JP6752711B2 - How to build a heat exchange system, controller, and neural network - Google Patents

Description

The present invention relates to a heat exchange system, a controller, and a method for constructing a neural network.

For example, Patent Document 1 includes a neural network that learns operation characteristics due to a difference in refrigeration cycle, and a control means that controls the operation of an air conditioner based on the learning result of the neural network. The device is disclosed.

Japanese Unexamined Patent Publication No. 5-10568

However, in the above-mentioned Patent Document 1, for example, control of each of a plurality of devices by a plurality of PID controllers 1 is not considered, and therefore, one PID control of each PID control for each of the plurality of devices is performed. Points that affect other PID controls are not considered. Further, in Patent Document 1, for example, what kind of PID control is performed for a plurality of target values and what kind of learning is performed is not considered. Therefore, conventionally, it may not be possible to perform favorable operation control.

An object of the present invention is to provide a heat exchange system that performs suitable control for heat exchange, a controller that performs suitable control for heat exchange, and a method for constructing a neural network that performs suitable control for heat exchange.

(1) In order to achieve the above object, the heat exchange system according to the first aspect of the present invention is
A heat exchange system (for example, heat exchange system 100) that exchanges heat using a circulating heat medium (for example, a refrigerant).
A first adjusting device (for example, an inverter 102 or an evaporation pressure adjusting valve 109) for adjusting a first flow rate, which is a flow rate of the heat medium at the first position,
A second adjusting device (for example, an electronic expansion valve 106) that adjusts a second flow rate, which is the flow rate of the heat medium at the second position,
A first physical quantity (for example, the temperature of the blown air to be fed back) that changes according to the first flow rate is input, and the first physical quantity is set to a first target value (for example, blown air preset by an operation from the user or the like). The first feedback control unit (for example, PID control unit 11 or PID control unit) that performs the first feedback control (for example, PID control) that outputs the first operation amount to the first adjustment device so as to match the temperature set value). 12) and
A second physical quantity (for example, the degree of superheat to be fed back) that changes according to the second flow rate is input, and the second physical quantity is set to a second target value (for example, a superheat degree setting preset by an operation from the user or the like). It is a second feedback control (for example, PID control) that outputs a second operation amount to the second adjusting device so as to match the value), and affects the first feedback control (for example, PID control unit 13). A second feedback control unit (for example, PID control unit 13) that performs a second feedback control (which affects the PID control by the PID control unit 11 or 12 by changing the temperature of the blown air)
A plurality of state quantities that affect heat exchange by this heat exchange system other than the first physical quantity and the second physical quantity, the first target value, and the second target value are input, and the first physical quantity is input . A machine learning unit that controls the first adjusting device and the second adjusting device so that the physical quantity matches the first target value and the second physical quantity matches the second target value. For example, a plurality of input values including a blown air temperature set value, a superheat degree set value, and a state quantity that affects the heat exchange are input, and an operation quantity is calculated based on the input input values, and the calculated operation is performed. A neural network 15) that outputs an amount to the inverter 102 or the evaporative pressure regulating valve 109 and the electronic expansion valve 106, and the first manipulated quantity and the second manipulated quantity are used as teacher signals for the first manipulated quantity and the first manipulated quantity. learning to reduce the second operation amount have rows (e.g., learning performed as each operation amount can output a manipulated variable such that 0 from the PID controller 11 to 13), on the basis of the inputs and results of the learning , A machine learning unit (for example, a neural work 15) that outputs a third operation amount for operating the first adjustment device and a fourth operation amount for operating the second adjustment device .
A first adder that adds the first manipulated variable and the third manipulated variable and supplies it to the first adjusting device.
A second adder that adds the second manipulated variable and the fourth manipulated variable and supplies it to the second adjusting device.
Equipped with a,
The first feedback control unit performs the first feedback control with the same parameters regardless of the value set as the first target value.
The second feedback control unit performs the second feedback control with the same parameters regardless of the value set as the second target value.
It is a heat exchange system.

(2) The heat exchange system of (1) above is
A compressor that compresses the heat medium (for example, compressor 101) and
An inverter (for example, an inverter 102) that adjusts the output of the compressor,
A condenser (for example, condenser 103) that condenses the heat medium compressed by the compressor, and
An expansion valve (for example, an electronic expansion valve 106) that expands the heat medium in which the condenser is condensed,
An evaporator (for example, an evaporator 108) that exchanges heat by evaporating the heat medium expanded by the expansion valve.
A regulating valve (for example, an evaporation pressure regulating valve 109) for adjusting the flow rate of the heat medium evaporated by the evaporator is provided.
One of the first adjusting device and the second adjusting device has the inverter or the adjusting valve, and the other has the expansion valve.
The machine learning unit has a neural network (for example, a neural network 15).
You may do so.

(3) The controller according to the second aspect of the present invention is
A first adjusting device (for example, an inverter 102 or an evaporation pressure adjusting valve 109) that adjusts a first flow rate, which is a flow rate at a first position of a heat medium (for example, a refrigerant) that circulates and is used for heat exchange, and the heat. A controller (for example, controller 120) that controls a second adjusting device (for example, the electronic expansion valve 106) that adjusts the second flow rate, which is the flow rate at the second position of the medium.
A first physical quantity (for example, the temperature of the blown air to be fed back) that changes according to the first flow rate is input, and the first physical quantity is set to a first target value (for example, blown air preset by an operation from the user or the like). The first feedback control unit (for example, PID control unit 11 or PID control unit) that performs the first feedback control (for example, PID control) that outputs the first operation amount to the first adjustment device so as to match the temperature set value). 12) and
A second physical quantity (for example, the degree of superheat to be fed back) that changes according to the second flow rate is input, and the second physical quantity is set to a second target value (for example, a superheat degree setting preset by an operation from the user or the like). It is a second feedback control (for example, PID control) that outputs a second operation amount to the second adjusting device so as to match the value), and affects the first feedback control (for example, PID control unit 13). A second feedback control unit (for example, PID control unit 13) that performs a second feedback control (which affects the PID control by the PID control unit 11 or 12 by changing the temperature of the blown air)
A plurality of state quantities other than the first physical quantity and the second physical quantity, the first target value, and the second target value are input so that the first physical quantity matches the first target value. In addition, a machine learning unit that controls the first adjusting device and the second adjusting device (for example, blown air temperature setting value and overheating degree setting) so that the second physical quantity matches the second target value. A plurality of input values including the value and the state quantity affecting the heat exchange are input, the operation quantity is calculated based on the input input value, and the calculated operation quantity is combined with the inverter 102 or the evaporation pressure adjusting valve 109. A neural network 15) that outputs to the electronic expansion valve 106, and learning to reduce the first operation amount and the second operation amount by using the first operation amount and the second operation amount as a teacher signal (for example, PID). There line learning) performed as the operation amount can output a manipulated variable such that 0 from the control unit 11 to 13, based on the input and results of the learning, for operating the first adjustment device A machine learning unit (for example, a neural work 15) that outputs a third operation amount and a fourth operation amount for operating the second adjustment device, and
A first adder that adds the first manipulated variable and the third manipulated variable and supplies it to the first adjusting device.
A second adder that adds the second manipulated variable and the fourth manipulated variable and supplies it to the second adjusting device.
Equipped with a,
The first feedback control unit performs the first feedback control with the same parameters regardless of the value set as the first target value.
The second feedback control unit performs the second feedback control with the same parameters regardless of the value set as the second target value.
It is a controller.

(4) The method for constructing a neural network according to the third aspect of the present invention is
A first adjusting device (for example, an inverter 102 or an evaporation pressure adjusting valve 109) for adjusting a first flow rate, which is a flow rate at a first position of a heat medium (for example, a refrigerant) that circulates and is used for heat exchange, and the heat. A method of constructing a neural network (for example, a neural network 15) that controls a second adjusting device (for example, an electronic expansion valve 106) that adjusts a second flow rate that is a flow rate at a second position of a medium (a type of production method). ) And
A first physical quantity (for example, the feedback air temperature) that changes according to the first flow rate is input, and the first physical quantity is set to a first target value (for example, blown air preset by an operation from the user or the like). The first operation amount (for example, the operation amount from the PID control unit 11 or the PID control unit 12) input to the first adjusting device by the first feedback control (for example, PID control) approaching the temperature set value) and the above-mentioned A second physical quantity (for example, a feedback degree of superheat) that changes according to the second flow rate is input, and this second physical quantity is set as a second target value (for example, a superheat degree set value preset by an operation from the user or the like). ) Is a second feedback control (for example, PID control) that affects the first feedback control (for example, the amount of operation from the PID control unit 13 also changes the blown air temperature, and the PID control unit 11 or 12 The second operation amount (for example, the operation amount from the PID control unit 13) input to the second adjustment device by the second feedback control (which affects the PID control by the PID control unit 13) and the first operation amount as a teacher signal. And the learning is performed so as to output a third operation amount for operating the first adjustment device and a fourth operation amount for operating the second adjustment device so as to reduce the second operation amount. It is provided with a step to be performed by a network (for example, learning performed so that each operation amount from the PID control units 11 to 13 can be output to be 0).
The neural network inputs a plurality of state quantities other than the first physical quantity and the second physical quantity, the first target value, and the second target value, and the first physical quantity is the first target. to match the value, and such that said second physical quantity is equal to the second target value, the first adjustment device and controls the second adjusting device (e.g., a neural network 15, the outlet air A plurality of input values including a temperature set value, a superheat degree set value, and a state amount affecting the heat exchange are input, an operation amount is calculated based on the input input value, and the calculated operation amount is calculated by the inverter 102. Or output to the evaporation pressure adjusting valve 109 and the electronic expansion valve 106),
The first operation amount and the third operation amount are added and supplied to the first adjustment device.
The second operation amount and the fourth operation amount are added and supplied to the second adjusting device.
This is a method for constructing a neural network.

(5) According to the above configuration , even if a certain feedback control (first feedback control) is affected by another feedback control (second feedback control), the influence is reduced (ideally). (Eliminate), suitable control for heat exchange can be performed.

Further, since the machine learning unit or the neural network learns a plurality of values set as target values, even if feedback control is performed with the same parameters regardless of the target value , suitable control for heat exchange is performed. It can be performed.

Contact name computer (e.g., controller 120) a first feedback controller and the second feedback control section and the machine learning unit (or neural network) at the machine learning unit of the (or neural network) to operate as The program also allows suitable control over heat exchange.

According to the present invention, suitable control regarding heat exchange can be performed.

It is a block diagram of the heat exchange system which concerns on one Embodiment of this invention. It is an example of the block diagram which shows a part of the controller of FIG. It is a figure which shows the structural example of a neural network. It is a figure which shows the transition of the air blowing temperature and the degree of superheat when the neural network is used, and when it is not used (the blowing air temperature set value = 20 ° C.). It is a figure which shows the transition of the air blowing temperature and the degree of superheat when the neural network is used, and when it is not used (the blowing air temperature set value = 10 ° C.).

[Heat exchange system 100]
The heat exchange system 100 according to the embodiment of the present invention is configured as a direct expansion type cooling device that cools (cools) the room to be cooled.

As shown in FIG. 1, the heat exchange system 100 includes a circulation path L1, a bypass path L2, a compressor 101, an evaporator 102, a condenser 103, a cooling water channel 104, a cooling water flow rate adjusting valve 105, and the like. Electronic expansion valve 106, blower 107, evaporator 108, evaporation pressure adjusting valve 109, capacity adjusting valve 110, liquid cooling valve 111, operation unit 112, controller 120, pressure sensors SP1 to SP3, The temperature sensors ST1 to ST4 are provided.

Further, the heat exchange system 100 is roughly composed of an outdoor unit Y and an indoor unit Z. The outdoor unit Y is equipped with a compressor 101, an inverter 102, and the like, and the indoor unit Z is equipped with a blower 107, an evaporator 108, and the like. The indoor unit Z includes a suction port Z1 for sucking in ambient air and an outlet Z2 for blowing out cooled air as described later.

The circulation path L1 is a path for circulating the refrigerant. A compressor 101, a condenser 103, an electronic expansion valve 106, an evaporator 108, and an evaporation pressure adjusting valve 109 are arranged in the middle of the circulation path L1.

The compressor 101 has a suction port for sucking the refrigerant (steam) flowing through the circulation path L1. The compressor 101 compresses the refrigerant sucked from the suction port to a high temperature and a high pressure. The compressor 101 also has a function of circulating the refrigerant. The compressor 101 compresses the refrigerant by a motor.

A pressure sensor SP1 and a temperature sensor ST1 are provided in the vicinity of the suction port of the compressor. The pressure sensor SP1 detects the pressure of the refrigerant (also referred to as the compressor suction pressure) near the suction port of the compressor 101, and supplies the detected pressure (pressure value) to the controller 120. The temperature sensor ST1 detects the temperature of the refrigerant (also referred to as the compressor suction temperature) in the vicinity of the suction port of the compressor 101, and supplies the detected temperature (temperature value) to the controller 120.

The inverter 102 adjusts the output of the compressor 101 (the flow rate of the refrigerant discharged from the compressor 101). The inverter 102 adjusts the output of the compressor 101 by adjusting the rotation speed of the motor of the compressor 101.

The condenser 103 condenses (cools and liquefies) the refrigerant compressed by the compressor 101. A cooling water channel 104 through which cooling water flows passes inside the condenser 103, and the refrigerant is condensed by the cooling water passing through the cooling water channel 104.

The cooling water flow rate adjusting valve 105 is arranged in the middle of the cooling water channel 104, and adjusts the flow rate of the cooling water flowing through the cooling water channel 104 according to the opening degree thereof.

A pressure sensor SP2 is provided in the vicinity of the outlet of the condenser 103 (the outlet of the condensed refrigerant). The pressure sensor SP2 detects the pressure (also referred to as condensation pressure) of the refrigerant condensed by the condenser 103, and supplies the detected pressure (pressure value) to the controller 120.

The electronic expansion valve 106 adjusts the flow rate of the refrigerant condensed by the condenser 103 according to the opening degree thereof, and rapidly depressurizes the refrigerant. Due to this pressure drop, the temperature of the refrigerant also drops.

The blower 107 blows the air in the room to be cooled to the evaporator 108.

The evaporator 108 evaporates the refrigerant whose temperature has been lowered by the electronic expansion valve 106. Evaporation of the refrigerant by the evaporator 108 removes heat from the air blown by the blower 107. That is, the evaporator 108 cools the air by exchanging heat between the blown air and the evaporating refrigerant.

The blower 107 and the evaporator 108 are mounted on the indoor unit Z. By the operation of the blower 107, the air near the suction port Z1 of the room to be cooled is sucked into the indoor unit Z from the suction port Z1, the sucked air is cooled by the evaporator 108, and the cooled air is blown out Z2. It is blown out from and supplied to the room to be cooled. As a result, the room to be cooled is cooled.

A temperature sensor ST2 is provided in the vicinity of the suction port Z1. The temperature sensor ST2 detects the temperature near the suction port Z1 (also referred to as the suction air temperature), and supplies the detected temperature (temperature value) to the controller 120. The suction air temperature when the blower 107 is operating is the temperature of the air (air before cooling) sucked from the suction port Z1.

A temperature sensor ST3 is provided in the vicinity of the outlet Z2. The temperature sensor ST3 detects the temperature near the outlet Z2 (also referred to as the blown air temperature), and supplies the detected temperature (temperature value) to the controller 120. The temperature of the blown air when the blower 107 is operating can be said to be the temperature of the air (air blown out from the blowout port Z2) after being cooled by the evaporator 108.

A pressure sensor SP3 and a temperature sensor ST4 are provided in the vicinity of the refrigerant outlet of the evaporator 108. The refrigerant outlet is an outlet from which the refrigerant evaporated by the evaporator 108 exits the evaporator. The pressure sensor SP3 detects the pressure (also referred to as evaporation pressure) of the refrigerant evaporated by the evaporator 108, and supplies the detected pressure (pressure value) to the controller 120. The temperature sensor ST4 detects the temperature of the refrigerant evaporated by the evaporator 108 (also referred to as the refrigerant vapor temperature), and supplies the detected temperature (temperature value) to the controller 120.

The evaporation pressure adjusting valve 109 adjusts the flow rate (refrigerant pressure) of the refrigerant evaporated by the evaporator 108. The refrigerant (steam) that has passed through the evaporation pressure adjusting valve 109 returns to the compressor 101.

The bypass path L2 is a path for bypassing a part of the refrigerant circulating in the circulation path L1. The bypass path L2 has paths L21 to L23. One end of the path L21 is connected to a branch point P1 on the circulation path L1. The branch point P1 is located between the compressor 101 and the condenser 103. One end of the path L22 is connected to a branch point P2 on the circulation path L1. The branch point P2 is located between the condenser 103 and the electronic expansion valve 106. A capacity adjusting valve 110 and a liquid cooling valve 111 are arranged in the middle of each of the paths L21 and 22. The other end of the path L21 and the other end of the path L22 are connected to the confluence P3 on the bypass path L2. The path L23 connects the confluence point P3 and the confluence point P4 of the circulation path L1. The confluence point P4 is located between the compressor 101 and the evaporation pressure regulating valve 109.

The capacity adjusting valve 110 and the liquid cooling valve 111 are basically closed, but the evaporation pressure detected by the pressure sensor SP3 is preset according to the upper limit of the compressor suction pressure (according to the specifications of the compressor 101). It is opened when the value) is exceeded. When the capacity adjusting valve 110 is opened, a part of the refrigerant compressed by the compressor 101 (the part is also referred to as a first refrigerant) flows into the path L21. When the liquid cooling valve 111 is opened, a part of the refrigerant condensed by the condenser 103 (the part is also referred to as a second refrigerant) flows into the path L22.

The capacity adjusting valve 110 adjusts the flow rate of the first refrigerant flowing through the path L21 according to the opening degree thereof. The liquid cooling valve 111 adjusts the flow rate of the second refrigerant flowing through the path L22 according to the opening degree thereof. The first refrigerant and the second refrigerant merge at the path L23 and are returned to the circulation path L1 via the confluence point P4.

When the capacity adjusting valve 110 and the liquid cooling valve 111 are closed, the refrigerant circulates in the order of compressor 101 → condenser 103 → electronic expansion valve 106 → evaporator 108 → evaporation pressure regulating valve 109 → compressor 101. It circulates on L1. An accumulator or the like for storing the refrigerant may be provided in the middle of the circulation path L1.

When the capacity adjusting valve 110 and the liquid cooling valve 111 are open, a part of the refrigerant circulating in the circulation path L1 flows into the bypass path L2 (paths L21 to L23) and is bypassed. By such a bypass, the compressor suction pressure and the compressor suction temperature can be reduced.

The operation unit 112 includes a remote controller and the like that accept operations from the user. For example, the user operates the operation unit 112 and inputs desired values for the condensation pressure, the blown air temperature, and the degree of superheat in the heat exchange system 100. The degree of superheat is the degree of superheat of the refrigerant evaporated by the evaporator 108, and is calculated by the refrigerant steam temperature − saturation temperature (saturation temperature at the evaporation pressure of the refrigerant). Each input value is supplied to the controller 120.

The controller 120 is composed of various computers such as a PLC (programmable logic controller). For example, the controller 120 has a ROM (Read Only Memory) for storing programs and data, and a non-volatile storage capable of changing (including adding and deleting) stored information (programs and data) such as a hard disk and a flash memory. A controller based on a device (hereinafter, may be simply referred to as a non-volatile storage device) and a program stored in the ROM or the non-volatile storage device, and using data stored in the ROM or the non-volatile storage device. It includes a CPU (Central Processing Unit) that actually executes the processing executed by the 120, and a RAM (Random Access Memory) that is the main memory of the CPU.

The controller 120 stores each of the above values supplied from the operation unit 112 in a predetermined storage area of the non-volatile storage device. As a result, each of the above values is preset as a set value. In addition, each set value is also referred to as a condensing pressure set value, a blown air temperature set value, and a superheat degree set value, respectively. The set value is used as a target value in PID control and the like described later. At least one of the condensation pressure and the degree of superheat may be set as a fixed value by the manufacturer of the heat exchange system 100.

The controller 120 controls the inverter 102, the cooling water flow rate adjusting valve 105, the electronic expansion valve 106, the evaporation pressure adjusting valve 109, the capacity adjusting valve 110, and the liquid cooling valve 111 to control the condensing pressure, the compressor suction pressure, and the compressor. The suction temperature, the blown air temperature, and the degree of superheat are controlled, and the condensing pressure, the blown air temperature, and the degree of superheat are brought close to (including matching and maintaining) each of the above set values. The condensing pressure can be controlled by controlling the opening degree (flow rate of cooling water) of the cooling water flow rate adjusting valve 105. The compressor suction pressure can be controlled by controlling the opening degree (flow rate of the refrigerant) of the capacity adjusting valve 110. The compressor suction temperature can be controlled by controlling the opening degree (flow rate of the refrigerant) of the liquid cooling valve 111. By controlling the inverter 102 or the evaporation pressure adjusting valve 109 (opening degree), the flow rate of the refrigerant can be controlled, and therefore the temperature of the blown air can be controlled. The degree of superheat can be controlled by controlling the opening degree (flow rate of the refrigerant) of the electronic expansion valve 106.

The controller 120 controls the control target by supplying, for example, an operation amount capable of taking a numerical value in a predetermined range to the control target such as the inverter 102 and the cooling water flow rate adjusting valve 105. The operation amount has a common numerical range of 0 to 100 [%] for each control target, but the operation based on the operation amount of the same numerical value is basically different depending on the control target.

In the inverter 102, the frequency of the output AC power is controlled by the amount of operation. The frequency is controlled according to the operation amount 0 to 100. The inverter 102 outputs the AC power of the minimum frequency (the minimum frequency that the inverter 102 can output) when the operation amount is 0 [%], and the maximum frequency (the inverter 102) when the operation amount is 100 [%]. Outputs AC power (maximum frequency that can be output). Therefore, as the amount of operation increases, the frequency increases. The rotation speed of the motor of the compressor 101 is controlled by the frequency of the AC power. Therefore, the output of the compressor 101 (resulting in the flow rate of the refrigerant output from the compressor 101 and the temperature of the blown air) is controlled by the amount of operation.

The opening degree of each of the cooling water flow rate adjusting valve 105, the electronic expansion valve 106, the evaporation pressure adjusting valve 109, the capacity adjusting valve 110, and the liquid cooling valve 111 is controlled by the amount of operation. The operation amount indicates the ratio of the opening degree. Specifically, when the operation amount is 0 [%], the opening degree is 0 (the state in which the valve is closed). When the operation amount is 100 [%], the opening degree is fully open. When the operation amount is 20 [%], the opening degree is 20% of the fully opened opening degree.

For example, the controller 120 calculates the operation amount by PID (Proportional-Integral-Differential) based on the deviation between the condensing pressure set value and the condensing pressure (feedback value) detected by the pressure sensor SP2. The operation amount is output to the cooling water flow rate adjusting valve 105 (PID control). By such PID control, the opening degree of the cooling water flow rate adjusting valve 105 (in other words, so that the condensing pressure detected by the pressure sensor SP2 approaches the condensing pressure set value (target value) (so that the deviation becomes 0)). Then, the condensation pressure) is controlled.

For example, the controller 120 monitors whether the evaporation pressure detected by the pressure sensor SP3 exceeds the upper limit of the compressor suction pressure. When the evaporation pressure exceeds the upper limit value, the controller 120 sets the upper limit value of the compressor suction pressure as a set value, or may set a predetermined value (which may be set by operating the operation unit 112). (It may be set as a fixed value in advance) is set as the set value of the compressor suction temperature. The controller 120 calculates the operation amount A by PID based on the deviation between the upper limit value of the compressor suction pressure set as the set value and the compressor suction pressure (feedback value) detected by the pressure sensor SP1. The operation amount A is output to the capacity adjusting valve 110 (PID control). The controller 120 calculates the operation amount B by PID based on the deviation between the predetermined value set as the set value and the compressor suction temperature (feedback value) detected by the temperature sensor ST1, and the calculated operation amount B. Is output to the liquid cooling valve 111 (PID control). By these PID controls, the compressor suction pressure and the compressor suction temperature detected by the pressure sensor SP1 and the temperature sensor ST1 each approach the set values (target values) (each deviation becomes 0). , The opening degree of the capacity adjusting valve 110 (in other words, the compressor suction pressure) and the opening degree of the liquid cooling valve 111 (in other words, the compressor suction temperature) are controlled. The controller 120 supplies the inverter 102 with an operation amount that makes the output of the compressor 101 the lowest value (the rotation speed of the motor is the lowest).

Further, the controller 120 controls the blown air temperature by controlling the inverter 102 or the evaporation pressure adjusting valve 109. Further, the controller 120 controls the degree of superheat by controlling the electronic expansion valve 106. The details of this control will be described with reference to FIG.

[Controller 120 when controlling the temperature of blown air and the degree of superheat]
As shown in FIG. 2, the controller 120 that controls the inverter 102, the electronic expansion valve 106, and the evaporation pressure adjusting valve 109 includes PID control units 11 to 13 and a neural network 15. For example, the configuration shown in FIG. 2 is realized by operating the PLC or the like as the PID control units 11 to 13 and the neural network 15, respectively.

The PID control unit 11 operates when the evaporation pressure (a) detected by the pressure sensor SP3 is equal to or less than the upper limit value (b) of the compressor suction pressure (when a ≦ b). The deviation between the blown air temperature set value and the blown air temperature (feedback value) detected by the temperature sensor ST3 is input to the PID control unit 11, and the PID control unit 11 inputs the PID based on the input deviation. The operation amount is calculated and the calculated operation amount is output to the inverter 102. In this way, the PID control unit 11 controls the inverter 102 so that the blown air temperature detected by the temperature sensor ST3 approaches the blown air temperature set value (target value) (so that the deviation becomes 0). To do.

The PID control unit 12 operates when the evaporation pressure (a) detected by the pressure sensor SP3 is higher than the upper limit value (b) of the compressor suction pressure (when a> b). The deviation between the blown air temperature set value and the blown air temperature (feedback value) detected by the temperature sensor ST3 is input to the PID control unit 12, and the PID control unit 12 inputs the PID based on the input deviation. The operation amount is calculated and the calculated operation amount is output to the evaporation pressure adjusting valve 109. In this way, the PID control unit 11 has the evaporation pressure adjusting valve so that the blown air temperature detected by the temperature sensor ST3 approaches the blown air temperature set value (target value) (so that the deviation becomes 0). Control 109.

The deviation between the superheat degree set value and the superheat degree (feedback value) derived from the evaporation pressure and the refrigerant steam temperature detected by the pressure sensor SP3 and the temperature sensor ST4 is input to the PID control unit 13. The controller 120 derives the saturation temperature (saturation temperature at that evaporation pressure) based on the evaporation pressure (may be calculated, or can be obtained by referring to a table that defines the relationship between the evaporation pressure and the saturation temperature). The superheat degree is obtained by subtracting the saturation temperature from the refrigerant steam temperature (the superheat degree is fed back). The PID control unit 13 calculates the operation amount by the PID based on the input deviation, and outputs the calculated operation amount to the electronic expansion valve 106. In this way, the PID control unit 13 controls the electronic expansion valve 106 so that the superheat degree acquired above approaches the superheat degree set value (target value) (so that the deviation becomes 0).

As shown in FIG. 3, the neural network 15 has an input layer, an intermediate layer, and an output layer. The input value to the input layer of the neural network 15 x _i, the coupling weight w _ij between the input layer and the hidden _layer, and enter the s _j of the connection weights of the intermediate layers and the output layer w _jk, the element of the intermediate layer Then, the input s _j to the element of the intermediate layer can be expressed by the following mathematical formula (1), and the output q _j from the intermediate layer to the output layer can be expressed by the following mathematical formulas (2) and (3). The output y _k of 15 can be expressed by the following mathematical formula (4). Here, the subscripts i, j, and k are the element numbers of the input layer, the intermediate layer, and the output layer, respectively, and n and m are the number of inputs and the number of intermediate elements, respectively. In the following formula, x ₀ = 1, s ₀ = 1, and q ₀ = 1.

The input layer has an input value x _i , a blown air temperature set value (for example, x ₁ ), a superheat degree set value (for example, x ₂ ), and various states that affect heat exchange by the heat exchange system 100. Amounts (eg, various state quantities that affect the cooling characteristics of the heat exchange system 100) (feedback values) are entered. For example, as the state quantity of the is detected compressor suction pressure detected by the pressure sensor SP1 _{(e.g., x} 3), the suction air temperature detected by the temperature sensor ST2 _{(e.g., x} 4), the pressure sensor SP2 Condensation pressure (eg x ₅ ), last three operations (eg x _6, x _7, x ₈ ) input to the electronic expansion valve 106, inverter 102 (when a ≤ b) or evaporation pressure control valve The operation amount (for example, x _9, x _10, x ₁₁ ) of the past three times input in 109 (in the case of a> b) is input to the neural network 15.

The past operation amount (x _{6 to} x ₁₁ ) is the sum of the operation amount from the neural network 15 and the operation amount from any of the PID control units 11 to 13 (details will be described later). Either one of the operation amount from the neural network 15 and the operation amount from any of the PID control units 11 to 13 may be used as the past operation amount input to the neural network 15. The past operation amount input to the neural network 15 may be each operation amount one or more times in the past, or may be an average of the operation amounts of a plurality of times in the past. In addition, at least one such as the amount of change in the manipulated variable (current manipulated variable-previous manipulated variable), the first derivative (time derivative) of the time change of the past multiple manipulated variables, and the second derivative (time derivative). It may be one. As described above, what is input to the neural network 15 as the past manipulated variable is only one or more values indicating the history of the manipulated variable to the controlled object.

The plurality of input values input to the neural network 15 may be those that operate the neural network 15 appropriately (those that output a suitable operation amount). The state quantity includes a physical quantity, a manipulated quantity, and the like, and also includes a differential value due to a change amount, a time, and the like. The input value input to the neural network 15 can be appropriately controlled by not including the physical quantities (blown air temperature, superheat degree) fed back to the PID control units 11 to 13.

The neural network 15 derives the output y _k based on the input values (x _{1 to} x ₁₁ ). For example, the output y _k is calculated by each of the above equations. The neural network 15 outputs one of the outputs y _{k to} the inverter 102 (when a ≦ b) or the evaporation pressure regulating valve 109 (when a> b) as the manipulated variable y ₁ , and outputs the other one to the inverter 102 (when a ≦ b). The electronic expansion valve 106 is set as the operation amount y ₂ . The manipulated variable y ₁ is the manipulated variable that brings the blown air temperature closer to the blown air temperature set value (the manipulated variable that is controlled to match the two), and the manipulated variable y ₂ brings the superheat degree closer to the superheat degree set value. The operation amount (the operation amount controlled so as to match the two).

Such a neural network 15 uses the manipulated variables output from the PID control units 11 to 13 as teacher signals, and performs machine learning so as to make these manipulated variables zero. For example, the neural network 15 performs machine learning so that the operation amounts (teacher signals) from the PID control units 11 to 13 can be output to be zero. The machine learning is performed by, for example, the error back propagation method, and the weight correction by the error back propagation method is performed by the gradient method. In this way, the result of machine learning is reflected in the neural network 15. The neural network 15 derives the manipulated variable based on the result of machine learning. For example, the neural network 15 may stop learning after delivering the heat exchange system 100 to the user and then learning to some extent, or may always learn.

The inverter 102, and the operation amount from the PID controller 11, an operation amount _{y 1} from the neural network 15, the sum of the input. The evaporation pressure control valve 109, and the operation amount from the PID controller 12, an operation amount y ₁ from the neural network 15, the sum of the input. The sum of the manipulated variable from the PID control unit 13 and the manipulated variable y ₂ from the neural network 15 is input to the electronic expansion valve 106. In this way, the inverter 102 (in other words, the output (flow rate of the refrigerant) or the blown air temperature of the compressor 101) is controlled by the neural network 15 and the PID control unit 11. The evaporation pressure regulating valve 109 (in other words, the blown air temperature) is controlled by the neural network 15 and the PID control unit 12. The electronic expansion valve 106 (in other words, the degree of superheat) is controlled by the neural network 15 and the PID control unit 13. However, the neural network 15 learns to set each operation amount output from the PID control units 11 to 13 to 0. Therefore, in an ideal state in which learning by the neural network 15 has progressed, the blown air temperature of the inverter 102, the evaporation pressure adjusting valve 109, and the electronic expansion valve 106 matches the blown air temperature set value only with the neural network 15. , The degree of superheat is controlled to match the set value of the degree of superheat.

In this embodiment, even if the blown air temperature set value is changed by the user, the various parameters set in the PID control units 11 to 13 are not changed. That is, in this embodiment, the same PID control is performed regardless of the blown air temperature set value. Further, one neural network 15 performs learning with various blown air temperature set values.

The neural network 15 may control the blower 107 (the amount of air to be blown) according to the control of the inverter 102 or the evaporation pressure adjusting valve 109. For example, when the flow rate of the refrigerant is increased by controlling the inverter 102 or the like (when the air to be blown is cooled more), the blower 107 may be controlled to increase the air blown in order to increase the cooling capacity. ..

[Other examples of manipulated variable]
The manipulated variable output by the PID control units 11 to 13 and the manipulated variable output from the neural network 15 were previously input to the control target (inverter 102, electronic expansion valve 106, evaporation pressure adjusting valve 109). It may be represented by the amount of change from the manipulated variable. In this case, for example, the manipulated variable MV input to the inverter 102 is the manipulated variable MV _-1 input last time, the manipulated variable ΔMV _PID (change amount) output from the PID control unit 11, and the neural network 15. It is the total value of the manipulated variable ΔMV _NN (change amount) (manipulation amount y ₁ ). The total value (operation amount MV) becomes the operation amount MV _-1 at the next input of the operation amount. The controller 120 may hold the previous operation amount in the RAM or the like. The operation amount output in the PID control for controlling the control target other than the inverter 102, the electronic expansion valve 106, and the evaporation pressure adjusting valve 109 may also be represented by the amount of change from the previous operation amount.

[Effect of this embodiment]
(1) The control of the inverter 102 or the evaporation pressure regulating valve 109 and the control of the electronic expansion valve 106 influence each other. For example, if the output of the compressor 101 is changed by controlling the inverter 102 (changing the manipulated variable), or if the opening degree is changed by controlling the evaporation pressure adjusting valve 109 (changing the manipulated variable), the flow rate of the refrigerant changes. It is changed, and as a result, the degree of superheat, which is a physical quantity controlled by the control of the electronic expansion valve 106, also changes (resulting in the need to further control the electronic expansion valve 106). Similarly, when the opening degree is changed by controlling the electronic expansion valve 106 (changing the operation amount), the flow rate of the refrigerant is changed, and as a result, it is controlled by the control of the inverter 102 or the evaporation pressure adjusting valve 109. The temperature of the blown air, which is a physical quantity, also changes (as a result, it becomes necessary to further control the inverter 102 or the evaporation pressure regulating valve 109). Therefore, when the control for the inverter 102 or the evaporation pressure adjusting valve 109 and the control for the electronic expansion valve 106 are individually performed (for example, the control is performed only by the PID control units 11 to 13, the inverter 102, the evaporation pressure adjusting valve 109, and the electron. (For example, when a PID control unit and a neural network are prepared for each of the expansion valves 106), control may not be successful. However, in this embodiment, since one neural network 15 learns the operation amount output from the PID control units 11 to 13 as a teacher signal (so that the operation amount becomes 0), the learning progresses sufficiently. After that, the influence of the control by the neural network 15 becomes large (ideally, the control by the PID control units 11 to 13 disappears), and the control on the inverter 102 or the evaporation pressure adjusting valve 109 and the control on the electronic expansion valve 106 become large. It is possible to reduce (ideally eliminate) the influence of and on each other. Therefore, it is possible to make the blown air temperature and the degree of superheat more stable in the vicinity of each set value with the neural network than without the neural network (when the PID controls influence each other). That is, by providing the neural network 15, the influence of each PID control can be reduced, and the blown air temperature and the degree of superheat can be suitably controlled. In other words, suitable control for heat exchange can be performed. It should be noted that such an effect can be said even when one PID control affects the other PID control, but the other PID control does not affect one PID control.

(2) For example, when controlling the operation of the heat exchange system 100 only by PID control without using a neural network, a plurality of types of PID control units (parameters) having different parameters for each value that can be set as the blown air temperature set value. It is necessary to prepare (including the case of changing only). On the other hand, in this embodiment, the neural network 15 learns the operation amount from the PID control units 11 to 13 at various blown air temperature target values as a teacher signal, and controls the neural network 15 and PID. Units 11 to 13 cooperate to control. Therefore, it is not necessary to prepare a plurality of parameters for each of the PID control units 11 to 13 for each value that can be set as the blown air temperature setting value (each of a plurality of types of PID control units 11 to 13 for each value that can be set). It is not necessary to prepare them), and PID control units 11 to 13 may be prepared one by one. Even with such a configuration, the inverter 102, the evaporation pressure adjusting valve 109, and the electronic expansion valve 106 can be suitably controlled. Specifically, regardless of the value of the blown air temperature set value, the blown air temperature and the degree of superheat can be changed in the vicinity of the blown air temperature set value and the superheat degree set value, respectively.

(3) In the control by the controller 120, the difference in the transition of the degree of superheat and the blown air temperature between the control with and without the neural network 15 is shown in FIGS. 4 to 5. The neural network 15 here has been learned to some extent. In each figure, the broken line shows the transition when there is no neural network (without NN), and the solid line shows the transition when there is a neural network (with NN). The control conditions are superheat degree set value = 15 ° C., blown air temperature set value = 20 ° C. (Fig. 4), 10 ° C. (Fig. 5), PID proportional gain = 0.1, PID integration time = 60 seconds. Is. Further, the magnitude relationship between the evaporation pressure (a) detected by the pressure sensor SP3 and the upper limit value (b) of the compressor suction pressure is b> a. Therefore, the control targets here are the electronic expansion valve 106 and the evaporation pressure adjusting valve 109, the PID control unit 11 does not operate, the PID control units 12 and 13 operate, and the neural network 15 operates the electronic expansion valve 106 and The amount of operation is supplied to the evaporation pressure adjusting valve 109. For example, in FIGS. 4 to 5, one same PID control unit 12 (no parameter change, etc.) and one same PID control unit 13 (no parameter change, etc.) are used.

The degree of superheat ((A) in each of FIGS. 4 to 5) and the temperature of the blown air ((B) in each of FIGS. 4 to 5) without the neural network 15 are not stable. On the other hand, when the neural network 15 is present, the degree of superheat and the blown air temperature can be controlled more stably near each set value than when the neural network 15 is not provided, regardless of the blown air temperature set value (FIG. 4). ~ Fig. 5).

As shown in the results of FIGS. 4 to 5, control by the PID control unit 12 and control by the PID control unit 13 are performed by performing control in cooperation with the neural network 15 and the PID control units 12 and 13. Can be reduced from affecting each other, and the degree of superheat and the temperature of the blown air can be stably changed in the vicinity of each set value. Further, even if the parameters of the PID control units 12 and 13 are one type and the set value of the blown air temperature is changed, the degree of superheat and the temperature of the blown air are stably changed near each set value. Can be done. In addition, referring to the transition of the superheat degree and the blown air temperature without the neural network in FIGS. 4 and 5, the transition of the superheat degree and the blown air temperature is not stable in any case. Therefore, the PID control units 12 and 13 used here must be suitable for any of the cases where the blown air temperature set value is 20 ° C. (FIG. 4) and 10 ° C. (FIG. 5). Conceivable. However, by using the neural network 15, the degree of superheat and the temperature of the blown air can be stably changed in the vicinity of each set value. It should be noted that such an advantage is obtained when the magnitude relationship between the evaporation pressure (a) detected by the pressure sensor SP3 and the upper limit value (b) of the compressor suction pressure is b ≦ a (that is, the inverter 102). It is thought that the same can be obtained when controlling).

(4) In relation to the above (2), if PID control units 11 to 13 having different parameters are provided for each possible value of the blown air temperature set value, the labor for setting the parameter for each set value is extremely large. However, by setting each parameter of the PID control units 11 to 13 to one type as in the present embodiment, it is possible to reduce such labor. That is, the controller 120 (particularly, the portion that controls the inverter 102, the electronic expansion valve 106, and the evaporation pressure adjusting valve 109) can be constructed relatively easily.

(5) Normally, the degree of superheat is not changed, but the same can be said for the degree of superheat. In the configuration of the present embodiment, even if the degree of superheat is changed, the temperature of the blown air and the degree of superheat are respectively. Can be stably changed in the vicinity of each set value.

(6) Among the devices such as the valves included in the heat exchange system 100, in the above embodiment, the inverter 102, the evaporation pressure adjusting valve 109, and the electronic expansion valve 106 are connected by the neural network 15 and the PID control units 11 to 13. Control. Thereby, the blown air temperature and the like can be effectively controlled.

[Modifications]
The present invention is not limited to the above embodiment, and the above embodiment can be appropriately modified. An example thereof is shown below as a modified example, and at least a part of the modified example below can be combined. Further, the configuration disclosed in the present specification may be omitted in any configuration as long as there is no technical problem.

(Modification example 1)
The blown air temperature input and set from the operation unit 112 may be set to room temperature, suction air temperature, or the like. When setting the room temperature, a temperature sensor for detecting the room temperature may be provided, and the room temperature detected by the temperature sensor may be used as a feedback value for PID control or the like. When the suction air temperature (the temperature is also close to room temperature) is set, the suction air temperature detected by the temperature sensor ST2 may be used as a feedback value for PID control or the like.

(Modification 2)
The neural network 15 may use the manipulated variable output from the PID control units 11 to 13 as a teacher signal for learning to reduce the manipulated variable (it does not have to be learning to set the manipulated variable to 0). Even in such a case, after the learning has progressed sufficiently, the influence of the control by the neural network 15 can be increased, and suitable control for heat exchange can be performed.

(Modification 3)
The neural network 15 replaces or in addition to at least one of the inverter 102, the electronic expansion valve 106, and the evaporation pressure regulating valve 109, and the cooling water flow rate regulating valve 105 and the blower 107. , At least one of the capacity adjusting valve 110 and the liquid cooling valve 111 may be controlled. For example, the neural network 15 has an operation amount of supplying the capacitance adjusting valve 110 from the PID control unit that controls PID to the capacitance adjusting valve 110 and an operation of supplying the liquid cooling valve 111 from the PID control unit that controls PID to the liquid cooling valve 111. Machine learning is performed using the amount as a teacher signal, and various input values (various state quantities that affect the cooling characteristics of the heat exchange system 100, etc.) including the compressor suction pressure set value and the compressor suction temperature set value are used. Based on this, the operation amount is output to each of the capacity adjusting valve 110 and the liquid cooling valve 111. The opening degree of the capacity adjusting valve 110 and the liquid cooling valve 111 is controlled by each operation amount from each PID control unit and each operation amount from the neural network 15.

(Modification example 4)
In the above embodiment, for each of the blown air temperature and the degree of superheat, one PID control unit 11 to 13 is provided (with one set of parameters) for all the settable numerical values, but it can be set. All the numerical values may be divided into a plurality of ranges, and one PID control unit 11 to 13 may be provided for each of the plurality (2 to 3) ranges. The neural network 15 may also be provided one for each of the plurality of ranges. The plurality of ranges are, for example, a range in which the shape of the refrigeration cycle represented as a Moriel diagram changes (a range including a plurality of numerical values for which PID control parameters need to be changed). Further, when the evaporation pressure (a) detected by the pressure sensor SP3 is equal to or less than the upper limit value (b) of the compressor suction pressure (when a ≦ b) and vice versa (when a> b). , And different neural networks 15 may be prepared. Since the control target is different between the case of a ≦ b and the case of a> b (one is the inverter 102 and the other is the evaporation pressure adjusting valve 109), the case of a ≦ b and the case of a> b are different. There is a possibility that the blown air temperature and the degree of superheat can be controlled more accurately by switching between the two neural networks 15.

(Modification 5)
Instead of the neural network 15, another machine learning device may be provided.

(Modification 6)
The heat exchange system 100 may perform heating (in this case, the circulation direction of the refrigerant is reversed). The heat exchange system 100 may use any heat exchangeable heat medium. Further, the target to be cooled or heated by the heat exchange system 100 is not limited to air, but may be another gas, liquid, or the like. That is, the target may be a fluid. For example, the heat exchange system 100 may be used in a freezer. The heat exchange system is not limited to the direct expansion type cooling system, and may be another heat exchange system. In this case, the control target by the neural network 15 and the PID control units 11 to 13 is also changed as appropriate. However, the control target is an adjusting device (for example, an inverter that controls a compressor, various valves provided in the flow path of the heat medium) that adjusts the flow rate of the refrigerant (heat medium) used in the heat exchange system. Is desirable. Further, it is desirable that the control target is an adjusting device (for example, an inverter, an expansion valve) that affects the degree of heat exchange (the degree of change in the temperature of the fluid due to heat exchange).

(Modification 7)
Various values used in the controller 120 (for example, various values input to the PID control units 11 to 13, the neural network 15, etc.) may be normalized as appropriate. For example, the blown air temperature set value input to the neural network 15 may be normalized, while the blown air temperature set value input to the PID control unit 11 may not be normalized (however, both values are the same). Blow-out air temperature set value).

(Modification 8)
The PID control is a feedback control in which a predetermined physical quantity is fed back to bring the physical quantity closer to a target value, but the PID control may be changed to another feedback control. Other feedback controls include PI (Proportional-Integral) control.

(Modification 9)
The physical quantity (temperature or pressure) detected by each of the above sensors may be calculated by calculation. For example, the compressor suction temperature may be calculated based on the compressor suction pressure or the like.

L1 ... Circulation path, L2 ... Bypass path, 11-13 ... PID control unit, 15 ... Neural network, 100 ... Heat exchange system, 101 ... Compressor, 102 ... Inverter, 103 ... Condenser, 104 ... Cooling water channel, 105 ... Cooling water flow control valve, 106 ... Electronic expansion valve, 107 ... Blower, 108 ... Evaporator, 109 ... -Evaporation pressure control valve, 110 ... Capacity control valve, 111 ... Liquid cooling valve, 112 ... Operation unit, 120 ... Controller, SP1 to SP3 ... Pressure sensor, ST1 to ST4 ... Temperature sensor

Claims (5)

A heat exchange system that exchanges heat using a circulating heat medium.
A first adjusting device that adjusts the first flow rate, which is the flow rate of the heat medium at the first position, and
A second adjusting device that adjusts the second flow rate, which is the flow rate of the heat medium at the second position,
A first feedback control is performed in which a first physical quantity that changes according to the first flow rate is input, and a first operation amount is output to the first adjusting device so that the first physical quantity matches the first target value. 1 Feedback control unit and
It is a second feedback control that inputs a second physical quantity that changes according to the second flow rate and outputs a second manipulated quantity to the second adjusting device so that the second physical quantity matches the second target value. , A second feedback control unit that performs a second feedback control that affects the first feedback control,
A plurality of state quantities that affect heat exchange by this heat exchange system other than the first physical quantity and the second physical quantity, the first target value, and the second target value are input, and the first physical quantity is input . A machine learning unit that controls the first adjusting device and the second adjusting device so that the physical quantity matches the first target value and the second physical quantity matches the second target value. there, have rows learning to reduce the first operation amount and the second operation amount of the first operation amount and the second operation amount as a teacher signal, based on the input and results of the learning, the first A machine learning unit that outputs a third operation amount for operating the adjustment device and a fourth operation amount for operating the second adjustment device .
A first adder that adds the first manipulated variable and the third manipulated variable and supplies it to the first adjusting device.
A second adder that adds the second manipulated variable and the fourth manipulated variable and supplies it to the second adjusting device.
Equipped with a,
The first feedback control unit performs the first feedback control with the same parameters regardless of the value set as the first target value.
The second feedback control unit performs the second feedback control with the same parameters regardless of the value set as the second target value.
Heat exchange system. A compressor that compresses the heat medium and
An inverter that adjusts the output of the compressor and
A condenser that condenses the heat medium compressed by the compressor, and
An expansion valve that expands the heat medium condensed by the condenser, and
An evaporator that exchanges heat by evaporating the heat medium expanded by the expansion valve.
A regulating valve for adjusting the flow rate of the heat medium evaporated by the evaporator is provided.
One of the first adjusting device and the second adjusting device has the inverter or the adjusting valve, and the other has the expansion valve.
The machine learning unit has a neural network.
The heat exchange system according to claim 1 . The first adjusting device has the inverter and the adjusting valve, and the second adjusting device has the expansion valve.
The first physical quantity is the temperature adjusted by this heat exchange system, and the second physical quantity is the degree of superheat of the heat medium.
The neural network inputs the temperature set value, the superheat degree set value, and the state quantity,
The first feedback control unit
A third feedback control unit that operates when the evaporation pressure is equal to or less than the upper limit of the suction pressure of the compressor and outputs a fifth operation amount for controlling the inverter.
It is provided with a fourth feedback control unit that operates when the evaporation pressure is larger than the upper limit of the suction pressure of the compressor and outputs a sixth operation amount for controlling the adjusting valve.
The neural network
As the third operation amount, the seventh operation amount for controlling the inverter is adjusted when the evaporation pressure is equal to or less than the compressor suction pressure upper limit value, and when the evaporation pressure is larger than the compressor suction pressure upper limit value. The eighth operation amount for controlling the valve is output, respectively.
The first adder includes a third adder and a fourth adder.
The third adder adds the fifth manipulated variable and the seventh manipulated variable and supplies the inverter to the inverter.
The fourth adder adds the sixth manipulated variable and the eighth manipulated variable and supplies them to the regulating valve.
The second adder adds the second manipulated variable and the fourth manipulated variable and supplies the second adder to the expansion valve.
The state quantity includes the compressor suction pressure, temperature, condensing pressure, the past operation amount input to the expansion valve, the past operation amount input to the inverter or the adjustment valve, and the past operation amount input to the expansion valve. Includes the amount of manipulation of, the derivative of these values, the amount of change, or the average value,
The heat exchange system according to claim 2. A first adjusting device that adjusts the first flow rate, which is the flow rate of the heat medium that circulates and is used for heat exchange at the first position, and a second that adjusts the second flow rate, which is the flow rate of the heat medium at the second position. 2 A controller that controls the adjusting device and
A first feedback control is performed in which a first physical quantity that changes according to the first flow rate is input, and a first operation amount is output to the first adjusting device so that the first physical quantity matches the first target value. 1 Feedback control unit and
It is a second feedback control that inputs a second physical quantity that changes according to the second flow rate and outputs a second manipulated quantity to the second adjusting device so that the second physical quantity matches the second target value. , A second feedback control unit that performs a second feedback control that affects the first feedback control,
A plurality of state quantities other than the first physical quantity and the second physical quantity, the first target value, and the second target value are input so that the first physical quantity matches the first target value. In addition, it is a machine learning unit that controls the first adjusting device and the second adjusting device so that the second physical quantity matches the second target value, and is the first operating amount and the first. 2 manipulated variable have line learning to reduce the first operation amount and the second operation amount as a teacher signal, based on the input and results of the learning, the third operation amount for operating the first adjustment device And a machine learning unit that outputs a fourth operation quantity for operating the second adjusting device, and
A first adder that adds the first manipulated variable and the third manipulated variable and supplies it to the first adjusting device.
A second adder that adds the second manipulated variable and the fourth manipulated variable and supplies it to the second adjusting device.
Equipped with a,
The first feedback control unit performs the first feedback control with the same parameters regardless of the value set as the first target value.
The second feedback control unit performs the second feedback control with the same parameters regardless of the value set as the second target value.
controller. A first adjusting device that adjusts the first flow rate, which is the flow rate of the heat medium that circulates and is used for heat exchange at the first position, and a second that adjusts the second flow rate, which is the flow rate of the heat medium at the second position. It is a method of constructing a neural network that controls two adjusting devices.
A first manipulated quantity input to the first adjusting device by a first feedback control in which a first physical quantity that changes according to the first flow rate is input and the first physical quantity is brought closer to the first target value, and the second The second adjustment device is a second feedback control that inputs a second physical quantity that changes according to the flow rate and brings the second physical quantity closer to the second target value, and is a second feedback control that affects the first feedback control. The third operation amount and the second adjustment for operating the first adjustment device so as to reduce the first operation amount and the second operation amount by using the second operation amount input to A step of causing the neural network to perform training so as to output a fourth operation quantity for operating the device is provided.
A plurality of state quantities other than the first physical quantity and the second physical quantity, the first target value, and the second target value are input so that the first physical quantity matches the first target value. In addition, the first adjusting device and the second adjusting device are controlled so that the second physical quantity matches the second target value .
The first operation amount and the third operation amount are added and supplied to the first adjustment device.
The second operation amount and the fourth operation amount are added and supplied to the second adjusting device.
How to build a neural network.