A multi-split air conditioning system and a fault detection method

Abstract

The embodiment of the application discloses a multi-split air conditioning system and a fault detection method, relates to the technical field of multi-split air conditioners, and is used for improving the efficiency and accuracy of electronic expansion valve fault detection and positioning. The multi-split air conditioning system comprises an outdoor unit, at least one indoor unit, an electronic expansion valve arranged between each indoor unit and the outdoor unit, and a controller configured to acquire real-time operation data of the multi-split air conditioning system, perform fault identification on the real-time operation data through a preset graph attention model, and obtain a fault identification result. The fault identification result includes that the electronic expansion valve in the multi-split air conditioning system is in a fault operation state or a normal operation state. In the case that the electronic expansion valve in the multi-split air conditioning system is in the fault operation state, the controller performs fault positioning on the electronic expansion valve through the preset graph attention model to determine the indoor unit serial number of the electronic expansion valve.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of multi-connected air conditioning, and particularly relates to a multi-connected air conditioning system and a fault detection method. BACKGROUND

[0002] As an important component in the multi-connected air conditioning system, the electronic expansion valve plays a key role in refrigerant distribution and on-demand operation of the multi-connected air conditioning system. If the electronic expansion valve has problems such as abnormal opening degree and inability to completely close, the refrigeration and cooling effect of the multi-connected air conditioning system will be reduced, and the user experience will be affected.

[0003] At present, most of the online fault detection and diagnosis technologies for the multi-connected air conditioning system are still in the theoretical research stage, and machine learning algorithms are mostly used to achieve this. However, this method requires a large amount of data for model training. For the multi-connected air conditioning system, there are many designed working conditions and environmental factors. In the actual use process, it is impossible to exhaust the data of this working condition, causing distortion of the model. SUMMARY

[0004] The present application provides a multi-connected air conditioning system and a fault detection method for improving the efficiency and accuracy of electronic expansion valve fault detection and positioning.

[0005] In order to achieve the above purpose, the technical scheme is as follows.

[0006] In a first aspect, the embodiments of the present application provide a multi-connected air conditioning system, comprising: an outdoor unit; at least one indoor unit, each indoor unit being provided with an electronic expansion valve between the indoor unit and the outdoor unit; a controller configured to: obtain real-time running data of the multi-connected air conditioning system; perform fault identification on the real-time running data through a preset graph attention model to obtain a fault identification result, the fault identification result including whether the electronic expansion valve in the multi-connected air conditioning system is in a fault running state or a normal running state; and in the case that the electronic expansion valve in the multi-connected air conditioning system is in the fault running state, performing fault positioning on the electronic expansion valve through the preset graph attention model to determine the indoor unit number of the electronic expansion valve.

[0007] The technical scheme provided by the embodiments of the present application at least brings the following beneficial effects: by inputting the obtained real-time running data of the multi-connected air conditioning system into the preset graph attention model, the fault identification result of the electronic expansion valve in the multi-connected air conditioning system is obtained, and the fault detection efficiency of the electronic expansion valve is improved. At the same time, when it is determined that the electronic expansion valve in the multi-connected air conditioning system is in the fault running state, the electronic expansion valve is positioned through the preset graph attention model, and the indoor unit number of the first expansion valve is obtained. In this way, the dependence on prior knowledge and the data stream preprocessing process is reduced, and the accuracy of the fault positioning result of the electronic expansion valve is improved.

[0008] In some embodiments, the real-time operation data includes real-time operation data of the outdoor unit and real-time operation data of the indoor unit; wherein the real-time operation data of the outdoor unit includes compressor frequency, pressure value, pressure top temperature value, exhaust gas superheat value, defrosting temperature value and ambient temperature value; the real-time operation data of the indoor unit includes variable frequency heat dissipation value, liquid pipe temperature value, gas pipe temperature value, return air temperature value and outlet air temperature value.

[0009] In some embodiments, the controller is further configured to: obtain normal operation data and fault operation data of the multi-split air conditioning system; wherein the normal operation data is operation data of the multi-split air conditioning system when the electronic expansion valve is in a normal operation state; the fault operation data is operation data of the multi-split air conditioning system when the electronic expansion valve is in a fault operation state; and train the initial graph attention model according to the normal operation data and the fault operation data to obtain the preset graph attention model after training.

[0010] In some embodiments, the controller is configured to, before training the initial graph attention model according to the normal operation data and the fault operation data, further configured to: convert the normal operation data and the fault operation data into a graph structure through a preset graph neural network model; the graph structure includes original node feature vectors and connection relationships between the original node feature vectors; wherein the original node feature vectors are vector representations of the normal operation data and the fault operation data.

[0011] In some embodiments, the controller is further configured to: perform Pearson correlation coefficient calculation on different data information between two original node feature vectors; if the Pearson correlation coefficient is greater than a preset threshold, the two original node feature vectors are connected by an edge; if the Pearson correlation coefficient is less than the preset threshold, the two original node feature vectors are not connected by an edge.

[0012] In some embodiments, the controller is configured to train the initial graph attention model according to the normal operation data and the fault operation data, and is specifically configured to: convert the original node feature vectors into target node feature vectors, and assign attention coefficients to the target node feature vectors; perform aggregation operation on the target node feature vectors according to the attention coefficients; update the preset graph attention model through iteration of the preset graph neural network model to determine whether the original node feature vectors meet a preset convergence condition; and in the case where the original node feature vectors meet the preset convergence condition, update the attention coefficients to obtain influence relationship between the original node feature vectors.

[0013] In a second aspect, the embodiments of the present application provide a fault detection method of a multi-split air conditioning system. The method is applied to the multi-split air conditioning system, and the multi-split air conditioning system includes an outdoor unit and a plurality of indoor units. An electronic expansion valve is arranged between each indoor unit and the outdoor unit. The method includes: obtaining real-time operation data of the multi-split air conditioning system; performing fault identification on the real-time operation data by using a preset graph attention model to obtain a fault identification result. The fault identification result includes that the electronic expansion valve in the multi-split air conditioning system is in a fault operation state or a normal operation state; and in the case that the electronic expansion valve in the multi-split air conditioning system is in the fault operation state, performing fault positioning on the electronic expansion valve by using the preset graph attention model to determine an indoor unit serial number of the electronic expansion valve.

[0014] In a third aspect, the embodiments of the present application provide a controller. The controller includes: one or more processors; and one or more memories. The one or more memories are configured to store computer program codes. The computer program codes include computer instructions. When the one or more processors execute the computer instructions, the controller performs any one of the fault detection methods of the multi-split air conditioning system provided in the second aspect.

[0015] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium. The computer readable storage medium includes computer instructions. When the computer instructions are executed on a computer, the computer executes the method provided in the second aspect and possible implementation manners.

[0016] In a fifth aspect, the embodiments of the present application provide a computer program product. The computer program product can be directly loaded into a memory and contains software codes. The computer program product can be loaded and executed by a computer to implement the method provided in the second aspect and possible implementation manners.

[0017] It should be noted that the computer instructions described above can be stored in the computer readable storage medium in whole or in part. The computer readable storage medium can be packaged together with the processor of the controller or packaged separately from the processor of the controller, and the present application does not limit the computer readable storage medium.

[0018] The beneficial effects of the second aspect to the fifth aspect described in the present application can be analyzed with reference to the beneficial effects of the first aspect, which will not be described here. BRIEF DESCRIPTION OF DRAWINGS

[0019] The accompanying drawings are included to provide a further understanding of the technical solutions of the present application, and constitute a part of the specification. The drawings, together with the embodiments of the present application, are used to explain the technical solutions of the present application, and do not constitute a limitation on the technical solutions of the present application.

[0020] Figure 1 A structural schematic diagram of a multi-split air conditioning system provided by the embodiments of the present application is shown in the following figure.

[0021] Figure 2 A setting position diagram of an electronic expansion valve provided for an embodiment of the present application;

[0022] Figure 3 Another structure diagram of a multi-split air conditioning system provided for an embodiment of the present application;

[0023] Figure 4 A principle diagram of a refrigeration cycle provided for an embodiment of the present application;

[0024] Figure 5 A structure diagram of a controller provided for an embodiment of the present application;

[0025] Figure 6 A hardware configuration block diagram of a multi-split air conditioning system provided for an embodiment of the present application;

[0026] Figure 7 A setting position diagram of a temperature sensor provided for an embodiment of the present application;

[0027] Figure 8 An interaction diagram of a controller and a terminal device provided for an embodiment of the present application;

[0028] Figure 9 A management interface diagram of a terminal device provided for an embodiment of the present application;

[0029] Figure 10 Another management interface diagram of a terminal device provided for an embodiment of the present application;

[0030] Figure 11 Another management interface diagram of a terminal device provided for an embodiment of the present application;

[0031] Figure 12 A fault detection method flow diagram of a multi-split air conditioning system provided for an embodiment of the present application;

[0032] Figure 13 A model training method flow diagram provided for an embodiment of the present application;

[0033] Figure 14 Another model training method flow diagram provided for an embodiment of the present application;

[0034] Figure 15 Another fault detection method flow diagram of a multi-split air conditioning system provided for an embodiment of the present application. DETAILED DESCRIPTION

[0035] With reference to the drawings, the technical solutions in the embodiments of the present application will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative effort should fall within the scope of the present application.

[0036] It should be noted that all the direction indications (such as up, down, left, right, front, back, and the like) in the embodiments of the present application are only used to explain the relative position relationship, movement condition and the like between components in a certain posture (as shown in the drawings), and if the certain posture changes, the direction indications also change accordingly.

[0037] The terms "first", "second", "third", "fourth", "fifth", "sixth" and the like in the description of the embodiments of the present application are used to describe various technical features, and do not indicate or imply relative importance or imply the number of the technical features indicated. Therefore, a feature defined with "first", "second", "third", "fourth", "fifth", "sixth" and the like can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified and limited, the meaning of "a plurality of" is two or more.

[0038] In the description of the present application, it should be noted that unless otherwise specified and limited, the terms "connected", "connected" should be understood in a broad sense, for example, can be fixedly connected, can be detachably connected, or integrally connected. For a person of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances. In addition, when describing the pipeline, the "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.

[0039] In the embodiments of the present application, the words "exemplary" or "for example" are used to represent an example, illustration or description. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the words "exemplary" or "for example" are intended to present the relevant concept in a specific manner.

[0040] For the convenience of understanding, first, some terms or basic concepts of technology related to the embodiments of the present application are simply introduced and explained.

[0041] Graph Attention Networks (GAT): A graph neural network model that uses a self-attention mechanism. The network calculates the attention of each adjacent node relative to a certain node in the graph in a manner similar to self-attention in neural network models, and then combines the features of the node itself and the attention features as the features of the node. Based on this, the node is classified and other tasks are performed.

[0042] Graph Neural Networks (GNN): A connection model that obtains the dependency relationship in a graph through information transmission between nodes in the network. GNN updates the state of a node by its neighbors at any depth, and this state can represent state information.

[0043] Pearson correlation coefficient: Also known as Pearson product-moment correlation coefficient, it is used to measure the correlation (linear correlation) between two variables X and Y, and its value is between -1 and 1.

[0044] As described in the background, the current online fault detection and diagnosis technology of the multi-split air conditioning system is still in the theoretical research stage, and machine learning algorithms are mostly used to achieve this. However, this method requires a large amount of data for model training. For multi-split air conditioning systems, there are many designed working conditions and environmental factors. In the actual use process, it is impossible to exhaust the data of this working condition, causing the model to be distorted.

[0045] Therefore, the embodiments of the present application provide a multi-split air conditioning system, which comprises: an outdoor unit; at least one indoor unit, each indoor unit being provided with an electronic expansion valve between the indoor unit and the outdoor unit; and a controller configured to: obtain real-time running data of the multi-split air conditioning system; perform fault identification on the real-time running data through a preset graph attention model to obtain a fault identification result, the fault identification result including whether the electronic expansion valve in the multi-split air conditioning system is in a fault running state or a normal running state; and in the case that the electronic expansion valve in the multi-split air conditioning system is in the fault running state, perform fault positioning on the electronic expansion valve through the preset graph attention model to determine the indoor unit number of the electronic expansion valve.

[0046] In this way, the fault detection efficiency of the electronic expansion valve is improved, and the accuracy of the fault positioning result of the electronic expansion valve is also improved.

[0047] To further describe the scheme of the present application, reference can be made to Figure 1 , Figure 1 A structure diagram of a multi-split air conditioning system according to an exemplary embodiment of the present application is provided.

[0048] As Figure 1As shown, the multi-split air conditioning system 10 includes an outdoor unit 11, a throttling device 12, multiple indoor units 13, and a controller 14. Figure 1 (Not shown in the image).

[0049] The throttling device 12 includes multiple electronic expansion valves 121, each corresponding to an indoor unit 13. The outdoor unit 11 is connected to the multiple indoor units 13 by pipes, and each pipe between the indoor unit 13 and the outdoor unit 11 is equipped with an electronic expansion valve 121. These pipes, also referred to as gas-liquid pipes, include: a gas pipe 15 for transporting gaseous refrigerant and a liquid pipe 16 for transporting two-phase refrigerant.

[0050] For example, such as Figure 2 The diagram shown is a schematic of the location of an electronic expansion valve provided in accordance with an exemplary embodiment of this application. The electronic expansion valve 121 can be installed on the liquid pipe 16, and a throttle valve can also be installed on the liquid pipe 16. One end of the liquid pipe 16 can be connected to the indoor heat exchanger 131 described below. Similarly, one end of the gas pipe 15 can also be connected to the indoor heat exchanger 131 described below.

[0051] In addition, the outdoor unit 11, the throttling device 12, and multiple indoor units 13 are all connected to the controller. Figure 1 (not shown in the image), and performs related operations according to the controller's instructions.

[0052] The outdoor unit 11 is usually installed outdoors to assist in heat exchange in the indoor environment.

[0053] The throttling device 12 is used to regulate the fluid flow rate in the air conditioning gas-liquid pipe and to regulate the refrigerant flow rate. Multiple electronic expansion valves 121 are used to regulate the refrigerant supply in the pipes, and these multiple electronic expansion valves 121 can operate independently of the multiple indoor units 13 (e.g., Figure 1 As shown), it can also belong to multiple parts of indoor unit 13 (such as...). Figure 3 (as shown), Figure 3 This is a schematic diagram of another multi-split air conditioning system provided in accordance with an exemplary embodiment of this application. The multiple indoor units 13 can be wall-mounted or floor-standing units; this embodiment does not impose any limitations on this.

[0054] It should be noted that, Figure 1 or Figure 3 The number of electronic expansion valves and indoor units shown are merely examples and do not constitute a specific limitation on the embodiments of this application.

[0055] Taking multiple electronic expansion valves operating independently of multiple indoor units 13 as an example, Figure 4 A schematic diagram of the refrigeration cycle principle of a multi-split air conditioning system is shown.

[0056] likeFigure 4 As shown, the multi-split air conditioning system includes an outdoor unit 11, a throttling device 12, a plurality of indoor units 13, and a controller 14 (not shown). Figure 4

[0057] The outdoor unit 11 includes a compressor 111, an outdoor heat exchanger 112, an accumulator 113, and a four-way valve 114. In some embodiments, the outdoor unit 11 further includes one or more of the following:

[0058] An outdoor fan, and an outdoor fan motor.

[0059] In some embodiments, the throttling device 12 is configured to regulate the flow rate of the fluid in the gas pipe 15 and the liquid pipe 16 in the multi-split air conditioning system.

[0060] The indoor unit 13 includes an indoor heat exchanger 131, a display 132, and an indoor fan 133. In some embodiments, the indoor unit 13 further includes an indoor fan motor.

[0061] In some embodiments, the compressor 111 is configured between the throttling device 12 and the accumulator 113, for compressing the refrigerant delivered by the accumulator 113 and delivering the compressed refrigerant to the throttling device 12 via the four-way valve 114. The compressor 111 can be an inverter compressor with variable capacity based on inverter-based speed control.

[0062] In some embodiments, the controller 14 can obtain the working current value (also referred to as the compressor current value) of the compressor 111 at each time point.

[0063] In some embodiments, one end of the outdoor heat exchanger 112 is connected to the accumulator 113 via the four-way valve 114, and the other end is connected to the throttling device 12. The outdoor heat exchanger 112 has a first port for allowing the refrigerant to flow between the outdoor heat exchanger 112 and the suction port of the compressor 111 via the accumulator 113, and has a second port for allowing the refrigerant to flow between the outdoor heat exchanger 112 and the throttling device 12. The outdoor heat exchanger 112 exchanges heat between the outdoor air and the heat transfer pipe connected between the first port and the second port, and functions as a condenser in the cooling cycle.

[0064] In some embodiments, one end of the accumulator 113 is connected to the compressor 111, and the other end is connected to the outdoor heat exchanger 112 via the four-way valve 114. In the accumulator 113, the refrigerant flowing from the outdoor heat exchanger 112 to the compressor 111 via the four-way valve 114 is separated into gaseous refrigerant and liquid refrigerant. And, the gaseous refrigerant is mainly supplied to the suction port of the compressor 111 from the accumulator 113.

[0065] ​In some embodiments, the four ports of the four-way valve 114 are connected to the compressor 111, the outdoor heat exchanger 112, the accumulator 113, and the plurality of electronic expansion valves 121, respectively. The four-way valve 114 is used to achieve mutual conversion between cooling and heating by changing the flow direction of the refrigerant in the system piping.

[0066] In some embodiments, the outdoor fan generates an air flow of outdoor air through the outdoor heat exchanger 112 to facilitate heat exchange between the refrigerant flowing in the heat transfer pipe between the first and second ports and the outdoor air.

[0067] In some embodiments, the outdoor fan motor is used to drive or change the rotation speed of the outdoor fan.

[0068] In some embodiments, the electronic expansion valve 121 has a function of expanding and depressurizing the refrigerant flowing therethrough, and can be used to adjust the supply amount of the refrigerant in the piping. If the opening degree of the electronic expansion valve 121 is reduced, the flow path resistance of the refrigerant passing through the electronic expansion valve 121 is increased. If the opening degree of the electronic expansion valve 121 is increased, the flow path resistance of the refrigerant passing through the electronic expansion valve 121 is reduced.

[0069] In some embodiments, the indoor heat exchanger 131 has a third port for passing liquid refrigerant between the electronic expansion valve 121, and has a fourth port for passing gaseous refrigerant between the discharge port of the compressor 111. The indoor heat exchanger 131 exchanges heat between the refrigerant flowing in the heat transfer pipe connected between the third and fourth ports and the indoor air.

[0070] In some embodiments, the indoor fan 133 generates an air flow of indoor air through the indoor heat exchanger 131 to facilitate heat exchange between the refrigerant flowing in the heat transfer pipe between the third and fourth ports and the indoor air.

[0071] In some embodiments, the indoor fan motor is used to drive or change the rotation speed of the indoor fan 133.

[0072] In some embodiments, the display 132 is used to display the indoor temperature or the current operation mode.

[0073] In the embodiments shown in this application, controller 14 refers to a device that can generate operation control signals according to instruction opcodes and timing signals, instructing the multi-split air conditioning system to execute control commands. Exemplarily, the controller 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 does not impose any limitations on this.

[0074] In addition, the controller 14 can be used to control the operation of each component inside the multi-split air conditioning system 10 so that each component of the multi-split air conditioning system 10 can operate to achieve the predetermined functions of the multi-split air conditioning system.

[0075] In some embodiments, the multi-split air conditioning system 10 is also equipped with a remote control, which has the function of communicating with the controller 14, for example, using infrared or other communication methods. The remote control is used by the user to perform various controls on the multi-split air conditioning system, realizing interaction between the user and the multi-split air conditioning system 10.

[0076] Reference Figure 5 This is a schematic diagram of the structure of a controller provided in an embodiment of this application. Figure 5 As shown, the controller 14 includes an outdoor control module 141 and an indoor control module 142. The outdoor control module 141 includes a first memory 1411, and the indoor control module 142 includes a second memory 1421. The indoor control module 142 is connected to the outdoor control module 141 via wired or wireless communication. The outdoor control module 141 can be installed in the outdoor unit 11 or independently of the outdoor unit 11, and is used to control the outdoor unit 11 to perform related operations. The indoor control module 142 can be installed in the indoor unit 13 or independently of the indoor unit 13, and is used to control the components of the indoor unit 13 and the throttling device 12 to perform related operations. It should be understood that the above module division is only functional; the outdoor control module 141 and the indoor control module 142 can also be integrated into one module. The first memory 1411 and the second memory 1421 can also be integrated into one memory.

[0077] In some embodiments, the first memory 1411 is used to store applications and data related to the outdoor unit 11, and the outdoor control module 141 performs various functions and data processing of the multi-split air conditioning system by running the applications and data stored in the memory 1411.

[0078] In some embodiments, the second memory 1421 is configured to store application programs and data related to the plurality of indoor units 13 and the plurality of electronic expansion valves 121, and the indoor control module 142 performs various functions and data processing of the multi-split air conditioning system by running the application programs and data stored in the second memory 1421. The second memory 1421 mainly includes a program storage area and a data storage area. In some examples, the second memory 1421 is also configured to store the correspondence between the addresses of the indoor units 13 and the addresses of the electronic expansion valves 121.

[0079] In some embodiments, the outdoor control module 141 is in communication connection with the outdoor unit 11, and is configured to control the outdoor unit to perform relevant operations according to user instructions or system default instructions. Optionally, the outdoor control module 141 can control the rotation speed of the outdoor fan according to the air conditioning operation mode selected by the user.

[0080] Optionally, the outdoor control module 141 can also obtain the outdoor temperature according to user instructions or system instructions, and store the obtained outdoor temperature to the first memory 1411.

[0081] Optionally, the outdoor control module 141 can also control the rotation of the four-way valve 114 in the outdoor unit 11 according to the air conditioning operation mode selected by the user, so as to realize the selection of the cooling or heating mode.

[0082] Optionally, the outdoor control module 141 can also control the operation mode, compressor frequency, etc. of the outdoor unit 11 during the address correction process.

[0083] In some embodiments, the indoor control module 142 is in communication connection with the indoor unit 13, and is configured to control the indoor unit 13 to perform relevant operations according to user instructions or system default instructions.

[0084] In some embodiments, the indoor control module 142 is in communication connection with the plurality of electronic expansion valves 121, and is configured to control the plurality of electronic expansion valves 121 to perform relevant operations according to user instructions or system default instructions.

[0085] It should be understood that the above description of the multi-split air conditioning system is based on the example that the throttling device 12 is independent of the plurality of indoor units 13. If the throttling device 12 is located within the plurality of indoor units 13, the above-mentioned cooling cycle principle of the multi-split air conditioning system is still applicable, and the following will not be repeated. Figure 4 The embodiments shown are based on the example that the throttling device 12 is independent of the plurality of indoor units 13. If the throttling device 12 is located within the plurality of indoor units 13, the above-mentioned cooling cycle principle of the multi-split air conditioning system is still applicable, and the following will not be repeated.

[0086] Figure 6 Fig. 2 shows a hardware configuration block diagram of a multi-split air conditioning system according to an exemplary embodiment of the present application. As shown in Fig. 2, the multi-split air conditioning system comprises an outdoor unit 11, a plurality of indoor units 13, a plurality of electronic expansion valves 121, a throttling device 12, an outdoor control module 141, and an indoor control module 142. Figure 6As shown, the multi-split air conditioning system 10 can further include one or more of: a plurality of first temperature sensors 101, a plurality of second temperature sensors 102, a plurality of third temperature sensors 103, a plurality of fourth temperature sensors 104, a fifth temperature sensor 105, a sixth temperature sensor 106, a plurality of first pressure sensors 107, a second pressure sensor 108, and a communicator 109.

[0087] In some embodiments, the plurality of first temperature sensors 101, the plurality of second temperature sensors 102, the plurality of third temperature sensors 103, the plurality of fourth temperature sensors 104, the plurality of fourth temperature sensors 105, the plurality of fourth temperature sensors 106, and the plurality of fourth temperature sensors 107 are connected to the controller 14, in combination Figure 2 As shown, the schematic diagram of the electronic expansion valve setting position is as follows: Figure 7 As shown, the first temperature sensor 101 can be arranged on the gas pipe 15 to detect the temperature value of the gas pipe 15, the second temperature sensor 102 can be arranged on the liquid pipe 16 to detect the temperature value of the liquid pipe 16, the third temperature sensor 103 can be arranged at the air outlet of the indoor unit 13 to detect the air outlet temperature value of the indoor unit 13, the fourth temperature sensor 103 can be arranged at the air return inlet of the indoor unit 13 to detect the air return temperature value of the indoor unit 13, the fifth temperature sensor 105 can be arranged at the compressor 111 of the outdoor unit 11 to detect the suction temperature value and the discharge temperature value of the compressor 111, the sixth temperature sensor 106 can be arranged at the refrigerant discharge pipe of the outdoor unit 11 to detect the discharge temperature value of the outdoor unit 11, and send the detected temperature value to the controller 14.

[0088] In some embodiments, the plurality of first pressure sensors 107 are each connected to the controller 14, the first pressure sensor 107 can be arranged in the indoor unit 13 to detect the high pressure value and the low pressure value of the indoor unit 13, the second pressure sensor 108 can be arranged at the refrigerant discharge pipe of the outdoor unit 11 to detect the discharge pressure value of the outdoor unit 11, and send the detected high pressure value and low pressure value to the controller 14.

[0089] In some embodiments, the communicator 109 is connected with the controller 14, and is configured to establish a communication connection with other network entities, such as a terminal device. The communicator 109 can include a radio frequency (RF) module, a cellular module, a wireless fidelity (Wi-Fi) module, a GPS module, and the like. Taking the RF module as an example, the RF module can be configured to receive and send signals, in particular, send the received information to the controller 14 for processing, and send the signals generated by the controller 14. Generally, the RF circuit can include, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (LNA), a duplexer, and the like.

[0090] For example, the multi-split air conditioning system 10 can receive a control instruction sent by a terminal device through the communicator 109, and perform corresponding processing according to the control instruction, so as to realize the interaction between the user and the multi-split air conditioning system 10.

[0091] Those skilled in the art can understand that, Figure 6 The hardware structure shown in the above figure does not constitute a limitation on the multi-split air conditioning system, and the multi-split air conditioning system can include more or fewer components than the figure, or combine certain components, or different component arrangements.

[0092] Figure 8 An interaction diagram between the controller 14 of a multi-split air conditioning system according to an exemplary embodiment of the present application and a terminal device 300 is shown in the figure.

[0093] As Figure 8 shown, the terminal device 300 can establish a communication connection with the controller 14 of the air conditioning system. Exemplarily, any known network communication protocol can be used to establish the communication connection. The above network communication protocol can be various wired or wireless communication protocols, such as Ethernet, universal serial bus (USB), FIREWIRE, any cellular network communication protocol (such as 3G / 4G / 5G), Bluetooth, wireless fidelity (Wi-Fi), NFC, or any other suitable communication protocol. The above communication connection can be a Bluetooth connection, NFC, zigbee, wireless fidelity (Wi-Fi), and the like. The embodiments of the present application do not make specific limitations on this.

[0094] It should be noted that, Figure 8The terminal device 300 shown is only one example of a terminal device. The terminal device 300 in this application can be a remote controller, a mobile phone, a tablet computer, a personal computer (PC), a personal digital assistant (PDA), a smart watch, a netbook, a wearable electronic device, an augmented reality (AR) device, a virtual reality (VR) device, a robot, etc. The specific form of the terminal device is not specially limited in this application.

[0095] For example, the terminal device 300 is a mobile phone. The user can download a smart home APP on the mobile phone. The smart home APP can be used to manage smart home devices. The embodiments of this application take the air conditioning system 10 as an example. Further, the user can select the air conditioning system 10 as an online device and select the control function to be performed on the air conditioning system 10 in the management options of the air conditioning system 10. For example, the control functions include starting, stopping, switching modes (such as monitoring mode, intrusion detection mode), etc.

[0096] In some embodiments, the running mode of the air conditioning system can be set through the terminal device 300. For example, as shown in the figure, Figure 9 The management interface 301 of the multi-split air conditioning system is displayed on the terminal device 300. The management interface 301 includes a "mode management" button 302. When the user clicks the "mode management" button 302 in the management interface 301, the terminal device pops up a running mode drop-down selection box 303 in the management interface 301. After the terminal device 300 detects the selection instruction of the user in the running mode drop-down selection box 303, it sends the instruction to the multi-split air conditioning system to complete the setting of the running mode. For example, the user can select the "cooling" mode.

[0097] In some embodiments, the user can start the fault detection function and the fault positioning function through the management interface of the terminal device 300. For example, as shown in the figure, Figure 10 The management interface 301 of the terminal device includes "fault detection" and "fault positioning" buttons, Figure 10 The "fault detection" button is in the off state as shown by the button 3041. The terminal device detects the user's click on the "fault detection" button and changes the state of the "fault detection" button to the on state as shown by 3042. The terminal device transmits the instruction to start fault detection to the controller of the multi-split air conditioning system, so that the multi-split air conditioning system enters the fault detection state.

[0098] In some embodiments, after detecting that there is an electronic expansion valve fault in the multi-split air conditioning system, the controller 14 sends a fault existing instruction to the terminal device 300 through the communicator 109. After receiving the fault existing instruction, the terminal device 300 displays prompt information such as "a fault of the electronic expansion valve is detected in the multi-split air conditioning system. Do you want to start fault positioning immediately?" on the management interface of the terminal device 300 to prompt the user to select whether to start the fault positioning function. If the user selects and clicks the icon of the "confirm" function button, it means that the user selects to start fault positioning immediately. The terminal device 300 sends a confirmation instruction to the controller 14 in response to the user's confirmation instruction. After receiving the confirmation instruction, the controller 14 starts the fault positioning function. Figure 11

[0099] In some embodiments, if the user selects and clicks the icon of the "cancel" function button, it means that the user selects to perform fault positioning at a later time. The user can click the icon of the "fault positioning" button at a later time to send a fault positioning instruction to the controller 14.

[0100] The embodiments provided by the present application will be described in detail below with reference to the accompanying drawings.

[0101] As shown in Figure 12 , the present application provides a fault detection method of a multi-split air conditioning system, which is applied to the controller of the multi-split air conditioning system as shown in Figure 1 , and the method comprises the following steps.

[0102] S101, the controller acquires real-time running data of the multi-split air conditioning system.

[0103] It should be noted that after the multi-split air conditioning system is started, the problem of the electronic expansion valve fault may occur. At this time, it is necessary to perform fault detection on all electronic expansion valves. However, since there are multiple electronic expansion valves in the multi-split air conditioning system, if the fault electronic expansion valve is not positioned, the fault electronic expansion valve cannot be found in time for early repair. Therefore, the real-time running data of the multi-split air conditioning system is acquired to perform fault detection and positioning on the electronic expansion valve.

[0104] In some embodiments, after the multi-split air conditioning system is started, the controller acquires the real-time running data of the multi-split air conditioning system periodically at a first preset time interval. Therefore, the real-time running data of the air conditioning system acquired by the controller in step S101 is the real-time running data acquired by the controller in one of the periods. In this way, the controller can perform the fault detection method provided by the present application once every time the real-time running data of the air conditioning system is acquired, so that the air conditioning system can be periodically detected for faults.

[0105] ​It should be noted that the first preset time length is set by the multi-split air conditioning system manufacturer and is pre-stored in the memory. The first preset time length of different manufacturers can be changed, and the present application does not limit this.

[0106] The real-time running data of the multi-split air conditioning system includes running data of the outdoor unit and running data of each indoor unit.

[0107] Specifically, the running data of the outdoor unit includes compressor frequency, pressure value, pressure top temperature value, exhaust gas superheat value, defrosting temperature value, and ambient temperature value.

[0108] Further, the running data of each indoor unit includes variable frequency heat dissipation value, liquid pipe temperature value, gas pipe temperature value, return air temperature value, and outlet air temperature value.

[0109] S102, the controller identifies the fault by inputting the real-time running data into the preset graph attention model, and obtains a fault identification result.

[0110] Optionally, the graph attention model includes a fault detection model and a fault positioning model. The fault detection model is used to detect whether the electronic expansion valve is in a fault running state. The fault positioning model is used to locate the indoor unit serial number of the electronic expansion valve in the fault running state.

[0111] In some embodiments, when the controller obtains the real-time running data of the multi-split air conditioning system, the electronic expansion valve is identified by inputting the real-time running data into the preset graph attention model, and a fault identification result is obtained.

[0112] The normal running data is the running data of the multi-split air conditioning system when the electronic expansion valve is in a normal running state; and the fault running data is the running data of the multi-split air conditioning system when the electronic expansion valve is in a fault running state. The fault identification result includes that the electronic expansion valve in the multi-split air conditioning system is in a fault running state or a normal running state.

[0113] Optionally, the preset graph attention model can be a fault detection model in a graph attention model (GAT).

[0114] Illustratively, the controller inputs the real-time running data of the multi-split air conditioning system into the fault detection GAT model to identify the fault of the electronic expansion valve, and obtains a fault identification result.

[0115] In some embodiments, the GAT model is obtained by training through the controller. Figure 13 A GAT model training method flowchart is provided for the embodiments of the present application. As shown in the figure, the method includes: Figure 13

[0116] ​S11, the controller acquires normal operation data and fault operation data of the multi-split air conditioning system.

[0117] In some embodiments, training the GAT model can improve the accuracy of the electronic expansion valve fault identification result and the fault positioning result.

[0118] Further, the controller inputs the normal operation data and the fault operation data of the multi-split air conditioning system into the GAT model for training, which can enable the GAT model to acquire the operation data features of the electronic expansion valve in the normal operation state and the operation data features of the electronic expansion valve in the fault operation state.

[0119] Therefore, the presence or absence of fault operation data in the real-time operation data of the multi-split air conditioning system can be more accurately identified.

[0120] It can be understood that the normal operation data is the multi-split air conditioning system operation data when the electronic expansion valve is in the normal operation state. The fault operation data is the multi-split air conditioning system operation data when the electronic expansion valve is in the fault operation state.

[0121] The data content contained in the normal operation data and the fault operation data is consistent with the above-mentioned real-time operation data, which will not be described here.

[0122] Optionally, the training principle of the GAT model can be a graph neural network model (GNN).

[0123] S12, the controller trains the initial graph attention model according to the normal operation data and the fault operation data to obtain a preset graph attention model after training.

[0124] In some embodiments, before the controller inputs the normal operation data and the fault operation data into the initial graph attention model to obtain the preset graph attention model after training, the controller converts the normal operation data and the fault operation data into a graph structure through the GNN model, performs a node aggregation operation on the graph structure, iteratively updates the GAT model parameters, and finally outputs the fault detection and positioning result of the electronic expansion valve.

[0125] Figure 14 Another method flowchart for training the GAT model provided by the embodiments of the present application is provided. As shown in the method, the method comprises: Figure 14

[0126] S21, the controller converts the normal operation data and the fault operation data into a graph structure through a preset graph neural network model.

[0127] ​In some embodiments, before the controller trains the initial graph attention model according to the normal operation data and the fault operation data, the controller converts the normal operation data and the fault operation data into a graph structure through a preset graph neural network model.

[0128] The graph structure includes original node feature vectors and connection relationships between the original node feature vectors. The original feature vectors are vector representations of the normal operation data and the fault operation data.

[0129] Optionally, the connection relationships between the original node feature vectors can be exhibited in the form of edges, and the edges are represented in the form of an adjacency matrix.

[0130] In some embodiments, Pearson correlation coefficient calculation is performed on different data information between two original node feature vectors. If the Pearson correlation coefficient is greater than a preset threshold, the two original node feature vectors are connected by an edge, that is, the two original node feature vectors are related. If the Pearson correlation coefficient is less than the preset threshold, the two original node feature vectors are not connected by an edge, that is, the two original node feature vectors are not related.

[0131] Optionally, the preset threshold can be 0.5.

[0132] For example, if the Pearson correlation coefficient A is greater than the preset threshold, that is, A>0.5, the two original node feature vectors are connected by an edge. If A<0.5, the two original node feature vectors are not connected by an edge.

[0133] For example, the graph structure can be represented in the form shown in formula (1):

[0134] AM = [[S1, T1], [S2, T2], [S3, T3], …, [Sn, Tn]] Formula (1) n n

[0135] Wherein, AM is an adjacency matrix, S and T are different operation data, and n is the position of the operation data.

[0136] In some embodiments, after the controller converts the normal operation data and the fault operation data into a graph structure, the controller converts the original node feature vectors into target node feature vectors according to the graph structure, and assigns attention to the target node feature vectors.

[0137] Optionally, after the controller converts the normal operation data and the fault operation data into a graph structure, the controller inputs the original node feature vectors into an attention layer to obtain target node feature vectors through linear transformation.

[0138] ​​Furthermore, attention coefficients are assigned to the feature vectors of the target nodes through self-attention, as shown in equations (2) and (3):

[0139] x = {x1, x1, ..., x1} n} Formula (2)

[0140] x' = {x'1, x'2, ..., x'} n} Formula (3)

[0141] Where x is the original node feature vector and x' is the target node feature vector.

[0142] It should be noted that Self-attention is an important mechanism in neural network models, used to perform self-attention calculation on each element in the input sequence and obtain the self-attention representation of each element.

[0143] Optionally, the range matrix W of the target node feature vector satisfies the relationship between the attention coefficient e and the given relationship. ij The linear transformation is shown in formulas (4) and (5):

[0144] W∈R F’*F Formula (4)

[0145]

[0146] Where R is the feature space of the target node feature vector, F' is the dimension of the target node feature vector, F is the dimension of the original node feature vector, which is the influence coefficient of the target node feature vector i on the target node feature vector j, and a represents a single-layer feedforward neural network with a numerical output.

[0147] In some embodiments, once the attention coefficients are determined, the feature vectors of the target nodes are aggregated based on the attention coefficients.

[0148] For example, the feature vector V of the target node will be... i The surrounding related nodes {V a V b , ..., V n Weighted to V i This is an aggregation of the feature vectors of the target node.

[0149] It should be noted that, in this embodiment of the application, the feature vectors of all target nodes can be aggregated, and the aggregation method is shown in formula (6):

[0150] M i =G({w j *x j , j∈Ni}) Equation (6)

[0151] wherein G() is an aggregation function, w j is a weight, x j is a target node feature vector, N i is a set of node feature vectors connected to the target node feature vector.

[0152] In some embodiments, when the aggregation operation of the target node feature vector is completed, the preset graph attention model is updated through iteration of the preset graph neural network model, and the controller determines whether the original node feature vector reaches a preset convergence condition.

[0153] In some embodiments, when the controller determines that the original node feature vector does not reach the preset convergence condition, the attention coefficient is updated to obtain the influence relationship between the original node feature vectors.

[0154] Optionally, the updating method of the attention coefficient is shown in Equation (7):

[0155] L t+1 = F w (L t , x) Equation (7)

[0156] wherein L is the original node feature vector, F w is a compression mapping.

[0157] In some embodiments, when the controller determines that the original node feature vector does not reach the preset convergence condition, the training of the initial graph attention model is stopped to obtain the preset graph attention model after the training is completed.

[0158] Optionally, the judgment method of whether the state vector of the original node feature vector x reaches the convergence condition is shown in Equation (8):

[0159] ||x t -x t-1 |<ε F Equation (8)

[0160] In some embodiments, when the GAT model stops updating, the controller inputs the real-time running data of the multi-online air conditioning system into the GAT model to output real-time results. By comparing the running state of the electronic expansion valve and the indoor unit serial number in the real-time results with those in the model results, it is confirmed whether the GAT model is successfully trained.

[0161] For example, if the running state of the electronic expansion valve in the real-time results matches that in the model results, and the indoor unit serial number in the real-time results matches that in the model results, it is confirmed that the GAT model is successfully trained.

[0162] For example, if the running state of the electronic expansion valve in the real-time result matches the running state of the electronic expansion valve in the model result, but the indoor unit serial number in the real-time result does not match the indoor unit serial number in the model result, the GAT model is updated in parameters.

[0163] For example, if the running state of the electronic expansion valve in the real-time result does not match the running state of the electronic expansion valve in the model result, and the indoor unit serial number in the real-time result also does not match the indoor unit serial number in the model result, the GAT model is updated in parameters.

[0164] In some embodiments, after the GAT model stops updating, the real-time running data of the multi-split air conditioning system is input into a fault detection model in the GAT model, and after parameter updating through back propagation, a fault identification result is obtained.

[0165] S103, in the case that the electronic expansion valve in the multi-split air conditioning system is in a fault running state, the controller locates the fault of the electronic expansion valve through a preset graph attention model to determine the indoor unit serial number of the electronic expansion valve.

[0166] In some embodiments, when the fault identification result is that the electronic expansion valve in the multi-split air conditioning system is in a fault running state, each original node feature vector is input into a fault locating model in the GAT model to locate the indoor unit serial number of the electronic expansion valve in the fault running state.

[0167] The technical scheme provided by the embodiments of the present application at least brings the following beneficial effects: by inputting the obtained real-time running data of the multi-split air conditioning system into the preset graph attention model, a fault identification result of the electronic expansion valve in the multi-split air conditioning system is obtained, and the fault detection efficiency of the electronic expansion valve is improved. At the same time, when it is determined that the electronic expansion valve in the multi-split air conditioning system is in a fault running state, the preset graph attention model is used to locate the electronic expansion valve, and the indoor unit serial number of the first expansion valve is obtained. In this way, the dependence on prior knowledge and the data stream preprocessing process is reduced, and the accuracy of the fault locating result of the electronic expansion valve is improved.

[0168] In some embodiments, when the fault identification result is that the electronic expansion valve in the multi-split air conditioning system is in a normal running state, the real-time running data of the multi-split air conditioning system is obtained again after a second preset time period.

[0169] It should be noted that the second preset time period is set by the manufacturer of the multi-split air conditioning system and is pre-stored in the memory. The second preset time period of different manufacturers can vary, and the present application does not limit this.

[0170] In some embodiments, the controller updates the GAT model by actual operation data and states of the multi-connected air conditioning system in multiple stages and actual operation data and states of other units, so as to continuously update the GAT model.

[0171] In some embodiments, the above method can also be implemented through a flowchart as shown in Figure 15 The method includes the following steps as shown in Figure 15

[0172] S1, the controller acquires normal operation data and fault operation data of the multi-connected air conditioning system.

[0173] It should be noted that the normal operation data and fault operation data of the multi-connected air conditioning system contain the same content as the real-time operation data described above, which will not be repeated here.

[0174] The normal operation data is the operation data of the multi-connected air conditioning system when the electronic expansion valve is in a normal operation state, and the fault operation data is the operation data of the multi-connected air conditioning system when the electronic expansion valve is in a fault operation state.

[0175] S2, the controller inputs the normal operation data and the fault operation data into the initial GAT model.

[0176] Optionally, the initial GAT model is trained by the GNN model.

[0177] It should be noted that the training method of the GNN model on the initial GAT model is described in the above step S21, which will not be repeated here.

[0178] In some embodiments, the controller first inputs the normal operation data and the fault operation data into the fault detection model of the initial GAT model, converts the normal operation data and the fault operation data into a graph structure by the GNN model, and determines whether the electronic expansion valve of the multi-connected air conditioning system is in a fault operation state according to the graph structure.

[0179] Further, when the controller determines that the electronic expansion valve of the multi-connected air conditioning system is in a fault operation state, the controller inputs the normal operation data and the fault operation data into the fault positioning model of the initial GAT model, and locates the indoor unit serial number of the electronic expansion valve in the fault operation state.

[0180] S3, when the fault state of the electronic expansion valve output by the initial GAT model is incorrect, or the fault state of the electronic expansion valve output by the GAT model is correct but the indoor unit serial number of the electronic expansion valve in the fault operation state is incorrect, the controller updates the parameters of the initial GAT model.

[0181] ​S4, when the electronic expansion valve fault state output by the initial GAT model and the indoor unit serial number of the electronic expansion valve in the fault operation state are both correct, the initial GAT model is successfully trained into a preset GAT model.

[0182] S5, the controller inputs the real-time operation data of the multi-split air conditioner into the preset GAT model.

[0183] In some embodiments, the controller first inputs the real-time operation data into a fault detection model of the preset GAT model, converts the real-time operation data into a graph structure through a GNN model, and determines whether the electronic expansion valve of the multi-split air conditioner system is in a fault operation state according to the graph structure.

[0184] Further, when the controller determines that the electronic expansion valve of the multi-split air conditioner system is in a fault operation state, the controller inputs the real-time operation data into a fault positioning model of the preset GAT model to locate the indoor unit serial number of the electronic expansion valve in the fault operation state.

[0185] S6, when the electronic expansion valve fault state output by the preset GAT model and the indoor unit serial number of the electronic expansion valve in the fault operation state are both correct, the controller controls the preset GAT model to learn; when the electronic expansion valve fault state output by the initial GAT model is incorrect, or the electronic expansion valve fault state output by the GAT model is correct but the indoor unit serial number of the electronic expansion valve in the fault operation state is incorrect, the controller controls to execute the above step S3.

[0186] The embodiment of the present application also provides a computer readable storage medium, which comprises computer execution instructions, and when the computer execution instructions run on the computer, the computer executes the method provided by the above embodiment.

[0187] The embodiment of the present application also provides a computer program product, which can be directly loaded into the memory and contains software codes, and the computer program product can realize the method provided by the above embodiment after being loaded and executed by the computer.

[0188] Those skilled in the art should realize that, in one or more of the above examples, the functions described in the present application can be realized by hardware, software, firmware or any combination thereof. When realized by 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 a computer storage medium and a communication medium, wherein the communication medium includes any medium facilitating the transmission of computer programs from one place to another. The storage medium can be any available medium accessible by a general or special purpose computer.

[0189] Through the description of the above embodiments, those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.

[0190] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only exemplary, such as the division of modules or units, which is only a logical function division, and actual implementation can have another division manner. For example, a plurality of units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms. The units described as separate components can be or can not be physically separated, and the components shown as units can be one physical unit or a plurality of physical units, that is, they can be located in one place or distributed in a plurality of different places. Some or all units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0191] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit. When the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application essentially or say the parts that make contributions to the prior art or all or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium, including a plurality of instructions to make a device (which can be a single chip, a chip, etc.) or a processor execute all or part of the steps of the method of each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a ROM, a RAM, a magnetic disk or an optical disk, and various storage medium that can store program codes.

[0192] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any change or replacement within the technical scope disclosed in the present application should be covered in 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. A multi-split air conditioning system, characterized in that, Comprise: An outdoor unit; At least one indoor unit, each indoor unit being provided with an electronic expansion valve between the indoor unit and the outdoor unit; A controller configured to: Obtain real-time operation data of the multi-split air conditioning system; The real-time operation data includes: real-time operation data of the outdoor unit and real-time operation data of the indoor unit; wherein the real-time operation data of the outdoor unit includes compressor frequency, pressure value, pressure top temperature value, exhaust gas superheat value, defrosting temperature value and ambient temperature value; the real-time operation data of the indoor unit includes variable frequency heat dissipation value, liquid pipe temperature value, gas pipe temperature value, return air temperature value and outlet air temperature value; Through a preset graph attention model, the real-time operation data is identified to obtain a fault identification result, the fault identification result including that the electronic expansion valve in the multi-split air conditioning system is in a fault operation state or a normal operation state; wherein the preset graph neural network model is trained by converting normal operation data and fault operation data into a graph structure, the graph structure including original node feature vectors and connection relationships between the original node feature vectors, the connection relationships being determined by Pearson correlation coefficient calculation; In the case that the electronic expansion valve in the multi-split air conditioning system is in a fault operation state, the preset graph attention model is used to locate the fault of the electronic expansion valve to determine the indoor unit serial number of the electronic expansion valve.

2. The multi-split air conditioning system according to claim 1, wherein, The controller is further configured to: Obtain normal operation data and fault operation data of the multi-split air conditioning system; wherein the normal operation data is operation data of the multi-split air conditioning system when the electronic expansion valve is in a normal operation state; the fault operation data is operation data of the multi-split air conditioning system when the electronic expansion valve is in a fault operation state; According to the normal operation data and the fault operation data, an initial graph attention model is trained to obtain the preset graph attention model after training.

3. The multi-split air conditioning system according to claim 2, wherein, Before the controller is configured to train the initial graph attention model according to the normal operation data and the fault operation data, the controller is further configured to: Convert the normal operation data and fault operation data into a graph structure by a preset graph neural network model; the graph structure includes original node feature vectors and connection relationships between the original node feature vectors; wherein the original node feature vectors are vector representations of the normal operation data and the fault operation data.

4. The multi-split air conditioning system according to claim 3, wherein, The controller is further configured to: Perform Pearson correlation coefficient calculation on different data information between two original node feature vectors; If the Pearson correlation coefficient is greater than a preset threshold, the two original node feature vectors are connected by an edge; If the Pearson correlation coefficient is less than the preset threshold, the two original node feature vectors are not connected by an edge.

5. The multi-split air conditioning system according to claim 3, wherein, The controller is configured to train the initial graph attention model according to the normal operation data and the fault operation data, and is specifically configured to: convert the original node feature vector into a target node feature vector, and assign an attention coefficient to the target node feature vector; perform an aggregation operation on the target node feature vector according to the attention coefficient; update the preset graph attention model through iteration of the preset graph neural network model to determine whether the original node feature vector reaches a preset convergence condition; update the attention coefficient in a case where the original node feature vector does not reach the preset convergence condition; stop training the initial graph attention model in a case where the original node feature vector reaches the preset convergence condition, to obtain the preset graph attention model after training is completed.

6. A method for detecting a failure of a multi VRF system, the method comprising: The method is applied to a multi-split air conditioning system, the multi-split air conditioning system includes an outdoor unit and a plurality of indoor units, an electronic expansion valve is arranged between each indoor unit and the outdoor unit, and the method includes: obtaining real-time operation data of the multi-split air conditioning system; the real-time operation data includes real-time operation data of the outdoor unit and real-time operation data of the indoor unit; wherein the real-time operation data of the outdoor unit includes a compressor frequency, a pressure value, a pressure top temperature value, an exhaust gas superheat value, a defrosting temperature value and an ambient temperature value; the real-time operation data of the indoor unit includes a variable frequency heat dissipation value, a liquid pipe temperature value, a gas pipe temperature value, a return air temperature value and an outlet air temperature value; performing fault identification on the real-time operation data through a preset graph attention model to obtain a fault identification result, the fault identification result including that an electronic expansion valve in the multi-split air conditioning system is in a fault operation state or a normal operation state; wherein the preset graph attention model is trained by converting normal operation data and fault operation data into a graph structure through a preset graph neural network model, the graph structure including original node feature vectors and a connection relationship between the original node feature vectors, and the connection relationship being determined through Pearson correlation coefficient calculation; in a case where the electronic expansion valve in the multi-split air conditioning system is in a fault operation state, performing fault positioning on the electronic expansion valve through the preset graph attention model to determine an indoor unit serial number of the electronic expansion valve.

7. The method of claim 6, wherein, The method further includes: obtaining normal operation data and fault operation data of the multi-split air conditioning system; wherein the normal operation data is operation data of the multi-split air conditioning system when the electronic expansion valve is in a normal operation state, and the fault operation data is operation data of the multi-split air conditioning system when the electronic expansion valve is in a fault operation state; training an initial graph attention model according to the normal operation data and the fault operation data to obtain the preset graph attention model after training is completed.

8. The method of claim 7, wherein, The method further includes: convert the original node feature vector into a target node feature vector, and assign an attention coefficient to the target node feature vector; perform an aggregation operation on the target node feature vector according to the attention coefficient; update the preset graph attention model through iteration of the preset graph neural network model to determine whether the original node feature vector reaches a preset convergence condition; In a case where the original node feature vector does not reach a preset convergence condition, the attention coefficient is updated; In a case where the original node feature vector reaches the preset convergence condition, the training of the initial graph attention model is stopped to obtain the preset graph attention model after the training is completed.

Images