Air conditioning system and control method thereof

Abstract

The invention provides an air conditioner system and a control method thereof, particularly relates to the technical field of air conditioners, and aims to improve the accuracy of air conditioner fault diagnosis. The air conditioning system comprises a controller and a controller, wherein the controller is configured to obtain current operation data of the air conditioning system; inputting the current operation data into a target fault diagnosis model, and performing fault diagnosis on the air conditioning system through the target fault diagnosis model to obtain a fault diagnosis result of the air conditioning system; under the condition that the similarity between the historical operation data set and the normal operation data in the source sample data set is below a preset value, the target fault diagnosis model is obtained through training in the following mode that the source fault diagnosis model is trained based on the historical operation data set of the air conditioner system, and the target fault diagnosis model is obtained.

Description

Technical Field

[0001] This application relates to the field of air conditioning technology, and in particular to an air conditioning system and its control method. Background Technology

[0002] With continuous technological advancements, air conditioning fault diagnosis models have achieved remarkable results in practical applications. These models enable the analysis and processing of operational data from air conditioning systems, accurately identifying any abnormalities in their operation and thus predicting potential faults.

[0003] However, due to the wide distribution of air conditioning systems and the significant differences in operating conditions of air conditioning systems in different regions, the accuracy of general air conditioning fault diagnosis models for diagnosing faults in air conditioning systems under different operating conditions is low, while high-precision air conditioning fault diagnosis models have poor generalization, resulting in air conditioning fault detection models not being applicable to fault diagnosis of air conditioning systems under various operating conditions.

[0004] Therefore, improving the accuracy of air conditioner fault diagnosis is an urgent problem to be solved. Summary of the Invention

[0005] This application provides an air conditioning system and its control method to improve the accuracy of air conditioning fault diagnosis.

[0006] To achieve the above objectives, the embodiments of this application adopt the following technical solutions:

[0007] In a first aspect, an air conditioning system is provided, comprising:

[0008] The controller is configured as follows:

[0009] Obtain the current operating data of the air conditioning system;

[0010] The current operating data is input into the target fault diagnosis model, and the air conditioning system is diagnosed through the target fault diagnosis model to obtain the fault diagnosis results of the air conditioning system.

[0011] When the similarity between the normal operating data in the historical operating dataset and the source sample dataset is below a preset value, the target fault diagnosis model is trained in the following way:

[0012] The source fault diagnosis model is trained based on the historical operation dataset of the air conditioning system to obtain the target fault diagnosis model; the source sample data includes normal operation data and abnormal operation data of multiple air conditioning systems; the source fault diagnosis model is pre-trained based on the source sample data and is used to diagnose faults based on the operation data of the air conditioning system.

[0013] The technical solution provided in this application offers at least the following advantages: This application considers that when the similarity between the historical operating dataset and the normal operating data in the source sample dataset is below a preset value, it indicates that the historical operating data of the air conditioning system is dissimilar to the modeling data of the source fault diagnosis model. In this case, if the source fault diagnosis model is used to diagnose the air conditioning system, the resulting fault diagnosis may differ significantly from the actual fault condition of the air conditioning system. Therefore, this application trains the source fault diagnosis model based on the historical operating dataset of the air conditioning system to obtain a target fault diagnosis model that is more suitable for the current air conditioning system. Thus, diagnosing the air conditioning system using the target fault diagnosis model can improve the accuracy of the fault diagnosis results.

[0014] In some embodiments, the above-mentioned training of the source fault diagnosis model based on the historical operation dataset to obtain the target fault diagnosis model corresponding to the air conditioning system includes: obtaining sample labels corresponding to the historical operation data in the historical operation dataset; establishing a training sample set based on the historical operation dataset and the sample labels corresponding to the historical operation data in the historical operation dataset; and training the preset layer of the source fault diagnosis model based on the training sample set to obtain the target fault diagnosis model, wherein the preset layer is the last q layer in the p layer of the source fault diagnosis model, and p and q are positive integers.

[0015] In some embodiments, the sample labels corresponding to the historical operating data in the aforementioned historical operating dataset are determined by the following steps: based on the numerical range of the outdoor temperature values ​​in the historical operating data, the historical operating data is divided into multiple first-category historical operating data; based on the numerical range of the load rate in the first-category historical operating data, the first-category historical operating data is divided into multiple second-category historical operating data; based on the numerical range of the average power in the second-category historical operating data, the second-category historical operating data is divided into low-power mode sample data, medium-power mode sample data, and high-power mode sample data; the sample labels corresponding to the low-power mode sample data and the high-power mode sample data are determined to be abnormal labels, and the sample labels corresponding to the medium-power mode sample data are determined to be normal labels.

[0016] In some embodiments, when the similarity between the historical operating data of the air conditioning system and the normal sample data in the sample data is above a preset value, the target fault diagnosis model is the source fault diagnosis model.

[0017] In some embodiments, the above-mentioned source fault diagnosis model is obtained by the following steps: obtaining a source sample dataset; and training an initial source fault diagnosis model based on the source sample dataset to obtain a source fault diagnosis model.

[0018] Secondly, embodiments of this application provide a control method for an air conditioning system, the method comprising:

[0019] Obtain the current operating data of the air conditioning system;

[0020] The current operating data is input into the target fault diagnosis model, and the air conditioning system is diagnosed through the target fault diagnosis model to obtain the fault diagnosis results of the air conditioning system.

[0021] When the similarity between the normal operating data in the historical operating dataset and the source sample dataset is below a preset value, the target fault diagnosis model is trained in the following way:

[0022] The source fault diagnosis model is trained based on the historical operation dataset of the air conditioning system to obtain the target fault diagnosis model; the source sample data includes normal operation data and abnormal operation data of multiple air conditioning systems; the source fault diagnosis model is pre-trained based on the source sample data and is used to diagnose faults based on the operation data of the air conditioning system.

[0023] Thirdly, embodiments of this application provide a controller, including: one or more processors; one or more memories; wherein the one or more memories are used to store computer program code, the computer program code including computer instructions, and when the one or more processors execute the computer instructions, the controller executes the control method of the air conditioning system provided in the second aspect.

[0024] Fourthly, embodiments of this application provide a computer-readable storage medium including computer instructions that, when controlled on a computer, cause the computer to execute the control method for the air conditioning system provided in the second aspect and possible implementations.

[0025] Fifthly, embodiments of the present invention provide a computer program product that can be directly loaded into a memory and contains software code. After being loaded and executed by a computer, the computer program product can implement the control method of the air conditioning system provided in the second aspect and possible implementations.

[0026] It should be noted that the aforementioned computer instructions may be stored, in whole or in part, on a computer-readable storage medium. This computer-readable storage medium may be packaged together with the controller's processor, or it may be packaged separately from the controller's processor; this application does not impose any limitations on this.

[0027] The beneficial effects described in aspects two through five of this application can be referred to the analysis of the beneficial effects of aspect one, and will not be repeated here. Attached Figure Description

[0028] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of the present invention and do not constitute a limitation on the technical solutions of the present invention.

[0029] Figure 1 A schematic diagram of an air conditioning system provided in this application embodiment. Figure 1 ;

[0030] Figure 2 A schematic diagram of an air conditioning system provided in this application embodiment. Figure 2 ;

[0031] Figure 3 A schematic diagram of an air conditioning system provided in this application embodiment. Figure 3 ;

[0032] Figure 4 A schematic diagram of an air conditioning system provided in this application embodiment. Figure 4 ;

[0033] Figure 5 A schematic diagram of an air conditioning system provided in this application embodiment. Figure 5 ;

[0034] Figure 6 A schematic diagram of the hardware structure of a controller provided in an embodiment of this application;

[0035] Figure 7 A schematic diagram of model training provided for an embodiment of this application;

[0036] Figure 8 This is a schematic diagram illustrating an application scenario of an air conditioning system provided in an embodiment of this application;

[0037] Figure 9 This is a schematic diagram of the structure of a cloud server provided in an embodiment of this application;

[0038] Figure 10 A flowchart illustrating a control method for an air conditioning system provided in an embodiment of this application;

[0039] Figure 11 This is a schematic diagram illustrating the interaction between an air conditioning system and a cloud server, provided as an embodiment of this application. Detailed Implementation

[0040] 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, and 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.

[0041] In the description of this invention, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0042] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0043] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0044] The terms “comprising” and “having”, and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0045] Furthermore, in the embodiments of this application, the words "exemplary" or "for example" are used to indicate that they are examples, illustrations, or descriptions. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0046] As described in the background section, with continuous technological advancements, air conditioning fault diagnosis models have achieved significant results in practical applications. These models allow for the analysis and processing of operating data from air conditioning systems, accurately identifying whether abnormalities have occurred and enabling fault prediction. However, due to the wide distribution of air conditioning systems and the significant differences in operating conditions across different regions, general-purpose fault diagnosis models have low accuracy in diagnosing faults under various conditions. Furthermore, high-precision fault diagnosis models suffer from poor generalization, rendering them unsuitable for diagnosing faults in air conditioning systems under diverse operating conditions. Therefore, improving the accuracy of air conditioning fault diagnosis is a pressing issue that needs to be addressed.

[0047] In view of this, this application considers that when the similarity between the historical operating dataset and the normal operating data in the source sample dataset is below a preset value, it indicates that the historical operating data of the air conditioning system is dissimilar to the modeling data of the source fault diagnosis model. In this case, if the source fault diagnosis model is used to diagnose the air conditioning system, the resulting fault diagnosis may differ significantly from the actual fault condition of the air conditioning system. Therefore, this application trains the source fault diagnosis model based on the historical operating dataset of the air conditioning system to obtain a target fault diagnosis model more suitable for the current air conditioning system. Thus, using the target fault diagnosis model to diagnose the air conditioning system can improve the accuracy of the fault diagnosis results.

[0048] The air conditioning system provided in this application embodiment can be a split-type wall-mounted air conditioning system, a split-type floor-standing air conditioning system, a ceiling-mounted air conditioning system, an embedded air conditioning system, a central air conditioning system, or a multi-split air conditioning system. This application embodiment does not impose any restrictions on this. The number of air conditioning systems can be one, two, or more; each air conditioning system can have one, two, or more outdoor units. This application embodiment does not impose any restrictions on this.

[0049] To further describe the technical solutions of the embodiments of this application, as follows: Figure 1 The diagram shown is a structural schematic of an air conditioning system provided in an embodiment of this application.

[0050] like Figure 1 As shown, the air conditioning system 1 includes at least one outdoor unit 10 and at least one indoor unit 20.

[0051] The following is a detailed explanation of the settings and functions of each component of the outdoor unit.

[0052] In some embodiments, such as Figure 2As shown, the outdoor unit 10 includes an outdoor heat exchanger 101. One end of the outdoor heat exchanger 101 is connected to the compressor via a four-way reversing valve, and the other end is connected to each indoor heat exchanger via connecting pipes. The outdoor heat exchanger 101 is used to facilitate heat exchange between the refrigerant flowing in the heat transfer tubes of the outdoor heat exchanger 101 and the outdoor air.

[0053] In some embodiments, such as Figure 3 As shown, the outdoor unit 10 also includes a compressor 102. The compressor 102 is located between each indoor heat exchanger and the outdoor heat exchanger 101, and is used to provide power for the refrigerant circulation. Taking the refrigeration cycle as an example, the compressor 102 delivers the compressed refrigerant to the outdoor heat exchanger 101 via a four-way reversing valve 103.

[0054] In some embodiments, such as Figure 4 As shown, the outdoor unit 10 also includes a four-way reversing valve 103. The four ports of the four-way reversing valve 103 are respectively connected to the exhaust port of the compressor 102, the outdoor heat exchanger 101, the suction port of the compressor 102, and each indoor heat exchanger. The four-way reversing valve 103 is used to switch between cooling mode and heating mode by changing the flow direction of refrigerant in the system piping.

[0055] In some embodiments, continue as follows Figure 4 As shown, the outdoor unit 10 also includes an outdoor throttling device 104. The outdoor throttling device 104 is located between the outdoor heat exchanger 101 and each indoor heat exchanger, and has the effect of throttling the refrigerant flowing through it to reduce pressure, thereby regulating the refrigerant flow rate in the refrigerant path. Optionally, the outdoor throttling device 104 can be an electronic expansion valve. If the outdoor throttling device 104 decreases its opening, the flow resistance of the refrigerant through it increases. If the outdoor throttling device 104 increases its opening, the flow resistance decreases. Thus, even if the states of other components in the circuit remain unchanged, the refrigerant flow rate to the indoor heat exchanger 201 or the outdoor heat exchanger 101 will change when the opening of the outdoor throttling device 104 changes.

[0056] In some embodiments, continue as follows Figure 4 As shown, the outdoor unit 10 also includes a gas-liquid separator 105. The gas-liquid separator 105 is connected to the suction port of the compressor 102 and is used to contain the refrigerant in the liquid return section of the refrigerant passage to prevent liquid slugging of the compressor 102.

[0057] The following is a detailed explanation of the settings and functions of each component of the indoor unit.

[0058] In some embodiments, such as Figure 5As shown, the indoor unit 20 includes an indoor heat exchanger 201. The indoor heat exchanger 201 is used to exchange heat between the refrigerant flowing in the heat transfer tubes of the indoor heat exchanger 201 and the indoor air.

[0059] In some embodiments, continue as follows Figure 5 As shown, the indoor unit 20 also includes an indoor throttling device 202. The indoor throttling device 202 is connected to the indoor heat exchanger 201 and is used to throttle the refrigerant flowing through it, thereby reducing pressure and regulating the refrigerant flow rate in the refrigerant path. Optionally, the indoor throttling device 202 can be an electronic expansion valve.

[0060] In this application, the air conditioning system executes a refrigeration cycle by using a compressor 102, an outdoor heat exchanger 101, a throttling device 104, an indoor heat exchanger 201, and a four-way reversing valve 103 as a refrigerant circulation loop. The refrigeration cycle includes a series of processes involving compression, condensation, expansion, and evaporation, and supplies refrigerant to the conditioned and heat-exchanged air.

[0061] Indoor heat exchanger 201 and outdoor heat exchanger 101 are used as condensers or evaporators. When indoor heat exchanger 201 is used as a condenser, air conditioning system 1 is used as a heater in heating mode. When indoor heat exchanger 201 is used as an evaporator, air conditioning system 1 is used as a cooler in cooling mode.

[0062] In some embodiments, the air conditioning system 1 further includes a controller 300. For example... Figure 6 As shown, the controller 300 is electrically connected to the outdoor heat exchanger 101, the compressor 102, the four-way reversing valve 103, the outdoor throttling device 104, the gas-liquid separator 105, the indoor heat exchanger 201, and the indoor throttling device 202.

[0063] In some embodiments, the controller 300 may be used to perform the following steps S1-S2:

[0064] S1. Obtain the current operating data of the air conditioning system.

[0065] Optionally, the current operating data of the air conditioning system may include the outdoor temperature at the location of the air conditioning system, the load rate of the air conditioning system, and the operating power of the air conditioning system.

[0066] As one possible implementation, the air conditioning system 1 has its operating data stored in its storage device, and the controller 300 can obtain the current operating data of the air conditioning system 1 from the storage device.

[0067] S2. Input the current operating data into the target fault diagnosis model, and use the target fault diagnosis model to diagnose the air conditioning system and obtain the fault diagnosis results of the air conditioning system.

[0068] In this case, when the similarity between the normal operation data in the historical operation dataset and the source sample dataset is below a preset value, the target fault diagnosis model is trained by the following steps: the source fault diagnosis model is trained based on the historical operation dataset of the air conditioning system to obtain the target fault diagnosis model.

[0069] The source sample data includes normal operation data and abnormal operation data corresponding to multiple air conditioning systems; the source fault diagnosis model is pre-trained based on the source sample data and is used to diagnose faults based on the operation data of the air conditioning system.

[0070] Optionally, the historical operating dataset of the air conditioning system can be the operating data of the air conditioning system over a period of time, such as a year's operating data. The operating data may include the outdoor temperature at the location of the air conditioning system, the load rate of the air conditioning system, and the operating power of the air conditioning system.

[0071] As one possible implementation, the air conditioning system 1 has its operating data stored in its storage device, and the controller 300 can obtain the historical operating data of the air conditioning system 1 from the storage device.

[0072] As another possible implementation, the air conditioning system 1 is connected to a cloud server, which stores the operating data of the air conditioning system 1. The controller 300 can obtain the historical operating data of the air conditioning system 1 from the cloud server.

[0073] As one possible approach, similarity is determined by the Maximum Mean Discrepancy (MMD), which can be calculated using the following formula:

[0074]

[0075] Where X is the historical running dataset; Y is the set of normally running data in the source sample dataset; k is the kernel function; and m, n, i, and j are positive integers.

[0076] Optionally, the kernel function can be a Gaussian kernel function.

[0077] Understandably, the smaller the MMD, the higher the similarity between the two datasets; conversely, the larger the MMD, the lower the similarity between the two datasets.

[0078] Optionally, the preset value is P. If 0 ≤ MMD ≤ P, it indicates that the historical operating data of the air conditioning system is similar to the normal operating data in the source sample dataset; if MMD > P, it indicates that the historical operating data of the air conditioning system is not similar to the normal operating data in the source sample dataset.

[0079] For example, 0 < P < 10, and P and Q are positive integers.

[0080] As another possible implementation, similarity can be determined using the Jaccard similarity coefficient.

[0081] As can be seen from the above embodiments, when the similarity between the historical operating dataset and the normal operating data in the source sample dataset is below a preset value, it indicates that the historical operating data of the air conditioning system is dissimilar to the modeling data of the source fault diagnosis model. In this case, if the source fault diagnosis model is used to diagnose the air conditioning system, the resulting fault diagnosis may differ significantly from the actual fault condition of the air conditioning system. Therefore, this application trains the source fault diagnosis model based on the historical operating dataset of the air conditioning system to obtain a target fault diagnosis model more suitable for the current air conditioning system. Thus, using the target fault diagnosis model to diagnose the air conditioning system can improve the accuracy of the fault diagnosis results.

[0082] In some embodiments, the above-mentioned target fault diagnosis model can be obtained by the controller 300 by specifically executing the following steps S21-S23:

[0083] S21. Obtain the sample labels corresponding to the historical running data in the historical running dataset.

[0084] As one possible implementation, the sample labels corresponding to the historical running data in the aforementioned historical running dataset are determined by the controller 300 by performing the following steps S211-S214:

[0085] S211. Based on the range of outdoor temperature (OT) values ​​in historical operating data, the historical operating data is divided into multiple first-category historical operating data.

[0086] For example, based on the outdoor temperature value (OT), the historical operating data is divided into eight categories: [-∞℃, -20℃], [-20℃, -10℃], [-10℃, 0℃], [0℃, 10℃], [10℃, 20℃], [20℃, 30℃], [30℃, 40℃], and [40℃, ∞℃].

[0087] S212. Based on the numerical range of the load rate in the historical operating data of the first category, divide the historical operating data of the first category into multiple historical operating data of the second category.

[0088] For example, based on the load factor (LR), the historical operating data of each first category is divided into four second categories of historical operating data: [0, 25%], [25%, 50%], [50%, 75%], and [75%, 100%].

[0089] S213. Based on the numerical range of the average power in the historical operating data of the second category, the historical operating data of the second category is divided into low power mode sample data, medium power mode sample data and high power mode sample data.

[0090] As one possible implementation, based on the power mean of the historical running data of each group of the second category, the 32 groups of historical running data of the second category are clustered into low power mode sample data, medium power mode sample data and high power mode sample data using the k-means clustering algorithm.

[0091] As another possible implementation, based on the power mean of the historical running data of each group of the second category, the K-medoids clustering algorithm is used to cluster the 32 groups of historical running data of the second category into low power mode sample data, medium power mode sample data and high power mode sample data.

[0092] S214. Determine that the sample labels corresponding to the low-power mode sample data and the high-power mode sample data are abnormal labels, and the sample labels corresponding to the medium-power mode sample data are normal labels.

[0093] As shown in the above embodiments, outdoor temperature and air conditioning load rate can reflect the operating condition of the air conditioning system. Historical operating data is divided and categorized into different intervals based on different outdoor temperature and load rates, with each interval representing historical operating data under similar conditions. This makes the sample labels corresponding to historical data under different operating conditions more accurate and consistent. Simultaneously, based on the average power in the historical operating data within different intervals, the second category of historical operating data is clustered into low-power mode sample data, medium-power mode sample data, and high-power mode sample data. The low-power and high-power mode sample data reflect malfunctions in the air conditioning system under corresponding outdoor temperature and load rate conditions, while the medium-power mode sample data reflects normal operation of the air conditioning system under corresponding outdoor temperature and load rate conditions.

[0094] S22. Establish a training sample set based on the historical running dataset and the sample labels corresponding to the historical running data in the historical running dataset.

[0095] Optionally, a set of historical running data corresponds to a sample label.

[0096] S23. Based on the training sample set, train the preset layer of the source fault diagnosis model to obtain the target fault diagnosis model.

[0097] For example, such as Figure 7As shown, the source fault diagnosis model includes: layer 1, ..., layer pq, and preset layer; the target fault diagnosis model also includes: layer 1, ..., layer pq, and preset layer. Layers 1 to pq of the target fault diagnosis model are copied from layers 1 to pq of the source fault diagnosis model. The last q layers (preset layers) of the target fault diagnosis model are trained based on the training sample set, i.e., fine-tuned q layers.

[0098] If the source fault diagnosis model has p layers, the preset layer can be the last 3 layers of the source fault diagnosis model.

[0099] As can be seen from the above embodiments, by training the last q layers in the p layer of the source fault diagnosis model, that is, freezing the first (pq) layers of the source fault diagnosis model, on the one hand, the general features learned by the first (pq) layers in the training data can be preserved, and on the other hand, training only the last q layers can reduce the training parameters and improve the training speed of the model.

[0100] In some embodiments, when the similarity between the historical operating data of the air conditioning system and the normal sample data in the sample data is above a preset value, the target fault diagnosis model is the source fault diagnosis model.

[0101] As can be seen from the above embodiments, when the similarity between the historical operating dataset and the normal operating data in the source sample dataset is below a preset value, it indicates that the historical operating data of the air conditioning system is similar to the modeling data of the source fault diagnosis model. Therefore, there is no need to adjust the source fault diagnosis model; directly using the source fault diagnosis model to diagnose the air conditioning system will yield a relatively accurate diagnostic result.

[0102] As one possible implementation, the controller 300 trains a preset layer of the source fault diagnosis model based on a training sample set, including the following steps S231-S235:

[0103] S231. Input the historical operating data from the training sample set into the preset layer of the source fault diagnosis model. Obtain the fault diagnosis prediction results of the historical operating data in the training sample set.

[0104] S232. Determine the loss value based on the actual fault diagnosis results indicated by the sample labels corresponding to the historical operating data in the training sample set and the predicted fault diagnosis results of the historical operating data.

[0105] S233. Determine whether the source fault diagnosis model has converged based on the loss value of the training sample set.

[0106] S234. If not, then based on the loss value of the training sample set, update the weight parameters in the source fault diagnosis model, fine-tune the source fault diagnosis model, and continue to execute step S231 to input the historical running data in the training sample set into the source fault diagnosis model. Obtain the fault diagnosis prediction results of the historical running data in the training sample set.

[0107] S235. If so, use the fine-tuned source fault diagnosis model as the target fault diagnosis model.

[0108] In some embodiments, after obtaining the target fault diagnosis model, the controller 300 can also be used to test the target fault diagnosis model using a historical running dataset. It should be noted that the historical running data used to train the source fault diagnosis model is different from the historical running data used to test the target fault diagnosis model. For example, the historical running data used to train the source fault diagnosis model may be 80% of the running data in the historical running dataset, while the historical running data used to test the target fault diagnosis model may be the remaining 20% ​​of the data in the historical running dataset.

[0109] In some embodiments, the controller 300 can also be used to acquire a source sample dataset; and based on the source sample dataset, to train an initial source fault diagnosis model to obtain a source fault diagnosis model.

[0110] As one possible implementation, the controller 300 trains the initial source fault diagnosis model based on the source training sample set, including the following steps S5-S9:

[0111] S5. Input the historical operating data from the source training sample set into the initial source fault diagnosis model. Obtain the fault diagnosis prediction results of the historical operating data in the source training sample set.

[0112] S6. Determine the loss value based on the actual fault diagnosis results indicated by the sample labels corresponding to the historical operating data in the source training sample set and the predicted fault diagnosis results of the historical operating data.

[0113] S7. Determine whether the initial source fault diagnosis model has converged based on the loss value of the source training sample set.

[0114] S8. If not, update the weight parameters in the initial source fault diagnosis model based on the loss value of the source training sample set, and continue to execute step S5 to input the historical running data in the source training sample set into the initial source fault diagnosis model. Obtain the fault diagnosis prediction results of the historical running data in the source training sample set.

[0115] S9. If so, adjust the source fault diagnosis model.

[0116] The aforementioned controller 300 refers to a device that can generate operation control signals based on instruction opcodes and timing signals, instructing the air conditioning system 1 to execute control commands. Exemplarily, the controller 300 can be a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), a microprocessor, a microcontroller, a programmable logic device (PLD), or any combination thereof. The controller can also be other devices with processing functions, such as circuits, devices, or software modules; this application embodiment does not impose any limitations on this.

[0117] In some embodiments, the controller 300 can be a microcontroller unit (MCU). An MCU, also known as a single-chip microcomputer, is a chip-level computer that integrates a central processing unit (CPU) with appropriately reduced frequency and specifications, along with peripheral interfaces such as memory, timer, USB, A / D converter, UART, PLC, DMA, and even LCD driver circuitry, all onto a single chip. This allows for different combinations of control for various applications.

[0118] In addition, the controller 300 can be used to control the operation of each component in the air conditioning system 1 so that each component of the air conditioning system 1 can operate to achieve each predetermined function of the air conditioning system 1.

[0119] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the air conditioning system. In other embodiments of this application, the air conditioning system may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0120] Figure 8 This is a schematic diagram illustrating an application scenario of an air conditioning system provided in an embodiment of this application. For example... Figure 8 As shown, the application scenario includes multiple air conditioning systems, such as the first air conditioning system 100 and the second air conditioning system 200, as well as a cloud server 400.

[0121] The cloud server 400 has a communication connection with the first air conditioning system 100 and the second air conditioning system 200.

[0122] In some embodiments, the first air conditioning system 100 and the second air conditioning system 200 are devices for adjusting and controlling parameters such as temperature, humidity, and flow rate of the ambient air inside a building or structure.

[0123] In some embodiments, cloud server 400 may be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and big data servers. This application does not impose any special restrictions on the specific form of cloud server 400.

[0124] In some embodiments, the first air conditioning system 100 and the second air conditioning system 200 can send their own operating data to the cloud server 400 so that the cloud server 400 can diagnose the faults of the first air conditioning system 100 and the second air conditioning system 200 based on the operating data of the first air conditioning system 100 and the second air conditioning system 200.

[0125] Figure 9 The diagram shown is a structural schematic of a cloud server provided in this application. Figure 9 As shown, the cloud server 500 may include a communication device 501, a memory 502, and a processor 503.

[0126] In some embodiments, the communication device 501 is used to establish a communication connection with the air conditioning system, and through the communication device 501, the cloud server 500 can obtain the historical operating data of the air conditioning system.

[0127] In other embodiments, the communication device 501 can also be used to obtain the current operating data of the air conditioning system.

[0128] The communication device 501 may include a radio frequency (RF) module, a cellular module, a wireless fidelity (WIFI) module, and a GPS module, etc. Taking the RF module as an example, the RF module can be used for signal reception and transmission. Specifically, it sends the received information to the processor 503 for processing; in addition, it sends the signals generated by the processor 503 out. Typically, the RF circuit may include, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (LNA), a duplexer, etc.

[0129] In some embodiments, memory 502 can be used to store historical operating data of the air conditioning system.

[0130] In other embodiments, memory 502 may also be used to store source fault diagnosis models and source sample datasets.

[0131] The processor 503 executes various functions and data processing of the cloud server 500 by running software programs or data stored in the memory 502. The memory 502 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. The memory 502 stores an operating system that enables the processor 503 to run. In this application, the memory 502 may store the operating system and various application programs, and may also store code that executes the control method of the air conditioning system provided in the embodiments of this application.

[0132] In some embodiments, the processor 503 is used to acquire the current operating data of the air conditioning system; input the current operating data into the target fault diagnosis model, and perform fault diagnosis on the air conditioning system through the target fault diagnosis model to obtain the fault diagnosis result of the air conditioning system; when the similarity between the normal operating data in the historical operating dataset and the source sample dataset is below a preset value, the target fault diagnosis model is trained in the following way: the source fault diagnosis model is trained based on the historical operating dataset of the air conditioning system to obtain the target fault diagnosis model; the source sample data includes normal operating data and abnormal operating data corresponding to multiple air conditioning systems; the source fault diagnosis model is pre-trained based on the source sample data and is used to perform fault diagnosis based on the operating data of the air conditioning system.

[0133] The processor 503 executes the air conditioning system control method of this application and its beneficial effects can be referred to the steps and beneficial effects of the controller 300 executing the air conditioning system control method of this application, which will not be repeated here.

[0134] It should be noted that the cloud server 500 can establish communication connections with multiple air conditioning systems. It can perform fault diagnosis on the target air conditioning system in the cloud through the target fault diagnosis model, or it can send the target fault diagnosis model to the target air conditioning system, and the target air conditioning system can perform fault diagnosis on its own through the target fault diagnosis model.

[0135] Those skilled in the art will understand that Figure 9 The hardware structure shown does not constitute a limitation on cloud servers. Cloud servers may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0136] This application embodiment considers the wide distribution of air conditioning systems and the significant differences in operating conditions across different latitudes. A source fault detection model trained on limited labeled data in the cloud cannot meet the needs of fault diagnosis in various scenarios. This application embodiment stores the source fault diagnosis model only as a source model, then collects a small amount of labeled normal and fault data from air conditioning systems under different operating conditions and uploads it to the cloud for fine-tuning of the source model. This enables personalized training of fault diagnosis models for air conditioning systems in different types of regions, improving the generalization and versatility of the source fault diagnosis model. Finally, the fine-tuned target fault diagnosis model under different operating conditions is uploaded and deployed on the target air conditioning system to improve the accuracy and generalization of the fault diagnosis model.

[0137] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0138] like Figure 10 As shown in the figure, this application provides a control method for an air conditioning system, applied to a cloud server processor. The method includes the following steps:

[0139] S101. Obtain the current operating data of the air conditioning system.

[0140] S102. Input the current operating data into the target fault diagnosis model, and perform fault diagnosis on the air conditioning system through the target fault diagnosis model to obtain the fault diagnosis result of the air conditioning system.

[0141] In some embodiments, when the similarity between the normal operation data in the historical operation dataset and the source sample dataset is below a preset value, the target fault diagnosis model is trained in the following way: the source fault diagnosis model is trained based on the historical operation dataset of the air conditioning system to obtain the target fault diagnosis model; the source sample data includes normal operation data and abnormal operation data corresponding to multiple air conditioning systems; the source fault diagnosis model is pre-trained based on the source sample data and is used to diagnose faults based on the operation data of the air conditioning system.

[0142] In other embodiments, when the similarity between the historical operating data of the air conditioning system and the normal sample data in the sample data is above a preset value, the target fault diagnosis model is the source fault diagnosis model.

[0143] As one possible implementation, the above-mentioned training of the source fault diagnosis model based on the historical operation dataset to obtain the target fault diagnosis model corresponding to the air conditioning system can be specifically implemented in the following steps: obtaining the sample labels corresponding to the historical operation data in the historical operation dataset; establishing a training sample set based on the historical operation dataset and the sample labels corresponding to the historical operation data in the historical operation dataset; and training the preset layer of the source fault diagnosis model based on the training sample set to obtain the target fault diagnosis model.

[0144] In some embodiments, the sample labels corresponding to the historical operating data in the historical operating dataset are determined by the following steps: based on the numerical range of the outdoor temperature values ​​in the historical operating data, the historical operating data is divided into multiple first categories of historical operating data; based on the numerical range of the load rate in the first category of historical operating data, the historical operating data in the first category is divided into multiple second categories of historical operating data; based on the numerical range of the average power in the second category of historical operating data, the historical operating data in the second category is divided into low-power mode sample data, medium-power mode sample data, and high-power mode sample data; the sample labels corresponding to the low-power mode sample data and the high-power mode sample data are determined to be abnormal labels, and the sample labels corresponding to the medium-power mode sample data are determined to be normal labels.

[0145] Figure 10 The illustrated embodiments offer at least the following advantages: This application considers that when the similarity between the historical operating dataset and the normal operating data in the source sample dataset is below a preset value, it indicates that the historical operating data of the air conditioning system is dissimilar to the modeling data of the source fault diagnosis model. In this case, if the source fault diagnosis model is used to diagnose the air conditioning system, the resulting fault diagnosis may differ significantly from the actual fault condition of the air conditioning system. Therefore, this application trains the source fault diagnosis model based on the historical operating dataset of the air conditioning system to obtain a target fault diagnosis model more suitable for the current air conditioning system. Thus, using the target fault diagnosis model to diagnose the air conditioning system can improve the accuracy of the fault diagnosis results.

[0146] In some embodiments, the above-mentioned source fault diagnosis model is obtained by the following steps: obtaining a source sample dataset; and training an initial source fault diagnosis model based on the source sample dataset to obtain a source fault diagnosis model.

[0147] Figure 11 This application provides an interactive diagram of an air conditioning system and a cloud server. The following is a combination of... Figure 11 The control method of the air conditioning system of this application is described in detail.

[0148] The cloud server stores source sample data and source fault diagnosis models.

[0149] Determine whether the historical operating data of the air conditioning system is similar to the source sample data;

[0150] If the historical operating data of the air conditioning system is similar to the source sample data, then the air conditioning system is diagnosed using the source fault diagnosis model.

[0151] If the historical operating data of the air conditioning system is not similar to the source sample data, the source fault diagnosis model is fine-tuned, and the air conditioning system is diagnosed through the fine-tuned source fault diagnosis model.

[0152] As can be seen, the above mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, the embodiments of this application provide corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the modules and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software-driven hardware manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0153] This application embodiment can divide the controller into functional modules according to the above method example. For example, each function can be divided into a separate functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0154] This application also provides a computer-readable storage medium including computer-executable instructions that, when run on a computer, cause the computer to execute any of the air conditioning system control methods provided in the above embodiments.

[0155] This application also provides a computer program product containing computer execution instructions, which, when run on a computer, causes the computer to execute any of the air conditioning system control methods provided in the above embodiments.

[0156] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer-executable instructions. When these computer-executable instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer-executable instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer-executable instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks, SSDs).

[0157] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0158] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely illustrative descriptions of the application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

[0159] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An air conditioning system, characterized in that, include: The controller is configured as follows: Obtain the current operating data of the air conditioning system; The current operating data is input into the target fault diagnosis model, and the air conditioning system is diagnosed through the target fault diagnosis model to obtain the fault diagnosis result of the air conditioning system. When the similarity between the historical operating dataset and the source sample dataset and the normal operating data is below a preset value, the target fault diagnosis model is trained in the following way: The source fault diagnosis model is trained based on the historical operation dataset of the air conditioning system to obtain the target fault diagnosis model; the source sample data includes normal operation data and abnormal operation data corresponding to multiple air conditioning systems; the source fault diagnosis model is pre-trained based on the source sample data and is used to diagnose faults based on the operation data of the air conditioning system.

2. The air conditioning system according to claim 1, characterized in that, The step of training the source fault diagnosis model based on the historical operating dataset to obtain the target fault diagnosis model corresponding to the air conditioning system includes: Obtain the sample labels corresponding to the historical running data in the historical running dataset; A training sample set is established based on the historical running dataset and the sample labels corresponding to the historical running data in the historical running dataset; Based on the training sample set, the preset layer of the source fault diagnosis model is trained to obtain the target fault diagnosis model.

3. The air conditioning system according to claim 2, characterized in that, The sample labels corresponding to the historical running data in the historical running dataset are determined by the following steps: Based on the range of outdoor temperature values ​​in the historical operation data, the historical operation data is divided into multiple first categories of historical operation data. Based on the numerical range of the load rate in the historical operating data of the first category, the historical operating data of the first category is divided into multiple historical operating data of the second category; Based on the numerical range of the average power in the historical operating data of the second category, the historical operating data of the second category is divided into low power mode sample data, medium power mode sample data and high power mode sample data. The sample labels corresponding to the low-power mode sample data and the high-power mode sample data are determined to be abnormal labels, while the sample labels corresponding to the medium-power mode sample data are determined to be normal labels.

4. The air conditioning system according to claim 1, characterized in that, If the similarity between the historical operating data and the normal sample data of the air conditioning system is above a preset value, the target fault diagnosis model is the source fault diagnosis model.

5. The air conditioning system according to claim 1, characterized in that, The source fault diagnosis model is obtained through the following steps: Obtain the source sample dataset; Based on the source sample dataset, the initial source fault diagnosis model is trained to obtain the source fault diagnosis model.

6. A control method for an air conditioning system, characterized in that, include: Obtain the current operating data of the air conditioning system; The current operating data is input into the target fault diagnosis model, and the air conditioning system is diagnosed through the target fault diagnosis model to obtain the fault diagnosis result of the air conditioning system. When the similarity between the historical operating dataset and the source sample dataset and the normal operating data is below a preset value, the target fault diagnosis model is trained in the following way: The source fault diagnosis model is trained based on the historical operation dataset of the air conditioning system to obtain the target fault diagnosis model; the source sample data includes normal operation data and abnormal operation data corresponding to multiple air conditioning systems; the source fault diagnosis model is pre-trained based on the source sample data and is used to diagnose faults based on the operation data of the air conditioning system.

7. The method according to claim 6, characterized in that, The step of training the source fault diagnosis model based on the historical operating dataset to obtain the target fault diagnosis model corresponding to the air conditioning system includes: Obtain the sample labels corresponding to the historical running data in the historical running dataset; A training sample set is established based on the historical running dataset and the sample labels corresponding to the historical running data in the historical running dataset; Based on the training sample set, the preset layer of the source fault diagnosis model is trained to obtain the target fault diagnosis model, wherein the preset layer is the source fault diagnosis model.

8. The method according to claim 7, characterized in that, The sample labels corresponding to the historical running data in the historical running dataset are determined by the following steps: Based on the range of outdoor temperature values ​​in the historical operation data, the historical operation data is divided into multiple first categories of historical operation data. Based on the numerical range of the load rate in the historical operating data of the first category, the historical operating data of the first category is divided into multiple historical operating data of the second category; Based on the numerical range of the average power in the historical operating data of the second category, the historical operating data of the second category is divided into low power mode sample data, medium power mode sample data and high power mode sample data. The sample labels corresponding to the low-power mode sample data and the high-power mode sample data are determined to be abnormal labels, while the sample labels corresponding to the medium-power mode sample data are determined to be normal labels.

9. The method according to claim 6, characterized in that, If the similarity between the historical operating data and the normal sample data of the air conditioning system is above a preset value, the target fault diagnosis model is the source fault diagnosis model.

10. The method according to claim 6, characterized in that, The source fault diagnosis model is obtained through the following steps: Obtain the source sample dataset; Based on the source sample dataset, the initial source fault diagnosis model is trained to obtain the source fault diagnosis model.

Images