JP7575936B2 - Refrigeration System - Google Patents

Description

An embodiment of the present invention relates to a refrigerator system.

To keep food at low temperatures in the refrigerator or freezer, refrigerators use a refrigerant to exchange heat. In this case, frost can form on the cooling device, reducing cooling performance.

JP 2016-191497 A Japanese Patent Application Publication No. 06-273033 JP 2007-225155 A

The problem that this invention aims to solve is to provide a refrigerator system that can perform appropriate defrosting control to remove frost that has built up on the refrigerator's cooler.

The refrigerator system of the embodiment includes a refrigerator main body equipped with a cooler, and a prediction unit that predicts the color of ice that has formed on the cooler based on a learned learning model trained with a plurality of learning information for learning a learning model that receives as input information at least one of food ingredient information indicating the food ingredients inside the refrigerator at each time since the last defrosting, the temperature inside the refrigerator, the door opening/closing history, the outside air temperature, and the outside air humidity, and that receives as output information the color of ice that has formed on the cooler corresponding to the input information, and a control unit that controls a defrosting operation of the cooler so as to enhance defrosting capacity when the prediction unit predicts that the color of the ice is transparent compared to when the prediction unit predicts that the color of the ice is white .

FIG. 1 is a diagram showing an example of a configuration of a refrigerator system according to an embodiment. FIG. 1 is a diagram illustrating an example of a configuration of a refrigerator main body according to an embodiment. 3 is a cross-sectional view of the refrigerator body of one embodiment taken along line F2-F2 in FIG. 2. FIG. 1 is a diagram illustrating an example of a configuration of a refrigeration cycle device according to an embodiment. FIG. 2 is a block diagram showing a part of a configuration related to control of a refrigerator main body according to an embodiment. FIG. 2 is a diagram illustrating an example of a configuration of a control device according to an embodiment. FIG. 4 is a diagram showing an example of teacher data according to an embodiment; FIG. 4 is a diagram showing a processing flow of the refrigerator system according to one embodiment. FIG. 13 is a diagram illustrating an example of a configuration of a control device according to a sixth modified example of an embodiment.

Below, a refrigerator system according to an embodiment will be described with reference to the drawings. In the following description, components having the same or similar functions will be given the same reference numerals. Duplicate descriptions of these components may be omitted. In this specification, left and right are defined based on the direction in which the refrigerator is viewed from a user standing in front of the refrigerator, and the side closer to the user standing in front of the refrigerator as viewed from the refrigerator is defined as the "front" and the side furthest from the user is defined as the "rear."

"Based on XX" means "based on at least XX" and may include cases where it is based on other elements in addition to XX. "Based on XX" is not limited to the direct use of XX, but may also include cases where it is based on XX that has been calculated or processed. "XX or YY" is not limited to either XX or YY, but may include both XX and YY. "XX" and "YY" are any element (e.g. any information).

<Embodiment>
(Configuration of Refrigerator System)
Fig. 1 is a diagram showing an example of the configuration of a refrigerator system 200 according to an embodiment. The refrigerator system 200 includes a refrigerator main body 1 and a control device 2. The control device 2 is, for example, a server connected to the refrigerator main body 1 via a network. The refrigerator system 200 is a system that can perform appropriate defrosting control for removing frost adhering to a cooler of the refrigerator main body 1 by controlling the refrigerator main body 1 with the control device 2.

(Configuration of refrigerator body)
A refrigerator body 1 of one embodiment will be described with reference to Fig. 2 to Fig. 8. Fig. 2 is a front view showing the refrigerator body 1. Fig. 3 is a cross-sectional view taken along line F2-F2 of the refrigerator body 1 shown in Fig. 2. As shown in Fig. 2 and Fig. 3, the refrigerator body 1 includes, for example, a housing 10, a plurality of doors 20, an operation panel 30, a flow path forming component 40, a cooling unit 50, and a control unit 100.

The housing 10 has an upper wall 11, a lower wall 12, left and right side walls 13, 14, and a rear wall 15. The upper wall 11 and the lower wall 12 extend approximately horizontally. The left and right side walls 13, 14 rise upward from the left and right ends of the lower wall 12 and are connected to the left and right ends of the upper wall 11. The rear wall 15 rises upward from the rear end of the lower wall 12 and is connected to the rear end of the upper wall 11.

As shown in FIG. 3, the housing 10 has, for example, an inner box 10a, an outer box 10b, and a heat insulating section 10c. The inner box 10a is a member that forms the inner surface of the housing 10. The outer box 10b is a member that forms the outer surface of the housing 10. The outer box 10b is formed to be slightly larger than the inner box 10a, and is disposed outside the inner box 10a. Between the inner box 10a and the outer box 10b, a heat insulating section 10c including a foamed heat insulating material such as urethane foam is provided.

The housing 10 has a plurality of storage compartments 17 inside. The plurality of storage compartments 17 include, for example, a refrigerator compartment 17A, a chilled compartment 17B, a vegetable compartment 17C, an ice-making compartment 17D, a small freezer compartment 17E, and a main freezer compartment 17F. For example, the refrigerator compartment 17A is located at the top, the vegetable compartment 17C is located below the refrigerator compartment 17A, the ice-making compartment 17D and the small freezer compartment 17E are located below the vegetable compartment 17C, and the main freezer compartment 17F is located below the ice-making compartment 17D and the small freezer compartment 17E. However, the arrangement of the storage compartments 17 is not limited to the above example. The housing 10 has an opening on the front side of each storage compartment 17 that allows food to be put in and taken out of each storage compartment 17.

Chilled compartment 17B is provided below a portion of refrigerator compartment 17A. Chilled compartment 17B is at least partially separated from refrigerator compartment 17A by, for example, a shelf or a wall. Chilled compartment 17B is cooled to a lower temperature than refrigerator compartment 17A because it is located lower than refrigerator compartment 17A and cold air can easily flow in there, and because it is located closer to a refrigeration cooler 61 (described below) (an example of a cooler) than refrigerator compartment 17A.

The housing 10 has a first partition wall 18 and a second partition wall 19. The first partition wall 18 and the second partition wall 19 are partition walls that are approximately horizontal. The first partition wall 18 is located between the refrigerator compartment 17A (chilled compartment 17B) and the vegetable compartment 17C, and separates the refrigerator compartment 17A (chilled compartment 17B) and the vegetable compartment 17C. On the other hand, the second partition wall 19 is located between the vegetable compartment 17C and the ice making compartment 17D and the small freezer compartment 17E, and separates the vegetable compartment 17C from the ice making compartment 17D and the small freezer compartment 17E. The second partition wall 19 contains, for example, a foam insulation material and has thermal insulation properties. The first partition wall 18 is formed of, for example, a synthetic resin, and has lower thermal insulation properties than the second partition wall 19.

The openings of the storage compartments 17 are closed by the doors 20. The doors 20 include, for example, left and right refrigerator compartment doors 20Aa and 20Ab that close the opening of the refrigerator compartment 17A, a chilled compartment door 20B that closes the opening of the chilled compartment 17B, a vegetable compartment door 20C that closes the opening of the vegetable compartment 17C, an ice-making compartment door 20D that closes the opening of the ice-making compartment 17D, a small freezer compartment door 20E that closes the opening of the small freezer compartment 17E, and a main freezer compartment door 20F that closes the opening of the main freezer compartment 17F. The chilled compartment door 20B is located further inside the refrigerator compartment 17A than the refrigerator compartment doors 20Aa and 20Ab (see FIG. 3). The chilled compartment door 20B may be, for example, a type that is integral with the chilled compartment container 36B (described below) and is pulled forward integrally with the chilled compartment container 36B, or a type that is opened by rotating around a hinge that is provided adjacent to the chilled compartment 17B.

As shown in FIG. 2, the operation panel 30 is provided on the door 20 (left refrigerator door 20Aa), for example. The operation panel 30 accepts user input operations related to changing the set temperature zone of the refrigerator body 1 and changing the operation mode. The operation panel 30 includes, for example, a button 31 for starting and stopping the control mode of "quick chill" or "thawing" described later, and a button 32 for starting and stopping the control mode of "special chill". The "quick chill" control mode is a control mode in which the air temperature of the chilled compartment 17B is lowered more quickly than in the control mode of chilled with normal settings. In the "special chill" control mode, the chilled compartment 17B is cooled in a low temperature zone for a period of time and the chilled compartment 17B is cooled in a high temperature zone for a period of time that is repeated in sequence. However, the operation of selecting the start and stop of these various control modes may be input from a user's mobile terminal or smart speaker via a network instead of the operation panel 30.

As shown in FIG. 3, multiple shelves 35 are provided in the refrigerator compartment 17A. Multiple containers 36 include a chilled compartment container 36B provided in the chilled compartment 17B, first and second vegetable compartment containers 36Ca, 36Cb provided in the vegetable compartment 17C, an ice compartment container (not shown) provided in the ice compartment 17D, a small freezer container 36E provided in the small freezer compartment 17E, and first and second main freezer containers 36Fa, 36Fb provided in the main freezer compartment 17F. Here, "container" also includes a shallow member such as a tray.

The flow path forming part 40 is disposed within the housing 10. The flow path forming part 40 includes a first duct part 41 and a second duct part 42.

The first duct part 41 is provided along the rear wall 15 of the housing 10 and extends vertically. The first duct part 41 extends, for example, from the rear of the lower end of the vegetable compartment 17C to the rear of the upper end of the refrigerator compartment 17A. Between the first duct part 41 and the rear wall 15 of the housing 10, a first duct space D1 is formed, which is a passage through which cold air (air) flows. The first duct part 41 has a plurality of refrigerator compartment cold air outlets 41a, a chilled compartment cold air outlet 41b, and a cold air return port 41c. The plurality of refrigerator compartment cold air outlets 41a are provided at a plurality of height positions above the chilled compartment 17B. The plurality of refrigerator compartment cold air outlets 41a open to the refrigerator compartment 17A. The cold air flowing through the first duct space D1 is blown out from the refrigerator compartment cold air outlet 41a to the refrigerator compartment 17A. The chilled compartment cold air outlet 41b opens into the chilled compartment 17B. The cold air flowing through the first duct space D1 is blown out from the chilled compartment cold air outlet 41b into the chilled compartment 17B. The cold air return port 41c opens into the vegetable compartment 17C. The cold air that has passed through the vegetable compartment 17C returns to the first duct space D1 from the cold air return port 41c.

The second duct part 42 is provided along the rear wall 15 of the housing 10 and extends vertically. The second duct part 42 extends, for example, from the rear of the main freezer 17F to the rear of the upper ends of the ice-making chamber 17D and the small freezer 17E. Between the second duct part 42 and the rear wall 15 of the housing 10, a second duct space D2 is formed, which is a passage through which cold air (air) flows. The second duct part 42 has a cold air outlet 42a and a cold air return port 42b. The cold air outlet 42a opens into the ice-making chamber 17D and the small freezer 17E. The cold air flowing through the second duct space D2 is blown out from the cold air outlet 42a into the ice-making chamber 17D and the small freezer 17E. The cold air return port 42b opens into the main freezer 17F. The cold air that passes through the main freezer compartment 17F returns to the second duct space D2 through the cold air return port 42b.

The cooling section (cooling unit) 50 cools a plurality of storage compartments 17 (refrigeration compartment 17A, chilled compartment 17B, vegetable compartment 17C, ice making compartment 17D, small freezer compartment 17E, and main freezer compartment 17F). The cooling section 50 includes, for example, a first cooling module 60, a second cooling module 70, a compressor 80, and a refrigeration cycle device 90 (FIG. 4). Here, "cooling" means a state in which a refrigerant is supplied from the compressor 80 to a cooler (a refrigeration cooler 61 (an example of a cooler) or a freezer cooler 71 (an example of a cooler) described later) corresponding to each storage compartment 17. However, "cooling" is not limited to the case in which the refrigeration fan 62 or the freezer fan 72 described later is driven. For example, "cooling" also includes a case where the refrigerant is sent from the compressor 80 to the refrigeration cooler 61 while the refrigeration fan 62 is stopped, and the temperature of the refrigeration compartment 17B is reduced by heat transfer between the refrigeration cooler 61 and the refrigeration compartment 17B.

The first cooling module 60 includes, for example, a refrigeration cooler 61 and a refrigeration fan 62. The refrigeration cooler 61 is arranged in the first duct space D1. The refrigeration cooler 61 is supplied with refrigerant compressed by the compressor 80 and cools the cold air flowing through the first duct space D1. The refrigeration cooler 61 is arranged, for example, at a height corresponding to the chilled chamber 17B.

The refrigeration fan 62 is provided, for example, in the cold air return port 41c of the first duct part 41. When the refrigeration fan 62 is driven, the air in the vegetable compartment 17C flows into the first duct space D1 from the cold air return port 41c. The air that flows into the first duct space D1 flows upward in the first duct space D1 and is cooled by the refrigeration cooler 61. The cold air cooled by the refrigeration cooler 61 is blown out from the multiple refrigerator compartment cold air outlets 41a into the refrigerator compartment 17A and from the chilled compartment cold air outlet 41b into the chilled compartment 17B. The cold air blown out to the refrigerator compartment 17A and the chilled compartment 17B flows through the refrigerator compartment 17A and the chilled compartment 17B, respectively, and then returns to the cold air return port 41c again via, for example, the vegetable compartment 17C. This causes the cold air flowing through the refrigerator compartment 17A, chilled compartment 17B, and vegetable compartment 17C to circulate within the refrigerator body 1, cooling the refrigerator compartment 17A, chilled compartment 17B, and vegetable compartment 17C.

On the other hand, the second cooling module 70 includes, for example, a refrigeration cooler 71 and a refrigeration fan 72. The refrigeration cooler 71 is arranged in the second duct space D2. The refrigeration cooler 71 is supplied with refrigerant compressed by the compressor 80 and cools the cold air flowing through the second duct space D2.

The freezing fan 72 is provided, for example, in the cold air return port 42b of the second duct part 42. When the freezing fan 72 is driven, the air in the main freezing chamber 17F flows into the second duct space D2 from the cold air return port 42b. The air that flows into the second duct space D2 flows upward in the second duct space D2 and is cooled by the freezing cooler 71. The cold air cooled by the freezing cooler 71 flows into the ice making chamber 17D, the small freezing chamber 17E, and the main freezing chamber 17F from the cold air outlet 42a. The cold air that flows into the ice making chamber 17D and the small freezing chamber 17E flows through the ice making chamber 17D and the small freezing chamber 17E, respectively, and then returns to the cold air return port 42b again via the main freezing chamber 17F. This causes the cold air flowing through the ice making compartment 17D, the small freezer compartment 17E, and the main freezer compartment 17F to circulate within the refrigerator body 1, cooling the ice making compartment 17D, the small freezer compartment 17E, and the main freezer compartment 17F.

The compressor 80 is provided, for example, in a machine room at the bottom of the refrigerator body 1. The compressor 80 compresses the refrigerant gas used to cool the storage chamber 17. The refrigerant gas compressed by the compressor 80 is sent to the refrigeration cooler 61 and the freezing cooler 71 via the condenser 91 and the like.

(Refrigeration cycle device)
4 is a diagram showing the configuration of a refrigeration cycle device 90. The refrigeration cycle device 90 includes, in the order of refrigerant flow, a condenser 91, a dryer 92, a three-way valve 93, and capillary tubes 94 and 95. More specifically, the condenser 91 and the dryer 92 are connected in this order to the high-pressure discharge port of the compressor 80 via a connecting pipe 96. The three-way valve 93 is connected to the discharge side of the dryer 92. The three-way valve 93 has one inlet to which the dryer 92 is connected, and two outlets.

One of the two outlets of the three-way valve 93 is connected in turn to a capillary tube 94 on the refrigeration side and a refrigeration cooler 61. The refrigeration cooler 61 is connected to the compressor 80 via a refrigeration side suction pipe 97, which is a connecting pipe. The other of the two outlets of the three-way valve 93 is connected in turn to a capillary tube 95 on the freezing side and a refrigeration cooler 71. The refrigeration cooler 71 is connected to the compressor 80 via a refrigeration side suction pipe 98, which is a connecting pipe. A check valve 99 is provided between the refrigeration cooler 71 and the compressor 80 to prevent the refrigerant from the refrigeration cooler 61 from flowing back to the refrigeration cooler 71.

The refrigerant circulating through the refrigeration cycle device 90 is compressed by the compressor 80 to become a high-temperature, high-pressure gaseous refrigerant, which flows through the flow path A. This gaseous refrigerant is dissipated heat by the condenser 91 to become a medium-temperature, high-pressure liquid refrigerant. The liquid refrigerant, from which impurities such as dirt and moisture have been removed by passing through the dryer 92, enters the capillary tube 94 (or capillary tube 95) while being throttled and controlled by the three-way valve 93. At this time, the medium-temperature, high-pressure liquid refrigerant in the capillary tube 94 (or capillary tube 95) is decompressed while exchanging heat with the refrigerant in the refrigeration side suction pipe 97 (or the freezing side suction pipe 98). The decompressed refrigerant then evaporates while passing through the refrigeration cooler 61 (or the freezing cooler 71), thereby cooling the refrigeration cooler 61 (or the freezing cooler 71).

The low-temperature, low-pressure gaseous refrigerant then flows into the refrigeration side suction pipe 97 (or the freezing side suction pipe 98). The temperature of the refrigerant gas immediately after it flows into the refrigeration side suction pipe 97 (or the freezing side suction pipe 98) is low, at around -10°C. As this refrigerant gas passes through the refrigeration side suction pipe 97 (or the freezing side suction pipe 98), it exchanges heat with the refrigerant in the capillary tube 94 (or the capillary tube 95), and is eventually heated to about room temperature. This refrigerant gas is then sucked back into the compressor 80, completing the circulation of the refrigerant.

FIG. 5 is a block diagram showing a part of the configuration related to the control of the refrigerator main body 1. The control unit 100 may be configured by a computer including a microcomputer and a timer (not shown) that measures time. The control unit 100 may be a software function unit realized by executing a computer program by one or more hardware processors such as a CPU (Central Processing Unit), or may be realized by hardware (e.g., a circuit unit; circuitry) such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a PLD (Programmable Logic Device). Note that all or a part of the control unit 100 may be realized by a combination of a software function unit and hardware. In the above-mentioned refrigeration cycle device 90, the three-way valve 93 is controlled by the control unit 100 (see FIG. 5) to select, for example, one of the flow paths B and C. The flow path B is a flow path that supplies the refrigerant to the refrigeration cooler 61. The flow path C is a flow path that supplies the refrigerant to the refrigeration cooler 71. These two flow paths B and C join at a joining point D. The refrigerant flows from the joining point D in the direction of the arrow E and returns to the compressor 80.

The control unit 100 controls the entire refrigerator body 1. The control unit 100 is connected to the refrigeration fan 62, the freezer fan 72, the compressor 80, the three-way valve 93, the operation panel 30, the outside air temperature sensor 111, the refrigerator compartment temperature sensor 112, the chilled compartment temperature sensor 113, the refrigerator compartment door switch 114, the chilled compartment door switch 115, the interior camera 116, and the memory unit 117. Each of the refrigeration fan 62, the freezer fan 72, the compressor 80, the three-way valve 93, the operation panel 30, the outside air temperature sensor 111, the refrigerator compartment temperature sensor 112, the chilled compartment temperature sensor 113, the refrigerator compartment door switch 114, the chilled compartment door switch 115, and the interior camera 116 is an example of a detection unit, and detects the state of the cooler or the refrigerator body 1.

The outside air temperature sensor 111 is exposed, for example, on the surface of the refrigerator body 1 and detects the air temperature outside the refrigerator body 1. The refrigerator compartment temperature sensor 112 is exposed, for example, in the refrigerator compartment 17A and detects the air temperature in the refrigerator compartment 17A.

The chilled compartment temperature sensor 113 is exposed to the chilled compartment 17B and detects the temperature related to the chilled compartment 17B. The "temperature related to the chilled compartment" is, for example, one or more of the temperature of the food stored in the chilled compartment 17B (for example, the surface temperature of the food), the air temperature in the chilled compartment 17B, or the temperature of the member stored in the chilled compartment 17B (for example, the chilled compartment container 36B). For example, the chilled compartment temperature sensor 113 is a contact-type temperature sensor that detects the temperature of the food stored in the chilled compartment 17B. However, instead of providing the chilled compartment temperature sensor 113, for example, the control unit 101a may estimate the air temperature of the chilled compartment 17B based on the detection result of the refrigerator compartment temperature sensor 112 and the correspondence relationship between the air temperature of the refrigerator compartment 17A and the air temperature of the chilled compartment 17B that is obtained in advance. For convenience, the air temperature in the refrigerator compartment 17A will be referred to as the "refrigerator compartment temperature," the air temperature in the chilled compartment 17B as the "chilled compartment temperature," and the air temperature in the main freezer compartment 17F as the "freezer compartment temperature."

The refrigerator compartment door switch 114 is provided, for example, between the housing 10 and the refrigerator compartment doors 20Aa and 20Ab, and detects whether the refrigerator compartment doors 20Aa and 20Ab are open or closed. The chilled compartment door switch 115 is provided, for example, adjacent to the chilled compartment door 20B, and detects whether the chilled compartment door 20B is open or closed.

The interior camera 116 is installed, for example, in the refrigerator compartment 17A and detects the entry and removal of food from the refrigerator compartment 17A or chilled compartment 17B.

The storage unit 117 is realized by a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), a ROM (read-only memory), a RAM (random access memory), or the like. The storage unit 117 stores information necessary for implementing each control mode such as "quick chill", "thaw", and "special chill" (e.g., setting information for implementation time and standby time, etc.).

As described above, the control unit 100 alternates between flow paths B and C of the refrigerant by controlling the three-way valve 93. When the refrigerant flows through flow path B, the storage compartments 17 (refrigerator compartment 17A, chilled compartment 17B, vegetable compartment 17C) in the refrigeration temperature range are cooled. When the refrigerant flows through flow path C, the storage compartments 17 (ice-making compartment 17D, small freezer compartment 17E, main freezer compartment 17F) in the freezing temperature range are cooled. For example, the control unit 100 flows the refrigerant through flow path B for 20 minutes to cool the storage compartments 17 in the refrigeration temperature range (so-called refrigeration operation), and flows the refrigerant through flow path C for 40 minutes to cool the storage compartments 17 in the freezing temperature range (so-called freezing operation).

The control unit 100 executes "refrigeration operation" and "freezing operation" as basic operations of the refrigerator main body 1. "Refrigeration operation" refers to an operation in which the three-way valve 93 is switched and liquid refrigerant is supplied from the compressor 80 to the refrigeration cooler 61. On the other hand, "freezing operation" refers to an operation in which the three-way valve 93 is switched and liquid refrigerant is supplied from the compressor 80 to the refrigeration cooler 71.

The control unit 100 controls the cooling unit 50, for example, by alternating between refrigeration operation and freezing operation, so that the storage compartments 17 in the refrigeration temperature zone (refrigeration compartment 17A, chilled compartment 17B, vegetable compartment 17C) and the storage compartments 17 in the freezing temperature zone (ice-making compartment 17D, small freezer compartment 17E, main freezer compartment 17F) are kept at their respective set temperature zones. For example, the control unit 100 alternates between cooling the storage compartments 17 in the refrigeration temperature zone for a predetermined time (e.g., 20 minutes) and cooling the storage compartments 17 in the freezing temperature zone for another predetermined time (e.g., 40 minutes).

The control unit 100 may terminate the refrigeration operation and start the freezing operation even during the above-mentioned predetermined time, for example, when the refrigerator compartment temperature reaches the lower limit of the set temperature range of the refrigerator compartment 17A (or when the chilled compartment temperature reaches the lower limit of the set temperature range of the chilled compartment 17B) or when the freezer compartment temperature reaches the upper limit of the set temperature range of the main freezer compartment 17F during the refrigeration operation. The control unit 100 may terminate the freezing operation and start the refrigeration operation even during the above-mentioned predetermined time, for example, when the freezer compartment temperature reaches the lower limit of the set temperature range of the main freezer compartment 17F during the refrigeration operation or when the refrigerator compartment temperature reaches the upper limit of the set temperature range of the refrigerator compartment 17A (or when the chilled compartment temperature reaches the upper limit of the set temperature range of the chilled compartment 17B).

Here, during refrigeration operation, the air temperature in storage compartment 17 in the refrigeration temperature range drops, but the air temperature in storage compartment 17 in the freezing temperature range rises. On the other hand, during freezing operation, the air temperature in storage compartment 17 in the freezing temperature range drops, but the air temperature in storage compartment 17 in the refrigeration temperature range rises. For this reason, the air temperatures in storage compartment 17 in the refrigeration temperature range and the air temperature in storage compartment 17 in the freezing temperature range repeatedly rise and fall in a sawtooth pattern. Refrigeration operation and freezing operation in the refrigeration cycle device 90 are repeated alternately.

The control unit 100 also performs defrosting operations for the cooler under the control of the control unit 101a of the control device 2 described below.

(Control device)
6 is a diagram showing an example of the configuration of the control device 2. The control device 2 includes a control unit 101a, a prediction unit 101b, a determination unit 101c, and a storage unit 101d, which are configured by a computer including a microcomputer and a timer 101a1 that measures time. Each of the control unit 101a, the prediction unit 101b, and the determination unit 101c may be a software function unit realized by executing a computer program by one or more hardware processors such as a CPU (Central Processing Unit), or may be realized by hardware (e.g., a circuit unit; circuitry) such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a PLD (Programmable Logic Device). All or a part of the control unit 101a, the prediction unit 101b, and the determination unit 101c may be realized by a combination of a software function unit and hardware.

The prediction unit 101b predicts the frosting state of at least one of the refrigeration cooler 61 or the freezing cooler 71 (an example of a cooler) based on at least one of food ingredient information indicating the food ingredients in the refrigerator, the refrigerator temperature, the door opening/closing history, the outdoor air temperature, and the outdoor air humidity. An example of the outdoor air temperature is the temperature detected by the ambient temperature sensor of the refrigerator body 1. An example of the outdoor air humidity is the humidity detected by the ambient humidity sensor of the refrigerator body 1. Examples of the interior temperature are the refrigerator compartment temperature sensor, the vegetable compartment temperature sensor, the freezer compartment temperature sensor, the ice compartment temperature sensor, the freezer compartment humidity sensor, and the vegetable compartment humidity sensor. For example, the prediction unit 101b predicts the frosting state of at least one of the frosting of the refrigeration cooler 61 or the freezing cooler 71 based on at least one of the following: food ingredient information indicating the food ingredients in the refrigerator from time t0 to time t when the previous defrosting was performed, the refrigerator temperature from time t0 to time t when the previous defrosting was performed, the door opening and closing history from time t0 to time t when the previous defrosting was performed, the outdoor air temperature from time t0 to time t when the previous defrosting was performed, or the outdoor air humidity from time t0 to time t when the previous defrosting was performed. Examples of the frosting state include the amount of frosting of the refrigeration cooler 61, the amount of frosting of the freezing cooler 71, the state regarding the type of ice attached to the refrigeration cooler 61, the state regarding the type of ice attached to the freezing cooler 71, the state regarding the distribution of ice attached to the refrigeration cooler 61, and the state regarding the distribution of ice attached to the freezing cooler 71.

In addition, the prediction unit 101b may predict the time when the predicted frost state reaches a predetermined state when the predicted frost state has not reached a predetermined state. For example, the prediction unit 101b may predict the frost state regarding the frost of at least one of the refrigeration cooler 61 or the freezing cooler 71 based on at least one of the food information indicating the food in the refrigerator from the time when the last defrosting was performed until the time when the time t has elapsed, the temperature in the refrigerator, the door opening and closing history, the outdoor air temperature, or the outdoor air humidity, and when the predicted frost state has not reached a predetermined state, the prediction unit 101b may predict the time when the frost state reaches a predetermined state based on at least one of the food information indicating the food in the refrigerator from the time when the last defrosting was performed until the time when the time t has elapsed, the temperature in the refrigerator, the door opening and closing history, the outdoor air temperature, or the outdoor air humidity. Note that the prediction unit 101b may predict the initial frost state using, for example, at least one of the food information indicating the food inside the refrigerator from the time when the refrigerator system 200 was started up to the time when the time t has elapsed, the food temperature inside the refrigerator, the door opening/closing history, the outdoor temperature, or the outdoor humidity, instead of at least one of the food information indicating the food inside the refrigerator from the time when the previous defrosting was performed until the time when the time t has elapsed, the food temperature inside the refrigerator, the door opening/closing history, the outdoor temperature, or the outdoor humidity.

For example, if the time t is 168 hours, which corresponds to one week, then at least one of the food information, the temperature inside the refrigerator, the door opening/closing history, the outdoor temperature, and the outdoor humidity indicating the food in the refrigerator from the time of the last defrosting to the time when the time t has elapsed may be at least one of the food information, the temperature inside the refrigerator, the door opening/closing history, the outdoor temperature, and the outdoor humidity indicating the food in the refrigerator from the time when the last defrosting was performed to the time when the time t has elapsed, when the timer is set to zero at the time when the last defrosting was performed and the measurement of the time elapsed is started, until the time when the measurement of the time t has elapsed. Also, if the time t is 168 hours, which corresponds to one week, then at least one of the food information, the temperature inside the refrigerator, the door opening/closing history, the outdoor temperature, and the outdoor humidity indicating the food in the refrigerator from the time when the last defrosting was performed to the time when the time t has elapsed may be at least one of the food information, the temperature inside the refrigerator, the door opening/closing history, the outdoor temperature, and the outdoor humidity indicating the food in the refrigerator from the time when the last defrosting was performed to the time when the time t has elapsed, when the time when the last defrosting was performed is, for example, 8:00 a.m. on December 14th, until 8:00 a.m. on December 21st, which is 168 hours after that time. If a notification is issued to prompt the user to defrost when it is predicted that the frost state has reached a predetermined state, the predetermined frost state is, for example, a frost state determined in advance by the user that occurs prior to the frost state that requires the refrigerator system 200 to defrost. If a notification is not issued to prompt the user to defrost when it is predicted that the frost state has reached a predetermined state, and the refrigerator system 200 performs defrosting based on the prediction, the predetermined frost state is, for example, a frost state determined in advance by the user that requires the refrigerator system 200 to defrost.

The prediction of the frost state regarding frost formation on at least one of the refrigeration cooler 61 or the refrigeration cooler 71 by the prediction unit 101b described above may be performed, for example, using a trained model. The parameters of the trained model that specifies the frost state are determined, for example, as follows. Note that here, the determination of the parameters of the trained model for predicting the frost state regarding frost formation on the refrigeration cooler 71 will be described.

The parameters of the trained model that identifies the frost state are determined by applying training data to a trained model whose parameters have not yet been determined. The training data used to determine the parameters of the trained model that identifies the frost state includes input data and output data that corresponds to the input data.

7 is a diagram showing an example of teacher data for identifying parameters of a learning model used by the prediction unit 101b to predict the frosting state of the refrigeration cooler 71. For example, the memory unit 101d stores this teacher data. The teacher data is actual data showing the past frosting state of the refrigeration cooler 71. When predicting the frosting state of the refrigeration cooler 71, the teacher data includes at least one of the following items (the temperature inside the refrigerator in the teacher data shown in FIG. 7) as input data: food material information showing the food material inside the refrigerator at each time from the time of the previous defrosting to the time when the time t has elapsed, the temperature inside the refrigerator, the door opening/closing history, the outside air temperature, and the outside air humidity at each time from the time of the previous defrosting to the time when the time t has elapsed; the frosting state of the refrigeration cooler 71 corresponding to the input data (the amount of frost in the teacher data shown in FIG. 7) as output data; and includes multiple data in which the input data and the output data corresponding to the input data are associated. In the example shown in FIG. 7, the data in which the input data and the output data corresponding to the input data are associated is 10,000 sets of data.

For example, consider the case where parameters in a learning model for a refrigeration cooler 71 are determined using teacher data consisting of 10,000 sets of data shown in FIG. 7. In this case, the teacher data is divided into, for example, training data, evaluation data, and test data. Examples of the ratios of training data, evaluation data, and test data include 70%, 15%, 15%, 95%, 2.5%, and 2.5%. For example, suppose that teacher data of data #1 to #10000 is divided into data #1 to #7000 as training data, data #7001 to #8500 as evaluation data, and data #8501 to #10000 as 15% test data. In this case, data #1, which is training data, is input into a neural network, which is a learning model. The input data of the training data is input to the neural network, and each time the frosting state of the refrigeration cooler 71 is output from the neural network (in this case, each time each of the data #1 to #7000 is input to the neural network), the parameters indicating the weighting of the data connections between the nodes are changed by, for example, performing backpropagation in response to the output (i.e., the model of the neural network is changed). In this way, the training data is input to the neural network to adjust the parameters.

Next, the input data of the evaluation data (data #7001 to #8500) is input in order to the neural network whose parameters have been changed by the training data. The neural network outputs the frosting state of the refrigeration cooler 71 according to the input evaluation data. Here, if the data output by the neural network differs from the output data associated with the input data in FIG. 7, the parameters are changed so that the output of the neural network becomes the output data associated with the input data in FIG. 7. In this way, the neural network (i.e., the learning model) whose parameters have been determined is a trained model for predicting the frosting state of the refrigeration cooler 71 (hereinafter referred to as the first trained model).

Next, as a final check, input data of the test data (data #8501 to #10000) is input in order to the neural network of the first trained model. The neural network of the first trained model outputs the frosting state of the refrigeration cooler 71 according to the input test data. If the frosting state of the refrigeration cooler 71 output by the neural network of the first trained model for all test data matches the frosting state of the refrigeration cooler 71 associated with the input data in FIG. 7, the neural network of the first trained model is the desired model. Also, if the frosting state of the refrigeration cooler 71 output by the neural network of the first trained model does not match the frosting state of the refrigeration cooler 71 associated with the input data in FIG. 7 for even one of the test data, the parameters of the trained model are determined using new teacher data. The above-mentioned determination of the parameters of the trained model is repeated until a first trained model having the desired parameters is obtained. When a first trained model having the desired parameters is obtained, the first trained model is recorded, for example, in the memory unit 101d.

The control unit 101a controls the defrosting operation for the cooler (at least one of the refrigeration cooler 61 or the freezing cooler 71) based on the frosting state predicted by the prediction unit 101b. Examples of the defrosting operation include rotating a fan (the refrigeration fan 62 in the case of the refrigeration cooler 61, and the freezing fan 72 in the case of the freezing cooler 71) that sends air to the cooler, increasing the rotation speed of the fan, operating a heater that warms the cooler, and increasing the temperature of the heater. For example, the control unit 101a controls the defrosting operation via the control unit 100 based on the amount of frost predicted by the prediction unit 101b. Specifically, the determination unit 101c determines the content of the defrosting operation by the fan or heater based on the frosting state predicted by the prediction unit 101b. More specifically, the determination unit 101c determines that the cooler is made easier to warm as the amount of frost predicted by the prediction unit 101b increases. The control unit 101a controls the fan or heater based on the content of the defrosting operation determined by the determination unit 101c. That is, the control unit 101a transmits to the control unit 100 a control command that makes it easier for the cooler to warm up as the amount of frost predicted by the prediction unit 101b increases. That is, the control unit 101a transmits to the control unit 100 at least one of a control command to increase the rotation speed of the fan, a control command to lengthen the rotation time of the fan, a control command to increase the temperature of the heater, or a control command to lengthen the operation time of the heater as the amount of frost increases. Then, the control unit 100 controls the fan or heater according to the control command transmitted by the control unit 101a. However, the control unit 101a transmits to the control unit 100 a control command to control the fan or heater so that the cooler does not warm up beyond a threshold, i.e., so that the temperature inside the cabinet does not warm up beyond a predetermined temperature.

(Processing performed by the refrigerator system)
Next, a process performed by the refrigerator system 200 of an embodiment will be described. FIG. 8 is a diagram showing an example of a process flow of the refrigerator system 200 of an embodiment. Here, the process flow of the refrigerator system 200 shown in FIG. 8 will be described. It is assumed that a first trained model in which parameters are determined using teacher data prepared using data such as past performance data is already prepared before predicting the frosting state related to the frosting of the cooler. It is also assumed that the cooler is a refrigeration cooler 71, and the frosting state related to the frosting of the cooler is the amount of frost. It is also assumed that the temperature inside the freezer has been measured by a sensor since the previous prediction of the frosting state, and the temperature inside the freezer has been recorded in the memory unit 101d together with the time.

The prediction unit 101b predicts the frosting state of at least one of the refrigeration cooler 61 or the freezing cooler 71 (examples of coolers) based on at least one of food ingredient information indicating the food ingredients in the cabinet, the cabinet temperature, the door opening/closing history, the outside air temperature, or the outside air humidity (step S1). For example, the prediction unit 101b inputs the cabinet temperature stored in the memory unit 101d to the first trained model together with the elapsed time since the previous prediction of the frosting state. The first trained model outputs the amount of frosting in the freezing cooler 71 at the time indicated by the elapsed time.

The control unit 101a controls the defrosting operation for the cooler based on the frost state predicted by the prediction unit 101b (step S2). For example, the control unit 101a controls the defrosting operation via the control unit 100 based on the amount of frost predicted by the prediction unit 101b. Specifically, the control unit 101a transmits to the control unit 100 a control command that makes it easier for the cooler to warm up as the amount of frost predicted by the prediction unit 101b increases. That is, the control unit 101a transmits to the control unit 100 at least one of a control command to increase the number of rotations of the fan, a control command to lengthen the rotation time of the fan, a control command to increase the temperature of the heater, or a control command to lengthen the operation time of the heater as the amount of frost increases. Then, the control unit 100 controls the fan and the heater according to the control command transmitted by the control unit 101a. However, the control unit 101a transmits to the control unit 100 a control command to control the fan and the heater so that the cooler does not warm up beyond a threshold value, that is, so that the temperature inside the cabinet does not warm up beyond a predetermined temperature. The control command for the control unit 101a to control the fan and heater so that the temperature does not exceed a predetermined temperature is, for example, obtained by the control unit 101a from a sensor during control to obtain information indicating the inside temperature, and when the inside temperature reaches a threshold value, the control unit 101a stops sending the control command. Then, when the inside temperature falls to the predetermined temperature, the control unit 101a again sends the control command to the control unit 100.

In the above explanation, defrosting of the refrigeration cooler 71 was described as a specific example, but as mentioned above, the cooler to be defrosted may be the refrigeration cooler 61, or it may be both the refrigeration cooler 71 and the refrigeration cooler 61.

(advantage)
The refrigerator system 200 according to one embodiment has been described above. In the refrigerator system 200, the prediction unit 101b predicts a frosting state related to frosting of the cooler based on at least one of food material information indicating food materials in the refrigerator, the refrigerator temperature, the door opening/closing history, the outside air temperature, and the outside air humidity. The control unit 101a controls a defrosting operation for the cooler based on the frosting state predicted by the prediction unit 101b. The refrigerator system 200 allows the cooler of the refrigerator main body 1 to be defrosted appropriately.

Below, we will explain some modified examples. Note that the configuration of each modified example other than that described below is the same as the above embodiment.

First Modification of the Embodiment
In a first modified example of the embodiment, the control unit 101a may control the defrosting operation based on the state related to the type of ice predicted by the prediction unit 101b. For example, when frost begins to form on the cooler, white low-density ice adheres to the cooler, but when the frost melts and freezes again, high-density transparent ice adheres to the cooler. In other words, since high-density ice is harder to melt than low-density ice, the determination unit 101c determines to perform at least one of the following controls when the prediction unit 101b predicts that the type of ice is, for example, transparent: increase the number of rotations of the fan, lengthen the rotation time of the fan, increase the temperature of the heater, or lengthen the operation time of the heater, compared to when the ice is white. Based on the determination by the determination unit 101c, the control unit 101a may transmit to the control unit 100 at least one of the following control commands: increase the number of rotations of the fan, lengthen the rotation time of the fan, increase the temperature of the heater, or lengthen the operation time of the heater, compared to when the ice is, for example, transparent or white. The color of ice may be determined by, for example, the prediction unit 101b using a trained model whose parameters are determined using teacher data, in the same manner as in the case of determining the parameters of the first trained model. In this case, an example of the teacher data includes at least one of the following: food information indicating the food in the refrigerator at each time from the time of the last defrosting until the time when the time t has elapsed; the refrigerator temperature; the door opening/closing history; the outside air temperature; and the outside air humidity; and the color of the ice attached to the cooler corresponding to the input data is output data, and includes a plurality of data in which the input data and the output data corresponding to the input data are associated with each other. However, the control unit 101a transmits to the control unit 100 a control command to control the fan and heater so that the cooler does not heat up beyond a threshold value, that is, so that the inside temperature does not heat up beyond a predetermined temperature. The control unit 101a controls the fan and heater so that the temperature does not heat up beyond a predetermined temperature, for example, by acquiring information indicating the inside temperature from a sensor during control by the control unit 101a, and temporarily stopping the transmission of the control command when the inside temperature reaches a threshold value. Then, when the temperature inside the refrigerator drops to a predetermined temperature, the control unit 101a may again transmit a control command to the control unit 100.

(advantage)
The refrigerator system 200 according to the first modified example of the embodiment has been described above. With this refrigerator system 200, it is possible to prevent the temperature of the cooler from being increased more than necessary when ice melts easily. In other words, with this refrigerator system 200, it is possible to prevent the temperature inside the refrigerator from being increased more than necessary.

<Second Modification of the Embodiment>
In a second modified embodiment, the prediction unit 101b may predict the frosting state of the frosting of the freezer cooler 71 based on at least the food ingredient information indicating the food ingredients in the refrigerator. For example, the prediction unit 101b may predict the frosting state based on at least the type of food ingredients included in the food ingredient information indicating the food ingredients in the refrigerator, or the information indicating the food ingredient classification to which the food ingredients belong. The information indicating the type of food ingredients or the food ingredient classification to which the food ingredients belong may be obtained by, for example, the prediction unit 101b analyzing an image taken by an in-fridge camera. Specifically, the prediction unit 101b may predict that the amount of frosting of the cooler is large when the food ingredients are relatively water-rich, and may predict that the amount of frosting of the cooler is small when the food ingredients are relatively water-rich. This prediction is based on the idea that when food ingredients with a high water content are placed in the refrigerator, the humidity increases and frosting is likely to occur. Note that this prediction may be made by, for example, the prediction unit 101b using a trained model whose parameters are determined using teacher data, as in the case of determining the parameters of the first trained model. An example of teacher data in this case would be data in which input data includes at least information indicating the type of ingredients contained in the ingredient information indicating the ingredients in the refrigerator at each time from the time of the last defrosting to the time a time t has elapsed, or information indicating the ingredient classification to which the ingredients belong, and output data includes the frost state of the cooler corresponding to the input data, and the input data and the output data corresponding to the input data are associated with each other.

(advantage)
The above describes the refrigerator system 200 according to the second modified example of the embodiment. The refrigerator system 200 can improve the accuracy of prediction of the frosting state of the cooler by the prediction unit 101b compared to a case where food ingredient information indicating the food ingredients in the refrigerator is not used.

<Third Modification of the Embodiment>
In a third modified example of the embodiment, the prediction unit 101b may predict at least a state related to the distribution of ice attached to the cooler as the frosting state of the cooler. Examples of the state related to the distribution of ice attached to the cooler include the apparent ice adhesion ratio when viewed from six directions, i.e., up, down, front, back, left, and right, and the ice adhesion location on the cooler. Note that, as in the case of determining the parameters of the first trained model, this prediction may be performed, for example, by the prediction unit 101b using a trained model whose parameters are determined using teacher data. In this case, an example of the teacher data is information including at least the type of ingredients included in the ingredient information indicating the ingredients in the refrigerator at each time from the time of the previous defrosting to the time a time t has elapsed, or information indicating the ingredient classification to which the ingredients belong, as input data, and the state related to the distribution of ice attached to the cooler corresponding to the input data is output data, and includes a plurality of data in which the input data and the output data corresponding to the input data are associated with each other. In addition, the determination unit 101c determines to perform control according to the prediction result by the prediction unit 101b. For example, as the prediction unit 101b predicts that the proportion of ice that appears to adhere to the cooler increases when viewed from six directions, i.e., up, down, front, back, left, and right, or when the prediction unit 101b predicts that ice will adhere to a part of the cooler where ice is difficult to melt, the decision unit 101c decides to perform control to make ice melt more easily. Specifically, for example, when the prediction unit 101b predicts that frost has formed on the right side of the cooler, the decision unit 101c may decide to rotate the fan on the right side of the cooler. Then, the control unit 101a may control the defrosting operation determined by the decision unit 101c based on the state regarding the distribution of ice predicted by the prediction unit 101b. However, the control unit 101a transmits a control command to the control unit 100 to control the fan and heater so that the cooler does not heat up beyond a threshold value, i.e., so that the temperature inside the cooler does not heat up beyond a predetermined temperature. The control command for controlling the fan or heater so that the temperature does not exceed a predetermined temperature is, for example, obtained by the control unit 101a from a sensor during control, and once the temperature inside the cabinet reaches a threshold, the control unit 101a stops transmitting the control command. Then, once the temperature inside the cabinet falls to the predetermined temperature, the control unit 101a transmits the control command again to the control unit 100.

(advantage)
The refrigerator system 200 according to the second modified example of the embodiment has been described above. The refrigerator system 200 can improve the accuracy of the prediction of the frosting state of the cooler by the prediction unit 101b compared to a case where food ingredient information indicating the food ingredients in the refrigerator is not used. Also, the control unit 101a can efficiently defrost the cooler without increasing the temperature in the refrigerator more than necessary.

<Fourth Modification of the Embodiment>
In a fourth modified example of the embodiment, the prediction unit 101b may predict the time when the frost state predicted by the prediction unit 101b has not reached the predetermined state. In addition, this prediction may be performed by the prediction unit 101b using a trained model whose parameters are determined using teacher data, as in the case of determining the parameters of the first trained model. In this case, an example of the teacher data is a data set including at least information including the type of ingredients included in the food ingredient information indicating the ingredients in the refrigerator at each time from the time when the previous defrosting was performed to the time when the time t has elapsed, or information indicating the food ingredient classification to which the ingredients belong, as input data, and the elapsed time when the floor state of the cooler corresponding to the input data reaches a predetermined state as output data, and the input data and the output data corresponding to the input data are associated with each other. The control unit 101a may change the content or timing of the defrosting operation for the cooler depending on the time when the frost state predicted by the prediction unit 101b reaches the predetermined state. For example, the determination unit 101c determines the content or timing of the defrosting operation for the cooler according to the time when the frost state predicted by the prediction unit 101b reaches a predetermined state. Then, the control unit 101a generates a control command according to the result determined by the determination unit 101c and transmits the generated control command to the control unit 100.

(advantage)
The refrigerator system 200 according to the fourth modified example of the embodiment has been described above. With this refrigerator system 200, it is possible to know the timing when the frosting state reaches a predetermined state, and to prevent unnecessary frosting operations from being performed.

Fifth Modification of the Embodiment
In the fifth modified embodiment, the control unit 101a may change the timing of the defrosting operation for the cooler to before or after the special time period when the time when the frost state predicted by the prediction unit 101b reaches a predetermined state is a special time period. Examples of the special time period include a time period when the door of the refrigerator body 1 is frequently opened and closed, and the temperature inside the refrigerator is likely to rise, and a time period when the temperature inside the refrigerator is not stable. Specifically, the special time period includes a time period when the door is frequently opened and closed, such as breakfast, lunch, and dinner, a time period when quick freezing or putting high-temperature foods into the freezer, and a time period when the cooler must be kept at a low temperature, such as when special chill is set. The special time period can be obtained, for example, from the daily door opening and closing history and the refrigerator control history (the history of the operation mode operation by the user).

(advantage)
The above describes the refrigerator system 200 according to the fifth modified example of the embodiment. The refrigerator system 200 can prevent the internal temperature from rising to a predetermined temperature or higher.

Sixth Modification of the Embodiment
In a sixth modified example of the embodiment, the control device 2 may include a learning unit 101e that additionally learns a learned model (an example of a learned model) as shown in Fig. 9. For example, the detection unit (i.e., the refrigeration fan 62, the freezer fan 72, the compressor 80, the three-way valve 93, the operation panel 30, the outside air temperature sensor 111, the refrigerator compartment temperature sensor 112, the chilled compartment temperature sensor 113, the refrigerator compartment door switch 114, the chilled compartment door switch 115, the interior camera 116, etc.) detects the state of the cooler or the refrigerator main body 1 during operation of the refrigerator main body 1. The memory unit 101d stores the detection result detected by the detection unit. The learning unit 101e may update the parameters of the learned model by using the detection result detected by the detection unit during operation of the refrigerator main body 1 stored in the memory unit 101d as new teacher data. In other words, the learning unit 101e may additionally learn a learning model based on the content of the defrosting operation by the control unit 101a determined based on the frost state predicted by the prediction unit 101b and the detection results detected by the detection unit (i.e., the refrigeration fan 62, the freezing fan 72, the compressor 80, the three-way valve 93, the operation panel 30, the outside air temperature sensor 111, the refrigerator compartment temperature sensor 112, the chilled compartment temperature sensor 113, the refrigerator compartment door switch 114, the chilled compartment door switch 115, the interior camera 116, etc.) in response to the defrosting operation.

Seventh Modification of the Embodiment
In the seventh modified embodiment, the input data of the teacher data may include at least one of the following information obtained from a server that manages the application program, in addition to or instead of the above-mentioned input data: information related to various sensors such as an infrared sensor, a light sensor, an odor sensor (gas sensor), a pressure sensor attached to a door or a door pocket, and a current detector; information related to electrical components such as a damper; and information obtained from a server that manages an application program, such as an installation location. The infrared distance sensor, the light sensor, and the pressure sensor detect how much food is in the refrigerator, and information related to the amount of food in the refrigerator can be added to the input data. For example, when circulating the air in the refrigerator to defrost the cooler, if there is a large amount of food, the air in the refrigerator does not flow easily, and the air circulation becomes poor. As a result, the time required to defrost the cooler may be longer. In other words, the information detected by the infrared distance sensor, the light sensor, and the pressure sensor affects the defrosting of the cooler, and by using the detection results as input data, the refrigerator system 200 can predict the frosting state more accurately. Specifically, for example, based on the idea that when air circulation becomes poor, the temperature difference between the inside of the refrigerator and the cooler increases, and the amount of frost on the cooler increases, the detection results of the infrared distance sensor, the light sensor, and the pressure sensor can be used to predict the amount of frost on the cooler. Also, for example, when high-temperature food is placed in the refrigerator, the temperature inside the refrigerator rises, the temperature difference between the cooler and the inside of the refrigerator increases, and the amount of frost increases. Since the infrared heat sensor can measure the temperature of the food inside the refrigerator, the refrigerator system 200 can predict the frost state more accurately by adding the detection result of the infrared heat sensor to the input data in this case as well. Also, the odor sensor can determine the type of food by detecting the odor. For example, it is possible to distinguish between leafy vegetables and root vegetables based on the detection result of the odor sensor. Since the amount of moisture contained in the food differs between leafy vegetables and root vegetables, when the food is placed in the refrigerator, the amount of moisture released from the food into the refrigerator also differs. When compared with the same volume, leafy vegetables release more moisture into the refrigerator than root vegetables. Therefore, when the amount of leafy vegetables is larger than the amount of root vegetables as ingredients in the refrigerator, the amount of frost on the cooler may increase. Therefore, in this case, too, by adding the detection result of the odor sensor to the input data, the refrigerator system 200 can predict the frost state more accurately.

(advantage)
The above describes the refrigerator system 200 according to the seventh modified example of the embodiment. The refrigerator system 200 can increase the amount of input information related to the output data output by the trained model, and as a result, it is expected that the prediction by the prediction unit 101b will become more accurate.

Eighth Modification of the Embodiment
In an eighth modified example of the embodiment, the control unit 100 provided in the refrigerator main body 1 may perform some or all of the processing performed by the control device 2 on behalf of the control device 2.

(advantage)
The above describes the refrigerator system 200 according to the eighth modified example of the embodiment. With this refrigerator system 200, there is a possibility that the refrigerator system 200 can be realized with a different configuration, for example, with only the refrigerator main body 1.

Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. These embodiments and their modifications are within the scope of the invention and its equivalents as set forth in the claims, as well as the scope and gist of the invention.

1...refrigerator body, 2...control device, 10...casing, 20...multiple doors, 30...operation panel, 40...flow path forming parts, 50...cooling section, 62...refrigeration fan, 72...freezing fan, 80...compressor, 93...three-way valve, 100...control section, 101a...control section, 101a1...timer, 101b...prediction section, 101c...determination section, 101d...storage section, 101e...learning section, 111...outdoor air temperature sensor, 112...refrigerator compartment temperature sensor, 113...chilled compartment temperature sensor, 114...refrigerator compartment door switch, 115...chilled compartment door switch, 116...interior camera, 117...storage section.

Claims (3)

A refrigerator body having a cooling device;
a prediction unit that predicts the color of ice that has formed on the cooling unit based on a learning model that has been trained using a plurality of pieces of learning information for learning a learning model that uses as input information at least one of food ingredient information indicating the food ingredients in the cooling unit at each time since the previous defrosting, the temperature inside the cooling unit, the door opening/closing history, the outside air temperature, and the outside air humidity, and that outputs information indicating the color of ice that has formed on the cooling unit corresponding to the input information;
a control unit that controls a defrosting operation of the cooling device so as to increase a defrosting capacity when the prediction unit predicts that the color of the ice is transparent compared to when the prediction unit predicts that the color of the ice is white;
A refrigerator system comprising: a fan for changing an airflow rate to the cooler for performing a defrosting operation on the cooler or a heater for changing an ambient temperature of the cooler;
A determination unit that determines the content of a defrosting operation to be performed by the fan or the heater based on the color of the ice predicted by the prediction unit;
Further equipped with
The control unit is
controlling the fan or the heater based on the content of the defrosting operation determined by the determination unit;
2. The refrigerator system of claim 1 . A detection unit that detects a state of the cooler or the refrigerator main body;
A learning unit that additionally learns the learning model after the learning;
Further equipped with
The prediction unit is
predicting a color of the ice using the trained model;
The learning unit is
additionally learning the learning model after learning based on the content of the defrosting operation by the control unit determined based on the color of the ice predicted by the prediction unit and the detection result detected by the detection unit in response to the defrosting operation.
3. A refrigerator system according to claim 1 or 2 .

Images