WO2023084608A1

WO2023084608A1 - Control device and control method

Info

Publication number: WO2023084608A1
Application number: PCT/JP2021/041210
Authority: WO
Inventors: 守濱田; 寛光穂苅
Original assignee: 三菱電機株式会社
Priority date: 2021-11-09
Filing date: 2021-11-09
Publication date: 2023-05-19
Also published as: JPWO2023084608A1

Abstract

In the present invention, a selection unit (103) selects, from among a plurality of indoor units each being assigned an air conditioning space representing a target for air conditioning, an indoor unit serving as a representative indoor unit representing the plurality of indoor units, such selection being on the basis of a required air conditioning capacity which is the air conditioning capacity required by each indoor unit. A setting unit (105) uses a trained model (110) obtained through machine learning to set a target temperature for either an evaporation temperature or a condensation temperature at which the air conditioning capacity of the representative indoor unit can be matched to the required air conditioning capacity of the representative indoor unit. In a case where, for each indoor unit of the plurality of indoor units excluding the representative indoor unit, either the evaporation temperature or the condensation temperature in each indoor unit matches the target temperature, a calculation unit (106) calculates either the degree of superheating or the degree of supercooling with which the air conditioning capacity of each indoor unit matches the required air conditioning capacity of the indoor unit.

Description

Control device and control method

The present disclosure relates to control of an air conditioner.

As a technology related to the present disclosure, there is a technology disclosed in Patent Document 1. Patent Literature 1 discloses an air conditioner in which a plurality of indoor units are connected to one outdoor unit.
Patent Document 1 discloses a configuration that achieves both energy saving and comfort. Specifically, in Patent Literature 1, the air conditioning capacity required for each indoor unit (hereinafter referred to as the required air conditioning capacity) is obtained. Then, the target evaporation temperature and the target degree of superheat are set according to the maximum requested air-conditioning capacity among the acquired requested air-conditioning capacities. Then, when the cooling temperature (indoor temperature) of an indoor unit other than the indoor unit with the maximum required air conditioning capacity falls below the target cooling temperature (target indoor temperature) by a predetermined value or more, the target degree of superheat of the corresponding indoor unit is changed.

JP 2018-138841 A

In the technique disclosed in Patent Literature 1, searching for the target evaporation temperature is started after each indoor unit is operated. Therefore, the target evaporating temperature is searched for during operation of each indoor unit. Therefore, with the technique of Patent Document 1, it takes time to set the target evaporation temperature. When the required air conditioning capacity of any indoor unit fluctuates while searching for the target evaporating temperature, intermittent operation occurs in the corresponding indoor unit, resulting in a problem of reduced operating efficiency.
Further, in the technique of Patent Document 1, the target degree of superheat of the indoor unit is changed according to the decrease in the cooling temperature (indoor temperature). Therefore, it is difficult to match the air conditioning capacity of the indoor unit with the required air conditioning capacity. Therefore, the technique of Patent Document 1 has a problem that intermittent operation occurs in one of the indoor units, resulting in a decrease in operating efficiency.

One of the main purposes of this disclosure is to solve the above problems. Specifically, the main purpose of the present disclosure is to prevent the occurrence of intermittent operation in each indoor unit and to prevent a decrease in operating efficiency.

A control device according to the present disclosure includes:
A representative of the plurality of indoor units, each of which is assigned an air-conditioned space to be air-conditioned, based on the required air-conditioning capacity, which is the air-conditioning capacity required for each indoor unit. a selection unit that selects an indoor unit to be used as a representative indoor unit;
A learning model obtained by machine learning is used to set a target temperature of either the evaporating temperature or the condensing temperature that allows the air conditioning capacity of the representative indoor unit to match the required air conditioning capacity of the representative indoor unit. a setting unit;
For each indoor unit of the plurality of indoor units excluding the representative indoor unit, when either the evaporation temperature or the condensation temperature in each indoor unit is made to match the target temperature, the air conditioning capacity of each indoor unit is a calculating unit that calculates either the degree of superheating or the degree of supercooling that matches the required air conditioning capacity of the indoor unit.

In the present disclosure, an appropriate target temperature (evaporating temperature or condensing temperature) is set early by using a learning model. In addition, in the present disclosure, based on the set target temperature, the degree of superheating and the degree of supercooling of each indoor unit whose air conditioning capability matches the required air conditioning capability of the indoor unit are calculated. That is, in the present disclosure, an appropriate degree of superheating or supercooling that does not cause intermittent operation in each indoor unit is calculated for each indoor unit.
Therefore, according to the present disclosure, it is possible to prevent the occurrence of intermittent operation in each indoor unit and prevent a decrease in operating efficiency.

1 is a diagram showing a configuration example of an air conditioning system according to Embodiment 1; FIG. 2 is a diagram showing a functional configuration example of a control device according to Embodiment 1; FIG. 2 is a diagram showing a hardware configuration example of a control device according to Embodiment 1; FIG. 4 is a flowchart showing an operation example of the control device in the operation phase (during cooling operation) according to the first embodiment; 4 is a flowchart showing an operation example of the control device in the operation phase (at the time of heating operation) according to the first embodiment; 4 is a flowchart showing an operation example of the control device in a learning phase (during cooling operation) according to the first embodiment; 4 is a flowchart showing an operation example of the control device in a learning phase (during heating operation) according to Embodiment 1; 9 is a flowchart showing an example of the operation of the control device in the operation phase (during cooling operation) according to the second embodiment; 9 is a flowchart showing an example of the operation of the control device in the operation phase (during heating operation) according to the second embodiment; 9 is a flowchart showing an example of the operation of the control device in the learning phase (during cooling operation) according to the second embodiment; 9 is a flowchart showing an example of the operation of a control device in a learning phase (during heating operation) according to Embodiment 2;

An embodiment will be described below with reference to the drawings. In the following description of the embodiments and drawings, the same reference numerals denote the same or corresponding parts.

Embodiment 1.
*** Configuration description ***
FIG. 1 shows a configuration example of an air conditioning system 500 according to this embodiment.
Air conditioning system 500 according to the present embodiment has control device 100 and air conditioner 400 .

The air conditioner 400 is composed of one outdoor unit 200 and a plurality of indoor units 300 . A plurality of indoor units 300 are connected to one outdoor unit 200 . Although three indoor units 300 are connected to the outdoor unit 200 in FIG. 1, the number of indoor units 300 may be other than three.
The outdoor unit 200 is installed outside the building.
Each indoor unit 300 is installed inside the building. Each indoor unit 300 is individually assigned a space (for example, a room) to be air-conditioned. The space to be air-conditioned that is assigned to each indoor unit 300 is hereinafter referred to as an air-conditioned space.

A control device 100 controls an outdoor unit 200 and a plurality of indoor units 300 .
The control device 100 has a learning phase and an operation phase as operation phases.
The control device 100 performs machine learning in the learning phase. Also, in the operation phase, the control device 100 controls the outdoor unit 200 and the plurality of indoor units 300 using the results of machine learning.
In the learning phase, the control device 100 collects operating data from the outdoor unit 200 and the multiple indoor units 300 . Then, the control device 100 performs machine learning using the collected driving data, and generates a learning model that reflects the results of the machine learning.
Also in the operation phase, the control device 100 collects operation data from the outdoor unit 200 and the plurality of indoor units 300 . The control device 100 also applies the operation data to the learning model to generate control target values for controlling the operation of the outdoor unit 200 and the plurality of indoor units 300 . Then, the control device 100 outputs the control target value to the outdoor unit 200 and each indoor unit 300 to control the operation of the outdoor unit 200 and each indoor unit 300 . In FIG. 1, the operating data are represented by solid-line arrows. Also, the control target value is indicated by a dashed arrow.
Details of the operation data and the control target value will be described later. Also, the details of the machine learning method and the application method of the learning model will be described later.
The operating procedure of the control device 100 corresponds to the control method.

FIG. 2 shows an example of the functional configuration of the control device 100. As shown in FIG. FIG. 3 shows a hardware configuration example of the control device 100. As shown in FIG.
First, a hardware configuration example of the control device 100 will be described with reference to FIG.

Control device 100 according to the present embodiment is a computer.
The control device 100 includes a processor 901, a main storage device 902, an auxiliary storage device 903, a communication device 904, and an input/output device 905 as hardware.
As shown in FIG. 2, the control device 100 includes a collection unit 101, an estimation unit 102, a selection unit 103, a learning unit 104, a setting unit 105, a calculation unit 106, a control unit 107, and an operation data storage unit 108 as a functional configuration. Prepare.
The functions of the collecting unit 101, the estimating unit 102, the selecting unit 103, the learning unit 104, the setting unit 105, the calculating unit 106, and the control unit 107 are implemented by, for example, programs.
Auxiliary storage device 903 stores programs for realizing the functions of collection unit 101 , estimation unit 102 , selection unit 103 , learning unit 104 , setting unit 105 , calculation unit 106 and control unit 107 .
These programs are loaded from the auxiliary storage device 903 to the main storage device 902 . Then, the processor 901 executes these programs to perform operations of the collecting unit 101, the estimating unit 102, the selecting unit 103, the learning unit 104, the setting unit 105, the calculating unit 106, and the control unit 107, which will be described later.
FIG. 3 schematically shows a state in which the processor 901 is executing a program that implements the functions of the collection unit 101, the estimation unit 102, the selection unit 103, the learning unit 104, the setting unit 105, the calculation unit 106, and the control unit 107. represent.
The driving data storage unit 108 is realized by the auxiliary storage device 903, for example.

Next, an example of the functional configuration of the control device 100 will be described with reference to FIG.

The collection unit 101 uses the communication device 904 to collect operation data from the outdoor unit 200 and each indoor unit 300 .
When the air conditioner 400 is performing cooling operation, the collection unit 101 collects the outside air temperature, the set temperature in each air-conditioned space, the measured temperature measured in each air-conditioned space, and the temperature of each indoor unit 300 as operation data. , the degree of superheat of each indoor unit 300, and the operating status value of each indoor unit 300 are obtained.
On the other hand, when the air conditioner 400 is performing heating operation, the operating data include the outside air temperature, the measured temperature measured in each air-conditioned space, the condensation temperature of each indoor unit 300, and the degree of supercooling of each indoor unit 300. , to obtain the operating status value of each indoor unit 300 .
In addition, below, the measured temperature of each indoor unit 300 is also called room temperature. Evaporation temperature is also called ET. The degree of superheat is also called SH. The condensation temperature is also called CT. The degree of supercooling is also called SC.
The operating status value is a value that indicates the operating status of each indoor unit 300 for a predetermined period of time (for example, 10 minutes; the same applies hereinafter). The operating status value is a value between 0 and 1.0. The operating status value is the ratio of the time during which the indoor unit 300 is operating to the predetermined time. When the indoor unit 300 continues to operate for the predetermined time (when intermittent operation does not occur), the operation status value is 1.0. If the indoor unit 300 continues to stop for the predetermined time, the operating status value is 0. If the intermittent operation occurs in the indoor unit 300 for the predetermined time and the time during which the indoor unit 300 is operating is half of the predetermined time, the operation status value is 0.5.
Intermittent operation refers to an operating state in which the indoor unit 300 intermittently repeats operation and stop.
The operating status value is also called Thermo ON/OFF.
The collection unit 101 outputs the collected driving data to the estimation unit 102 .

The estimation unit 102 estimates the required air conditioning capacity of each indoor unit 300 using the operating data. The required air conditioning capacity is the air conditioning capacity required for each indoor unit 300 . That is, the required air conditioning capacity is the air conditioning capacity required in the air conditioning space of each indoor unit 300 . In other words, the required air-conditioning capacity is the air-conditioning capacity required to match the room temperature of the air-conditioned space of each indoor unit 300 with the set temperature. The required air conditioning capacity is also called load.
When the air conditioner 400 performs cooling operation, the required air conditioning capacity is the cooling capacity required of each indoor unit 300 (hereinafter referred to as the required cooling capacity). On the other hand, when the air conditioner 400 performs the heating operation, the required air conditioning capacity is the heating capacity required of each indoor unit 300 (hereinafter referred to as required heating capacity).
The details of the method of estimating the required air conditioning capacity will be described later.
The estimation unit 102 notifies the selection unit 103 of the required air conditioning capacity of each indoor unit 300 .
The estimation unit 102 also stores the driving data in the driving data storage unit 108 .

The selection unit 103 selects the indoor unit 300 with the maximum required air conditioning capacity from among the plurality of indoor units 300 as the learning indoor unit or the representative indoor unit.
In the learning phase, the selection unit 103 selects the indoor unit 300 with the maximum required air conditioning capacity as the learning indoor unit. On the other hand, in the operation phase, the selection unit 103 selects the indoor unit 300 with the maximum required air conditioning capacity as the representative indoor unit.
The learning indoor unit is the indoor unit 300 representing the plurality of indoor units 300 in the learning phase. The learning indoor unit is the indoor unit 300 used for machine learning in the learning unit 104, which will be described later.
A representative indoor unit is an indoor unit 300 that represents a plurality of indoor units 300 in the operation phase. The representative indoor unit is the indoor unit 300 used for setting the target temperature in the setting unit 105, which will be described later.
In the learning phase, the selection unit 103 notifies the learning unit 104 of the indoor unit 300 selected as the learning indoor unit. On the other hand, in the operation phase, the selection unit 103 notifies the setting unit 105 of the indoor unit 300 selected as the representative indoor unit. Further, the selection unit 103 notifies the calculation unit 106 of the required cooling capacity of the indoor units 300 other than the representative indoor unit.

The learning unit 104 acquires operation data of the learned indoor unit from the operation data storage unit 108 .
Then, the learning unit 104 performs machine learning using the operation data of the learning indoor unit, and generates the learning model 110 in which the result of the machine learning is reflected.
The learning unit 104 learns an evaporating temperature or a condensing temperature that does not cause intermittent operation of the learning indoor unit for a predetermined time. That is, the learning unit 104 learns the evaporating temperature or the condensing temperature that can match the air conditioning capacity of the learned indoor unit with the required air conditioning capacity of the learned indoor unit. When the air conditioner 400 is performing cooling operation, the learning unit 104 learns the evaporation temperature that allows the cooling capacity of the learned indoor unit to match the required cooling capacity of the learned indoor unit. When the air conditioner 400 is performing heating operation, the learning unit 104 learns the condensing temperature that allows the heating capacity of the learned indoor unit to match the required heating capacity of the learned indoor unit.
When the air conditioner 400 is performing cooling operation, the learning unit 104 calculates the set temperature of the air-conditioned space of the learning indoor unit, the measured temperature measured in the air-conditioned space of the learning indoor unit, and the temperature of the learning indoor unit. The relationship between the operating status value, the evaporating temperature measured by the learning indoor unit, the outside air temperature, the degree of superheat measured by the learning indoor unit, and the cooling capacity of the learning indoor unit is learned. More specifically, the learning unit 104 uses the operating data of the learning indoor unit to determine the input (set temperature, measured temperature, operating status value, evaporating temperature, outside temperature, degree of superheat) and output (cooling of the learning indoor unit). ability) is calculated. Then, the learning unit 104 accumulates the calculated correlation formula in the learning model 110 .
As will be described later, when the required cooling capacity of the representative indoor unit is given, the setting unit 105 sets the measured temperature, the set temperature, the operating status value, the evaporation temperature, the outside air temperature, the degree of superheat, and the required cooling capacity of the representative indoor unit. is applied to the learning model 110 (correlation formula), it is possible to derive the target evaporating temperature at which the cooling capacity of the representative indoor unit matches the required cooling capacity.
Further, when the air conditioner 400 is performing heating operation, the learning unit 104 determines the set temperature of the air-conditioned space of the learning indoor unit, the measured temperature measured in the air-conditioned space of the learning indoor unit, It learns the relationship between the operating status value of the machine, the condensing temperature measured by the learning indoor unit, the outside air temperature, the degree of subcooling measured by the learning indoor unit, and the heating capacity of the learning indoor unit. More specifically, the learning unit 104 uses the operating data of the learning indoor unit to obtain input (set temperature, measured temperature, operating status value, condensation temperature, outside air temperature, degree of supercooling) and output (learning indoor unit heating capacity). Then, the learning unit 104 accumulates the calculated correlation formula in the learning model 110 .
As will be described later, when the required heating capacity of the representative indoor unit is given, the setting unit 105 sets the measured temperature, the set temperature, the operating status value, the condensing temperature, the outside air temperature, the degree of subcooling, and the required heating capacity of the representative indoor unit. By applying the capacity to the learning model 110 (correlation formula), it is possible to derive the target condensing temperature at which the heating capacity of the representative indoor unit matches the required heating capacity.

The setting unit 105 acquires the operating data of the representative indoor unit from the operating data storage unit 108 .
Then, the setting unit 105 applies the operating data of the representative indoor unit to the learning model 110 to set the target temperature.
The setting unit 105 sets the target evaporation temperature as the target temperature when the air conditioner 400 is performing cooling operation. The setting unit 105 uses the learning model 110 to set, as the target evaporating temperature, a target evaporating temperature at which intermittent operation does not occur in the representative indoor unit for a predetermined time. That is, the setting unit 105 uses the learning model 110 to set the target evaporation temperature as the target evaporation temperature at which the cooling capacity of the representative indoor unit can match the required cooling capacity of the representative indoor unit for a predetermined time. . More specifically, the setting unit 105 determines the set temperature in the air-conditioned space of the representative indoor unit, the measured temperature measured in the air-conditioned space of the representative indoor unit, the operating status value, and the temperature measured in the representative indoor unit. The obtained evaporation temperature, the outside air temperature, the fixed degree of superheat, and the required cooling capacity of the representative indoor unit are applied to the learning model 110 (correlation formula) to obtain the target evaporation temperature.
On the other hand, when the air conditioner 400 is performing heating operation, the setting unit 105 sets the target condensing temperature as the target temperature. The setting unit 105 uses the learning model 110 to set, as the target condensation temperature, a target temperature at which intermittent operation does not occur in the representative indoor unit for a predetermined time. That is, the setting unit 105 uses the learning model 110 to set the target condensing temperature at which the heating capacity of the representative indoor unit matches the required heating capacity of the representative indoor unit for a predetermined time. Set as temperature. More specifically, the setting unit 105 determines the set temperature in the air-conditioned space of the representative indoor unit, the measured temperature measured in the air-conditioned space of the representative indoor unit, the operating status value of the representative indoor unit, and the The condensing temperature measured by the machine, the outside air temperature, the fixed degree of subcooling, and the required cooling capacity of the representative indoor unit are applied to the learning model 110 to obtain the target condensing temperature.
The setting unit 105 notifies the calculation unit 106 of the set target temperature (target evaporation temperature or target condensation temperature) and a fixed degree of superheat (in the case of cooling operation) or a fixed degree of supercooling (in the case of heating operation). do.

The calculation unit 106 converts the degree of superheating (in the case of cooling operation) or the degree of supercooling (in the case of heating operation) of the representative indoor unit to the fixed degree of superheating or the fixed degree of supercooling used to set the target temperature. set.
Further, the calculation unit 106 calculates the degree of superheat (in the case of cooling operation) or the degree of supercooling (in the case of heating operation) of each indoor unit 300 other than the representative indoor unit based on the target temperature (target evaporating temperature or target condensing temperature). calculated by More specifically, when the air conditioner 400 is performing cooling operation, the calculation unit 106 matches the evaporation temperature of each indoor unit 300 with the target evaporation temperature. A degree of superheat at which the cooling capacity of each indoor unit 300 matches the required cooling capacity of the indoor unit 300 for a predetermined time is calculated. The calculation unit 106 calculates the degree of superheat of each indoor unit 300 in which intermittent operation does not occur in each indoor unit 300 . In addition, when the air conditioner 400 is performing the heating operation, the calculation unit 106 calculates, for each indoor unit 300, when the condensing temperature in each indoor unit 300 matches the target condensing temperature. The degree of supercooling at which the heating capacity of each indoor unit 300 matches the required heating capacity of the indoor unit 300 is calculated. That is, the calculation unit 106 calculates the degree of supercooling of each indoor unit 300 at which intermittent operation does not occur in each indoor unit 300 .

The control unit 107 generates a control target value based on the target evaporation temperature and the degree of superheat of each indoor unit 300 when the air conditioner 400 is performing cooling operation. Then, the control unit 107 outputs the generated control target value to the outdoor unit 200 and each indoor unit 300 .
Further, when the air conditioner 400 is performing heating operation, the control unit 107 generates a control target value based on the target condensing temperature and the degree of supercooling of each indoor unit 300 . Then, the control unit 107 outputs the generated control target value to the outdoor unit 200 and each indoor unit 300 .
The control unit 107 controls the operation of the outdoor unit 200 and each indoor unit 300 by outputting the control target value.

***Description of operation***
Next, an operation example of the control device 100 according to this embodiment will be described.
First, an operation example of the control device 100 in the operation phase will be described with reference to FIGS. 4 and 5. FIG.
It is assumed that the learning model 110 has already been generated at the start of the flow of FIGS. 4 and 5 .
FIG. 4 shows an operation example of the control device 100 when the air conditioner 400 is performing cooling operation.
FIG. 5 shows an operation example of the control device 100 when the air conditioner 400 is performing heating operation.

First, referring to FIG. 4, the operation of the control device 100 during cooling operation will be described.

In FIG. 4, first, in step S101, the collection unit 101 collects driving data.
The collection unit 101 collects, as operating data, the outside air temperature, the set temperature in each air-conditioned space, the measured temperature in each air-conditioned space, the evaporation temperature of each indoor unit 300, the degree of superheat of each indoor unit 300, and the temperature of each indoor unit 300. Get the health status value of . The collection unit 101 acquires the set temperature, the measured temperature, the degree of superheat, and the operating status value from each indoor unit 300 . The evaporation temperature is collectively managed by the outdoor unit 200 . Therefore, the collection unit 101 acquires the evaporation temperature from the outdoor unit 200 . The collection unit 101 also acquires the outside air temperature from the outdoor unit 200 .
The collection unit 101 outputs the collected driving data to the estimation unit 102 .

Next, in step S102 , the estimation unit 102 estimates the required cooling capacity of each indoor unit 300 .
The estimation unit 102 estimates the required cooling capacity L1 [kW] of each indoor unit 300, for example, according to the following equation (1).
L1 = (measured temperature - ET - SH) x Thermo formula (1)
ET is the evaporation temperature. SH is the degree of superheat. Thermo is an operational status value.
The estimation unit 102 notifies the selection unit 103 of the required cooling capacity L1 of each indoor unit 300 .

Next, in step S103, the selection unit 103 selects a representative indoor unit.
Specifically, the selection unit 103 selects the indoor unit 300 with the maximum required cooling capacity L1 as the representative indoor unit.
The selecting unit 103 notifies the setting unit 105 of the selected representative indoor unit and the requested cooling capacity L1 of the representative indoor unit.
Further, the selection unit 103 notifies the calculation unit 106 of the required cooling capacity L1 of the indoor units 300 other than the representative indoor unit.

Next, in step S104, the setting unit 105 sets the target evaporation temperature.
The setting unit 105 acquires the operating data of the representative indoor unit from the operating data storage unit 108 . Specifically, the setting unit 105 acquires the measured temperature, the set temperature, the operating status value, the outside air temperature, and the evaporating temperature of the representative indoor unit from the operation data storage unit 108 .
Then, the setting unit 105 acquires the measured temperature, the set temperature, the operating status value, the outside temperature, and the evaporation temperature of the representative indoor unit acquired from the operation data storage unit 108, the degree of superheat of the fixed value, and the required cooling capacity of the representative indoor unit. and L1 are applied to the learning model 110 to set the target evaporating temperature.
The setting unit 105 applies the minimum allowable superheating degree (for example, SH=2K) as the fixed value of the superheating degree.
The target evaporating temperature set by the setting unit 105 is the highest evaporating temperature at which intermittent operation does not occur in the representative indoor unit for a predetermined time, that is, the highest evaporating temperature at which the cooling capacity of the representative indoor unit does not fall short of the required cooling capacity L1. is the evaporation temperature. That is, the target evaporating temperature is the highest evaporating temperature among the evaporating temperatures at which the cooling capacity of the representative indoor unit can be matched with the required cooling capacity L1. "The cooling capacity of the representative indoor unit can be matched with the required cooling capacity L1" means matching the measured temperature in the air-conditioned space of the representative indoor unit with the set temperature in the air-conditioned space of the representative indoor unit. means.
The setting unit 105 uses the correlation equation of the learning model 110 to identify the highest evaporating temperature among the evaporating temperatures that allow the cooling capacity of the representative indoor unit to match the required cooling capacity L1, and sets the identified evaporating temperature to Set the target evaporation temperature.
The setting unit 105 notifies the calculation unit 106 of the target evaporation temperature and the fixed superheat degree (for example, SH=2K).

Next, in step S105, the calculation unit 106 sets the target degree of superheat of the representative indoor unit.
Specifically, the calculation unit 106 sets the degree of superheat of the fixed value notified from the setting unit 105 as the target degree of superheat of the representative indoor unit.

Next, in step S106, the calculation unit 106 calculates the degree of superheat of each indoor unit 300 other than the representative indoor unit according to the above equation (1).
Specifically, the calculation unit 106 sets the required cooling capacity L1 of each indoor unit 300 notified from the selection unit 103 to L1 in Equation (1). Further, the calculation unit 106 sets the measured temperature measured in the air-conditioned space of each indoor unit 300 as the measured temperature in Equation (1). Further, the calculation unit 106 sets the target evaporation temperature notified from the setting unit 105 to ET in the formula (1). Further, the calculation unit 106 sets Thermo in Equation (1) to a value of 1 (Thermo=1) in common among the indoor units 300 . Calculation unit 106 then calculates SH that makes the right side and left side of Equation (1) match as the target degree of superheat. Note that the calculation unit 106 acquires the measured temperature of each indoor unit 300 from the operating data storage unit 108 .
The target degree of superheat calculated by the calculation unit 106 is the degree of superheat at which intermittent operation does not occur in each indoor unit 300 for a predetermined time when the evaporation temperature in each indoor unit 300 is matched with the target evaporation temperature, that is, It is the degree of superheat that allows the cooling capacity of each indoor unit 300 to match the required cooling capacity L1 of each indoor unit 300 . "Matching the cooling capacity of each indoor unit 300 with the required cooling capacity L1 of each indoor unit 300" means that the measured temperature in the air-conditioned space of each indoor unit 300 is set to the set temperature in the air-conditioned space of each indoor unit 300. means to match
The calculation unit 106 notifies the control unit 107 of the target evaporation temperature and the target degree of superheat of each indoor unit 300 including the representative indoor unit.

Next, in step S107, the control unit 107 generates a control instruction.
The control unit 107 determines the operating speed of the compressor based on the target evaporation temperature. The operating speed of the compressor is adjusted by the outdoor unit 200 .
Also, the control unit 107 determines the degree of opening of the indoor expansion valve for each indoor unit 300 based on the target degree of superheat of each indoor unit 300 . The opening degree of the indoor expansion valve is a different value for each indoor unit 300 . The degree of opening of the indoor expansion valve is adjusted by each indoor unit 300 .
The control unit 107 generates a control instruction to the outdoor unit 200 indicating the determined operating rotation speed of the compressor. Further, the control unit 107 generates a control instruction to each indoor unit 300 indicating the determined degree of opening of the indoor expansion valve.

Next, the control unit 107 outputs respective control instructions to the outdoor unit 200 and each indoor unit 300 .
The operation of the outdoor unit 200 and the plurality of indoor units 300 is controlled by the outdoor unit 200 and the indoor units 300 operating according to the control instructions.
That is, the control unit 107 can match the cooling capacity of each indoor unit 300 with the required cooling capacity of each indoor unit 300 by outputting the control instruction, and the intermittent operation of each indoor unit 300 does not occur.

Next, the operation of the control device 100 during heating operation will be described with reference to FIG.

In FIG. 5, first, in step S201, the collection unit 101 collects driving data.
The collection unit 101 collects, as operating data, the outside air temperature, the set temperature in each air-conditioned space, the measured temperature in each air-conditioned space, the condensation temperature of each indoor unit 300, the degree of supercooling of each indoor unit 300, each indoor unit 300 operating status values are obtained. The collection unit 101 acquires the set temperature, the measured temperature, the degree of supercooling, and the operating status value from each indoor unit 300 . The condensation temperature is collectively managed by the outdoor unit 200 . Therefore, the collection unit 101 acquires the condensation temperature from the outdoor unit 200 . The collection unit 101 also acquires the outside air temperature from the outdoor unit 200 .
The collection unit 101 outputs the collected driving data to the estimation unit 102 .

Next, in step S202 , the estimation unit 102 estimates the required heating capacity of each indoor unit 300 .
The estimation unit 102 estimates the required heating capacity L2 [kW] of each indoor unit 300, for example, according to the following equation (2).
L2 = (CT-SC-room temperature) x Thermo formula (2)
CT is the condensation temperature. SC is the degree of supercooling. Thermo is an operational status value.
The estimation unit 102 notifies the selection unit 103 of the required heating capacity L2 of each indoor unit 300 .

Next, in step S203, the selection unit 103 selects a representative indoor unit.
Specifically, the selection unit 103 selects the indoor unit 300 with the maximum required heating capacity L2 as the representative indoor unit.
The selecting unit 103 notifies the setting unit 105 of the selected representative indoor unit and the requested heating capacity L2 of the representative indoor unit.
Further, the selection unit 103 notifies the calculation unit 106 of the required heating capacity L2 of the indoor units 300 other than the representative indoor unit.

Next, in step S204, the setting unit 105 sets the target condensing temperature.
The setting unit 105 acquires the operating data of the representative indoor unit from the operating data storage unit 108 . Specifically, the setting unit 105 acquires the measured temperature, set temperature, operating status value, outside air temperature, and condensing temperature of the representative indoor unit from the operation data storage unit 108 .
Then, the setting unit 105 acquires the measured temperature, the set temperature, the operating status value, the outside air temperature, and the condensing temperature of the representative indoor unit acquired from the operation data storage unit 108, the degree of subcooling of the fixed value, and the required heating of the representative indoor unit. L2 and L2 are applied to the learning model 110 to set the target condensing temperature.
The setting unit 105 applies the minimum allowable degree of supercooling (for example, SC=5K) as the fixed value of the degree of supercooling.
Note that the target condensing temperature set by the setting unit 105 is the lowest condensing temperature at which intermittent operation does not occur in the representative indoor unit for a predetermined time, that is, the lowest temperature at which the heating capacity of the representative indoor unit does not fall short of the required heating capacity L2. is the condensation temperature. That is, the target condensing temperature is the lowest condensing temperature among the condensing temperatures that allow the heating capacity of the representative indoor unit to match the required heating capacity L2. "The heating capacity of the representative indoor unit can be matched with the required heating capacity L2" means matching the measured temperature in the air-conditioned space of the representative indoor unit with the set temperature in the air-conditioned space of the representative indoor unit. means.
The setting unit 105 uses the correlation equation of the learning model 110 to identify the lowest condensation temperature among the condensation temperatures that allow the heating capacity of the representative indoor unit to match the required heating capacity L2, and sets the identified condensation temperature to Set target condensing temperature.
The setting unit 105 notifies the calculation unit 106 of the target condensing temperature and a fixed degree of supercooling (for example, SC=5K).

Next, in step S205, the calculation unit 106 sets the target degree of supercooling of the representative indoor unit.
Specifically, the calculation unit 106 sets the degree of supercooling of the fixed value notified from the setting unit 105 as the target degree of supercooling of the representative indoor unit.

Next, in step S206, the calculation unit 106 calculates the degree of supercooling of each indoor unit 300 other than the representative indoor unit according to the above equation (2).
Specifically, the calculation unit 106 sets the required heating capacity L2 of each indoor unit 300 notified from the selection unit 103 to L2 of the equation (2). In addition, the calculation unit 106 sets the measured temperature measured in the air-conditioned space of each indoor unit 300 as the measured temperature in Equation (2). Further, the calculation unit 106 sets the target condensing temperature notified from the setting unit 105 to CT in Equation (2). Further, the calculation unit 106 sets Thermo in Equation (2) to a value of 1 (Thermo=1) in common among the indoor units 300 . Calculation unit 106 then calculates SC that matches the right side and left side of equation (2) as the target degree of supercooling. Note that the calculation unit 106 acquires the measured temperature of each indoor unit 300 from the operating data storage unit 108 .
The target supercooling degree calculated by the calculating unit 106 is the degree of supercooling that does not cause intermittent operation in each indoor unit 300 for a predetermined time when the condensation temperature in each indoor unit 300 is matched with the target condensation temperature. That is, it is the degree of subcooling that allows the heating capacity of each indoor unit 300 to match the required heating capacity L2 of each indoor unit 300 . "Match the heating capacity of each indoor unit 300 with the required heating capacity L2 of each indoor unit 300" means that the measured temperature in the air-conditioned space of each indoor unit 300 is set to the set temperature in the air-conditioned space of each indoor unit 300. means to match
The calculation unit 106 notifies the control unit 107 of the target condensation temperature and the target supercooling degree of each indoor unit 300 including the representative indoor unit.

Next, in step S207, the control unit 107 generates a control instruction.
The control unit 107 determines the operating speed of the compressor based on the target evaporation temperature. The operating speed of the compressor is adjusted by the outdoor unit 200 .
Also, the control unit 107 determines the degree of opening of the indoor expansion valve for each indoor unit 300 based on the target degree of subcooling of each indoor unit 300 . The opening degree of the indoor expansion valve is a different value for each indoor unit 300 . The degree of opening of the indoor expansion valve is adjusted by each indoor unit 300 .
The control unit 107 generates a control instruction to the outdoor unit 200 indicating the determined operating rotation speed of the compressor. Further, the control unit 107 generates a control instruction to each indoor unit 300 indicating the determined degree of opening of the indoor expansion valve.

Next, the control unit 107 outputs respective control instructions to the outdoor unit 200 and each indoor unit 300 .
The operation of the outdoor unit 200 and the plurality of indoor units 300 is controlled by the outdoor unit 200 and the indoor units 300 operating according to the control instructions.
That is, the control unit 107 can match the heating capacity of each indoor unit 300 with the required heating capacity of each indoor unit 300 by outputting the control instruction, and the intermittent operation of each indoor unit 300 does not occur.

Next, an operation example of the control device 100 in the learning phase will be described with reference to FIGS. 6 and 7. FIG.
For example, the control device 100 sets a period of one month as the period of the learning phase, and repeats the flow of FIG. 6 or 7 for one month.
FIG. 6 shows an operation example of the control device 100 when the air conditioner 400 is performing cooling operation.
FIG. 7 shows an operation example of the control device 100 when the air conditioner 400 is performing heating operation.

First, referring to FIG. 6, the operation of the control device 100 during cooling operation will be described.

In FIG. 6, first, in step S301, the collection unit 101 collects driving data.
The collection unit 101 collects, as operating data, the outside air temperature, the set temperature in each air-conditioned space, the measured temperature in each air-conditioned space, the evaporation temperature of each indoor unit 300, the degree of superheat of each indoor unit 300, and the temperature of each indoor unit 300. Get the health status value of .
In the learning phase, the outdoor unit 200 randomly changes the evaporation temperature. Also, each indoor unit 300 fixes the degree of superheat to a minimum value (for example, SH=2K).
The collection unit 101 outputs the collected driving data to the estimation unit 102 .

Next, in step S302 , the estimation unit 102 estimates the required cooling capacity of each indoor unit 300 .
The estimating unit 102 estimates the required cooling capacity L1 of each indoor unit 300 according to Equation (1) described above.
As described above, in the learning phase, the outdoor unit 200 randomly changes the evaporating temperature. Estimate L1.
The estimation unit 102 notifies the selection unit 103 of the required cooling capacity L1 of each indoor unit 300 for each evaporation temperature.

Next, in step S303, the selection unit 103 selects the learning indoor unit.
Specifically, the selecting unit 103 selects the indoor unit 300 with the maximum required cooling capacity L1 among the required cooling capacities L1 notified from the estimating unit 102 as the learning indoor unit.
The selection unit 103 notifies the learning unit 104 of the selected learning indoor unit.

Next, in step S304, the learning unit 104 sets a fixed degree of superheat.
Specifically, the learning unit 104 sets the degree of superheat to a minimum value (for example, SH=2K).

Next, in step S305, the learning unit 104 performs machine learning.
The learning unit 104 uses the operating data of the learning indoor unit and the fixed superheat degree set in step S304 to determine the highest evaporating temperature at which intermittent operation does not occur in the learning indoor unit, that is, the cooling capacity of the learning indoor unit. The highest evaporating temperature among the evaporating temperatures that can match the maximum required cooling capacity L1 is learned. "The cooling capacity of the learning indoor unit can match the maximum required cooling capacity L1" means that when the learning indoor unit is operated at the maximum required cooling capacity L1, the measurement in the air-conditioned space of the learning indoor unit It means matching the temperature with the set temperature in the air-conditioned space of the learning indoor unit.
As described above, the learning unit 104 uses the operating data of the learning indoor unit as machine learning to obtain input (set temperature, measured temperature, operation status value, evaporation temperature, outside temperature, degree of superheat) and output (learning indoor unit (air conditioner cooling capacity).
The learning unit 104 may perform either supervised learning or unsupervised learning.

Finally, in step S306, the learning unit 104 generates the learning model 110 that reflects the results of the machine learning performed in step S305.

In FIG. 7, first, in step S401, the collection unit 101 collects driving data.
The collection unit 101 collects, as operating data, the outside air temperature, the set temperature in each air-conditioned space, the measured temperature in each air-conditioned space, the condensation temperature of each indoor unit 300, the degree of supercooling of each indoor unit 300, each indoor unit 300 operating status values are obtained.
In the learning phase, the outdoor unit 200 randomly changes the condensation temperature. Also, each indoor unit 300 fixes the degree of supercooling to a minimum value (for example, SC=5K).
The collection unit 101 outputs the collected driving data to the estimation unit 102 .

Next, in step S402 , the estimation unit 102 estimates the required heating capacity of each indoor unit 300 .
The estimation unit 102 estimates the required heating capacity L2 of each indoor unit 300 according to the above equation (2).
As described above, in the learning phase, the outdoor unit 200 randomly changes the condensing temperature. Estimate L2.
The estimation unit 102 notifies the selection unit 103 of the required heating capacity L2 of each indoor unit 300 for each condensation temperature.

Next, in step S403, the selection unit 103 selects the learning indoor unit.
Specifically, the selecting unit 103 selects the indoor unit 300 with the maximum required heating capacity L2 among the required heating capacities L2 notified from the estimating unit 102 as the learning indoor unit.
The selection unit 103 notifies the learning unit 104 of the selected learning indoor unit.

Next, in step S404, the learning unit 104 sets a fixed degree of supercooling.
Specifically, the learning unit 104 sets the degree of supercooling to a minimum value (for example, SC=5K).

Next, in step S405, the learning unit 104 performs machine learning.
The learning unit 104 uses the operating data of the learning indoor unit and the fixed value of supercooling degree set in step S404 to determine the lowest condensing temperature at which intermittent operation does not occur in the learning indoor unit, that is, the heating capacity of the learning indoor unit. is the lowest condensing temperature among the condensing temperatures that can match the maximum required heating capacity L2. "The heating capacity of the learning indoor unit can be matched to the maximum required heating capacity L2" means that when the learning indoor unit is operated at the maximum required heating capacity L2, the measurement in the air conditioning space of the learning indoor unit It means matching the temperature with the set temperature in the air-conditioned space of the learning indoor unit.
As described above, as machine learning, the learning unit 104 uses the operating data of the learning indoor unit to obtain input (set temperature, measured temperature, operation status value, condensing temperature, outside air temperature, degree of supercooling) and output (learning Calculate the correlation with the heating capacity of the indoor unit).
The learning unit 104 may perform either supervised learning or unsupervised learning.

Finally, in step S406, the learning unit 104 generates the learning model 110 that reflects the results of the machine learning performed in step S405.

***Description of the effect of the embodiment***
As described above, in the present embodiment, an appropriate target temperature (evaporating temperature or condensing temperature) is set early by using a learning model. Further, in the present embodiment, an appropriate degree of superheating or supercooling that does not cause intermittent operation in each indoor unit is calculated for each indoor unit based on the set target temperature.
Therefore, according to the present embodiment, it is possible to prevent the occurrence of intermittent operation in each indoor unit and prevent the deterioration of the operating efficiency. In addition, since intermittent operation of each indoor unit is prevented, fluctuations in room temperature are suppressed and comfort is improved.

Further, according to the present embodiment, it is possible to operate the indoor unit at an appropriate evaporation temperature that does not cause insufficient cooling capacity.
Similarly, according to the present embodiment, it is possible to operate the indoor unit at an appropriate condensation temperature that does not cause insufficient heating capacity.
Therefore, according to the present embodiment, it is possible to save energy while maintaining comfort.

Also, in the present embodiment, machine learning is performed using only the parameters of the learning indoor unit, which is the indoor unit with the maximum required cooling capacity or required heating capacity. Therefore, according to this embodiment, machine learning can be completed in a short time.

Embodiment 2.
In Embodiment 1, the setting unit 105 sets the target evaporation temperature by applying a fixed value of the degree of superheat (for example, SH=2K) to the learning model 110 . Then, the calculation unit 106 sets the degree of superheat of the fixed value as the target degree of superheat of the representative indoor unit. Further, in Embodiment 1, the setting unit 105 applies a fixed degree of supercooling (for example, SC=5K) to the learning model 110 to set the target condensing temperature. Then, the calculation unit 106 sets the fixed value of the subcooling degree as the target cooling degree of the representative indoor unit.
In the present embodiment, an example will be described in which setting unit 105 uses learning model 110 to derive a target evaporating temperature and an appropriate degree of superheat as the target degree of superheat of the representative indoor unit. Further, in the present embodiment, an example will be described in which the setting unit 105 uses the learning model 110 to derive the target condensing temperature and an appropriate supercooling degree as the target supercooling degree of the representative indoor unit.

In this embodiment, differences from the first embodiment will be mainly described.
Matters not described below are the same as those in the first embodiment.

*** Configuration description ***
A configuration example of an air conditioning system 500 according to the present embodiment is as shown in FIG.
FIG. 2 shows an example of the functional configuration of the control device 100 according to this embodiment. However, as described below, the operations of the learning unit 104, the setting unit 105, and the calculation unit 106 are different from those in the first embodiment.
A hardware configuration example of the control device 100 according to the present embodiment is as shown in FIG.

***Description of operation***
8 and 9 show an operation example in the operation phase of the control device 100 according to this embodiment.
FIG. 8 shows an operation example of the control device 100 when the air conditioner 400 is performing cooling operation. FIG. 8 corresponds to FIG. 4 shown in the first embodiment.
FIG. 9 shows an operation example of the control device 100 when the air conditioner 400 is performing heating operation. FIG. 9 corresponds to FIG. 5 shown in the first embodiment.

First, referring to FIG. 8, the operation of the control device 100 during cooling operation will be described.

　Steps S101 to S103 in FIG. 8 are the same as those shown in FIG. Therefore, the description is omitted.

In step S115, the setting unit 105 sets the target evaporation temperature and the target degree of superheat of the representative indoor unit.
The setting unit 105 acquires the operating data of the representative indoor unit from the operating data storage unit 108 . Specifically, the learning unit 104 acquires the measured temperature, set temperature, operating status value, outside air temperature, and evaporating temperature of the representative indoor unit from the operating data storage unit 108 .
Then, the setting unit 105 stores the measured temperature, set temperature, operating status value, outside temperature, and evaporation temperature of the representative indoor unit acquired from the operation data storage unit 108, and the required cooling capacity L1 of the representative indoor unit in the learning model 110. Apply to set the target evaporating temperature and the target superheat of the representative indoor unit.
The target evaporating temperature and target degree of superheat set in step S115 allow the cooling capacity of the representative indoor unit to match the required cooling capacity L1 for a predetermined period of time, and further minimize the power consumption of the representative indoor unit. It is the combination of the optimum evaporation temperature and superheating temperature and superheating degree. That is, the target evaporating temperature and target degree of superheat set in step S115 are a combination of the highest evaporating temperature and the highest degree of superheat that satisfy these conditions.
The setting unit 105 notifies the calculation unit 106 of the set target evaporation temperature and the target superheat degree of the representative indoor unit.

Steps S106 to S108 are the same as those shown in FIG. Therefore, the description is omitted.

　Steps S201 to S203 in FIG. 9 are the same as those shown in FIG. Therefore, the description is omitted.

In step S215, the setting unit 105 sets the target condensation temperature and the target supercooling degree of the representative indoor unit.
The setting unit 105 acquires the operating data of the representative indoor unit from the operating data storage unit 108 . Specifically, the learning unit 104 acquires the measured temperature, set temperature, operating status value, outside air temperature, and condensing temperature of the representative indoor unit from the operating data storage unit 108 .
Then, the setting unit 105 stores the measured temperature, the set temperature, the operating status value, the outside temperature, and the condensation temperature of the representative indoor unit acquired from the operation data storage unit 108, and the required heating capacity L2 of the representative indoor unit in the learning model 110. Apply to set the target condensing temperature and the target subcooling degree of the representative indoor unit.
The target condensing temperature and the target supercooling degree set in step S215 can match the heating capacity of the representative indoor unit with the required heating capacity L2 for a predetermined time, and furthermore, can minimize the power consumption of the representative indoor unit. It is the optimum combination of the condensation temperature and the degree of supercooling among the condensation temperature and the degree of supercooling. That is, the target condensing temperature and the target supercooling degree set in step S215 are a combination of the highest condensing temperature and the highest supercooling degree that satisfy these conditions.
The setting unit 105 notifies the calculation unit 106 of the set target condensation temperature and the target degree of subcooling of the representative indoor unit.

Steps S206 to S208 are the same as those shown in FIG. Therefore, the description is omitted.

Next, an operation example in the learning phase of control device 100 according to the present embodiment will be described with reference to FIGS. 10 and 11. FIG.
In this embodiment as well, the control device 100 repeats the flow of FIG. 10 or 11 for one month, for example, with a period of one month as the period of the learning phase.
FIG. 10 shows an operation example of the control device 100 when the air conditioner 400 is performing cooling operation.
FIG. 11 shows an operation example of the control device 100 when the air conditioner 400 is performing heating operation.

First, referring to FIG. 10, the operation of the control device 100 during cooling operation will be described.

In FIG. 10, in step S311, the collection unit 101 collects driving data.
The collection unit 101 collects, as operating data, the outside air temperature, the set temperature in each air-conditioned space, the measured temperature in each air-conditioned space, the evaporation temperature of each indoor unit 300, the degree of superheat of each indoor unit 300, and the temperature of each indoor unit 300. and the power consumption value of each indoor unit 300 are acquired.
Unlike step S301 in FIG. 6, the collection unit 101 also acquires the power consumption value of each indoor unit 300 in step S311.
In step S311, the outdoor unit 200 randomly changes the evaporation temperature, and each indoor unit 300 randomly changes the degree of superheat. In step S301 of FIG. 6, each indoor unit 300 fixes the degree of superheat to the minimum value (for example, SH=2K), but in step S311, each indoor unit 300 randomly changes the degree of superheat.
The collection unit 101 outputs the collected driving data to the estimation unit 102 .

Next, in step S312 , the estimation unit 102 estimates the required cooling capacity of each indoor unit 300 .
The estimating unit 102 estimates the required cooling capacity L1 of each indoor unit 300 according to Equation (1) described above.
As described above, in the learning phase, the outdoor unit 200 randomly changes the evaporating temperature, and each indoor unit 300 randomly changes the degree of superheat. The required cooling capacity L1 of each indoor unit 300 is estimated for each combination.
The estimation unit 102 notifies the selection unit 103 of the required cooling capacity L1 of each indoor unit 300 for each combination of the evaporation temperature and the degree of superheat.

　Step S303 is the same as that shown in FIG. Therefore, the description is omitted.

Next, in step S315, the learning unit 104 performs machine learning.
The learning unit 104 uses the operation data of the learned indoor unit to match the cooling capacity of the learned indoor unit with the maximum required cooling capacity L1, and furthermore, the evaporation temperature that can minimize the power consumption of the learned indoor unit. Learn the optimum combination of evaporation temperature and degree of superheat within the degree of superheat. That is, the learning unit 104 learns the combination of the highest evaporation temperature and the highest degree of superheat that satisfy such conditions.
The learning unit 104 may perform either supervised learning or unsupervised learning.

　Step S306 is the same as that shown in FIG. Therefore, the description is omitted.

In FIG. 11, in step S411, the collection unit 101 collects driving data.
The collection unit 101 collects, as operating data, the outside air temperature, the set temperature in each air-conditioned space, the measured temperature in each air-conditioned space, the condensation temperature of each indoor unit 300, the degree of supercooling of each indoor unit 300, each indoor unit 300 and the power consumption value of each indoor unit 300 are acquired.
Unlike step S401 in FIG. 7, the collection unit 101 also acquires the power consumption value of each indoor unit 300 in step S411.
In step S411, the outdoor unit 200 randomly changes the condensation temperature, and each indoor unit 300 randomly changes the supercooling degree. In step S401 of FIG. 7, each indoor unit 300 fixes the degree of supercooling to the minimum value (for example, SC=5K), but in step S411, each indoor unit 300 randomly changes the degree of supercooling.
The collection unit 101 outputs the collected driving data to the estimation unit 102 .

Next, in step S412 , the estimation unit 102 estimates the required heating capacity of each indoor unit 300 .
The estimation unit 102 estimates the required heating capacity L2 of each indoor unit 300 according to the above equation (2).
As described above, in the learning phase, the outdoor unit 200 randomly changes the condensing temperature, and each indoor unit 300 randomly changes the degree of supercooling. The required heating capacity L2 of each indoor unit 300 is estimated for each combination of degrees.
The estimation unit 102 notifies the selection unit 103 of the required heating capacity L2 of each indoor unit 300 for each combination of the condensation temperature and the degree of supercooling.

　Step S403 is the same as that shown in FIG. Therefore, the description is omitted.

Next, in step S415, the learning unit 104 performs machine learning.
The learning unit 104 uses the operation data of the learning indoor unit to match the heating capacity of the learning indoor unit to the maximum required heating capacity L2, and furthermore, the condensing temperature that can minimize the power consumption of the learning indoor unit. Learn the optimum combination of condensing temperature and degree of supercooling among degrees of supercooling. That is, the learning unit 104 learns the combination of the highest condensing temperature and the highest degree of supercooling that satisfy such conditions.
The learning unit 104 may perform either supervised learning or unsupervised learning.

　Step S406 is the same as that shown in FIG. Therefore, the description is omitted.

***Description of the effect of the embodiment***
According to this embodiment, the target degree of superheat of the representative indoor unit can be set to an appropriate degree of superheat. Similarly, according to the present embodiment, the target degree of supercooling of the representative indoor unit can be set to an appropriate degree of supercooling.
Moreover, according to this embodiment, the power consumption of each indoor unit can be suppressed.

Embodiment 3.
In Embodiments 1 and 2, the example in which the required cooling capacity L1 is calculated according to the formula (1) has been described. Further, in Embodiments 1 and 2, the example in which the required heating capacity L2 is calculated according to the equation (2) has been described.
However, the required cooling capacity L1 may be calculated using another formula instead of the formula (1). Moreover, the required heating capacity L2 may be calculated using another formula instead of the formula (2).
In this embodiment, an example using formulas other than formulas (1) and (2) will be described.

*** Configuration description ***
A configuration example of an air conditioning system 500 according to the present embodiment is as shown in FIG.
FIG. 2 shows an example of the functional configuration of the control device 100 according to this embodiment. However, as described below, the operations of the estimation unit 102, the learning unit 104, and the calculation unit 106 are different from those in the first embodiment.
A hardware configuration example of the control device 100 according to the present embodiment is as shown in FIG.

***Description of operation***
Also in this embodiment, the control device 100 performs the operations shown in FIGS. 4 and 5 as operations in the operation phase.

In step S101, the collection unit 101 collects driving data. In the present embodiment, the collecting unit 101 collects values described later in addition to the values collected in the first embodiment.

Also, in the present embodiment, the estimation unit 102 estimates the required cooling capacity L1 [kW] of each indoor unit 300 in accordance with the following equations (3) and (4) in step S102.
L1 = (heo-hei) × Gr formula (3)
Gr = 7.59 × 10 ^-4 × Cv × ((Ph-Ps) × ρl × 1000) ^0.5
Formula (4)

Here, hei [kJ/kg] in equation (3) is the enthalpy at the inlet of the evaporator of the indoor unit 300 . In addition, heo [kJ/kg] in equation (3) is the enthalpy at the outlet of the evaporator of the indoor unit 300 . hei is determined from the outlet temperature of the condenser of the outdoor unit 200 . heo is determined from the outlet temperature of the condenser of the outdoor unit 200 and Ps.
Also, Gr [kg/s] in Equation (3) is the amount of refrigerant flowing through the indoor unit 300 . Gr can be calculated by Equation (4).
In Equation (4), Ph [MPa] is the high pressure value (the refrigerant discharge pressure value in the compressor). Ps [MPa] is the low pressure value (refrigerant suction pressure value in the compressor). Cv is an index representing the ease of fluid flow. Cv can be obtained from the indoor unit expansion valve opening Li [Pulse]. The relationship between Li and Cv is determined by the expansion valve. It is assumed that the control device 100 holds data indicating the relationship between Li and Cv in advance as a database.
ρl in equation (4) is the density of the refrigerant at the inlet of the expansion valve. ρl is determined from the outlet temperature of the condenser of the outdoor unit 200 .
The collection unit 101 collects Ph, Ps, Li, and the outlet temperature of the condenser as operating data in addition to the values shown in the first embodiment.
The estimating unit 102 calculates hei in Equation (3) from the outlet temperature of the condenser obtained as the operating data. The estimating unit 102 also calculates heo from the condenser outlet temperature and Ps obtained as the operating data. Also, the estimating unit 102 obtains Cv in Equation (4) using Li obtained as the operating data and data in the database. The estimating unit 102 also calculates ρl in Equation (4) from the outlet temperature of the condenser obtained as the operating data.
Estimating section 102 then calculates Gr according to equation (4). Further, estimation unit 102 calculates required cooling capacity L1 from calculated Gr, hei, and heo according to equation (3).

Steps S103 to S105 in FIG. 4 are the same as those shown in the first embodiment. Therefore, the description is omitted.

In step S106, the calculation unit 106 calculates the degree of superheat of each indoor unit 300 other than the representative indoor unit according to the formula (1) shown in the first embodiment.
In this embodiment, the value set for L1 in Equation (1) is the value of L1 calculated according to Equation (3) in step S102.
Other values of formula (1) are the same as those shown in the first embodiment.

Steps S107 and S108 in FIG. 4 are the same as those shown in the first embodiment. Therefore, the description is omitted.

In step S201, the collection unit 101 collects driving data. In the present embodiment, the collecting unit 101 collects values described later in addition to the values collected in the first embodiment.

In the present embodiment, estimation unit 102 estimates required heating capacity L2 [kW] of each indoor unit 300 in step S202 according to the following equations (5) and (6).
L2 = (hco-hci) × Gr Formula (5)
Gr = 7.59 × 10 ^-4 × Cv × ((Ph-Ps) × ρl × 1000) ^0.5
Formula (6)

Here, hci [kJ/kg] in equation (5) is the inlet enthalpy of the condenser of the indoor unit 300 . Also, hco [kJ/kg] in equation (5) is the enthalpy at the outlet of the condenser of the indoor unit 300 . hci is determined from the inlet temperature of the evaporator of the outdoor unit 200 and Ph. hco is determined from the inlet temperature of the evaporator of the outdoor unit 200 .
In addition, Gr [kg/s] in Equation (5) is the amount of refrigerant flowing through indoor unit 300 . Gr can be calculated by Equation (6). ρl in equation (6) is determined from the inlet temperature of the evaporator of the outdoor unit 200 . Other values in equation (6) are the same as in equation (4).
The collection unit 101 collects Ph, Ps, Li, and the inlet temperature of the evaporator as operation data in addition to the values shown in the first embodiment.
The estimation unit 102 calculates hci in Equation (5) from the evaporator inlet temperature and Ph obtained as the operating data. The estimating unit 102 also calculates hco from the inlet temperature of the evaporator obtained as the operating data. Also, the estimating unit 102 obtains Cv in Equation (6) using Li obtained as the operating data and data in the database. The estimating unit 102 also calculates ρl in Equation (6) from the evaporator inlet temperature obtained as the operating data.
Estimating section 102 then calculates Gr according to equation (6). Further, estimation unit 102 calculates requested heating capacity L2 from calculated Gr, hci, and hco according to equation (5).

Steps S203 to S205 in FIG. 5 are the same as those shown in the first embodiment. Therefore, the description is omitted.

In step S206, the calculation unit 106 calculates the degree of supercooling of each indoor unit 300 other than the representative indoor unit according to the formula (2) shown in the first embodiment.
Note that in the present embodiment, the value set for L2 in Equation (2) is the value of L2 calculated according to Equation (5) in step S202.
Other values of formula (2) are the same as those shown in the first embodiment.

Steps S207 and S208 in FIG. 5 are the same as those shown in the first embodiment. Therefore, the description is omitted.

Also in this embodiment, the control device 100 performs the operations shown in FIGS. 6 and 7 as operations in the learning phase.

In step S301, the collection unit 101 collects driving data. The collection unit 101 collects the same driving data as that collected in step S101 in this embodiment.

Also, in step S302, the estimation unit 102 estimates the required cooling capacity L1 of each indoor unit 300 for each evaporating temperature according to Equation (3).
The operations after step S303 are the same as those shown in the first embodiment. Therefore, the description is omitted.

In step S401, the collection unit 101 collects driving data. The collection unit 101 collects the same driving data as that collected in step S201 in this embodiment.

Also, in step S402, the estimation unit 102 estimates the required heating capacity L2 of each indoor unit 300 according to the equation (2) for each condensing temperature.
The operations after step S403 are the same as those shown in the first embodiment. Therefore, the description is omitted.

***Description of the effect of the embodiment***
In the present embodiment, required cooling capacity L1 is calculated using equations (3) and (4) instead of equation (1). By using equations (3) and (4), the required cooling capacity L1 can be calculated more accurately than when using equation (1). Similarly, in the present embodiment, required heating capacity L2 is calculated using equations (5) and (6) instead of equation (2). By using the equations (5) and (6), the required heating capacity L2 can be calculated more accurately than when using the equation (2).
Therefore, according to the present embodiment, each indoor unit 300 can be controlled more precisely than in the first embodiment.

Although the first to third embodiments have been described above, two or more of these embodiments may be combined for implementation.
Alternatively, one of these embodiments may be partially implemented.
Alternatively, two or more of these embodiments may be partially combined for implementation.
Also, the configurations and procedures described in these embodiments may be changed as necessary.

*** Supplementary explanation of hardware configuration ***
Finally, a supplementary description of the hardware configuration of the control device 100 will be given.
A processor 901 shown in FIG. 3 is an IC (Integrated Circuit) that performs processing.
The processor 901 is a CPU (Central Processing Unit), a DSP (Digital Signal Processor), or the like.
The main memory device 902 shown in FIG. 3 is a RAM (Random Access Memory).
The auxiliary storage device 903 shown in FIG. 3 is a ROM (Read Only Memory), flash memory, HDD (Hard Disk Drive), or the like.
The communication device 904 shown in FIG. 3 is an electronic circuit that performs data communication processing.
The communication device 904 is, for example, a communication chip or a NIC (Network Interface Card).
The input/output device 905 is a keyboard, mouse, display, and the like.

The auxiliary storage device 903 also stores an OS (Operating System).
At least part of the OS is executed by the processor 901 .
The processor 901 executes a program that implements the functions of the collection unit 101, the estimation unit 102, the selection unit 103, the learning unit 104, the setting unit 105, the calculation unit 106, and the control unit 107 while executing at least part of the OS.
Task management, memory management, file management, communication control, and the like are performed by the processor 901 executing the OS.
In addition, at least one of information, data, signal values, and variable values indicating the processing results of the collection unit 101, the estimation unit 102, the selection unit 103, the learning unit 104, the setting unit 105, the calculation unit 106, and the control unit 107, It is stored in at least one of a main memory device 902, an auxiliary memory device 903, a register in the processor 901, and a cache memory.
Also, a program that implements the functions of the collection unit 101, the estimation unit 102, the selection unit 103, the learning unit 104, the setting unit 105, the calculation unit 106, and the control unit 107 can be a magnetic disk, a flexible disk, an optical disk, a compact disk, a Blu-ray ( It may be stored in a portable recording medium such as a registered trademark) disk, DVD, or the like. A portable recording medium in which a program for realizing the functions of the collection unit 101, the estimation unit 102, the selection unit 103, the learning unit 104, the setting unit 105, the calculation unit 106, and the control unit 107 is stored may be distributed.

In addition, the “units” of the collecting unit 101, the estimating unit 102, the selecting unit 103, the learning unit 104, the setting unit 105, the calculating unit 106, and the control unit 107 are replaced with “circuit”, “process”, “procedure”, or “processing”. Or you may read it as "circuitry".
Also, the control device 100 may be realized by a processing circuit. The processing circuits are, for example, logic ICs (Integrated Circuits), GAs (Gate Arrays), ASICs (Application Specific Integrated Circuits), and FPGAs (Field-Programmable Gate Arrays).
In this case, the collection unit 101, the estimation unit 102, the selection unit 103, the learning unit 104, the setting unit 105, the calculation unit 106, and the control unit 107 are each implemented as part of the processing circuit.
In this specification, the general concept of processors and processing circuits is referred to as "processing circuitry."
Thus, processors and processing circuitry are each examples of "processing circuitry."

100 control device, 101 collection unit, 102 estimation unit, 103 selection unit, 104 learning unit, 105 setting unit, 106 calculation unit, 107 control unit, 108 operation data storage unit, 110 learning model, 200 outdoor unit, 300 indoor unit, 400 air conditioner, 500 air conditioning system, 901 processor, 902 main storage device, 903 auxiliary storage device, 904 communication device, 905 input/output device.

Claims

A representative of the plurality of indoor units, each of which is assigned an air-conditioned space to be air-conditioned, based on the required air-conditioning capacity, which is the air-conditioning capacity required for each indoor unit. a selection unit that selects an indoor unit to be used as a representative indoor unit;
A learning model obtained by machine learning is used to set a target temperature of either the evaporating temperature or the condensing temperature that allows the air conditioning capacity of the representative indoor unit to match the required air conditioning capacity of the representative indoor unit. a setting unit;
For each indoor unit of the plurality of indoor units excluding the representative indoor unit, when either the evaporation temperature or the condensation temperature in each indoor unit is made to match the target temperature, the air conditioning capacity of each indoor unit is A control device having a calculation unit that calculates either the degree of superheating or the degree of supercooling that matches the required air conditioning capacity of the indoor unit.
The control device further
an estimating unit for estimating a required air conditioning capacity of each indoor unit of the plurality of indoor units;
The selection unit
The control device according to claim 1, wherein an indoor unit having a maximum required air conditioning capacity is selected from among the plurality of indoor units as the representative indoor unit.
The setting unit
Evaporating temperature and condensing temperature that can match the air conditioning capacity of the learning indoor unit, which is the indoor unit with the largest required air conditioning capacity among the plurality of indoor units in the learning phase, to the required air conditioning capacity of the learning indoor unit. 3. The control device according to claim 2, wherein the target temperature is set using a learning model obtained by learning either one.
The control device further
a collection unit that collects the measured temperature measured in the air-conditioned space of each indoor unit;
an estimating unit for estimating the required air conditioning capacity of each indoor unit;
The calculation unit
For each indoor unit of the plurality of indoor units excluding the representative indoor unit, using the required air conditioning capacity of each indoor unit, the measured temperature of the air-conditioned space of each indoor unit, and the target temperature, each indoor unit 2. The control device according to claim 1, which calculates either the degree of superheat or the degree of supercooling.
The setting unit
A set temperature in the air-conditioned space of the representative indoor unit, a measured temperature measured in the air-conditioned space of the representative indoor unit, an operation status value indicating the operation status of the representative indoor unit, and the temperature measured by the representative indoor unit. 2. The target temperature is set by applying one of the obtained evaporation temperature and condensation temperature, the outside air temperature, and either the fixed degree of superheat or the fixed degree of subcooling to the learning model. Control device as described.
The control device further
The representative indoor unit is controlled using the target temperature and either the fixed degree of superheating or the fixed degree of supercooling, and for each indoor unit of the plurality of indoor units excluding the representative indoor unit 6. The control device according to claim 5, further comprising a control unit that controls each indoor unit using the target temperature and one of the degree of superheating and the degree of supercooling calculated for each indoor unit.
The setting unit
Evaporation temperature and condensation that can match the air conditioning capacity of the representative indoor unit to the required air conditioning capacity of the representative indoor unit and minimize the power consumption of the representative indoor unit using the learning model 2. The control device according to claim 1, which sets a target temperature for any of the temperatures.
The setting unit
In the learning phase, the air conditioning capacity of the learning indoor unit, which is the indoor unit with the largest required air conditioning capacity among the plurality of indoor units, can be matched with the required air conditioning capacity of the learning indoor unit, and Using the learning model obtained by learning either the combination of the evaporation temperature and the degree of superheat and the combination of the condensation temperature and the degree of supercooling that can minimize the power consumption, the air conditioning capacity of the representative indoor unit is improved. Derive a combination of the evaporation temperature and the degree of superheat and the combination of the condensation temperature and the degree of supercooling that can match the required air conditioning capacity of the representative indoor unit and minimize the power consumption of the representative indoor unit. 8. The control device according to claim 7, wherein one of the derived evaporation temperature and condensation temperature is set as the target temperature.
The control device further
The representative indoor unit is controlled using the target temperature and either the degree of superheating or the degree of supercooling derived from the learning model, and for each indoor unit of the plurality of indoor units excluding the representative indoor unit, The control device according to claim 8, further comprising a control unit that controls each indoor unit using the target temperature and one of the degree of superheating and the degree of supercooling calculated for each indoor unit.
The setting unit
Evaporation temperature at which the cooling capacity of the representative indoor unit can be matched with the cooling capacity required for the representative indoor unit when the plurality of indoor units perform cooling operation for air conditioning of each air conditioning space. set the target temperature of
When the plurality of indoor units perform heating operation for air conditioning of each air-conditioned space, the condensing temperature at which the heating capacity of the representative indoor unit can be matched with the heating capacity required of the representative indoor unit. set the target temperature of
The calculation unit
When the plurality of indoor units performs cooling operation for air conditioning of each air-conditioned space, for each indoor unit of the plurality of indoor units excluding the representative indoor unit, the evaporation temperature in each indoor unit is set to the above Calculate the degree of superheat at which the cooling capacity of each indoor unit matches the cooling capacity required for the indoor unit when the target temperature is matched,
When the plurality of indoor units perform heating operation for air conditioning of each air-conditioned space, for each indoor unit of the plurality of indoor units excluding the representative indoor unit, the condensing temperature in each indoor unit is set to the above 2. The control device according to claim 1, wherein when the temperature is matched with the target temperature, the degree of supercooling is calculated such that the heating capacity of each indoor unit matches the heating capacity required of the indoor unit.
A computer selects a plurality of indoor units, each of which is assigned an air-conditioned space to be air-conditioned, based on the required air-conditioning capacity, which is the air-conditioning capacity required for each indoor unit. Select the indoor unit that represents the machine as the representative indoor unit,
A target of either an evaporating temperature or a condensing temperature that allows the computer to match the air conditioning capacity of the representative indoor unit with the required air conditioning capacity of the representative indoor unit using a learning model obtained by machine learning. set the temperature and
When the computer matches one of the evaporation temperature and the condensation temperature in each indoor unit to the target temperature for each indoor unit of the plurality of indoor units excluding the representative indoor unit, the air in each indoor unit A control method for calculating either the degree of superheating or the degree of supercooling at which the air conditioning capacity matches the required air conditioning capacity of the indoor unit.