Air conditioner and control method thereof
By dynamically optimizing the hyperparameters of the air conditioner fault detection model, and using preset relationship adjustment models to adapt to different fault conditions, the problem of manual and insufficient model generalization capabilities of hyperparameter adjustment in the existing technology is solved, and more accurate and flexible fault detection is achieved.
Patent Information
- Application Number
- CN202311593781.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-24
- Publication Date
- 2025-05-27
AI Technical Summary
The existing air conditioner fault detection model needs to manually adjust hyperparameters when facing different fault conditions, which limits the accuracy and generalization capabilities of the model and cannot adapt to the complex working environment and working conditions of the air conditioner.
By obtaining the hyperparameters preset by the fault detection model, the operating data of the air conditioner and the preset relationship between the hyperparameters and performance indicators, the hyperparameters are dynamically optimized to adjust the fault detection model, so that it can adapt to various fault types more intelligently and flexibly.
The automatic hyperparameter optimization of the air conditioner fault detection model is realized, which broadens the scope of application of fault detection, so that the air conditioner can accurately detect faults in various actual scenarios, and improves the intelligence level and stability of fault detection.
Smart Images
Figure CN120043204A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of air conditioners, and particularly to an air conditioner and its control method. Background Art
[0002] With the development of the economic society, air conditioners are increasingly widely used in various places such as entertainment, home, and work.
[0003] In recent years, with the complexity of the air conditioner structure and the diversification of application scenarios, the types of faults that occur in air conditioners also show an increasingly diverse trend. In order to improve the accuracy and efficiency of air conditioner fault diagnosis, researchers have become a current research hotspot by using methods such as machine learning and data mining to construct fault detection models.
[0004] However, the current fault detection models often need to manually adjust hyperparameters when facing different fault situations, which limits the accuracy and generalization ability of the models, thus affecting their performance in actual applications. Due to the certain complexity of the working environment and operating conditions of air conditioners, traditional fixed hyperparameter settings often cannot adapt to all situations. Therefore, it is necessary to solve the problem of how to automatically adjust hyperparameters to improve the intelligent level and generalization ability of the fault detection models. Summary of the Invention
[0005] This application provides an air conditioner and its control method for expanding the applicable range of air conditioner fault detection.
[0006] To achieve the above object, this application adopts the following technical solutions.
[0007] In a first aspect, an embodiment of this application provides an air conditioner, which includes: at least one outdoor unit; at least one indoor unit; a controller configured to: obtain a first hyperparameter preset by a fault detection model, operation data of the air conditioner, and a preset relationship between the hyperparameters of the fault detection model and the performance indicators of the fault detection model; determine a target hyperparameter according to the preset relationship to optimize the first hyperparameter; adjust the fault detection model according to the target hyperparameter to obtain an adjusted target fault detection model, and the target fault detection model is used to perform fault detection on the air conditioner based on the operation data of the air conditioner.
[0008] The technical solution provided by the embodiments of the present application at least brings the following beneficial effects: The embodiments of the present application provide an air conditioner, which can dynamically optimize the first hyperparameter of a fault detection model based on a preset relationship between the hyperparameters of the fault detection model and the performance indicators of the fault detection model. In this way, the fault detection model can be adjusted and improved under different fault conditions, making the fault detection model more intelligent and flexible to adapt to various fault types. Thus, by performing dynamic optimization according to the above relationship, the applicable range of air conditioner fault detection is broadened, enabling the air conditioner to accurately detect faults under various actual scenarios.
[0009] In some embodiments, the performance indicator includes a first accuracy rate or a second accuracy rate; wherein, the performance indicator is used to indicate the model performance when the fault detection model detects the operation data set of the air conditioner; the operation data set includes multiple samples obtained based on the operation data of the air conditioner, and the multiple samples include normal samples and fault samples; the first accuracy rate is the ratio of the number of detection results indicating that the air conditioner is normal to the number of normal samples; the second accuracy rate is the ratio of the number of detection results indicating that the air conditioner has a fault to the number of fault samples; the preset relationship includes a positive correlation between the hyperparameters of the fault detection model and the first accuracy rate, or a negative correlation between the hyperparameters of the fault detection model and the second accuracy rate.
[0010] In some embodiments, the controller is configured to determine the target hyperparameter according to the preset relationship to optimize the first hyperparameter, and is specifically configured to: obtain the model information of the air conditioner; determine the value range of the hyperparameter corresponding to the model information according to the model information; find the target hyperparameter within the value range according to the preset relationship to optimize the first hyperparameter.
[0011] In some embodiments, the controller is configured to find the target hyperparameter within the value range according to the preset relationship and a preset search algorithm. The controller is specifically configured to: obtain the operation data of the air conditioner in the historical time period; selection step: select a second hyperparameter from the value range and adjust the fault detection model based on the second hyperparameter to obtain an adjusted first fault detection model; detection step: input the operation data of the air conditioner in the historical time period into the first fault detection model to obtain an expected performance indicator; range update step: in the case that the expected performance indicator is outside the preset range, determine a new value range of the hyperparameter based on the second hyperparameter and the preset relationship; repeatedly execute the selection step, the detection step, and the range update step until the obtained expected performance indicator is within the preset range; determine the second hyperparameter of the second fault detection model as the target hyperparameter; wherein, the expected performance indicator obtained based on the second fault detection model is within the preset range.
[0012] In some embodiments, the fault detection model is a fault detection model established based on a Bayesian network.
[0013] In a second aspect, an embodiment of the present application provides a control method for an air conditioner. The method is applied to the air conditioner and includes: obtaining a first hyperparameter preset by the fault detection model, the operating data of the air conditioner, and a preset relationship between the hyperparameters of the fault detection model and the performance indicators of the fault detection model; determining target hyperparameters according to the preset relationship to optimize the first hyperparameters; and adjusting the fault detection model according to the target hyperparameters to obtain an adjusted target fault detection model, where the target fault detection model is used to perform fault detection on the air conditioner based on the operating data of the air conditioner.
[0014] In some embodiments, the performance indicators include a first accuracy rate or a second accuracy rate; wherein, the performance indicators are used to indicate the model performance when the fault detection model detects the operating data set of the air conditioner; the operating data set includes multiple samples obtained based on the operating data of the air conditioner, and the multiple samples include normal samples and fault samples; the first accuracy rate is the ratio of the number of detection results indicating that the air conditioner is normal to the number of normal samples; the second accuracy rate is the ratio of the number of detection results indicating that the air conditioner has a fault to the number of fault samples; the preset relationship includes a positive correlation between the hyperparameters of the fault detection model and the first accuracy rate, or a negative correlation between the hyperparameters of the fault detection model and the second accuracy rate.
[0015] In some embodiments, determining the target hyperparameters according to the preset relationship to optimize the first hyperparameters includes: obtaining the model information of the air conditioner; determining the value range of the hyperparameters corresponding to the model information according to the model information; and finding the target hyperparameters within the value range according to the preset relationship to optimize the first hyperparameters.
[0016] In some embodiments, finding the target hyperparameters within the value range according to the preset relationship includes: obtaining the operating data of the air conditioner in a historical time period; a selection step: selecting a second hyperparameter from the value range and adjusting the fault detection model based on the second hyperparameter to obtain an adjusted first fault detection model; a detection step: inputting the operating data of the air conditioner in the historical time period into the first fault detection model to obtain an expected performance indicator; a range update step: in the case where the expected performance indicator is outside the preset range, determining a new value range of the hyperparameters based on the second hyperparameter and the preset relationship; repeatedly executing the selection step, the detection step, and the range update step until the obtained expected performance indicator is within the preset range; and determining the second hyperparameter of the second fault detection model as the target hyperparameter; wherein, the expected performance indicator obtained based on the second fault detection model is within the preset range.
[0017] In some embodiments, the fault detection model is a fault detection model established based on a Bayesian network.
[0018] In a third aspect, an embodiment of the present application provides a controller, including: one or more processors; one or more memories; wherein, the one or more memories are used to store computer program code, and the computer program code includes computer instructions. When the one or more processors execute the computer instructions, the controller executes any one of the air conditioner control methods provided in the second aspect.
[0019] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which includes computer instructions. When the computer instructions run on a computer, the computer is caused to execute any one of the air conditioner control methods provided in the second aspect.
[0020] In a fifth aspect, an embodiment of the present invention provides a computer program product, which 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 any one of the air conditioner control methods provided in the second aspect.
[0021] It should be noted that the above computer instructions can be stored in whole or in part on a computer-readable storage medium. Among them, the computer-readable storage medium can be packaged together with the processor of the controller or separately packaged from the processor of the controller. The present application does not make any limitation in this regard.
[0022] For the beneficial effects described in the second to fifth aspects of the present application, reference can be made to the analysis of the beneficial effects in the first aspect, and details will not be elaborated here. Description of the Drawings
[0023] The drawings are used to provide a further understanding of the technical solutions of the present invention, and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solutions of the present invention and do not constitute a limitation to the technical solutions of the present invention.
[0024] Figure 1 It is a schematic diagram of the composition of an air conditioner provided by an embodiment of the present application;
[0025] Figure 2 It is a schematic structural diagram of an air conditioner provided by an embodiment of the present application;
[0026] Figure 3 It is a hardware configuration block diagram of an air conditioner provided by an embodiment of the present application;
[0027] Figure 4 It is a flowchart of an air conditioner control method provided by an embodiment of the present application;
[0028] Figure 5Schematic diagram of a preset relationship between hyperparameters of a fault detection model provided by an embodiment of the present application and performance indicators of the fault detection model;
[0029] Figure 6 Schematic diagram of another preset relationship between hyperparameters of a fault detection model provided by an embodiment of the present application and performance indicators of the fault detection model;
[0030] Figure 7 Schematic diagram of another preset relationship between hyperparameters of a fault detection model provided by an embodiment of the present application and performance indicators of the fault detection model;
[0031] Figure 8 Schematic diagram of another preset relationship between hyperparameters of a fault detection model provided by an embodiment of the present application and performance indicators of the fault detection model;
[0032] Figure 9 Schematic diagram of another preset relationship between hyperparameters of a fault detection model provided by an embodiment of the present application and performance indicators of the fault detection model;
[0033] Figure 10 Schematic diagram of another preset relationship between hyperparameters of a fault detection model provided by an embodiment of the present application and performance indicators of the fault detection model;
[0034] Figure 11 Schematic diagram of a fault detection model provided by an embodiment of the present application in three - dimensional space;
[0035] Figure 12 Flowchart of another control method for an air conditioner provided by an embodiment of the present application;
[0036] Figure 13 Flowchart of a process from establishing a fault detection model of an air conditioner to performing fault detection on the air conditioner provided by an embodiment of the present application;
[0037] Figure 14 Flowchart of another control method for an air conditioner provided by an embodiment of the present application;
[0038] Figure 15 Flowchart of another control method for an air conditioner provided by an embodiment of the present application;
[0039] Figure 16 Flowchart of another control method for an air conditioner provided by an embodiment of the present application. Detailed implementation manners
[0040] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.
[0041] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative positional relationship and movement conditions between components in a specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.
[0042] The terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise specified, the meaning of "a plurality" is two or more.
[0043] In the description of the present application, it should be noted that unless otherwise clearly specified and defined, the terms "connected" and "connected" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations. In addition, when describing pipelines, the terms "connected" and "connected" used in the present application have the meaning of conducting. The specific meaning needs to be understood in combination with the context.
[0044] In the embodiments of the present application, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, using words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner.
[0045] To expand the applicable scope of air conditioner fault detection, the embodiments of the present application provide a control method for an air conditioner, which can dynamically optimize the first hyperparameter of the fault detection model based on the preset relationship between the hyperparameters of the fault detection model and the performance indicators of the fault detection model. In this way, the fault detection model can be adjusted and improved under different fault conditions, making the fault detection model more intelligent and flexible to adapt to various fault types. Thus, by dynamically optimizing according to the above relationship, the applicable scope of air conditioner fault detection is broadened, enabling the air conditioner to accurately detect faults in various actual scenarios.
[0046] In this application, the air conditioner performs a refrigeration cycle by using a compressor, a condenser, an electronic expansion valve, an evaporator, and a four-way valve as a refrigerant circulation circuit. The refrigeration cycle includes a series of processes involving compression, condensation, expansion, and evaporation, and supplies refrigerant to the air that has been conditioned and heat-exchanged.
[0047] The compressor compresses the refrigerant gas in a high-temperature and high-pressure state and discharges the compressed refrigerant gas. The discharged refrigerant gas flows into the condenser. The condenser condenses the compressed refrigerant into a liquid phase, and heat is released to the surrounding environment through the condensation process.
[0048] The electronic expansion valve expands the liquid-phase refrigerant in a high-temperature and high-pressure state condensed in the condenser into a low-pressure liquid-phase refrigerant. The evaporator evaporates the refrigerant expanded in the electronic expansion valve and returns the refrigerant gas in a low-temperature and low-pressure state to the compressor. The evaporator can achieve a refrigeration effect by using the latent heat of evaporation of the refrigerant for heat exchange with the material to be cooled. Throughout the cycle, the air conditioner can adjust the temperature of the indoor space.
[0049] The outdoor unit of the air conditioner refers to the part of the refrigeration cycle including the compressor and the outdoor heat exchanger. The indoor unit of the air conditioner includes an indoor heat exchanger, and the expansion valve can be provided in the indoor unit or the outdoor unit.
[0050] The indoor heat exchanger and the outdoor heat exchanger serve as condensers or evaporators. When the indoor heat exchanger serves as a condenser, the air conditioner serves as a heater in the heating mode. When the indoor heat exchanger serves as an evaporator, the air conditioner serves as a cooler in the cooling mode.
[0051] Figure 1 FIG. is a schematic diagram of the composition of an air conditioner provided according to an exemplary embodiment of the present application. As Figure 1 shown, the air conditioner 100 includes an indoor unit 101, an outdoor unit 102, and a controller 103 ( Figure 1 not shown in the figure).
[0052] In some embodiments, the number of indoor units 101 can be one or multiple. The number of outdoor units 102 can be one or multiple, and the embodiments of the present application do not make any restrictions.
[0053] Taking the indoor unit 101 as an indoor wall-mounted unit as an example, the indoor wall-mounted unit is usually installed on the indoor wall surface or the like. Again, the indoor cabinet unit is also a form of the indoor unit.
[0054] The outdoor unit 102 is usually set outdoors, can be connected to the indoor unit 101, and is used for heat exchange in the indoor environment. In addition, the outdoor unit 102 is usually located outdoors on the opposite side of the wall surface from the indoor unit 101.
[0055] In some embodiments, the controller 103 refers to a device that can generate operation control signals according to instruction operation codes and timing signals to instruct the air conditioner 100 to execute control instructions. Exemplarily, the controller can be a central processing unit (CPU), a general-purpose processor, 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, and the embodiments of the present application do not impose any restrictions thereon.
[0056] In addition, both the outdoor unit 102 and the indoor unit 101 are communicatively connected to the controller 103 ( Figure 1 not shown in the figure) and perform related operations according to the instructions of the controller 103.
[0057] Taking the number of indoor units 101 as three and the number of outdoor units as one as an example, Figure 2 FIG. 1 is a schematic structural diagram of an air conditioner provided according to an exemplary embodiment of the present application. As Figure 2 shown, the air conditioner 100 includes an indoor heat exchanger 201, an indoor throttling device 202, a compressor 203, a four-way valve 204, an outdoor heat exchanger 205, an outdoor throttling device 206, and a gas-liquid separator 207.
[0058] Among them, the indoor heat exchanger 201 and the indoor throttling device 202 belong to a part of the indoor unit 101, and the compressor 203, the four-way valve 204, the outdoor heat exchanger 205, the outdoor throttling device 206, and the gas-liquid separator 207 belong to a part of the outdoor unit 102.
[0059] In some embodiments, the indoor heat exchanger 201 has a first inlet and outlet for allowing liquid refrigerant to flow between it and the indoor throttling device 202, and has a second inlet and outlet for allowing gaseous refrigerant to flow between it and the suction port of the compressor 203. The indoor heat exchanger 201 exchanges heat between the refrigerant flowing in the heat transfer pipe connected between the first inlet and outlet and the second inlet and outlet and the indoor air.
[0060] In some embodiments, the indoor throttling device 202 is used to adjust the refrigerant flow rate in the air conditioner pipeline. For example, the indoor throttling device 202 is an electronic expansion valve, which has the function of expanding and decompressing the refrigerant flowing through the electronic expansion valve, and can be used to adjust the supply amount of the refrigerant in the pipeline. If the opening of the electronic expansion valve is reduced, the flow resistance of the refrigerant passing through the electronic expansion valve increases. If the opening of the electronic expansion valve is increased, the flow resistance of the refrigerant passing through the electronic expansion valve decreases. Thus, when the opening of the electronic expansion valve changes, the refrigerant flow rate flowing to the indoor heat exchanger 201 will change.
[0061] In some embodiments, the compressor 203 is arranged between the outdoor throttling device 206 and the gas-liquid separator 207, and is used to compress the refrigerant delivered by the gas-liquid separator 207 and deliver the compressed refrigerant to the outdoor throttling device 206 via the four-way valve 204.
[0062] In some embodiments, the four ports of the four-way valve 204 are respectively connected to the compressor 203, the outdoor heat exchanger 205, the gas-liquid separator 207, and the indoor heat exchanger 201. The four-way valve 204 is used to realize the mutual conversion between refrigeration and heating by changing the flow direction of the refrigerant in the system pipeline.
[0063] In some embodiments, the outdoor heat exchanger 205 has a third inlet / outlet for the refrigerant to flow between the four-way valve 204 and the discharge port of the compressor 203, and has a fourth inlet / outlet for the refrigerant to flow between the outdoor heat exchanger 205 and the outdoor throttling device 206. The outdoor heat exchanger 205 exchanges heat between the refrigerant flowing in the heat transfer pipe connected between the third inlet / outlet and the fourth inlet / outlet and the outdoor air.
[0064] In some embodiments, the outdoor throttling device 206 is used to adjust the refrigerant flow rate in the air conditioner pipeline. For example, the outdoor throttling device 202 is an electronic expansion valve, which has the function of expanding and decompressing the refrigerant flowing through the electronic expansion valve, and can be used to adjust the supply amount of the refrigerant in the pipeline.
[0065] In some embodiments, the outlet of the gas-liquid separator 207 is connected to the inlet of the compressor 203, and the inlet of the gas-liquid separator 207 is connected to the S port of the four-way valve 204. In the gas-liquid separator 207, taking the refrigeration cycle as an example, the refrigerant flowing from the indoor heat exchanger 201 to the compressor 203 via the four-way valve 204 is separated into gaseous refrigerant and liquid refrigerant. And, mainly gaseous refrigerant is supplied from the gas-liquid separator 207 to the discharge port of the compressor 203.
[0066] In addition, the controller 103 can be used to control the operation of each component inside the air conditioner 100, so that each component of the air conditioner 100 operates to realize each predetermined function of the air conditioner.
[0067] In some embodiments, the controller 103 may be integrated into the outdoor unit 102, that is, the outdoor unit 102 can control the operation of each component in the air conditioner 100.
[0068] In some embodiments, the air conditioner 100 is also attached with a remote controller, which has the function of communicating with the controller 103 using, for example, infrared rays or other communication methods. The remote controller is used for various controls of the air conditioner by the user, realizing the interaction between the user and the air conditioner 100.
[0069] Figure 3 This is a hardware configuration block diagram of an air conditioner provided by this application according to an exemplary embodiment. As Figure 3 shown, the air conditioner 100 may further include one or more of the following: a communicator 301 and a memory 302.
[0070] In some embodiments, the communicator 301 is used to establish a communication connection with other network entities, such as establishing a communication connection with a terminal device. The communicator 301 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. In particular, the received information is sent to the controller 14 for processing; in addition, the signal generated by the controller 103 is sent out. Usually, 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.
[0071] In some embodiments, the memory 302 can be used to store software programs and data. The controller 103 executes various functions and data processing of the air conditioner 100 by running the software programs or data stored in the memory 302. The memory 302 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. The memory 302 stores an operating system that enables the air conditioner 100 to run. In this application, the memory 302 can store the operating system and various application programs, and can also store the code for executing the control method of the air conditioner provided by the embodiments of this application.
[0072] Those skilled in the art can understand that Figure 3 the hardware structure shown in does not constitute a limitation on the air conditioner. The air conditioner may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements.
[0073] The following specifically introduces the embodiments provided by the present application in conjunction with the accompanying drawings of the specification.
[0074] As Figure 4 shown, the embodiment of the present application provides a control method for an air conditioner, and the method includes the following steps:
[0075] S101. The controller obtains the first hyperparameters preset by the fault detection model, the operation data of the air conditioner, and the preset relationship between the hyperparameters of the fault detection model and the performance indicators of the fault detection model.
[0076] Among them, the fault detection model is used to detect the fault conditions of the air conditioner. Optionally, the fault conditions of the air conditioner may include: excessive refrigerant charge in the air conditioner, insufficient refrigerant charge in the air conditioner, indoor unit blockage, outdoor unit blockage, etc.
[0077] The hyperparameters of the fault detection model refer to the parameters set before training the fault detection model, rather than the parameters learned through training data. Generally, it is necessary to optimize the preset first hyperparameters to obtain a set of optimal hyperparameters (i.e., the target hyperparameters in the following steps) to improve the performance and effect of the fault detection model.
[0078] Optionally, the hyperparameters include but are not limited to learning rate, weight of the regularization term, number of network layers, number of hidden units, etc.
[0079] The performance indicators of the fault detection model are used to indicate the model performance when the fault detection model detects the operation dataset of the air conditioner. Among them, the operation dataset includes multiple samples obtained based on the operation data of the air conditioner, and the multiple samples include normal samples and fault samples.
[0080] Exemplarily, when obtaining multiple samples in the operation dataset of the air conditioner, first clean and preprocess the operation data of the air conditioner, including processing default values, outliers, duplicate values, etc., to determine the quality and integrity of the operation data. Further, extract feature variables from the processed operation data. For example, the feature variables may include the operation mode of the air conditioner, energy consumption fluctuation conditions, etc. Further, according to the required sample types, label the operation data to obtain multiple normal samples and multiple fault samples.
[0081] In some embodiments, the performance indicators may include: the first accuracy rate or the second accuracy rate. Among them, the first accuracy rate is the ratio of the number of detection results indicating that the air conditioner is normal to the number of normal samples. The second accuracy rate is the ratio of the number of detection results indicating that the air conditioner has a fault to the number of fault samples.
[0082] Exemplarily, the operation dataset of the air conditioner includes 50 samples. Among them, the number of normal samples is 35, and the number of faulty samples is 15. When detecting this operation dataset of the air conditioner, the number of detection results indicating that the air conditioner is normal is 33, and the number of detection results indicating that the air conditioner is faulty is 13. Then, the first accuracy rate is 33 / 35, and the second accuracy rate is 13 / 15.
[0083] In some embodiments, the preset relationship between the hyperparameters of the fault detection model and the performance metrics of the fault detection model may include: a positive correlation between the hyperparameters of the fault detection model and the first accuracy rate of the fault detection model, or a negative correlation between the hyperparameters of the fault detection model and the second accuracy rate of the fault detection model.
[0084] Exemplarily, the preset relationship between the hyperparameters of the fault detection model and the performance metrics of the fault detection model is as Figure 5 shown. As Figure 5 shown, as the hyperparameters increase, the first accuracy rate also increases. As the hyperparameters increase, the second accuracy rate decreases.
[0085] In addition, the above preset relationship is obtained through experimental methods. For example, fault detection experiments are conducted on the air conditioner under different fault conditions, so as to obtain the preset relationship between the hyperparameters and the performance metrics of the fault detection model.
[0086] Exemplarily, as Figure 6 , Figure 7 , Figure 8 , Figure 9 and Figure 10 shown, under different fault conditions, whether it is a one-to-three, one-to-five, one-to-seven or one-to-nine air conditioner, as the hyperparameters increase, the first accuracy rate all increases, and the second accuracy rate all decreases.
[0087] In some embodiments, the fault detection model may be a fault detection model established based on a Bayesian network.
[0088] Exemplarily, the fault detection model established based on a Bayesian network is as shown in the following formula (1).
[0089]
[0090] Among them, x i is a feature variable; a i is the coefficient corresponding to the feature variable; where i = 1, 2,..., n.
[0091] Optionally, the feature variables may be the evaporation temperature, condensation temperature, suction temperature, discharge temperature, evaporation pressure, condensation pressure, compressor power, etc. of the air conditioner.
[0092] In addition, for the fault detection model shown in the above formula (1), when considering the two-dimensional case, its decision boundary can be regarded as an elliptical region. In this two-dimensional space, the interior of the ellipse represents that the air conditioner is detected as normal, while the exterior of the ellipse represents that the air conditioner is detected as faulty.
[0093] When considering the three-dimensional case, the decision boundary becomes an ellipsoidal region. When considering the case of four dimensions and above, the decision boundary presents as a high-dimensional ellipsoidal shape. In other words, at different dimensions, the decision boundaries formed by the fault detection model exhibit different geometric shapes, which are used to distinguish the normal state and the faulty state of the air conditioner.
[0094] Exemplarily, in the three-dimensional space, as Figure 11 shown, the interior of the ellipsoid represents that the air conditioner is detected as normal, while the exterior of the ellipsoid represents that the air conditioner is detected as faulty.
[0095] S102. The controller determines the target hyperparameters according to a preset relationship to optimize the first hyperparameters.
[0096] In some embodiments, as Figure 12 shown, step S102 can be implemented as the following steps.
[0097] S1021. The controller obtains the model information of the air conditioner.
[0098] Optionally, the model information of the air conditioner can exist in the form of a model code.
[0099] In some embodiments, the controller can obtain the model information of the air conditioner through a sensor for scanning the model information of the air conditioner. Or the user can manually input the model information of the air conditioner through a terminal device or the remote controller of the air conditioner, and then the controller can obtain the signal information of the air conditioner. Alternatively, the controller can obtain the model information of the air conditioner from the memory.
[0100] S1022. The controller determines the value range of the hyperparameters corresponding to the model information according to the model information.
[0101] In some implementations, there is a corresponding relationship between the model information of the air conditioner and the value range of the hyperparameters.
[0102] Exemplarily, when the model information of the air conditioner is used to represent that the air conditioner is a household multi-connected unit, the value range of the hyperparameters corresponding to the model information is (25, 50). When the model information of the air conditioner is used to represent that the air conditioner is a modular multi-connected unit, the value range of the hyperparameters corresponding to the model information is (75, 200).
[0103] Among them, the household multi-connected air conditioner is usually used in households or small shopping mall places, and the modular multi-connected air conditioner is usually used in commercial buildings, office buildings or large places.
[0104] S1023. The controller searches for the target hyperparameter within the value range according to the preset relationship to optimize the first hyperparameter.
[0105] In some embodiments, the hyperparameters of the fault detection model within the value range also have a positive correlation with the first accuracy rate of the fault detection model, and the hyperparameters of the fault detection model within the value range also have a negative correlation with the second accuracy rate of the fault detection model.
[0106] In some embodiments, the controller can select a second hyperparameter from the value range of the hyperparameters and adjust the fault detection model based on the second hyperparameter.
[0107] Further, the controller inputs the operation data of the air conditioner in the historical time period into the first fault detection model adjusted based on the second hyperparameter to obtain the preset performance index.
[0108] Still further, the controller determines whether the expected performance index is within the preset range. In addition, for the specific description of the preset range, reference can be made to the specific description of the preset range in step S204 below, which will not be elaborated herein in this application.
[0109] If so, the second hyperparameter is the target hyperparameter. If not, based on the second hyperparameter and the preset relationship, determine the value range of the new hyperparameter. And based on the value range of the new hyperparameter, repeat the above steps until the obtained expected performance index is within the preset range. When the expected performance index is within the preset range, the second hyperparameter of the second fault detection model at this time is determined as the target hyperparameter.
[0110] In addition, for the specific search process of the target hyperparameter, reference can be made to the Figure 14 specific description of the embodiments shown below, which will not be elaborated herein in this application.
[0111] S103. The controller adjusts the fault detection model according to the target hyperparameter to obtain the adjusted target fault detection model.
[0112] Among them, the target fault detection model is used to perform fault detection on the air conditioner based on the operation data of the air conditioner.
[0113] In some embodiments, the controller inputs the operation data of the air conditioner obtained in real time into the target fault detection model, and the target fault detection model outputs the detection result. The detection result includes a detection result indicating that the air conditioner is normal or a detection result indicating that the air conditioner has a fault.
[0114] The following is combined withFigure 13 The illustrated embodiment exemplarily introduces the complete process from establishing a fault detection model for an air conditioner to performing fault detection on the air conditioner.
[0115] First, a modeling experiment is conducted to establish a fault detection model based on preset first hyperparameters. Further, the operation data of the air conditioner is detected based on the fault detection model to obtain the performance metrics of the fault detection model. Through a preset relationship, the target hyperparameters of the fault detection model are determined to optimize the first hyperparameters. Still further, the fault detection model is adjusted based on the target hyperparameters to obtain a target fault detection model, and then the operation data of the air conditioner is detected based on the target fault detection model to determine whether the air conditioner has a fault. If a fault occurs, measures can be taken in a timely manner for repair and maintenance to restore the air conditioner to normal. In this way, by establishing and optimizing the fault detection model, air conditioner faults can be discovered and solved in a timely manner, improving the stability and reliability of the air conditioner, while reducing maintenance costs and time.
[0116] Based on Figure 4 the illustrated embodiment, the present application provides a control method for an air conditioner according to an exemplary embodiment, which can dynamically optimize the first hyperparameters of the fault detection model based on a preset relationship between the hyperparameters of the fault detection model and the performance metrics of the fault detection model. This method can adjust and improve the fault detection model under different fault conditions, making the fault detection model more intelligent and flexible in adapting to various fault types. In this way, by performing dynamic optimization according to the above relationship, the applicable range of air conditioner fault detection is broadened, enabling the air conditioner to accurately detect faults in various actual scenarios.
[0117] In some embodiments, in order to find the target hyperparameters within the value range, as Figure 14 illustrated, the control method of the air conditioner further includes the following steps.
[0118] S201. The controller obtains the operation data of the air conditioner within a historical time period.
[0119] Optionally, the historical time period can be a time period of the past year or more than one year.
[0120] In some embodiments, the controller can obtain the operation data of the air conditioner within a historical time period from a memory.
[0121] Optionally, the operation data includes the operation data of the air conditioner in the cooling mode and the operation data in the heating mode, and the data volume of the operation data in the cooling mode and the data volume of the operation data in the heating mode are both greater than or equal to 1000 pieces.
[0122] S202. The controller selects a second hyperparameter from within the value range and adjusts the fault detection model based on the second hyperparameter to obtain an adjusted first fault detection model.
[0123] Optionally, the second hyperparameter can be a hyperparameter located at the middle position of the value range.
[0124] S203. The controller inputs the operation data of the air conditioner in the historical time period into the first fault detection model to obtain an expected performance index.
[0125] In some embodiments, the controller can obtain an operation data set of the air conditioner based on the operation data of the air conditioner in the historical time period. Further, the controller inputs the operation data set of the air conditioner into the first fault detection model, that is, the controller uses the first fault detection model to detect the operation data set of the air conditioner to obtain an expected performance index.
[0126] In addition, for the expected performance index, reference can be made to the specific description of the performance index in step S101 above, and details are not elaborated herein.
[0127] Optionally, the expected performance index can include a first accuracy rate or a second accuracy rate.
[0128] S204. When the expected performance index is outside the preset range, the controller determines the value range of the new hyperparameter based on the second hyperparameter and the preset relationship.
[0129] Wherein, the preset range is (n, m), and m ≥ n.
[0130] It can be understood that the preset range is a value range of a relatively optimal performance index obtained through a large number of experiments. If the expected performance index is outside the preset range, it means that the second hyperparameter is not appropriate and a new hyperparameter needs to be determined as the target hyperparameter.
[0131] In some embodiments, according to the different parameters specifically included in the expected performance index, the situation where the expected performance index is outside the preset range includes multiple situations, and the preset relationship can also include different relationships.
[0132] In one example, when the expected performance index includes the first accuracy rate, the situation where the expected performance index is outside the preset range includes: the situation where the first accuracy rate is greater than the upper limit value of the preset range, or the situation where the first accuracy rate is less than the lower limit value of the preset range. The preset relationship includes: the positive correlation relationship between the hyperparameter of the fault detection model and the first accuracy rate of the fault detection model.
[0133] In another example, when the expected performance metric includes a second accuracy rate, the cases where the expected performance metric is outside the preset range include: the case where the second accuracy rate is greater than the upper limit value of the preset range, or the case where the second accuracy rate is less than the lower limit value of the preset range. The preset relationship: includes the negative correlation relationship between the hyperparameters of the fault detection model and the second accuracy rate of the fault detection model.
[0134] In some embodiments, when the expected performance metric is outside the preset range, the controller divides the value range of the hyperparameters into two sub-value ranges. Among them, the upper limit value of one sub-value range is the second hyperparameter, and the lower limit value is the lower limit value of the value range before division. The upper limit value of the other sub-value range is the upper limit value of the value range before division, and the lower limit value is the second hyperparameter.
[0135] Exemplarily, the value range of the hyperparameters before division is (25, 50), and the second hyperparameter is 45. When the expected performance metric is outside the preset range, the value range (25, 50) is divided into the sub-value range (25, 45) and the sub-value range (45, 50).
[0136] Furthermore, the controller selects at least one sub-value range from the two sub-value ranges as the value range of the new hyperparameters according to the preset relationship.
[0137] Exemplarily, the following uses the parameters specifically included in the expected performance metric to exemplarily illustrate the determination process of the value range of the new hyperparameters:
[0138] (1) The expected performance metric includes the first accuracy rate
[0139] In some embodiments, when the first accuracy rate is greater than the upper limit value of the preset range, the controller determines the sub-value range with the upper limit value being the second hyperparameter and the lower limit value being the lower limit value of the value range before division as the value range of the new hyperparameters.
[0140] It can be understood that if the first accuracy rate is greater than the upper limit value of the preset range, it means that the first accuracy rate is relatively high and the first accuracy rate needs to be reduced. Since there is also a positive correlation between the hyperparameters within the value range of the fault detection model and the first accuracy rate of the fault detection model, when the first accuracy rate is relatively high, hyperparameters less than the second hyperparameter can be selected to reduce the first accuracy rate.
[0141] In some embodiments, when the first accuracy rate is less than the lower limit value of the preset range, the controller determines the sub-value range with the upper limit value being the upper limit value of the value range before division and the lower limit value being the second hyperparameter as the value range of the new hyperparameters.
[0142] Similarly, if the first accuracy rate is less than the lower limit of the preset range, it indicates that the first accuracy rate is relatively small and needs to be increased. Since the hyperparameters within the value range of the fault detection model are also positively correlated with the first accuracy rate of the fault detection model, when the first accuracy rate is small, hyperparameters greater than the second hyperparameter can be selected to increase the first accuracy rate.
[0143] (2) The expected performance indicators include the second accuracy rate
[0144] In some embodiments, when the second accuracy rate is greater than the upper limit of the preset range, the controller determines the sub-value range with the upper limit being the upper limit of the value range before division and the lower limit being the second hyperparameter as the new value range of the hyperparameters.
[0145] It can be understood that if the second accuracy rate is greater than the upper limit of the preset range, it indicates that the second accuracy rate is relatively large and needs to be decreased. Since the hyperparameters within the value range of the fault detection model are also negatively correlated with the second accuracy rate of the fault detection model, when the second accuracy rate is large, hyperparameters greater than the second hyperparameter can be selected to decrease the second accuracy rate.
[0146] In some embodiments, when the second accuracy rate is less than the lower limit of the preset range, the controller determines the sub-value range with the upper limit being the second hyperparameter and the lower limit being the lower limit of the value range before division as the value range of the new hyperparameters.
[0147] Similarly, if the second accuracy rate is less than the lower limit of the preset range, it indicates that the second accuracy rate is relatively small and needs to be increased. Since the hyperparameters within the value range of the fault detection model are also negatively correlated with the second accuracy rate of the fault detection model, when the second accuracy rate is small, hyperparameters less than the second hyperparameter can be selected to increase the second accuracy rate.
[0148] S205. The controller repeatedly executes step S202, step S203, and step S204 until the obtained expected performance indicator is within the preset range.
[0149] Exemplarily, taking the case where the expected performance index includes the first accuracy rate as an example, when the first accuracy rate is greater than the upper limit value of the preset range, the controller selects a new second hyperparameter within the value range of the new hyperparameter and adjusts the fault detection model based on the new second hyperparameter. Further, the controller inputs the operation data of the air conditioner in the historical time period into the adjusted fault detection model to obtain a new expected performance index. Still further, the controller determines whether the new expected performance index is outside the preset range. If so, based on the new second hyperparameter and the preset relationship, the value range of the new hyperparameter is determined again. Thus, steps S202, S203, and S204 are repeatedly executed until the obtained expected performance index is within the preset range.
[0150] S206. The controller determines the second hyperparameter of the second fault detection model as the target hyperparameter.
[0151] Among them, the expected performance index obtained based on the second fault detection model is within the preset range.
[0152] It can be understood that the preset range is a range of values of a relatively optimal performance index obtained through a large number of experiments. If the expected performance index obtained based on the second fault detection model is within the preset range, it indicates that the expected performance index is relatively optimal. Therefore, the second hyperparameter of the second fault detection model at this time can be determined as the target hyperparameter.
[0153] Optionally, the second fault detection model can be the first fault detection model in step S202, or a fault detection model different from the first fault detection model obtained after being adjusted based on the new second hyperparameter.
[0154] In one example, if the controller executes steps S202 and S203 for the first time and the obtained expected performance index is within the preset range, the second fault detection model is the first fault detection model in step S202.
[0155] In another example, if the controller does not execute steps S202 and S203 for the first time, the second fault detection model is a fault detection model different from the first fault detection model obtained after being adjusted based on the new second hyperparameter.
[0156] Next, in combination with the Figure 15 shown embodiment, the complete process of determining the target hyperparameter is exemplarily introduced in the case where the expected performance index includes the first accuracy rate.
[0157] As Figure 15 shown, the process starts.
[0158] Step a1. The controller obtains the operation data of the air conditioner in the historical time period.
[0159] Step a2: The controller selects a second hyperparameter from within the value range and adjusts the fault detection model based on the second hyperparameter to obtain an adjusted first fault detection model.
[0160] Step a3: The controller inputs the operation data of the air conditioner within the historical time period into the first fault detection model to obtain a first accuracy rate.
[0161] Determine whether the first accuracy rate is within the preset range.
[0162] If so, execute the following step a7.
[0163] If not, execute the following step a4.
[0164] Step a4: Determine whether the first accuracy rate is greater than the upper limit value of the preset range.
[0165] If so, execute the following step a5 and the above step a2.
[0166] If not, execute the following step a6 and the above step a2.
[0167] Step a5: The controller determines a sub-value range with the upper limit value as the second hyperparameter and the lower limit value as the lower limit value of the value range before division as the new value range of the hyperparameter.
[0168] Step a6: The controller determines a sub-value range with the upper limit value as the upper limit value of the value range before division and the lower limit value as the second hyperparameter as the value range of the new hyperparameter.
[0169] Step a7: The controller determines the second hyperparameter of the second fault detection model as the target hyperparameter.
[0170] The following combines with the Figure 16 illustrated embodiment to exemplarily introduce the completion process of determining the target hyperparameter when the expected performance index includes the second accuracy rate.
[0171] As Figure 16 shown, the process starts.
[0172] Step b1: The controller obtains the operation data of the air conditioner within the historical time period.
[0173] Step b2: The controller selects a second hyperparameter from within the value range and adjusts the fault detection model based on the second hyperparameter to obtain an adjusted first fault detection model.
[0174] Step b3: The controller inputs the operation data of the air conditioner within the historical time period into the first fault detection model to obtain a second accuracy rate.
[0175] Determine whether the second accuracy rate is within the preset range.
[0176] If so, execute the following step b7.
[0177] If not, execute the following step b4.
[0178] Step b4: Determine whether the second accuracy rate is greater than the upper limit value of the preset range.
[0179] If so, execute the following step b5 and the above step b2.
[0180] If not, execute the following step b6 and the above step b2.
[0181] Step b5: The controller determines the sub-value range with the upper limit value being the upper limit value of the value range before division and the lower limit value being the second hyperparameter as the value range of the new hyperparameter.
[0182] Step b6: The controller determines the sub-value range with the upper limit value being the second hyperparameter and the lower limit value being the lower limit value of the value range before division as the value range of the new hyperparameter.
[0183] Step b7: The controller determines the second hyperparameter of the second fault detection model as the target hyperparameter.
[0184] Those skilled in the art should be able to realize that in one or more of the above examples, the functions described in the present invention can be implemented by hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. The computer-readable medium includes computer storage media and communication media, where the communication media includes any medium that facilitates the transmission of a computer program from one place to another. The storage media can be any available medium accessible by a general-purpose or special-purpose computer.
[0185] Through the description of the above embodiments, those skilled in the art can clearly understand that for the convenience and conciseness of description, only the above division of each functional module is used as an example for illustration. In actual applications, the above functions can be allocated to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.
[0186] The above is only the specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
Claims
1. An air conditioner, characterized in that, it includes: at least one outdoor unit; at least one indoor unit; a controller configured to: obtain a first hyperparameter preset in a fault detection model, operation data of the air conditioner, and a preset relationship between the hyperparameters of the fault detection model and the performance index of the fault detection model; determine a target hyperparameter according to the preset relationship to optimize the first hyperparameter; adjust the fault detection model according to the target hyperparameter to obtain an adjusted target fault detection model, and the target fault detection model is used to perform fault detection on the air conditioner based on the operation data of the air conditioner.
2. The air conditioner according to claim 1, characterized in that, the performance index includes a first accuracy rate or a second accuracy rate; wherein, the performance index is used to indicate the model performance when the fault detection model detects an operation data set of the air conditioner; the operation data set includes multiple samples obtained based on the operation data of the air conditioner, and the multiple samples include normal samples and fault samples; the first accuracy rate is the ratio between the number of detection results indicating that the air conditioner is normal and the number of normal samples; the second accuracy rate is the ratio between the number of detection results indicating that the air conditioner has a fault and the number of fault samples; the preset relationship includes a positive correlation relationship between the hyperparameters of the fault detection model and the first accuracy rate, or a negative correlation relationship between the hyperparameters of the fault detection model and the second accuracy rate.
3. The air conditioner according to claim 1 or 2, characterized in that, the controller, configured to determine the target hyperparameter according to the preset relationship to optimize the first hyperparameter, is specifically configured to: obtain the model information of the air conditioner; determine the value range of the hyperparameters corresponding to the model information according to the model information; find the target hyperparameter within the value range according to the preset relationship to optimize the first hyperparameter.
4. The air conditioner according to claim 3, characterized in that, the controller, configured to find the target hyperparameter within the value range according to the preset relationship, is specifically configured to: obtain the operation data of the air conditioner in a historical time period; selection step: select a second hyperparameter from the value range and adjust the fault detection model based on the second hyperparameter to obtain an adjusted first fault detection model; detection step: input the operation data of the air conditioner in the historical time period into the first fault detection model to obtain an expected performance index; range update step: in the case that the expected performance index is outside the preset range, determine a new value range of the hyperparameters based on the second hyperparameter and the preset relationship; repeatedly execute the selection step, the detection step, and the range update step until the obtained expected performance index is within the preset range; determine the second hyperparameter of the second fault detection model as the target hyperparameter; wherein, the expected performance index obtained based on the second fault detection model is within the preset range.
5. The air conditioner according to claim 1 or 2, characterized in that, the fault detection model is a fault detection model established based on a Bayesian network.
6. A control method for an air conditioner, characterized in that, the method includes: obtaining a first hyperparameter preset in the fault detection model, operation data of the air conditioner, and a preset relationship between the hyperparameters of the fault detection model and the performance indicators of the fault detection model; determining target hyperparameters according to the preset relationship to optimize the first hyperparameters; adjusting the fault detection model according to the target hyperparameters to obtain an adjusted target fault detection model, and the target fault detection model is used to perform fault detection on the air conditioner based on the operation data of the air conditioner.
7. The method according to claim 6, characterized in that, the performance indicators include a first accuracy rate or a second accuracy rate; wherein, the performance indicators are used to indicate the model performance when the fault detection model detects the operation data set of the air conditioner; the operation data set includes a plurality of samples obtained based on the operation data of the air conditioner, and the plurality of samples include normal samples and fault samples; the first accuracy rate is the ratio between the number of detection results indicating that the air conditioner is normal and the number of normal samples; the second accuracy rate is the ratio between the number of detection results indicating that the air conditioner has a fault and the number of fault samples; the preset relationship includes a positive correlation between the hyperparameters of the fault detection model and the first accuracy rate, or a negative correlation between the hyperparameters of the fault detection model and the second accuracy rate.
8. The method according to claim 6 or 7, characterized in that, the determining target hyperparameters according to the preset relationship to optimize the first hyperparameters includes: obtaining the model information of the air conditioner; determining the value range of the hyperparameters corresponding to the model information according to the model information; searching for the target hyperparameters within the value range according to the preset relationship to optimize the first hyperparameters.
9. The method according to claim 8, characterized in that, the searching for the target hyperparameters within the value range according to the preset relationship includes: obtaining the operation data of the air conditioner in a historical time period; selection step: selecting second hyperparameters from the value range and adjusting the fault detection model based on the second hyperparameters to obtain an adjusted first fault detection model; detection step: inputting the operation data of the air conditioner in the historical time period into the first fault detection model to obtain an expected performance indicator; range update step: in the case that the expected performance indicator is outside the preset range, determining a new value range of the hyperparameters based on the second hyperparameters and the preset relationship; repeatedly executing the selection step, the detection step, and the range update step until the obtained expected performance indicator is within the preset range; determining the second hyperparameters of the second fault detection model as the target hyperparameters; wherein, the expected performance indicator obtained based on the second fault detection model is within the preset range.
10. The method according to claim 6 or 7, characterized in that, the fault detection model is a fault detection model established based on a Bayesian network.