick frost samples in each characteristic dimension reflecting frost formation, and then obtain the sample center of the second frost layer type.

[0057] Optionally, at S122, the processor analyzes each sample and the sample center using Euclidean distance to obtain a similarity score between each sample and the second frost layer type, including:

[0058] score=S1 / (S0+S1).

[0059] Among them, score is the similarity score, S0 is the Euclidean distance from the sample to the sample center of the first frost layer type, and S1 is the Euclidean distance from the sample to the sample center of the second frost layer type.

[0060] The above formula can be used to determine which type of frost layer the sample is closest to. The larger the score value, the thicker the frost layer and the closer it is to the second frost layer type.

[0061] Combine Figure 4 As shown, the embodiment of the present disclosure provides another method for identifying the frost state of an air conditioner, including:

[0062] S101: The processor divides samples in the initial sample set into a first frost layer type and a second frost layer type according to the original frost layer labels of the samples.

[0063] S102, the processor determines a similarity score between each sample in the initial sample set and a second frost layer type; wherein the frost thickness of the second frost layer type is greater than the frost thickness of the first frost layer type.

[0064] S113: The processor determines a critical sample according to the similarity score.

[0065] S123: The processor updates the original frost layer labels of the critical samples to corresponding similarity scores to form a new sample set.

[0066] S104: The processor inputs the new sample set into the neural network for training to obtain a frost layer type recognition model.

[0067] After obtaining a similarity score between each sample in the initial sample set and the frost thickness of the second frost layer type, critical samples are determined based on the similarity scores of the samples in the initial sample set. Critical samples are samples with similarity scores between thick frost and thin frost, making them difficult to classify. Alternatively, samples with similarity scores between 0.4 and 0.6 are determined as critical samples.

[0068] The original frost labels of critical samples are 1 or 0, and the original frost labels are updated to the corresponding similarity scores. For example, if a critical sample originally has a frost label of 0 and a similarity score of 0.5, the frost score of the critical sample is updated to 0.5. The labels of non-critical samples remain 0 or 1. This method updates the original frost scores of critical samples that are difficult to classify to similarity scores, giving them fuzzy labels. This breaks the limitations of traditional labels, strengthens the model's generalization ability for critical frost states, and accurately reflects the gradual characteristics of frost.

[0069] Combine Figure 5 As shown, the embodiment of the present disclosure provides another method for identifying the frost state of an air conditioner, including:

[0070] S101: The processor divides samples in the initial sample set into a first frost layer type and a second frost layer type according to the original frost layer labels of the samples.

[0071] S102, the processor determines a similarity score between each sample in the initial sample set and a second frost layer type; wherein the frost thickness of the second frost layer type is greater than the frost thickness of the first frost layer type.

[0072] S103: The processor updates the original frost layer labels of the samples in the initial sample set according to the similarity scores to form a new sample set.

[0073] S105: The processor normalizes the sample data in the new sample set.

[0074] S104: The processor inputs the new sample set into the neural network for training to obtain a frost layer type recognition model.

[0075] Before a new sample set is fed into the neural network, the sample data in the new sample set is normalized. Normalization uses maximum normalization to map the original data range to [-1, 1]. This helps eliminate the influence of different dimensions or magnitudes on data analysis.

[0076] Optionally, the neural network used in this embodiment is a fully connected neural network consisting of three fully connected layers. A new sample set is input into the neural network for training, with the first two layers of neurons using the ReLU activation function and the last layer of neurons using the Sigmoid activation function.

[0077] Each neuron in a fully connected layer is connected to all neurons in the previous layer, meaning that the output of each neuron in the previous layer is passed to every neuron in the current layer. Its main function is to integrate and transform the features extracted by the previous layer, thereby completing complex nonlinear mapping.

[0078] The input vector of the fully connected layer is , where x i is the output of the i-th neuron in the previous layer. The weight matrix is ​​W, which is an m×n matrix, where W ij Represents the weight from the jth neuron in the previous layer to the ith neuron in the current layer; the bias vector is , where b i is the bias of the i-th neuron in the current layer. The activation function is , the input z of the i-th neuron in the current layer i It can be expressed as , written in matrix form as The output of the i-th neuron in the current layer is It can be expressed as , written in matrix form as After the calculation of the fully connected layer and the activation function, the model calculation results will be classified by the Sigmoid classifier. The calculation formula of Sigmoid is , and output the probability of belonging to two categories of labels.

[0079] Through multiple iterations of training, the model fully learns sample characteristics, particularly those of critical samples, improving its ability to detect frost levels. This process leverages the advantage of fuzzy labels, which more accurately reflect the true state of category boundaries, enhancing the model's generalization capabilities at critical points. Compared to models trained with traditional labels, this model more fully learns the uncertainty of samples at critical points, significantly improving detection accuracy.

[0080] Combine Figure 6 As shown, the embodiment of the present disclosure provides another method for identifying the frost state of an air conditioner, including:

[0081] S101: The processor divides samples in the initial sample set into a first frost layer type and a second frost layer type according to the original frost layer labels of the samples.

[0082] S102, the processor determines a similarity score between each sample in the initial sample set and a second frost layer type; wherein the frost thickness of the second frost layer type is greater than the frost thickness of the first frost layer type.

[0083] S103: The processor updates the original frost layer labels of the samples in the initial sample set according to the similarity scores to form a new sample set.

[0084] S104: The processor inputs the new sample set into the neural network for training to obtain a frost layer type recognition model.

[0085] S106: The processor obtains the operating data, working condition data, and indoor and outdoor environment data of the air conditioner in real time.

[0086] S107 , the processor inputs the operation data, the working condition data, and the indoor and outdoor environment data into the frost type recognition model to obtain the real-time frost state of the air conditioner.

[0087] S108: When the real-time frost layer state meets the defrosting conditions, the processor controls the air conditioner to perform defrosting.

[0088] The trained frost type recognition model is deployed in the processor. During air conditioner operation, sensors collect real-time operating data, operating condition data, and indoor and outdoor environmental data. This data is then fed into the frost type recognition model. The frost type recognition model outputs the results, which in turn provide the air conditioner's real-time frost status. If the real-time frost status meets the defrost conditions, the air conditioner is controlled to defrost. Otherwise, further operating data, operating condition data, and indoor and outdoor environmental data are collected and analyzed using the frost type recognition model. This allows precise control of air conditioner defrost, minimizing the discrepancy between defrost decisions and actual demand, thereby reducing energy waste and minimizing large fluctuations in indoor temperature.

[0089] Optionally, to accurately determine the severity of frost on the air conditioner's outdoor unit heat exchanger, a result queue with a fixed capacity of a preset number of results is set as an accumulator. Initially, the queue is empty and equipped with a counter whose initial value is set to 0. Each time the frost layer type recognition model outputs a real-time frost probability result, the counter is incremented by 1 and the result is added to the queue. Optionally, the preset number of results is 30.

[0090] When the counter value is less than or equal to the preset number, the new frost probability result is directly stored in the queue. At the same time, the processor accumulates the new results. For example, if the first frost probability result is 0.3, the accumulator value is 0.3. The second result is 0.2, the accumulator value is updated to 0.3 + 0.2 = 0.5, and so on.

[0091] When the counter value exceeds the preset number, the queue always retains the latest preset number of results, following the first-in, first-out principle. The oldest frost probability result at the head of the queue is first removed and subtracted from the accumulator. The newly generated real-time frost probability result is then added to the tail of the queue and accumulated in the accumulator. For example, if the current accumulator value is 5.5, the value at the head of the queue is 0.1, and the new frost probability result is 0.4, 0.1 is first subtracted from the accumulator to get 5.4, and the new result 0.4 is added to the accumulator, updating the accumulator value to 5.8.

[0092] Combine Figure 7 As shown, in S107, the processor obtains the real-time frost state of the air conditioner, including:

[0093] S117: The processor obtains the frost probability output by the frost layer type recognition model.

[0094] S127: When the accumulated value of the frost probability is greater than the probability threshold, the processor determines that the real-time frost layer state is a severe frost state.

[0095] The frost layer type recognition model outputs a frost probability within the interval [0, 1]. A pre-set probability threshold is used to obtain the frost probability and compare it with the threshold. As you can see, the frost layer gradually thickens, so the frost probability output by the frost layer type recognition model may not reach the defrost probability at each occurrence. Therefore, the frost probabilities output by the frost layer type recognition model are accumulated. If the accumulated value exceeds the probability threshold, the real-time frost layer status is determined to be severe.

[0096] Optionally, the probability threshold is 20. If the cumulative probability value output by the frost layer type identification model is greater than 20, the air conditioner is determined to be severely frosted, and a defrost operation is automatically triggered immediately. If the cumulative value does not exceed 20, the system continues to process, accumulate, and compare subsequent real-time frost probability results according to the above-mentioned first-in, first-out method until the cumulative sum exceeds the threshold or other termination conditions such as the end of air conditioning operation are met.

[0097] This approach avoids unnecessary defrosting operations, effectively improving energy efficiency. It also utilizes only existing sensor data, eliminating the need for additional costs like frost thickness sensors. This reduces equipment procurement and subsequent maintenance costs, facilitating the large-scale application of this technology in various scenarios.

[0098] By building a lightweight network to output the probability of heavy frost, the results are accumulated in a first-in, first-out manner. When the accumulated frost probability exceeds a probability threshold, the system determines that the frost is severe and automatically defrosts the device. Compared to traditional binary classification methods, this embodiment can quantify the degree of heavy frost, improve the accuracy of defrost decisions, and balance real-time performance with low computational overhead.

[0099] Optionally, at S107, the processor obtains the real-time frost state of the air conditioner, including:

[0100] The processor obtains the frost probability output by the frost layer type recognition model.

[0101] The processor determines that the real-time frost layer state is a severe frost state when the accumulated value of the frost probability is greater than the probability threshold and the number of frost probabilities reaches a preset number.

[0102] The frost layer type identification model outputs a frost probability within the interval [0, 1]. A probability threshold is pre-set, the frost probability is obtained, and the frost probability is compared with the probability threshold. It is understandable that the frost layer gradually thickens, so the frost probability output by the frost layer type identification model may not reach the defrost probability each time. Therefore, the frost probabilities output by the frost layer type identification model are accumulated. Throughout this process, the processor monitors the counter value in real time to count the number of frost probabilities output by the frost layer type identification model. When the counter reaches the preset number, the sum of the probability accumulation values ​​in the accumulator is compared with the probability threshold each time the accumulator is updated. When the accumulated frost probability value is greater than the probability threshold and the number of frost probabilities reaches the preset number, the real-time frost layer state is determined to be severe frost.

[0103] Thus, when determining whether the real-time frost state is severe, in addition to determining whether the accumulated frost probability value is greater than the probability threshold, it is also necessary to determine whether the number of frost probabilities output by the frost type recognition model reaches a preset number. This ensures the number of frost probabilities involved in the judgment, making the frost probabilities involved in the judgment more stable and thus facilitating accurate judgment of the frost state.

[0104] Combine Figure 8 As shown, an embodiment of the present disclosure provides a device 80 for identifying the frost layer of an air conditioner, comprising: a division module 81, a determination module 82, an update module 83 and a training module 84. The division module 81 is configured to divide the samples in the initial sample set into a first frost layer type and a second frost layer type according to the original frost layer labels. The determination module 82 is configured to determine a similarity score between each sample in the initial sample set and the second frost layer type; wherein the frost thickness of the second frost layer type is greater than the frost thickness of the first frost layer type. The update module 83 is configured to update the original frost layer labels of the samples in the initial sample set according to the similarity scores to form a new sample set. The training module 84 is configured to input the new sample set into a neural network for training to obtain a frost layer type recognition model.

[0105] Using the device 80 for identifying frost layers on air conditioners provided by the embodiment of the present disclosure, the samples in the initial sample set are first divided into a first frost layer type and a second frost layer type, and the thickness of the frost layer of the samples is preliminarily distinguished in the original manner. Then, the similarity score between each sample in the initial sample set and the second frost layer type is determined to determine the degree of proximity between the sample and thick frost. The original frost layer labels of the samples in the initial sample set are updated based on the similarity scores, and the frost layer thickness of the samples is distinguished more finely using the similarity scores, thereby forming a new sample set. The new sample set is input into a neural network for training, and finally a frost layer recognition model is obtained. The frost layer recognition model can output the probability that the frost layer is thick frost, and then identify whether it has reached the level that requires defrosting, and more accurately judge the frost layer state.

[0106] Combine Figure 9 As shown, an embodiment of the present disclosure provides a device 90 for identifying frost in an air conditioner, comprising a processor 91 and a memory 92. Optionally, the device 90 may further include a communication interface 93 and a bus 94. The processor 91, communication interface 93, and memory 92 may communicate with each other via bus 94. The communication interface 93 may be used for information transmission. The processor 91 may invoke logic instructions in the memory 92 to execute the method for identifying frost in an air conditioner according to the above embodiment.

[0107] In addition, the logic instructions in the memory 92 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.

[0108] Memory 92, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as program instructions / modules corresponding to the methods in the embodiments of the present disclosure. Processor 91 executes the program instructions / modules stored in memory 92 to perform functional applications and data processing, thereby implementing the method for identifying frost in an air conditioner in the above-mentioned embodiments.

[0109] The memory 92 may include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function; the data storage area may store data generated based on the use of the terminal device. Furthermore, the memory 92 may include high-speed random access memory and non-volatile memory.

[0110] Combine Figure 10As shown, an embodiment of the present disclosure provides an air conditioner 100, comprising: an air conditioner body, and the above-mentioned device 80 (90) for identifying the frost layer of the air conditioner. The device 80 (90) for identifying the frost layer of the air conditioner is installed on the air conditioner body. The installation relationship described here is not limited to placement inside the air conditioner body, but also includes installation connections with other components of the air conditioner 100, including but not limited to physical connections, electrical connections or signal transmission connections. It can be understood by those skilled in the art that the device 80 (90) for identifying the frost layer of the air conditioner can be adapted to a feasible product body, thereby realizing other feasible embodiments.

[0111] An embodiment of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to execute the above-mentioned method for identifying the frost layer of an air conditioner.

[0112] The technical solutions of the embodiments of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes one or more instructions for causing a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present disclosure. The aforementioned storage medium may be a non-transitory storage medium, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, among other media capable of storing program code.

[0113] The foregoing description and accompanying drawings sufficiently illustrate the embodiments of the present disclosure to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, process, and other changes. The embodiments are merely representative of possible variations. Unless expressly required, individual components and functions are optional, and the order of operations may vary. Portions and features of some embodiments may be included in or substituted for portions and features of other embodiments. As used in this application, the term "and / or" is intended to encompass any and all possible combinations of one or more of the associated listed items. Furthermore, when used in this application, the term "comprise" and its variations "comprises" and / or "comprising" refer to the presence of the stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. Without further limitation, an element defined by the phrase "comprises a..." does not preclude the presence of additional identical elements in the process, method, or apparatus that includes the element. In this document, each embodiment may focus on the differences from other embodiments, and similar parts between the embodiments can be referenced. For methods, products, etc. disclosed in the embodiments, if they correspond to the method section disclosed in the embodiments, then the relevant parts can be referenced to the description of the method section.

[0114] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software may depend on the specific application and design constraints of the technical solution. The technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the embodiments of the present disclosure. The technicians will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0115] In the embodiments disclosed herein, the disclosed methods and products (including but not limited to devices and equipment) can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units may be merely a logical functional division. In actual implementation, other divisions may be used, such as combining or integrating multiple units or components into another system, or omitting or disabling some features. Furthermore, the coupling or direct coupling or communication connection shown or discussed between each other may be through some interface, or indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to implement the embodiments according to actual needs. Furthermore, the functional units in the embodiments disclosed herein may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit.

[0116] The flowcharts and block diagrams in the accompanying drawings show the possible implementation architectures, functions and operations of the systems, methods and computer program products according to the embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of the code, and the module, program segment or part of the code contains one or more executable instructions for implementing the specified logical functions. In some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, or they can sometimes be executed in the opposite order, which can depend on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in an order different from that disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, or they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified function or action, or may be implemented by a combination of dedicated hardware and computer instructions.

Claims

1. A method for identifying frost layer of an air conditioner, characterized in that: include: The samples in the initial sample set are divided into the first frost layer type and the second frost layer type according to the original frost layer labels; determining a similarity score between each sample in the initial sample set and the second frost layer type; wherein the frost thickness of the second frost layer type is greater than the frost thickness of the first frost layer type; updating the original frost layer labels of the samples in the initial sample set according to the similarity scores to form a new sample set; The new sample set is input into a neural network for training to obtain a frost layer type recognition model.

2. The method according to claim 1, characterized in that Determining a similarity score between each sample in the initial sample set and the second frost layer type includes: determining sample centers of the first frost layer type and the second frost layer type; Each sample in the initial sample set and the sample center are analyzed to obtain a similarity score between each sample in the initial sample set and the second frost layer type.

3. The method according to claim 2, characterized in that The determining of the sample centers of the first frost layer type and the second frost layer type includes: Calculate the average value of the data of samples of the same frost layer type on multiple characteristic dimensions reflecting frost formation; The obtained average value is used as the sample center of the corresponding frost layer type.

4. The method according to claim 1, wherein Updating the original frost layer labels of the samples in the initial sample set according to the similarity scores includes: determining a critical sample according to the similarity score; The original frost layer label of the critical sample is updated to the corresponding similarity score.

5. The method according to claim 1, wherein After forming a new sample set and before inputting the new sample set into the neural network for training, the method further includes: Normalization is performed on the sample data in the new sample set.

6. The method according to any one of claims 1 to 5, characterized in that After obtaining the frost layer type identification model, the method further includes: Real-time acquisition of operating data, operating condition data, and indoor and outdoor environmental data of the air conditioner; Inputting the operation data, the working condition data and the indoor and outdoor environment data into the frost type identification model to obtain the real-time frost state of the air conditioner; When the real-time frost layer state meets the defrosting condition, the air conditioner is controlled to perform defrosting.

7. The method according to claim 6, characterized in that The method of obtaining the real-time frost state of the air conditioner includes: Obtaining the frost probability output by the frost layer type identification model; When the accumulated value of the frost probability is greater than a probability threshold, it is determined that the real-time frost layer state is a severe frost state.

8. A device for identifying frost layer of an air conditioner, comprising a processor and a memory storing program instructions, characterized in that: The processor is configured to execute the method for identifying frost layer of an air conditioner according to any one of claims 1 to 7 when running the program instructions.

9. An air conditioner, characterized in that: include: Air conditioner body; The device for identifying frost layer of an air conditioner according to claim 8 is installed on the air conditioner body.

10. A computer-readable storage medium storing program instructions, characterized in that: When the program instructions are executed, the computer is used to execute the method for identifying frost layer of an air conditioner according to any one of claims 1 to 7.

Images