Estimation device

Abstract

To provide a technique for improving estimation accuracy of energy consumption amount in an air conditioner.SOLUTION: An estimation device 100 that estimates energy consumption amount of an air conditioner 900 includes: an acquisition section that acquires information related to a configuration of the air conditioner 900 used for estimation of the energy consumption amount and stored in a storage section and information on a building in which the air conditioner 900 is provided; a first estimation section 103 that estimates indoor environment information on the basis of a first learning model learned through machine learning for estimating the information on the indoor environment where the air conditioner 900 is provided; a second estimation section 104 that estimates the number of persons present in a room on the basis of a second learning model learned through machine learning for estimating the number of persons present in the room; and a third estimation section 105 that estimates new energy consumption amount on the basis of a third learning model learned through machine learning for estimating the energy consumption amount of the air conditioner 900.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an estimation device. [Background technology]

[0002] There is growing interest in energy conservation (hereinafter referred to as "energy conservation") in the construction industry. Air conditioning systems account for 40% of the energy consumption in buildings. Therefore, the impact of energy conservation effects related to air conditioning systems in large buildings such as buildings is significant, and there are also high expectations for the energy conservation effects of air conditioning systems.

[0003] In order to improve the energy-saving effect of air conditioners, the amount of energy consumed by the air conditioners may be estimated using the measured amount of energy consumed by the air conditioners (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-2776 Summary of the Invention [Problem to be solved by the invention]

[0005] When performing the above-described control, it is necessary to collect data for estimation over a long period of time, and there is also the problem that the estimation accuracy decreases when the set temperature is changed. The present invention has been made to solve the above-mentioned problems, and has an object to propose a technique for improving the accuracy of estimating the amount of energy consumed by an air conditioner. [Means for solving the problem]

[0006] In order to achieve the above object, the present invention provides the following means. An estimation device according to one embodiment of the present invention is an estimation device that estimates the energy consumption of an air conditioning unit, and is characterized by including: an acquisition unit that acquires input information stored in a memory unit used for estimating the energy consumption, the input information being information related to the configuration of the air conditioning unit and information related to the building in which the air conditioning unit is installed; a first estimation unit that estimates the indoor environment information by inputting the information related to the configuration of the air conditioning unit and information related to the building acquired by the acquisition unit into a trained first learning model that has undergone machine learning to estimate indoor environment information in which the air conditioning unit is installed; a second estimation unit that estimates the number of occupants in the room by inputting the information related to the configuration of the air conditioning unit and information related to the building acquired by the acquisition unit into a trained second learning model that has undergone machine learning to estimate the number of occupants in the room in which the air conditioning unit is installed; and a third estimation unit that estimates a new energy consumption amount by inputting the information related to the configuration of the air conditioning unit, information related to the building, the indoor environment information estimated by the first estimation unit, and the number of occupants estimated by the second estimation unit into a trained third learning model that has undergone machine learning to estimate the energy consumption of the air conditioning unit.

[0007] According to the estimation device of the first aspect of the present invention, the first estimation unit estimates indoor environment information, the second estimation unit estimates the number of occupants, and the third estimation unit can estimate the energy consumption of the air conditioner based on at least the estimated indoor environment information and the estimated number of occupants.

[0008] In the first aspect of the invention, the information about the building preferably includes at least information about the height of the building. In this way, the information about the building is information about the building that the air conditioning device is installed in. The information about the building preferably includes at least information about the building height, information about the height of each floor, and the amount of sunlight radiation on each floor.

[0009] In the first aspect of the invention, the building preferably has at least two or more floors, and the at least two or more floors have air conditioning devices with similar configurations. In this way, by having the same air conditioning equipment configuration for each floor, when estimating the energy consumption of an air conditioning equipment installed on the floor being estimated, it is possible to base it on information about air conditioning equipment installed on a floor other than the floor being estimated. [Effects of the Invention]

[0010] According to the estimation device of the present invention, control can be performed that makes it easy to improve the estimation accuracy of the amount of energy consumed by an air conditioner. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram illustrating a configuration of an estimation device according to a first embodiment of the present invention. [Figure 2] 10 is a flowchart illustrating an estimation process of the estimation device. [Figure 3] 10 is a flowchart illustrating the estimation process of a third estimation unit. DETAILED DESCRIPTION OF THE INVENTION

[0012] [First embodiment] An estimation device 100 according to a first embodiment of the present invention will be described with reference to Fig. 1 to Fig. 3. The estimation device 100 of this embodiment is a device that estimates the amount of energy consumed by an air conditioner 900.

[0013] The air conditioner 900 is configured to take in indoor air, cool and heat it (hereinafter also referred to as "cooling, etc."), and supply the cooled, etc. air into the room. In this embodiment, the air conditioner 900 will be described as supplying cooled air into the room of a building. In this embodiment, the air conditioner 900 will be described as being installed in a building.

[0014] The building has multiple floors, and it is preferable that at least two or more floors are equipped with air conditioners 900. It is preferable that the configuration of the air conditioners 900 on some floors be the same as the configuration of the air conditioners 900 on the remaining floors. Furthermore, it is preferable that the configuration of the air conditioners 900 on all floors be the same.

[0015] The configuration of the air conditioner 900 refers to the combination of the number of indoor units and the number of outdoor units. Note that in this embodiment, the configuration of the air conditioner 900 may also be the combination of the number of indoor units and outdoor units provided on one floor. Furthermore, the configuration of the air conditioner 900 may also include information on the locations of the indoor units and outdoor units. The outdoor unit is configured to exchange heat between a refrigerant and outside air or water. The outdoor unit is connected to the indoor unit through piping through which the refrigerant flows. The outdoor unit is also equipped with a compressor (not shown) that increases the pressure of the refrigerant and an outdoor heat exchanger (not shown) that cools the drawn-in air through heat exchange.

[0016] The indoor unit is configured to discharge cooled air into the room. The indoor unit is connected to the other indoor units and the outdoor unit via pipes through which a refrigerant flows. The indoor unit is also provided with an air conditioning fan (not shown) that discharges the air that has undergone heat exchange, and an indoor heat exchanger (not shown) that cools the discharged air through heat exchange.

[0017] The estimation device 100 of this embodiment is connected to the air conditioning device 900, the detection device 500 (described later), and the communication device 501 (described later) via known wired or wireless information communication means so as to be able to transmit information. The estimation device 100 is also connected to the air conditioning device 900, the detection device 500, and the communication device 501 via a known wireless communication network or a combination of a wireless communication network and a wired communication network so as to be able to transmit information.

[0018] The estimation device 100 estimates the amount of energy consumed by the air conditioning device 900. The estimation device 100 is an information processing device such as a server having a CPU (Central Processing Unit), ROM, RAM, an input / output interface, etc. As shown in Fig. 1 , a program stored in the storage device such as the ROM causes the CPU, ROM, RAM, and input / output interface to cooperate with each other to function as at least an acquisition unit 101, a storage unit 102, a first estimation unit 103, a second estimation unit 104, and a third estimation unit 105.

[0019] The acquisition unit 101 is connected to the air conditioning device 900, the detection device 500, and the communication device 501 so that information can be communicated with them. The acquisition unit 101 is configured to acquire information about the air conditioning device 900, detected information, communication information, and information about the building (hereinafter also referred to as various types of information) at predetermined time intervals. In this embodiment, the predetermined time interval is described as one hour. Note that the predetermined time may be a time interval other than the above.

[0020] The information relating to the air conditioner 900 is information acquired from the air conditioner 900. In this embodiment, the information relating to the air conditioner 900 preferably includes the configuration of the air conditioner 900, the operating state of the air conditioner 900, and the amount of energy consumed. The operating state of the air conditioner 900 preferably includes various control information such as the on / off state of the air conditioner 900, the operating settings, and the set temperature.

[0021] The detected information is information acquired from the detection device 500. In this embodiment, the detected information preferably includes information about the indoor environment, information about the outdoor air environment, and information about the people in the room. Each of these pieces of information will be described later.

[0022] The communication information is information acquired from the communication device 501. In this embodiment, the communication information preferably includes estimated information on the outdoor air environment corresponding to the next time interval and information on the building. The information on the outdoor air environment corresponding to the next time interval will be described later.

[0023] The information about the building is information about the building in which the air conditioner 900 is installed. The information about the building preferably includes at least information about the height of the building, information about the height of each floor, and the sunlight radiation temperature on each floor.

[0024] The solar radiation temperature is a value that indicates the temperature inside a room that changes due to the heat radiated from the sun. The storage unit 102 is an information storage medium configured to store various types of information. The various types of information stored preferably include information about the air conditioner 900, detected information, and communication information. The storage unit 102 may be a flash memory such as an SD memory card, or may be a recording medium in another format.

[0025] The first estimation unit 103 is configured to estimate information about the next indoor environment at predetermined time intervals. In this embodiment, the first estimation unit 103 is configured to input the following information to the first learning model.

[0026] The information input to the first learning model includes information about the air conditioning device 900, information about the indoor environment acquired last time, information about the outdoor environment acquired last time, and information about the building. In addition to the above information, information about the indoor environment acquired this time and information about the outdoor environment acquired this time may also be included.

[0027] The first learning model is a model that has been trained by machine learning. A known learning method can be used as the machine learning. The machine learning is preferably ensemble learning. Furthermore, it is more preferable that the machine learning is gradient boosting regression trees (XGBOOST, eXtreme Gradient Boosting).

[0028] The information relating to the indoor environment is information indicating the indoor environment. In this embodiment, the information relating to the indoor environment preferably includes the indoor temperature and the indoor humidity.

[0029] The information relating to the outdoor air environment is information indicating outdoor air environment information. In this embodiment, the information relating to the outdoor air environment preferably includes outdoor air temperature, outdoor air humidity, weather, probability of precipitation, cloud cover, wind speed, and amount of solar radiation.

[0030] Next, the relationship between the current time, the previous time, and the next time will be described. The estimation device 100 performs calculation processing at predetermined time intervals. In this embodiment, the predetermined time interval is preferably one hour ago.

[0031] The present time refers to the timing at which the estimation device 100 performs the calculation process. Hereinafter, the timing will also be referred to as the next timing. The previous time is a time when the estimation device 100 is not performing calculation processing, and is the time immediately before the current time.

[0032] The next time is a time when the estimation device 100 is not performing calculation processing, and is a time after the current first point. The second estimation unit 104 is configured to estimate the next number of occupants at predetermined time intervals. In this embodiment, the second estimation unit 104 is configured to input the following information to the second learning model.

[0033] The information input into the second learning model is information regarding the date corresponding to the next time, information regarding the outdoor air environment corresponding to the next time, information regarding the date obtained this time, information regarding the occupants obtained this time, information regarding the outdoor air environment obtained this time, information regarding the date obtained last time, information regarding the occupants obtained last time, information regarding the outdoor air environment obtained last time, and information regarding the building.

[0034] The information about the occupants includes the number of occupants and details about the occupants. The number of occupants is information indicating the number of people in the room. The details about the occupants preferably include information about the group to which the occupants belong, the amount of clothing worn by the occupants, and the amount of exercise performed by the occupants.

[0035] The second learning model is a model that has been trained by machine learning. A known learning method can be used as the machine learning. The machine learning is preferably ensemble learning. Furthermore, it is more preferable that the machine learning is gradient boosting regression trees (XGBOOST, eXtreme Gradient Boosting).

[0036] The third estimation unit 105 is configured to estimate the next energy consumption amount at predetermined time intervals. In this embodiment, the third estimation unit 105 is configured to input the following information to the third learning model.

[0037] The information input to the third learning model includes the currently acquired information about the air conditioner 900, the previously acquired information about the indoor environment, the previously acquired information about the outdoor environment, the previously acquired amount of energy consumption, the estimated information about the next indoor environment, the estimated number of people in the room, and information about the building. In addition to the above information, the currently acquired information about the indoor environment, the currently acquired information about the outdoor environment, and the currently acquired amount of energy consumption may also be included.

[0038] The third learning model is a model that has been trained by machine learning. A known learning method can be used as the machine learning. The machine learning is preferably a neural network.

[0039] The detection device 500 is a device communicably connected to the estimation device 100, and is configured to detect information related to the indoor environment, information related to the outdoor air environment, information related to the date, and information related to the occupants of the room. In this embodiment, it is preferable that one detection device 500 is provided in each room. Note that a plurality of detection devices 500 may be provided in one room, or no detection device 500 may be provided at all.

[0040] The communication device 501 is a device communicably connected to the estimation device 100 and configured to acquire information about the building and estimated information about the outdoor air environment corresponding to the next time interval. In this embodiment, the communication device 501 is a device connectable to a network, and includes a personal computer or the like.

[0041] The estimated information about the outdoor environment for the next time interval is preferably information about the next outdoor environment, and may be a weather forecast, which preferably includes estimates of outdoor temperature, outdoor humidity, weather, precipitation probability, cloud cover, wind speed, and solar radiation.

[0042] Next, the operation of the estimation device 100 configured as described above will be described. First, the air conditioner 900 will be described, second, the estimation device 100 will be described, and third, a learning method for the learning model will be described.

[0043] The mechanism by which the air conditioner 900 cools a room will now be described. The air conditioner 900 rotates the air conditioning fan to draw indoor air into the indoor unit. The drawn-in air is cooled in the indoor heat exchanger. Specifically, the temperature of the drawn-in air is lowered as heat is absorbed by the refrigerant circulating between the indoor unit and the outdoor unit. The refrigerant that absorbed the heat then releases the heat to the outside air in the outdoor heat exchanger. The refrigerant that released the heat then absorbs heat from the drawn-in air again in the heat exchanger. In other words, the drawn-in air is cooled by the refrigerant.

[0044] The cooled air is discharged into the room by the air conditioning fan. The air discharged into the room is warmed by heat radiated from the people and electronic devices in the room. The warmed air is then drawn into the air conditioner 900 again.

[0045] Next, the detection device 500 will be described. When the detection device 500 is started, it performs a process of detecting detected information. Specifically, it performs a process of detecting information related to the indoor environment, information related to the outdoor air environment, information related to the date, and information related to the occupants of the room. The detected information is communicated to the estimation device 100.

[0046] Next, the control of the estimation device 100 will be described with reference to Fig. 2. When control in the estimation device 100 is started, the acquisition unit 101 performs a process of acquiring various pieces of information from the air conditioning device 900, the detection device 500, and the communication device 501 at predetermined time intervals (S1). In this embodiment, the predetermined time interval is preferably one hour.

[0047] Once the various types of information are stored, the first estimation unit 103 performs a process of estimating at least one piece of information relating to the next indoor environment based on the input data 1 (S2). The estimated information relating to the next indoor environment is stored in the storage unit 102.

[0048] The input data 1 includes information about the air conditioner 900 that has been acquired this time, information about the indoor environment that has been acquired last time, information about the outdoor air environment that has been acquired last time, and information about the building.

[0049] Once the information about the next estimated indoor environment is stored, the second estimation unit 104 performs a process of estimating the next number of occupants based on input data 2 (S3). The next estimated number of occupants is stored in the memory unit 102.

[0050] Input data 2 includes information regarding the date of the next corresponding time, information regarding the outdoor air environment corresponding to the next corresponding time, information regarding the date acquired this time, information regarding the occupants acquired this time, information regarding the outdoor air environment acquired this time, information regarding the date acquired last time, information regarding the occupants acquired last time, information regarding the outdoor air environment acquired last time, and information regarding the building.

[0051] When the information about the estimated next occupants is stored, the third estimation unit 105 performs a process of estimating the next energy consumption amount for each piece of information about the estimated next indoor environment based on the input data 3, the information about the estimated next indoor environment, and the estimated number of occupants for the next time (S3). The information about the estimated next energy consumption amount is stored in the storage unit 102.

[0052] The input data 3 includes information about the air conditioner 900 acquired this time, information about the indoor environment acquired last time, information about the outdoor air environment acquired last time, and the amount of energy consumption acquired last time. The information about the estimated next energy consumption amount stored in S3 may be transmitted to an information terminal or the like of a person who manages the building.

[0053] Next, we will explain the machine learning of the first learning model, the second learning model, and the third learning model. In this embodiment, the present invention will be described as being applied to an example in which machine learning of a first learning model, a second learning model, and a third learning model is performed in an information processing device different from the estimation device 100. The first learning model, the second learning model, and the third learning model that have undergone machine learning are stored in the storage unit 102 before being controlled by the estimation device 100.

[0054] Furthermore, after control by the estimation device 100 is performed, a first learning model, a second learning model, and a third learning model that have been further subjected to machine learning may be stored in the storage unit 102. In this case, the previously stored first learning model, second learning model, and third learning model are replaced with the first learning model, second learning model, and third learning model that have been further subjected to machine learning.

[0055] Note that machine learning of the first learning model, the second learning model, and the third learning model may be performed in different information processing devices as described above, or may be performed in the estimation device 100. When machine learning is performed in the estimation device 100, a machine learning unit that performs machine learning is provided in the estimation device 100. Furthermore, machine learning for one of the first learning model, the second learning model, and the third learning model may be performed in a different information processing device, and machine learning for the other may be performed in the estimation device 100.

[0056] The specific machine learning in the first learning model, the second learning model, and the third learning model can use well-known supervised learning, and the specific content of the calculation processing in supervised learning is not limited.

[0057] Furthermore, the teacher data for the first learning model and the teacher data for the second learning model can be created using known methods, and there are no specific limitations on the creation method.

[0058] According to the estimation device 100 having the above configuration, the third estimation unit 105 can estimate the energy consumption of the air conditioner 900 based on at least the estimated indoor environment information and the estimated number of occupants, as shown in Fig. 3. Therefore, it is easier to estimate the energy consumption with higher accuracy than when it is based on acquired information.

[0059] Furthermore, the estimation device 100 can estimate the amount of energy consumption based on information about the building. By making an estimation based on information about the building, it is easy to estimate the amount of energy consumption with high accuracy. Specifically, by estimating the amount of energy consumption based on the amount of sunlight radiation, it is possible to make an estimation with high accuracy.

[0060] Furthermore, by having the same configuration of air conditioning devices 900 for each floor, when estimating the energy consumption of air conditioning devices 900 installed on the floor being estimated, it is possible to base it on information about air conditioning devices 900 installed on floors other than the floor being estimated.

[0061] In addition, there is no need to build a learning model for each layer, which makes it easier to reduce the accumulation of learning data. The technical scope of the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the present invention. For example, the present invention is not limited to applications of the above-described embodiments, and may be applied to embodiments in which these embodiments are appropriately combined, and is not particularly limited. [Explanation of symbols]

[0062] 100...estimation device, 101...acquisition unit, 102...storage unit, 103...first estimation unit, 104...second estimation unit, 105...third estimation unit, 500...detection device, 501...communication device, 900...air conditioning device.

Claims

1. An estimation device for estimating an amount of energy consumed by an air conditioner, an acquisition unit that acquires input information stored in a storage unit used for estimating the amount of energy consumption, the input information being information related to the configuration of the air conditioning device and information related to the building in which the air conditioning device is installed; a first estimation unit that estimates the indoor environment information by inputting information related to the configuration of the air conditioning device and information related to the building acquired by the acquisition unit into a trained first learning model that has undergone machine learning to estimate the indoor environment information in which the air conditioning device is installed; a second estimation unit that estimates the number of occupants in a room in which the air conditioning unit is installed by inputting information related to the configuration of the air conditioning unit and information related to the building acquired by the acquisition unit into a trained second learning model that has undergone machine learning to estimate the number of occupants in the room in which the air conditioning unit is installed; a third estimation unit that estimates a new amount of energy consumption by inputting information related to the configuration of the air conditioning unit, information about the building, the indoor environment information estimated by the first estimation unit, and the number of occupants estimated by the second estimation unit into a trained third learning model that has undergone machine learning to estimate the amount of energy consumption of the air conditioning unit; An estimation device comprising:

2. 2. The estimation device according to claim 1, wherein the information about the building includes at least information about the height of the building.

3. 3. The estimation device according to claim 1, wherein the building has a plurality of floors, and at least two or more of the floors have the same configuration of the air conditioning devices.

Images