REFRIGERATION CYCLE CONTROL DEVICE, AND REFRIGERATION CYCLE DEVICE

A refrigeration cycle state predicting device according to the present disclosed technique is a refrigeration cycle state predictor constituting a refrigeration cycle device, and includes a predictor to calculate a prediction value of a future state of the refrigeration cycle device on the basis of an input provisional operation amount array.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description
CROSS REFERENCE TO RELATED APPLICATION

This application is a Continuation of PCT International Application No. PCT/JP2022/020601, filed on May 18, 2022, which is hereby expressly incorporated by reference into the present application.

TECHNICAL FIELD

The present disclosed technique relates to a refrigeration cycle control device, and a refrigeration cycle device.

BACKGROUND ART

The refrigeration cycle device is a device using a refrigeration cycle, and a room air conditioner functioning as a cooler and a heater, and an industrial cooling device correspond to the refrigeration cycle device. In the refrigeration cycle device, a phenomenon in which frost adheres to a running heat exchanger (referred to as “frosting”) may occur.

Frosting reduces efficiency of heat exchange. For this reason, the refrigeration cycle device generally performs a defrosting operation when a frosting amount is large.

For example, Patent Literature 1 discloses a refrigeration cycle device in which a heat exchanger includes a plurality of fins disposed at predetermined intervals from each other and a plurality of heat transfer tubes penetrating the plurality of fins, heat is exchanged between air and a refrigerant flowing through the heat transfer tubes using the plurality of fins, a thickness of frost adhering to the fins is estimated from a frost density output by a density estimation unit of the refrigeration cycle device, and control is performed so as not to change an opening degree of an expansion portion on the basis of the estimated thickness of frost. With this configuration, the refrigeration cycle device according to Patent Literature 1 can operate equipment while maintaining coefficient of performance (COP) even under a frosting condition.

CITATION LIST Patent Literature

  • Patent Literature 1: JP 2012-42207 A

SUMMARY OF INVENTION Technical Problem

An object of the present disclosed technique is to improve the refrigeration cycle device exemplified in Patent Literature 1. An object of the present disclosed technique is to provide a refrigeration cycle device that reduces an unnecessary defrosting operation as much as possible as compared with a conventional device.

Solution to Problem

A refrigeration cycle controller according to the present disclosed technique is a refrigeration cycle controller constituting a refrigeration cycle device, wherein, the refrigeration cycle device includes a plurality of indoor equipments having different refrigerant systems in the same room, or includes an indoor equipment having a plurality of heat exchangers having different refrigerant systems, the refrigeration cycle controller comprising: operation amount calculator to calculate a provisional operation amount array by calculation in such a manner as to satisfy a constraint condition; and a refrigeration cycle state predictor including a predictor to calculate a prediction value in a future state of the refrigeration cycle device by referring to the input provisional operation amount array, wherein, the operation amount calculator includes, an evaluator to calculate a value of an evaluation function used to calculate the provisional operation amount array, and an optimization calculator to obtain a solution of the provisional operation amount array minimizing the value of the evaluation function, and wherein, the evaluation function includes the following penalty term (Vp),

V p = t = 0 t = k switch ( t ) N d ( t ) T W T WN d ( t ) dt Equation l wherein k switch ( t ) = { 1 , N d ( t ) T N d ( t ) 2 0 , otherwise N d ( t ) = [ n 1 ( t ) n 2 ( t ) n d ( t ) ] for i = 1 to d , n i ( t ) = { 0 , i th refrigerant system is in normal operation mode 1 , i th refrigerant system is in defrosting operation mode

where W represents a weighting coefficient matrix representing a weight for each refrigerant system.

Advantageous Effects of Invention

The refrigeration cycle device according to the present disclosed technique includes the refrigeration cycle state predicting device having the above configuration, and therefore does not need to perform an unnecessary defrosting operation by referring to a prediction value of a future state of the refrigeration cycle device.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a block diagram illustrating a functional configuration of a refrigeration cycle device including a refrigeration cycle control device according to a first embodiment.

FIG. 2 is a flowchart illustrating a processing content of the refrigeration cycle control device according to the first embodiment.

FIG. 3 is a block diagram illustrating a functional configuration of a refrigeration cycle device including a refrigeration cycle control device according to a second embodiment.

FIG. 4 is a block diagram illustrating a functional configuration of a refrigeration cycle device including a refrigeration cycle control device according to a third embodiment.

FIG. 5 is a flowchart illustrating a processing content of the refrigeration cycle control device according to the third embodiment.

FIG. 6 is a block diagram illustrating a functional configuration of a refrigeration cycle device including a refrigeration cycle control device according to a fourth embodiment.

FIG. 7 is a block diagram illustrating a functional configuration of a refrigeration cycle device including a refrigeration cycle control device according to a fifth embodiment.

FIG. 8 is a block diagram illustrating a functional configuration of a refrigeration cycle device including a refrigeration cycle control device according to a seventh embodiment.

FIG. 9 is a flowchart illustrating a processing content of the refrigeration cycle control device according to the seventh embodiment.

FIG. 10 is an explanatory diagram illustrating a method for determining an operation amount by a refrigeration cycle control device 3 of the refrigeration cycle control device according to the seventh embodiment.

FIG. 11 is a flowchart illustrating a processing content for determining an operation amount by the refrigeration cycle control device 3 of the refrigeration cycle control device according to the seventh embodiment.

DESCRIPTION OF EMBODIMENTS

A refrigeration cycle device according to the present disclosed technique means a device using a refrigeration cycle. The refrigeration cycle device is specifically a room air conditioner or an industrial cooling device. In general, a vapor compression refrigeration cycle is most often used in a room air conditioner and an industrial cooling device. The refrigeration cycle device includes four-component parts of a compressor, a condenser, an expansion valve (capillary tube), and an evaporator. The inside of the refrigeration cycle device is in a sealed state, and a refrigerant circulates in a fixed direction with a state change, and cooling is thereby achieved.

First Embodiment

FIG. 1 is a block diagram illustrating a functional configuration of a refrigeration cycle device including a refrigeration cycle control device 3 according to a first embodiment.

As illustrated in FIG. 1, the refrigeration cycle device includes an operation unit 1, a measurement unit 2, a refrigeration cycle control device 3, and a control unit 4.

As illustrated in FIG. 1, the refrigeration cycle control device 3 includes a refrigeration cycle state predicting device 5 and an operation amount calculating unit 6. In the refrigeration cycle control device 3 according to the first embodiment, the refrigeration cycle state predicting device 5 includes a prediction unit 9 including a model storage unit 10. In the refrigeration cycle control device 3 according to the first embodiment, the operation amount calculating unit 6 includes an evaluation unit 7 and an optimization calculating unit 8.

<<Operation Unit 1 of Refrigeration Cycle Device>>

The operation unit 1 constituting the refrigeration cycle device is a component that sends operation information to the refrigeration cycle control device 3. Here, the operation information means information for calculating an “operation amount” described later.

Specifically, the operation information is information such as an operation start signal, an operation stop signal, an operation mode (a mode such as a normal mode, an energy saving mode, a sleep mode, or a defrosting mode), a set temperature, or a set air volume.

<<Measurement Unit 2 of Refrigeration Cycle Device>>

The measurement unit 2 constituting the refrigeration cycle device is a component that sends a measurement amount to the refrigeration cycle control device 3. Here, the measurement amount means an internal or external state of the refrigeration cycle device, obtained by measurement. Examples of the state of the refrigeration cycle device include a frosting amount, a frost layer thickness, a frost layer density, a frost layer surface temperature, and a heat exchanger surface temperature. The measurement amount is an amount such as an indoor temperature, an outdoor temperature, an evaporation temperature, an evaporation pressure, an evaporator fan air volume, an evaporator fan rotational speed, an evaporator fan current value, a dry bulb temperature of evaporator intake air, a humidity of evaporator intake air, or a dry bulb temperature of condenser intake air in addition to a value obtained by measuring the internal state exemplified above in an observable case.

Specifically, a means for measuring the measurement amount may be a sensor such as a temperature sensor, a pressure sensor, an air volume sensor, a humidity sensor, or a temperature and humidity sensor.

When the measurement amount is a frost layer thickness, a measurement method may be a method for analyzing an image acquired by a camera.

When the measurement amount is an outdoor temperature or an outdoor humidity, the measurement method may be a method for acquiring information announced by the Meteorological Agency by a means such as Internet communication.

<<Refrigeration Cycle Control Device 3 of Refrigeration Cycle Device>>

The refrigeration cycle control device 3 constituting the refrigeration cycle device is a component to calculate an operation amount on the basis of the operation information sent from the operation unit 1 and the measurement amount sent from the measurement unit 2.

Specifically, the operation amount is an amount such as a compressor frequency, an expansion valve opening degree, an evaporator fan air volume, or a condenser fan air volume.

A detailed processing content of the refrigeration cycle control device 3 will be apparent from the following description.

<<Control Unit 4 of Refrigeration Cycle Device>>

The control unit 4 constituting the refrigeration cycle device is a component that controls an actuator (not illustrated) of the refrigeration cycle device on the basis of the operation amount sent from the refrigeration cycle control device 3.

Here, the actuator is specifically an element of the refrigeration cycle device, such as a compressor, an expansion valve, an evaporator fan, or a condenser fan.

<<Refrigeration Cycle State Predicting Device 5 of Refrigeration Cycle Control Device 3>>

The refrigeration cycle state predicting device 5 constituting the refrigeration cycle control device 3 is a component to calculate a prediction value of a state at a future time on the basis of a provisional operation amount array sent from the operation amount calculating unit 6 described later. Here, the provisional operation amount array is a data array in which provisional values of operation amounts at times (time k+1, . . . , time k+m) from a time advanced by one sampling time (time k+1) to a future time advanced by m sampling time (time k+m) are arranged in time series at the current time (time k). The provisional value is a temporary value that is not formally determined but is determined for a while at the current time (time k), and an actual operation amount does not necessarily coincide with this provisional value. The number of elements (m) in the provisional operation amount array is a predetermined parameter.

A state in which the refrigeration cycle state predicting device 5 calculates a prediction value is synonymous with a state in a state equation, a state vector, or the like in the modern control theory. Specifically, the state assumed by the present disclosed technique is a state such as a frosting amount, a frost layer thickness, a frost layer density, a frost layer surface temperature, or a heat exchanger surface temperature. The frosting amount, the frost layer thickness, the frost layer density, the frost layer surface temperature, and the heat exchanger surface temperature are all states related to frosting, and thus are referred to as “frosting-related state” in the present specification.

The prediction value of a state calculated by the refrigeration cycle state predicting device 5 is sent to the operation amount calculating unit 6 again. That is, in the refrigeration cycle control device 3, as illustrated in FIG. 1, a loop is formed by the refrigeration cycle state predicting device 5 and the operation amount calculating unit 6.

As illustrated in FIG. 1, the refrigeration cycle state predicting device 5 includes the prediction unit 9 including the model storage unit 10.

The prediction unit 9 is a component that actually calculates a prediction value of a state. More specifically, the prediction unit 9 calculates a prediction value of a future state on the basis of an input provisional operation amount array. The prediction unit 9 may store artificial intelligence that has performed learning in the model storage unit 10, for example. Specifically, the artificial intelligence that has performed learning may be implemented by a mathematical model that has performed learning (also referred to as “learning model”). For learning of the artificial intelligence, for example, an algorithm such as a gradient boosting tree, a linear regression, a neural network, a random forest, or a k-approximate method may be used. A training data set used for learning may be actual data obtained by actually operating the refrigeration cycle device, simulation data obtained by simulation, or data obtained by combining the actual data and the simulation data.

In general, when a real-time simulation is performed for various operation scenarios of the refrigeration cycle device, a large-scale calculation resource suitable for the simulation, such as a supercomputer is required. By including the learning model, the refrigeration cycle state predicting device 5 according to the present disclosed technique can implement the function of the prediction unit 9 with a small-scale calculation resource such as a general PC.

<<Operation Amount Calculating Unit 6 of Refrigeration Cycle Control Device 3>>

The operation amount calculating unit 6 constituting the refrigeration cycle control device 3 is a component that obtains a provisional operation amount array by calculation. In the provisional operation amount array obtained by the operation amount calculating unit 6, a value at each time is determined in such a manner as to satisfy a constraint condition.

A typical example of the constraint condition assumed by the present disclosed technique is “A frosting amount does not exceed a predetermined threshold”. In this case, the state in which the refrigeration cycle state predicting device 5 calculates a prediction value is a frosting amount. A set of the constraint condition and the state may be appropriately determined depending on specifications required for the refrigeration cycle device. Note that the state in which the refrigeration cycle state predicting device 5 calculates a prediction value is not limited to the frosting amount. The state in which the refrigeration cycle state predicting device 5 calculates a prediction value may be another frosting-related state.

As described above, the operation amount calculating unit 6 includes the evaluation unit 7 and the optimization calculating unit 8.

The evaluation unit 7 is a component for calculating a value (hereinafter, referred to as “evaluation value”) of an evaluation function used in the calculation of the provisional operation amount array performed by the operation amount calculating unit 6. The evaluation function used in the operation amount calculating unit 6 is synonymous with an evaluation function used in the modern control theory, and is a scalar function such as an energy function. The provisional operation amount array is obtained in such a manner as to minimize the evaluation function. The evaluation function used in the operation amount calculating unit 6 may include, for example, a term of capability of the refrigeration cycle device (cooling capability or heating capability). The evaluation function used in the operation amount calculating unit 6 preferably includes a term related to consumed energy or coefficient of performance (COP). In the evaluation function, as for the term related to consumed energy, as consumed energy increases, a value of the evaluation function also increases. The evaluation function used in the operation amount calculating unit 6 preferably includes a term related to a difference (error) between a set temperature and an actual indoor temperature. Each term included in the evaluation function is multiplied by a weighting coefficient, integrated with time in a case of analog, and a sum of values at times is calculated in a case of digital.

The evaluation function preferably further includes the above-described constraint condition as a penalty term or a barrier term.

The optimization calculating unit 8 is a component that solves a conditional minimization problem. Specifically, the optimization calculating unit 8 is a component that obtains a solution of a provisional operation amount array minimizing the above-described evaluation function under the above-described constraint condition.

In solving the conditional minimization problem, the optimization calculating unit 8 may use an algorithm of, for example, random search, grid search, a Nelder-Mead method, covariance matrix adaptation evolution strategy (CMA-ES), or a black box function optimization method such as Bayesian optimization. In addition, in solving the conditional minimization problem, the optimization calculating unit 8 may use an algorithm of, for example, a gradient method such as sequential quadratic programming (SQP).

The operation amount calculating unit 6 does not necessarily need to strictly solve the conditional minimization problem. The operation amount calculating unit 6 may compare evaluation values of candidates of the provisional operation amount array prepared in advance and generated on the basis of a predetermined rule, and may select a candidate having the smallest evaluation value.

<<Processing Content of Refrigeration Cycle Control Device 3>>

FIG. 2 is a flowchart illustrating a processing content of the refrigeration cycle control device 3 according to the first embodiment. FIG. 2 illustrates steps S1 to S6 as processing steps performed by the refrigeration cycle control device 3.

In FIG. 2, step S1 described as “Acquire operation information and measurement value” is a processing step in which the refrigeration cycle control device 3 acquires operation information sent from the operation unit 1 and a measurement amount sent from the measurement unit 2. After step S1 is performed, the processing process proceeds to step S2.

In FIG. 2, step S2 described as “Determine provisional operation amount” is a processing step in which the operation amount calculating unit 6 constituting the refrigeration cycle control device 3 obtains a provisional operation amount array by calculation. After step S2 is performed, the processing process proceeds to step S3.

In FIG. 2, step S3 described as “Estimate predicted state amount” is a processing step in which the refrigeration cycle state predicting device 5 constituting the refrigeration cycle control device 3 calculates a prediction value of a state at a future time on the basis of the provisional operation amount array sent from the operation amount calculating unit 6. After step S3 is performed, the processing process proceeds to step S4.

In FIG. 2, step S4 described as “Calculate evaluation value” is a processing step in which the evaluation unit 7 of the operation amount calculating unit 6 constituting the refrigeration cycle control device 3 calculates an evaluation value used in calculation of the provisional operation amount array. After step S4 is performed, the processing process proceeds to step S5.

In FIG. 2, step S5 described as “Iteration is ended?” is a processing step in which the optimization calculating unit 8 of the operation amount calculating unit 6 constituting the refrigeration cycle control device 3 determines whether or not an end condition is satisfied in an iterative calculation for obtaining a solution of the provisional operation amount array minimizing an evaluation function under a constraint condition.

In general, when a value of a parameter at which the evaluation function is an extreme value (a local maximum value or a local minimum value) is numerically obtained by an iterative calculation using a gradient of a steepest descent method or the like, the end condition of the iterative calculation is that an absolute value of the gradient is sufficiently small and is equal to or less than a threshold (ε) prepared in advance. In addition, in order to cope with, for example, a phenomenon that a solution does not converge in the iterative calculation, when the number of iterations exceeds a threshold (kmax) prepared in advance, the calculation processing is generally forcibly terminated. The optimization calculating unit 8 in step S5 may also determine whether or not the end condition is satisfied using such a threshold (ε, kmax).

In step S5, if it is determined that the end condition is not satisfied (NO), the processing process proceeds to step S2. In step S5, if it is determined that the end condition is satisfied (YES), the processing process proceeds to step S6.

In FIG. 2, step S6 described as “Determine operation amount” is a processing step in which the refrigeration cycle control device 3 determines the provisional operation amount array obtained by the iterative calculation as a series of operation amounts to be used for actual control and outputs the determined provisional operation amount array to the control unit 4. After step S6 is performed, the processing step is ended.

As described above, the refrigeration cycle device according to the first embodiment has the above configuration, and therefore can continue a normal operation while keeping the frosting-related state at a value equal to or lower than the threshold as long as the operation amount does not exceed a limit thereof, that is, when the actuator has sufficient capability. The present disclosed technique is desirably applied to a refrigeration cycle device such as a refrigeration cycle device including a plurality of outdoor units or a refrigeration cycle device using a dual on-defrost circuit using a heat exchanger divided into upper and lower portions (refer to sixth and seventh embodiments).

Second Embodiment

A refrigeration cycle device according to a second embodiment is a modification of the refrigeration cycle device according to the present disclosed technique. In the second embodiment, the same reference signs as those used in the first embodiment are used unless otherwise specified. In the second embodiment, description overlapping with the first embodiment is appropriately omitted.

FIG. 3 is a block diagram illustrating a functional configuration of the refrigeration cycle device including a refrigeration cycle control device 3 according to the second embodiment. As illustrated in FIG. 3, the refrigeration cycle control device 3 according to the second embodiment includes an integration calculating unit 11 in a prediction unit 9 of a refrigeration cycle state predicting device 5 in addition to the configuration illustrated in the first embodiment.

<<Integration Calculating Unit 11>>

In general, dynamic behavior of a system (particularly a linear time-invariant system) is represented by a differential equation for a state (generally referred to as a “state equation”). An integration calculation is used when a solution of the state equation is numerically obtained. If an initial state (x(0)) is known and a time differential value (dx(t)/dt) of a state at each future time is known, a value (x(t)) of a state at any future time can be simply obtained by an integration calculation.

It is considered that dynamic behavior of the refrigeration cycle device according to the present disclosed technique is also represented by the state equation.

The integration calculating unit 11 that is a component added in the second embodiment is a component that performs an integration calculation on a time differential value of a state (the time derivative of a state is hereinafter referred to as “state amount time differential value”) at each future time predicted by the prediction unit 9 and calculates a value of a state at any future time.

Note that, strictly speaking, the integration calculation performed by the integration calculating unit 11 is processing on the premise of not continuous time but discrete time.

In addition, an actual refrigeration cycle device is not a linear time-invariant system. Therefore, for the nonlinear system, linearization around an equilibrium state (also referred to as “linearization around an equilibrium point”) is generally performed, and an approximate linear time-invariant system is used.

Since the learning model included in the prediction unit 9 calculates a prediction value of a future state on the basis of the input provisional operation amount array, the learning model is the state equation itself when the refrigeration cycle device is a linear time-invariant system. It can be said that a mathematical model included in the prediction unit 9 is a mathematical model capable of reproducing dynamic behavior of the refrigeration cycle device that is a nonlinear system by learning.

The mathematical model included in the prediction unit 9 according to the second embodiment may perform learning in such a manner that an initial state (x(0)) is known and a state amount time differential value at each future time is output from time series data of the operation amount input to the refrigeration cycle device.

As described above, the refrigeration cycle device according to the second embodiment has the above configuration, and therefore exhibits similar effects to those described in the first embodiment.

Third Embodiment

A refrigeration cycle device according to a third embodiment is a modification of the refrigeration cycle device according to the present disclosed technique. In the third embodiment, the same reference signs as those used in the above-described embodiments are used unless otherwise specified. In the third embodiment, description overlapping with the above-described embodiments is appropriately omitted.

FIG. 4 is a block diagram illustrating a functional configuration of the refrigeration cycle device including a refrigeration cycle control device 3 according to the third embodiment. As illustrated in FIG. 4, the refrigeration cycle control device 3 according to the third embodiment includes an estimation unit 12 in a refrigeration cycle state predicting device 5 in addition to the configuration illustrated in the first embodiment.

<<Estimation Unit 12>>

In general, when a controller that determines an operation amount on the basis of a state is designed, all the states to be used by the controller can be desirably observed. However, there is a case where the state cannot be observed, for example, because a sensor cannot be disposed due to environmental or hardware limitations. In order to cope with such a situation, a state estimator (also referred to as “observer”) that estimates a state is generally known.

Also in the technical field of the refrigeration cycle device, a situation may occur in which some states (for example, frost layer thickness) cannot be observed due to environmental or hardware limitations.

The estimation unit 12 that is a component added in the second embodiment is a component that estimates an unobservable state.

Examples of the unobservable state include a frost layer thickness, enthalpy, and dryness.

When the unobservable state is a frost layer thickness, the estimation unit 12 may estimate the frost layer thickness without using a sensor that directly observes the frost layer thickness, for example, using a method disclosed in JP 2011-127853 A.

FIG. 5 is a flowchart illustrating a processing content of the refrigeration cycle control device 3 according to the third embodiment. As illustrated in FIG. 5, processing steps of the refrigeration cycle control device 3 according to the third embodiment include a processing step (S10) described as “Calculate estimation value” in addition to the processing steps (S1 to S6) of the refrigeration cycle control device 3 according to the first embodiment.

In FIG. 5, step S10 described as “Calculate estimation value” is a processing step performed by the estimation unit 12. The estimation value in “Calculate estimation value” described in FIG. 5 means an estimation value in an unobservable state.

As illustrated in FIG. 5, step S10 is performed after step S1. After step S10 is performed, step S2 is performed.

As described above, the refrigeration cycle device according to the third embodiment has the above configuration, and therefore exhibits similar effects to those described in the first embodiment also when a state of interest cannot be directly observed.

Fourth Embodiment

A refrigeration cycle device according to a fourth embodiment is a modification of the refrigeration cycle device according to the third embodiment. In the fourth embodiment, the same reference signs as those used in the above-described embodiments are used unless otherwise specified. In the fourth embodiment, description overlapping with the above-described embodiments is appropriately omitted.

FIG. 6 is a block diagram illustrating a functional configuration of the refrigeration cycle device including a refrigeration cycle control device 3 according to the fourth embodiment. As illustrated in FIG. 6, the refrigeration cycle control device 3 according to the fourth embodiment includes a second model storage unit 13 in an estimation unit 12.

As a specific aspect of the estimation unit 12, it is conceivable that the estimation unit 12 has a second learning model that has performed learning in such a manner as to output an estimation value of an unobservable state separately from a learning model in a prediction unit 9. The estimation unit 12 according to the fourth embodiment includes the second model storage unit 13 that stores the second learning model.

For the learning of the second learning model, for example, an algorithm such as a gradient boosting tree, a linear regression, a neural network, a random forest, or a k-approximation method may be used.

A training data set used for learning of the second learning model may be actual data obtained by actually operating the refrigeration cycle device, simulation data obtained by simulation, or data obtained by combining the actual data and the simulation data.

As described above, in general, when a real-time simulation is performed for various operation scenarios of the refrigeration cycle device, a large-scale calculation resource suitable for the simulation, such as a supercomputer is required. By including the second learning model, the refrigeration cycle control device 3 according to the present disclosed technique can implement the function of the estimation unit 12 with a small-scale calculation resource such as a general PC.

As described above, the refrigeration cycle device according to the fourth embodiment has the above configuration, and therefore exhibits similar effects to those described in the third embodiment.

Fifth Embodiment

A refrigeration cycle device according to a fifth embodiment is a modification of the refrigeration cycle device according to the fourth embodiment. In the fifth embodiment, the same reference signs as those used in the above-described embodiments are used unless otherwise specified. In the fifth embodiment, description overlapping with the above-described embodiments is appropriately omitted.

FIG. 7 is a block diagram illustrating a functional configuration of the refrigeration cycle device including a refrigeration cycle control device 3 according to the fifth embodiment. As illustrated in FIG. 7, the refrigeration cycle control device 3 according to the fifth embodiment is an aspect in a case where it is not necessary to store a learning model in a prediction unit 9, and does not have to include a model storage unit 10.

As described above, in general, dynamic behavior of a linear time-invariant system is represented by the state equation. When the refrigeration cycle device can be approximated to a linear time-invariant system by linearization around an equilibrium point or the like, the prediction unit 9 only needs to include the state equation as a mathematical model.

As described above, the refrigeration cycle device according to the fifth embodiment has the above configuration, and therefore exhibits similar effects to those described in the third embodiment similarly to the fourth embodiment.

Sixth Embodiment

A refrigeration cycle device according to a sixth embodiment is a modification of the refrigeration cycle device according to the first embodiment. In the sixth embodiment, the same reference signs as those used in the above-described embodiments are used unless otherwise specified. In the sixth embodiment, description overlapping with the above-described embodiments is appropriately omitted.

The refrigeration cycle device according to the sixth embodiment has the same functional configuration as the functional configuration of the refrigeration cycle device according to the first embodiment illustrated in FIG. 1. The refrigeration cycle device according to the sixth embodiment includes a plurality of indoor units having different refrigerant systems in the same room, or includes an indoor unit having a plurality of heat exchangers having different refrigerant systems. As described above, as an example of such a configuration, a dual on-defrost circuit using a heat exchanger divided into upper and lower portions is known.

The refrigeration cycle device according to the sixth embodiment has a planning problem of how to plan an execution period of a defrosting operation by observing a frosting-related state in each refrigerant system.

The refrigeration cycle device according to the sixth embodiment uses an operation amount including an operation mode (a mode such as a normal mode, an energy saving mode, a sleep mode, or a defrosting mode). In order to simplify the description, it is assumed that each refrigerant system can select only one of the normal operation mode and the defrosting operation mode as the operation mode. That is, the refrigeration cycle control device 3 according to the sixth embodiment instructs each of the refrigerant systems to perform either the normal operation mode or the defrosting operation mode.

In the refrigeration cycle device according to the sixth embodiment, an evaluation function handled by the evaluation unit 7 preferably includes a penalty term that imposes a penalty for overlapping of the defrosting operation modes at the same time. As for the penalty term, the larger the number of refrigerant systems in which the defrosting operation modes overlap with each other, the larger the penalty. In addition, as for the penalty term, the longer a time during which the defrosting operation modes overlap with each other, the larger the penalty. For example, as the penalty term (Vp), the following mathematical equation can be considered.

V p = t = 0 t = k switch ( t ) N d ( t ) T W T WN d ( t ) dt Equation l wherein k switch ( t ) = { 1 , N d ( t ) T N d ( t ) 2 0 , otherwise N d ( t ) = [ n 1 ( t ) n 2 ( t ) n d ( t ) ] for i = 1 to d , n i ( t ) = { 0 , i th refrigerant system is in normal operation mode 1 , i th refrigerant system is in defrosting operation mode

Here, the subscript p of Vp representing the penalty term is derived from the initial letter of Penalty.

In Equation (1), kswitch(t) is a function serving as a switch that takes 1 when the defrosting operation modes of the refrigerant systems overlap with each other and takes 0 otherwise.

In Equation (1), W represents a weighting coefficient matrix representing a weight for each refrigerant system. The superscript T represents a transposition operation. W of the weight matrix may be a diagonal matrix in which components other than diagonal components take 0.

In Equation (1), ni(t) is a function that takes 1 when the i-th refrigerant system is in a defrosting operation mode and takes 0 when the i-th refrigerant system is in a normal operation mode.

As described above, the refrigeration cycle device according to the present disclosed technique can continue a normal operation while keeping the frosting-related state at a value equal to or lower than the threshold as long as the operation amount does not exceed a limit thereof, that is, when the actuator has sufficient capability.

The refrigeration cycle control device 3 according to the sixth embodiment determines that a condition for performing a normal operation (hereinafter, referred to as “normal operation condition”) is not satisfied for a refrigerant system in which capability of the actuator is insufficient and at least one of the frosting-related states exceeds the threshold, and performs control to switch the operation mode to a defrosting operation mode.

As described above, the refrigeration cycle device according to the sixth embodiment has the above configuration, and therefore exhibits an effect of observing the frosting-related state in each refrigerant system, performing a defrosting operation on a refrigerant system requiring the defrosting operation, and minimizing the number of refrigerant systems that perform the defrosting operation at the same time. Due to this effect, the refrigeration cycle device according to the sixth embodiment can continue a normal operation by any one of the refrigerant systems, and therefore comfort is not lost. In addition, the refrigeration cycle device according to the sixth embodiment can minimize the number of refrigerant systems to be in a defrosting operation mode at the same time, and therefore can be expected to contribute to energy saving.

Seventh Embodiment

A refrigeration cycle device according to a seventh embodiment is a modification of the refrigeration cycle device according to the sixth embodiment. In the seventh embodiment, the same reference signs as those used in the above-described embodiments are used unless otherwise specified. In the seventh embodiment, description overlapping with the above-described embodiments is appropriately omitted.

In the sixth embodiment, the evaluation unit 7 introduces the penalty term exemplified in mathematical equation (1) into an evaluation function to be used, and minimizes the number of refrigerant systems to be in the defrosting operation mode at the same time. The number of refrigerant systems to be in the defrosting operation mode at the same time can also be minimized by another method. The seventh embodiment minimizes the number of refrigerant systems to be in the defrosting operation mode at the same time without using the evaluation function.

FIG. 8 is a block diagram illustrating a functional configuration of the refrigeration cycle device including a refrigeration cycle control device 3 according to the seventh embodiment. As illustrated in FIG. 8, the refrigeration cycle control device 3 according to the seventh embodiment is an aspect in a case where the operation amount calculating unit 6 does not need the evaluation unit 7 and the optimization calculating unit 8.

FIG. 9 is a flowchart illustrating a processing content of the refrigeration cycle control device 3 according to the seventh embodiment. As illustrated in FIG. 9, processing steps of the refrigeration cycle control device 3 according to the seventh embodiment include steps S1, S12, S3, and S16.

The processing steps of the refrigeration cycle control device 3 according to the seventh embodiment are performed in the order (sequence) illustrated in FIG. 9. Steps S1 and S3 in the seventh embodiment have the same content as steps S1 and S3 (refer to FIG. 2) described in the first embodiment.

In FIG. 9, step S12 described as “Determine provisional operation amount” is a processing step in which the operation amount calculating unit 6 constituting the refrigeration cycle control device 3 generates a provisional operation amount array.

At the stage of step S12, the operation amount calculating unit 6 provisionally generates a provisional operation amount array indicating that all the refrigerant systems are in a normal operation mode at all times.

In FIG. 9, step S16 described as “Determine operation amount” is a processing step in which the operation amount calculating unit 6 determines an operation amount by a method described later.

FIG. 10 is an explanatory diagram illustrating a method for determining an operation amount by the refrigeration cycle control device 3 of the refrigeration cycle control device 3 (mainly the operation amount calculating unit 6) according to the seventh embodiment. Specifically, FIG. 10 is a graph illustrating a time course of a predicted frosting amount with the horizontal axis representing time and the vertical axis representing a predicted frosting amount. In the graph illustrated in FIG. 10, a time course of a predicted frosting amount for each of four refrigerant systems described as “refrigerant system 1”, “refrigerant system 2”, “refrigerant system 3”, and “refrigerant system 4” is exemplified.

In the graph illustrated in FIG. 10, Fn appearing in the title of the vertical axis represents an upper limit value of a frosting amount determined not to allow a more frosting amount. That is, in the graph illustrated in FIG. 10, the predicted frosting amount on the vertical axis is normalized by the frosting amount upper limit value (Fn). Therefore, in the graph illustrated in FIG. 10, a horizontal line with “1” on the vertical axis represents a boundary where a frosting amount exceeds the upper limit value. It is assumed that the normal operation condition described above is that the predicted frosting amount does not exceed the frosting amount upper limit value (Fn) in each refrigerant system.

In the graph illustrated in FIG. 10, t1_1 described on the horizontal axis represents a time at which a frosting amount is predicted to reach the upper limit value in the refrigerant system 1. t2_1 represents a time at which a frosting amount is predicted to reach the upper limit value in the refrigerant system 2, and t3_1 represents a time at which a frosting amount is predicted to reach the upper limit value in the refrigerant system 3.

In the graph illustrated in FIG. 10, tT described on the horizontal axis represents the furthest future time assumed by the refrigeration cycle control device 3.

“Rd” described in the upper right part of the graph illustrated in FIG. 10 represents a set of refrigerant systems in which it is estimated that a frosting amount does not exceed the frosting amount upper limit value (Fn). FIG. 10 illustrates a situation in which the refrigerant system 1, the refrigerant system 2, and the refrigerant system 3 are included in the set of Rd.

FIG. 11 is a flowchart illustrating a processing content for determining an operation amount by the refrigeration cycle control device 3 (mainly the operation amount calculating unit 6) according to the seventh embodiment. As illustrated in FIG. 11, processing steps of the refrigeration cycle control device 3 according to the seventh embodiment include steps S21 to S25.

In FIG. 11, step S21 described as “Acquire predicted frosting amount” is a processing step performed by the prediction unit 9. In step S21, the prediction unit 9 calculates a prediction value of a frosting amount for each refrigerant system. In step S21, the prediction unit 9 is ready to perform plotting for the graph illustrated in FIG. 10.

In FIG. 11, step S22 described as “Initialize defrosting operable time” is a processing step performed by the operation amount calculating unit 6. In step S22, the operation amount calculating unit 6 initializes a defrosting operable time (td). An initial value of the defrosting operable time (td) may be, for example, tT. The concept of the defrosting operable time (td) will be apparent from description of subsequent processing steps.

In FIG. 11, step S23 described as “Rd is not empty?” is a processing step performed by the operation amount calculating unit 6. In step S23, the operation amount calculating unit 6 checks whether or not Rd is an empty set.

If Rd is an empty set, the processing step is ended. If Rd is not an empty set, the processing step proceeds to step S24.

In FIG. 11, step S24 described as “Determine defrosting period of refrigerant system i” is a processing step performed by the operation amount calculating unit 6. In step S24, the operation amount calculating unit 6 determines a period during which the defrosting operation is to be performed for a refrigerant system included in Rd.

When determining the period during which the defrosting operation is to be performed, the operation amount calculating unit 6 needs to grasp in advance a time length during which each refrigerant system continues the defrosting operation. “td_1” appearing on the horizontal axis of the graph in FIG. 10 represents a time length during which the refrigerant system 1 continues the defrosting operation by a double-headed arrow line segment. Similarly, “td_2” represents a time length during which the refrigerant system 2 continues the defrosting operation, and “td_3” represents a time length during which the refrigerant system 3 continues the defrosting operation.

The operation amount calculating unit 6 first determines the period during which the defrosting operation is to be performed for a refrigerant system having the slowest (or longest) time at which a frosting amount exceeds the frosting amount upper limit value (Fn) among the refrigerant systems included in Rd. For example, in the case illustrated in FIG. 10, the refrigerant system 3 has the slowest (or longest) time at which a frosting amount exceeds the frosting amount upper limit value (Fn) among the refrigerant systems included in Rd. When the defrosting operation start time (“t3_2” in FIG. 10, in which the subscript “3” indicates the refrigerant system 3 and the subscript “_2” indicates the defrosting operation start time) is set to a time at which a frosting amount is predicted to reach an upper limit value (“t3_1” in FIG. 10) for the refrigerant system 3, the operation amount calculating unit 6 first checks whether refrigerant systems performing the defrosting operation overlap with each other. As illustrated in FIG. 10, for the refrigerant system 3, refrigerant systems in which the defrosting operation is performed do not overlap with each other even when the defrosting operation start time (t3_2) is set to the time (t3_1) at which the frosting amount is predicted to reach the upper limit value. Therefore, the operation amount calculating unit 6 determines the period during which the defrosting operation is to be performed as from t3_1 to t3_1+td_3.

In FIG. 11, step S25 described as “Update defrosting operable time and Rd” is a processing step performed by the operation amount calculating unit 6. In step S25, the operation amount calculating unit 6 updates the defrosting operable time (td) and updates Rd.

In the case illustrated in FIG. 10, the operation amount calculating unit 6 substitutes t3_2 for td, that is, the defrosting operation start time (t3_2) for the refrigerant system 3 determined in the immediately preceding step S24 for td. As described above, the defrosting operable time (td) indicates that “By this time, even if the defrosting operation is performed, refrigerant systems in which the defrosting operation is performed do not overlap with each other”.

In the refrigerant system 3, the defrosting operation is started at the defrosting operation start time (t3_2), and therefore the refrigerant system 3 no longer belongs to the set Rd. In the case illustrated in FIG. 10, the operation amount calculating unit 6 deletes the refrigerant system 3 from Rd in step S25.

As illustrated in FIG. 11, after step S25, the processing step proceeds to step 23 again.

In the case illustrated in FIG. 10, when the defrosting operation start time (“t2_2” in FIG. 10, in which the first subscript “2” indicates the refrigerant system 2 and the subscript “_2” indicates the defrosting operation start time) is set to a time at which a frosting amount is predicted to reach an upper limit value (“t2” in FIG. 10) for the refrigerant system 2, in step S24 of the second cycle, the operation amount calculating unit 6 checks whether refrigerant systems performing the defrosting operation overlap with each other. In the case illustrated in FIG. 10, when the defrosting operation start time (t2_2) is set to a time at which a frosting amount is predicted to reach an upper limit value (t2_1) for the refrigerant system 2, the defrosting operation period of the refrigerant system 2 overlaps with that of the refrigerant system 3.

Therefore, the operation amount calculating unit 6 makes the defrosting operation start time (t2_2) of the refrigerant system 2 slightly earlier than the time (t2_1) at which a frosting amount is predicted to reach an upper limit value in such a manner that the refrigerant systems performing the defrosting operation do not overlap with each other. The operation amount calculating unit 6 determines the defrosting operation start time on the basis of, for example, the following mathematical equation.

for i = 1 to d , ( 2 ) t i , 2 = min ( t i _ 1 , t d - t d_i )

In the case illustrated in FIG. 10, the operation amount calculating unit 6 determines t2_2 of the defrosting operation start time for the refrigerant system 2 as td−td_2.

In the seventh embodiment, particularly in the case illustrated in FIG. 10, the defrosting operation is determined on the basis of the “defrosting amount”. However, the present disclosed technique is not limited thereto. The refrigeration cycle device according to the present disclosed technique may determine the defrosting operation on the basis of another frosting-related state.

As described above, the refrigeration cycle device according to the seventh embodiment has the above configuration, and therefore exhibits the effects described in the sixth embodiment without performing the calculation based on the evaluation function.

INDUSTRIAL APPLICABILITY

The present disclosed technique can be applied to a room air conditioner and an industrial cooling device, and has industrial applicability.

REFERENCE SIGNS LIST

1: operation unit, 2: measurement unit, 3: refrigeration cycle control device (refrigeration cycle controller), 4: control unit, 5: refrigeration cycle state predicting device (refrigeration cycle state predictor), 6: operation amount calculating unit (operation amount calculator), 7: evaluation unit (evaluator), 8: optimization calculating unit (optimization calculator), 9: prediction unit (predictor), 10: model storage unit, 11: integration calculating unit, 12: estimation unit, 13: second model storage unit

Claims

1. A refrigeration cycle controller constituting a refrigeration cycle device, wherein, V p = ∫ t = 0 t = ∞ k switch ( t ) ⁢ N d ( t ) T ⁢ W T ⁢ WN d ( t ) ⁢ dt Equation ⁢ l wherein k switch ( t ) = { 1, N d ( t ) T ⁢ N d ( t ) ≥ 2 0, otherwise N d ( t ) = [ n 1 ( t ) n 2 ( t ) ⋮ n d ( t ) ] for ⁢ i = 1 ⁢ to ⁢ d, n i ( t ) = { 0, i th ⁢ refrigerant ⁢ system ⁢ is ⁢ in ⁢ normal ⁢ operation ⁢ mode 1, i th ⁢ refrigerant ⁢ system ⁢ is ⁢ in ⁢ defrosting ⁢ operation ⁢ mode where W represents a weighting coefficient matrix representing a weight for each refrigerant system.

the refrigeration cycle device includes a plurality of indoor equipments having different refrigerant systems in the same room, or includes an indoor equipment having a plurality of heat exchangers having different refrigerant systems,
the refrigeration cycle controller comprising:
an operation amount calculator to calculate a provisional operation amount array by calculation in such a manner as to satisfy a constraint condition; and
a refrigeration cycle state predictor including a predictor to calculate a prediction value in a future state of the refrigeration cycle device by referring to the input provisional operation amount array,
wherein,
the operation amount calculator includes, an evaluator to calculate a value of an evaluation function used to calculate the provisional operation amount array, and an optimization calculator to obtain a solution of the provisional operation amount array minimizing the value of the evaluation function,
and wherein,
the evaluation function includes the following penalty term (Vp),

2. A refrigeration cycle controller constituting a refrigeration cycle device, wherein, if Rd is an empty set, the processing operation is ended, and if Rd is not an empty set, the processing operation proceeds to operation (d);

the refrigeration cycle device includes a plurality of indoor equipments having different refrigerant systems in the same room, or includes an indoor equipment having a plurality of heat exchangers having different refrigerant systems,
the refrigeration cycle controller comprising:
an operation amount calculator to calculate a provisional operation amount array by calculation in such a manner as to satisfy a constraint condition; and
a refrigeration cycle state predictor including a predictor to calculate a prediction value in a future state of the refrigeration cycle device by referring to the input provisional operation amount array,
wherein,
the operation amount calculator executes the following operation of,
(a) provisionally generates a provisional operation amount array indicating that all the refrigerant systems are in a normal operation mode at all times;
(b) initializes a defrosting operable time (td);
(c) by defining a set Rd being the set of refrigerant systems in which the amount of frost is estimated not to exceed the upper limit of the amount of frost (Fn), checks whether or not Rd is an empty set, and,
(d) determines the period during which the defrosting operation is to be performed for a refrigerant system having the slowest time at which a frosting amount exceeds the frosting amount upper limit value (Fn) among the refrigerant systems included in Rd;
(e) updates the defrosting operable time (td) and updates Rd.

3. A refrigeration cycle device comprising the refrigeration cycle controller according to claim 1.

Patent History
Publication number: 20250020376
Type: Application
Filed: Sep 30, 2024
Publication Date: Jan 16, 2025
Applicant: Mitsubishi Electric Corporation (Tokyo)
Inventors: Rin ITO (Tokyo), Koki NAKANE (Tokyo), Kosuke TANAKA (Tokyo), Ryuichi TAKEMURA (Tokyo), Toshisada MARIYAMA (Tokyo)
Application Number: 18/901,906
Classifications
International Classification: F25B 49/00 (20060101); F25B 47/02 (20060101);