State determination device, state determination method, failure estimation device, failure estimation method, air conditioning system, and air conditioner

Abstract

This state determination device comprises: a current detection unit that detects a measured current value, which is a measured value of a drive current of a compressor; and an estimation unit that estimates an estimated current value, which is an estimated value of the drive current of the compressor. Furthermore, the state determination device comprises a determination unit that calculates the ratio between the measured current value and the estimated current value, and determines the state of the compressor by using the calculated ratio. As a result, the state of the compressor can be determined even during normal operation without requiring any special operation or device.

Description

State determination device, state determination method, failure estimation device, failure estimation method, air conditioning system, and air conditioner

[0001] The present invention relates to a state determination device, a state determination method, a failure estimation device, a failure estimation method, an air conditioning system, and an air conditioner.

[0002] Various methods have been proposed for diagnosing a compressor that circulates refrigerant in a refrigerant circuit of an air conditioner, including a method of diagnosing a compressor failure by performing an oil amount adjustment operation that is different from the normal operation of the refrigerant circuit (Patent Document 1) and a method of diagnosing a compressor failure by incorporating a special detection device in an inverter circuit that drives the compressor (Patent Document 2).

[0003] The method of Patent Document 1 takes into consideration that the drive current of the compressor changes depending on the condition of the oil that lubricates the compressor, and performs an oil amount adjustment operation to adjust the oil amount to suppress variations in the operating state quantity due to the condition of the oil.The method then acquires the current compressor current value after performing the oil amount adjustment operation, compares the current compressor current value with the compressor current value under normal conditions, and diagnoses whether or not the compressor is faulty based on the comparison result.

[0004] In addition, the method of Patent Document 2 uses a detection device to frequency analyze the drive current waveform in the inverter circuit, compares the analysis results under normal conditions with the current analysis results, and diagnoses whether or not the compressor is faulty based on the comparison results.

[0005] JP 2014-98515 A JP 2017-221023 A

[0006] The diagnostic method of Patent Document 1 requires that an oil amount adjustment operation, which is different from normal operation, be performed in order to suppress variations in the operating state quantity due to the oil condition. To perform the oil amount adjustment operation, normal operation must be interrupted. This makes it impossible to diagnose the compressor while normal operation is being performed, and the interruption of normal operation during compressor diagnosis may cause discomfort to users.

[0007] Furthermore, the diagnostic method of Patent Document 2 requires that a detection device be incorporated into the inverter circuit that drives the compressor, making it difficult to diagnose the compressor in an existing air conditioner.

[0008] The present invention has been made in consideration of the above problems, and an object of the present invention is to provide a state determination device that can determine the state of a compressor even during normal operation without requiring any special operation or device.

[0009] A state determination device according to one aspect includes a current detection unit that detects an actual measured current value, which is a measured value of a drive current of a compressor, and an estimation unit that estimates an estimated current value, which is an estimated value of the drive current of the compressor. The state determination device further includes a determination unit that calculates a ratio between the actual measured current value and the estimated current value and determines the state of the compressor using the calculated ratio.

[0010] One aspect is that the compressor status can be determined during normal operation without requiring special operations or equipment.

[0011] FIG. 1 is an explanatory diagram showing an example of an air conditioning system according to a first embodiment. FIG. 2 is an explanatory diagram showing an example of an air conditioner according to the first embodiment. FIG. 3 is a Mollier diagram showing a refrigeration cycle of an air conditioner. FIG. 4 is a block diagram showing an example of a centralized controller according to the first embodiment. FIG. 5 is an explanatory diagram showing an example of a normal distribution of ratios. FIG. 6 is an explanatory diagram showing an example of a relationship between ratios, compressor states, and protection control details. FIG. 7 is a flowchart showing an example of processing operations of a learning unit involved in learning processing according to the first embodiment. FIG. 8 is a flowchart showing an example of processing operations of a centralized controller involved in state determination processing according to the first embodiment. FIG. 9 is a block diagram showing an example of an air conditioning system according to a second embodiment. FIG. 10 is a block diagram showing an example of a centralized controller according to the second embodiment. FIG. 11 is a block diagram showing an example of a centralized controller according to a third embodiment. FIG. 12 is an explanatory diagram showing an example of a normal distribution of ratios. FIG. 13 is a flowchart showing an example of processing operations of a learning unit involved in learning processing according to the third embodiment. FIG. 14 is a flowchart showing an example of processing operations of a centralized controller involved in failure determination processing according to the third embodiment. FIG. 15 is a block diagram showing an example of a centralized controller according to a fourth embodiment.

[0012] Hereinafter, with reference to the drawings, an embodiment of an air conditioning system including a state determination device and the like disclosed in the present application will be described in detail. Note that the disclosed technology is not limited to these embodiments. Furthermore, each embodiment described below may be modified as appropriate within the scope of not causing any contradiction.

[0013] <Configuration of Air Conditioning System> Fig. 1 is an explanatory diagram showing an example of an air conditioning system 100 of Example 1. The air conditioning system 100 shown in Fig. 1 includes an air conditioner 1, a server device 110, and a communication network 120. The air conditioner 1 is an air conditioner installed in each location. The server device 110 is a server that generates and manages a learning model of the air conditioner 1 in the air conditioning system 100. The communication network 120 is, for example, a communication network such as the Internet.

[0014] <Configuration of Air Conditioner> Fig. 2 is an explanatory diagram showing an example of an air conditioner 1 of Example 1. The air conditioner 1 shown in Fig. 2 has one outdoor unit 2, N indoor units 3, and a centralized controller 7 (N is a natural number equal to or greater than 2). The outdoor unit 2 is connected to each indoor unit 3 in parallel by liquid pipes 4 and gas pipes 5. The outdoor unit 2 and the indoor units 3 are connected by refrigerant piping such as the liquid pipes 4 and gas pipes 5, thereby forming a refrigerant circuit 6 of the air conditioner 1.

[0015] <Configuration of Outdoor Unit> The outdoor unit 2 has a compressor 11, a four-way valve 12, an outdoor heat exchanger 13, an outdoor unit expansion valve 14, a first shut-off valve 15, a second shut-off valve 16, an accumulator 17, an outdoor unit fan 18, and a first control unit 19. The compressor 11, the four-way valve 12, the outdoor heat exchanger 13, the outdoor unit expansion valve 14, the first shut-off valve 15, the second shut-off valve 16, and the accumulator 17 are connected to each other by refrigerant pipes described in detail below to form an outdoor-side refrigerant circuit that forms part of the refrigerant circuit 6.

[0016] The compressor 11 is, for example, a high-pressure vessel type variable capacity compressor whose operating capacity can be varied in response to the drive of a motor (not shown) whose rotation speed is controlled by an inverter circuit 11A. The refrigerant discharge side of the compressor 11 is connected to the first port 12A of the four-way valve 12 by a refrigerant pipe 21. The refrigerant suction side of the compressor 11 is also connected to the refrigerant outlet side of the accumulator 17 by a refrigerant pipe 22.

[0017] The four-way valve 12 is a valve for switching the flow direction of refrigerant in the refrigerant circuit 6, and includes first to fourth ports 12A to 12D. The first port 12A is connected to a discharge pipe of the compressor 11 by a refrigerant pipe 21. The second port 12B is connected to one refrigerant inlet / outlet of the outdoor heat exchanger 13 by a refrigerant pipe 23. The third port 12C is connected to a suction port of the compressor 11 by a refrigerant pipe 26. The fourth port 12D is connected to the second stop valve 16 by a refrigerant pipe 24.

[0018] The outdoor heat exchanger 13 is a heat exchanger that exchanges heat between the refrigerant and outside air that is drawn into the outdoor unit 2 by the rotation of the outdoor unit fan 18. One refrigerant inlet / outlet of the outdoor heat exchanger 13 is connected to the second port 12B of the four-way valve 12 by a refrigerant pipe 23. The other refrigerant inlet / outlet of the outdoor heat exchanger 13 is connected to the first stop valve 15 by a refrigerant pipe 25. The outdoor heat exchanger 13 functions as a condenser when the air conditioner 1 is performing cooling operation, and functions as an evaporator when the air conditioner 1 is performing heating operation.

[0019] The outdoor unit expansion valve 14 is provided in the refrigerant piping 25 and is an electronic expansion valve driven by a pulse motor (not shown). The outdoor unit expansion valve 14 adjusts the amount of refrigerant flowing into or out of the outdoor heat exchanger 13 by adjusting its opening depending on the number of pulses applied to the pulse motor. When the air conditioner 1 is performing heating operation, the opening of the outdoor unit expansion valve 14 is adjusted so that the refrigerant superheat on the refrigerant suction side of the compressor 11 becomes a target suction refrigerant superheat. When the air conditioner 1 is performing cooling operation, the opening of the outdoor unit expansion valve 14 is fully open.

[0020] The accumulator 17 is connected to the third port 12C of the four-way valve 12 by a refrigerant pipe 26. Furthermore, the accumulator 17 is connected to a refrigerant inlet of the compressor 11 by a refrigerant pipe 22. The accumulator 17 separates the refrigerant that has flowed into the accumulator 17 from the refrigerant pipe 26 into gas refrigerant and liquid refrigerant, and allows only the gas refrigerant to be drawn into the compressor 11.

[0021] The outdoor unit fan 18 is made of a resin material and is disposed near the outdoor heat exchanger 13. The outdoor unit fan 18 generates an air flow that takes in outside air from an air inlet (not shown) into the outdoor unit 2 in response to the rotation of a fan motor (not shown), and discharges the outside air that has exchanged heat with the refrigerant in the outdoor heat exchanger 13 to the outside of the outdoor unit 2 from an air outlet (not shown).

[0022] In addition, a plurality of sensors are arranged in the outdoor unit 2. A discharge pressure sensor 31 that detects the pressure of the refrigerant discharged from the compressor 11 as a discharge pressure value, and a discharge temperature sensor 32 that detects the temperature of the refrigerant discharged from the compressor 11, i.e., the discharge temperature, are arranged in the refrigerant pipe 21. A suction pressure sensor 33 that detects the pressure of the refrigerant sucked into the compressor 11 as a suction pressure value, and a suction temperature sensor 34 that detects the temperature of the refrigerant sucked into the compressor 11, i.e., the suction temperature, are arranged near the refrigerant inlet of the accumulator 17 in the refrigerant pipe 26.

[0023] A refrigerant temperature sensor 35 is disposed in the refrigerant piping 25 between the outdoor heat exchanger 13 and the outdoor unit expansion valve 14 to detect the temperature of the refrigerant flowing into the outdoor heat exchanger 13 or the temperature of the refrigerant flowing out of the outdoor heat exchanger 13. An outdoor air temperature sensor 36 is disposed near an air inlet (not shown) of the outdoor unit 2 to detect the temperature of the outdoor air flowing into the outdoor unit 2, i.e., the outdoor air temperature.

[0024] The first control unit 19 controls the entire outdoor unit 2. As shown in FIG. 1 , the first control unit 19 has a first communication unit 19A, a first input unit 19B, a first output unit 19C, and a first input / output unit 19D. The first communication unit 19A is an interface such as a universal asynchronous receiver transmitter (UART) that communicates with the second control unit 50 in the indoor unit 3 or with the centralized controller 7 via the communication network 8. The first input unit 19B is an interface that inputs sensor values ​​detected by the discharge pressure sensor 31, discharge temperature sensor 32, suction pressure sensor 33, suction temperature sensor 34, refrigerant temperature sensor 35, and outdoor air temperature sensor 36. The first output unit 19C is an interface that outputs signals that control the outdoor unit fan 18, the outdoor unit expansion valve 14, and the four-way valve 12. The first input / output unit 19D is an interface that outputs a signal to control the inverter circuit 11A and also inputs a signal to detect the output of the inverter circuit 11A, for example, an actual measured current value, which is the actual measured drive current value (actual measured value) flowing through the inverter circuit 11A when the compressor 11 is operating.

[0025] 2, the indoor unit 3 has an indoor heat exchanger 51, an indoor unit expansion valve 52, a shut-off valve 53, a shut-off valve 54, an indoor unit fan 55, a second control unit 50, and a remote control 70 (see FIG. 1). The indoor heat exchanger 51, the indoor unit expansion valve 52, the shut-off valve 53, and the shut-off valve 54 are connected to each other by refrigerant pipes 56, 57 described later, and constitute an indoor unit refrigerant circuit that forms part of the refrigerant circuit 6.

[0026] The indoor heat exchanger 51 is a heat exchanger that exchanges heat between the refrigerant and indoor air that is drawn into the indoor unit 3 from an air inlet (not shown) by the rotation of the indoor unit fan 55. One refrigerant inlet / outlet of the indoor heat exchanger 51 is connected to a shut-off valve 53 via a refrigerant pipe 56. The other refrigerant inlet / outlet of the indoor heat exchanger 51 is connected to a shut-off valve 54 via a refrigerant pipe 57. The indoor heat exchanger 51 functions as a condenser when the air conditioner 1 is performing heating operation. On the other hand, the indoor heat exchanger 51 functions as an evaporator when the air conditioner 1 is performing cooling operation.

[0027] The indoor unit expansion valve 52 is an electronic expansion valve provided in the refrigerant piping 56. When the indoor heat exchanger 51 functions as an evaporator, i.e., when the indoor unit 3 is performing cooling operation, the opening degree of the indoor unit expansion valve 52 is adjusted so that the refrigerant superheat degree at the refrigerant outlet (shutoff valve 54 side) of the indoor heat exchanger 51 becomes a target refrigerant superheat degree. Also, when the indoor heat exchanger 51 functions as a condenser, i.e., when the indoor unit 3 is performing heating operation, the opening degree of the indoor unit expansion valve 52 is adjusted so that the refrigerant subcooling degree at the refrigerant outlet (shutoff valve 53 side) of the indoor heat exchanger 51 becomes a target refrigerant subcooling degree. Here, the target refrigerant superheat degree and the target refrigerant subcooling degree are the refrigerant superheat degree and the refrigerant subcooling degree required for the indoor unit 3 to exhibit sufficient cooling or heating capacity.

[0028] The indoor unit fan 55 is made of a resin material and is disposed near the indoor heat exchanger 51. The indoor unit fan 55 is rotated by a fan motor (not shown) to draw indoor air into the indoor unit 3 through an air inlet (not shown) and generate an air flow for discharging the indoor air that has exchanged heat with the refrigerant in the indoor heat exchanger 51 into the room through an air outlet (not shown).

[0029] Various sensors are provided in the indoor unit 3. A liquid-side refrigerant temperature sensor 61 is disposed in the refrigerant piping 56, between the indoor heat exchanger 51 and the indoor unit expansion valve 52, to detect the temperature of the refrigerant flowing into or out of the indoor heat exchanger 51. A gas-side refrigerant temperature sensor 62 is disposed in the refrigerant piping 57 to detect the temperature of the refrigerant flowing out of or into the indoor heat exchanger 51. A suction temperature sensor 63 is disposed near an air inlet (not shown) of the indoor unit 3 to detect the temperature of the indoor air flowing into the indoor unit 3, i.e., the suction temperature.

[0030] The second control unit 50 controls the entire indoor unit 3. As shown in FIG. 1 , the second control unit 50 has a second communication unit 50A, a third communication unit 50B, a second input unit 50C, and a second output unit 50D. The second communication unit 50A is an interface such as a UART that communicates with the first control unit 19 in the outdoor unit 2 or with the centralized controller 7 via the communication network 8. The third communication unit 50B communicates with the remote control 70 via remote communication such as infrared communication. The second input unit 50C is an interface that inputs sensor values ​​detected by the liquid-side refrigerant temperature sensor 61, the gas-side refrigerant temperature sensor 62, and the suction temperature sensor 63. The second output unit 50D is an interface that outputs signals that control the indoor unit fan 55 and the indoor unit expansion valve 52.

[0031] <Operation of Refrigerant Circuit> Next, the flow of refrigerant in the refrigerant circuit 6 and the operation of each part during air conditioning operation of the air conditioner 1 in this embodiment will be described.

[0032] When the air conditioner 1 performs heating operation, the four-way valve 12 is switched so that the first port 12A and the fourth port 12D are connected and the second port 12B and the third port 12C are connected. As a result, the refrigerant circuit 6 becomes a heating cycle in which each indoor heat exchanger 51 functions as a condenser and the outdoor heat exchanger 13 functions as an evaporator. For ease of explanation, the flow of refrigerant during heating operation is indicated by solid arrows in Figure 2.

[0033] When the compressor 11 is driven with the refrigerant circuit 6 in the above state, the refrigerant discharged from the compressor 11 flows through the refrigerant pipe 21 and into the four-way valve 12, then flows from the four-way valve 12 through the refrigerant pipe 24 and into the gas pipe 5 via the second shut-off valve 16. The refrigerant flowing through the gas pipe 5 is diverted to each indoor unit 3 via each shut-off valve 54. The refrigerant that has flowed into each indoor unit 3 flows through each refrigerant pipe 57 and into each indoor heat exchanger 51. The refrigerant that has flowed into each indoor heat exchanger 51 condenses by exchanging heat with indoor air drawn into each indoor unit 3 by the rotation of each indoor unit fan 55. In other words, each indoor heat exchanger 51 functions as a condenser, and the indoor air heated by the refrigerant in each indoor heat exchanger 51 is blown into the room through an air outlet (not shown), thereby heating the room in which each indoor unit 3 is installed.

[0034] The refrigerant that flows from each indoor heat exchanger 51 into each refrigerant pipe 56 is decompressed by passing through each indoor unit expansion valve 52, the opening of which is adjusted so that the degree of refrigerant subcooling at the refrigerant outlet side of each indoor heat exchanger 51 becomes the target degree of refrigerant subcooling. Here, the target degree of refrigerant subcooling is determined based on the cooling capacity required by each indoor unit 3.

[0035] The refrigerant decompressed by each indoor unit expansion valve 52 flows from each refrigerant pipe 56 through each shut-off valve 53 into the liquid pipe 4. The refrigerant that joins in the liquid pipe 4 flows into the outdoor unit 2 through the first shut-off valve 15. The refrigerant that flows into the first shut-off valve 15 of the outdoor unit 2 flows through the refrigerant pipe 25 and is decompressed while passing through the outdoor unit expansion valve 14. The refrigerant that is decompressed by the outdoor unit expansion valve 14 flows through the refrigerant pipe 25 into the outdoor heat exchanger 13, and evaporates through heat exchange with outside air that flows in from an inlet (not shown) of the outdoor unit 2 due to the rotation of the outdoor unit fan 18. The refrigerant that flows out from the outdoor heat exchanger 13 into the refrigerant pipe 26 flows in this order: the four-way valve 12, the refrigerant pipe 26, the accumulator 17, and the refrigerant pipe 22. The refrigerant that has flowed in is then drawn into the compressor 11 and compressed again, and flows out into the refrigerant pipe 24 via the first port 12A and the fourth port 12D of the four-way valve 12.

[0036] When the air conditioner 1 is in cooling operation, the four-way valve 12 is switched so that the first port 12A and the second port 12B are in communication, and the third port 12C and the fourth port 12D are in communication. As a result, the refrigerant circuit 6 becomes a cooling cycle in which each indoor heat exchanger 51 functions as an evaporator and the outdoor heat exchanger 13 functions as a condenser. For ease of explanation, the flow of refrigerant during cooling operation is indicated by dashed arrows in Figure 2.

[0037] When the compressor 11 is driven in the refrigerant circuit 6, the refrigerant discharged from the compressor 11 flows through the refrigerant pipe 21 and into the four-way valve 12, and then flows from the four-way valve 12 through the refrigerant pipe 26 and into the outdoor heat exchanger 13. The refrigerant that has flowed into the outdoor heat exchanger 13 condenses by exchanging heat with outdoor air that has been drawn into the outdoor unit 2 by the rotation of the outdoor unit fan 18. In other words, the outdoor heat exchanger 13 functions as a condenser, and the indoor air heated by the refrigerant in the outdoor heat exchanger 13 is blown outside through an air outlet (not shown).

[0038] The refrigerant that flows from the outdoor heat exchanger 13 into the refrigerant piping 25 passes through the outdoor unit expansion valve 14, which is fully open, and is decompressed. The refrigerant decompressed by the outdoor unit expansion valve 14 flows through the liquid pipe 4 via the first shut-off valve 15 and is diverted to each indoor unit 3. The refrigerant that flows into each indoor unit 3 flows through the refrigerant piping 56 via each shut-off valve 53 and is decompressed by passing through the indoor unit expansion valve 52, which has an opening adjusted to such an extent that the refrigerant subcooling degree becomes the target refrigerant subcooling degree at the refrigerant outlet of the indoor heat exchanger 51. The refrigerant decompressed by the indoor unit expansion valve 52 flows through the refrigerant piping 56 and flows into the indoor heat exchanger 51, where it exchanges heat with indoor air that flows in from an intake port (not shown) of the indoor unit 3 due to the rotation of the indoor unit fan 55, and evaporates. In other words, each indoor heat exchanger 51 functions as an evaporator, and the indoor air cooled by the refrigerant in each indoor heat exchanger 51 is blown into the room from an air outlet (not shown), thereby cooling the room in which each indoor unit 3 is installed.

[0039] The refrigerant flowing from the indoor heat exchanger 51 to the gas pipe 5 via the shut-off valve 54 flows into the refrigerant pipe 24 via the second shut-off valve 16 of the outdoor unit 2 and flows into the fourth port 12D of the four-way valve 12. The refrigerant that has flowed into the fourth port 12D of the four-way valve 12 flows from the third port 12C into the refrigerant inlet side of the accumulator 17. The refrigerant that has flowed into the refrigerant inlet side of the accumulator 17 flows through the refrigerant pipe 22, is sucked into the compressor 11, and is compressed again.

[0040] 3 is a Mollier diagram showing the refrigeration cycle of the air conditioner 1. During cooling operation of the air conditioner 1, the outdoor heat exchanger 13 functions as a condenser, and the indoor heat exchanger 51 functions as an evaporator. During heating operation of the air conditioner 1, the outdoor heat exchanger 13 functions as an evaporator, and the indoor heat exchanger 51 functions as a condenser.

[0041] The compressor 11 compresses the low-temperature, low-pressure gas refrigerant flowing in from the evaporator and discharges the high-temperature, high-pressure gas refrigerant (refrigerant in the state of point B in Figure 3). The temperature of the gas refrigerant discharged from the compressor 11 is the discharge temperature, and the discharge temperature is detected by the discharge temperature sensor 32.

[0042] The condenser condenses the high-temperature, high-pressure gas refrigerant from the compressor 11 by heat exchange with air. During this process, after the gas refrigerant has changed to liquid due to latent heat change, the temperature of the liquid refrigerant drops due to sensible heat change, resulting in a supercooled state (point C in Figure 3). The temperature at which the gas refrigerant changes to liquid due to latent heat change is the high-pressure saturation temperature, and the temperature of the supercooled refrigerant at the condenser outlet is the heat exchanger outlet temperature. The high-pressure saturation temperature is calculated using the discharge pressure value (pressure value P2, labeled "HPS" in Figure 3) detected by the discharge pressure sensor 31. The heat exchanger outlet temperature is detected by the refrigerant temperature sensor 35. Using a Mollier diagram, the discharge specific enthalpy can be calculated based on the discharge pressure value and discharge temperature. For ease of explanation, the air conditioner 1 is illustrated as being equipped with a discharge temperature sensor 32 that detects the discharge temperature. However, even if the discharge temperature cannot be detected, the discharge specific enthalpy can be calculated based on the discharge pressure value P2 and the discharge superheat. However, in this case, the discharge superheat is an estimated value obtained by conducting tests in advance, and if the state of the refrigerant circuit 6 differs between when the test was conducted and when the refrigerant circuit 6 is actually in operation, the accuracy of calculating the discharge specific enthalpy may be reduced compared to when the actually measured discharge temperature is used. For this reason, it is desirable to install a discharge temperature sensor 32 and calculate the discharge superheat using the actually measured discharge temperature.

[0043] The expansion valve reduces the pressure of the low-temperature, high-pressure refrigerant flowing out from the condenser to form a gas-liquid two-phase refrigerant in which gas and liquid are mixed (refrigerant in the state of point D in FIG. 3).

[0044] The evaporator evaporates the two-phase gas-liquid refrigerant that has flowed in through heat exchange with air. After the two-phase gas-liquid refrigerant is completely converted into gas refrigerant through latent heat change in the evaporator, the temperature of the gas refrigerant rises through sensible heat change, becoming superheated (point A in FIG. 3 ), and is then drawn into the compressor 11. The temperature at which the liquid refrigerant is undergoing a phase change into gas refrigerant through latent heat change is the low-pressure saturation temperature. The low-pressure saturation temperature is determined using the suction pressure value (pressure value P1 indicated as "LPS" in FIG. 3 ) detected by the suction pressure sensor 33. The temperature of the refrigerant that has been superheated in the evaporator and drawn into the compressor 11 is the suction temperature. The suction temperature is detected by the suction temperature sensor 34. Using a Mollier diagram, the suction specific enthalpy can be calculated based on the suction pressure value and the suction temperature. For ease of explanation, the air conditioner 1 is shown equipped with an intake temperature sensor 34 that detects intake temperature. However, even if the intake temperature cannot be detected, the intake specific enthalpy can be calculated based on the intake pressure value P1 and the intake superheat. However, the intake superheat in this case is an estimated value obtained by conducting a test or the like in advance. If the state of the refrigerant circuit 6 differs between the test and actual operation, the accuracy of the calculation of the intake specific enthalpy may be reduced compared to when the actually measured intake temperature is used. For this reason, it is desirable to install an intake temperature sensor 34 and calculate the intake superheat using the actually measured intake temperature.

[0045] The degree of refrigerant subcooling when the refrigerant is in a subcooled state when it flows out of the condenser can be calculated by subtracting the refrigerant temperature at the refrigerant outlet of the heat exchanger functioning as a condenser (the heat exchange outlet temperature) from the high-pressure saturation temperature. The degree of refrigerant superheat when the refrigerant is in a superheated state when it flows out of the evaporator can be calculated by subtracting the suction temperature from the low-pressure saturation temperature.

[0046] <Configuration of Centralized Controller> FIG. 4 is a block diagram showing an example of the centralized controller 7 according to the first embodiment. The centralized controller 7 shown in FIG. 4 includes a first communication unit 71, a second communication unit 72, a storage unit 73, and a control unit 74. The first communication unit 71 is an interface such as a UART that communicates with the server device 110 via the communication network 120. The second communication unit 72 is an interface such as a UART that communicates with the first control unit 19 in the outdoor unit 2 or the second control unit 50 in the indoor unit 3 via the communication network 8. The storage unit 73 is, for example, a flash memory, and stores various information, programs, and the like. The storage unit 73 stores a learning model 73A generated by the server device 110.

[0047] The control unit 74 functions as a state determination device that determines the state of the compressor 11. The control unit 74 has a detection unit 741, an acquisition unit 742, a calculation unit 743, an estimation unit 744, a determination unit 745, and a protection control unit 746. The detection unit 741 acquires, from the first control unit 19 of the outdoor unit 2, for example, a measured current value, which is the current value flowing through the primary side of the inverter circuit 11A when the compressor 11 is driven. Furthermore, the acquisition unit 742 acquires, in addition to the above-mentioned measured current value, characteristic quantities related to the drive of the compressor 11, which will be described below. The characteristic quantities related to the drive of the compressor 11 include the rotation speed of the compressor 11, refrigerant state quantities related to the drive of the compressor 11, and state quantities (coefficients) related to the efficiency of the compressor 11. The state quantities of the refrigerant related to the operation of the compressor 11 are the sensor values ​​of the discharge pressure sensor 31, discharge temperature sensor 32, suction pressure sensor 33, suction temperature sensor 34, refrigerant temperature sensor 35, and outside air temperature sensor 36 in the outdoor unit 2 from the first control unit 19. Furthermore, the acquisition unit 742 acquires the sensor values ​​of the liquid side refrigerant temperature sensor 61, gas side refrigerant temperature sensor 62, and suction temperature sensor 63 from the second control unit 50 of each indoor unit 3 as the state quantities of the refrigerant related to the operation of the compressor 11.

[0048] The acquisition unit 742 also acquires a state quantity (coefficient) related to the efficiency of the compressor 11 from the learning model 73A. The learning model 73A estimates the state quantity (coefficient) related to the efficiency of the compressor 11 using the rotation speed of the compressor 11 and a state quantity of the refrigerant related to the operation of the compressor 11. The estimation of the state quantity (coefficient) related to the efficiency of the compressor 11 will be described in detail later.

[0049] That is, the acquisition unit 742 periodically acquires the measured current value, the rotation speed of the compressor 11, the state quantities of the refrigerant related to the operation of the compressor 11, and the state quantities (coefficients) related to the efficiency of the compressor 11.

[0050] The calculation unit 743 calculates a refrigerant suction density, which is the density of the refrigerant sucked into the compressor 11, using the acquired state quantities of the refrigerant related to the operation of the compressor 11. Specifically, the calculation unit 743 refers to a table that associates the refrigerant suction density with the suction pressure value and suction temperature of the compressor 11, and acquires the refrigerant suction density corresponding to the suction pressure value and the suction temperature. Note that the table is obtained in advance by performing tests or the like, and is stored in the storage unit 73 of the centralized controller 7 (not shown in FIG. 4 ).

[0051] The calculation unit 743 calculates a specific enthalpy difference, which is the difference between the discharge specific enthalpy, which is the specific enthalpy of the refrigerant discharged from the compressor 11, and the suction specific enthalpy, which is the specific enthalpy of the refrigerant drawn into the compressor 11. Specifically, the calculation unit 743 calculates the specific enthalpy difference by subtracting the suction specific enthalpy from the discharge specific enthalpy. The calculation unit 743 calculates the discharge specific enthalpy based on the Mollier diagram shown in FIG. 3, the discharge pressure value, and the discharge temperature. Furthermore, the calculation unit 743 calculates the suction specific enthalpy based on the Mollier diagram shown in FIG. 3, the suction pressure value, and the suction temperature. The Mollier diagram shown in FIG. 3 is calculated in advance through testing or the like and is stored in the memory unit 73 of the centralized controller 7 (not shown in FIG. 4). The memory unit 73 stores, for example, the discharge pressure value, the discharge temperature, the suction pressure value, the suction temperature, the suction specific enthalpy, the discharge specific enthalpy, and the like in the Mollier diagram shown in FIG. 3.

[0052] The estimation unit 744 uses the feature quantity to estimate an estimated current value as a drive current value corresponding to the operating load of the compressor 11. The estimation unit 744 estimates the estimated current value using the refrigerant suction density (represented as p in the following formula), the rotation speed of the compressor 11, the specific enthalpy difference, and a state quantity (coefficient, represented as k in the following formula) related to the efficiency of the compressor 11.

[0053] Formula: Estimated current value = k × p × (rotational speed of compressor 11) × specific enthalpy difference

[0054] The estimation unit 744 estimates the estimated current value each time the acquisition unit 742 acquires the rotation speed of the compressor 11 and the state quantity of the refrigerant related to the operation of the compressor 11 (for example, every 5 minutes).

[0055] The determination unit 745 compares the estimated current value estimated by the estimation unit 744 with the actual current value detected by the detection unit 741, and calculates, for example, the ratio between the estimated current value and the actual current value (actual current value ÷ estimated current value). The determination unit 745 determines the state of the compressor 11 based on the calculated ratio. The states of the compressor 11 include normal, deterioration, and failure. Normal refers to a state in which the compressor 11 is operating normally. Failure refers to a state in which an abnormality has occurred in the compressor 11. Deterioration refers to a state in which the performance of the compressor 11 is reduced compared to when the compressor 11 is normal, and there is a possibility that the compressor 11 will fail in the near future (e.g., one to two months from now). In other words, the determination unit 745 constantly determines the state of the compressor 11 based on the calculated ratio, even during normal operation, such as heating operation, cooling operation, or dehumidification operation.

[0056] Next, a method for the determination unit 745 to determine the state of the compressor 11 using the ratio described above will be described. The determination unit 745 determines the state of the compressor 11 by checking which range (hereinafter referred to as the determination range) of the normal distribution described below the calculated ratio falls within. FIG. 5 is an explanatory diagram showing an example of a normal distribution of ratios. The determination unit 745 determines the state of the compressor 11 based on whether the ratio falls within the determination range in the normal distribution. Here, the determination range has a first predetermined range and a second predetermined range. As shown in FIG. 5, the first predetermined range is, for example, μ±1σ, where σ is the standard deviation of the occurrence of the statistically calculated ratio and μ is the mean value of the occurrence of the ratio. In the first predetermined range, the ratio falls within the range of 0.8 to 1.2. If the ratio is within the μ±1σ range, the state of the compressor 11 can be estimated to be normal. Approximately 68% of the calculated ratio falls within the μ±1σ range. On the other hand, the remaining approximately 32% of the calculated ratio falls outside the μ±1σ range. Furthermore, the situation in which the calculated ratio frequently falls outside the specified range is unlikely to occur when the compressor 11 is operating normally, and it can be assumed that the compressor 11 has begun to deteriorate or has deteriorated to a certain extent.

[0057] Furthermore, the second predetermined range is, for example, μ±2σ, as shown in FIG. 5 , where σ is the standard deviation of the occurrence of the statistically calculated ratio and μ is the mean value of the occurrence of the ratio. That is, the second predetermined range is set from the first upper limit obtained by adding the predetermined value σ to the upper limit of the first predetermined range to the first lower limit obtained by subtracting the predetermined value σ from the lower limit of the first predetermined range. In the second predetermined range, the ratio falls within the range of μ±2σ, with approximately 95% of the calculated ratio falling within the range. On the other hand, the remaining approximately 5% of the calculated ratio falls outside the range of μ±2σ. Frequent occurrences of the calculated ratio falling outside the second predetermined range are unlikely to occur when the compressor 11 is operating normally, and it can be assumed that an abnormality has occurred in the compressor 11, i.e., that the compressor 11 is malfunctioning. Furthermore, when the ratio is within the second predetermined range, it can be assumed that the compressor 11 is degraded, except when the ratio is within the first predetermined range.

[0058] The states of the compressor 11 corresponding to the ratios corresponding to the first predetermined range (μ±1σ) and the second predetermined range (μ±2σ) shown in FIG. 5 are as shown in FIG. 6 . FIG. 6 is an explanatory diagram showing an example of the relationship between the ratio, the state of the compressor 11, and the protection control content. When the ratio is less than 0.6, the ratio is outside the second predetermined range, and the state of the compressor 11 is abnormal. When the ratio is 0.6≦ratio<0.8, the ratio is outside the first predetermined range but within the second predetermined range, and the state of the compressor 11 is degraded. When the ratio is 0.8≦ratio<1.2, the ratio is within the first predetermined range, and the state of the compressor 11 is normal. When the ratio is 1.2≦ratio<1.4, the ratio is outside the first predetermined range but within the second predetermined range, and the state of the compressor 11 is degraded. When the ratio is ≧1.4, the ratio is outside the second predetermined range, and the state of the compressor 11 is abnormal.

[0059] That is, the determination unit 745 determines whether the calculated ratio is within a first predetermined range. If the calculated ratio is within the first predetermined range, as shown in Figures 5 and 6, the determination unit 745 determines that the compressor 11 is normal.

[0060] Furthermore, when the calculated ratio is not within the first predetermined range, i.e., when the ratio is outside the first predetermined range, the determination unit 745 determines whether the calculated ratio is within a second predetermined range. As shown in Figures 5 and 6, when the calculated ratio is outside the first predetermined range but within the second predetermined range, the determination unit 745 determines that the compressor 11 is deteriorated. Furthermore, when the calculated ratio is not within the second predetermined range, i.e., when the ratio is outside the second predetermined range, the determination unit 745 determines that the compressor 11 is faulty.

[0061] In this embodiment, when the calculated ratio frequently falls outside the second predetermined range and the actual measurement average value μ falls outside the range of ±2σ, the compressor 11 is determined to be faulty. The narrower the second predetermined range, the higher the sensitivity for determining that the compressor 11 is faulty, but the greater the possibility of misjudging a compressor 11 that is not faulty. Therefore, in this embodiment, the ±2σ range is used, which provides a good balance between the sensitivity and accuracy of the fault determination. However, if higher sensitivity is desired, the predetermined range can be changed to ±3σ, and the setting can be changed.

[0062] The determination unit 745 sequentially calculates the ratio between the estimated current value and the actually measured current value (periodically, for example, every five minutes, when the rotation speed of the compressor 11 and the state quantities of the refrigerant related to the operation of the compressor 11 are acquired), calculates the average value of the ratio for a predetermined period of time, for example, for one hour, and determines whether the calculated average value of the ratio is within a first predetermined range. If the average value of the ratio is within the first predetermined range, the determination unit 745 determines that the compressor 11 is normal.

[0063] Furthermore, when the average value of the ratio is outside the first predetermined range, the determination unit 745 determines whether the average value of the ratio is within a second predetermined range. When the average value of the ratio is within the second predetermined range, the determination unit 745 determines that the compressor 11 is deteriorated. When the average value of the ratio is outside the second predetermined range, the determination unit 745 determines that the compressor 11 is malfunctioning.

[0064] When the determination unit 745 determines that the compressor 11 is deteriorated, the protection control unit 746 controls the compressor 11 to reduce the load on the compressor 11 according to the ratio. Methods for reducing the load on the compressor 11 according to the ratio include, for example, limiting the upper limit rotation speed of the compressor 11 according to the ratio, and limiting the pressure ratio of the compressor 11 (the ratio between the discharge pressure detected by the discharge pressure sensor 31 and the suction pressure detected by the suction pressure sensor) according to the ratio. Furthermore, when the determination unit 745 determines that the compressor 11 is faulty, the protection control unit 746 stops the operation of the compressor 11.

[0065] The protective control of the protection control unit 746 will now be described with reference to FIG. 6 . In this embodiment, the protective controls include an upper limit rotation speed reduction control that reduces the upper limit rotation speed of the compressor 11, and a pressure ratio reduction control that reduces the pressure ratio between the discharge pressure and the suction pressure of the compressor 11. When it is determined that the compressor 11 is degraded, at least one of these protective controls is executed. FIG. 6 shows an example of a table referenced by the protection control unit 746 when performing the upper limit rotation speed reduction control and the pressure ratio reduction control. Depending on the ratio, a lower limit rotation speed reduction rate indicating the degree to which the upper limit rotation speed of the compressor 11 is reduced, the upper limit rotation speed (unit: rps) of the compressor 11, a lower limit pressure ratio reduction rate indicating the degree to which the upper limit value of the pressure ratio of the compressor 11 is reduced, and the upper limit pressure ratio of the compressor 11 are respectively defined.

[0066] If the calculated ratio is 0.8≦ratio<less than 1.2 (a value within a first predetermined range), the protection control unit 746 determines that the compressor 11 is normal, and does not reduce the upper limit rotation speed or upper limit pressure ratio of the compressor 11 by referring to the table in Fig. 6. Furthermore, if the calculated ratio is less than 0.6 or equal to or greater than 1.4 (a value outside a second predetermined range), the protection control unit 746 determines that the compressor 11 is faulty, and stops the compressor 11 by referring to the table in Fig. 6.

[0067] If the calculated ratio is 0.6≦ratio<0.8 or 1.2≦ratio<1.4 (a value outside the first predetermined range and within the second predetermined range), the protection control unit 746 determines that the compressor 11 is deteriorated and reduces the upper limit rotation speed and the upper limit pressure ratio of the compressor 11 by referring to the table in Fig. 6. Specifically, if the calculated ratio is 0.7≦ratio<0.8 or 1.2≦ratio<1.3, the protection control unit 746 refers to the table in Fig. 6 and executes at least one of upper limit rotation speed reduction control in which the upper limit rotation speed reduction rate is 80% and the upper limit rotation speed is 96 rps, and pressure ratio reduction control in which the upper limit pressure ratio reduction rate is 80% and the upper limit pressure ratio is 12.0. Furthermore, if the calculated ratio is less than 0.6≦ratio<0.7, or if the calculated ratio is 1.3≦ratio<1.4, the protection control unit 746 refers to the table in FIG. 6 and executes at least one of upper limit rotation speed reduction control in which the upper limit rotation speed reduction rate is set to 70% and the upper limit rotation speed to 84 rps, and pressure ratio reduction control in which the upper limit pressure ratio reduction rate is set to 70% and the upper limit pressure ratio is set to 10.5.

[0068] As described above, when the protection control unit 746 determines that the compressor 11 is deteriorated, it executes at least one of the protective controls of the upper limit rotation speed reduction control and the pressure ratio reduction control to reduce the load on the compressor 11. By performing such protective control, the rate of deterioration of the compressor 11 is slowed, thereby lengthening the time until a failure occurs. Lengthening the time until the compressor 11 fails allows a replacement compressor 11 to be arranged and obtained before the failure occurs, and also allows the compressor 11 replacement work to be scheduled in advance. Therefore, compared to arranging for a replacement compressor 11 and scheduling the replacement work after the compressor 11 fails, the compressor 11 can be replaced more quickly and before the compressor 11 fails and stops. This avoids the inconvenience to users of the air conditioning being unable to operate between the compressor 11 failure and replacement.

[0069] When performing protection control, the upper limit rotation speed reduction rate and the upper limit pressure ratio reduction rate are increased as the ratio decreases from the lower limit of the first predetermined range (=0.9) or increases from the upper limit of the first predetermined range (=1.2). When it is determined that the compressor 11 is degraded but the degree of degradation is small (in this embodiment, when the ratio is 0.7≦ratio<0.8 or 1.2≦ratio<1.3), the upper limit rotation speed reduction rate and the upper limit pressure ratio reduction rate are minimized, thereby suppressing the degradation of the compressor 11 and minimizing the decline in air conditioning capacity. When the degree of degradation of the compressor 11 is large (in this embodiment, when the ratio is 0.6≦ratio<0.7 or 1.3≦ratio<1.4), the upper limit rotation speed reduction rate and the upper limit pressure ratio reduction rate are increased to maximize the rate of degradation of the compressor 11.

[0070] <Configuration of Server Device> As shown in FIG. 1 , the server device 110 includes a storage unit 111 and a learning unit 112. The storage unit 111 periodically collects, for example, the rotation speed, discharge pressure, and suction pressure of the compressor 11 as feature quantities from the centralized controller 7, for example, every five minutes, and stores the collected feature quantities. The learning unit 112 generates a learning model 73A for calculating a state quantity (coefficient) k related to the efficiency of the compressor 11 using feature quantities stored in the storage unit 111 over a certain period of time, for example, 30 days. Thereafter, the learning model 73A is updated every 30 days using the feature quantities acquired up to that point. The learning unit 112 provides the generated or updated learning model 73A to the centralized controller 7. The centralized controller 7 stores the learning model 73A received from the server device 110 in the memory unit 73.

[0071] <Learning Model> As described above, the state quantity (coefficient) k related to the efficiency of the compressor 11 can be calculated using the learning model 73A. The state quantity (coefficient) k related to the efficiency of the compressor 11 will now be described. The state quantity (coefficient) k related to the efficiency of the compressor 11 can be calculated, for example, using the volumetric efficiency, compression efficiency, and swept volume (displacement volume) of the compressor 11. However, the volumetric efficiency and compression efficiency of the compressor 11 vary depending on the rotation speed of the compressor 11 and the pressure ratio (ratio of high pressure to low pressure) in the refrigerant circuit 6. Furthermore, for example, the delivery note and specifications for the compressor 11 only list values ​​under limited operating conditions. Since there is no data for all expected conditions, it is difficult to store the values ​​in advance, for example, as a table in a memory unit. Furthermore, while the swept volume of the compressor 11 can be obtained for compressors manufactured in-house, it is often difficult to obtain the value when purchasing a compressor from another company because the delivery note and specifications often do not include the value. Furthermore, in the case of an existing air conditioner, there may be cases where the information and specifications of the installed compressor are unknown. Therefore, it is difficult to determine the state quantity (coefficient) k related to the efficiency of the compressor 11 from the volumetric efficiency, compression efficiency, and stroke volume of the compressor 11. Therefore, in this embodiment, the state quantity (coefficient) k related to the efficiency of the compressor 11 is determined using the learning model 73A based on the rotation speed, discharge pressure value, and suction pressure value of the compressor 11.

[0072] The learning model 73A is generated by multiple regression analysis, a type of regression analysis, using the rotation speed, discharge pressure, and suction pressure of the compressor 11 among multiple feature quantities. In multiple regression analysis, the regression equation obtained from multiple simulation results is selected to have the smallest P value (a value (predetermined weight parameter) indicating the degree of influence that the operating state quantity has on the accuracy of the generated learning model) and the largest possible corrected R2 (a value indicating the accuracy of the generated learning model) between 0.9 and 1.0. Here, the P value and corrected R2 are values ​​related to the accuracy of the learning model when generating the learning model using multiple regression analysis. The smaller the P value and the closer the corrected R2 is to 1.0, the higher the accuracy of the generated learning model.

[0073] The learning model 73A is a regression equation for calculating a state quantity (coefficient) k related to the efficiency of the compressor 11. The regression equation is, for example, (α1 × rotation speed of the compressor 11) + (α2 × discharge pressure value) + (α3 × suction pressure value) + α4. The coefficients α1 to α4 are determined when calculating the state quantity (coefficient) k related to the efficiency of the compressor 11. The control unit 74 calculates the state quantity (coefficient) k related to the efficiency of the compressor 11 at the current time by substituting the current rotation speed, discharge pressure value, and suction pressure value of the compressor 11 acquired by the acquisition unit 742 into the regression equation.

[0074] <Learning Process> Here, the learning process performed when the learning unit 112 generates the learning model 73A will be described. Fig. 7 is a flowchart showing an example of the processing operation of the learning unit 112 related to the learning process of the first embodiment. In Fig. 7, the learning unit 112 in the server device 110 periodically (e.g., every 5 minutes) acquires the rotation speed, discharge pressure value, and suction pressure value of the compressor 11 from the centralized controller 7 (step S11). Then, the learning unit 112 sequentially accumulates the acquired rotation speed, discharge pressure value, and suction pressure value of the compressor 11 in the accumulation unit 111.

[0075] The learning unit 112 generates a learning model 73A using the rotation speed, discharge pressure value, and suction pressure value of the compressor 11 for a certain period, for example, 30 days, acquired in step S11 (step S12). The learning unit 112 transmits the generated learning model 73A to the centralized controller 7 via the communication network 120 (step S13), thereby completing the processing operation shown in Fig. 7. The centralized controller 7 then stores the learning model 73A received via the communication network 120 in the memory unit 73.

[0076] <State Determination Process> Next, a state determination process that allows the centralized controller 7 to constantly determine the state of the compressor 11 will be described. FIG. 8 is a flowchart showing an example of the processing operation of the centralized controller 7 related to the state determination process in the first embodiment. In FIG. 8, the control unit 74 in the centralized controller 7 determines whether or not the current time is an acquisition timing (step S21). The acquisition timing is every five minutes as described above. If the current time is an acquisition timing (step S21: Yes), the acquisition unit 742 in the control unit 74 acquires the rotation speed, discharge pressure value, suction pressure value, and actual measured current value of the compressor 11 (step S22). The discharge pressure value is detected by the discharge pressure sensor 31, and the suction pressure value is detected by the suction pressure sensor 33. The actual measured current value is detected by the detection unit 741 from the inverter circuit 11A of the compressor 11.

[0077] The calculation unit 743 in the control unit 74 uses the learning model 73A to calculate the state quantity (coefficient) k related to the efficiency of the compressor 11 (step S23). The learning model 73A calculates the state quantity (coefficient) k related to the efficiency of the compressor 11 by substituting the rotation speed, discharge pressure value, and suction pressure value of the compressor 11 into a regression equation.

[0078] The estimator 744 in the controller 74 estimates the estimated current value of the compressor 11 using the calculated state variable (coefficient) k related to compressor efficiency, the refrigerant suction density of the compressor 11, the rotation speed of the compressor 11, the discharge specific enthalpy, and the suction specific enthalpy (step S24). The calculator 743 references a table that associates the refrigerant suction density with the suction pressure value and suction temperature of the compressor 11, and obtains the refrigerant suction density corresponding to the suction pressure value and the suction temperature. The calculator 743 calculates the discharge specific enthalpy based on the Mollier diagram shown in FIG. 3, the discharge pressure value, and the discharge temperature. Furthermore, the calculator 743 calculates the suction specific enthalpy based on the Mollier diagram shown in FIG. 3, the suction pressure value, and the suction temperature. Then, the estimation unit 744 estimates the estimated current value of the compressor 11 using (state quantity (coefficient) k related to efficiency) x (refrigerant suction density p) x (rotation speed of the compressor 11) x (specific enthalpy difference (discharge specific enthalpy - suction specific enthalpy)).

[0079] The determination unit 745 in the control unit 74 divides the actual current value by the estimated current value to calculate the current ratio (step S25). The determination unit 745 determines whether the ratio for a predetermined time period has been calculated (step S26). The predetermined time period is, for example, one hour. If the determination unit 745 has not calculated the ratio for the predetermined time period (step S26: No), the process returns to step S21. If the determination unit 745 has calculated the ratio for the predetermined time period (step S26: Yes), the determination unit 745 calculates an average ratio from the ratio for the predetermined time period (step S27).

[0080] Then, the determination unit 745 determines whether the calculated actual average value of the ratio is within a first predetermined range (step S28). If the calculated actual average value of the ratio is within the first predetermined range (step S28: Yes), the determination unit 745 determines that the compressor 11 is normal (step S29), and ends the processing operation shown in FIG.

[0081] Furthermore, if the calculated actual average value of the ratio is outside the first predetermined range (step S28: No), the determination unit 745 determines whether the calculated actual average value of the ratio is within the second predetermined range (step S30). If the calculated actual average value of the ratio is within the second predetermined range (step S30: Yes), the determination unit 745 determines that the compressor 11 is degraded (step S31). If the determination unit 745 determines that the compressor 11 is degraded, the protection control unit 746 executes protection control of the compressor 11 according to the ratio calculated in step S29 (step S32) and terminates the processing operation shown in FIG. 8. In this way, when it is determined that the compressor 11 is degraded, control is performed to reduce the load on the compressor 11, thereby slowing the rate of deterioration of the compressor 11 and lengthening the time until failure. Lengthening the time until failure of the compressor 11 allows for the arrangement and procurement of a replacement compressor 11 before failure and allows for advance scheduling of compressor replacement work. Therefore, compared with the case where a replacement compressor 11 is arranged and a replacement work schedule is drawn up after the compressor 11 breaks down, the compressor 11 can be replaced quickly and before the compressor 11 breaks down and stops, so the user is not inconvenienced by being unable to operate the air conditioning during the period from the compressor 11 breaking down to the replacement.

[0082] Furthermore, if the calculated measured average value of the ratio is outside the second predetermined range (step S30: No), the determination unit 745 determines that the compressor 11 is faulty (step S33). If the protection control unit 746 determines that the compressor 11 is faulty, it stops the operation of the compressor 11 (step S34), outputs a notification of the compressor 11 fault (step S35), and terminates the processing operation shown in FIG. 8. If the calculated measured average value of the ratio is outside the second predetermined range, the compressor 11 is still actually operating, so the measured average value of the calculated ratio can be calculated. However, the ratio value is a value that would not be possible if the compressor 11 were normal (as mentioned above, ratio values ​​outside the second predetermined range account for 5% of the total, which is a rare value), and it can be inferred that the compressor 11 has suffered significant damage. If the compressor 11 continues to operate under such conditions, for example, the shaft of the motor housed inside the compressor 11 may be damaged, causing tiny powdery fragments (iron powder) from the chipped shaft to be ejected into the refrigerant circuit 6 along with the refrigerant, circulating within the refrigerant circuit 6 and potentially damaging the devices and components that make up the refrigerant circuit 6, such as becoming lodged in the expansion valve. In the present invention, by stopping the compressor 11 and reporting a fault when the ratio is outside the second predetermined range, replacement of the compressor 11 can be promoted before damage to the devices and components that make up the refrigerant circuit 6 occurs. The notification output is an alarm output that reports a compressor 11 fault to the display of a remote control used to operate the air conditioner 1 or the display of a PC or mobile device used by the administrator managing the air conditioner 1. As a result, the air conditioner 1 can determine the condition of the compressor 11 even during normal operation without requiring any special operation or device.

[0083] If the determination unit 745 has not calculated the ratio for the predetermined time period (step S26: No), the process returns to step S21 to determine whether or not the current time is the acquisition timing. If the control unit 74 has not calculated the ratio for the predetermined time period (step S21: No), the process returns to step S21 to determine whether or not the current time is the acquisition timing.

[0084] <Effects of Example 1> In the air conditioner 1 of Example 1, the state of the compressor 11 can be determined using only feature quantities that can be detected in a normal air conditioner, so no special operation is required for the determination, and no special device is required. Moreover, the air conditioner 1 always calculates the ratio between the actual measured current value of the compressor 11 and the estimated current value, and uses the calculated ratio to determine the state of the compressor 11, so the state of the compressor 11 can be accurately and consistently determined. As a result, the state of the compressor 11 can be accurately and consistently determined simply and at low cost.

[0085] In the air conditioner 1, if the ratio is a value within a first predetermined range, the compressor 11 is determined to be normal, and if the ratio is a value outside the first predetermined range and is a value within a second predetermined range, the compressor 11 is determined to be deteriorated. As a result, the normality or deterioration state of the compressor 11 can be accurately and consistently determined. In other words, by predicting the deterioration of the compressor 11, it is possible to reduce the load on the compressor 11 before the compressor 11 breaks down, thereby lengthening the time until the compressor 11 breaks down.

[0086] In the air conditioner 1, if the ratio is outside the second predetermined range, it is determined that the compressor 11 is malfunctioning. As a result, it is possible to accurately and consistently determine whether the compressor 11 is malfunctioning.

[0087] When the air conditioner 1 determines that the compressor 11 is degraded, it controls the compressor 11 to reduce the load on the compressor 11 according to the ratio. As a result, the time until the compressor 11 breaks down can be extended, and work related to replacing the compressor 11 can be used during that time to replace the compressor 11 before it breaks down, so the user is not inconvenienced by being unable to operate the air conditioning system between the compressor 11 breaking down and being replaced.

[0088] Conventional air conditioners have detected a malfunction of the compressor 11 by detecting an abnormal current with a single threshold value (for example, 50 A or more). For example, under high load (high rotation) operation, a malfunction of the compressor 11 can be detected and stopped using the single threshold value, but under low load (low rotation) operation, a malfunction of the compressor 11 cannot be detected and stopped for the following reasons.

[0089] For example, even if a load of approximately 20 A under normal conditions drops to 30 A due to deterioration of the compressor 11, this is below the threshold and therefore a malfunction of the compressor 11 cannot be detected. Furthermore, if the compressor 11 continues to operate with a measured current of 30 A, for example, the motor shaft stored inside the compressor 11 may be damaged, causing tiny powdery fragments (iron powder) from a chipped portion of the shaft to be ejected into the refrigerant circuit 6 along with the refrigerant and circulate within the refrigerant circuit 6, potentially damaging the devices and components that make up the refrigerant circuit 6, such as by becoming lodged in the valve portion of the expansion valve. In this case, replacing the compressor 11 does not remove the powdery debris remaining in the refrigerant circuit 6, and damage caused by this powdery debris may occur. The air conditioner 1 of the first embodiment can detect abnormalities in the compressor 11 early, allowing the compressor 11 to be replaced before a large amount of the powdery debris enters the refrigerant circuit 6, thereby preventing damage to the devices and components that make up the refrigerant circuit 6.

[0090] In the air conditioning system 100 of Example 1, a case has been exemplified in which a learning model 73A learned by the learning unit 112 in the server device 110 is generated, and the generated learning model 73A is provided to the centralized controller 7. However, a generation unit that generates a learning model may be provided in the centralized controller 7, and an embodiment of this will be described below as Example 2.

[0091] Fig. 9 is a block diagram showing an example of an air conditioning system 100A of Example 2. Note that the same components of the air conditioning system 100 of Example 1 and the air conditioning system 100A of Example 2 are assigned the same reference numerals, and descriptions of the overlapping components and operations will be omitted. The air conditioning system 100A shown in Fig. 9 does not include a server device 110 and is composed of an air conditioner 1 including an outdoor unit 2, an indoor unit 3, and a central controller 7A. The air conditioner 1 shown in Fig. 9 is an air conditioner that cannot communicate with the server device 110.

[0092] Fig. 10 is a block diagram illustrating an example of a centralized controller 7A according to Example 2. The control unit 74 in the centralized controller 7A illustrated in Fig. 10 includes a detection unit 741, an acquisition unit 742, a calculation unit 743, an estimation unit 744, a determination unit 745, and a protection control unit 746, as well as a generation unit 747 that generates a learning model 73A that determines a state quantity (coefficient) k related to the efficiency of the compressor 11.

[0093] The generation unit 747 periodically, for example, every five minutes, collects the rotation speed, discharge pressure value, and suction pressure value of the compressor 11 as feature quantities and stores the collected feature quantities. The generation unit 747 uses the accumulated rotation speed, discharge pressure value, and suction pressure value of the compressor 11 for a certain period, for example, 30 days, to generate a learning model 73A that calculates a state quantity (coefficient) k related to the efficiency of the compressor 11. The generation unit 747 stores the generated learning model 73A in the storage unit 73.

[0094] <Effects of Example 2> Even in an air conditioner 1 that cannot communicate with the server device 110 of Example 2, the state of the compressor 11 can be determined using only feature quantities that can be detected in a normal air conditioner, so no special operation is required for the determination, and no special device is required. Moreover, the air conditioner 1 always calculates the ratio between the actual measured current value of the compressor 11 and the estimated current value, and uses the calculated ratio to determine the state of the compressor 11, so the state of the compressor 11 can be accurately and consistently determined. As a result, the state of the compressor 11 can be accurately and consistently determined simply and at low cost.

[0095] In the first and second embodiments described above, the discharge specific enthalpy is calculated using the discharge pressure and the discharge temperature. However, there may be cases where the discharge temperature cannot be detected in an air conditioner. Therefore, the following describes a case where the discharge enthalpy can be calculated from the discharge pressure even if the discharge temperature cannot be detected.

[0096] As described above, in each embodiment, the air conditioner 1 is equipped with a discharge temperature sensor 32. However, the air conditioner 1 may not be equipped with a discharge temperature sensor 32. In this case, even if the air conditioner 1 does not have a discharge temperature sensor 32, the discharge specific enthalpy can be calculated using the discharge pressure value as long as the discharge superheat degree is known, as shown in FIG. 3 . However, in this case, the discharge superheat degree is an estimated value obtained by conducting a test or the like in advance. If the state of the refrigerant circuit 6 differs between the test and actual operation, the accuracy of the discharge specific enthalpy calculation may be reduced compared to when the actual discharge temperature is used. Therefore, it goes without saying that the accuracy of the discharge specific enthalpy calculation is improved when the discharge temperature sensor 32 is installed and the discharge superheat degree is calculated using the actual discharge temperature.

[0097] In addition, in the first and second embodiments, the suction specific enthalpy is calculated using the suction pressure and the suction temperature. However, there may be cases where the suction temperature cannot be detected in an air conditioner. Therefore, the following describes a case where the suction enthalpy can be calculated from the suction pressure even if the suction temperature cannot be detected.

[0098] As described above, in each embodiment, the air conditioner 1 is equipped with an intake temperature sensor 34. However, there are cases where the intake temperature sensor 34 is not equipped. In this case, even if the air conditioner 1 is not equipped with an intake temperature sensor 34, as shown in FIG. 3, the intake specific enthalpy can be calculated using the intake pressure value as long as the intake superheat degree is known. However, in this case, the intake superheat degree is an estimated value obtained by performing a test or the like in advance. If the state of the refrigerant circuit 6 differs between the test and actual operation, the accuracy of the intake specific enthalpy calculation may be reduced compared to when the actually measured intake temperature is used. Therefore, it goes without saying that the accuracy of the intake specific enthalpy calculation is improved when the intake temperature sensor 34 is installed and the actually measured intake temperature is used to calculate the intake superheat degree.

[0099] It is also possible that an air conditioner may be equipped with only one of the discharge temperature sensor 32 and the intake temperature sensor 34, but the estimated value of the discharge superheat degree is more likely to vary compared to the intake superheat degree. Therefore, if only one of these is to be obtained, it is preferable to install only the discharge temperature sensor 32 and obtain the discharge temperature.

[0100] In addition, the control unit 74 in the centralized controller 7A in the air conditioner 1 of Example 2 has been exemplified as performing the function of the state determination device. However, it does not have to be the control unit 74 in the centralized controller 7A, and the function of the state determination device may be performed, for example, in the first control unit 19 in the outdoor unit 2 or the second control unit 50 in the indoor unit 3, and this can be changed as appropriate.

[0101] Software that executes the functions of the status determination device may be installed in the centralized controller 7A of the air conditioner 1, and the functions of the status determination device may be executed by the control unit 74 in the centralized controller 7A, and this can be modified as appropriate. In this case, by installing software that executes the functions of the status determination device in an existing air conditioner 1 that does not have a learning unit, the mechanism required to determine the status of the compressor 11 can be easily installed even in an existing air conditioner 1.

[0102] The air conditioners 1 of Examples 1 and 2 illustrate a state determination device that determines the state of the compressor 11 when N indoor units 3 are connected to one outdoor unit 2. However, the state of the compressor 11 can also be determined in the same manner as in Examples 1 and 2 for an air conditioner 1 in which one outdoor unit 2 and one indoor unit 3 are connected.

[0103] In this embodiment, the learning model 73A is generated by the server device 110 or the centralized controller 7A, but the user may calculate the learning model 73A from simulation results. Furthermore, in this embodiment, the learning model 73A is generated using multiple regression analysis, but the learning model 73A may be generated using machine learning techniques such as SVR (Support Vector Regression) and NN (Neural Network), which are capable of performing general regression analysis. In this case, when selecting features, instead of the P value or corrected R2 used in multiple regression analysis, a general technique for selecting features to improve the accuracy of the learning model 73A (such as forward feature selection or backward feature elimination) may be used.

[0104] In the air conditioning system 100 of Example 1, a state quantity related to the efficiency of the compressor 11 is calculated using the learning model 73A acquired from the server device 110, and an estimated current value is estimated using the calculated state quantity related to the efficiency of the compressor 11. However, this is not limited to this and can be modified as appropriate. For example, the server device 110 may have a storage unit (not shown) that pre-stores drive current values ​​for each model of compressor 11 in association with feature quantities related to the operation of the compressor 11. Examples of feature quantities related to the operation of the compressor 11 include the rotation speed, discharge pressure, discharge temperature, suction pressure, and suction temperature of the compressor 11.

[0105] The estimation unit 744 in the centralized controller 7 transmits the detected feature quantities related to the operation of the compressor 11 to the server device 110 along with model information related to the model, such as the model name of the installed compressor 11. The server device 110 references the received model information, accesses data corresponding to the model of the compressor 11 from data stored in a storage unit that associates model-specific feature quantities with drive current values, reads out the drive current value corresponding to the received feature quantities related to the operation of the compressor 11 from the storage unit, and transmits the read drive current value to the estimation unit 744 in the centralized controller 7 as an estimated current value.

[0106] When the server device 110 reads out from the storage unit the drive current values ​​corresponding to the received feature quantities related to the drive of the compressor 11, it may read out all drive current values ​​corresponding to values ​​of the feature quantities related to the drive of the compressor 11 that are within a predetermined range, calculate the average value of all the read drive current values, and transmit this average value as an estimated current value to the estimation unit 744 in the centralized controller 7. Here, the predetermined range is a range centered on the received feature quantity related to the drive of the compressor 11, for example, a range of ±5% from the center value.

[0107] The operating state of the compressor 11 when the feature value is within a range of ±5% from the center value is considered to be close to the operating state of the compressor 11 when the received feature value is used. If a disturbance occurs when the estimator 744 in the centralized controller 7 acquires a feature value related to the operation of the compressor 11, for example, if the air conditioning load temporarily increases, there is a possibility of an erroneous determination if the state of the compressor 11 is determined using only the drive current value corresponding to the feature value as the estimated current value. In contrast, if the average value of the drive current values ​​when the feature value is within a range of ±5% from the center value is used as the estimated current value, the influence of the above-mentioned disturbance can be reduced and erroneous determination when determining the state of the compressor 11 can be prevented. The feature value related to the operation of the compressor 11 is within

[0108] As described above, by acquiring the estimated current value from the server device 110, the estimating unit 744 does not need to calculate the estimated current value, and therefore the processing load can be reduced.

[0109] The same components as those in the air conditioner of the first embodiment are denoted by the same reference numerals, and redundant explanations of the components and operations will be omitted.

[0110] <Configuration of Centralized Controller> FIG. 11 is a block diagram showing an example of a centralized controller 7 according to a third embodiment. The centralized controller 7 shown in FIG. 11 includes a first communication unit 71, a second communication unit 72, a storage unit 73, and a control unit 74. The first communication unit 71 is an interface such as a UART that communicates with the server device 110 via the communication network 120. The second communication unit 72 is an interface such as a UART that communicates with the first control unit 19 in the outdoor unit 2 or the second control unit 50 in the indoor unit 3 via the communication network 8. The storage unit 73 is, for example, a flash memory, and stores various types of information, programs, and the like. The storage unit 73 stores a learning model 73A generated by the server device 110.

[0111] The control unit 74 functions as a failure estimation device that estimates a failure of the compressor 11. The control unit 74 has a detection unit 741, an acquisition unit 742, a calculation unit 743, an estimation unit 744, and a determination unit 745. The detection unit 741 acquires, from the first control unit 19 of the outdoor unit 2, for example, a measured current value, which is the current value flowing through the primary side of the inverter circuit 11A when the compressor 11 is driven. Furthermore, the acquisition unit 742 acquires, in addition to the above-mentioned measured current value, characteristic quantities related to the drive of the compressor 11, which will be described below. The characteristic quantities related to the drive of the compressor 11 include the rotation speed of the compressor 11, state quantities of the refrigerant related to the drive of the compressor 11, and state quantities (coefficients) related to the efficiency of the compressor 11. The state quantities of the refrigerant related to the operation of the compressor 11 are the sensor values ​​of the discharge pressure sensor 31, discharge temperature sensor 32, suction pressure sensor 33, suction temperature sensor 34, refrigerant temperature sensor 35, and outside air temperature sensor 36 in the outdoor unit 2 from the first control unit 19. Furthermore, the acquisition unit 742 acquires the sensor values ​​of the liquid side refrigerant temperature sensor 61, gas side refrigerant temperature sensor 62, and suction temperature sensor 63 from the second control unit 50 of each indoor unit 3 as the state quantities of the refrigerant related to the operation of the compressor 11.

[0112] The acquisition unit 742 also acquires a state quantity (coefficient) related to the efficiency of the compressor 11 from the learning model 73A. The learning model 73A estimates the state quantity (coefficient) related to the efficiency of the compressor 11 using the rotation speed of the compressor 11 and a state quantity of the refrigerant related to the operation of the compressor 11. The estimation of the state quantity (coefficient) related to the efficiency of the compressor 11 will be described in detail later.

[0113] That is, the acquisition unit 742 periodically acquires the measured current value, the rotation speed of the compressor 11, the state quantities of the refrigerant related to the operation of the compressor 11, and the state quantities (coefficients) related to the efficiency of the compressor 11.

[0114] The calculation unit 743 calculates a refrigerant suction density, which is the density of the refrigerant sucked into the compressor 11, using the acquired state quantities of the refrigerant related to the operation of the compressor 11. Specifically, the calculation unit 743 refers to a table that associates the refrigerant suction density with the suction pressure value and suction temperature of the compressor 11, and acquires the refrigerant suction density corresponding to the suction pressure value and the suction temperature. Note that the table is obtained in advance by performing tests or the like, and is stored in the storage unit 73 of the centralized controller 7 (not shown in FIG. 11 ).

[0115] The calculation unit 743 calculates a specific enthalpy difference, which is the difference between the discharge specific enthalpy, which is the specific enthalpy of the refrigerant discharged from the compressor 11, and the suction specific enthalpy, which is the specific enthalpy of the refrigerant drawn into the compressor 11. Specifically, the calculation unit 743 calculates the specific enthalpy difference by subtracting the suction specific enthalpy from the discharge specific enthalpy. The calculation unit 743 calculates the discharge specific enthalpy based on the Mollier diagram shown in FIG. 3, the discharge pressure value, and the discharge temperature. Furthermore, the calculation unit 743 calculates the suction specific enthalpy based on the Mollier diagram shown in FIG. 3, the suction pressure value, and the suction temperature. The Mollier diagram shown in FIG. 3 is calculated in advance through testing or the like and is stored in the memory unit 73 of the centralized controller 7 (not shown in FIG. 4). The memory unit 73 stores, for example, the discharge pressure value, the discharge temperature, the suction pressure value, the suction temperature, the suction specific enthalpy, the discharge specific enthalpy, and the like in the Mollier diagram shown in FIG. 3.

[0116] The estimation unit 744 uses the feature quantity to estimate an estimated current value as a drive current value corresponding to the operating load of the compressor 11. The estimation unit 744 estimates the estimated current value using the refrigerant suction density (represented as p in the following formula), the rotation speed of the compressor 11, the specific enthalpy difference, and a state quantity (coefficient, represented as k in the following formula) related to the efficiency of the compressor 11.

[0117] Formula: Estimated current value = k × p × (rotational speed of compressor 11) × specific enthalpy difference

[0118] The estimation unit 744 estimates the estimated current value each time the acquisition unit 742 acquires the rotation speed of the compressor 11 and the state quantity of the refrigerant related to the operation of the compressor 11 (for example, every 5 minutes).

[0119] The determination unit 745 determines whether the compressor 11 is faulty based on the comparison result between the estimated current value estimated by the estimation unit 744 and the actual measured current value detected by the detection unit 741. The determination unit 745 determines that the compressor 11 is not faulty if the ratio between the estimated current value and the actual measured current value, for example, (actual measured current value divided by estimated current value), is within a predetermined range. The ratio is the ratio between the estimated current value and the actual measured current value. The determination unit 745 also determines that the compressor 11 is faulty if the ratio between the estimated current value and the actual measured current value is outside the predetermined range.

[0120] FIG. 12 is an explanatory diagram showing an example of a normal distribution of ratios. A compressor 11 failure is determined based on whether the ratio falls within a predetermined range in the normal distribution. Here, the predetermined range is, for example, μ±2σ, as shown in FIG. 12 , where σ is the standard deviation of the occurrence of the statistically calculated ratios and μ is the mean value of the occurrence of the ratios. In the predetermined range μ±2σ, the ratio falls within the range of 0.6 to 1.4. Approximately 95% of the calculated ratio falls within the μ±2σ range. Meanwhile, the remaining approximately 5% of the calculated ratio falls outside the μ±2σ range. A situation in which the calculated ratio falls outside the predetermined range is unlikely to occur when the compressor 11 is operating normally, and it can be assumed that an abnormality has occurred in the compressor 11, i.e., that the compressor 11 is malfunctioning.

[0121] In this embodiment, when the calculated ratio frequently falls outside the predetermined range and the actual measurement average value μ falls outside the range of ±2σ, the compressor 11 is determined to be faulty. The smaller the value of the predetermined range, the higher the sensitivity for determining that the compressor 11 is faulty, but the greater the possibility of misjudging a compressor 11 that is not faulty. Therefore, in this embodiment, the range of ±2σ is used, which provides a good balance between the sensitivity and accuracy of fault determination. However, if higher sensitivity is desired, the predetermined range can be set to ±1σ, and if higher accuracy is desired, the predetermined range can be set to ±3σ.

[0122] The determination unit 745 sequentially calculates the ratio between the estimated current value and the actually measured current value (each time the rotation speed of the compressor 11 or a state quantity of the refrigerant related to the operation of the compressor 11 is acquired), calculates the average value of the ratio for a predetermined time period, for example, for one hour, and determines whether the calculated average value of the ratio is within a predetermined range. If the average value of the ratio is within the predetermined range, the determination unit 745 determines that the compressor 11 is not malfunctioning, i.e., is normal. Furthermore, if the average value of the ratio is outside the predetermined range, the determination unit 745 determines that the compressor 11 is malfunctioning.

[0123] <Configuration of Server Device> The server device 110 includes a storage unit 111 and a learning unit 112. The storage unit 111 periodically collects, for example, the rotation speed, discharge pressure, and suction pressure of the compressor 11 as feature quantities from the centralized controller 7, for example, every five minutes, and stores the collected feature quantities. The learning unit 112 generates a learning model 73A for calculating a state quantity (coefficient) k related to the efficiency of the compressor 11 using feature quantities stored in the storage unit 111 over a certain period of time, for example, 30 days. Thereafter, the learning model 73A is updated every 30 days using the feature quantities acquired up to that point. The learning unit 112 provides the generated or updated learning model 73A to the centralized controller 7. The centralized controller 7 stores the learning model 73A received from the server device 110 in the memory unit 73.

[0124] <Learning Model> As described above, the state quantity (coefficient) k related to the efficiency of the compressor 11 can be calculated using the learning model 73A. The state quantity (coefficient) k related to the efficiency of the compressor 11 will now be described. The state quantity (coefficient) k related to the efficiency of the compressor 11 can be calculated, for example, using the volumetric efficiency, compression efficiency, and swept volume (displacement volume) of the compressor 11. However, the volumetric efficiency and compression efficiency of the compressor 11 vary depending on the rotation speed of the compressor 11 and the pressure ratio (ratio of high pressure to low pressure) in the refrigerant circuit 6. Furthermore, for example, the delivery note and specifications for the compressor 11 only list values ​​under limited operating conditions. Since there is no data for all expected conditions, it is difficult to store the values ​​in advance, for example, as a table in a memory unit. Furthermore, while the swept volume of the compressor 11 can be obtained for compressors manufactured in-house, it is often difficult to obtain the value when purchasing a compressor from another company because the delivery note and specifications often do not include the value. Furthermore, in the case of an existing air conditioner, there may be cases where the information and specifications of the installed compressor are unknown. Therefore, it is difficult to determine the state quantity (coefficient) k related to the efficiency of the compressor 11 from the volumetric efficiency, compression efficiency, and stroke volume of the compressor 11. Therefore, in this embodiment, the state quantity (coefficient) k related to the efficiency of the compressor 11 is determined using the learning model 73A based on the rotation speed, discharge pressure value, and suction pressure value of the compressor 11.

[0125] The learning model 73A is generated by multiple regression analysis, a type of regression analysis, using the rotation speed, discharge pressure, and suction pressure of the compressor 11 among multiple feature quantities. In multiple regression analysis, the regression equation obtained from multiple simulation results is selected to have the smallest P value (a value (predetermined weight parameter) indicating the degree of influence that the operating state quantity has on the accuracy of the generated learning model) and the largest possible corrected R2 (a value indicating the accuracy of the generated learning model) between 0.9 and 1.0. Here, the P value and corrected R2 are values ​​related to the accuracy of the learning model when generating the learning model using multiple regression analysis. The smaller the P value and the closer the corrected R2 is to 1.0, the higher the accuracy of the generated learning model.

[0126] The learning model 73A is a regression equation for calculating a state quantity (coefficient) k related to the efficiency of the compressor 11. The regression equation is, for example, (α1 × rotation speed of the compressor 11) + (α2 × discharge pressure value) + (α3 × suction pressure value) + α4. The coefficients α1 to α4 are determined when calculating the state quantity (coefficient) k related to the efficiency of the compressor 11. The control unit 74 calculates the state quantity (coefficient) k related to the efficiency of the compressor 11 at the current time by substituting the current rotation speed, discharge pressure value, and suction pressure value of the compressor 11 acquired by the acquisition unit 742 into the regression equation.

[0127] <Learning Process> Here, the learning process performed when the learning unit 112 generates the learning model 73A will be described. Fig. 13 is a flowchart showing an example of the processing operation of the learning unit 112 related to the learning process of Example 3. In Fig. 13, the learning unit 112 in the server device 110 periodically (e.g., every 5 minutes) acquires the rotation speed, discharge pressure value, and suction pressure value of the compressor 11 from the centralized controller 7 (step S11A). Then, the learning unit 112 sequentially accumulates the acquired rotation speed, discharge pressure value, and suction pressure value of the compressor 11 in the accumulation unit 111.

[0128] The learning unit 112 generates a learning model 73A using the rotation speed, discharge pressure value, and suction pressure value of the compressor 11 for a certain period, for example, 30 days, acquired in step S11A (step S12A). The learning unit 112 transmits the generated learning model 73A to the centralized controller 7 via the communication network 120 (step S13A), thereby completing the processing operation shown in Fig. 13. The centralized controller 7 then stores the learning model 73A received via the communication network 120 in the memory unit 73.

[0129] <Fault Determination Process> Next, a fault determination process that allows the centralized controller 7 to constantly determine whether the compressor 11 has a fault will be described. FIG. 14 is a flowchart illustrating an example of the processing operation of the centralized controller 7 related to the fault determination process of the third embodiment. In FIG. 14, the control unit 74 in the centralized controller 7 determines whether the current time is the acquisition timing (step S21A). The acquisition timing is every five minutes as described above. If the current time is the acquisition timing (step S21A: Yes), the acquisition unit 742 in the control unit 74 acquires the rotation speed, discharge pressure value, suction pressure value, and actual measured current value of the compressor 11 (step S22A). The discharge pressure value is detected by the discharge pressure sensor 31, and the suction pressure value is detected by the suction pressure sensor 33. The actual measured current value is detected by the detection unit 741 from the inverter circuit 11A of the compressor 11.

[0130] The calculation unit 743 in the control unit 74 uses the learning model 73A to calculate the state quantity (coefficient) k related to the efficiency of the compressor 11 (step S23A). The learning model 73A calculates the state quantity (coefficient) k related to the efficiency of the compressor 11 by substituting the rotation speed, discharge pressure value, and suction pressure value of the compressor 11 into a regression equation.

[0131] The estimator 744 in the controller 74 estimates the estimated current value of the compressor 11 using the calculated compressor efficiency-related state variable (coefficient) k, the refrigerant suction density of the compressor 11, the compressor 11 rotation speed, the discharge specific enthalpy, and the suction specific enthalpy (step S24A). Although not shown, the memory 73 pre-stores a table correlating the refrigerant suction density with the suction pressure value and suction temperature of the compressor 11. The calculator 743 references this table to obtain the refrigerant suction density corresponding to the suction pressure value and suction temperature. The calculator 743 calculates the discharge specific enthalpy based on the Mollier diagram shown in FIG. 3, the discharge pressure value, and the discharge temperature. Furthermore, the calculator 743 calculates the suction specific enthalpy based on the Mollier diagram shown in FIG. 3, the suction pressure value, and the suction temperature. Then, the estimation unit 744 estimates the estimated current value of the compressor 11 using (state quantity (coefficient) k related to efficiency) x (refrigerant suction density p) x (rotation speed of the compressor 11) x (discharge specific enthalpy - suction specific enthalpy).

[0132] The determination unit 745 in the control unit 74 divides the actual current value by the estimated current value to calculate the current ratio (step S25A). The determination unit 745 determines whether the ratio for a predetermined time period has been calculated (step S26A). The predetermined time period is, for example, one hour. If the determination unit 745 has not calculated the ratio for the predetermined time period (step S26A: No), the process returns to step S21A. If the determination unit 745 has calculated the ratio for the predetermined time period (step S26A: Yes), the determination unit 745 calculates an average ratio from the ratio for the predetermined time period (step S27A).

[0133] The determination unit 745 then determines whether the calculated average value of the ratio is outside a predetermined range (step S29A). If the calculated average value of the ratio is outside the predetermined range (step S29A: Yes), the determination unit 745 determines that the compressor 11 has failed (step S30A), outputs a notification of the compressor 11 failure (step S31A), and terminates the processing operation shown in FIG. 14. The notification output is an alarm output that notifies the user of the compressor 11 failure on the display of a remote control that operates the air conditioner 1 or on the display of a PC or mobile device of the administrator who manages the air conditioner 1. As a result, the air conditioner 1 can determine the state of the compressor 11 and estimate a failure even during normal operation without requiring any special operation or device.

[0134] If the calculated average value of the ratio is within a predetermined range (step S29A: No), the determination unit 745 determines that the compressor 11 is normal (step S32A) and ends the processing operation shown in Fig. 14. Furthermore, if the current time is not the acquisition timing (step S21A: No), the control unit 74 returns to the processing of step S21A to determine whether the current time is the acquisition timing.

[0135] <Effects of Example 3> In the air conditioner 1 of Example 3, a failure of the compressor 11 can be determined using only feature quantities that can be detected in a normal air conditioner, so no special operation is required for the determination, and no special device is required. Moreover, the air conditioner 1 always determines whether or not there is a failure of the compressor 11 using the actual measured current value and estimated current value of the compressor 11, so it can always accurately determine whether there is a failure of the compressor 11. As a result, it is possible to always accurately determine whether there is a failure of the compressor 11, simply and at low cost.

[0136] Conventional air conditioners have detected a malfunction of the compressor 11 by detecting an abnormal current with a single threshold value (for example, 50 A or more). For example, under high load (high rotation) operation, a malfunction of the compressor 11 can be detected and stopped using the single threshold value, but under low load (low rotation) operation, a malfunction of the compressor 11 cannot be detected and stopped for the following reasons.

[0137] For example, even if a load of approximately 20 A under normal conditions drops to 30 A due to deterioration of the compressor 11, the measured current is below the threshold and therefore cannot be detected as a compressor 11 failure. Furthermore, if the compressor 11 continues to operate with the measured current at 30 A, for example, the motor shaft stored inside the compressor 11 may be damaged, causing tiny powdery fragments (iron powder) from a chipped portion of the shaft to be ejected into the refrigerant circuit 6 along with the refrigerant and circulate within the refrigerant circuit 6, potentially damaging the devices and components that make up the refrigerant circuit 6, such as by becoming lodged in the valve portion of the expansion valve. In this case, replacing the compressor 11 does not remove the powdery debris remaining in the refrigerant circuit 6, and damage caused by this powdery debris may occur. The air conditioner 1 of Example 3 allows for early detection of an abnormality in the compressor 11, allowing the compressor 11 to be replaced before a large amount of the powdery debris enters the refrigerant circuit 6, thereby preventing damage to the devices and components that make up the refrigerant circuit 6.

[0138] In the air conditioning system 100 of Example 3, a learning model 73A is generated by learning in the learning unit 112 in the server device 110, and the generated learning model 73A is provided to the centralized controller 7. However, a generation unit that generates a learning model may be provided in the centralized controller 7, and such an embodiment will be described below as Example 4.

[0139] Note that the same components of the air conditioning system 100 of Example 3 and the air conditioning system 100A of Example 4 are denoted by the same reference numerals, and descriptions of the overlapping components and operations will be omitted. The air conditioning system 100A shown in Fig. 9 does not include a server device 110, and is composed of an air conditioner 1 including an outdoor unit 2, an indoor unit 3, and a central controller 7A. The air conditioner 1 shown in Fig. 9 is an air conditioner that cannot communicate with the server device 110.

[0140] Fig. 15 is a block diagram illustrating an example of a centralized controller 7A according to Example 4. The control unit 74 in the centralized controller 7A illustrated in Fig. 15 includes a detection unit 741, an acquisition unit 742, a calculation unit 743, an estimation unit 744, and a determination unit 745, as well as a generation unit 746A that generates a learning model 73A that determines a state quantity (coefficient) k related to the efficiency of the compressor 11.

[0141] The generation unit 746A periodically, for example, every five minutes, collects the rotation speed, discharge pressure value, and suction pressure value of the compressor 11 as feature quantities and accumulates the collected feature quantities. The generation unit 746A uses the accumulated rotation speed, discharge pressure value, and suction pressure value of the compressor 11 for a certain period, for example, 30 days, to generate a learning model 73A that calculates a state quantity (coefficient) k related to the efficiency of the compressor 11. The generation unit 746A stores the generated learning model 73A in the storage unit 73.

[0142] <Effects of Example 4> Even in an air conditioner 1 that cannot communicate with the server device 110 of Example 4, a malfunction of the compressor 11 can be determined using only feature quantities that can be detected by a normal air conditioner, so no special operation is required for the determination, and no special device is required. Moreover, the air conditioner 1 always determines whether or not there is a malfunction of the compressor 11 using the actual measured current value and estimated current value of the compressor 11, so that a malfunction of the compressor 11 can always be accurately determined. As a result, a malfunction of the compressor 11 can always be accurately determined simply and inexpensively.

[0143] In the third and fourth embodiments described above, the discharge specific enthalpy is calculated using the discharge pressure and the discharge temperature. However, there may be cases where the discharge temperature cannot be detected in an air conditioner. Therefore, the following describes a case where the discharge enthalpy can be calculated from the discharge pressure even if the discharge temperature cannot be detected.

[0144] As described above, in each embodiment, the air conditioner 1 is equipped with a discharge temperature sensor 32. However, the air conditioner 1 may not be equipped with a discharge temperature sensor 32. In this case, even if the air conditioner 1 does not have a discharge temperature sensor 32, the discharge specific enthalpy can be calculated using the discharge pressure value as long as the discharge superheat degree is known, as shown in FIG. 3 . However, in this case, the discharge superheat degree is an estimated value obtained by conducting a test or the like in advance. If the state of the refrigerant circuit 6 differs between the test and actual operation, the accuracy of the discharge specific enthalpy calculation may be reduced compared to when the actual discharge temperature is used. Therefore, it goes without saying that the accuracy of the discharge specific enthalpy calculation is improved when the discharge temperature sensor 32 is installed and the discharge superheat degree is calculated using the actual discharge temperature.

[0145] In addition, in Examples 3 and 4, the suction specific enthalpy is calculated using the suction pressure and the suction temperature. However, there may be cases where the suction temperature cannot be detected in an air conditioner. Therefore, the following describes a case where the suction enthalpy can be calculated from the suction pressure even if the suction temperature cannot be detected.

[0146] As described above, in each embodiment, the air conditioner 1 is equipped with an intake temperature sensor 34. However, there are cases where the intake temperature sensor 34 is not equipped. In this case, even if the air conditioner 1 is not equipped with an intake temperature sensor 34, as shown in FIG. 3, the intake specific enthalpy can be calculated using the intake pressure value as long as the intake superheat degree is known. However, in this case, the intake superheat degree is an estimated value obtained by performing a test or the like in advance. If the state of the refrigerant circuit 6 differs between the test and actual operation, the accuracy of the intake specific enthalpy calculation may be reduced compared to when the actually measured intake temperature is used. Therefore, it goes without saying that the accuracy of the intake specific enthalpy calculation is improved when the intake temperature sensor 34 is installed and the actually measured intake temperature is used to calculate the intake superheat degree.

[0147] It is also possible that an air conditioner may be equipped with only one of the discharge temperature sensor 32 and the intake temperature sensor 34, but the estimated value of the discharge superheat degree is more likely to vary compared to the intake superheat degree. Therefore, if only one of these is to be obtained, it is preferable to install only the discharge temperature sensor 32 and obtain the discharge temperature.

[0148] In addition, the control unit 74 in the centralized controller 7A in the air conditioner 1 of Example 4 has been exemplified as performing the function of the failure estimation device. However, it does not have to be the control unit 74 in the centralized controller 7A, and the function of the failure estimation device may be performed, for example, in the first control unit 19 in the outdoor unit 2 or the second control unit 50 in the indoor unit 3, and this can be changed as appropriate.

[0149] Software that executes the functions of the failure estimation device may be installed in the centralized controller 7A of the air conditioner 1, and the functions of the failure estimation device may be executed by the control unit 74 in the centralized controller 7A, and this can be changed as appropriate. In this case, by installing software that executes the functions of the failure estimation device in an existing air conditioner 1 that does not have a learning unit, the mechanism required to determine deterioration of the compressor 11 can be easily installed even in an existing air conditioner 1.

[0150] The air conditioners 1 of Examples 3 and 4 illustrate examples of a failure estimation device that estimates a failure in the compressor 11 when N indoor units 3 are connected to one outdoor unit 2. However, even for an air conditioner 1 in which one outdoor unit 2 and one indoor unit 3 are connected, a failure in the compressor 11 can be estimated using the same method as in Examples 3 and 4.

[0151] In this embodiment, the learning model 73A is generated by the server device 110 or the centralized controller 7A, but the user may calculate the learning model 73A from simulation results. Furthermore, in this embodiment, the learning model 73A is generated using multiple regression analysis, but the learning model 73A may be generated using machine learning techniques such as SVR (Support Vector Regression) and NN (Neural Network), which are capable of performing general regression analysis. In this case, when selecting features, instead of the P value or corrected R2 used in multiple regression analysis, a general technique for selecting features to improve the accuracy of the learning model 73A (such as forward feature selection or backward feature elimination) may be used.

[0152] Furthermore, the components of each unit shown in the figure do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each unit is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0153] Furthermore, the various processing functions performed by each device may be executed in whole or in part on a CPU (Central Processing Unit) (or a microcomputer such as an MPU (Micro Processing Unit) or an MCU (Micro Controller Unit)). Needless to say, the various processing functions may be executed in whole or in part on a program analyzed and executed by the CPU (or a microcomputer such as an MPU or MCU), or on hardware using wired logic.

[0154] REFERENCE SIGNS LIST 1 Air conditioner 2 Outdoor unit 3 Indoor unit 7 Centralized controller 11 Compressor 13 Outdoor heat exchanger 14 Outdoor unit expansion valve 31 Discharge pressure sensor 32 Discharge temperature sensor 33 Suction pressure sensor 34 Suction temperature sensor 73A Learning model 74 Control unit 741 Detection unit 742 Acquisition unit 743 Calculation unit 744 Estimation unit 745 Determination unit 746 Protection control unit 747 Generation unit

Claims

1. A state determination device comprising: a current detection unit that detects an actual measured current value, which is an actual measurement value of the drive current of a compressor; an estimation unit that estimates an estimated current value, which is an estimated value of the drive current of the compressor; and a determination unit that calculates the ratio between the actual measured current value and the estimated current value, and determines the state of the compressor using the calculated ratio.

2. A condition determination device as described in claim 1, characterized in that a second predetermined range is set from a first upper limit value obtained by adding a predetermined value to the upper limit value of a first predetermined range to a first lower limit value obtained by subtracting the predetermined value from the lower limit value of the first predetermined range, and the determination unit determines that the compressor is normal if the ratio is a value within the first predetermined range, and determines that the compressor is deteriorated if the ratio is a value outside the first predetermined range and is a value within the second predetermined range.

3. The state determination device according to claim 2, characterized in that the determination unit determines that the compressor is faulty when the ratio is outside the second predetermined range.

4. The state determination device according to claim 2, further comprising a protection control unit that controls the compressor to reduce the load on the compressor in accordance with the ratio when the determination unit determines that the compressor is deteriorated.

5. A state determination device as described in any one of claims 1 to 4, characterized in that it has an acquisition unit that acquires feature quantities related to the operation of the compressor, and the estimation unit estimates the estimated current value using the feature quantities related to the operation of the compressor acquired by the acquisition unit.

6. The state determination device according to claim 5, characterized in that the characteristic quantities related to the operation of the compressor include the measured current value, the rotation speed of the compressor, a state quantity of the refrigerant related to the operation of the compressor, and a state quantity related to the efficiency of the compressor.

7. A state determination device as described in claim 6, characterized in that it has a calculation unit that uses state quantities of refrigerant related to the operation of the compressor to calculate an suction density, which is the density of the refrigerant sucked into the compressor, and a specific enthalpy difference, which is the difference between the discharge specific enthalpy and the suction specific enthalpy, and the estimation unit estimates the estimated current value using the suction density, the rotation speed of the compressor, the specific enthalpy difference, and state quantities related to the efficiency of the compressor.

8. A state determination device as described in claim 7, characterized in that it has a learning unit that generates a learning model for determining a state quantity related to the efficiency of the compressor using the rotation speed of the compressor and a state quantity of the refrigerant related to the operation of the compressor.

9. The state determination device according to claim 8, wherein the state quantities of the refrigerant related to the operation of the compressor include the discharge pressure and suction pressure of the compressor.

10. The state determination device according to claim 9, wherein the state quantity of the refrigerant related to the operation of the compressor further includes a discharge temperature of the compressor.

11. The state determination device according to claim 10, wherein the state quantity of the refrigerant related to the operation of the compressor further includes the intake temperature of the compressor.

12. The state determination device described in claim 11, characterized in that the acquisition unit periodically acquires the current value of the compressor, the rotation speed of the compressor, and the refrigerant state quantity related to the compressor, respectively; the estimation unit estimates the estimated current value of the compressor each time the acquisition unit acquires the rotation speed of the compressor and the refrigerant state quantity; and the determination unit calculates the ratio between the estimated current value of the compressor and the actual measured current value of the compressor, and determines the state of the compressor using an average value of the ratio calculated over a predetermined period of time.

13. A state determination method comprising the steps of: detecting an actual current value, which is an actual measurement value of the drive current of a compressor; estimating an estimated current value, which is an estimated value of the drive current of the compressor; calculating a ratio between the actual current value and the estimated current value; and determining the state of the compressor using the calculated ratio.

14. An air conditioning system comprising: an air conditioner having a compressor driven by an inverter circuit, an outdoor unit equipped with an outdoor heat exchanger and an expansion valve, and an indoor unit connected to the outdoor unit; and a server device capable of communicating with the air conditioner, wherein the air conditioner has: a current detection unit that detects an actual measured current value that is an actual measurement value of the drive current of the compressor; an estimation unit that estimates an estimated current value that is an estimate of the drive current of the compressor; and a determination unit that calculates the ratio between the actual measured current value and the estimated current value and determines the state of the compressor using the calculated ratio.

15. The air conditioning system described in claim 14, characterized in that the server device has a learning unit that generates a learning model for determining a state quantity related to the efficiency of the compressor using the rotation speed of the compressor and a state quantity of the refrigerant related to the operation of the compressor, and the estimation unit determines the state quantity related to the efficiency of the compressor using the learning model acquired from the server device, and estimates the estimated current value using the determined state quantity related to the efficiency of the compressor.

16. The air conditioning system described in claim 15, characterized in that the server device has a memory unit in which drive current values ​​are stored in correspondence with features related to the drive of each compressor for each different compressor model, and the estimation unit reads out from the memory unit the drive current value corresponding to the detected feature related to the drive of the compressor for the compressor model installed in the air conditioner, and uses the read drive current value as the estimated current value.

17. The air conditioning system described in claim 16, characterized in that when the estimation unit reads out from the memory unit the drive current values ​​corresponding to the detected features related to the operation of the compressor, it reads out all drive current values ​​corresponding to values ​​of the features related to the operation of the compressor that are within a predetermined range, and sets the average value of the read drive current values ​​as the estimated current value.

18. An air conditioner having a compressor driven by an inverter circuit, an outdoor unit equipped with an outdoor heat exchanger and an expansion valve, and an indoor unit connected to the outdoor unit, characterized in having: a current detection unit that detects an actual measured current value that is an actual measurement value of the drive current of the compressor; an estimation unit that estimates an estimated current value that is an estimated value of the drive current of the compressor; and a determination unit that calculates the ratio between the actual measured current value and the estimated current value and determines the state of the compressor using the calculated ratio.

19. A fault estimation device comprising: a detection unit that detects an actual measured current value, which is an actual measured drive current value that drives a compressor; an acquisition unit that acquires feature quantities related to the drive of the compressor, including the actual measured current value; an estimation unit that uses the feature quantities to estimate a drive current value corresponding to the operating load of the compressor as an estimated current value; and a judgment unit that judges the state of the compressor using the estimated current value estimated by the estimation unit and the actual measured current value detected by the detection unit.

20. The fault estimation device described in claim 19, characterized in that the acquisition unit acquires the actual measured current value, the compressor rotation speed, a refrigerant state quantity related to the operation of the compressor, and a state quantity related to the efficiency of the compressor as the feature quantities.

21. A failure estimation device according to claim 20, further comprising a calculation unit that calculates a refrigerant suction density, which is the density of the refrigerant sucked into the compressor, and a specific enthalpy difference, which is the difference between the discharge specific enthalpy, which is the specific enthalpy of the refrigerant discharged from the compressor, and the suction specific enthalpy, which is the specific enthalpy of the refrigerant sucked into the compressor, using the refrigerant state quantities related to the operation of the compressor acquired by the acquisition unit; wherein the estimation unit estimates the estimated current value using the refrigerant suction density, the rotation speed of the compressor, the specific enthalpy difference, and state quantities related to the efficiency of the compressor; and wherein the determination unit determines whether or not the compressor has failed based on a comparison result between the estimated current value estimated by the estimation unit and the actual measured current value detected by the detection unit.

22. The fault estimation device according to claim 21, further comprising a learning model for determining a state quantity related to the efficiency of the compressor using the rotation speed of the compressor and a state quantity of the refrigerant related to the operation of the compressor.

23. The fault estimation device according to claim 22, wherein the state quantities of the refrigerant related to the operation of the compressor include the discharge pressure and suction pressure of the compressor.

24. The fault estimation device according to claim 23, wherein the state quantity of the refrigerant related to the operation of the compressor includes a discharge temperature of the compressor.

25. The fault estimation device according to claim 24, wherein the state quantity of the refrigerant related to the operation of the compressor includes the intake temperature of the compressor.

26. A failure estimation device as described in any one of claims 19 to 25, characterized in that it determines that the compressor is not at fault when the ratio between the estimated current value and the actual measured current value is within a predetermined range, and determines that the compressor is at fault when the ratio between the estimated current value and the actual measured current value is outside the predetermined range.

27. The failure estimation device described in claim 26, characterized in that the acquisition unit periodically acquires the actual measured current value, the compressor rotation speed, and the refrigerant state quantity related to the operation of the compressor, the estimation unit estimates the estimated current value each time the acquisition unit acquires the compressor rotation speed and the refrigerant state quantity related to the operation of the compressor, and the determination unit calculates a comparison result between the estimated current value and the actual measured current value each time the estimated current value is estimated, calculates an average value of the comparison results for a predetermined period of time, and determines whether or not the compressor has failed based on the calculated average value.

28. A failure estimation method characterized in that a failure estimation device that estimates a compressor failure performs the following steps: detecting an actual measured current value, which is the actual measured drive current value that drives the compressor; acquiring feature quantities related to the drive of the compressor, including the actual measured current value; estimating a drive current value corresponding to the operating load of the compressor as an estimated current value using the feature quantities related to the drive of the compressor; and determining the state of the compressor using the estimated current value and the detected actual measured current value.

29. An air conditioning system comprising: an air conditioner having an outdoor unit equipped with a compressor, an outdoor heat exchanger, and an expansion valve, and an indoor unit connected to the outdoor unit; and a server device capable of communicating with the air conditioner, wherein the server device has a learning unit that generates a learning model for determining a state quantity related to the efficiency of the compressor using the rotation speed of the compressor and a state quantity of the refrigerant related to the operation of the compressor, and the air conditioner has: a detection unit that detects an actual measured current value that is the actual driving current value driving the compressor; an acquisition unit that acquires the actual measured current value of the compressor, the rotation speed of the compressor, the state quantity of the refrigerant related to the operation of the compressor, and the state quantity related to the efficiency of the compressor; an estimation unit that estimates a driving current value as an estimated current value in accordance with the operating load of the compressor using the rotation speed of the compressor, the state quantity of the refrigerant related to the operation of the compressor acquired by the acquisition unit, and the state quantity related to the efficiency of the compressor determined by the learning model; and a determination unit that determines the state of the compressor using the estimated current value estimated by the estimation unit and the actual measured current value detected by the detection unit.

30. An air conditioner having an outdoor unit equipped with a compressor, an outdoor heat exchanger, and an expansion valve, and an indoor unit connected to the outdoor unit, comprising: a detection unit that detects an actual measured current value, which is an actual measured drive current value that drives the compressor; an acquisition unit that acquires the actual measured current value of the compressor, the rotation speed of the compressor, and state quantities of the refrigerant related to the drive of the compressor; a learning model that is generated using the rotation speed of the compressor and the state quantities of the refrigerant related to the drive of the compressor acquired by the acquisition unit, and that determines state quantities related to the efficiency of the compressor; an estimation unit that estimates a drive current value as an estimated current value in accordance with the operating load of the compressor, using the rotation speed of the compressor and the state quantities of the refrigerant related to the drive of the compressor acquired by the acquisition unit, and the state quantities related to the efficiency of the compressor determined by the learning model; and a determination unit that determines the state of the compressor using the estimated current value estimated by the estimation unit and the actual measured current value detected by the detection unit.

Images