Inference device, air conditioning system, and learning device

Abstract

To provide an inference device, etc. that can perform at least one of the start and stop of operation of an air conditioning device on a condition suitable for an individual user without troubling the user to wear a wearable device.SOLUTION: An inference device 200 comprises an inference data acquisition unit 210 for acquiring inference data generated from pulse wave data of a human body detected by a Doppler sensor without contact with the human body, and an inference unit 220 for inferring operation contents of an air conditioning device 2 from the inference data. The inferred operation contents of the air conditioning device 2 include one or both of the start and stop of operation of the air conditioning device 2.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present disclosure relates to an inference device, an air conditioning system, and a learning device. [Background technology]

[0002] An electronic device is known that includes a wearable device having a blood flow sensor that detects blood flow in the human body, a communication unit that communicates with an air conditioner that can adjust at least the air temperature, and a control unit that, when the communication unit receives the detection result from the blood flow sensor, controls the communication unit to send a control signal that controls the air conditioner based on the detection result (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-148361 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the technology disclosed in Patent Document 1 requires the user to wear a wearable device, which is inconvenient.

[0005] The present disclosure has been made to solve these problems, and its purpose is to provide an inference device, an air conditioning system, and a learning device that can at least start and stop the operation of an air conditioner under conditions that are suited to an individual user, without the user having to go through the hassle of wearing a wearable device. [Means for solving the problem]

[0006] The inference device of the present disclosure includes a data acquisition unit that acquires inference data generated from pulse wave data of the human body detected by a Doppler sensor without contacting the human body, and an inference unit that infers the operating details of the air conditioning device from the inference data, and the operating details include one or both of starting and stopping the operation of the air conditioning device.

[0007] The air conditioning system of the present disclosure comprises the above-mentioned inference device and the air conditioning device, wherein the air conditioning device comprises an air conditioning means for conditioning the air in the target space and a control unit for controlling the air conditioning means, and the control unit starts and / or stops the operation of the air conditioning means in accordance with the operation content output from the inference unit.

[0008] The learning device according to the present disclosure includes a data acquisition unit that acquires learning data and a model generation unit that uses the learning data to generate a trained model, wherein the learning data includes data generated from pulse wave data of a human body detected non-contact with the human body by a Doppler sensor and operation details of an air conditioning device, and the trained model is a model for inferring the operation details of the air conditioning device from data generated from the pulse wave data, and the operation details include one or both of starting and stopping operation of the air conditioning device. [Effects of the Invention]

[0009] The inference device, air conditioning system, and learning device disclosed herein have the advantage of allowing the user to start and / or stop the operation of an air conditioning device under conditions that are suitable for the individual user, without the user having to go through the hassle of wearing a wearable device. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram schematically illustrating the overall configuration of an air conditioning system according to a first embodiment. [Figure 2] 1 is a diagram showing the configuration of an air conditioning device included in an air conditioning system according to a first embodiment. [Figure 3]2 is a diagram showing the configuration of an indoor unit control unit of the air conditioner according to the first embodiment. FIG. [Figure 4] 2 is a diagram showing the configuration of a measuring device provided in the air conditioning system according to the first embodiment. FIG. [Figure 5] 1 is a block diagram showing a configuration of an inference device according to a first embodiment. [Figure 6] 4 is a flowchart showing an example of the operation of the inference device according to the first embodiment. [Figure 7] 1 is a block diagram showing a configuration of a learning device according to a first embodiment. [Figure 8] FIG. 2 is a diagram illustrating an example of a neural network in the learning device according to the first embodiment. [Figure 9] 4 is a flowchart showing an example of the operation of the learning device according to the first embodiment. [Figure 10] FIG. 10 is a block diagram showing the configuration of another example of the learning device according to the first embodiment. [Figure 11] 10 is a flowchart showing an example of operation of another example of the learning device according to the first embodiment. FIG. [Figure 12] 1 is a diagram illustrating an example of a configuration for realizing the functions of an inference device and a learning device according to a first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] Embodiments for implementing an inference device, an air conditioning system, and a learning device according to the present disclosure will be described with reference to the accompanying drawings. In each drawing, identical or corresponding parts are designated by the same reference numerals, and redundant explanations are appropriately simplified or omitted. For convenience, the following description may express the positional relationship of each structure based on the illustrated state. Note that the present disclosure is not limited to the following embodiments, and any combination of the embodiments, any modification of any component of each embodiment, or any omission of any component of each embodiment are possible within the scope of the present disclosure.

[0012] Embodiment 1 A first embodiment of the present disclosure will be described with reference to FIGS. 1 to 12. FIG. 1 is a diagram schematically illustrating the overall configuration of an air conditioning system. FIG. 2 is a diagram illustrating the configuration of an air conditioner provided in the air conditioning system. FIG. 3 is a diagram illustrating the configuration of an indoor unit control unit of the air conditioner. FIG. 4 is a diagram illustrating the configuration of a measurement device provided in the air conditioning system. FIG. 5 is a block diagram illustrating the configuration of an inference device. FIG. 6 is a flow diagram illustrating an example of the operation of the inference device. FIG. 7 is a block diagram illustrating the configuration of a learning device. FIG. 8 is a diagram illustrating an example of a neural network in the learning device. FIG. 9 is a flow diagram illustrating an example of the operation of the learning device. FIG. 10 is a block diagram illustrating the configuration of another example of the learning device. FIG. 11 is a flow diagram illustrating an example of the operation of another example of the learning device. FIG. 12 is a diagram illustrating an example of a configuration for implementing the functions of the inference device and the learning device.

[0013] The air conditioning system 1 according to the present disclosure can be applied to various buildings such as a detached house, an apartment building, an office building, etc. In the following embodiment, an example in which the air conditioning system is applied to a detached house will be described.

[0014] FIG. 1 is a diagram showing the overall configuration of an air conditioning system 1 according to this embodiment. The air conditioning system 1 is an example of an air conditioning system according to the present disclosure. The air conditioning system 1 includes an air conditioner 2, a measuring device 3, a router 4, and a server 5. As shown in FIG. 1, the air conditioner 2 is installed in a house. The house is, for example, a typical detached residential building. The air conditioner 2 is a heat pump type air conditioning facility that uses, for example, HFC (hydrofluorocarbon) or the like as a refrigerant. The air conditioner 2 is equipped with a vapor compression type refrigeration cycle. The air conditioner 2 operates by obtaining power from a commercial power source, a power generation facility, a power storage facility, or the like (not shown).

[0015] The air conditioner 2 and the measurement device 3 are communicatively connected to a router 4. The communication between the air conditioner 2 and the router 4 may be wireless or wired. Similarly, the communication between the measurement device 3 and the router 4 may be wireless or wired. The router 4 is connected to a communication network such as the Internet. The air conditioner 2 and the measurement device 3 can communicate with a server 5 via the router 4 and the communication network. The server 5 is installed, for example, outside the house.

[0016] 2 is a diagram showing an air conditioner 2 according to this embodiment. The air conditioner 2 is equipment that conditions an indoor space 71, which is a space to be air-conditioned. Air conditioning in this disclosure refers to adjusting the temperature, humidity, cleanliness, airflow, etc. of the air in the space to be air-conditioned. That is, air conditioning in this disclosure may specifically include, for example, heating, cooling, dehumidification, humidification, air purification, etc.

[0017] The air conditioner 2 includes an outdoor unit 11, an indoor unit 13, and a remote controller 55. The outdoor unit 11 is provided in an outdoor space 72 outside the house. The indoor unit 13 is provided in an indoor space 71 inside the house. The remote controller 55 is operated by a user. The outdoor unit 11 and the indoor unit 13 are connected via a refrigerant piping 61 and a communication line 63. The refrigerant piping 61 is a piping through which a refrigerant flows. In addition, various signals are transferred between the outdoor unit 11 and the indoor unit 13 via the communication line 63.

[0018] The air conditioner 2 conditions the indoor space 71 by blowing out conditioned air from the indoor unit 13. Specifically, for example, the air conditioner 2 cools the indoor space 71 by blowing out cold air from the indoor unit 13. The air conditioner 2 also heats the indoor space 71 by blowing out warm air from the indoor unit 13.

[0019] The outdoor unit 11 includes a compressor 21, a four-way valve 22, an outdoor heat exchanger 23, an expansion valve 24, an outdoor blower 26, and an outdoor unit control unit 51. The indoor unit 13 includes an indoor heat exchanger 25, an indoor blower 27, and an indoor unit control unit 53. The refrigerant piping 61 connects the compressor 21, the four-way valve 22, the outdoor heat exchanger 23, the expansion valve 24, and the indoor heat exchanger 25 in a ring shape. This forms a refrigeration cycle.

[0020] The compressor 21 compresses the refrigerant and circulates it within the refrigerant pipe 61. Specifically, the compressor 21 compresses a low-temperature, low-pressure refrigerant and discharges the high-temperature, high-pressure refrigerant to the four-way valve 22. The compressor 21 is equipped with an inverter circuit that can change the operating capacity according to the drive frequency. The operating capacity is the amount of refrigerant that the compressor 21 delivers per unit time. The compressor 21 changes the operating capacity according to instructions from the outdoor unit control unit 51.

[0021] The four-way valve 22 is installed on the discharge side of the compressor 21. The four-way valve 22 switches the flow direction of the refrigerant in the refrigerant pipe 61 depending on whether the air conditioner 2 is operating in cooling or dehumidifying operation, or in heating operation.

[0022] The outdoor heat exchanger 23 is a first heat exchanger that exchanges heat between the refrigerant flowing through the refrigerant piping 61 and the air in the outdoor space 72 that is outside the space to be air-conditioned. The outdoor blower 26 is provided near the outdoor heat exchanger 23. The outdoor blower 26 is a first blower that sends air from the outdoor space 72 to the outdoor heat exchanger 23. When the outdoor blower 26 starts blowing air, negative pressure is generated inside the outdoor unit 11, and air from the outdoor space 72 is sucked in. The sucked air is supplied to the outdoor heat exchanger 23, where it exchanges heat with the cold and hot heat supplied by the refrigerant flowing through the refrigerant piping 61, and is then blown out into the outdoor space 72.

[0023] The expansion valve 24 is installed between the outdoor heat exchanger 23 and the indoor heat exchanger 25, and reduces the pressure of the refrigerant flowing through the refrigerant pipe 61 to expand it. The expansion valve 24 is, for example, an electronic expansion valve whose opening degree is changeable. The expansion valve 24 changes its opening degree in accordance with instructions from the outdoor unit control unit 51 to adjust the pressure of the refrigerant.

[0024] The indoor heat exchanger 25 is a second heat exchanger that exchanges heat between the refrigerant flowing through the refrigerant piping 61 and the air in the indoor space 71. The indoor blower 27 is provided beside the indoor heat exchanger 25. The indoor blower 27 is a second blower that sends air in the indoor space 71 to the indoor heat exchanger 25. When the indoor blower 27 starts blowing air, negative pressure is generated inside the indoor unit 13, and air in the indoor space 71 is sucked in. The sucked air is supplied to the indoor heat exchanger 25, where heat is exchanged between the air and the cold or hot air supplied from the refrigerant flowing through the refrigerant piping 61, and then blown out into the indoor space 71.

[0025] The air that has undergone heat exchange in the indoor heat exchanger 25 is supplied to the indoor space 71 as conditioned air. This conditions the indoor space 71. The greater the amount of heat exchanged between the refrigerant and the air in the indoor heat exchanger 25, the higher the air conditioning capacity of the air conditioner 2. Hereinafter, the air conditioning capacity during cooling will be referred to as the cooling capacity, and the air conditioning capacity during heating will be referred to as the heating capacity.

[0026] The compressor 21, four-way valve 22, outdoor heat exchanger 23, expansion valve 24, and outdoor blower 26 in the outdoor unit 11, and the indoor heat exchanger 25 and indoor blower 27 in the indoor unit 13 are collectively referred to as an air conditioning unit. The air conditioning unit functions as air conditioning means for conditioning the indoor space 71.

[0027] The indoor unit 13 further includes a first temperature detection unit 41, a second temperature detection unit 44, a third temperature detection unit 45, a humidity detection unit 42, and an infrared detection unit 43. The first temperature detection unit 41 and the humidity detection unit 42 detect the temperature and humidity of the air drawn into the indoor unit 13. The second temperature detection unit 44 and the third temperature detection unit 45 detect the temperature of the refrigerant in the indoor heat exchanger 25. For example, during cooling, the indoor heat exchanger 25 functions as an evaporator, and the second temperature detection unit 44 detects the evaporation temperature Te, and the third temperature detection unit 45 detects the evaporator outlet temperature Teo. The infrared detection unit 43 detects the surface temperatures of the floor, people, etc. in the indoor space 71.

[0028] The air conditioner 2 also includes other detectors (not shown) in addition to those described above. Specifically, the air conditioner 2 includes a discharge pressure detector installed on the discharge side of the compressor 21 and detecting the pressure of the refrigerant discharged from the compressor 21, a suction pressure detector installed on the suction side of the compressor 21 and detecting the pressure of the refrigerant drawn into the compressor 21, a discharge temperature detector installed on the discharge side of the compressor 21 and detecting the temperature of the refrigerant drawn into the compressor 21, a suction temperature detector installed on the suction side of the compressor 21 and detecting the temperature of the refrigerant drawn into the compressor 21, an outdoor temperature detector that detects the temperature of the outdoor air, an outdoor heat exchanger temperature detector that detects the temperature of the outdoor heat exchanger 23, and the like.

[0029] The detection results by each detection unit, including the first temperature detection unit 41, are transmitted to the indoor unit control unit 53. The indoor unit control unit 53 transmits the received detection results to the outdoor unit control unit 51 via the communication line 63.

[0030] The indoor unit control unit 53 controls the operation of the indoor unit 13. As shown in Fig. 3, the indoor unit control unit 53 includes a control unit 101, a storage unit 102, and a communication unit 104. These units are connected via a bus 109.

[0031] The control unit 101 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). The CPU is also called a central processing unit, central arithmetic unit, processor, microprocessor, microcomputer, or DSP (Digital Signal Processor). In the control unit 101, the CPU reads out programs and data stored in the ROM and uses the RAM as a work area to perform overall control of the outdoor unit control unit 51.

[0032] The storage unit 102 is a non-volatile semiconductor memory such as a flash memory, an EPROM (Erasable Programmable ROM), or an EEPROM (Electrically Erasable Programmable ROM), and serves as a so-called secondary storage device or auxiliary storage device. The storage unit 102 stores programs and data used by the control unit 101 to perform various processes, as well as data generated or acquired by the control unit 101 as a result of performing various processes.

[0033] The communication unit 104 is an interface for communicating with the outdoor unit control unit 51 and the remote controller 55. The communication unit 104 receives operation information accepted by the remote controller 55 from the remote controller 55. The communication unit 104 transmits notification information to the remote controller 55 for notifying the user. The communication unit 104 also transmits operation commands for the indoor unit 13 to the outdoor unit control unit 51 via the communication line 63. The communication unit 104 receives status information indicating the status of the outdoor unit 11 from the outdoor unit control unit 51 via the communication line 63.

[0034] The outdoor unit control unit 51 includes a CPU, ROM, RAM, a communication interface, and a readable / writable nonvolatile semiconductor memory, all of which are not shown. In the outdoor unit control unit 51, the CPU controls the operation of the outdoor unit 11 by executing a control program stored in the ROM while using the RAM as a work memory.

[0035] The indoor unit control unit 53 is connected to the outdoor unit control unit 51 via a communication line 63, which is a wired, wireless, or other communication medium. The outdoor unit control unit 51 cooperates with the outdoor unit control unit 51 by sending and receiving various signals via the communication line 63, and controls the entire air conditioner 2. In this way, the indoor unit control unit 53 functions as a control device that controls the air conditioner 2.

[0036] The outdoor unit control unit 51 and the indoor unit control unit 53 control the operation of the air conditioner 2 based on the detection results of the first temperature detection unit 41, the infrared detection unit 43, and other detection units, and on setting information for the air conditioner 2 set by the user. Specifically, the outdoor unit control unit 51 controls the drive frequency of the compressor 21, the switching of the four-way valve 22, the rotation speed of the outdoor blower 26, and the opening degree of the expansion valve 24. The indoor unit control unit 53 controls the rotation speed of the indoor blower 27. Alternatively, the outdoor unit control unit 51 may control the rotation speed of the indoor blower 27. Alternatively, the indoor unit control unit 53 may control the drive frequency of the compressor 21, the switching of the four-way valve 22, the rotation speed of the outdoor blower 26, or the opening degree of the expansion valve 24. In this way, the outdoor unit control unit 51 and the indoor unit control unit 53 output various operation commands to various devices in response to operation commands given to the air conditioner 2.

[0037] A remote controller 55 is disposed in the indoor space 71. The remote controller 55 transmits and receives various signals to and from the indoor unit control unit 53 provided in the indoor unit 13. A user of the air conditioner 2 operates the remote controller 55 to input operation commands to the air conditioner 2. Examples of operation commands include a command to switch between operation and stop, a command to switch operation modes (auto, cooling, dehumidification, heating, etc.), a command to switch target temperatures, a command to switch target humidity, a command to switch airflow volume, a command to switch airflow direction, or a command to switch timers. The air conditioner 2 starts operation in accordance with the input operation command.

[0038] Next, the "cooling" and "dehumidification" operation modes will be described. When the outdoor unit control unit 51 receives an operation command for "cooling" or "dehumidification," it switches the flow path of the four-way valve 22 so that the refrigerant discharged from the compressor 21 flows into the outdoor heat exchanger 23, opens the expansion valve 24, and drives the compressor 21 and the outdoor blower 26. Furthermore, when the indoor unit control unit 53 receives an operation command for "cooling" or "dehumidification," it drives the indoor blower 27.

[0039] When the compressor 21 is driven, the refrigerant discharged from the compressor 21 passes through the four-way valve 22 and flows into the outdoor heat exchanger 23. The refrigerant that flows into the outdoor heat exchanger 23 exchanges heat with outdoor air drawn in from the outdoor space 72, condenses, and liquefies, and then flows into the expansion valve 24. The refrigerant that flows into the expansion valve 24 is decompressed by the expansion valve 24 and then flows into the indoor heat exchanger 25. The refrigerant that flows into the indoor heat exchanger 25 exchanges heat with indoor air drawn in from the indoor space 71 and evaporates. The refrigerant then passes through the four-way valve 22 and is sucked back into the compressor 21. As the refrigerant flows in this manner, the indoor air drawn in from the indoor space 71 is cooled by the indoor heat exchanger 25. The amount of heat exchanged between the refrigerant and the indoor air in the indoor heat exchanger 25 is called the cooling capacity. Of the cooling capacity, the portion that lowers the temperature of the indoor air is called the sensible heat capacity, and the portion that removes moisture from the indoor air (dehumidifies) is called the latent heat capacity.

[0040] Dehumidification operations include "weak cooling dehumidification," "partial cooling dehumidification," and "reheat dehumidification." The "weak cooling dehumidification" operation mode is a first dehumidification mode with a lower cooling capacity and a higher dehumidification capacity than "cooling." When the control unit 101 receives an operation command for "weak cooling dehumidification," it circulates the refrigerant in the same direction as "cooling." Then, the control unit 101 reduces the rotation speed of the indoor blower 27 compared to "cooling." In other words, the control unit 101 reduces the amount of air sent by the indoor blower 27 to the indoor heat exchanger 25 in "weak cooling dehumidification" compared to "cooling."

[0041] Generally, the greater the airflow rate of the indoor blower 27, the higher the evaporation temperature of the refrigerant in the indoor heat exchanger 25, resulting in a more efficient refrigeration cycle. Therefore, in "cooling," the air conditioner 2 can save energy by operating at a large airflow rate that does not cause noise. In contrast, in "weak cooling and dehumidification," the control unit 101 reduces the airflow rate of the indoor blower 27 compared to "cooling," thereby lowering the evaporation temperature of the refrigerant. This reduces the sensible heat capacity and increases the latent heat capacity of the indoor heat exchanger 25. Therefore, the sensible heat ratio decreases. As a result, the room temperature Ti is less likely to decrease and the indoor humidity RHi is more likely to decrease in "weak cooling and dehumidification" than in "cooling."

[0042] The "partial cooling dehumidification" operation mode is a second dehumidification mode in which the evaporation temperature of the refrigerant on the inlet side of the indoor heat exchanger 25 is lowered below the dew point temperature of the air, and the degree of superheat of the refrigerant on the outlet side of the indoor heat exchanger 25 is increased. When the control unit 101 receives an operation command for "partial temperature dehumidification," it circulates the refrigerant in the same direction as in "cooling." Then, the control unit 101 controls the opening of the expansion valve 24 to an opening such that the evaporation temperature of the refrigerant at the inlet through which the refrigerant flows into the indoor heat exchanger 25 is lower than the dew point temperature of the air.

[0043] In "cooling" and "weak cooling and dehumidification," the control unit 101 controls the opening of the expansion valve 24 to such an extent that the refrigerant becomes a saturated gas at the refrigerant outlet of the indoor heat exchanger 25, that is, so that the degree of superheat near the refrigerant outlet of the indoor heat exchanger 25 approaches zero. This allows the latent heat capacity of the air conditioner 2 to be output efficiently. In contrast, in "partial cooling and dehumidification," the control unit 101 controls the opening of the expansion valve 24 so that the evaporation temperature of the refrigerant near the refrigerant inlet of the indoor heat exchanger 25 becomes lower than the dew point temperature of the air drawn into the indoor heat exchanger 25.

[0044] Specifically, in "partial cooling dehumidification," the control unit 101 narrows the opening of the expansion valve 24 more than in "cooling" and "weak cooling dehumidification." This lowers the evaporation temperature of the refrigerant near the inlet of the indoor heat exchanger 25, and since much of the refrigerant evaporates near the inlet of the indoor heat exchanger 25, the degree of superheat near the outlet of the indoor heat exchanger 25 increases. As a result, it is possible to dehumidify the air at a low temperature on the inlet side of the indoor heat exchanger 25, and the air is not cooled too much on the outlet side. In "partial cooling dehumidification," the room temperature Ti is less likely to decrease and the indoor humidity RHi is more likely to decrease than in "weak cooling dehumidification."

[0045] The "reheat dehumidification" operating mode is a third dehumidification mode that reduces humidity while suppressing a decrease in temperature in the indoor space 71. When this "reheat dehumidification" operating mode is possible, the air conditioner 2 has two heat exchangers as the indoor heat exchanger 25 and an expansion valve between these heat exchangers. When the control unit 101 receives a "reheat dehumidification" operating command, it circulates the refrigerant in the same direction as "cooling." The control unit 101 then appropriately closes an expansion valve (not shown) between the two heat exchangers in the indoor heat exchanger 25. By narrowing the opening of this expansion valve, the heat exchanger located upstream of the expansion valve functions as a condenser that condenses the refrigerant and heats the air. Meanwhile, the heat exchanger located downstream of the expansion valve functions as an evaporator that evaporates the refrigerant and reduces the humidity of the air. Because the "reheat dehumidification" operating mode heats the air while reducing humidity, the room temperature Ti is less likely to decrease and the indoor humidity RHi is more likely to decrease in "reheat dehumidification" than in other dehumidification modes.

[0046] Next, the "heating" operation mode will be described. When the outdoor unit control unit 51 receives a "heating" operation command, it switches the flow path of the four-way valve 22 so that the refrigerant discharged from the compressor 21 flows into the indoor heat exchanger 25, opens the expansion valve 24, and drives the compressor 21 and the outdoor blower 26. Furthermore, when the indoor unit control unit 53 receives a "heating" operation command, it drives the indoor blower 27.

[0047] When the compressor 21 is driven, the refrigerant discharged from the compressor 21 passes through the four-way valve 22 and flows into the indoor heat exchanger 25. The refrigerant that flows into the indoor heat exchanger 25 exchanges heat with indoor air drawn from the indoor space 71, condenses, and liquefies, and then flows into the expansion valve 24. The refrigerant that flows into the expansion valve 24 is decompressed by the expansion valve 24 and then flows into the outdoor heat exchanger 23. The refrigerant that flows into the outdoor heat exchanger 23 exchanges heat with outdoor air drawn from the outdoor space 72 and evaporates, then passes through the four-way valve 22 and is sucked back into the compressor 21. As the refrigerant flows in this manner, the indoor air drawn from the indoor space 71 is heated by the indoor heat exchanger 25. The amount of heat exchanged between the refrigerant and the indoor air in the indoor heat exchanger 25 is called the heating capacity.

[0048] The measuring device 3 is placed on furniture such as a table in the indoor space 71. As shown in FIG. 4, the measuring device 3 includes an environmental sensor 31 and a biological sensor 32. The environmental sensor 31 detects the temperature and humidity of the air at the location of the measuring device 3. The biological sensor 32 detects biological information of a human body without contacting the human body. In this embodiment, the biological sensor 32 is a microwave Doppler sensor. A microwave Doppler sensor is a sensor that can detect the moving speed of an object using frequency changes in reflected waves due to the Doppler effect. The biological sensor 32, which is a microwave Doppler sensor, detects minute movements on the surface of the human body due to heartbeats, thereby detecting the human body's pulse wave as biological information. Note that the measuring device 3 only needs to include at least the biological sensor 32; the environmental sensor 31 is not necessary. In this case, the temperature sensor and humidity sensor built into the air conditioner 2 can be used.

[0049] The measurement device 3 further includes an inference data generation unit 33. The inference data generation unit 33 generates inference data from the detection data of the environmental sensor 31 and the biosensor 32. The inference data generated by the inference data generation unit 33 includes, for example, human pulse wave data detected by a microwave Doppler sensor, which is the biosensor 32. The inference data generated by the inference data generation unit 33 may also include room temperature data detected by the environmental sensor 31, i.e., room temperature data of the indoor space 71 in which the air conditioner 2 is installed.

[0050] The inference data generation unit 33 may calculate a stress index of the human body from the human pulse wave, and include the calculated stress index data in the inference data instead of the pulse wave data. The human body maintains a constant body temperature through a balance between heat generation due to metabolism and heat dissipation to the surrounding environment; the less heat dissipation relative to heat generation, the higher the body temperature and the more heat dissipation, and the more the body temperature drops and the more the person feels cold. Heat dissipation includes heat dissipation due to the temperature difference between the skin and the air, and heat dissipation due to moisture evaporation through sweating and breathing. When the air temperature is high, the amount of sweat increases, increasing heat dissipation due to moisture evaporation, but when the air humidity is high, sweat does not evaporate easily, which is thought to cause discomfort.

[0051] The autonomic nervous system is divided into sympathetic and parasympathetic nerves, and the skin blood vessels and sweat glands involved in thermoregulation as mentioned above are primarily controlled by the sympathetic nerve. Furthermore, the heart rate also increases when the sympathetic nerve is excited. To capture the activity of the sympathetic and parasympathetic nerves, the fluctuations in the heartbeat intervals of pulse wave data are frequency analyzed to obtain the power spectral density, and the value obtained by integrating the low frequency band (hereinafter referred to as LF; Low Frequency) and the value obtained by integrating the high frequency band (hereinafter referred to as HF; High Frequency) are used. HF reflects respiratory fluctuations and serves as an index of parasympathetic nerve activity. LF reflects blood pressure fluctuations and is affected by both the sympathetic and parasympathetic nerves, so LF / HF is used as an index of sympathetic nerve activity. Therefore, the inference data generation unit 33 calculates LF / HF from the human pulse wave and uses it as a stress index. An increase in LF / HF is considered to indicate an increase in stress.

[0052] The measurement device 3 transmits the inference data generated by the inference data generation unit 33 to the server 5 via the router 4 and the communication network. The server 5 receives the inference data transmitted from the measurement device 3. The server 5 includes an inference device 200 and a trained model storage unit 300. An example of the configuration of the inference device 200 is shown in FIG. 5.

[0053] The trained model storage unit 300 stores trained models. The trained models stored in the trained model storage unit 300 are used to infer the operation details of the air conditioner 2 from the inference data. The operation details of the air conditioner 2, which are the inference results, include at least one or both of starting and stopping the operation of the air conditioner 2. The trained models stored in the trained model storage unit 300 are generated, for example, by a learning device 400, which will be described later. The trained model storage unit 300 may be provided in another server device, etc., that is capable of communicating with the server 5.

[0054] The inference data acquisition unit 210 of the inference device 200 acquires input data for the inference device 200. The input data is inference data generated by the inference data generation unit 33 of the measurement device 3. In other words, the input data includes data generated from human pulse wave data detected by a Doppler sensor without contacting the human body. As described above in detail, the input data may include the human pulse wave data itself. Alternatively, the input data may include a stress index calculated from the human pulse wave data. Furthermore, the input data may further include room temperature data for the indoor space 71 in which the air conditioner 2 is installed.

[0055] The inference unit 220 of the inference device 200 uses the learned model stored in the learned model storage unit 300 to infer the operation details of the air conditioner 2 from the input data acquired by the inference data acquisition unit 210. As described above, the operation details of the air conditioner 2, which are the inference results, include at least one or both of starting and stopping the operation of the air conditioner 2. The inference unit 220 can output the operation details of the air conditioner 2 inferred from the input data by inputting the input data acquired by the inference data acquisition unit 210 into the learned model. In this way, the inference unit 220 outputs the operation details of the air conditioner 2 from the input data acquired by the inference data acquisition unit 210, using the learned model for inferring the operation details of the air conditioner 2 from the input data.

[0056] The operation details of the air conditioner 2 output from the inference unit 220 are transmitted from the inference device 200 to the air conditioner 2. The outdoor unit control unit 51 and indoor unit control unit 53 of the air conditioner 2 control the air conditioning means based on the operation details of the air conditioner 2 transmitted from the inference device 200. That is, the outdoor unit control unit 51 and the indoor unit control unit 53 start and / or stop the operation of the air conditioning means in accordance with the operation details of the air conditioner 2 transmitted from the inference device 200. Note that the operation details of the air conditioner 2 inferred by the inference device 200 may further include at least one of the set temperature, set humidity, and air volume of the air conditioner 2.

[0057] Next, an example of the operation of the air conditioning system 1 including the inference device 200, air conditioner 2, and measurement device 3 configured as described above will be described with reference to the flow diagram in Fig. 6. First, in step S11, the inference data acquisition unit 210 of the inference device 200 acquires input data. In the following step S12, the inference unit 220 of the inference device 200 inputs the input data acquired in step S11 into the trained model. In further following step S13, the inference unit 220 outputs operation content data of the air conditioner 2, which is the inference result obtained by inputting the input data into the trained model in step S12.

[0058] The data output from the inference unit 220 in step S13 is input to the air conditioner 2. After step S13, the air conditioner 2 performs the process of step S14. In step S14, the outdoor unit control unit 51 and the indoor unit control unit 53 of the air conditioner 2 control the air conditioning means based on the operation details of the air conditioner 2. In other words, they issue a command to start or stop operation of the air conditioner 2. When the process of step S14 is completed, the series of operations ends.

[0059] Next, a configuration example of a learning device 400 according to this embodiment will be described with reference to Fig. 7. The learning device 400 according to this embodiment learns the operation details of the air conditioner 2. The learned operation details of the air conditioner 2 are used to control the air conditioner 2, as described above. As shown in the figure, the learning device 400 includes a learning data acquisition unit 410 and a model generation unit 420.

[0060] The learning data acquisition unit 410 acquires learning data. The learning data includes data generated from human pulse wave data detected by the biosensor 32 of the measurement device 3, i.e., the Doppler sensor, without contacting the human body, and the operation details of the air conditioner 2. The operation details of the air conditioner 2 are so-called correct answer data. The learning data is data in which the data generated from the human pulse wave data detected by the Doppler sensor without contacting the human body and the operation details of the air conditioner 2, which are the correct answer data, are associated with each other.

[0061] Here, the data generated from the human pulse wave data detected by the Doppler sensor without contacting the human body may be the human pulse wave data itself or the stress index data described above. In addition, the operation details of the air conditioner 2, which are the correct data, include one or both of starting and stopping the operation of the air conditioner 2.

[0062] The model generation unit 420 learns the operation details of the air conditioner 2 from the above-mentioned learning data created based on a combination of data generated from the human pulse wave data and the operation details of the air conditioner 2. In other words, the model generation unit 420 uses the learning data acquired by the learning data acquisition unit 410 to generate a learned model that infers the optimal operation details of the air conditioner 2 from the data generated from the human pulse wave data.

[0063] The learning algorithm used by the model generation unit 420 can be a known algorithm such as supervised learning. As an example, a case where a neural network is applied will be described. The model generation unit 420 learns the operation details of the air conditioner 2, for example, by so-called supervised learning according to a neural network model. Here, supervised learning refers to a method in which pairs of input and result (label) data are provided to the learning device 400, and the learning device 400 learns the features of the learning data and infers the result from the input.

[0064] A neural network is composed of an input layer consisting of multiple neurons, an intermediate layer (hidden layer) consisting of multiple neurons, and an output layer consisting of multiple neurons. The intermediate layer may be one layer or two or more layers.

[0065] For example, in a three-layer neural network as shown in Figure 8, when multiple inputs are input to the input layer (X1-X3), the values ​​are multiplied by weight W1 (w11-w16) and input to the middle layer (Y1-Y2), and the result is further multiplied by weight W2 (w21-w26) and output from the output layer (Z1-Z3). This output result changes depending on the values ​​of weights W1 and W2.

[0066] In the present disclosure, the neural network learns the operation details of the air conditioner 2 by so-called supervised learning based on the above-mentioned learning data created based on a combination of data generated from human pulse wave data acquired by the learning data acquisition unit 410 and the operation details of the air conditioner 2. In other words, the neural network learns by inputting data generated from human pulse wave data into the input layer and adjusting the weights W1 and W2 so that the results output from the output layer approach the operation details of the air conditioner 2.

[0067] The model generation unit 420 generates and outputs a trained model by performing the above-described learning. The trained model storage unit 300 stores the trained model output from the model generation unit 420. As described above, the trained model storage unit 300 may be provided in, for example, the server 5 or the learning device 400.

[0068] Next, an example of the operation of the learning device 400 configured as described above will be described with reference to the flow diagram of Fig. 9. First, in step S21, the learning data acquisition unit 410 acquires learning data. Note that the data generated from the human pulse wave data included in the learning data and the data on the operation details of the air conditioner 2 are acquired simultaneously, but it is sufficient if these data are input in an associated manner, and the data generated from the human pulse wave data and the data on the operation details of the air conditioner 2 may be acquired at different times.

[0069] After step S21, the learning device 400 then performs the process of step S22. In step S22, the model generation unit 420 uses the learning data acquired in step S21 to learn the operation details of the air conditioner 2 through so-called supervised learning, and generates a trained model. In the following step S23, the trained model storage unit 300 stores the trained model generated in step S22. When the process of step S23 is completed, the series of operations ends.

[0070] The model generation unit 420 may learn the operation details of the air conditioners 2 according to learning data created for multiple air conditioners 2. The model generation unit 420 may acquire learning data from multiple air conditioners 2 used in the same area, or may learn the operation details of the air conditioners 2 using learning data collected from multiple air conditioners 2 operating independently in different areas. It is also possible to add or remove air conditioners 2 from which learning data is collected as the target. Furthermore, the learning device 400 that has learned the operation details of an air conditioner 2 for a certain air conditioner 2 may be applied to another air conditioner 2, and the operation details of the air conditioner 2 for that other air conditioner 2 may be re-learned and updated.

[0071] Furthermore, the learning algorithm used in the model generation unit 420 can be deep learning, which learns to extract the feature quantities themselves, or other known methods such as genetic programming or functional logic programming can be used to perform machine learning. Furthermore, the learning device 400 and the inference device 200 are used to learn the operation details of the air conditioner 2, which are used to control the air conditioner 2, but they may be separate devices from the air conditioner 2, or may be built into the air conditioner 2.

[0072] The learning data may further include room temperature data of the space in which the air conditioner 2 is installed. In this case, the learned model generated by the model generation unit 420 is a model for inferring the operation details of the air conditioner 2 from the room temperature data and data generated from the pulse wave data. Furthermore, the operation details of the air conditioner 2 may further include at least one of the set temperature, set humidity, and air volume of the air conditioner 2.

[0073] Next, another example of a learning device 400 according to this embodiment will be described with reference to Figures 10 and 11. As shown in Figure 10, the learning device 400 includes a learning data acquisition unit 410 and a model generation unit 420.

[0074] The learning data acquisition unit 410 acquires learning data. The learning data includes data generated from human pulse wave data detected by the biosensor 32 of the measurement device 3, i.e., the Doppler sensor, without contacting the human body, and operation details of the air conditioner 2. The learning data includes data generated from human pulse wave data detected by the Doppler sensor without contacting the human body, and operation details of the air conditioner 2 under conditions indicated by the data generated from the human pulse wave data.

[0075] Here, the data generated from the human pulse wave data detected by the Doppler sensor without contacting the human body may be the human pulse wave data itself or the stress index data described above. Furthermore, the operation details of the air conditioner 2 include one or both of starting and stopping the operation of the air conditioner 2.

[0076] The model generation unit 420 learns the operation details of the air conditioner 2 from the above-mentioned learning data including data generated from the human pulse wave data and the operation details of the air conditioner 2 under the conditions indicated by the data generated from the human pulse wave data. In other words, the model generation unit 420 uses the learning data acquired by the learning data acquisition unit 410 to generate a learned model that infers the optimal operation details of the air conditioner 2 from the data generated from the human pulse wave data.

[0077] The learning algorithm used by the model generation unit can be a known algorithm such as reinforcement learning. As an example, we will explain the application of reinforcement learning. In reinforcement learning, an agent (acting subject) in a certain environment observes the current state (environmental parameters) and decides on the action to take. The environment changes dynamically depending on the agent's actions, and the agent is given a reward according to the environmental changes. The agent repeats this process and learns the action course that will obtain the most reward through a series of actions. Q-learning and TD-learning are known as representative reinforcement learning methods. For example, in the case of Q-learning, the general update formula for the action value function Q(s, a) is expressed as [Equation 1].

[0078]

number

[0079] In [Equation 1], st represents the state of the environment at time t, and at represents the action at time t. The action at changes the state to st+1. rt+1 represents the reward obtained due to the change in state, γ represents the discount rate, and α represents the learning coefficient. Note that γ is in the range of 0<γ≦1, and α is in the range of 0<α≦1. Driving or stopping becomes the action at, biological information and environmental information become the state st, and the best action at in the state st at time t is learned.

[0080] The update formula expressed by [Equation 1] increases the action value Q if the action value Q of the action a with the highest Q value at time t+1 is greater than the action value Q of the action a executed at time t, and decreases the action value Q in the opposite case. In other words, the action value function Q(s, a) is updated so that the action value Q of the action a at time t approaches the best action value at time t+1. As a result, the best action value in a certain environment is propagated sequentially to the action values ​​in previous environments.

[0081] When generating a trained model by reinforcement learning in this way, the model generation unit 420 includes a reward calculation unit 421 and a function update unit 422. The reward calculation unit 421 calculates a reward based on data generated from human pulse wave data, for example, a stress index. The reward calculation unit 421 calculates a reward r based on the stress index. For example, if the stress index decreases, the reward r is increased (for example, a reward of "1" is given), and on the other hand, if the stress index increases, the reward r is decreased (for example, a reward of "-1" is given).

[0082] The function update unit 422 updates the function for determining the operation details of the air conditioner 2 in accordance with the reward calculated by the reward calculation unit 421, and outputs the updated function to the learned model storage unit 300. For example, in the case of Q-learning, the action value function Q(st,at) expressed by [Equation 1] is used as a function for calculating operation or stop. The learning described above is executed repeatedly. The learned model storage unit 300 stores the action value function Q(st,at) updated by the function update unit 422, i.e., the learned model.

[0083] Next, the learning process performed by the learning device 400 will be described with reference to Fig. 11. In step S31, the learning data acquisition unit 410 acquires, as learning data, data generated from human pulse wave data and the operation details of the air conditioner 2 under the conditions indicated by the data generated from the human pulse wave data.

[0084] In step S32, the model generation unit 420 calculates a reward based on the data generated from the human pulse wave data and the operation details of the air conditioner 2 under the conditions indicated by the data generated from the human pulse wave data. Specifically, the reward calculation unit 421 determines whether to increase the reward (step S33) or decrease the reward (step S34) based on the stress index.

[0085] If the remuneration calculation unit 421 determines that the remuneration should be increased, it increases the remuneration in step S33. On the other hand, if the remuneration calculation unit 421 determines that the remuneration should be decreased, it decreases the remuneration in step S34.

[0086] In step S35, the function update unit 422 updates the action value function Q(st,at) stored in the trained model storage unit 300 and expressed by [Equation 1], based on the reward calculated by the reward calculation unit 421.

[0087] The learning device 400 repeatedly executes the above steps S31 to S35, and stores the generated action-value function Q(st,at) as a learned model.

[0088] With the inference device 200, air conditioning system 1, and learning device 400 configured as described above, the operation details of the air conditioning device 2 can be inferred from inference data generated from pulse wave data of a human body detected by a Doppler sensor without contacting the human body, and the user can start and / or stop the operation of the air conditioning device under conditions suitable for the individual user without the hassle of wearing a wearable device.

[0089] FIG. 12 is a diagram showing an example of a configuration for realizing the respective functions of the inference device 200 and the learning device 400 in this embodiment. The respective functions of the inference device 200 and the learning device 400 are realized, for example, by a processing circuit. The processing circuit may include a processor 501 and a memory 502. The processing circuit may be dedicated hardware 503. A portion of the processing circuit may be formed as dedicated hardware 503, and the processing circuit may further include a processor 501 and a memory 502. In the example shown in the figure, a portion of the processing circuit is formed as dedicated hardware 503. Furthermore, in the example shown in the figure, the processing circuit further includes a processor 501 and a memory 502.

[0090] The processing circuit, part of which is at least one dedicated hardware 503, may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof. When the processing circuit includes at least one processor 501 and at least one memory 502, the functions of the inference device 200 and the learning device 400 are realized by software, firmware, or a combination of software and firmware.

[0091] The software and firmware are written as programs and stored in memory 502. Processor 501 realizes the functions of each unit by reading and executing the programs stored in memory 502. Processor 501 is also called a CPU (Central Processing Unit), central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, or DSP. Examples of memory 502 include non-volatile or volatile semiconductor memory such as RAM, ROM, flash memory, EPROM, and EEPROM, as well as magnetic disks, flexible disks, optical disks, compact disks, minidisks, and DVDs.

[0092] In this way, the processing circuits of inference device 200 and learning device 400 can use hardware, software, firmware, or a combination of these to realize the respective functions of inference device 200 and learning device 400. When the processing circuits of inference device 200 and learning device 400 each include at least processor 501 and memory 502, processor 501 executes a program stored in memory 502 in inference device 200 and learning device 400, and the hardware and software of inference device 200 and learning device 400 work together to realize the functions of the various parts of inference device 200 and learning device 400.

[0093] In the present disclosure, the embodiments may be combined in any manner without departing from the spirit of the present disclosure. Examples of various aspects of the present disclosure are summarized below as appendices. (Appendix 1) a data acquisition unit that acquires inference data generated from pulse wave data of the human body detected by a Doppler sensor without contacting the human body; an inference unit that infers the operation details of the air conditioner from the inference data, An inference device in which the operation content includes one or both of starting and stopping the operation of the air conditioner. (Appendix 2) An inference device as described in Appendix 1, wherein the data for inference includes stress index data of the human body generated from the pulse wave data. (Appendix 3) An inference device according to claim 1 or 2, wherein the data for inference further includes room temperature data of a space in which the air conditioning device is installed. (Appendix 4) 4. The inference device according to claim 1, wherein the operation details further include at least one of the set temperature, set humidity, and air volume of the air conditioner. (Appendix 5) An inference device according to any one of Supplementary Note 1 to Supplementary Note 4; the air conditioning device, The air conditioning device an air conditioning unit for conditioning the air in the target space; a control unit that controls the air conditioning means, The control unit starts and / or stops the operation of the air conditioning unit in accordance with the operation content output from the inference unit. (Appendix 6) a data acquisition unit that acquires learning data; a model generation unit that generates a trained model using the training data, the learning data includes data generated from pulse wave data of the human body detected by a Doppler sensor without contacting the human body, and operation details of the air conditioner; the trained model is a model for inferring the operation details of the air conditioner from data generated from the pulse wave data, The operation content of the learning device includes one or both of starting and stopping the operation of the air conditioner. (Appendix 7) 7. An inference device as described in Appendix 6, wherein the learning data includes stress index data of the human body generated from the pulse wave data. (Appendix 8) The learning data further includes room temperature data of a space in which the air conditioning device is installed, The inference device according to claim 6 or 7, wherein the trained model is a model for inferring the operation details of the air conditioning device from data generated from the pulse wave data and the room temperature data. [Explanation of symbols]

[0094] 1. Air conditioning system 2 Air conditioner 3. Measuring equipment 4. Router 5 Server 11 Outdoor unit 13 Indoor unit 21 Compressor 22 Four-way valve 23 Outdoor heat exchanger 24 Expansion valve 25 Indoor heat exchanger 26 Outdoor blower 27 Indoor fan 31 Environmental Sensors 32 Biometric Sensors 33 Inference data generation unit 41 First temperature detection unit 42 Humidity detection unit 43 Infrared detector 44 Second temperature detection unit 45 Third temperature detection unit 51 Outdoor unit control unit 53 Indoor unit control unit 55 Remote Controller 61 Refrigerant piping 63 Communication Line 71 Indoor space 72 Outdoor Space 101 Control section 102 Storage section 104 Communications Department 109 Bus 200 Reasoning device 210 Inference data acquisition unit 220 Reasoning Department 300 Trained model memory unit 400 Learning Device 410 Learning data acquisition unit 420 Model Generation Unit 421 Remuneration Calculation Department 422 Function Update Section 501 processor 502 memory 503 Dedicated Hardware

Claims

1. a data acquisition unit that acquires inference data generated from pulse wave data of the human body detected by a Doppler sensor without contacting the human body; an inference unit that infers the operation details of the air conditioner from the inference data, An inference device in which the operation content includes one or both of starting and stopping the operation of the air conditioner.

2. The inference device according to claim 1 , wherein the data for inference includes stress index data of the human body generated from the pulse wave data.

3. The inference device according to claim 1 , wherein the inference data further includes room temperature data of a space in which the air conditioner is installed.

4. The inference device according to claim 1 , wherein the operation details further include at least one of a set temperature, a set humidity, and an air volume of the air conditioner.

5. An inference device according to any one of claims 1 to 4; the air conditioning device, The air conditioning device an air conditioning means for conditioning the air in the target space; a control unit that controls the air conditioning means, The control unit starts and / or stops the operation of the air conditioning unit in accordance with the operation content output from the inference unit.

6. a data acquisition unit that acquires learning data; a model generation unit that generates a trained model using the training data, the learning data includes data generated from pulse wave data of the human body detected by a Doppler sensor without contacting the human body, and operation details of the air conditioner; the trained model is a model for inferring the operation details of the air conditioner from data generated from the pulse wave data, The operation content of the learning device includes one or both of starting and stopping the operation of the air conditioner.

7. The learning device according to claim 6 , wherein the learning data includes stress index data of the human body generated from the pulse wave data.

8. The learning data further includes room temperature data of a space in which the air conditioning device is installed, The learning device according to claim 6 or 7, wherein the trained model is a model for inferring the operation details of the air conditioning device from data generated from the pulse wave data and the room temperature data.

Images