Air conditioning equipment control device, air conditioning equipment control method, and program

Abstract

This air conditioning equipment control device comprises: an acquisition unit that acquires a measurement value including a temperature that is measured at a predetermined position of a facility, and a setting condition of the air conditioning equipment including a set temperature of an air conditioner; a prediction unit that predicts a future heat load corresponding to each of a plurality of setting condition candidates, by using a model in which the measurement value and the setting condition are used as explanatory variables and a heat load due to heat generation of a heating element is used as a response variable; and a control unit that selects a setting condition candidate from among the setting condition candidates, in which setting condition candidate a prediction value that satisfies an environmental condition that defines a reference value or a reference range of the heat load is obtained, and controls the air conditioning equipment on the basis of the selected setting candidate condition.

Description

Air conditioning equipment control device, air conditioning equipment control method, and program

[0001] This application claims priority to Japanese Patent Application No. 2023-133892, filed on August 21, 2023, the contents of which are incorporated herein by reference.

[0002] In facilities (e.g., server rooms) where many heat-generating elements such as IT devices (e.g., servers and communication devices) are installed, air conditioning equipment is installed to cool the heat-generating elements (see Patent Documents 1 and 2). In such air conditioning equipment, the setting conditions (e.g., temperature) are generally changed according to the season, taking into account factors such as power efficiency.

[0003] Japanese Patent No. 5801732 Japanese Patent Laid-Open No. 2023-26104

[0004] In conventional air conditioning equipment control methods, the set conditions remain constant throughout a given season. However, the heat load of the heating element (such as the rise in room temperature due to heat generated by the heating element) changes not only with the season but also with the time of day. Therefore, in reality, the ideal set conditions change from moment to moment, even throughout the day.

[0005] An object of the present disclosure is to provide an air conditioning equipment control device, an air conditioning equipment control method, and a program that can operate the air conditioning equipment under optimal setting conditions in accordance with the heat load that changes from moment to moment.

[0006] According to one aspect of the present disclosure, a control device for an air conditioning equipment is a control device for an air conditioning equipment including an air conditioner installed in a facility where multiple heat generating elements are arranged, and includes: an acquisition unit that acquires measurement values ​​including temperatures measured at predetermined locations in the facility and setting conditions of the air conditioning equipment including the set temperatures of the air conditioners; a prediction unit that predicts future heat loads corresponding to multiple setting condition candidates using a model in which the measurement values ​​and the setting conditions are explanatory variables and the heat load due to the heat generated by the heat generating elements is a target variable; and a control unit that selects, from the setting condition candidates, a setting condition candidate that has obtained a predicted value that satisfies environmental conditions that specify a reference value or reference range for the heat load, and controls the air conditioning equipment based on the selected candidate setting conditions.

[0007] According to one aspect of the present disclosure, a method for controlling air conditioning equipment is a method for controlling air conditioning equipment including an air conditioner installed in a facility where multiple heat generating elements are arranged, and includes the steps of: acquiring measurement values ​​including a temperature measured at a predetermined position in the facility and setting conditions of the air conditioning equipment including a set temperature of the air conditioner; predicting future heat loads corresponding to multiple setting condition candidates using a model in which the measurement values ​​and the setting conditions are explanatory variables and the heat load due to the heat generated by the heat generating elements is a target variable; and selecting, from the setting condition candidates, a setting condition candidate that has obtained a predicted value that satisfies environmental conditions that specify a reference value or reference range for the heat load, and controlling the air conditioning equipment based on the selected candidate setting conditions.

[0008] According to one aspect of the present disclosure, the program causes a control device of an air conditioning equipment including an air conditioner installed in a facility where multiple heat generating elements are arranged to perform the following steps: acquiring measurement values ​​including temperatures measured at predetermined locations in the facility and setting conditions of the air conditioning equipment including the set temperatures of the air conditioners; predicting future heat loads corresponding to multiple setting condition candidates using a model in which the measurement values ​​and the setting conditions are explanatory variables and the heat load due to the heat generated by the heat generating elements is a target variable; selecting, from the setting condition candidates, a setting condition candidate that has obtained a predicted value that satisfies the environmental conditions that specify the reference value or reference range of the heat load, and controlling the air conditioning equipment based on the selected candidate setting conditions.

[0009] According to the above aspect, the air conditioning equipment can be operated under optimal setting conditions in accordance with the heat load that changes from moment to moment.

[0010] 1 is a diagram showing the overall configuration of an air conditioning system according to a first embodiment; FIG. 2 is a block diagram showing the functional configuration of a control device according to the first embodiment; FIG. 3 is a flowchart showing an example of processing of a control device according to the first embodiment; FIG. 4 is a diagram showing an example of a model according to the first embodiment; and FIG. 5 is a diagram for explaining the function of a control device according to the first embodiment.

[0011] First Embodiment Hereinafter, an embodiment will be described in detail with reference to the drawings.

[0012] (Overall Configuration) Fig. 1 is a diagram showing the overall configuration of an air conditioning system according to Embodiment 1. As shown in Fig. 1, the air conditioning system 1 includes a plurality of heating elements 2, an air conditioning unit 3, and a control device 10.

[0013] The heat generating element 2 is, for example, IT equipment such as a server or communication device located within a facility R. The facility R may be a server room or the like. The heat generating element 2 has a fan 20 that takes in cold air from the outside and expels air heated by the heat generated by the heat generating element 2. As shown in FIG. 1 , the heat generating elements 2 are arranged in a row in the left-right direction of the paper and in a row in the up-down direction of the paper, with an aisle between them. In this case, the heat generating elements 2 are arranged so that the intake sides of the fans 20 of each heat generating element 2 face each other and the exhaust sides face each other in the up-down direction of the paper. The aisle where the intake sides of the fans 20 are located is also referred to as a cold aisle CA, and the aisle where the exhaust sides enter is also referred to as a hot aisle HA.

[0014] The heating element 2 is also provided with an intake temperature sensor 21 that measures the intake temperature of the heating element 2 (fan 20) and an exhaust temperature sensor 22 that measures the exhaust temperature. In other embodiments, instead of or in addition to the intake temperature sensor 21 and the exhaust temperature sensor 22, a temperature sensor that measures the temperature of the CPU, GPU, or the like (hereinafter collectively referred to as CPU temperature) of the heating element 2 may be provided.

[0015] The air conditioning equipment 3 has at least one air conditioner 30. The air conditioner 30 draws in air within the facility R, cools it using a heat exchanger (not shown), and discharges it into the facility R. The number of air conditioners 30 may be changed as desired depending on the size of the facility R, the number of heating elements 2, etc.

[0016] The air conditioner 30 is also provided with an intake temperature sensor 31 that measures the temperature of the intake air, and an exhaust temperature sensor 32 that measures the temperature of the exhaust air (cooled air).

[0017] The air conditioning equipment 3 may further include at least one humidifier 35. The humidifier 3 adjusts the humidity within the facility R. The number of humidifiers 35 may be changed as desired depending on the size of the facility R, etc.

[0018] The fan 20 of the heating element 2 may also be used as one of the air conditioning units 3 .

[0019] In addition, the facility R is provided with a temperature sensor 51 that measures the room temperature of the facility R, and a humidity sensor 52 that measures the humidity of the facility R.

[0020] The control device 10 controls the air conditioning equipment 3 so that the environment within the facility R satisfies predetermined environmental conditions. The environmental conditions are described in detail below, but are, for example, target values ​​for the room temperature of the facility R, the humidity of the facility R, and the like.

[0021] (Functional Configuration of Control Device) Fig. 2 is a block diagram showing the functional configuration of the control device according to the first embodiment. As shown in Fig. 2, the control device 10 includes a processor 11, a memory 12, a storage 13, and a communication interface 14.

[0022] The processor 11 operates in accordance with a predetermined program to cause the control device 10 to perform various functions. The functions of the processor 11 will be described later.

[0023] The memory 12 has a memory area necessary for the operation of the processor 11 .

[0024] The storage 13 is a so-called auxiliary storage device, such as a hard disk drive (HDD), a solid state drive (SSD), etc. The storage 13 stores data that each part of the processor 11 acquires, generates, and references during processing.

[0025] The communication interface 14 is an interface for transmitting and receiving various data (measurement values), control signals, etc. between each sensor and the air conditioning equipment 3 (air conditioner 30, humidifier 35).

[0026] Next, a description will be given of the functions of the processor 11. The processor 11 functions as an acquisition unit 111, a prediction unit 112, a control unit 113, and a learning unit 114 by operating in accordance with a program prepared in advance.

[0027] The acquisition unit 111 acquires the measurement values ​​of each sensor.

[0028] The sensor measurement value is, for example, a temperature measured at a predetermined position in the facility R. The temperature measured at a predetermined position in the facility R is at least one of the room temperature of the facility R measured by the temperature sensor 51, the intake temperature of the air conditioner 30 measured by the intake temperature sensor 31, the exhaust temperature of the air conditioner 30 measured by the exhaust temperature sensor 32, the intake temperature of the heating element 2 (fan 20) measured by the intake temperature sensor 21, the exhaust temperature of the heating element 2 (fan 20) measured by the exhaust temperature sensor 22, and the temperature of the heating element (CPU temperature). In addition, if the air conditioning equipment 3 has a humidifier 35, the sensor measurement value may further include the humidity of the facility R measured by the humidity sensor 52.

[0029] The prediction unit 112 predicts the future heat load corresponding to each of the multiple setting condition candidates using a model M that uses the measured values ​​and the setting conditions of the air conditioning equipment 3 as explanatory variables and the heat load due to the heat generation of the heating element 2 as a target variable. The model M is a model that has been trained in advance by the learning unit 114, which will be described later.

[0030] The setting conditions of the air conditioning equipment 3 include, for example, the set temperature of the air conditioners 30. The setting conditions of the air conditioning equipment 3 may also include the set air volume of the air conditioners 30. If the air conditioning equipment 3 has multiple air conditioners 30, the setting conditions may also include the number of operating air conditioners 30. In this case, the setting conditions include, in addition to the number of operating air conditioners 30, the ON (operating) / OFF (not operating) setting of each air conditioner 30, and the set temperature and set air volume of each operating air conditioner 30.

[0031] Furthermore, if the air conditioning equipment 3 has humidifiers 35, the setting conditions may include the number of humidifiers in operation. In this case, the setting conditions include the ON (operation) / OFF (non-operation) setting of each humidifier 35 in addition to the number of humidifiers in operation.

[0032] Furthermore, the set conditions may include the air volume or rotation speed of the fan 20 of each heating element 2 .

[0033] The thermal load includes at least one of the power usage effectiveness (PUE) of the facility R, which is affected by the heat generated by the heating element 2, the room temperature of the facility R, the intake temperature of the air conditioner 30, the exhaust temperature of the air conditioner 30, the intake temperature of the heating element 2 (fan 20), the exhaust temperature of the heating element 2 (fan 20), and the temperature (CPU temperature) of the heating element 2. Furthermore, if the air conditioning equipment 3 has a humidifier 35, the thermal load may further include the humidity of the facility R.

[0034] The control unit 113 selects, from among the candidate setting conditions, a candidate setting condition that has yielded a predicted value that satisfies the environmental conditions that specify the reference value of the thermal load, and controls the air conditioning equipment 3 based on the selected candidate setting condition.

[0035] The environmental conditions are the reference values ​​(upper or lower limit values) or reference ranges (ranges from the lower limit value to the upper limit value) of the heat loads described above. The reference values ​​or reference ranges of each heat load are set arbitrarily depending on the number, arrangement, specifications, etc. of the heating elements 2 and air conditioning equipment 3.

[0036] The learning unit 114 uses time-series data of the measurement values ​​of each sensor and the heat load calculated from the measurement values ​​and the set conditions of the air conditioning equipment 3 as learning data to machine-learn the model M. In the learning phase, the learning unit 114 learns the model M in advance using a predetermined number of time-series data and stores the learned model M in the storage 13.

[0037] Furthermore, after the learning phase is completed and the operation phase is started, the learning unit 114 re-learns and updates the model M. Specifically, the learning unit 114 calculates the actual value of the thermal load at a certain evaluation time from the measurement values ​​acquired at the evaluation time and the set conditions, and learns and updates the model M based on the error between the actual value and the predicted value of the thermal load at the evaluation time predicted in the past. The evaluation time is, for example, the current time, but may also be a time in the past (e.g., n minutes before the current time).

[0038] The predetermined program executed by the processor 11 is stored on a computer-readable recording medium. Computer-readable recording media include magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, and semiconductor memories. The computer program may be distributed to a computer via a communication line, and the computer that receives the distribution may execute the program. The program may also be a program for implementing some of the above-described functions. Furthermore, the program may be a so-called differential file (differential program) that can realize the above-described functions in combination with a program already stored in the computer system.

[0039] (Processing flow of the control device) Fig. 3 is a flowchart showing an example of processing of the control device according to the first embodiment. Hereinafter, details of the processing of the control device 10 will be described with reference to Fig. 3. It is assumed that the learning phase has been completed at the time of Fig. 3. Therefore, Fig. 3 shows processing in the operation phase of the control device 10.

[0040] The acquisition unit 111 acquires the measurement value at the current time from each sensor (step S01).

[0041] Next, the control device 10 controls the air conditioning equipment 3 (step S02) and updates the model M (step S03). Note that, although an example in which steps S02 and S03 are executed in parallel is shown in FIG. 3, this is not limiting. In other embodiments, step S03 may be executed after step S02, or step S02 may be executed after step S03.

[0042] First, we will explain the control process S02 of the air conditioning equipment 3. The prediction unit 112 reads out the model M recorded in the storage 13, inputs the measurement value at the current time acquired in step S01 and candidate setting conditions for the air conditioning equipment 3 to be applied from the current time onwards into the model M, and obtains a predicted value of the future heat load as an output (step S21).

[0043] FIG. 4 is a diagram illustrating an example of a model according to the first embodiment. As shown in FIG. 4 , the model M according to this embodiment is a neural network that receives explanatory variables as input and outputs a response variable. For simplicity, the explanatory variables input to the model M are three set conditions: the number of operating air conditioners, the set temperatures of the operating air conditioners 30, and the number of operating humidifiers; and two measured values: the room temperature and humidity of the facility R. The response variables output from the model M are predicted values ​​of the power usage efficiency (PUE), which represents the future (e.g., 30 minutes from now) heat load, the room temperature of the facility R, and the humidity of the facility R. The PUE is an index obtained by dividing the total power consumption within the facility R by the power consumption of the heating element 2. The smaller the PUE value, the higher the efficiency. The storage 13 of the control device 10 records the actual set values ​​for the number of operating air conditioners, the set temperatures of the air conditioners, and the number of operating humidifiers at each time, as well as the measured values ​​of the room temperature, humidity, and PUE. The trained model M recorded in the storage 13 is machine-learned using this past time-series data as training data in the training phase.

[0044] In the operation phase shown in FIG. 3 , in step S21, the current room temperature and humidity (measured values ​​acquired in step S01) and candidate setting conditions to be applied from the current time onward (combinations of the number of operating air conditioners, the set temperatures of the air conditioners, and the number of operating humidifiers) are input to model M as explanatory variables. The objective variables output from model M are predicted values ​​of future PUE, room temperature, and humidity when the candidate setting conditions are applied. In other words, the prediction unit 112 can predict how the future heat load will change when the setting conditions are changed. In this embodiment, the prediction unit 112 inputs multiple candidate setting conditions to model M and obtains, as output, a predicted value of the future heat load corresponding to each candidate setting condition.

[0045] Next, the control unit 113 selects optimal setting conditions that satisfy the environmental conditions of the facility R based on the setting condition candidates and the predicted value of the future heat load (step S22).

[0046] The environmental conditions of the facility R are, for example, a reference value (upper or lower limit) or a reference range (a range from the upper limit to the lower limit) of the thermal load. In the above example, the reference value of the room temperature (upper limit: X degrees or less) and the reference value of the humidity (lower limit: Y% or more) are set as the environmental conditions to be satisfied for the thermal load. The control unit 113 selects, as the optimal setting conditions, the candidate setting conditions that satisfy the environmental conditions for both the room temperature and humidity and minimize the PUE (step S22).

[0047] FIG. 5 is a diagram illustrating the functions of the control device according to the first embodiment. Furthermore, the control unit 113 may generate a response surface showing the sensitivity of the setting conditions to the thermal load in the current state (room temperature, humidity) as shown in FIG. 5 , and identify the optimal setting conditions using the response surface. FIG. 5 illustrates (a) a predicted PUE value, (b) a predicted room temperature value, and (c) a predicted humidity value corresponding to each combination of setting conditions. Note that FIG. 5 illustrates an example of a response surface showing the correspondence between the combination of the setting temperatures of two air conditioners 30 (air conditioners 1 and 2) and the thermal load, among the setting conditions. However, similar response surfaces are generated for other combinations of setting conditions.

[0048] For some thermal loads, a reference value (upper or lower limit) or a reference range (a range from upper limit to lower limit) is predefined as an environmental condition of the facility R. For example, a reference value C1 for room temperature (e.g., X degrees or less) and a reference value C2 for humidity (e.g., Y% or more) are predefined as environmental conditions. The control unit 113 narrows down the set condition candidates that satisfy both the room temperature and humidity environmental conditions based on the room temperature and humidity response surfaces shown in (b) and (c) of FIG. 5 and the environmental conditions (room temperature reference value C1 and humidity reference value C2). Furthermore, the control unit 113 references the PUE response surface shown in (a) of FIG. 5 and selects the set condition candidate that minimizes the PUE as the optimal set condition (step S22). The control unit 113 records the identified set condition and the corresponding predicted value of the thermal load in the storage 13.

[0049] Next, the control unit 113 outputs a control signal to each of the air conditioning equipment 3 (air conditioner 30, humidifier 35) based on the selected setting conditions (step S23).

[0050] By repeatedly executing the above process, the control device 10 can predict the heat load, which changes from moment to moment, and operate the air conditioning equipment 3 under optimal settings to minimize power (electricity consumption).

[0051] 3, the update process S03 of the model M in the learning unit 114 will be described. First, the learning unit 114 identifies the thermal load at the current time (time t) based on the measurement values ​​acquired in step S01 (step S31). Here, as in step S02, the thermal load is assumed to be the room temperature, humidity, and PUE of the facility R. The measurement values ​​acquired in step S01 may be used as they are for the room temperature and humidity. The PUE is calculated by measuring the power consumption of the heating element 2 and the power consumption of the entire facility R, for example, using a power meter (not shown).

[0052] Next, the learning unit 114 learns the error between the previously predicted "predicted value of the thermal load at time t" and the "actually measured value of the thermal load at time t" and updates the model M (step S32). For example, assume that the control unit 113 is predicting the thermal load 30 minutes from now. In this case, the learning unit 114 reads from the storage 13 the predicted value of the thermal load predicted by the control unit 113 30 minutes before time t (the current time) and determines the error between this predicted value and the actual measured value of the thermal load identified in step S31. Then, the learning unit 114 updates the model M by learning weights, for example, so as to minimize an evaluation function consisting of this error.

[0053] By constantly relearning and updating the model M in this way, the learning unit 114 can improve the accuracy of prediction of the thermal load by the model M.

[0054] (Actions and Effects) As described above, the control device 10 of the air conditioning equipment 3 according to this embodiment includes an acquisition unit 111 that acquires the measurement values ​​measured by each sensor and the setting conditions of the air conditioning equipment 3, a prediction unit 112 that predicts the future heat load corresponding to each of a plurality of setting condition candidates using a model M that uses the measurement values ​​and the setting conditions as explanatory variables and the heat load due to heat generation by the heating element 2 as a target variable, and a control unit 113 that selects, from the setting condition candidates, a setting condition candidate that has obtained a predicted value that satisfies the environmental conditions, and controls the air conditioning equipment 3 based on the selected candidate setting conditions.

[0055] In this way, the control device 10 can predict the heat load that changes from moment to moment, and operate the air conditioning equipment 3 under optimal setting conditions to minimize power (electricity consumption).

[0056] In addition, the control device 10 further includes a learning unit 114 that identifies the actual value of the thermal load at the time to be evaluated from the measurement values ​​and set conditions acquired at the time to be evaluated, and learns and updates the model M based on the error between the identified actual value of the thermal load and the predicted value of the thermal load at the time to be evaluated that was predicted in the past.

[0057] By doing this, the control device 10 can improve the accuracy of predicting the heat load using the model M the more the heating element 2 and the air conditioning equipment 3 are operated.

[0058] In addition, the measured values ​​are at least one of the room temperature of facility R, the intake temperature of air conditioner 30, the exhaust temperature of air conditioner 30, the intake temperature of heating element 2, the exhaust temperature of heating element 2, and the temperature of heating element 2 (CPU temperature).

[0059] In this way, the control device 10 can predict the heat load and select the optimal setting conditions by using various temperatures measured in the facility R. In other words, the control device 10 can use the temperature sensors already installed in the facility R, the heating element 2, and the air conditioning equipment 3, thereby reducing the cost of introducing new sensors.

[0060] The set conditions further include at least one of the number of operating air conditioners 30 , the air volume of the air conditioners 30 , and the air volume or rotation speed of the fan 20 of the heating element 2 .

[0061] In this way, the control device 10 can control the air conditioning equipment 3 by combining various setting conditions depending on the number and functions of the air conditioners 30, the specifications of the fans 20 of the heating elements 2, etc. Furthermore, by combining multiple setting conditions, the control device 10 can perform detailed control of the air conditioning equipment 3 so that the future heat load will be in a more appropriate state.

[0062] In addition, the thermal load includes at least one of the room temperature of facility R, the intake temperature of air conditioner 30, the exhaust temperature of air conditioner 30, the intake temperature of heating element 2, the exhaust temperature of heating element 2, and the temperature of heating element 2 (CPU temperature), as well as the PUE of facility R, and the control unit 113 selects setting condition candidates in which the predicted values ​​of the thermal load other than PUE satisfy the environmental conditions and the predicted value of PUE is minimized.

[0063] In this way, the control device 10 can control the air conditioning equipment 3 so that the future heat load is in a more appropriate state and the PUE is minimized.

[0064] Furthermore, the air conditioning equipment 3 further includes a plurality of humidifiers 35, the measurement values ​​and heat load further include the humidity of the facility R, and the setting conditions further include the number of humidifiers 35 in operation.

[0065] In this way, the control device 10 can control the air conditioning equipment 3 so that not only the temperature but also the humidity within the facility R is kept at an appropriate state.

[0066] Other Embodiments Although one embodiment has been described in detail above with reference to the drawings, the specific configuration is not limited to the above, and various design modifications are possible. That is, in other embodiments, the order of the above-described processes may be changed as appropriate. Furthermore, some processes may be executed in parallel.

[0067] <Additional Notes> The air conditioning equipment control device, the air conditioning equipment control method, and the program described in the above-described embodiments can be understood, for example, as follows.

[0068] (1) According to the first aspect, a control device 10 for an air conditioning equipment 3 including an air conditioner 30 installed in a facility R in which a plurality of heating elements 2 are arranged includes an acquisition unit 111 that acquires measurement values ​​including temperatures measured at predetermined positions in the facility R and setting conditions for the air conditioning equipment 3 including the set temperature of the air conditioner 30; a prediction unit 112 that predicts future heat loads corresponding to a plurality of candidate setting conditions using a model M in which the measurement values ​​and the setting conditions are explanatory variables and the heat load due to heat generation by the heating elements 2 is a target variable; and a control unit 113 that selects, from the candidate setting conditions, a candidate setting condition that has yielded a predicted value that satisfies the environmental conditions that specify the reference value or reference range of the heat load, and controls the air conditioning equipment 3 based on the selected candidate setting conditions.

[0069] In this way, the control device 10 can predict the heat load that changes from moment to moment, and operate the air conditioning equipment 3 under optimal setting conditions to minimize power (electricity consumption).

[0070] (2) According to the second aspect, the control device 10 of the air conditioning equipment 3 according to the first aspect further includes a learning unit 114 that identifies the actual value of the heat load at the time to be evaluated from the measurement values ​​and setting conditions acquired at the time to be evaluated, and learns and updates a model based on the error between the identified actual value of the heat load and the predicted value of the heat load at the time to be evaluated that was predicted in the past.

[0071] By doing this, the control device 10 can improve the accuracy of predicting the heat load using the model M the more the heating element 2 and the air conditioning equipment 3 are operated.

[0072] (3) According to the third aspect, in the control device 10 of the air conditioning equipment 3 relating to the first or second aspect, the temperature measured at a predetermined location in the facility R is at least one of the room temperature of the facility R, the intake temperature of the air conditioner 30, the exhaust temperature of the air conditioner 30, the intake temperature of the heating element 2, the exhaust temperature of the heating element 2, and the temperature of the heating element 2.

[0073] In this way, the control device 10 can predict the heat load and select the optimal setting conditions by using various temperatures measured in the facility R. In other words, the control device 10 can use the temperature sensors already installed in the facility R, the heating element 2, and the air conditioning equipment 3, thereby reducing the cost of introducing new sensors.

[0074] (4) According to a fourth aspect, in the control device 10 of the air conditioning equipment 3 relating to any one of the first to third aspects, the setting conditions further include at least one of the number of operating air conditioners 30 and the air volume of the air conditioners 30.

[0075] In this way, the control device 10 can control the air conditioning equipment 3 by combining various setting conditions depending on the number and functions of the air conditioners 30, the specifications of the fans 20 of the heating elements 2, etc. Furthermore, by combining multiple setting conditions, the control device 10 can perform detailed control of the air conditioning equipment 3 so that the future heat load will be in a more appropriate state.

[0076] (5) According to a fifth aspect, in the control device 10 of the air conditioning equipment 3 relating to any one of the first to fourth aspects, the air conditioning equipment 3 includes a fan 20 provided in the heating element 2, and the setting conditions further include the air volume or rotation speed of the fan 20.

[0077] (6) According to the sixth aspect, in the control device 10 of the air conditioning equipment 3 relating to any one of the first to fifth aspects, the thermal load includes at least one of the room temperature of the facility R, the intake temperature of the air conditioner 30, the exhaust temperature of the air conditioner 30, the intake temperature of the heating element 2, the exhaust temperature of the heating element 2, and the temperature of the heating element 2, as well as the power usage efficiency (PUE) of the facility R, and the control unit 113 selects a setting condition candidate that satisfies the environmental conditions for the predicted values ​​of the thermal load other than the power usage efficiency and minimizes the predicted value of the power usage efficiency.

[0078] In this way, the control device 10 can control the air conditioning equipment 3 so that the future heat load is in a more appropriate state and the PUE is minimized.

[0079] (7) According to a seventh aspect, in the control device 10 of the air conditioning equipment 3 relating to any one of the first to sixth aspects, the air conditioning equipment 3 further includes a plurality of humidifiers 35, the measured value and the heat load further include the humidity of the facility R, and the setting conditions further include the number of humidifiers in operation.

[0080] In this way, the control device 10 can control the air conditioning equipment 3 so that not only the temperature but also the humidity within the facility R is kept at an appropriate state.

[0081] (8) According to an eighth aspect, a method for controlling an air conditioning equipment 3 including an air conditioner 30 installed in a facility R in which a plurality of heating elements 2 are arranged includes the steps of: acquiring measurement values ​​including a temperature measured at a predetermined position in the facility R and setting conditions of the air conditioning equipment 3 including a set temperature of the air conditioner 30; predicting future heat loads corresponding to each of a plurality of candidate setting conditions using a model M in which the measurement values ​​and the setting conditions are explanatory variables and the heat load due to heat generation by the heating elements 2 is a target variable; and selecting, from the candidate setting conditions, a candidate setting condition that has obtained a predicted value that satisfies the environmental conditions that specify the reference value or reference range of the heat load, and controlling the air conditioning equipment 3 based on the selected candidate setting conditions.

[0082] (9) According to the ninth aspect, the program causes the control device 10 of the air conditioning equipment 3, including the air conditioner 30 installed in the facility R where a plurality of heating elements 2 are arranged, to execute the steps of acquiring measurement values ​​including temperatures measured at predetermined positions in the facility R and setting conditions of the air conditioning equipment 3 including the set temperature of the air conditioner 30; predicting future heat loads corresponding to each of a plurality of candidate setting conditions using a model M in which the measurement values ​​and the setting conditions are explanatory variables and the heat load due to the heat generation of the heating elements 2 is the target variable; and selecting, from the candidate setting conditions, a candidate setting condition that has obtained a predicted value that satisfies the environmental conditions that specify the reference value or reference range of the heat load, and controlling the air conditioning equipment 3 based on the selected candidate setting conditions.

[0083] According to the above aspect, the air conditioning equipment can be operated under optimal setting conditions in accordance with the heat load that changes from moment to moment.

[0084] REFERENCE SIGNS LIST 1 Air conditioning system 2 Heat generating element 3 Humidifier 10 Control device 11 Processor 111 Acquisition unit 112 Prediction unit 113 Control unit 114 Learning unit 12 Memory 13 Storage 14 Communication interface 20 Fan 21 Intake temperature sensor 22 Exhaust temperature sensor 30 Air conditioner 31 Intake temperature sensor 32 Exhaust temperature sensor 35 Humidifier 51 Temperature sensor 52 Humidity sensor CA Cold aisle HA Hot aisle M Model

Claims

1. A control device for an air conditioning system including an air conditioner installed in a facility where multiple heat generating elements are arranged, comprising: an acquisition unit that acquires measurement values ​​including temperature measured at a predetermined position in the facility and setting conditions of the air conditioning system including the set temperature of the air conditioner; a prediction unit that predicts the future heat load corresponding to each of multiple setting condition candidates using a model in which the measurement values ​​and the setting conditions are explanatory variables and the heat load due to the heat generated by the heat generating elements is a target variable; and a control unit that selects, from the setting condition candidates, a setting condition candidate that has obtained a predicted value that satisfies an environmental condition that specifies a reference value or reference range for the heat load, and controls the air conditioning system based on the selected candidate setting conditions.

2. The air conditioning equipment control device of claim 1, further comprising a learning unit that identifies an actual value of the thermal load at the evaluation time from the measurement value and the set conditions acquired at the evaluation time, and learns and updates the model based on the error between the identified actual value of the thermal load and a predicted value of the thermal load at the evaluation time predicted in the past.

3. The control device for an air conditioning system as described in claim 1, wherein the temperature measured at a specified location in the facility is at least one of the room temperature of the facility, the intake temperature of the air conditioner, the exhaust temperature of the air conditioner, the intake temperature of the heating element, the exhaust temperature of the heating element, and the temperature of the heating element.

4. The air conditioning equipment control device according to claim 1, wherein the set conditions further include at least one of the number of operating air conditioners and the air volume of the air conditioners.

5. The control device for an air conditioning system according to claim 1, wherein the air conditioning system includes a fan that the heat generating element has, and the setting conditions further include an air volume or a rotation speed of the fan.

6. The control device for an air conditioning equipment as described in claim 1, wherein the thermal load includes at least one of the room temperature of the facility, the intake temperature of the air conditioner, the exhaust temperature of the air conditioner, the intake temperature of the heating element, the exhaust temperature of the heating element, and the temperature of the heating element, and the power usage efficiency of the facility, and the control unit selects the setting condition candidate where the predicted value of the thermal load other than the power usage efficiency satisfies the environmental condition and the predicted value of the power usage efficiency is minimized.

7. The control device for an air conditioning system according to any one of claims 1 to 6, wherein the air conditioning system further includes a plurality of humidifiers, the measurement values ​​and the heat load further include humidity of the facility, and the setting conditions further include the number of humidifiers in operation.

8. A method for controlling an air conditioning system including an air conditioner installed in a facility in which a plurality of heat generating elements are arranged, comprising the steps of: acquiring measurement values ​​including a temperature measured at a predetermined position in the facility and setting conditions of the air conditioning system including a set temperature of the air conditioner; predicting future heat loads corresponding to each of a plurality of candidate setting conditions using a model in which the measurement values ​​and the setting conditions are explanatory variables and the heat load due to the heat generated by the heat generating elements is a target variable; and selecting, from the candidate setting conditions, a candidate setting condition that has obtained a predicted value that satisfies an environmental condition that specifies a reference value or reference range for the heat load, and controlling the air conditioning system based on the selected candidate setting conditions.

9. A program that causes a control device of an air conditioning equipment including an air conditioner installed in a facility where multiple heat generating elements are arranged to execute the steps of: acquiring measurement values ​​including temperature measured at a predetermined position in the facility and setting conditions of the air conditioning equipment including the set temperature of the air conditioner; predicting future heat loads corresponding to each of multiple setting condition candidates using a model in which the measurement values ​​and the setting conditions are explanatory variables and the heat load due to the heat generated by the heat generating elements is a target variable; and selecting, from the setting condition candidates, a setting condition candidate that has obtained a predicted value that satisfies the environmental conditions that specify the reference value or reference range of the heat load, and controlling the air conditioning equipment based on the selected candidate setting conditions.

Images