Multi-split air conditioning system, fault positioning method and fault diagnosis model training method
Patent Information
- Application Number
- CN202380013716.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-04-11
- Filing Date
- 2023-01-09
- Publication Date
- 2024-11-26
AI Technical Summary
The fault location efficiency of the electronic expansion valve in the multi-split air conditioning system is low, and it is difficult to quickly identify the faulty component, resulting in low fault location efficiency.
By monitoring the temperature difference between the liquid pipe and the air pipe of the indoor unit, the controller is used to obtain characteristic data, identify abnormal indoor units and determine the faulty electronic expansion valve, thereby improving the automation and efficiency of fault location.
The faulty electronic expansion valve can be accurately located without the need for professional maintenance personnel, improving the accuracy and automation of fault location and improving the fault location efficiency of the multi-split air conditioning system.
Smart Images

Figure CN119032239A_ABST
Abstract
Description
Multi-split air conditioning system, fault location method and fault diagnosis model training method
[0001] This application claims priority to the Chinese patent application with application number 202210386517.3 filed on April 11, 2022, and priority to the Chinese patent application with application number 202210682549.8 filed on June 16, 2022; the entire contents of which are incorporated by reference into this application. Technical Field
[0002] The present disclosure relates to the technical field of air conditioning, and in particular to a multi-split air conditioning system, a fault location method, and a fault diagnosis model training method. Background Art
[0003] A multi-split air conditioning system includes an outdoor unit and multiple indoor units, thereby regulating the temperature of multiple indoor spaces where the indoor units are located. Due to the large number of components in a multi-split air conditioning system, when a malfunction occurs, it is difficult to quickly locate the faulty component, resulting in low fault location efficiency.
[0004] Summary of the Invention
[0005] In one aspect, some embodiments of the present disclosure provide a multi-split air conditioning system. The multi-split air conditioning system includes an outdoor unit, multiple indoor units, multiple electronic expansion valves, and a controller. Each of the multiple indoor units is connected to a corresponding gas pipe and a liquid pipe, communicating with the outdoor unit via the gas pipe and the liquid pipe. The multiple electronic expansion valves correspond to the multiple indoor units. The multiple electronic expansion valves are respectively disposed in the liquid pipes connected to the corresponding indoor units and are configured to control the refrigerant output or refrigerant input of the corresponding indoor units.
[0006] The controller is configured to, in the event of an electronic expansion valve failure in a multi-split air conditioning system, obtain characteristic data for each of a plurality of indoor units. The characteristic data for each indoor unit includes a temperature difference between a temperature value of a liquid pipe connected to the indoor unit and a temperature value of a gas pipe connected to the indoor unit. Based on the characteristic data for each of the plurality of indoor units, an abnormal indoor unit among the plurality of indoor units is identified. The electronic expansion valve corresponding to the abnormal indoor unit is identified as a faulty electronic expansion valve.
[0007] In another aspect, some embodiments of the present disclosure provide a fault location method for a multi-split air conditioning system. The multi-split air conditioning system includes an outdoor unit, multiple indoor units, and multiple electronic expansion valves. Each of the multiple indoor units is connected to a corresponding gas pipe and a liquid pipe, communicating with the outdoor unit via the gas pipe and the liquid pipe. The multiple electronic expansion valves correspond to the multiple indoor units. The multiple electronic expansion valves are respectively disposed in the liquid pipes connected to the corresponding indoor units and are configured to control the refrigerant output or refrigerant input of the corresponding indoor units.
[0008] The method includes: when an electronic expansion valve failure occurs in a multi-split air conditioning system, obtaining characteristic data of each of a plurality of indoor units. The characteristic data of each indoor unit includes a temperature difference between a temperature value of a liquid pipe connected to the indoor unit and a temperature value of a gas pipe connected to the indoor unit. Based on the characteristic data of each of the plurality of indoor units, an abnormal indoor unit among the plurality of indoor units is identified. The electronic expansion valve corresponding to the abnormal indoor unit is identified as a faulty electronic expansion valve.
[0009] In another aspect, some embodiments of the present disclosure provide a method for training a fault diagnosis model for a multi-split air conditioning system. The training method includes: obtaining normal values and fault values of multiple operating data of a first multi-split air conditioning system. Based on the normal values and fault values of the multiple operating data of the first multi-split air conditioning system, determining a first feature offset space for the first multi-split air conditioning system. The first feature offset space includes the difference between the fault value and the normal value of each operating data in the multiple operating data of the first multi-split air conditioning system. Based on the difference between the normal values of the multiple operating data of the first multi-split air conditioning system and the normal values of the multiple operating data of a second multi-split air conditioning system, the first feature offset space is corrected to obtain a second feature offset space for the second multi-split air conditioning system. Based on the normal values of the multiple operating data of the second multi-split air conditioning system and the second feature offset space, the fault values of the multiple operating data of the second multi-split air conditioning system are determined. The fault diagnosis model is trained based on the normal values and fault values of the multiple operating data of the second multi-split air conditioning system. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] To more clearly illustrate the technical solutions of the present disclosure, the following briefly introduces the drawings required for use in some embodiments of the present disclosure. However, the drawings described below are only drawings of some embodiments of the present disclosure, and those skilled in the art can also derive other drawings based on these drawings. In addition, the drawings described below can be regarded as schematic diagrams and are not intended to limit the actual size of the products involved in the embodiments of the present disclosure, the actual process of the method, the actual timing of the signals, etc.
[0011] FIG1 is a structural diagram of a multi-split air conditioning system according to some embodiments;
[0012] FIG2 is another structural diagram of a multi-split air conditioning system according to some embodiments;
[0013] FIG3 is a block diagram of a controller according to some embodiments;
[0014] FIG4 is a diagram illustrating an interaction method between a controller and a terminal device according to some embodiments;
[0015] FIG5 is an interface diagram of a terminal device according to some embodiments;
[0016] FIG6 is a flow chart of a fault location method for a multi-split air conditioning system according to some embodiments;
[0017] FIG7 is another interface diagram of a terminal device according to some embodiments;
[0018] FIG8 is another interface diagram of a terminal device according to some embodiments;
[0019] FIG9 is a diagram illustrating the locations of a first temperature sensor and a second temperature sensor according to some embodiments;
[0020] FIG10 is another flow chart of a fault location method for a multi-split air conditioning system according to some embodiments;
[0021] FIG11 is another flow chart of a fault location method for a multi-split air conditioning system according to some embodiments;
[0022] FIG12 is another interface diagram of a terminal device according to some embodiments;
[0023] FIG13 is a flow chart of a fault diagnosis model training method for a multi-split air conditioning system according to some embodiments;
[0024] FIG14 is another flow chart of a fault diagnosis model training method for a multi-split air conditioning system according to some embodiments;
[0025] FIG15 is a structural diagram of a fault diagnosis model according to some embodiments;
[0026] FIG16 is another flowchart of a fault diagnosis model training method for a multi-split air conditioning system according to some embodiments;
[0027] FIG17 is another flowchart of a fault diagnosis model training method for a multi-split air conditioning system according to some embodiments;
[0028] FIG18 is a block diagram of an autoencoding model according to some embodiments. DETAILED DESCRIPTION
[0029] The following will be combined with the accompanying drawings to clearly and completely describe the technical solutions in some embodiments of the present disclosure. Obviously, the embodiments described are only some embodiments of the present disclosure, not all embodiments. Based on the embodiments provided by the present disclosure, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present disclosure.
[0030] Unless the context requires otherwise, throughout the specification and claims, the term "comprise" and its other forms, such as the third person singular form "comprises" and the present participle form "comprising", are to be interpreted as open and inclusive, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "example", "specific example" or "some examples" are intended to indicate that the specific features, structures, materials or characteristics associated with the embodiment or example are included in at least one embodiment or example of the present disclosure. The schematic representation of the above terms does not necessarily refer to the same embodiment or example. In addition, the specific features, structures, materials or characteristics may be included in any one or more embodiments or examples in any appropriate manner.
[0031] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present disclosure, unless otherwise specified, "plurality" means two or more.
[0032] When describing some embodiments, the expressions "coupled" and "connected" and their derivatives may be used. The term "connected" should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium. The term "coupled" indicates, for example, that two or more components are in direct physical or electrical contact. The term "coupled" or "communicatively coupled" may also refer to two or more components that are not in direct contact with each other, but still cooperate or interact with each other. The embodiments disclosed herein are not necessarily limited to the contents of this document.
[0033] “A and / or B” includes the following three combinations: A only, B only, and a combination of A and B.
[0034] The use of "configured to" herein is intended to be open and inclusive language that does not exclude devices being configured to perform additional tasks or steps.
[0035] Additionally, the use of “based on” is meant to be open and inclusive, as a process, step, calculation, or other action “based on” one or more stated conditions or values may, in practice, be based on additional conditions or values beyond those stated.
[0036] Some embodiments of the present disclosure provide a multi-split air-conditioning system 10. As shown in FIG1 , the multi-split air-conditioning system 10 includes an outdoor unit 11, a plurality of electronic expansion valves 12, a plurality of indoor units 13, and a controller 14 (not shown in FIG1 ). The outdoor unit 11 can be, for example, a device installed on the outside of a wall or on the roof of a house in the multi-split air-conditioning system 10. The outdoor unit 11 is mainly used to compress the refrigerant and drive the refrigerant to circulate in the multi-split air-conditioning system 10. The refrigerant is a substance that easily absorbs heat to become a gas and also easily releases heat to become a liquid. The indoor unit 13 can be, for example, a device installed indoors in the multi-split air-conditioning system 10. The indoor unit 13 is mainly used to transmit cold air or hot air to the indoor space where the indoor unit 13 is located to adjust the temperature of the indoor space.
[0037] In some embodiments, each of the plurality of indoor units 13 is connected to a corresponding gas pipe 15 and a liquid pipe 16, thereby communicating with the outdoor unit 11 via the gas pipe 15 and the liquid pipe 16. The gas pipe 15 is configured to transmit gaseous refrigerant or two-phase refrigerant (a refrigerant in both gaseous and liquid states) between the outdoor unit 11 and the indoor unit 13, and the liquid pipe 16 is configured to transmit liquid refrigerant or two-phase refrigerant between the outdoor unit 11 and the indoor unit 13. The gas pipe 15 and the liquid pipe 16 may be collectively referred to as a pipeline.
[0038] In some examples, multiple electronic expansion valves 12 correspond to multiple indoor units 13. The multiple electronic expansion valves 12 are respectively disposed in liquid pipes 16 connected to corresponding indoor units 13 and are configured to control the refrigerant output or refrigerant input of the corresponding indoor units 13.
[0039] It should be noted that the multiple electronic expansion valves 12 can be provided independently of the multiple indoor units 13 (as shown in FIG1 ), or can be provided as part of the multiple indoor units 13. The following embodiments are described using the example of multiple electronic expansion valves 12 being provided independently of the multiple indoor units 13.
[0040] In some embodiments, the outdoor unit 11 includes a compressor 111, an outdoor heat exchanger 112, a gas-liquid separator 113, a four-way valve 114, and an outdoor fan 115. The exhaust port of the compressor 111 is connected to the D end of the four-way valve 114, and the intake port of the compressor 111 is connected to the exhaust port of the gas-liquid separator 113. The intake port of the gas-liquid separator 113 is connected to the S end of the four-way valve 114. The C end of the four-way valve 114 is connected to the first end of the outdoor heat exchanger 112, and the E end of the four-way valve 114 is connected to each gas pipe 15. The second end of the outdoor heat exchanger 112 is connected to each liquid pipe 16.
[0041] In some embodiments, each of the multiple indoor units 13 includes an indoor heat exchanger 131, an indoor fan 132, an air outlet 133, and an air return outlet 134. A first end of the indoor heat exchanger 131 is connected to the corresponding air pipe 15, and a second end of the indoor heat exchanger 131 is connected to the corresponding liquid pipe 16. The air outlet 133 and the air return outlet 134 are connected to the indoor heat exchanger 131 via an air outlet duct and a return air duct, respectively.
[0042] It should be noted that the structure shown in FIG1 is an exemplary structure of a multi-split air conditioning system 10. The multi-split air conditioning system 10 may include more, fewer, or different components than those shown in FIG1 . For example, the outdoor unit 11 may further include an outdoor fan motor for driving the outdoor fan 115, and the indoor unit 13 may further include a display and an indoor fan motor for driving the indoor fan 132. When the indoor unit 13 includes a display, the display may be configured to display the temperature of the indoor space where the indoor unit 13 is located and / or the operating status of the multi-split air conditioning system 10.
[0043] In some embodiments, as shown in FIG2 , the controller 14 is coupled to the compressor 111, the four-way valve 114, and the outdoor fan 115 in the outdoor unit 11, to the plurality of electronic expansion valves 12, and to the indoor fans 132 in the plurality of indoor units 13. The controller 14 is configured to control the operating status of each component coupled to the controller 14.
[0044] The controller 14 is also coupled to multiple temperature sensors and multiple pressure sensors to obtain temperature values and pressure values of multiple components in the multi-split air conditioning system 10. The configuration of the multiple temperature sensors and the multiple pressure sensors will be described in subsequent embodiments.
[0045] In addition, the controller 14 is also coupled to the communicator 109 to establish a communication connection with other devices (e.g., terminal devices) through the communicator 109 to transmit and receive communication signals. For example, the communicator 109 may include a radio frequency (RF) device, a cellular device, a Wi-Fi device, a global positioning system (GPS) device, and the like.
[0046] In some examples, the controller 14 refers to a device that can generate an operation control signal based on an instruction opcode and a timing signal, thereby instructing the multi-split air conditioning system 10 to execute the control instruction. For example, the controller 14 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 14 can also be other devices with processing functions, such as circuits, devices, or software modules, which are not limited in the present embodiment.
[0047] For example, as shown in FIG3 , the controller 14 includes an outdoor controller 141 and an indoor controller 142. The outdoor controller 141 includes a first memory 1411 and is configured to control the outdoor unit 11 to perform related operations. The indoor controller 142 includes a second memory 1421 and is configured to control the multiple indoor units 13 and the multiple electronic expansion valves 12 to perform related operations. A wired or wireless communication connection is established between the indoor controller 142 and the outdoor controller 141. The outdoor controller 141 and the indoor controller 142 can be respectively attached to the outdoor unit 11 and the indoor unit 13, or they can be independently provided.
[0048] It should be noted that the structure of the controller 14 shown in FIG3 is an exemplary structure of the controller 14. For example, the outdoor controller 141 and the indoor controller 142 may be integrated into one controller. The first memory 1411 and the second memory 1421 may also be integrated into one memory.
[0049] In some embodiments, the first memory 1411 and the second memory 1421 are configured to store applications and data. The outdoor controller 141 and the indoor controller 142 respectively execute the applications and data stored in the first memory 1411 and the second memory 1421 to perform various functions and data processing of the multi-split air conditioning system 10. For example, the first memory 1411 and the second memory 1421 may include high-speed random access memory and non-volatile memory. For example, the first memory 1411 and the second memory 1421 may be disk storage devices, flash memory devices, etc.
[0050] In some embodiments, a user can control the operating state of the multi-split air conditioning system 10 through a device that has established a communication connection with the controller 14. For example, as shown in FIG4 , a communication connection exists between the controller 14 and the user's terminal device 300. This communication connection can be implemented using various wired or wireless communication technologies, such as Ethernet, Universal Serial Bus (USB), FireWire, any cellular network communication technology (such as 3G / 4G / 5G), Bluetooth, Wi-Fi, Near Field Communication (NFC), or any other suitable communication technology.
[0051] The terminal device 300 can be, for example, a remote control, a mobile phone, a tablet computer, a personal computer (PC), a personal digital assistant (PDA), a smart watch, a wearable electronic device, an augmented reality (AR) device, a virtual reality (VR) device, a robot, etc. Figure 4 and subsequent embodiments use the terminal device 300 as an example of a mobile phone. The present disclosure does not limit the specific form of the terminal device 300.
[0052] In this embodiment, as shown in FIG5 , a terminal device 300 displays a management interface 301 for the multi-split air conditioning system 10. This interface 301 includes a first button 302 for managing the operating status of the multi-split air conditioning system 10. In response to a user clicking on the first button 302 in the management interface 301, the terminal device 300 displays an operating status drop-down selection box 303 within the management interface 301. In response to the user selecting an operating status in the operating status drop-down selection box 303, the terminal device 300 sends an operating instruction corresponding to the selection to the multi-split air conditioning system 10, causing the multi-split air conditioning system 10 to operate according to the user's selected operating status.
[0053] In some embodiments, the multi-split air conditioning system 10 operates in a cooling mode to reduce the temperature of the indoor space. In the cooling mode, the controller 14 controls the compressor 111 to start operating and controls the four-way valve 114 so that the D end is connected to the C end and the S end is connected to the E end.
[0054] Thus, the compressor 111 begins to compress the refrigerant so that the refrigerant circulates in the multi-split air conditioning system 10. For example, the compressor 111 compresses the gaseous refrigerant into a high-temperature, high-pressure gaseous refrigerant, and drives the compressed refrigerant through the D and C ends of the four-way valve 114 to the first end of the outdoor heat exchanger 112, thereby entering the outdoor heat exchanger 112. After being liquefied into a low-temperature, low-pressure liquid refrigerant in the outdoor heat exchanger 112, the high-temperature, high-pressure gaseous refrigerant passes through the second end of the outdoor heat exchanger 112, the liquid pipe 16, and the multiple electronic expansion valves 12 to the second ends of the multiple indoor heat exchangers 131, thereby entering the multiple indoor heat exchangers 131. In any of the multiple indoor units 13, low-temperature, low-pressure liquid refrigerant is vaporized into gaseous refrigerant in the indoor heat exchanger 131 of that indoor unit. This vaporized gaseous refrigerant absorbs heat from the heat exchanger 131, lowering the temperature of the air inside the unit. The cooled gas is then transported to the outside of the unit through the air outlet 133 of the indoor unit, thereby lowering the temperature of the indoor space. The vaporized gaseous refrigerant then passes through the first end of the indoor heat exchanger 131 and the air pipe 15 to the four-way valve 114. It then passes through the E and S ends of the four-way valve 114 to the air intake of the gas-liquid separator 113. During the transfer from the indoor heat exchanger 131 to the gas-liquid separator 113, the gaseous refrigerant may condense to produce liquid. After the gas-liquid separator 113 separates the liquid, the gaseous refrigerant is fed into the compressor 111, achieving refrigerant recycling.
[0055] In other embodiments, the multi-split air conditioning system 10 operates in a heating mode to increase the temperature of the indoor space. Unlike the cooling mode, in the heating mode, the controller 14 controls the four-way valve 114 so that the D end is connected to the E end and the S end is connected to the C end.
[0056] In this way, the high-temperature, high-pressure gaseous refrigerant generated by the compression process in the compressor 111 passes through the D and E terminals of the four-way valve 114 and is then fed into the multiple indoor heat exchangers 131 via the gas pipe 15. Within each of the multiple indoor units 13, the high-temperature, high-pressure gaseous refrigerant is liquefied into a low-temperature, low-pressure liquid refrigerant in the indoor heat exchanger 131 of that unit. This releases heat to the surrounding area of the indoor heat exchanger 131, raising the temperature of the gas inside the unit. The heated gas is then transported to the outside of the unit through the air outlet 133 of the indoor unit, thereby raising the temperature of the indoor space. The low-temperature, low-pressure liquid refrigerant then flows out of the indoor heat exchanger 131 through the second terminal of the indoor heat exchanger 131, passes through the multiple electronic expansion valves 12 and the liquid pipe 16, and enters the outdoor heat exchanger 112. The low-temperature, low-pressure liquid refrigerant is vaporized into gaseous refrigerant in the outdoor heat exchanger 112 , and then transmitted to the gas-liquid separator 113 through the C end and the S end of the four-way valve 114 , and then returns to the compressor 111 .
[0057] In the above-mentioned cooling working state or heating working state, the outdoor fan 115 is configured to start working under the control of the controller 14 to discharge the heat generated by the liquefied refrigerant of the outdoor heat exchanger 112 or the cold generated by the vaporized refrigerant out of the outdoor unit 11; the indoor fan 132 is configured to start working under the control of the controller 14 to discharge the cold generated by the vaporized refrigerant of the indoor heat exchanger 131 or the heat generated by the liquefied refrigerant out of the indoor unit 13 to adjust the temperature of the indoor space where the indoor unit 13 is located.
[0058] In some embodiments, the return air vent 134 of the indoor unit 13 is configured to transport air from outside the indoor unit 13 to the interior of the indoor unit 13. This allows the air inside the indoor unit 13 to be cooled or heated by the indoor heat exchanger 131 before being transported to the exterior of the indoor unit 13 through the air outlet 133. In this way, the air in the indoor space where the indoor unit 13 is located can be circulated to cool or heat the air, thereby improving the cooling efficiency or heating efficiency of the multi-split air conditioning system 10.
[0059] It should be noted that when the multi-split air conditioning system 10 is operating in a cooling mode, the outdoor heat exchanger 112 may also be referred to as a condenser, and the indoor heat exchanger 131 may also be referred to as an evaporator. When the multi-split air conditioning system 10 is operating in a heating mode, the outdoor heat exchanger 112 may also be referred to as an evaporator, and the indoor heat exchanger 131 may also be referred to as a condenser. Furthermore, the pressure at the exhaust port of the compressor 111 may be referred to as an exhaust pressure, a condensing pressure, or a high-pressure pressure; and the pressure at the intake port of the compressor 111 may be referred to as an intake pressure, an evaporating pressure, or a low-pressure pressure.
[0060] In some embodiments, the electronic expansion valve 12 has the function of expanding and reducing the pressure of the refrigerant flowing through the electronic expansion valve 12, and can adjust the flow rate of the refrigerant in the pipeline. If the opening degree (the degree of opening) of the electronic expansion valve 12 decreases, the refrigerant flow rate through the electronic expansion valve 12 decreases. If the opening degree of the electronic expansion valve 12 increases, the refrigerant flow rate through the electronic expansion valve 12 increases. For example, if the multi-split air-conditioning system 10 is in a cooling working state, the multiple electronic expansion valves 12 are respectively located on the refrigerant input side of the corresponding indoor unit 13, so at this time, the multiple electronic expansion valves 12 respectively control the refrigerant input amount of the corresponding indoor unit 13; if the multi-split air-conditioning system 10 is in a heating working state, the multiple electronic expansion valves 12 are respectively located on the refrigerant output side of the corresponding indoor unit 13, so at this time, the multiple electronic expansion valves 12 respectively control the refrigerant output amount of the corresponding indoor unit 13.
[0061] As described in the background art, in the related art, if an electronic expansion valve malfunctions in a multi-split air conditioning system 10, professional maintenance personnel are required to observe and analyze the electronic expansion valves 12 in the entire multi-split air conditioning system 10, locate the faulty electronic expansion valve, and then repair or replace the faulty electronic expansion valve to eliminate the electronic expansion valve malfunction in the multi-split air conditioning system 10. Therefore, the fault location efficiency and automation level of the multi-split air conditioning system 10 in the related art are low.
[0062] In response to the above-mentioned technical problems existing in the related art, the inventors of the present disclosure have discovered through research that: when multiple electronic expansion valves 12 are operating normally, the temperature difference between the temperature value of the liquid pipe 16 connected to any indoor unit and the temperature value of the gas pipe 15 connected to the indoor unit is less than a certain threshold. When an electronic expansion valve failure occurs in the multi-split air conditioning system 10 (for example, the electronic expansion valve 12 cannot be opened normally), the temperature difference between the temperature value of the liquid pipe 16 connected to the indoor unit 13 corresponding to the failed electronic expansion valve and the temperature value of the gas pipe 15 connected to the indoor unit 13 is greater than the above-mentioned certain threshold. Therefore, by monitoring the characteristic data of multiple indoor units 13 during operation (for example, the temperature values of the liquid pipe 16 and the gas pipe 15 connected to the same indoor unit 13), it is possible to determine whether multiple electronic expansion valves 12 have failed.
[0063] Based on the above technical concept, some embodiments of the present disclosure provide a multi-split air conditioning system 10. In the event of an electronic expansion valve failure in the multi-split air conditioning system 10, the controller 14 in the multi-split air conditioning system 10 can obtain characteristic data of multiple indoor units 13 in the multi-split air conditioning system 10, and identify the abnormal indoor unit among the multiple indoor units 13 based on the characteristic data, and then determine the electronic expansion valve 12 corresponding to the abnormal indoor unit as a faulty electronic expansion valve to complete the positioning of the faulty electronic expansion valve. In this way, after an electronic expansion valve failure occurs in the multi-split air conditioning system 10, the faulty electronic expansion valve can be located without the need for professional maintenance personnel to manually identify the multiple electronic expansion valves 12 based on their personal experience, thereby improving the degree of automation of fault positioning in the multi-split air conditioning system 10 and improving the efficiency of positioning the faulty electronic expansion valve in the multi-split air conditioning system 10.
[0064] The following mainly describes the process of the controller 14 executing the fault location method with reference to the accompanying drawings.
[0065] In some embodiments, as shown in FIG. 6 , the controller 14 is configured to execute steps S101 to S103 described below.
[0066] S101 : When an electronic expansion valve failure occurs in a multi-split air conditioning system 10 , characteristic data of each of the plurality of indoor units 13 is obtained.
[0067] The characteristic data of each indoor unit includes a temperature difference between a temperature value of a liquid pipe 16 connected to the indoor unit and a temperature value of a gas pipe 15 connected to the indoor unit.
[0068] For example, the temperature difference may be a value obtained by subtracting the temperature of the gas pipe 15 connected to the indoor unit from the temperature of the liquid pipe 16 connected to the indoor unit. Alternatively, the temperature difference may be a value obtained by subtracting the temperature of the liquid pipe 16 connected to the indoor unit from the temperature of the gas pipe 15 connected to the indoor unit.
[0069] In some embodiments, a user can instruct the multi-split air conditioning system 10 to activate the fault detection function and the fault location function through a device that has established a communication connection with the controller 14, so that the controller 14 can begin detecting whether an electronic expansion valve fault has occurred and locating the faulty electronic expansion valve. It should be noted that the method by which the controller 14 detects whether an electronic expansion valve fault has occurred will be described in subsequent embodiments.
[0070] Taking the terminal device 300 as an example, as shown in Figures 7 and 8 , the management interface 301 of the terminal device 300 includes a second button 304 for turning on or off the fault detection function of the multi-split air conditioning system 10, and a third button 306 for turning on or off the fault location function of the multi-split air conditioning system 10. When the fault detection function (or fault location function) of the multi-split air conditioning system 10 is off, the second button 304 (or third button 306) is in a first display state 3041 (or third display state 3061); when the fault detection function (or fault location function) of the multi-split air conditioning system 10 is on, the second button 304 (or third button 306) is in a second display state 3042 (or fourth display state 3062). For example, when the second button 304 is in the first display state 3041, in response to a user clicking the second button 304, the terminal device 300 can send a command to the controller 14, instructing the controller 14 to turn on the fault detection function. After the fault detection function is turned on, the controller 14 may send an instruction to the terminal device 300 to instruct the terminal device 300 to switch the display state of the second button 304 to the second display state 3042 .
[0071] In some examples, if an electronic expansion valve fault is detected in the multi-split air conditioning system 10, the controller 14 may send an instruction to the terminal device 300 via the communicator 109, causing the terminal device 300 to display a prompt box 305 as shown in FIG8 . For example, the prompt message in prompt box 305 reads, "An electronic expansion valve fault has been detected in the multi-split air conditioning system. Do you want to perform fault location immediately?" to prompt the user to select whether to enable the fault location function. For example, in response to a user clicking the "OK" button in prompt box 305, the terminal device 300 may send an instruction to the controller 14, instructing the controller 14 to enable the fault location function. After the fault location function is enabled, the controller 14 may send an instruction to the terminal device 300, instructing the terminal device 300 to switch the display state of the third button 306 from the third display state 3061 to the fourth display state 3062.
[0072] In other embodiments, the multi-split air conditioning system 10 can automatically activate the fault detection function and the fault location function after operating for a preset time period. For example, the preset time period can be pre-set by the user or the manufacturer.
[0073] After the fault location function is turned on, the controller 14 starts to obtain characteristic data of a plurality of indoor units.
[0074] In some examples, as shown in FIG2 , the multi-split air conditioning system 10 includes multiple first temperature sensors 101 and multiple second temperature sensors 102 to respectively detect the temperature of the liquid pipe 16 and the temperature of the gas pipe 15 connected to the multiple indoor units 13. As shown in FIG9 , the multiple first temperature sensors 101 correspond to the indoor heat exchangers 131 of the multiple indoor units 13 and are disposed in the gas pipe 15 connected to the corresponding indoor heat exchangers 131 to detect the temperature of the corresponding gas pipe 15. The multiple second temperature sensors 102 correspond to the indoor heat exchangers 131 of the multiple indoor units 13 and are disposed in the liquid pipe 16 connected to the corresponding indoor heat exchangers 131 to detect the temperature of the corresponding liquid pipe 16. The controller 14 is coupled to the multiple first temperature sensors 101 and the multiple second temperature sensors 102 to obtain the temperature difference between the temperature of the liquid pipe 16 connected to each of the multiple indoor units 13 and the temperature of the gas pipe 15 connected to the corresponding indoor unit.
[0075] In some embodiments, the characteristic data of each indoor unit may further include at least one of the following: the exhaust pressure value of the compressor 111, the suction pressure value of the compressor 111, the exhaust temperature value of the compressor 111, the suction temperature value of the compressor 111, the outlet temperature value of the indoor unit's air outlet 133, or the return air temperature value of the indoor unit's return air outlet 134. In this embodiment, in addition to the plurality of first temperature sensors 101 and the plurality of second temperature sensors 102, the multi-split air conditioning system 10 includes a plurality of other temperature sensors and a plurality of pressure sensors to detect each of the aforementioned characteristic data.
[0076] In some examples, as shown in FIG2 , the multi-split air conditioning system 10 includes a first pressure sensor 107 and a second pressure sensor 108. The first pressure sensor 107 is disposed at the exhaust port of the compressor 111 and is configured to detect the exhaust pressure value of the compressor 111. The second pressure sensor 108 is disposed at the intake port of the compressor 111 and is configured to detect the intake pressure value of the compressor 111. The controller 14 is coupled to the first pressure sensor 107 and the second pressure sensor 108 to obtain the exhaust pressure value and the intake pressure value.
[0077] In some examples, the multi-split air conditioning system 10 includes a third temperature sensor 103, a fourth temperature sensor 104, multiple fifth temperature sensors 105, and multiple sixth temperature sensors 106. The third temperature sensor 103 is disposed at the exhaust port of the compressor 111 and is configured to detect the exhaust temperature of the compressor 111. The fourth temperature sensor 104 is disposed at the intake port of the compressor 111 and is configured to detect the intake temperature of the compressor 111. The multiple fifth temperature sensors 105 correspond to the multiple indoor units 13 and are disposed at the air outlets 133 of the corresponding indoor units 13 to detect the outlet air temperature of the air outlets 133. The multiple sixth temperature sensors 106 correspond to the multiple indoor units 13 and are disposed at the return air outlets 134 of the corresponding indoor units 13 to detect the return air temperature of the return air outlets 134. The controller 14 is coupled to the third temperature sensor 103 , the fourth temperature sensor 104 , the plurality of fifth temperature sensors 105 and the plurality of sixth temperature sensors 106 to obtain exhaust temperature values, intake temperature values, outlet temperature values and return air temperature values.
[0078] S102: Identify an abnormal indoor unit among the plurality of indoor units 13 based on the characteristic data of each indoor unit among the plurality of indoor units 13.
[0079] The abnormal indoor unit refers to an indoor unit 13 in an abnormal operating state, that is, an indoor unit 13 whose corresponding electronic expansion valve 12 is a faulty electronic expansion valve. Similarly, when the electronic expansion valve 12 corresponding to the indoor unit 13 is not a faulty electronic expansion valve, it can be understood that the indoor unit 13 is in a normal operating state.
[0080] It should be noted that a failure of an electronic expansion valve in the multi-split air conditioning system 10 can cause changes in the characteristic data of multiple indoor units 13. For example, when the multi-split air conditioning system 10 is operating normally, the temperature difference between the temperature value of the liquid pipe 16 connected to any indoor unit and the temperature value of the gas pipe 15 connected to the indoor unit is less than a certain threshold. If a failure of an electronic expansion valve 12 causes the electronic expansion valve 12 to fail to open normally, the amount of refrigerant in the indoor unit 13 corresponding to the electronic expansion valve 12 will be insufficient, resulting in the indoor unit 13 being unable to evaporate or condense the refrigerant normally. This is reflected in the temperature difference between the temperature value of the liquid pipe 16 connected to the indoor unit 13 and the temperature value of the gas pipe 15 connected to the indoor unit 13 increasing to exceed the certain threshold. Therefore, the controller 14 can determine whether any indoor unit is an abnormal indoor unit based on the characteristic data of the indoor unit.
[0081] Furthermore, since the various components of the multi-split air conditioning system 10 cooperate with each other to form an integrated whole, if an abnormal indoor unit exists in the multi-split air conditioning system 10, the operating data of other components in the entire multi-split air conditioning system 10 will also change. Therefore, any of the aforementioned operating data, such as the exhaust pressure value, intake pressure value, exhaust temperature value, intake temperature value, indoor unit outlet air temperature value, and indoor unit return air temperature value, can also be used as characteristic data of the indoor unit.
[0082] In some embodiments, the controller 14 may input characteristic data of each of the multiple indoor units into a fault identification model based on a deep neural network (DNN) to obtain a fault identification result. Deep neural network (DNN) is a technology in the field of machine learning (ML). Since the fault identification model based on the deep neural network (DNN) can output multiple labels, the fault identification result can indicate whether each of the multiple indoor units 13 is an abnormal indoor unit. In this way, the controller 14 can determine the abnormal indoor unit among the multiple indoor units 13 based on the fault identification result.
[0083] In some embodiments, a fault recognition model based on a deep neural network (DNN) is pre-stored in the memory of the controller 14. In some examples, the controller 14 or other processing-capable device can train the fault recognition model based on the deep neural network based on historical feature datasets of multiple indoor units 13 in normal operating states and historical feature datasets of multiple indoor units 13 in abnormal operating states to obtain a trained fault recognition model based on the deep neural network (DNN), and store the trained fault recognition model in the memory of the controller 14.
[0084] S103: Determine the electronic expansion valve corresponding to the abnormal indoor unit as a faulty electronic expansion valve.
[0085] In some embodiments, the controller 14 may determine one or more abnormal indoor units from the plurality of indoor units 13 , and further determine one or more electronic expansion valves 12 corresponding to the one or more abnormal indoor units as faulty electronic expansion valves.
[0086] The multi-split air conditioning system 10 provided by the embodiment of the present disclosure can obtain characteristic data of multiple indoor units 13 in the multi-split air conditioning system 10 when an electronic expansion valve failure occurs in the multi-split air conditioning system 10, and then determine the abnormal indoor unit among the multiple indoor units 13 based on the characteristic data of the multiple indoor units 13, and determine the electronic expansion valve 12 corresponding to the abnormal indoor unit as the faulty electronic expansion valve. In this way, the faulty electronic expansion valve in the multi-split air conditioning system 10 can be accurately located, eliminating the need for professional maintenance personnel to observe and analyze the faulty electronic expansion valve based on personal experience and consuming a large amount of time. This can improve the accuracy and automation of locating the faulty electronic expansion valve in the multi-split air conditioning system 10, and thus improve the fault locating efficiency of the multi-split air conditioning system 10.
[0087] The following is an exemplary description of a method for the controller 14 to detect whether an electronic expansion valve failure has occurred in the multi-split air conditioning system 10, mainly with reference to the accompanying drawings. As shown in FIG10 , the controller 14 can determine whether an electronic expansion valve failure has occurred in the multi-split air conditioning system 10 by executing steps S201 to S203.
[0088] S201 : Acquire the operating data of the multi-split air conditioning system 10 .
[0089] The aforementioned operating data is parameter information generated during the operation of the multi-split air conditioning system 10. The operating data may include, for example, at least one of the following: the operating current value of the compressor 111, the exhaust pressure value of the compressor 111, the exhaust temperature value of the compressor 111, the outlet air temperature values of the air outlets 133 of the multiple indoor units 13, and the return air temperature values of the return air outlets 134 of the multiple indoor units 13.
[0090] In some embodiments, as described above, after the controller 14 turns on the fault detection function, the controller 14 can obtain the operating data of the multi-split air conditioning system 10.
[0091] It should be noted that the examples of the above-mentioned characteristic data and the above-mentioned operating data in the embodiment of the present disclosure are exemplary. The characteristic data and the operating data may also include other data related to the operation of the multi-split air-conditioning system 10, which are not detailed here.
[0092] S202: Determine whether an electronic expansion valve failure occurs in the multi-split air conditioning system 10 based on the operating data.
[0093] In some embodiments, the controller 14 may input the operating data of the multi-split air conditioning system 10 into a fault diagnosis model based on a support vector machine (SVM) to obtain a fault diagnosis result indicating whether an electronic expansion valve fault has occurred in the multi-split air conditioning system 10. In this way, the controller 14 may determine whether an electronic expansion valve fault has occurred in the multi-split air conditioning system 10 based on the fault diagnosis result. For example, a support vector machine (SVM) is a generalized linear classifier that performs binary classification on data using a supervised learning approach. The support vector machine (SVM) has good binary classification performance. Therefore, a fault diagnosis model based on the support vector machine (SVM) may improve the accuracy of determining whether an electronic expansion valve fault has occurred in the multi-split air conditioning system 10.
[0094] In some embodiments, a fault diagnosis model based on a support vector machine (SVM) is pre-stored in the memory of the controller 14. In some examples, the controller 14 or another device with processing capabilities can train the fault diagnosis model based on a historical operating data set of the multi-split air conditioning system 10 without an electronic expansion valve fault and a historical operating data set of the multi-split air conditioning system 10 with an electronic expansion valve fault, thereby obtaining a trained fault diagnosis model based on the support vector machine (SVM), and store the trained fault diagnosis model in the memory of the controller 14.
[0095] In some embodiments, if the controller 14 determines that an electronic expansion valve failure occurs in the multi-split air conditioning system 10 , the controller 14 executes the above steps S101 to S103 .
[0096] In some embodiments, if the controller 14 determines that no electronic expansion valve failure occurs in the multi-split air conditioning system 10 , the controller 14 executes the following step S203 .
[0097] S203: Send a first prompt message.
[0098] The first prompt information is used to prompt that no electronic expansion valve failure occurs in the multi-split air-conditioning system 10 .
[0099] For example, when multiple indoor units 13 include displays, the first prompt information may be information such as text or images displayed on the display. Alternatively, when the indoor unit 13 includes a speaker, the first prompt information may be sound information emitted by the speaker. Alternatively, when the multi-split air conditioning system 10 is communicatively connected to a user's terminal device, the first prompt information may be information such as text, images, sounds, vibrations, etc. emitted by the terminal device. It should be noted that the first prompt information may also be other forms of information, such as lighting information, etc., which is not limited in this disclosure.
[0100] In some embodiments, after the controller 14 determines that a faulty electronic expansion valve has been detected, that is, after step S103, the controller 14 may issue a second prompt message to indicate that the faulty electronic expansion valve has failed. In some examples, as shown in FIG11 , the controller 14 issues the second prompt message by executing steps S301 to S303.
[0101] S301: After determining that a faulty electronic expansion valve is present, obtain the opening degree of the faulty electronic expansion valve.
[0102] As previously described, the controller 14 adjusts the refrigerant input or output to the indoor unit 13 corresponding to the electronic expansion valve 12 by adjusting the opening of the electronic expansion valve 12. Therefore, an abnormal operating state of an abnormal indoor unit is typically caused by an abnormal opening of the corresponding electronic expansion valve. Furthermore, the opening of the electronic expansion valve can indicate the severity of the fault.
[0103] S302: Determine the fault level of the faulty electronic expansion valve according to the opening degree of the faulty electronic expansion valve.
[0104] In some embodiments, the controller 14 may determine the fault level of the faulty electronic expansion valve based on a first correspondence between the opening degree of the faulty electronic expansion valve and the fault level. The first correspondence may be pre-stored in a memory of the controller 14 .
[0105] It should be noted that the normal opening degrees of the multiple electronic expansion valves 12 are related to the working state of the multi-split air-conditioning system 10 .
[0106] In some examples, if the multi-split air conditioning system 10 is in the heating operating mode, the normal opening degree of the multiple electronic expansion valves 12 should be 100% (i.e., the multiple electronic expansion valves 12 should be fully open). In this example, the smaller the opening degree of a faulty electronic expansion valve, the higher the fault level of the faulty electronic expansion valve. For example, when the multi-split air conditioning system 10 is in the heating operating mode, a first correspondence between the opening degree of a faulty electronic expansion valve and the fault level is shown in Table 1 below.
[0107] Table 1
[0108] Faulty electronic expansion valve opening fault level 75% to 99% Level 1 50% to 74% Level 2 24% to 49% Level 3 23% or less Level 4
[0109] Referring to Table 1, if the opening of the faulty electronic expansion valve is within the range of 75% to 99%, the controller 14 can determine that the fault level of the faulty electronic expansion valve is level one. If the opening of the faulty electronic expansion valve is within the range of 50% to 74%, the controller 14 can determine that the fault level of the faulty electronic expansion valve is level two. If the opening of the faulty electronic expansion valve is within the range of 24% to 49%, the controller 14 can determine that the fault level of the faulty electronic expansion valve is level three. If the opening of the faulty electronic expansion valve is below 23%, the controller 14 can determine that the fault level of the faulty electronic expansion valve is level four. It should be noted that, from "level one", "level two", "level three" to "level four", the fault level of the faulty electronic expansion valve increases successively, and the urgency of repairing the faulty electronic expansion valve also increases successively.
[0110] In other examples, if the multi-split air conditioning system 10 is in cooling mode, the normal opening of the multiple electronic expansion valves 12 should be between 12% and 13%. In this example, if the opening of a faulty electronic expansion valve is greater than 13%, the greater the opening of the faulty electronic expansion valve, the higher the fault level of the faulty electronic expansion valve, and thus the more urgent the need for repair of the faulty electronic expansion valve. For example, a first correspondence between the opening of a faulty electronic expansion valve and the fault level when the multi-split air conditioning system 10 is in cooling mode is shown in Table 2 below.
[0111] Table 2
[0112] Faulty electronic expansion valve opening fault level 75%-100% fourth level 50%-74% third level
[0113] 25%-49% Second level 14%-24% First level 12% or below
[0114] Referring to Table 2, if the opening of the faulty electronic expansion valve is within the range of 75% to 100%, the controller 14 may determine that the fault level of the faulty electronic expansion valve is level 4. If the opening of the faulty electronic expansion valve is within the range of 50% to 74%, the controller 14 may determine that the fault level of the faulty electronic expansion valve is level 3. If the opening of the faulty electronic expansion valve is within the range of 25% to 49%, the controller 14 may determine that the fault level of the faulty electronic expansion valve is level 2. If the opening of the faulty electronic expansion valve is within the range of 14% to 24%, or is below 12%, the controller 14 may determine that the fault level of the faulty electronic expansion valve is level 1.
[0115] S303: Send a second prompt message.
[0116] The second prompt information is used to prompt the fault level of the faulty electronic expansion valve.
[0117] In some embodiments, similar to the manner of issuing the first prompt message in step S203 above, the controller 14 may issue a second prompt message via a display, speaker, terminal device, etc. of the indoor unit 13 (e.g., an abnormal indoor unit). For example, taking the fault level of the faulty electronic expansion valve as level 4 and the faulty electronic expansion valve numbered 001 as an example, as shown in FIG12 , the second prompt message may be, for example, a text message such as "The electronic expansion valve numbered 001 has a serious fault. Immediate repair recommended!"
[0118] The multi-split air-conditioning system 10 in the above embodiment, after locating the faulty electronic expansion valve in the multi-split air-conditioning system 10, can determine the fault level of the faulty electronic expansion valve based on the first correspondence between the opening degree of the faulty electronic expansion valve and the fault level, and issue a second prompt information corresponding to the fault level of the faulty electronic expansion valve, so as to prompt the user to reasonably arrange the maintenance work of the faulty electronic expansion valve according to the fault level of the faulty electronic expansion valve, so as to avoid the electronic expansion valve failure causing a decrease in the cooling or heating effect of the multi-split air-conditioning system 10.
[0119] Some embodiments of the present disclosure also provide a fault locating method for a multi-split air conditioning system. The multi-split air conditioning system may be, for example, the multi-split air conditioning system 10 described above. The method may include, for example, the steps performed by the controller 14 described above. The beneficial effects of this method include at least those of the multi-split air conditioning system 10 described above, and are not further detailed here.
[0120] Since the operating data of different types of multi-split air-conditioning systems are different, in the related art, it is necessary to collect the operating data of various types of multi-split air-conditioning systems in normal operating conditions and the operating data in abnormal operating conditions (operating conditions where a fault occurs) as sample sets to establish a fault diagnosis model. In this way, the fault diagnosis model can be applied to various types of multi-split air-conditioning systems, thereby improving the adaptability and accuracy of fault diagnosis of multi-split air-conditioning systems using the fault diagnosis model. However, since it is necessary to collect operating data of multiple types of multi-split air-conditioning systems, the training method of this fault diagnosis model will result in an excessively long establishment cycle of the fault diagnosis model, which in turn leads to a low efficiency in fault diagnosis of multi-split air-conditioning systems using the fault diagnosis model.
[0121] Based on the above technical problems, some embodiments of the present disclosure also provide a method for training a fault diagnosis model for a multi-split air conditioning system. The method can construct a first feature offset space for the first multi-split air conditioning system based on the values of the operating data of the first multi-split air conditioning system in a normal operating state (i.e., the normal values of the operating data) and the values of the operating data when various faults occur (i.e., the fault values of the operating data), and based on the difference between the normal values of the operating data of the first multi-split air conditioning system and the normal values of the operating data of the second multi-split air conditioning system, correct the first feature offset space to obtain a second feature offset space for the second multi-split air conditioning system. Thus, based on the second feature offset space and the normal values of the operating data of the second multi-split air conditioning system, the fault values of the operating data of the second multi-split air conditioning system can be obtained to establish a fault diagnosis model for the second multi-split air conditioning system. In this way, in the process of establishing the fault diagnosis model for the second multi-split air conditioning system, it is not necessary to collect the operating data of the second multi-split air conditioning system under various fault operating states under actual working conditions, thereby shortening the establishment cycle of the fault diagnosis model and thereby improving the efficiency of fault diagnosis of the multi-split air conditioning system using the fault diagnosis model.
[0122] As shown in Figure 13, the fault diagnosis model training method for the multi-split air conditioning system includes steps S401 to S405. It should be noted that the method can be executed by the controller 14 described above, or by another device with processing capabilities, such as a server. The subsequent embodiments will use the controller 14 as an example to illustrate the method. The structure and operating principle of the multi-split air conditioning system can refer to the multi-split air conditioning system 10 described in the previous embodiment and will not be further described here.
[0123] S401: Acquire normal values and fault values of multiple operating data of a first multi-split air-conditioning system.
[0124] In some embodiments, the first multi-split air conditioning system may be referred to as an experimental multi-split air conditioning system. The memory of the first multi-split air conditioning system stores historical operating data generated when the first multi-split air conditioning system operates under various operating conditions.
[0125] The above-mentioned operating data includes, for example, the suction temperature value of the compressor 111, the exhaust temperature value of the compressor 111, the suction pressure value of the compressor 111, the exhaust pressure value of the compressor 111, the opening degree of multiple electronic expansion valves 12, the operating current value of the compressor 111, the ambient temperature value of the outdoor unit 11, the average value of the ambient temperature values of multiple indoor units, the suction superheat of the compressor 111, the exhaust superheat of the compressor 111, the operating temperature value of the outdoor unit 11 and the operating temperature values of multiple indoor units 13.
[0126] For example, suction superheat refers to the difference between the suction temperature of compressor 111 and the saturation temperature of the refrigerant at the suction pressure. Discharge superheat refers to the difference between the discharge temperature of compressor 111 and the saturation temperature of the refrigerant at the discharge pressure. The operating temperature of the outdoor unit 11 may be, for example, the operating temperature of the outdoor heat exchanger 112. The operating temperature of a particular indoor unit 13 may be, for example, the operating temperature of the indoor heat exchanger 131 of that particular indoor unit 13.
[0127] In some examples, as shown in FIG14 , the above step S101 may be specifically implemented as the following steps S4011 and S4012 .
[0128] S4011. Obtain historical operating data of the first multi-split air conditioning system.
[0129] The historical operation data includes normal operation data generated by the first multi-split air conditioning system in a normal operation state during the historical period and various fault operation data generated by the first multi-split air conditioning system when various faults occur. It should be noted that the length of the historical period is not limited in this disclosure.
[0130] S4012: Analyze the historical operation data to determine normal values of multiple operation data of the first multi-split air-conditioning system when the first multi-split air-conditioning system is in normal operation and fault values when a fault occurs in the first multi-split air-conditioning system.
[0131] It should be noted that the normal value of any one of the multiple operating data is different from the fault value of the same operating data. Therefore, the controller 14 can parse the historical operating data of the multi-split air conditioning system to distinguish between the multiple operating data generated by the multi-split air conditioning system during normal operation and the multiple operating data generated by the multi-split air conditioning system during a fault, thereby obtaining the normal value and the fault value of each of the multiple operating data. For example, possible faults of the multi-split air conditioning system include refrigerant leakage, excessive circulating refrigerant, or abnormal opening of the electronic expansion valve 12.
[0132] For example, the controller 14 may analyze historical operating data based on preset value ranges corresponding to the plurality of operating data. If the value of any of the plurality of operating data falls within the corresponding preset value range, the value of the operating data is determined to be a normal value. If the value of any of the operating data falls outside the corresponding preset value range, the value of the operating data is determined to be a fault value. The preset value range may be pre-set by a staff member or the manufacturer, for example.
[0133] It should be noted that when different faults occur in a multi-split air conditioning system, the fault values of the multiple operating data are generally different. When the controller 14 analyzes the historical operating data of the first multi-split air conditioning system, it can match the fault value of the first multi-split air conditioning system with the fault that occurred in the first multi-split air conditioning system.
[0134] S402: Determine a first characteristic offset space of the first multi-split air-conditioning system based on normal values and fault values of a plurality of operating data of the first multi-split air-conditioning system.
[0135] It should be noted that, because there are deviations between normal values and fault values of the plurality of operating data, a first characteristic offset space for the first multi-split air conditioning system can be constructed based on the offsets between the fault values and normal values of the plurality of operating data. The first characteristic offset space includes the offsets of each of the plurality of operating data of the first multi-split air conditioning system.
[0136] In some embodiments, the space composed of normal values of multiple operating data of the multi-split air-conditioning system can be called the feature space of the multi-split air-conditioning system.
[0137] For example, the first feature space of the first multi-split air-conditioning system can be represented as the following matrix:
[0138] X=[x1 x2…x n-1 x n ]
[0139] Among them, x j is a column vector of a certain running data, j ranges from 1 to n; n is an integer greater than 1.
[0140] For example, under a certain fault, the first characteristic offset space of the first multi-split air conditioning system can be expressed as:
[0141] Δxi=[Δxi1 Δxi2…Δxi n-1 Δxi n ]
[0142] Where Δxi j is the offset of a certain running data, the value of j ranges from 1 to n; n is an integer greater than 1.
[0143] S403: Based on the difference between the normal values of the plurality of operating data of the first multi-split air-conditioning system and the normal values of the plurality of operating data of the second multi-split air-conditioning system, correct the first characteristic offset space to obtain a second characteristic offset space of the second multi-split air-conditioning system.
[0144] It should be noted that if there is a difference between the normal values of multiple operating data of the first multi-split air conditioning system and the normal values of multiple operating data of the second multi-split air conditioning system, it can be considered that there is a difference between the models of the first multi-split air conditioning system and the second multi-split air conditioning system. For example, the model of the first multi-split air conditioning system is a cooling-only type, and the model of the second multi-split air conditioning system is a heat pump type.
[0145] Furthermore, because the thermophysical mechanisms that cause failures in various multi-split air conditioning systems are similar, different models of multi-split air conditioning systems typically have similar characteristic offset spaces. Therefore, based on the differences between the models of the first and second multi-split air conditioning systems, the first characteristic offset space can be modified to obtain a second characteristic offset space for the second multi-split air conditioning system.
[0146] Exemplarily, a second correspondence between the models of the multi-split air conditioning systems and the correction coefficients is pre-stored in the memory of the controller 14. The correction coefficient is, for example, the ratio between the characteristic offset space of the other multi-split air conditioning systems and the first characteristic offset space of the first multi-split air conditioning system. After determining the difference between the normal values of the multiple operating data of the first multi-split air conditioning system and the normal values of the multiple operating data of the second multi-split air conditioning system, the controller 14 can determine the model of the second multi-split air conditioning system based on the model of the first multi-split air conditioning system, and thus can determine the correction coefficient of the second characteristic offset space relative to the first characteristic offset space based on the model of the second multi-split air conditioning system and the second correspondence, and then correct the first characteristic offset space according to the correction coefficient to obtain the second characteristic offset space. For example, the controller 14 multiplies the correction coefficient by the first characteristic offset space to obtain the second characteristic offset space.
[0147] Exemplarily, the above second corresponding relationship can be obtained through experiments or simulations.
[0148] S404: Determine fault values of the plurality of operating data of the second multi-split air-conditioning system based on normal values of the plurality of operating data of the second multi-split air-conditioning system and the second characteristic offset space.
[0149] In some embodiments, the controller 14 can analyze the historical operating data of the second multi-split air-conditioning system in normal operating state during the historical period, determine the normal values of multiple operating data of the second multi-split air-conditioning system, and thus obtain the fault values of multiple operating data of the second multi-split air-conditioning system based on the normal values of the multiple operating data of the second multi-split air-conditioning system and the second characteristic offset space.
[0150] For example, the controller 14 may add normal values of the plurality of operating data of the second multi-split air-conditioning system to the second characteristic offset space to obtain fault values of the plurality of operating data of the second multi-split air-conditioning system.
[0151] S405: Training a fault diagnosis model based on normal values and fault values of a plurality of operation data of the second multi-split air-conditioning system.
[0152] In some embodiments, the controller 14 may use normal values and fault values of a plurality of operating data of the second multi-split air-conditioning system as a sample set to train the fault diagnosis model.
[0153] In some examples, the above-mentioned fault diagnosis model is a fault diagnosis model based on a one-dimensional convolutional neural network (CNN). Among them, the convolutional neural network is a multi-layer perceptron. The working principle of the convolutional neural network mainly involves three basic concepts: local receptive field, pooling, and shared weights. A one-dimensional convolutional neural network is a convolutional neural network with one-dimensional input data and has classification capabilities. Therefore, a one-dimensional convolutional neural network can be used for fault detection and diagnosis of multi-split air conditioning systems. Fault detection is the process of detecting whether a system has a fault, and fault diagnosis is to obtain detailed information such as the fault type and fault severity of the fault.
[0154] In some embodiments, as shown in FIG15 , the fault diagnosis model may include: an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer.
[0155] The convolutional layer is a key structure in convolutional neural networks. Its function is to convolve the input data from the input layer using a series of filters to form a series of feature maps of the original input data and output them. When a filter convolves the input data of the convolutional layer, a local receptive field repeatedly moves over the input data. The data in the local receptive field is dot-producted with the weight matrix in the filter, and a fixed bias value is added to form an output matrix, which is a feature map of the original input data. The convolution operation of the filter on the input has the characteristic of shared weights, that is, the weight matrix and bias used by a filter for the data in the local receptive field are the same.
[0156] For example, an activation function (e.g., a ReLU activation function) can be applied to the output of the convolutional layer to perform delinearization, thereby improving the nonlinear fitting ability of the convolutional layer. For example, the ReLU activation function can be expressed as the following formula (1):
[0157]
[0158] The output a of the i-th convolutional layeri It can be expressed as the following formula (2):
[0159]
[0160] in, is the convolution operation, σ is the ReLU activation function, b i is the bias value, w i is the weight matrix.
[0161] The output of the convolutional layer can be fed into a pooling layer. The pooling layer divides the input data into multiple pooled regions and summarizes each pooled region to form an output. The output of the pooling layer is a generalized and simplified feature map of its input.
[0162] The fully connected layer acts as a "classifier" within the convolutional neural network, mapping distributed features to the sample label space. In other words, the fully connected layer is used to integrate the output of the pooling layer to obtain the output of the fully connected layer, and then input the output of the fully connected layer into the output layer.
[0163] For example, a classification network (softmax) layer can be used as the output layer of the convolutional neural network. In this way, the classification result of the input data of the softmax layer can be determined based on the dimension where the maximum probability vector in the output of the softmax layer is located.
[0164] The presently disclosed embodiments provide a method for training a fault diagnosis model for a multi-split air conditioning system. Based on the normal values and fault values of multiple operating data from a first multi-split air conditioning system, a first feature offset space is obtained to characterize the offsets between the normal values and various fault values of the multiple operating data from the first multi-split air conditioning system. The first feature offset space is then modified based on the differences between the models of the second multi-split air conditioning system and the first multi-split air conditioning system to obtain a second feature offset space for the second multi-split air conditioning system. Consequently, based on the second feature offset space and the normal values of the multiple operating data from the second multi-split air conditioning system, the fault values of the multiple operating data from the second multi-split air conditioning system can be obtained. Furthermore, the fault diagnosis model can be trained using the fault and normal values of the multiple operating data from the second multi-split air conditioning system to obtain a trained fault diagnosis model suitable for the second multi-split air conditioning system. This allows the fault diagnosis model for the second multi-split air conditioning system to be established without collecting the fault values of the multiple operating data from the second multi-split air conditioning system under actual operating conditions when various faults occur. This shortens the model establishment cycle and improves the efficiency of fault diagnosis for the second multi-split air conditioning system using the fault diagnosis model.
[0165] In some embodiments, as shown in FIG16 , the above step S405 can be specifically implemented as the following steps S501 and S502 .
[0166] S501. Determine at least one first operating data from the plurality of operating data of the second multi-split air-conditioning system based on normal values and fault values of the plurality of operating data of the second multi-split air-conditioning system.
[0167] It should be noted that the multiple operating data of the second multi-split air conditioning system may contain some redundant operating data irrelevant to training the fault diagnosis model. To improve the training speed of the fault diagnosis model, data screening of the multiple operating data is required. For example, the at least one first operating data is common to multi-split air conditioning systems of different models and is highly relevant to training the fault diagnosis model.
[0168] For example, the controller 14 may calculate the difference between a normal value and a fault value for each of the plurality of operating data of the second multi-split air conditioning system, select the operating data corresponding to the difference greater than a preset threshold value among the plurality of operating data as candidate operating data, and then extract at least one operating data common to the various models of multi-split air conditioning systems from the candidate operating data as the first operating data. The preset threshold value may be pre-set by a staff member or the manufacturer.
[0169] It should be noted that the greater the difference between the normal value and the fault value of a certain operating data, the more the operating data can reflect that a fault has occurred in the second multi-split air-conditioning system. Therefore, using the operating data as the first operating data for training the fault detection model is conducive to improving the accuracy of the trained fault detection model in detecting faults.
[0170] S502: Train a fault diagnosis model based on a normal value and a fault value of at least one first operating data of the second multi-split air-conditioning system.
[0171] The fault diagnosis model training method for a multi-split air conditioning system in the above-described embodiment filters the plurality of operating data of the second multi-split air conditioning system to extract at least one first operating data from the plurality of operating data of the second multi-split air conditioning system. The normal values and fault values of the at least one first operating data are then used to train a fault diagnosis model. This streamlined training sample set for the fault diagnosis model improves the training speed of the fault diagnosis model, thereby improving the efficiency of fault diagnosis for the multi-split air conditioning system using the fault diagnosis model.
[0172] In some embodiments, as shown in FIG. 17 , step S502 may be specifically implemented as the following steps S601 to S603 .
[0173] S601: Train an autoencoder model based on a normal value and a fault value of at least one first operating data of a second multi-split air-conditioning system.
[0174] The autoencoder model is an unsupervised learning model. Exemplarily, as shown in Figure 18, the autoencoder includes an encoder and a decoder. The encoder is configured to encode high-dimensional input data (i.e., original feature data) into low-dimensional latent variables (i.e., compressed data), so that the autoencoder model learns data with a higher amount of information. The decoder is configured to restore the low-dimensional latent variables to the original dimensions of the high-dimensional input data. The process of encoding the original feature data by the encoder and then decoding it by the decoder is the data reconstruction process of the original feature data. Exemplarily, the process of training the autoencoder model may include the following steps 1 to 4.
[0175] Step 1: The controller 14 converts the normal value and the fault value of at least one first operation data into a feature vector matrix X = {X1, X2, ..., X n}, and input the feature vector matrix into the autoencoder model.
[0176] Among them, X1, X2, …, X n Indicates a normal value and a fault value of the at least one first operating data.
[0177] Step 2: The encoder compresses the original feature data to the hidden layer as shown in the following formula (3).
[0178] H=σ e (W e X+B e ) Formula (3)
[0179] Among them, σ e is the Sigmoid function, X is the original feature data set, W e is the encoder weight parameter, B e is the encoder bias, and H is the mapping of the original feature data in the hidden layer.
[0180] Step 3: The decoder reconstructs the original feature data in the manner shown in the following formula (4) and outputs the reconstructed feature data.
[0181]
[0182] Among them, σ d is the Sigmoid function, W d is the weight parameter of the decoder, B d is the decoder bias, is the original feature data after reconstruction.
[0183] Step 4: Train the autoencoder model to reduce the reconstruction error and obtain the corresponding W e 、W d and
[0184] For example, the calculation formula of the reconstruction error is shown in the following formula (5):
[0185]
[0186] in, is the softmax reconstruction error, X i is the original feature data, To reconstruct feature data.
[0187] S602. Input the normal value and the fault value of at least one first operating data of the second multi-split air-conditioning system into the trained autoencoder model to obtain the data reconstructed normal value and the data reconstructed fault value of at least one first operating data of the second multi-split air-conditioning system.
[0188] S603: Reconstructing a normal value based on the data and reconstructing a fault value based on the data, and training a fault diagnosis model.
[0189] The fault diagnosis model training method for the multi-split air-conditioning system in the above-mentioned embodiment reconstructs the normal value and fault value of at least one first operating data of the second multi-split air-conditioning system through an autoencoder, which can reduce the data noise of the normal value and fault value of the at least one first operating data (i.e., random errors in the data caused by measurement, etc.), thereby removing the data in the normal value and fault value of the at least one first operating data that has a misleading effect on the training fault detection model, thereby improving the accuracy of fault detection using the fault detection model.
[0190] Those skilled in the art will understand that the scope of the present invention is not limited to the above specific embodiments, and that certain elements of the embodiments may be modified and replaced without departing from the spirit of the present application. The scope of the present application is limited by the appended claims.
Claims
1. A multi-split air conditioning system, comprising: Outdoor unit; a plurality of indoor units, each of the plurality of indoor units being connected to a gas pipe and a liquid pipe, so as to be in communication with the outdoor unit through the gas pipe and the liquid pipe; a plurality of electronic expansion valves corresponding to the plurality of indoor units; the plurality of electronic expansion valves are respectively disposed in the liquid pipes connected to the corresponding indoor units and are configured to control the refrigerant output or refrigerant input of the corresponding indoor units; as well as The controller is configured as: In the event of an electronic expansion valve failure in the multi-split air conditioning system, characteristic data of each of the plurality of indoor units is obtained; the characteristic data of each indoor unit includes a temperature difference between a temperature value of the liquid pipe corresponding to the indoor unit and a temperature value of the gas pipe corresponding to the indoor unit; determining an abnormal indoor unit among the plurality of indoor units based on the characteristic data of each indoor unit among the plurality of indoor units; The electronic expansion valve corresponding to the abnormal indoor unit is determined to be a faulty electronic expansion valve.
2. The multi-split air conditioning system according to claim 1, wherein: The controller is configured to: Inputting the characteristic data of each indoor unit among the plurality of indoor units into a fault recognition model based on a deep neural network (DNN), and determining the abnormal indoor unit among the plurality of indoor units according to a fault recognition result of the fault recognition model; The outdoor unit includes a compressor; Each of the plurality of indoor units comprises: an air outlet and an air return outlet; The characteristic data of each indoor unit also includes at least one of the following: the exhaust pressure value of the compressor, the suction pressure value of the compressor, the exhaust temperature value of the compressor, the suction temperature value of the compressor, the outlet temperature value of the air outlet of the indoor unit, or the return air temperature value of the return air outlet of the indoor unit.
3. The multi-split air conditioning system according to claim 1 or 2, wherein: The outdoor unit includes a compressor; Each of the plurality of indoor units comprises: an air outlet and an air return outlet; The controller is further configured to: Acquiring operating data of the multi-split air conditioning system, the operating data including at least one of the following: an operating current value of the compressor, an exhaust pressure value of the compressor, an exhaust temperature value of the compressor, outlet air temperatures of the air outlets of the multiple indoor units, or return air temperatures of the return air outlets of the multiple indoor units; According to the operating data, it is determined whether the electronic expansion valve failure occurs in the multi-split air conditioning system.
4. The multi-split air conditioning system according to claim 3, wherein: The controller is configured to: The operating data is input into a fault diagnosis model based on a support vector machine (SVM), and according to a fault diagnosis result of the fault diagnosis model, it is determined whether the electronic expansion valve fault occurs in the multi-split air conditioning system.
5. The multi-split air conditioning system according to any one of claims 1 to 4, wherein: The controller is further configured to: In the case that the electronic expansion valve failure does not occur in the multi-split air-conditioning system, a first prompt message is issued, where the first prompt message is used to prompt that the electronic expansion valve failure does not occur in the multi-split air-conditioning system.
6. The multi-split air conditioning system according to any one of claims 1 to 5, wherein: The controller is further configured to: After determining the faulty electronic expansion valve, obtaining the opening degree of the faulty electronic expansion valve; determining a fault level of the faulty electronic expansion valve according to the opening degree of the faulty electronic expansion valve; A second prompt message is issued, where the second prompt message is used to prompt the fault level of the faulty electronic expansion valve.
7. A method for locating a fault in a multi-split air conditioning system, wherein: The multi-split air conditioning system comprises: Outdoor unit; a plurality of indoor units, each of the plurality of indoor units being connected to a gas pipe and a liquid pipe so as to communicate with the outdoor unit through the gas pipe and the liquid pipe; and a plurality of electronic expansion valves corresponding to the plurality of indoor units; the plurality of electronic expansion valves are respectively disposed in the liquid pipes connected to the corresponding indoor units and are configured to control the refrigerant output or refrigerant input of the corresponding indoor units; The method comprises: In the event of an electronic expansion valve failure in the multi-split air conditioning system, characteristic data of each of the plurality of indoor units is obtained; the characteristic data of each indoor unit includes a temperature difference between a temperature value of the liquid pipe corresponding to the indoor unit and a temperature value of the gas pipe corresponding to the indoor unit; determining an abnormal indoor unit among the plurality of indoor units based on the characteristic data of each indoor unit among the plurality of indoor units; The electronic expansion valve corresponding to the abnormal indoor unit is determined to be a faulty electronic expansion valve.
8. The method according to claim 7, wherein: Determining the abnormal indoor unit among the plurality of indoor units according to the characteristic data of each indoor unit among the plurality of indoor units includes: Inputting the characteristic data of each indoor unit among the plurality of indoor units into a fault recognition model based on a deep neural network (DNN), and determining the abnormal indoor unit among the plurality of indoor units according to a fault recognition result of the fault recognition model; The outdoor unit includes a compressor; Each of the plurality of indoor units comprises: an air outlet and an air return outlet; The characteristic data of each indoor unit also includes at least one of the following: the exhaust pressure value of the compressor, the suction pressure value of the compressor, the exhaust temperature value of the compressor, the suction temperature value of the compressor, the outlet temperature value of the air outlet of the indoor unit, or the return air temperature value of the return air outlet of the indoor unit.
9. The method according to claim 7 or 8, wherein The outdoor unit includes a compressor; Each of the plurality of indoor units comprises: an air outlet and an air return outlet; The method further comprises: Acquiring operating data of the multi-split air conditioning system, the operating data including at least one of the following: an operating current value of the compressor, an exhaust pressure value of the compressor, an exhaust temperature value of the compressor, outlet air temperatures of the air outlets of the multiple indoor units, or return air temperatures of the return air outlets of the multiple indoor units; According to the operating data, it is determined whether the electronic expansion valve failure occurs in the multi-split air conditioning system.
10. The method according to claim 9, wherein: Determining, based on the operating data, whether the electronic expansion valve failure occurs in the multi-split air conditioning system includes: The operating data is input into a fault diagnosis model based on a support vector machine (SVM), and according to a fault diagnosis result of the fault diagnosis model, it is determined whether the electronic expansion valve fault occurs in the multi-split air conditioning system.
11. The method according to any one of claims 7 to 10, further comprising: In the case that the electronic expansion valve failure does not occur in the multi-split air-conditioning system, a first prompt message is issued, where the first prompt message is used to prompt that the electronic expansion valve failure does not occur in the multi-split air-conditioning system.
12. The method according to any one of claims 7 to 11, further comprising: After determining the faulty electronic expansion valve, obtaining the opening degree of the faulty electronic expansion valve; determining a fault level of the faulty electronic expansion valve according to the opening degree of the faulty electronic expansion valve; A second prompt message is issued, where the second prompt message is used to prompt the fault level of the faulty electronic expansion valve.
13. A fault diagnosis model training method for a multi-split air conditioning system, comprising: Acquiring normal values and fault values of a plurality of operating data of the first multi-split air conditioning system; determining a first characteristic offset space of the first multi-split air conditioning system based on normal values and fault values of the plurality of operating data of the first multi-split air conditioning system, wherein the plurality of operating data are parameter information generated during operation of the multi-split air conditioning system, and the first characteristic offset space includes a difference between a fault value and a normal value of each of the plurality of operating data of the first multi-split air conditioning system; Correcting the first characteristic offset space based on differences between normal values of the plurality of operating data of the first multi-split air conditioning system and normal values of the plurality of operating data of the second multi-split air conditioning system to obtain a second characteristic offset space of the second multi-split air conditioning system; determining fault values of the plurality of operating data of the second multi-split air conditioning system based on normal values of the plurality of operating data of the second multi-split air conditioning system and the second characteristic offset space; A fault diagnosis model is trained based on normal values and fault values of the plurality of operating data of the second multi-split air-conditioning system.
14. The method according to claim 13, wherein: Training the fault diagnosis model based on the normal values and fault values of the plurality of operating data of the second multi-split air conditioning system includes: determining at least one first operating data from the plurality of operating data of the second multi-split air conditioning system based on normal values and fault values of the plurality of operating data of the second multi-split air conditioning system; The fault diagnosis model is trained according to the normal value and the fault value of the at least one first operating data of the second multi-split air-conditioning system.
15. The method according to claim 14, wherein Training the fault diagnosis model according to the normal value and the fault value of the first operating data of the second multi-split air conditioning system includes: training an autoencoder model based on a normal value and a fault value of the at least one first operating data of the second multi-split air conditioning system; Inputting a normal value and a fault value of the at least one first operating data of the second multi-split air-conditioning system into a trained autoencoder model to obtain a data-reconstructed normal value and a data-reconstructed fault value of the at least one first operating data of the second multi-split air-conditioning system; The fault diagnosis model is trained based on the data-reconstructed normal value and the data-reconstructed fault value.
16. The method according to any one of claims 13 to 15, wherein The multi-split air conditioning system comprises: Outdoor unit, including: a compressor configured to compress a refrigerant so as to circulate the refrigerant in the multi-split air conditioning system; a plurality of indoor units, connected to the outdoor unit; a plurality of electronic expansion valves corresponding to the plurality of indoor units and configured to control the refrigerant output or refrigerant input of the corresponding indoor units; The multiple operating data include the suction temperature value of the compressor, the exhaust temperature value of the compressor, the suction pressure value of the compressor, the exhaust pressure value of the compressor, the opening degree of the multiple electronic expansion valves, the operating current value of the compressor, the ambient temperature value of the outdoor unit, the average value of the ambient temperature values of the multiple indoor units, the suction superheat of the compressor, the exhaust superheat of the compressor, the operating temperature value of the outdoor unit and the operating temperature values of the multiple indoor units.
17. The method according to any one of claims 13 to 16, wherein Obtaining normal values and fault values of the plurality of operating data of the first multi-split air conditioning system includes: Acquiring historical operating data of the first multi-split air conditioning system; The historical operation data are analyzed to determine normal values of the multiple operation data of the first multi-split air-conditioning system when the first multi-split air-conditioning system is in a normal operating state and fault values when a fault occurs in the first multi-split air-conditioning system.
