Performance measurement method for air conditioner, performance measurement device for air conditioner, and control method for condition generator

Abstract

[Problem] To provide a performance measurement method for an air conditioner that improves the accuracy of performance measurement of a dynamically controlled air conditioner. [Solution] In this performance measurement method for an air conditioner 2: a future change in the air state of a virtual room, which is obtained by virtually reproducing an actual room, is found by emulating the air state of the virtual room using the air state of discharged air from the air conditioner 2; an air state delay in a test room relative to a configuration of a condition generator 4 when the future change in the air state of the virtual room is used for the configuration of the condition generator 4 is compensated for; the condition generator 4 is controlled using, as the configuration thereof, the future change in the air state of the virtual room that has undergone the delay compensation; and the performance of the air conditioner 2 is measured in the test room, which is set to a measurement condition generated by the condition generator 4.

Description

Air conditioner performance measurement method, air conditioner performance measurement device, and condition generator control method

[0001] The present invention relates to a method for measuring the performance of an air conditioner, an apparatus for measuring the performance of an air conditioner, and a method for controlling a condition generator.

[0002] A simple method for measuring the performance of an air conditioner is to operate the air conditioner under stable conditions by fixing the compressor rotation speed and not controlling the capacity of the air conditioner. However, when measuring performance under such a constant operating condition, i.e., a static operating condition, performance is not measured under actual operating conditions, and there is a risk of a discrepancy between the measurement results and performance in the actual usage environment.

[0003] Therefore, performance measurement while dynamically changing the operating state of an air conditioner has been considered. Performance measurement under such dynamic operating conditions is sometimes referred to as load-based performance measurement. According to the technology disclosed in Patent Document 1, a predetermined test environment such as a test room is prepared to measure the performance of an air conditioner under dynamic operating conditions. According to this technology, electrical equipment and a dummy, etc., placed in the test room are controlled to measure the performance of the air conditioner under dynamic operating conditions while taking into account the actual usage environment, such as heat generation from the electrical equipment and the ambient heat in the room.

[0004] Japanese Patent Application Laid-Open No. 2007-333557

[0005] In the technology disclosed in Patent Document 1, when performance measurements are carried out at different locations, it is necessary to create the same test room for each location in order to make the implementation environment the same. However, it is difficult to reproduce a completely identical test environment, including the external environment, etc., and there is a risk that the results of the performance measurements will vary.

[0006] Therefore, a method has been considered in which a condition generator consisting of an air conditioner different from the air conditioner being measured is installed in the test room. According to this method, a virtual test room (virtual room) is emulated based on the state of the air blown out from the air conditioner being measured, and the performance of the air conditioner being measured is measured while controlling the condition generator so that the air condition in the actual test room matches the air condition in the virtual room obtained by the emulation. This method can eliminate environmental factors such as the thermal capacity of the test room, enabling performance measurements of air conditioners under dynamic operating conditions while realizing a highly reproducible measurement environment.

[0007] However, even with this emulation method, delays can occur due to factors such as the control of the condition generator and the heat capacity of the air in the test room, which can cause the change (response) in the actual test room state to be delayed compared to the state of the virtual room obtained by the emulation. As a result, there is a problem in that these delays can affect the results of performance measurements of air conditioners under dynamic operating conditions.

[0008] The present invention has been made in consideration of the above-mentioned problems, and aims to provide an air conditioner performance measurement method, an air conditioner performance measurement device, and a condition generator control method that improve the accuracy of air conditioner performance measurement under dynamic operating conditions.

[0009] The air conditioner performance measurement method of the present invention measures the performance of an air conditioner that is dynamically operated in a test room equipped with a condition generator that generates measurement conditions. This air conditioner performance measurement method uses the air conditions of air blown out from the air conditioner to emulate the air conditions of a virtual room that virtually reproduces an actual room, thereby determining future changes in the air conditions of the virtual room, compensates for a delay in the air conditions of the test room relative to the setting of the condition generator when the future changes in the air conditions of the virtual room are used to set the condition generator, controls the condition generator using the delay-compensated future changes in the air conditions of the virtual room as the setting, and measures the performance of the air conditioner in the test room where the measurement conditions generated by the condition generator are met.

[0010] The air conditioner performance measuring device of the present invention measures the performance of an air conditioner that is dynamically operated in a test room equipped with a condition generator that generates measurement conditions. This air conditioner performance measuring device has an emulation unit that uses the air conditions of air blown out from the air conditioner to emulate the air conditions of a virtual room that virtually reproduces an actual room, thereby determining future changes in the air conditions of the virtual room, a delay compensation unit that compensates for a delay in the air conditions of the test room relative to the setting of the condition generator when the future changes in the air conditions of the virtual room are used to set the condition generator, and controls the condition generator using the delay-compensated future changes in the air conditions of the virtual room as the setting, and an air conditioning performance measurement unit that measures the performance of the air conditioner in a test room that meets the measurement conditions generated by the condition generator.

[0011] The condition generator control method of the present invention controls a condition generator that is installed in a test room and generates conditions for measuring the performance of a dynamically operated air conditioner. This condition generator control method uses the air condition of air blown out from the air conditioner to emulate the air condition of a virtual room that virtually reproduces an actual room, thereby determining future changes in the air condition of the virtual room, compensates for a delay in the air condition of the test room relative to the setting of the condition generator when the future changes in the air condition of the virtual room are used to set the condition generator, and controls the condition generator using the delay-compensated future changes in the air condition of the virtual room as a set value.

[0012] According to the air conditioner performance measurement method, air conditioner performance measurement device, and condition generator control method of the present invention, the delay in the change in the air condition of the test room relative to the set value input to the condition generator is compensated for in the air condition of the virtual room obtained by emulation. By performing such delay compensation, when measuring the performance of a dynamically controlled air conditioner, the difference between the air condition of the test room and the air condition of the virtual room becomes smaller, thereby improving the accuracy of performance measurement.

[0013] FIG. 1 is a schematic configuration diagram of a test room in which performance measurements of air conditioners common to each embodiment are performed. FIG. 2 is an explanatory diagram of the operation of a virtual room calculation unit of the first embodiment. FIG. 3 is an explanatory diagram of the operation of a virtual room calculation unit of a comparative example. FIG. 4 is a graph showing the performance of a delay model generated in a first example. FIG. 5 is a graph showing the performance of a compensation model generated. FIG. 6 is a graph showing the performance of a compensation model generated in a second example. FIG. 7 is a graph showing measurement results when the generated delay model is actually operated in a test room. FIG. 8 is an explanatory diagram of the operation of the virtual room calculation unit of the second embodiment.

[0014] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below with reference to the accompanying drawings. In the following description, the same components are designated by the same reference numerals, and redundant description will be omitted.

[0015] (First Embodiment) Fig. 1 is a schematic diagram of a test chamber 1. The illustrated test chamber 1 has a configuration common to all embodiments. According to this diagram, the test chamber 1 is provided with an air conditioner 2 to be measured, a measurement chamber 3 that collects air blown out from the air conditioner 2, and a condition generator 4 that takes in the air blown out from the measurement chamber 3 and generates conditions for performance measurement by making the air conditions (temperature and humidity) in the test chamber 1 into predetermined air conditions. As an example, the condition generator 4 is an air conditioner different from the air conditioner 2 and is capable of controlling temperature and humidity. The air conditioner 2 operates integrally with an outdoor unit 5, and the condition generator 4 operates integrally with an outdoor unit 6.

[0016] The air blown out from the air conditioner 2 is blown into the measurement chamber 3. Furthermore, part of the air blown out from the measurement chamber 3 is blown into the condition generator 4, and the rest is blown into the test room 1. In another embodiment, all of the air blown out from the measurement chamber 3 may be blown into the condition generator 4. The temperature and humidity of the air blown into the condition generator 4 are changed by the condition generator 4.

[0017] Two control units, an air conditioning performance measurement unit 7 that measures the performance of the air conditioner 2, and a virtual room calculation unit 8 that determines the target value of the condition generator 4, are integrated into the performance measurement device 9. The performance measurement device 9 is composed of a calculator (computer) that includes a central processing unit (CPU), a memory unit, etc. The performance measurement device 9 executes specific programs stored in the memory unit to perform processing to realize the functions of the air conditioning performance measurement unit 7 and the virtual room calculation unit 8. The air conditioning performance measurement unit 7 and the virtual room calculation unit 8 may be configured as a single computer, or may be configured as multiple computers.

[0018] The air conditioning performance measurement unit 7 is configured to calculate the air conditioning performance of the air conditioner 2 from the measured air condition in the test room 1 and the power consumption of the air conditioner 2. The air conditioning performance measurement unit 7 can perform dynamic performance measurement of the air conditioner 2 because the air conditioner 2 is dynamically controlled according to the measurement conditions.

[0019] The virtual room calculation unit 8 predicts the air conditions (temperature, humidity) of a virtual room (hereinafter referred to as the virtual room) through emulation using the measured air conditions of the air blown out from the air conditioner 2, and controls the condition generator 4 so that the air conditions of the test room 1 match the air conditions of the virtual room. In this way, by using the condition generator 4 to match the air conditions of the test room 1 with the air conditions of the virtual room, it is possible to suppress variations in the measurement results of air conditioning performance due to differences in the configurations of multiple test rooms 1 set up in different locations.

[0020] The entire test chamber 1 is covered with a thermal insulation material 10. The test chamber 1 has a two-layer structure in which an inner ceiling 12 is provided inside an outer wall 11 of the ceiling, and is configured so that air blown from the condition generator 4 can be taken in between the outer wall 11 and the inner ceiling 12. A plurality of holes are provided in the inner ceiling 12, and the air blown from the condition generator 4 is diffused throughout the interior of the test chamber 1 through these holes. Note that, although the inner ceiling 12 is provided in the test chamber 1 in this embodiment, the inner ceiling 12 may not be provided, and the air blown from the condition generator 4 may be blown directly or indirectly into the test chamber 1.

[0021] The air conditioner 2 is connected via piping or the like to an outdoor unit 5 provided outside the test chamber 1. The outdoor unit 5 is equipped with a compressor or the like, and the air conditioner 2 and the outdoor unit 5 operate together to blow out air at a predetermined temperature, humidity, and air volume from the air conditioner 2.

[0022] Near the intake port of the air conditioner 2, the room temperature T ｒ The wet-bulb temperature measured by the wet-bulb thermometer 22 and the indoor temperature T measured by the thermometer 21 are measured. ｒ Using this, indoor humidity x ｒ In addition, in the vicinity of the air outlet of the air conditioner 2, the outlet temperature T ｓ A thermometer 23 for measuring the humidity x of the air blown out is provided, and a hygrometer 24 for directly measuring the humidity is provided. Since there is a temperature and humidity distribution at the air outlet of the air conditioner 2, instead of a wet-bulb thermometer for measuring humidity by comparing it with the temperature measured by the thermometer, a hygrometer 24 is provided to directly measure the humidity x of the air blown out. ｓ is preferably measured.

[0023] The air conditioner 2 is controlled integrally with the outdoor unit 5, and the power consumption P used for air conditioning the test room 1 is calculated using a power meter 25 provided in the air conditioner 2 and a power meter 53 provided in the outdoor unit 5. A thermometer 51 and a wet-bulb thermometer 52 are provided near the outdoor unit 5, and they respectively measure the outdoor temperature T ｏ and outdoor humidity x ｏ Measure the outdoor temperature T ｏ and outdoor humidity x ｏ is not used in the first embodiment, but is used in the second embodiment. Note that a configuration is also possible in which a power meter is provided in only one of the air conditioner 2 and the outdoor unit 5.

[0024] A thermometer 31 and a wet-bulb thermometer 32 are provided inside the measurement chamber 3. An airflow meter 33 is provided at the outlet of the measurement chamber 3. The airflow meter 33 calculates the differential pressure between two pressure gauges provided along the air flow, and calculates the blown air volume G from the air conditioner 2 based on the differential pressure. ｓ Calculate.

[0025] Here, since there is a temperature and humidity distribution in the vicinity of the air outlet of the air conditioner 2, the temperature T ｓ and humidity x ｓ On the other hand, the temperature T obtained by the thermometer 31 and the wet-bulb thermometer 32 may have an error. ｓ ' and humidity x ｓ ' is the outlet temperature T measured by the thermometer 23 and the hygrometer 24 provided near the outlet of the air conditioner 2. ｓ and blowout humidity x ｓ In comparison, the measurement chamber 3 is located at a distance from the air outlet of the air conditioner 2, and the measurement chamber 3 itself has a heat capacity, so there may be an error from the actual state of the air blown out from the air conditioner 2. However, because the air blown out is in a stirred state within the measurement chamber 3, the state of the air blown out can be measured more stably.

[0026] Therefore, the temperature T obtained by the thermometer 31 and the wet-bulb thermometer 32 in the steady state is ｓ ' and humidity x ｓ ', the temperature T blown out from the air conditioner 2 measured by the thermometer 23 and the hygrometer 24 ｓ and blowout humidity x ｓ By calibrating the outlet temperature T ｓ and blowout humidity x ｓ Furthermore, when the measurement delay due to the distance between the air outlet of the air conditioner 2 and the measurement chamber 3 is considered to be relatively small, the temperature T blown out from the air conditioner 2 can be measured by using the thermometer 31 and the wet-bulb thermometer 32 to measure the air condition in the measurement chamber 3 without using the thermometer 23 and the hygrometer 24. ｓ and blowout humidity x ｓ You may ask for.

[0027] The air conditioning performance measurement unit 7 measures the indoor temperature T ｒ and indoor humidity x ｒ , the blowout temperature T obtained using the thermometer 23, the hygrometer 24, and the air volume meter 33 ｓ , blowing humidity x ｓ (The temperature T obtained by the thermometer 31 and the wet-bulb thermometer 32 ｓ ', humidity xｓ '), and the blown air volume G ｓ , and the power consumption P measured by the power meters 25 and 53 are input.

[0028] The air conditioning performance measurement unit 7 measures the indoor temperature T ｒ and indoor humidity x ｒ The enthalpy of the indoor air is calculated from the ｓ and blowout humidity x ｓ The air conditioning performance measurement unit 7 calculates the enthalpy of the air blown out from the air conditioner 2 using the above and calculates the difference between the enthalpies. ｓ Finally, the air conditioning performance measurement unit 7 calculates the air conditioning performance of the air conditioner 2 and the outdoor unit 5 by dividing the obtained air conditioning capacity by the power consumption P.

[0029] The virtual room calculation unit 8 calculates the blowout temperature T ｓ , blowing humidity x ｓ , and blown air volume G ｓ When the inputs of the above are received, the virtual room calculation unit 8 emulates the air condition of the virtual room using these inputs. Then, the virtual room calculation unit 8 determines a set value (set temperature T ｓｅｔ , set humidity x ｓｅｔ ) to the condition generator 4. In this way, the air conditioning performance of the air conditioner 2 can be measured in the test room 1 whose air condition matches that of the virtual room.

[0030] The condition generator 4 is configured by an air conditioner different from the air conditioner 2 to be measured, and operates in conjunction with an outdoor unit 6 connected via piping, etc. The condition generator 4 takes in part or all of the air blown out from the measurement chamber 3, and calculates the set temperature T ｓｅｔ , and set humidity x ｓｅｔ It operates according to the

[0031] Furthermore, thermometers, hygrometers, and wet-bulb thermometers other than the thermometers 21, 23, and 51, the hygrometer 24, and the wet-bulb thermometers 22 and 52 may be provided inside and outside the test chamber 1. By using the measurement results of a plurality of thermometers, hygrometers, and wet-bulb thermometers, the indoor temperature T ｒ and indoor humidity x ｒ , blowing temperature T ｓ and blowout humidity x ｓ , and the outdoor temperature T ｏ and outdoor humidity x ｏ can be obtained.

[0032] 2 is an explanatory diagram of the operation of the virtual room calculation unit 8 of this embodiment. In this diagram, the control flow is indicated by thin arrows, and the air flow is indicated by thick arrows (hatched with dots).

[0033] As indicated by the thick arrow, the air blown out from the air conditioner 2 passes through the measurement chamber 3, is changed to predetermined conditions (temperature, humidity) by the condition generator 4, and then is returned to the test room 1 and sucked into the air conditioner 2.

[0034] On the other hand, as indicated by the thin arrows, the air outlet temperature T ｓ , blowing humidity x ｓ and blown air volume G ｓ are input to the virtual room calculation unit 8. The virtual room calculation unit 8 has an emulation unit 81 and a delay compensation unit 82. Note that, in the present embodiment, an example has been described in which the emulation unit 81 and the delay compensation unit 82 are integrated to form the virtual room calculation unit 8, but the emulation unit 81 and the delay compensation unit 82 may be separate computers or software.

[0035] The emulation unit 81 emulates the measured blown air temperature T ｓ , blowing humidity x ｓ and blown air volume G ｓ and emulates the air condition in the virtual room to calculate the virtual room temperature T ｅｍｕ and virtual room humidity x ｅｍｕThis emulation reproduces the amount of heat entering and leaving the virtual room, specifically the incoming sunlight, the ventilation with the outside (including drafts), and the heat generated by the people and electrical appliances inside.

[0036] More specifically, the emulation unit 81 calculates the state of the air blown out from the air conditioner 2, that is, the blown out temperature T ｓ , blowing humidity x ｓ and blown air volume G ｓ Using various parameters, a differential equation is solved taking into account the heat flow in and out of the virtual room, and the time-dependent change in the air condition (temperature, humidity) of the virtual room is predicted. These future predicted values ​​of the air condition of the virtual room are corrected by a delay compensation unit 82 at the subsequent stage and then used to control the condition generator 4. These future predicted values ​​of the air condition of the virtual room indicate the air condition that changes over time, and are used to predict the virtual room temperature T ｅｍｕ and virtual room humidity x ｅｍｕ It is shown as follows.

[0037] The delay compensation unit 82 calculates the virtual room temperature T ｅｍｕ and virtual room humidity x ｅｍｕ In addition, similarly to the emulation unit 81, the blown-out temperature T ｓ , blowing humidity x ｓ and blown air volume G ｓ Accept.

[0038] Here, the setting input to the condition generator 4 (set temperature T ｓｅｔ , set humidity x ｓｅｔ ) with respect to the air condition in the test room 1 (room temperature T ｒ , indoor humidity x ｒ ) changes (responses) are delayed.

[0039] Therefore, the delay compensation unit 82 calculates the input blowout temperature T ｓ , blowing humidity x ｓ and blown air volume G ｓ The virtual room temperature T output from the emulation unit 81 is calculated by feedforward control using ｅｍｕ and virtual room humidity x ｅｍｕThe delay compensation unit 82 performs a correction to compensate for the delay in the set temperature T ｓｅｔ and set humidity x ｓｅｔ is output to the condition generator 4, which is controlled accordingly.

[0040] Furthermore, the condition generator 4 controls the air condition in the test chamber 1, and there is uncertainty in the control, which may result in a control error in the air condition. Therefore, the delay compensator 82 may be configured to compensate for such an error.

[0041] The condition generator 4 has a temperature adjustment unit 41 and a humidity adjustment unit 42, and the set temperature T ｓｅｔ and set humidity x ｓｅｔ By operating in this manner, the condition generator 4 operates based on the room temperature T ｒ and indoor humidity x ｒ , the virtual room temperature T ｅｍｕ and virtual room humidity x ｅｍｕ The temperature adjusting unit 41 and the humidity adjusting unit 42 are control units of the condition generator 4, and may be integrated into a condition generator control unit 43.

[0042] 3 is an explanatory diagram of the operation of the virtual room calculation unit 8 of the comparative example. Compared to the virtual room calculation unit 8 of the first embodiment shown in FIG. 2, the configuration shown in this figure omits the delay compensation unit 82 in the virtual room calculation unit 8.

[0043] Since the delay compensation unit 82 is omitted, the virtual room state (virtual room temperature T ｅｍｕ , virtual room humidity x ｅｍｕ ) is used as a set value to control the condition generator 4. In such a control, delays occur due to the mechanical control of the condition generator 4 and the heat capacity of the air in the test chamber 1, so the state in the test chamber 1 (room temperature T ｒ , indoor humidity x ｒ ) will be delayed relative to the set value.

[0044] Therefore, in this embodiment, a delay compensation unit 82 is provided after the emulation unit 81 as shown in FIG. 2, so that the delay-compensated set temperature T ｓｅｔ and set humidity x ｓｅｔ is used to control the condition generator 4. As a result, the air condition of the test room 1 (room temperature T ｒ , indoor humidity x ｒ ) is calculated by the emulation unit 81 based on the virtual room air condition (virtual room temperature T ｅｍｕ , virtual room humidity x ｅｍｕ ) can be approached.

[0045] The delay compensation unit 82 is configured, for example, in the following manner: First, a delay model is generated that indicates the delay of the air state in the test room 1 relative to the set value of the condition generator 4, and then a compensation model that compensates for that delay is generated. The delay compensation unit 82 is then configured using the compensation model generated in this manner.

[0046] In detail, first, a delay model is generated that models the delay that occurs when the delay compensation unit 82 is not present, as in the comparative example of FIG. 3. For example, the setting of the condition generator 4 (set temperature T ｓｅｔ , set humidity x ｓｅｔ ) historical data, virtual room air condition (virtual room temperature T ｅｍｕ , virtual room humidity x ｅｍｕ ) past data, air condition of test room 1 (room temperature T ｒ , indoor humidity x ｒ ) and the air condition of the air blown out from the air conditioner 2 (blowout temperature T ｓ , blowing humidity x ｓ , air volume G ｓ The delay model is constructed by machine learning using past data from the

[0047] Next, a compensation model that compensates for the delay is generated using the generated delay model. For example, if one of the inputs to the delay model is changed, the output value changes in response to the change in the input. One of the output values ​​that changes is the air condition of test room 1. Therefore, a change in the input is determined that minimizes the deviation of the air condition of test room 1 after the change from the air condition of the virtual room over time. For example, by using a genetic algorithm, such input changes can be repeated to determine the optimal change. Finally, a compensation model that compensates for the determined change can be generated.

[0048] In this way, a delay model is generated, and then a compensation model is generated, and the generated delay model can be used to configure the delay compensation unit 82. Note that in the above description, machine learning is used to generate the delay model and the compensation model, but this is not limited to machine learning, and a mathematical model that describes the phenomenon in the condition generator 4 may also be used.

[0049] Below, two specific examples of simulations for constructing a delay compensation unit 82 using machine learning will be described. In the first example, machine learning is performed using past measurement data, and the performance of the compensation model is verified by simulation, as shown in FIGS. 4 and 5. In the second example, machine learning is performed using an explicit parameter that represents the error between the air conditions in the virtual room and the indoor conditions of the test room 1, and the performance of the compensation model is verified by actual measurements, as shown in FIGS. 6 and 7. In both examples, the simulation was performed only for temperature, not humidity.

[0050] In the first example, the virtual room temperature T ｅｍｕ Indoor temperature T ｒ Then, a delay model was generated that shows the delay of the set temperature T ｓｅｔ The set temperature T ｓｅｔ The indoor temperature T obtained according to ｒ ' and the virtual room temperature T ｅｍｕ The difference between (T ｒ '-T ｓｅｔ The set temperature T ′) is minimized (smaller than or equal to a predetermined value).ｓｅｔ As a result, the virtual room temperature T ｅｍｕ is input and the set temperature T ｓｅｔ A delay model with output ' is generated.

[0051] 4 is a graph showing the validity of the delay model obtained by the machine learning of the first example. In this graph, the vertical axis shows the temperature normalized to a number between -1 and +1, and the horizontal axis shows the time in seconds. The dashed line shows the virtual room temperature T obtained by the emulation unit 81. ｅｍｕ The solid line indicates the room temperature T ｒ The thin dotted line shows the virtual room temperature T ｅｍｕ The indoor temperature T obtained by compensating for the delay ｒ In this figure, the area enclosed by the two-dot chain line rectangle in the upper part is shown enlarged in the lower part.

[0052] As shown in the figure, the indoor temperature T ｒ ' is the measured room temperature T ｒ In this way, since the two match most of the time, or the difference is equal to or less than a predetermined value, it can be understood that the delay model has been correctly constructed.

[0053] 5 is a graph showing the performance of a compensation model generated through machine learning. In this figure, the area enclosed by a two-dot chain rectangle in the upper part is shown enlarged in the lower part.

[0054] In this figure, as in FIG. 4, the virtual room temperature T ｅｍｕ For comparison with FIG. 4, the indoor temperature T ｒ Furthermore, the thin solid line indicates the virtual room temperature T ｅｍｕ The setting for the condition generator 4 (setting temperature T ｓｅｔ The thin dotted line indicates the set temperature T ｓｅｔ The temperature of the test room 1 (room temperature T ｒThe dashed line (virtual room temperature T ｅｍｕ ) and thin dotted line (indoor temperature T ｒ There are many overlapping parts with the dashed line, and the small "dots" that make up the thin dotted line that overlap with the dashed line are shown in white.

[0055] As shown in this figure, the virtual room temperature T ｅｍｕ Using the compensation model, the set temperature T ｓｅｔ ' is generated. Then, this set temperature T ｓｅｔ When the condition generator 4 is controlled by the delay model, the simulation result is obtained. ｒ ' is required.

[0056] Virtual room temperature T ｅｍｕ (shown by the dashed line) is the indoor temperature T obtained using the delay model. ｒ Since the delay compensation unit 82 is configured using this compensation model, the virtual room temperature T ｅｍｕ Indoor temperature T ｒ Since the delay is suppressed, the accuracy of the performance measurement of the air conditioner 2 using the virtual room can be improved.

[0057] In the second example, a delay model is generated by machine learning that explicitly uses, as a parameter, the error between the air condition of the virtual room and the indoor conditions of the test room 1. In other words, the compensation model generated based on this delay model can directly compensate for not only the delay but also the error.

[0058] FIG. 6 is a graph showing the performance of the compensation model generated through machine learning in the second example. In this figure, the temperature on the vertical axis is not normalized, as compared to the examples in FIGS. 4 and 5. The circles represent the set values ​​(set temperature T ｓｅｔ ) The solid line indicates this set value (set temperature T ｓｅｔ ) based on the indoor temperature T obtained using a delay model ｒ The dashed line indicates the virtual room temperature Tｅｍｕ Shows.

[0059] As shown in the figure, the indoor temperature T ｒ and the virtual room temperature T shown by the dashed line ｅｍｕ In this way, since the two are consistent for most of the time, or the difference is equal to or less than a predetermined value, it can be understood that the delay model has been correctly constructed.

[0060] FIG. 7 is a graph showing the measured temperature of the test chamber 1 when the delay compensation unit 82 generated using the compensation model generated through machine learning is actually used. The circles represent the set value (set temperature T ｓｅｔ The dashed line indicates the virtual room temperature T ｅｍｕ The triangle indicates this setting (setting temperature T ｓｅｔ ) The room temperature T measured in the test room 1 controlled based on ｒ Shows.

[0061] In this second example, the measured indoor temperature T ｒ and the virtual room temperature T shown by the dashed line ｅｍｕ In this way, since the two are the same for most of the time, or the difference is equal to or less than a predetermined value, it can be understood that the delay has been compensated for by the delay compensating unit 82.

[0062] According to the performance measurement method of the air conditioner 2 of the first embodiment, the air condition of the virtual room (virtual room temperature T ｅｍｕ , virtual room humidity x ｅｍｕ ), the delay compensation unit 82 adjusts the set value (virtual room temperature T ｅｍｕ , virtual room humidity x ｅｍｕ ) for the air condition of the test room 1 (room temperature T ｒ , indoor humidity x ｒ ) is compensated for. By performing such delay compensation, the air condition (room temperature T ｒ , indoor humidity x ｒ ) and the virtual room air condition (virtual room temperature T ｅｍｕ , virtual room humidity x ｅｍｕ) becomes smaller, the accuracy of the performance measurement of the air conditioner 2 can be improved.

[0063] According to the performance measurement method of the air conditioner 2 of the first embodiment, the delay compensation unit 82 measures the air condition of the air blown out from the air conditioner 2 (blowout temperature T ｓ , blowing humidity x ｓ , air volume G ｓ By performing such feedforward control, it is possible to improve the accuracy of the performance measurement of the air conditioner 2.

[0064] According to the performance measurement method of the air conditioner 2 of the first embodiment, the air condition of the virtual room reproduced by the condition generator 4 is the virtual room temperature T ｅｍｕ and virtual room humidity x ｅｍｕ In this way, by reproducing both the temperature and humidity of the virtual room in the test room 1, the reproducibility of the measurement conditions of the air conditioner 2 is increased, and the accuracy of the performance measurement of the air conditioner 2 can be improved.

[0065] According to the performance measurement method of the air conditioner 2 of the first embodiment, the virtual room state (virtual room temperature T ｅｍｕ , virtual room humidity x ｅｍｕ ) for the air condition of the test room 1 (room temperature T ｒ , indoor humidity x ｒ Then, a delay model is generated that indicates the delay in the change (response) of the virtual room temperature T ｅｍｕ ', virtual room humidity x ｅｍｕ ') is obtained, and a compensation model to be used for delay compensation is generated based on the change. In this way, after the delay model is generated, a compensation model is generated using the generated delay model, thereby simplifying the method of configuring the model.

[0066] According to the performance measurement method for the air conditioner 2 of the first embodiment, in compensating for the delay, the error of the air state of the test room 1 relative to the air state of the virtual room when the delay has been compensated for is further compensated for. In this way, by compensating for not only the delay but also the error, it is possible to improve the accuracy of the performance measurement of the air conditioner 2.

[0067] Second Embodiment In the first embodiment, the delay compensation unit 82 shown in FIG. 2 calculates the air condition (blowout temperature T ｓ , blowing humidity x ｓ , air volume G ｓ ), the virtual room air condition (virtual room temperature T ｅｍｕ , virtual room humidity x ｅｍｕ In the second embodiment, an example in which another parameter is used to improve the accuracy of delay compensation will be described.

[0068] 8 is an explanatory diagram of the operation of the virtual room calculation unit 8 of the second embodiment. According to this diagram, the delay compensation unit 82 of the virtual room calculation unit 8 receives the air condition of the air blown out from the air conditioner 2 (blowout temperature T ｓ , blowing humidity x ｓ , air volume G ｓ ), the air condition in the test room 1 (room temperature T ｒ , indoor humidity x ｒ ), and the air condition outside the test chamber 1 (outdoor temperature T ｏ , outdoor humidity x ｏ ) and the delay compensation unit 82 calculates the virtual room state (virtual room temperature T ｅｍｕ , virtual room humidity x ｅｍｕ ) is subjected to delay compensation, and the compensated value is used as the setting value of the condition generator 4 (setting temperature T ｓｅｔ , set humidity x ｓｅｔ ) is used as

[0069] The delay in the response of the air condition in the test room 1 to the set value input to the condition generator 4 is thought to be due to delays caused by the control of the condition generator 4 and the heat capacity of the air in the test room 1, as well as heat conduction / heat radiation from outside the test room 1 via the insulation material 10. In particular, since the outdoor unit 5 becomes hot, the outdoor temperature T ｏ and outdoor humidity x ｏ The feedforward using compensates for the delay due to heat conduction / heat radiation from outside the test chamber 1.

[0070] Furthermore, the delay compensation unit 82 determines the air condition in the test room 1 (room temperature T ｒ , indoor humidity x ｒ ) is received, the set value of the condition generator 4 (set temperature T ｓｅｔ , set humidity x ｓｅｔ ) with respect to the air condition in the test room 1 (room temperature T ｒ , indoor humidity x ｒ ) response delay / advance can be directly known.

[0071] Therefore, the air condition in the test room 1 (room temperature T ｒ , indoor humidity x ｒ ) is the set value of the condition generator 4 (set temperature T ｓｅｔ , set humidity x ｓｅｔ ), the compensation amount is further increased, and the air condition in the test room 1 (room temperature T ｒ , indoor humidity x ｒ ) is the set value of the condition generator 4 (set temperature T ｓｅｔ , set humidity x ｓｅｔ ) is ahead of the air condition in the test room 1 (room temperature T ｒ , indoor humidity x ｒ ) can be used for feedback control to optimally compensate for delays.

[0072] In this embodiment, the delay compensation unit 82 determines the air condition (blowout temperature T ｓ , blowing humidity x ｓ , air volume G ｓ ), the air condition in the test room 1 (room temperature T ｒ , indoor humidity x ｒ ), and the air condition outside the test room 1 (outdoor temperature T ｏ , outdoor humidity x ｏ Even in such cases, delay models and compensation models can be generated by machine learning by increasing the number of parameters handled during model generation.

[0073] According to the performance measurement method for the air conditioner 2 of the second embodiment, the delay compensation unit 82 further determines the air condition inside the test room 1 (room temperature T ｒ , indoor humidity xｒ The delay is compensated for by feedback control using the delay compensation function. By performing such feedback control, the amount of delay compensation can be appropriately adjusted, thereby further improving the accuracy of performance measurement of the air conditioner 2.

[0074] According to the performance measurement method for the air conditioner 2 of the second embodiment, the delay compensation unit 82 further determines the air condition outside the test room 1 (outdoor temperature T ｏ and outdoor humidity x ｏ ) is used to compensate for the delay. ｓ , blowing humidity x ｓ , air volume G ｓ ) is used, the amount of delay compensation can be appropriately adjusted, thereby further improving the accuracy of performance measurement of the air conditioner 2.

[0075] The present invention allows various embodiments and modifications without departing from the broad spirit and scope of the present invention. Furthermore, the above-described embodiments are intended to illustrate the present invention and do not limit the scope of the present invention. That is, the scope of the present invention is defined by the claims, not the embodiments. Various modifications made within the scope of the claims and within the scope of the meaning of the invention equivalent thereto are considered to be within the scope of the present invention.

[0076] REFERENCE SIGNS LIST 1 Test room 2 Air conditioner 3 Measurement chamber 4 Condition generator 5, 6 Outdoor unit 7 Air conditioning performance measurement unit 8 Virtual room calculation unit 9 Performance measurement device 21, 23, 31, 51 Thermometer 24 Hygrometer 22, 32, 52 Wet bulb thermometer 25, 53 Wattmeter 33 Air flow meter 81 Emulation unit 82 Delay compensation unit

Claims

1. A method for measuring the performance of an air conditioner that is dynamically operated in a test room equipped with a condition generator that generates measurement conditions, comprising the steps of: using the air condition of the air blown out from the air conditioner to emulate the air condition of a virtual room that virtually reproduces an actual room, thereby determining future changes in the air condition of the virtual room; compensating for a delay in the air condition of the test room relative to the setting of the condition generator when the future changes in the air condition of the virtual room are used to set the condition generator; controlling the condition generator using the future changes in the air condition of the virtual room, for which the delay has been compensated, as a setting; and measuring the performance of the air conditioner in the test room where the measurement conditions are generated by the condition generator.

2. The method for measuring performance of an air conditioner according to claim 1, wherein compensation for the delay is performed by feedforward control using the air condition of the air blown out from the air conditioner.

3. The method for measuring performance of an air conditioner according to claim 2, wherein said delay is further compensated for by feedback control using the air condition inside said test room.

4. The method for measuring performance of an air conditioner according to claim 2 or 3, wherein said delay is further compensated for by feedforward control using the air condition outside said test room.

5. The method for measuring performance of an air conditioner according to claim 1, wherein the air condition includes at least one of air temperature and humidity.

6. The method for measuring performance of an air conditioner according to claim 1, further comprising the step of compensating for an error in the air condition of the test room relative to the air condition of the virtual room for which the delay has been compensated.

7. The performance measurement method of claim 1, further comprising: generating a delay model indicative of the delay of the air conditions in the test room relative to the settings of the condition generator; varying the settings of the condition generator so that the delay becomes smaller; and generating a compensation model used to compensate for the delay based on the changes.

8. An air conditioner performance measuring device that measures the performance of an air conditioner that is dynamically operated in a test room equipped with a condition generator that generates measurement conditions, comprising: an emulation unit that determines future changes in the air condition of a virtual room that virtually reproduces an actual room by using the air condition of the air blown out from the air conditioner to emulate the air condition of the virtual room; a delay compensation unit that compensates for a delay in the air condition of the test room relative to the setting of the condition generator when the future changes in the air condition of the virtual room are used to set the condition generator, and controls the condition generator using the future changes in the air condition of the virtual room where the delay has been compensated for as the setting; and an air conditioner performance measuring unit that measures the performance of the air conditioner in the test room where the measurement conditions are generated by the condition generator.

9. A method for controlling a condition generator which is installed in a test room and generates conditions for measuring the performance of a dynamically operated air conditioner, comprising the steps of: emulating the air condition of a virtual room which virtually reproduces an actual room using the air condition of the air blown out from the air conditioner, determining future changes in the air condition of the virtual room; compensating for a delay in the air condition of the test room relative to the setting of the condition generator when the future changes in the air condition of the virtual room are used to set the condition generator; and controlling the condition generator using the future changes in the air condition of the virtual room, with the delay compensated for, as a set value.

Images