Inference device and learning device

Abstract

The inference device calculates an estimated value of the time required for frost melting relative to the defrost temperature information, wherein the defrost temperature information represents the temperature of the outdoor heat exchanger of the outdoor unit of the air-conditioning device or the state of change of the temperature, and the time required for frost melting is the period during which the temperature of the outdoor heat exchanger is stable within a first range. The inference device comprises: a first data acquisition unit, which acquires the defrost temperature information of the air-conditioning device; and an inference unit, which uses a learned model for inferring the time required for frost melting based on the defrost temperature information, and calculates the estimated value of the time required for frost melting relative to the defrost temperature information based on the defrost temperature information acquired by the first data acquisition unit.

Description

Technical Field

[0001] The present disclosure relates to an inference device and a learning device for an air conditioner having a defrost function. Background Art

[0002] When an air conditioner is operated for heating in winter, frost may form on the outdoor heat exchanger. Therefore, a technique has been proposed for performing a defrost operation to melt the frost on the outdoor heat exchanger using heat from the indoor heat exchanger mounted in the indoor unit (see, for example, Patent Document 1).

[0003] Patent Document 1: International Publication No. 2019 / 035195

[0004] In Patent Document 1, when a certain time has passed since the last defrosting operation ended and the heating operation was resumed and the temperature detected by the temperature sensor is below a predetermined value, it is considered that the conditions for starting the defrosting operation are met and the defrosting operation is started.

[0005] However, in such rule-based control, the defrosting operation may be started even though there is little frost on the outdoor heat exchanger, resulting in poor energy saving and comfort. Summary of the Invention

[0006] The present disclosure has been made to solve the above-mentioned problems, and an object of the present disclosure is to provide an estimating device and a learning device that can improve energy saving by appropriately determining a start time of a defrosting operation.

[0007] The inference device involved in the present disclosure calculates an estimated value of the time required for frost melting relative to defrost temperature information, wherein the defrost temperature information represents the temperature of an outdoor heat exchanger possessed by an outdoor unit of an air-conditioning device or a state of change in the temperature, and the time required for frost melting is a period during which the temperature of the outdoor heat exchanger is stable within a first range. The inference device comprises: a first data acquisition unit, which acquires the above-mentioned defrost temperature information of the above-mentioned air-conditioning device; and an inference unit, which uses a learned model for inferring the time required for frost melting based on the above-mentioned defrost temperature information, and calculates the estimated value of the above-mentioned time required for frost melting relative to the above-mentioned defrost temperature information based on the above-mentioned defrost temperature information acquired by the above-mentioned first data acquisition unit.

[0008] The learning device involved in the present disclosure generates a learned model, which calculates an estimated value of the time required for frost melting relative to defrost temperature information. The defrost temperature information represents the temperature of an outdoor heat exchanger possessed by an outdoor unit of an air-conditioning device or a state of change in the temperature. The time required for frost melting is a period during which the temperature of the outdoor heat exchanger is stable within a first range. The learning device comprises: a second data acquisition unit, which acquires learning data created based on a combination of the defrost temperature information and the actual value of the time required for frost melting; and a model generation unit, which generates a learned model that calculates the estimated value of the time required for frost melting based on the defrost temperature information of the air-conditioning device by using the learning data to learn in a manner such that the estimated value of the time required for frost melting relative to the defrost temperature information is close to the actual value of the time required for frost melting.

[0009] According to the estimation device and the learning device of the present disclosure, by estimating the time required for frost melting as the amount of frost on the outdoor heat exchanger, the start time of the defrosting operation is appropriately determined, thereby achieving improvement in energy saving. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is a schematic diagram showing an example of the configuration of a refrigerant circuit 100 of an air-conditioning apparatus 101 to which the estimation device 20 and the learning device 30 according to Embodiment 1 are applied.

[0011] Figure 2 This is a flowchart showing an example of the defrosting operation control of the air-conditioning apparatus 101 according to the first embodiment.

[0012] Figure 3 This is a diagram showing an example of changes over time in the defrosting temperature (defrosting temperature) θ during the defrosting operation.

[0013] Figure 4 This is a block diagram showing the configuration of the estimation device 20 according to the first embodiment.

[0014] Figure 5 This is a flowchart showing the flow of processing by the estimation device 20 according to the first embodiment.

[0015] Figure 6 This is a flowchart showing the flow of processing performed by the air-conditioning apparatus 101 according to Embodiment 1.

[0016] Figure 7 This is a block diagram showing the configuration of the learning device 30 according to the first embodiment.

[0017] Figure 8Schematically shows an example of a model of the neural network 34 included in the model generation unit 32 .

[0018] Figure 9 This is a flowchart showing the flow of processing by the learning device 30 according to the first embodiment. DETAILED DESCRIPTION

[0019] Hereinafter, the embodiments of the inference device 20 and the learning device 30 involved in the present disclosure will be described with reference to the accompanying drawings. The present disclosure is not limited to the following embodiments, and various modifications can be made without departing from the scope of the present disclosure. In addition, the present disclosure includes all combinations of combinable structures in the structures shown in the following embodiments and their modifications. In addition, in each figure, the structure marked with the same figure mark is the same or equivalent structure, which is common to the entire text of the specification. In addition, in each figure, there are cases where the relative size relationship or shape of each component is different from the actual structure.

[0020] Implementation method 1.

[0021] Hereinafter, description will be given of the estimation device 20 and the learning device 30 according to Embodiment 1. The estimation device 20 and the learning device 30 are mounted on the air-conditioning device 101 or are connected to the air-conditioning device 101 for use.

[0022] <Structure of Air Conditioning Device 101>

[0023] Figure 1 This is a schematic diagram showing an example of the configuration of a refrigerant circuit 100 of an air-conditioning apparatus 101 to which the estimation device 20 and the learning device 30 according to Embodiment 1 are applied.

[0024] like Figure 1 As shown, an air conditioning apparatus 101 includes an indoor unit 1 positioned in an indoor space to be air-conditioned, and an outdoor unit 2 positioned outdoors. Hereinafter, the environment in which the outdoor unit 2 is installed is referred to as the external air environment. The indoor unit 1 includes an indoor heat exchanger 5. Meanwhile, the outdoor unit 2 includes a compressor 3, a four-way valve 4, an outdoor heat exchanger 6, and an expansion valve 7. The compressor 3, the four-way valve 4, the outdoor heat exchanger 6, the expansion valve 7, and the indoor heat exchanger 5 are connected via refrigerant piping to form a refrigerant circuit 100.

[0025] The indoor heat exchanger 5 exchanges heat between the refrigerant flowing through the refrigerant piping disposed therein and the indoor air. Meanwhile, the outdoor heat exchanger 6 exchanges heat between the refrigerant flowing through the refrigerant piping disposed therein and the outdoor air. The indoor heat exchanger 5 and the outdoor heat exchanger 6 are, for example, fin-tube heat exchangers. Furthermore, the indoor heat exchanger 5 and the outdoor heat exchanger 6 may each be divided into multiple heat exchangers. In this case, these multiple heat exchangers are connected in series or in parallel.

[0026] The compressor 3 draws in the refrigerant flowing through the refrigerant circuit 100. The compressor 3 compresses the drawn-in refrigerant and discharges it. The compressor 3 is, for example, an inverter compressor. The refrigerant discharged from the compressor 3 flows into the indoor heat exchanger 5 or the outdoor heat exchanger 6.

[0027] The four-way valve 4 is a flow switching device configured to switch between a cooling operation for cooling the indoor space where the indoor unit 1 is arranged and a heating operation for heating the indoor space. Figure 1 Indicates the state in which the air conditioner 101 is performing heating operation. Figure 1 As shown, when the air conditioner 101 is in heating operation, the four-way valve 4 becomes Figure 1 In the state shown by the solid line, the refrigerant discharged from the compressor 3 flows into the indoor heat exchanger 5. At this time, the outdoor heat exchanger 6 functions as an evaporator and the indoor heat exchanger 5 functions as a condenser. On the other hand, when the air conditioner 101 is in cooling operation, the four-way valve 4 becomes Figure 1 In the state shown by the dotted line, the refrigerant discharged from the compressor 3 flows into the outdoor heat exchanger 6. At this time, the outdoor heat exchanger 6 functions as a condenser, and the indoor heat exchanger 5 functions as an evaporator. Alternatively, another flow switching device having the same function may be used in place of the four-way valve 4.

[0028] The expansion valve 7 is a decompression device for decompressing the refrigerant, and is composed of, for example, an electronic expansion valve. The expansion valve 7 is provided between the outdoor heat exchanger 6 and the indoor heat exchanger 5. Alternatively, another decompression device having the same function may be used in place of the expansion valve 7.

[0029] The refrigerant circuit 100 is filled with a refrigerant. The type of the refrigerant is not particularly limited, and may be R32, R410A, or the like.

[0030] The indoor unit 1 further includes an indoor blower 8 for ventilating the indoor heat exchanger 5. The indoor blower 8 is disposed on the upwind side of the indoor heat exchanger 5. Alternatively, the indoor blower 8 may be disposed on the downwind side of the indoor heat exchanger 5.

[0031] The outdoor unit 2 further includes an outdoor blower 9 for ventilating the outdoor heat exchanger 6. The outdoor blower 9 is arranged on the leeward side of the outdoor heat exchanger 6. Alternatively, the outdoor blower 9 may be arranged on the upwind side of the outdoor heat exchanger 6.

[0032] The outdoor fan 9 is equipped with a device that detects or estimates the current value used by the outdoor fan 9 for air supply. This device is referred to as a current measuring device 16. The current measuring device 16 is composed of, for example, a current sensor or a processor. The current information detected or estimated by the current measuring device 16 is output to the control unit 15 equipped in the outdoor unit 2. Alternatively, the current measuring device 16 may be provided in the control unit 15.

[0033] A temperature sensor 10 is mounted on the housing of the compressor 3 of the outdoor unit 2. The temperature sensor 10 detects the temperature of the compressor 3. The temperature sensor 10 can be mounted anywhere that can detect the temperature of the compressor 3. For example, the temperature sensor 10 can be installed on the refrigerant piping from the compressor 3 to the four-way valve 4, rather than on the housing of the compressor 3. The compressor temperature information detected by the temperature sensor 10 is output to the control unit 15.

[0034] A temperature sensor 11 is installed on the upwind side of the indoor air blower 8 of the indoor unit 1. The temperature sensor 11 detects the temperature of the air before it flows into the indoor heat exchanger 5, that is, the room temperature. In addition, the position of the temperature sensor 11 is not limited as long as it can detect the room temperature. Figure 1 The room temperature information detected by the temperature sensor 11 is output to the control unit 15.

[0035] A temperature sensor 12 is mounted on the wall of the refrigerant pipe of the indoor heat exchanger 5. The temperature sensor 12 detects the temperature of the indoor heat exchanger 5 when the indoor heat exchanger 5 functions as a condenser during heating, that is, the condensing temperature. The temperature sensor 12 can be located at any position that can detect the temperature of the indoor heat exchanger 5, and its location is not limited to Figure 1 The condensation temperature information detected by the temperature sensor 12 is output to the control unit 15.

[0036] The outdoor unit 2 is equipped with a temperature sensor 13 for measuring the temperature of the air ventilated to the outdoor heat exchanger 6 by the outdoor blower 9. Since the temperature sensor 13 measures the temperature of the air before passing through the outdoor heat exchanger 6, that is, the outside air temperature, it is installed on the upwind side of the outdoor heat exchanger 6. In addition, the location of the temperature sensor 13 is not limited to any location as long as it can detect the temperature of the air before passing through the outdoor heat exchanger 6. Figure 1The outside air temperature information detected by the temperature sensor 13 is output to the control unit 15.

[0037] A temperature sensor 14 is mounted on the wall of the refrigerant pipe of the outdoor heat exchanger 6. The temperature sensor 14 detects the temperature of the outdoor heat exchanger 6 when the outdoor heat exchanger 6 functions as an evaporator during heating, that is, the evaporation temperature. The temperature sensor 14 can be located at any position as long as it can estimate the temperature of the outdoor heat exchanger 6. Figure 1 The evaporation temperature information detected by the temperature sensor 14 is output to the control unit 15.

[0038] exist Figure 1 In the embodiment, five temperature sensors 10 to 14 are provided, but the number of temperature sensors 10 to 14 is not limited to Figure 1 The number of may be more or less than these. For example, when the outdoor heat exchanger 6 is divided into a plurality of heat exchangers, a plurality of temperature sensors 14 may be installed in the outdoor heat exchanger.

[0039] In addition, the type of sensor is not limited to Figure 1 For example, a humidity sensor 17 that measures the humidity of the external air environment where the outdoor unit 2 is installed, or an illuminance sensor 18 that measures the illuminance of the external air environment, can be installed in the outdoor unit 2. In this case, for example, both temperature information and humidity information of the external air environment can be obtained by the temperature sensor 13 and the humidity sensor 17. It is preferable that the illuminance measured by the illuminance sensor 18 is a value indicating the amount of sunlight on the casing of the outdoor unit 6. The information detected by these sensors 10 to 14 and 16 to 18 is collected by the control unit 15 equipped in the outdoor unit 2.

[0040] The control unit 15 is composed of a control substrate. The control substrate of the control unit 15 is equipped with a control device, a storage device, and a drive circuit. The control device is composed of, for example, dedicated hardware, a CPU (Central Processing Unit) or a microprocessor that executes programs stored in memory. The storage device is a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, or EPROM (Erasable Programmable ROM), or a disk such as a magnetic disk, floppy disk, or optical disk.

[0041] <Operation of Air Conditioning Device 101>

[0042] Next, Figure 1 The operation of the air conditioning device 101 will be briefly described.

[0043] As described above, refrigerant is sealed within the refrigerant circuit 100 and compressed by the compressor 3. During cooling, a refrigeration cycle is formed by the following cooling operation circuit. Specifically, the refrigerant compressed by the compressor 3 condenses and liquefies in the outdoor heat exchanger 6, expands in the expansion valve 7, and then evaporates in the indoor heat exchanger 5 before returning to the compressor 3. Furthermore, during defrost operation, a refrigeration cycle is formed by the cooling operation circuit.

[0044] On the other hand, during heating, a refrigeration cycle is formed by the following heating operation circuit: That is, the refrigerant compressed by the compressor 3 is condensed and liquefied in the indoor heat exchanger 5, expanded in the expansion valve 7, evaporated in the outdoor heat exchanger 6, and then returned to the compressor 3.

[0045] When cooling or heating is performed as described above, Figure 1 The air conditioner 101 controls each component so that the temperature detected by the indoor temperature sensor 11, that is, the room temperature, reaches a target value. Specifically, the air conditioner 101 controls the rotation speed of the compressor 3, the opening degree of the expansion valve 7, the air volume of the indoor fan 8, and the air volume of the outdoor fan 9.

[0046] This control is performed based on the temperatures detected by the temperature sensors 10 to 14 to control the cooling capacity or heating capacity of the air-conditioning apparatus 101. Such control is performed by the control unit 15 of the outdoor unit 2.

[0047] During heating operation, under conditions of low temperature and high humidity, frost may form on the outdoor heat exchanger 6, which functions as an evaporator. In this case, the ventilation resistance of the outdoor blower 9 increases, reducing the amount of heat exchanged in the outdoor heat exchanger 6 and lowering the heating capacity. Therefore, the air conditioner 101 performs a defrost operation to melt the frost on the outdoor heat exchanger 6.

[0048] <Defrosting operation>

[0049] Figure 2 This is a flowchart showing an example of the defrosting operation control of the air-conditioning apparatus 101 according to the first embodiment. Figure 2 The defrosting operation is a general operation example and is not limited to this. Figure 2 The defrost operation of the outdoor heat exchanger 6 is briefly described. The temperature of the refrigerant piping of the outdoor heat exchanger 6 detected by the temperature sensor 14 is referred to as the defrost temperature θ. Therefore, the defrost temperature θ is the temperature of the outdoor heat exchanger 6 when the outdoor heat exchanger 6 functions as an evaporator during heating.

[0050] like Figure 2As shown, in step S1, the control unit 15 determines whether a certain time has passed since the air conditioner 101 started or resumed heating operation. The certain time is a pre-set value. If it is determined that the certain time has not passed, the control unit 15 directly ends the operation. Figure 2 On the other hand, when it is determined that a certain time has passed, the control unit 15 proceeds to step S2.

[0051] In step S2, the control unit 15 determines whether the defrost temperature θ detected by the temperature sensor 14 is below a predetermined value. If it is determined that the defrost temperature θ is greater than the predetermined value, the control unit 15 directly ends the process. Figure 2 On the other hand, when it is determined that the defrost temperature θ is equal to or lower than the predetermined value, the control unit 15 proceeds to step S3.

[0052] In step S3, the control unit 15 determines that the conditions for starting the defrosting operation are met based on the determinations in steps S1 and S2, that is, that frost has formed on the outdoor heat exchanger 6. The control unit 15 temporarily stops the compressor 3 to start the defrosting operation.

[0053] In step S4, the controller 15 switches the four-way valve 4 to establish the aforementioned cooling operation circuit, restarting the compressor 3 to initiate the defrost operation. During the defrost operation, frost adhering to the outdoor heat exchanger 6 is melted using, for example, a reverse defrosting method that circulates refrigerant. During the defrost operation, the refrigerant, which has reached a high temperature and high pressure in the compressor 3, flows through the refrigerant piping to the outdoor heat exchanger 6. This refrigerant imparts heat to the frost adhering to the outdoor heat exchanger 6, melting the frost into water.

[0054] The defrosting operation continues until the defrost temperature θ reaches or exceeds the predetermined value. Therefore, in step S5, the control unit 15 determines whether the defrost temperature θ reaches or exceeds the predetermined value. If the defrost temperature θ is determined to be or exceeds the predetermined value in step S5, the control unit 15 proceeds to step S6.

[0055] In step S6, the control unit 15 determines that the defrosting operation termination condition is met, terminates the defrosting operation, and proceeds to step S7. When the defrosting operation is terminated, the control unit 15 first stops the compressor 3 and switches the four-way valve 4 to return to the above-mentioned heating operation circuit.

[0056] In step S7 , the control unit 15 restarts the compressor 3 to resume the heating operation.

[0057] Conventional defrost control has been implemented, but it sometimes initiates defrost operation even when no frost has formed, such as in severe winter or in cold climates, when the outside air temperature is low. In these cases, heating is temporarily suspended during defrost operation, resulting in an unnecessary drop in indoor temperature. Furthermore, additional energy is required, including energy required for the defrost operation and energy required to restore the indoor temperature, which has dropped during the defrost operation, to the set value.

[0058] Therefore, in the estimation device 20 according to the first embodiment, the learned model is used to obtain an estimated value Pest of the "time required for frost to melt" to be described later. The control unit 15 of the air conditioner 101 determines whether to start the cooling process based on the estimated value Pest of the "time required for frost to melt". Figure 2 This can avoid unnecessary defrosting operation when the outdoor heat exchanger 6 is not frosted.

[0059] <Time required for frost to melt>

[0060] Figure 3 : is a diagram showing an example of temporal change of the defrost temperature θ during the defrost operation. Figure 3 In the figure, the horizontal axis represents time and the vertical axis represents the defrost temperature θ. However, the temporal variation pattern of the defrost temperature θ varies depending on the values ​​of various parameters such as the outside air temperature, the amount of frost attached to the outdoor heat exchanger 6, and the number of compressors 3 driven during the defrost operation. Therefore, the temporal variation pattern of the defrost temperature θ is not limited to Figure 3 The following is an example of the situation shown in the following example. Figure 3 The situation shown is used as an example for explanation.

[0061] like Figure 3 As shown in the figure, when the defrost operation begins at time t0, the defrost temperature θ rises after the defrost operation begins. The refrigerant flowing through the refrigerant piping of the outdoor heat exchanger 6 then imparts heat to the frost adhering to the outdoor heat exchanger 6, gradually melting the frost. Consequently, the defrost temperature θ temporarily stabilizes near 0°C, then rises again after the frost on the outdoor heat exchanger 6 melts. As described above, the defrost operation ends when the defrost temperature θ reaches or exceeds the specified value.

[0062] In view of this, in embodiment 1, as Figure 3As shown, two temperatures θ1 and θ2 are specified near 0°C. At this time, temperature θ1 is set to a temperature lower than temperature θ2 (θ1<θ2). At this time, temperatures θ1 and θ2 are set, for example, so that the relationship θ1≤0<θ2 holds. Here, temperature θ1 is the temperature just before the defrost temperature θ is about to stabilize. Temperature θ2 is the temperature just before the defrost temperature θ rises again. The moment when the defrost temperature θ changes to temperature θ1 is set to t1, and the moment when the defrost temperature θ changes to temperature θ2 is set to t2. At this time, the period during which time t satisfies the relationship t1≤t≤t2 is the period during which the defrost temperature θ exists near the melting point, specifically, the period during which θ1≤θ≤θ2. This period is the period during which the refrigerant flowing in the refrigerant piping of the outdoor heat exchanger 6 imparts heat to the frost attached to the outdoor heat exchanger 6 to melt the frost. Hereinafter, this period will be referred to as the "time required for frost melting". Figure 3 Indicated by the reference symbol "P".

[0063] Therefore, assuming the range between temperatures θ1 and θ2 is the "first range," the frost-melting time P represents the period during which the defrost temperature θ, serving as defrost temperature information, remains stable within the first range. During the frost-melting time P, the frost adhering to the outdoor heat exchanger 6 transforms into water. Therefore, during the frost-melting time P, the heat energy of the refrigerant flowing through the refrigerant piping is consumed not as sensible heat from the change in the defrost temperature θ, but as latent heat from the change in the frost state from water. Therefore, the frost-melting time P represents the period during which the defrost temperature θ, serving as defrost temperature information, remains near its melting point, gradually melting the frost adhering to the outdoor heat exchanger 6. Therefore, during the frost-melting time P, the defrost temperature θ, serving as defrost temperature information, remains stable within the first range.

[0064] In addition, it is described above that the temperature θ1 and the temperature θ2 are appropriately set to, for example, the relationship θ1≤0<θ2. However, this is not limiting, and since it is also possible to consider that the melting point slightly fluctuates from 0°C due to the influence of impurities such as dirt, it can be set to any of the following relationships: θ1<0≤θ2, θ1<θ2≤0, or 0≤θ1<θ2. However, the temperatures θ1 and θ2 are set in the range of -20°C to +20°C, preferably in the range of -5°C to +5°C, or in the range of -10°C to +10°C. In any range, the "first range" is a range that includes the melting point or 0°C.

[0065] The time required for frost melting P varies depending on the amount of frost attached to the outdoor heat exchanger 6. When the amount of frost is large, it takes a long time to melt the frost, so the time required for frost melting P becomes longer. On the other hand, when the amount of frost is small, the time required for frost melting P becomes shorter. However, when the amount of frost is small and the time required for frost melting P is extremely short, after the defrosting temperature θ becomes higher than the temperature θ1, the frost will not melt as fast as before. Figure 3 It temporarily stabilizes near 0°C, but sometimes reaches a temperature above θ2.

[0066] Depending on the setting of the specified value for defrost temperature θ and the method for selecting temperatures θ1 and θ2, it is possible that at time t0 the defrost temperature θ is already θ1 ≤ θ < θ2. In this case, the time required for frost melting, P, is counted from time t0. In other words, let t1 = t0.

[0067] Furthermore, due to the stopping and starting of compressor 3 during defrost operation and the resulting series of transient phenomena, an unstable refrigerant state may occur, causing the defrost temperature θ to temporarily change from θ < θ1 to θ1 ≤ θ < θ2 and then again to θ < θ1. In this case, the second time the temperature changes to θ1 ≤ θ < θ2 is used as the starting point t1 for counting the required frost melting time P. If this phenomenon occurs multiple times, the final time the temperature changes to θ1 ≤ θ < θ2 is used as the starting point t1 for counting.

[0068] Furthermore, due to the stopping and starting of compressor 3 during defrost operation and the resulting series of transient phenomena, an unstable refrigerant state may occur, causing the defrost temperature θ to temporarily change from θ1 ≤ θ < θ2 to θ2 ≤ θ and then again to θ1 ≤ θ < θ2. In this case, the second time the temperature changes to θ2 ≤ θ is used as the end point t2 for counting the required frost melting time P. If this phenomenon occurs multiple times, the final time the temperature changes to θ2 ≤ θ is used as the end point t2 for counting.

[0069] Note that the method of determining the start and end points of counting the time required for frost melting P is merely an example and is not limited to the above.

[0070] The control unit 15 stores control values ​​such as the operating frequency of the compressor 3 , the current value of the outdoor blower 9 , detection values ​​of the temperature sensors 10 to 14 , and the time required for frost melting in a storage device provided therein.

[0071] <Inference stage>

[0072] Next, the inference device 20 according to the first embodiment will be described. The inference device 20 according to the first embodiment obtains the above-mentioned Figure 3The estimated value Pest of the frost melting time P shown in FIG. 1 is output to the air conditioning apparatus 101. The air conditioning apparatus 101 according to Embodiment 1 performs the operation only when the estimated value Pest of the frost melting time P output from the estimation device 20 is equal to or greater than a predetermined threshold value Th. Figure 2 Therefore, in the first embodiment, the defrosting operation is performed when all of the following three conditions (a) to (c) are satisfied. This can avoid unnecessary defrosting operation.

[0073] (a): The estimated value Pest of the time required for frost melting P is equal to or greater than a preset threshold value Th.

[0074] (b): A certain time has passed since the air conditioner 101 started the heating operation ( Figure 2 (Step S1 is "Yes").

[0075] (c): The current defrost temperature θ detected by the temperature sensor 14 is below a predetermined value ( Figure 2 Step S2 in the example is “Yes”).

[0076] Below, use Figure 4 The configuration of the estimation device 20 according to the first embodiment will be described. Figure 4 : is a block diagram showing the structure of the inference device 20 involved in the first embodiment. Figure 4 As shown, the estimation device 20 includes a first data acquisition unit 21 and an estimation unit 22. In addition, the estimation device 20 is connected to a learned model storage unit 33 or an external device 40.

[0077] In addition, the inference device 20 can also be used as Figure 1 The inference device 20 is provided as a component of the air conditioning device 101 shown. In this case, the inference device 20 is, for example, built into the outdoor unit 2 of the air conditioning device 101. Alternatively, the inference device 20 may be provided independently of the air conditioning device 101. For example, the inference device 20 may reside on a cloud server. In this case, the control unit 15 of the air conditioning device 101 is communicatively connected to the inference device 20.

[0078] The first data acquisition unit 21 acquires the defrost temperature θ detected by the temperature sensor 14 as defrost temperature information. As described above, the defrost temperature θ is the temperature of the outdoor heat exchanger 6 detected by the temperature sensor 14. The first data acquisition unit 21 may acquire the defrost temperature θ directly from the temperature sensor 14 or via the control unit 15.

[0079] The estimation unit 22 uses the learned model to obtain an estimated value Pest of the time required for frost melting. Specifically, the estimation unit 22 inputs the defrost temperature θ obtained by the first data acquisition unit 21 into the learned model to obtain an estimated value Pest of the time required for frost melting based on the defrost temperature θ.

[0080] The learned model is generated by the learning device 30 (described later) and stored in the learned model storage unit 33. The inference unit 22 obtains the learned model from the learned model storage unit 33. Alternatively, the inference unit 22 obtains the learned model from an external device 40 via a communication line such as the Internet. The external device 40 may be, for example, one or more other air conditioners, a cloud server, or the website of the manufacturer of the air conditioner 101.

[0081] Here, the hardware structure of the inference device 20 is described. The inference device 20 is composed of a processing circuit that realizes the functions of the first data acquisition unit 21 and the inference unit 22. The processing circuit is composed of dedicated hardware or a processor. Dedicated hardware is, for example, an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The processor executes the program stored in the memory. In addition, the inference device 20 has a storage unit (not shown) that stores the program and calculation results. The storage unit is composed of a memory. The memory is a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable ROM), or a disk such as a magnetic disk, a floppy disk, or an optical disk.

[0082] Next, use Figure 5 The flow of processing by the estimation device 20 will be described. Figure 5 This is a flowchart showing the flow of processing of the estimation device 20 according to Embodiment 1. When the air-conditioning device 101 performs heating operation, for example, Figure 5 Processing of the process.

[0083] like Figure 5 As shown, first, in step S21, the first data acquisition unit 21 acquires the current defrost temperature θ detected by the temperature sensor 14. The defrost temperature θ at this time is the temperature of the outdoor heat exchanger 6 of the air conditioner 101 when the outdoor heat exchanger 6 functions as an evaporator, and is the temperature before the defrost operation starts.

[0084] Next, in step S22 , the estimation unit 22 inputs the defrost temperature θ acquired in step S21 into the learned model stored in the learned model storage unit 33 to obtain an estimated value Pest of the frost melting time P. The learned model will be described later.

[0085] Next, in step S23 , the estimation unit 22 outputs the estimated value Pest of the frost-melting required time P obtained using the learned model to the control unit 15 of the air-conditioning apparatus 101 .

[0086] The control unit 15 of the air conditioner 101 receives the estimated value Pest of the time required for frost melting P from the estimation device 20, and performs Figure 6 processing. Figure 6 This is a flowchart showing the flow of processing performed by the air-conditioning apparatus 101 according to Embodiment 1.

[0087] like Figure 6 As shown, in step S31, the control unit 15 of the air conditioning device 101 determines whether the estimated value Pest of the frost-melting time P received from the estimation device 20 is less than a preset threshold value Th. If the estimated value Pest of the frost-melting time P is less than the threshold value Th, the control unit 15 proceeds to step S32. On the other hand, if the estimated value Pest of the frost-melting time P is greater than the threshold value Th, the control unit 15 proceeds to step S33.

[0088] In step S32, the control unit 15 determines that the defrosting operation is not to be performed and ends the operation immediately. Figure 6 processing.

[0089] In step S33, the control unit 15 proceeds to Figure 2 The process of processing. Figure 2 In the process, when the conditions of step S1 and step S2 are met, the control unit 15 performs the defrosting operation.

[0090] Thus, in the first embodiment, the air conditioner 101 compares the estimated value Pest of the frost melting time P output from the estimation device 20 with the threshold value Th. And, if the estimated value Pest of the frost melting time P is less than the threshold value Th, even if a certain time ( Figure 2 If S1 in the above example is “Yes”), the defrost temperature θ is below the specified value ( Figure 2 Thus, unnecessary defrosting operation can be avoided, power consumption can be reduced, and reduction in user comfort caused by the defrosting operation can be prevented.

[0091] On the other hand, when the estimated value Pest of the frost melting required time P is equal to or greater than the threshold value Th, after a certain time ( Figure 2 If S1 in the above example is “Yes”), the defrost temperature θ is below the specified value ( Figure 2 If the defrost start conditions are met during the conventional air conditioner defrost control and defrost initiation is determined by inference, the system switches to defrost operation. This allows defrost to be performed at the appropriate time when defrost operation is required.

[0092] As described above, in the first embodiment, the estimation device 20 obtains the estimated value Pest of the time required for frost melting P as the amount of frost to appropriately determine the start time of the defrosting operation, thereby achieving improvement in energy saving.

[0093] In the first embodiment, the defrost temperature θ is used as the defrost temperature information, but it does not need to be the value of the defrost temperature θ itself. The defrost temperature information may also be a value indicating the state of the temperature change of the outdoor heat exchanger 6. That is, the defrost temperature information may be, for example, at least one of the average value, cumulative value, integrated value, maximum value, or minimum value of the defrost temperature θ in the time interval from the end of the previous defrost operation to the present. In addition, the defrost temperature information may be as follows: Figure 3 The slope α of the defrost temperature θ relative to time t is shown in FIG. The slope α represents the ratio of the change in the defrost temperature θ relative to the change in time t. Therefore, Figure 3 As shown in FIG. 1 , during the time P required for frost melting, the slope α is 0 or substantially 0.

[0094] Furthermore, the estimation device 20 can use humidity information in addition to defrost temperature information to determine an estimated value Pest for the time required for frost melting. The higher the humidity in the external air environment, the greater the amount of frost accumulated on the outdoor heat exchanger 6. Therefore, the time required for frost melting P tends to increase with increasing humidity. In this case, the first data acquisition unit 21 acquires defrost temperature information from the temperature sensor 14 and acquires the humidity of the external air environment where the outdoor unit 2 is installed from the humidity sensor 17 as humidity information. In this case, the learned model is a learned model for estimating the time required for frost melting based on the defrost temperature and humidity information. The estimation unit 22 uses this learned model to determine an estimated value Pest for the time required for frost melting P.

[0095] Alternatively, the estimation device 20 can estimate the frost-melting time P using current information from the outdoor fan 9 in the outdoor unit 2 of the air conditioner 101 in addition to defrost temperature information. As the amount of frost on the outdoor heat exchanger 6 increases, ventilation performance deteriorates, increasing the load on the outdoor fan 9 and the current value. Therefore, as the current value relative to the rotational speed of the outdoor fan 9 increases, the likelihood of increased frost accumulation on the outdoor heat exchanger 6 increases, leading to a tendency for the frost-melting time P to increase. In this case, the first data acquisition unit 21 acquires defrost temperature information from the temperature sensor 14 and acquires the current value used by the outdoor fan 9 during ventilation as current information from the current measuring device 16. In this case, the learned model is used to estimate the frost-melting time based on the defrost temperature and humidity information. The estimation unit 22 uses this learned model to determine an estimated value Pest for the frost-melting time P.

[0096] Furthermore, the estimation device 20 can use the outside air temperature information in addition to the defrost temperature information to estimate the frost-melting time P. The higher the outside air temperature, the shorter the frost-melting time P. In this case, the first data acquisition unit 21 acquires the defrost temperature information from the temperature sensor 14 and the outside air temperature from the temperature sensor 13 as the outside air temperature information. In this case, the learned model is a learned model for estimating the frost-melting time based on the defrost temperature information and the outside air temperature information. The estimation unit 22 uses this learned model to determine an estimated value Pest for the frost-melting time P.

[0097] Furthermore, the estimation device 20 can use illumination information in addition to defrost temperature information to estimate the frost-melting time P. Generally, the greater the illumination, the shorter the frost-melting time P. In this case, the first data acquisition unit 21 acquires defrost temperature information from the temperature sensor 14 and acquires the illumination of the external air environment where the outdoor unit 2 is installed from the illumination sensor 18 as illumination information. In this case, the learned model is a learned model for estimating the frost-melting time based on the defrost temperature information and illumination information. The estimation unit 22 uses this learned model to determine an estimated value Pest for the frost-melting time P.

[0098] In this manner, the estimation device 20 can simply use at least one of the defrost temperature information, humidity information, current information, outside air temperature information, and illuminance information and, using the learned model, calculate the estimated value Pest for the time required for frost melting. The combination of this information can be modified as appropriate. Furthermore, the estimation device 20 can obtain this information directly from the sensors or through the control unit 15.

[0099] Furthermore, the estimation device 20 outputs an estimated value Pest of the frost-melting time P using the learned model acquired by the model generation unit 32 of the learning device 30. However, this is not limiting. As described above, the estimation device 20 may also obtain a learned model from the external device 40 and output an estimated value Pest of the frost-melting time P based on the learned model.

[0100] Learning Stage

[0101] Below, use Figure 7 The configuration of the learning device 30 according to the first embodiment will be described. Figure 7 1 is a block diagram showing the structure of the learning device 30 according to the first embodiment. Figure 7 As shown, the learning device 30 includes a second data acquisition unit 31 , a model generation unit 32 , and a learned model storage unit 33 .

[0102] The second data acquisition unit 31 acquires the defrost temperature θ and the measured value Pm of the time required for frost melting P as defrost temperature information. In addition, the second data acquisition unit 31 combines the defrost temperature θ with the measured value Pm of the time required for frost melting P to generate learning data. Here, the defrost temperature θ is the surface temperature of the outdoor heat exchanger 6 possessed by the air-conditioning device 101, and is a value detected by the temperature sensor 14. In addition, the measured value Pm of the time required for frost melting P is the period during which the temperature of the outdoor heat exchanger 6 is actually near the melting point during the defrosting operation of the air-conditioning device 101. Therefore, the measured value Pm of the time required for frost melting P is specifically the value of the defrost temperature θ detected by the temperature sensor 14 relative to the melting point. Figure 3 The duration of the period during which temperatures θ1 and θ2 satisfy the relationship θ1 ≤ θ ≤ θ2 is shown. The actual measured value Pm of the frost-melting time P is measured by the control unit 15 of the air conditioning unit 101. Specifically, the control unit 15 calculates the actual measured value Pm of the frost-melting time P based on the defrost temperature θ detected by the temperature sensor 14. The second data acquisition unit 31 can obtain the defrost temperature θ and the actual measured value Pm of the frost-melting time P from the control unit 15, or it can directly obtain the defrost temperature θ from the temperature sensor 14.

[0103] The model generation unit 32 learns the frost-melting time P based on learning data generated based on a combination of the defrost temperature θ output from the second data acquisition unit 31 and the measured value Pm of the frost-melting time P. Specifically, the model generation unit 32 generates a learned model for estimating the optimal frost-melting time P based on the defrost temperature θ of the air conditioner 101 and the measured value Pm of the frost-melting time P. The learning data here is data that associates the defrost temperature θ with the measured value Pm of the frost-melting time P.

[0104] While the example herein uses the defrost temperature θ as the defrost temperature information, the present invention is not limited to this example. The defrost temperature information may also be the current defrost temperature θ or a value indicating a changing state of the defrost temperature θ. Specifically, the defrost temperature information may be the current defrost temperature θ or at least one of the average value, cumulative value, integrated value, maximum value, and minimum value of the defrost temperature θ during the time period from the end of the previous defrost operation to the current time.

[0105] The learning device 30 is used to learn the frost-melting time P for the air conditioning device 101. The learning device 30 can be integrated into the air conditioning device 101 or installed independently of the air conditioning device 101. In this case, the learning device 30 is connected to the air conditioning device 101 via a network, for example, along with the inference device 20. Alternatively, the learning device 30 and the inference device 20 can be built into the air conditioning device 101. Furthermore, the learning device 30 and the inference device 20 can reside on a cloud server.

[0106] The model generation unit 32 can use a learning algorithm used for learning, such as trained learning, untrained learning, reinforcement learning, and other well-known algorithms. As an example, a case where a neural network is applied will be described.

[0107] Model generator 32 learns the frost-melting time P using, for example, a neural network model through so-called trained learning. Here, trained learning refers to a technique in which a set of input and result (label) data is given to learning device 30 to learn features present in the learning data and infer results based on the input.

[0108] A neural network consists of an input layer composed of multiple neurons, an intermediate layer (hidden layer) composed of multiple neurons, and an output layer composed of multiple neurons. The intermediate layer can be one or more than two layers.

[0109] Figure 8 : is a diagram schematically showing an example of a model of the neural network 34 provided in the model generation unit 32. Here, the neural network 34 of the model generation unit 32 is exemplified as follows: Figure 8 The following describes a three-layer neural network as an example. The neural network 34 has three input layers X1, X2, and X3, two intermediate layers Y1 and Y2, and three output layers Z1, Z2, and Z3.

[0110] At this time, if multiple inputs In1, In2, and In3 are input to the input layers X1, X2, and X3, respectively, the inputs In1, In2, and In3 are multiplied by a preset first weight W1 in the input layers X1, X2, and X3. Figure 8In the example, the first weight W1 of input layer X1 is either w11 or w12. Similarly, the first weight W1 of input layer X2 is either w13 or w14, and the first weight W1 of input layer X3 is either w15 or w16. The multiplication results obtained by multiplying by the first weight W1 are input to intermediate layers Y1 and Y2. Here, these multiplication results are referred to as the first multiplication results.

[0111] In the intermediate layers Y1 and Y2, the first multiplication result is multiplied by a preset second weight W2. Figure 8 In this example, the second weight W2 of intermediate layer Y1 is of three types: w21, w23, and w25. Similarly, the second weight W2 of intermediate layer Y2 is of three types: w22, w24, and w26. The multiplication result obtained by multiplying by the second weight W2 is input to output layers Z1, Z2, and Z3. Here, this multiplication result is referred to as the second multiplication result. The second multiplication result is output from output layers Z1, Z2, and Z3 as output results Out1, Out2, and Out3. The output results Out1, Out2, and Out3 vary depending on the values ​​of weights W1 and W2.

[0112] In the first embodiment, the neural network 34 learns the frost melting time P by so-called trained learning based on the learning data created based on the combination of the defrost temperature θ and the frost melting time Pm acquired by the second data acquisition unit 31 .

[0113] Specifically, neural network 34 learns the frost-melting time P for the defrost temperature θ using the following procedure. First, the first weight W1 and second weight W2 are adjusted so that the output results Out1, Out2, and Out3 from output layers Z1, Z2, and Z3, when the defrost temperature θ is input to input layers X1, X2, and X3, approximate the measured value Pm of the frost-melting time P. Thus, first weight W1 and second weight W2 are set, and the frost-melting time P for the defrost temperature θ is learned.

[0114] The model generation unit 32 generates and outputs a learned model of the frost melting required time P with respect to the defrost temperature θ by executing the above-described learning.

[0115] The learned model storage unit 33 stores the learned model output from the model generation unit 32 .

[0116] Here, the hardware structure of the learning device 30 is described. The learning device 30 is composed of a processing circuit that realizes the functions of the second data acquisition unit 31 and the model generation unit 32. The processing circuit is composed of dedicated hardware or a processor. Dedicated hardware is, for example, an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The processor executes the program stored in the memory. In addition, the learning device 30 has a storage device (not shown) that stores the program and calculation results. The storage device realizes the function of the learned model storage unit 33. The storage device is composed of a memory. The memory is a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable ROM), or a disk such as a magnetic disk, a floppy disk, or an optical disk.

[0117] Next, use Figure 9 The flow of processing by the learning device 30 will be described. Figure 9 This is a flowchart showing the flow of processing by the learning device 30 according to the first embodiment.

[0118] like Figure 9 As shown, in step S41, the second data acquisition unit 31 acquires the defrost temperature θ from the temperature sensor 14 and the measured value Pm of the frost-melting time P from the control unit 15. While the defrost temperature θ and the measured value Pm of the frost-melting time P are acquired simultaneously, this is not limiting. As long as the second data acquisition unit 31 can associate the defrost temperature θ with the measured value Pm of the frost-melting time P, the second data acquisition unit 31 may also acquire the defrost temperature θ and the measured value Pm of the frost-melting time P at different times.

[0119] Next, in step S42, the model generation unit 32 performs a learning process using the learning data. The learning data includes the measured value Pm of the defrost temperature θ and the frost-melting time P. More specifically, the learning data is created based on the combination of the measured value Pm of the defrost temperature θ and the frost-melting time P acquired by the second data acquisition unit 31. Using this learning data, the model generation unit 32 learns the frost-melting time P relative to the defrost temperature θ through so-called coached learning, generating a learned model.

[0120] Next, in step S43 , the learned model storage unit 33 stores the learned model generated by the model generation unit 32 .

[0121] In addition, in the first embodiment, the case where the learning algorithm used by the model generation unit 32 is taught learning is described, but the present invention is not limited to this. The learning algorithm used by the model generation unit 32 can also apply reinforcement learning, untaught learning, semi-taught learning, etc. in addition to taught learning.

[0122] Furthermore, when multiple air conditioners 101 exist in the same or different areas, the model generation unit 32 can learn the frost-melting time P based on the learning data generated for these multiple air conditioners 101. Specifically, the second data acquisition unit 31 can acquire learning data from multiple air conditioners 101 operating in the same area so that the model generation unit 32 can learn the frost-melting time P. Alternatively, the second data acquisition unit 31 can acquire learning data collected from multiple air conditioners 101 operating independently in different areas. Furthermore, the second data acquisition unit 31 can also add air conditioners 101 from which learning data is to be collected to the list mid-process, or conversely, remove air conditioners 101 from the list mid-process. Furthermore, the learning device 30 that has learned the frost-melting time P for one air conditioner 101 can be applied to a different air conditioner 101, and the frost-melting time P can be relearned and updated for the different air conditioner 101.

[0123] In addition, as a learning algorithm used in the model generation unit 32, deep learning that extracts learning feature quantities themselves can also be used, and machine learning can also be performed according to other well-known methods such as genetic programming, functional logic programming, support vector machines, etc.

[0124] As described above, the estimation device 20 according to Embodiment 1 includes a first data acquisition unit 21 for acquiring defrost temperature information from the air conditioner 101. Furthermore, the estimation unit 22 of the estimation device 20 uses a learned model and, based on the defrost temperature information acquired by the first data acquisition unit 21, calculates an estimated value Pest of the time required for frost melting P relative to the defrost temperature information. The air conditioner 101 compares the estimated value Pest of the time required for frost melting P calculated by the estimation device 20 with the threshold value Th. If the estimated value Pest is less than the threshold value Th, the air conditioner 101 does not start the defrost operation, even if a certain time has passed since the start of the heating operation and the defrost temperature θ is below a specified value. This prevents unnecessary defrost operations from being performed. As a result, the power consumption of the air conditioner 101 is reduced, and a decrease in user comfort caused by unnecessary defrost operations can be prevented.

[0125] On the other hand, when the estimated value Pest of the frost melting time P obtained by the estimation device 20 is equal to or greater than the threshold value Th, when a certain time ( Figure 2 If S1 in the above example is “Yes”), the defrost temperature θ is below the specified value ( Figure 2 (S2 in the embodiment is "Yes"), the air conditioner 101 performs a defrosting operation. Thus, defrosting can be performed at an appropriate timing when a defrosting operation is required.

[0126] Furthermore, the learning device 30 according to Embodiment 1 includes a second data acquisition unit 31 that acquires learning data created based on a combination of defrost temperature information and the measured value Pm of the frost-melting time required P. Furthermore, the model generation unit 32 of the learning device 30 uses the learning data to perform learning so that the estimated value Pest of the frost-melting time required P relative to the defrost temperature information approaches the measured value Pm of the frost-melting time required P. Thus, the model generation unit 32 generates a learned model that calculates the estimated value Pest of the frost-melting time required P based on the defrost temperature information of the air conditioner 101. Because the learning device 30 generates the learned model using learning data including the measured value Pm of the frost-melting time required P, it is possible to accurately calculate the estimated value Pest that is close to the measured value Pm of the frost-melting time required P.

[0127] As described above, according to the estimation device 20 and the learning device 30 according to the first embodiment, by obtaining the estimated value Pest of the frost melting time P as the amount of frost, the start time of the defrosting operation can be appropriately determined, thereby achieving improved energy saving.

[0128] Description of reference numerals:

[0129] 1…indoor unit; 2…outdoor unit; 3…compressor; 4…four-way valve; 5…indoor heat exchanger; 6…outdoor heat exchanger; 7…expansion valve; 8…indoor air blower; 9…outdoor air blower; 10…temperature sensor; 11…temperature sensor; 12…temperature sensor; 13…temperature sensor; 14…temperature sensor; 15…control unit; 16…current measuring device; 17…humidity sensor; 18…illuminance sensor; 20…inference device; 21…first data acquisition unit; 22…inference unit; 30…learning device; 31…second data acquisition unit; 32…model generation unit; 33…learned model storage unit; 34…neural network; 40…external device; 100…refrigerant circuit; 101…air conditioning unit.

Claims

1. An estimation device for obtaining an estimated value of a time required for frost melting relative to defrost temperature information indicating a temperature of an outdoor heat exchanger provided in an outdoor unit of an air conditioner or a state of change in the temperature, characterized in that: The frost melting time is a period during which the refrigerant flowing through the refrigerant pipe of the outdoor heat exchanger melts the frost attached to the outdoor heat exchanger and the temperature of the outdoor heat exchanger is stable within a first range. The inference device comprises: a first data acquisition unit that acquires the defrost temperature information of the air conditioner; and an estimating unit that uses a learned model for estimating a time required for frost melting based on the defrost temperature information and obtains an estimated value of the time required for frost melting relative to the defrost temperature information based on the defrost temperature information acquired by the first data acquiring unit; Whether to start defrost operation control in the air-conditioning apparatus is determined based on the estimated value.

2. The inference device according to claim 1, wherein: The defrost temperature information includes the defrost temperature representing the current temperature of the refrigerant piping of the outdoor heat exchanger, or at least any one of the average value, cumulative value, integral value, maximum value or minimum value of the defrost temperature in the time interval from the end of the last defrost operation to the present.

3. The inference device according to claim 1 or 2, characterized in that The learned model is a model that receives the defrost temperature information as input and outputs an estimated value of the time required for frost melting relative to the defrost temperature information. The learned model is a model obtained by learning based on learning data created based on a combination of the defrost temperature information and the actual value of the time required for frost melting so that the estimated value of the time required for frost melting relative to the defrost temperature information approaches the actual value of the time required for frost melting.

4. The inference device according to claim 1 or 2, characterized in that The defrost temperature information includes a slope indicating a ratio of a change in the defrost temperature information to a change in time.

5. The inference device according to claim 1 or 2, characterized in that The first data acquisition unit acquires the defrost temperature information and humidity information indicating the humidity of an external air environment where the outdoor unit of the air conditioner is installed. The estimating unit uses the learned model for estimating the time required for frost melting based on the defrost temperature information and the humidity information, and obtains an estimated value of the time required for frost melting based on the defrost temperature information and the humidity information acquired by the first data acquiring unit.

6. The inference device according to claim 1 or 2, characterized in that The first data acquisition unit acquires the defrost temperature information and current information indicating a current value used by an outdoor air blower included in an outdoor unit of the air conditioner. The estimating unit uses a learned model that estimates the time required for frost melting based on the defrost temperature information and the current information, and obtains an estimated value of the time required for frost melting based on the defrost temperature information and the current information acquired by the first data acquiring unit.

7. The inference device according to claim 1 or 2, characterized in that The first data acquisition unit acquires the defrost temperature information and outside air temperature information indicating the outside air temperature of an outside air environment where the outdoor unit of the air conditioner is installed. The estimating unit uses a learned model for estimating the time required for frost melting based on the defrost temperature information and the outside air temperature information to obtain an estimated value of the time required for frost melting based on the defrost temperature information and the outside air temperature information acquired by the first data acquiring unit.

8. The inference device according to claim 1 or 2, characterized in that The first data acquisition unit acquires the defrost temperature information and illuminance information indicating the illuminance of an external air environment where the outdoor unit of the air conditioner is installed. The estimating unit uses a learned model for estimating the time required for frost melting from the defrost temperature information and the illuminance information to obtain an estimated value of the time required for frost melting based on the defrost temperature information and the illuminance information acquired by the first data acquiring unit.

9. A learning device that generates a learned model for obtaining an estimated value of a time required for frost melting relative to defrost temperature information, the defrost temperature information indicating a temperature of an outdoor heat exchanger included in an outdoor unit of an air conditioner or a state of change in the temperature. The learning device is characterized in that The frost melting time is a period during which the refrigerant flowing through the refrigerant pipe of the outdoor heat exchanger melts the frost attached to the outdoor heat exchanger and the temperature of the outdoor heat exchanger is stable within a first range. The learning device comprises: a second data acquisition unit that acquires learning data created based on a combination of the defrost temperature information and an actual measurement value of the time required for frost melting; and The model generating unit generates a learned model for obtaining an estimated value of the time required for frost melting based on the defrost temperature information of the air conditioner by learning using the learning data in such a manner that the estimated value of the time required for frost melting relative to the defrost temperature information approaches the actual value of the time required for frost melting. Whether to start defrost operation control in the air-conditioning apparatus is determined based on the estimated value.

10. The learning device according to claim 9, characterized in that The learning data includes the defrost temperature information in a plurality of environments.

11. An air conditioning device, characterized in that: have: A refrigerant circuit, comprising a compressor, a condenser, an expansion valve, and an evaporator; a temperature sensor, detecting the evaporation temperature of the evaporator; a receiving unit that receives the estimated value of the time required for frost to melt from the estimating device according to claim 1 or 2; and Control Department, The control unit starts the defrosting operation of the evaporator when (a) the estimated value of the time required for the frost to melt is greater than a threshold value, (b) a certain time has passed since the start of the heating operation, and (c) the evaporation temperature detected by the temperature sensor is less than a specified value.

Images