JP2018055172A - Energy consumption monitoring unit, apparatus management system and program - Google Patents

Abstract

When a device is requested to change a setting that increases the amount of energy used, a peak time of energy usage accompanying a change in the set value is predicted, and an increase in the amount of energy used at the peak is suppressed.
When a change to lower the set temperature is requested during the cooling operation of the air conditioner, the power consumption after the setting change is based on the prediction model generated by the machine learning analysis and the changed setting value. The peak time prediction unit 122 that predicts the peak time of the amount, and the setting value at the peak time is calculated with reference to the setting return rate to reduce the amount of power used, and the calculated setting value is obtained at the peak time A set value model generation unit 123 that generates a set value model for controlling the air conditioning.
[Selection] Figure 3

Description

  The present invention relates to an energy usage monitoring device, a device management system, and a program, and more particularly, to a response when a setting change that increases energy usage is required for a device that consumes energy.

  In recent years, regarding the operation of air conditioners installed in building rooms, etc., an estimated model of power consumption, room temperature, etc. has been generated by machine learning using measured data on power consumption, set values such as room temperature, humidity, etc. And the system which estimates the driving | running state of an air conditioning based on this estimation model is proposed (for example, patent documents 1, 2).

  In addition, air conditioning may feel hot or cold depending on the conditions of the day, the number of people in the room, and the like. Therefore, the resident changes the set temperature for the air conditioner from the remote controller. In recent years, it has become possible to change from a personal computer (PC) or a mobile device such as a smartphone without moving to the place where the remote control is installed.

JP 2012-149839 A Special table 2013-514510 gazette JP 2007-047945 A Japanese Patent Application Laid-Open No. 08-075219 Special table 2014-531086 gazette

  In consideration of the comfort of the resident, it is preferable that the setting change from the own seat to the air conditioner can be easily performed. However, from the viewpoint of energy saving, the amount of energy used increases, which is not a preferable air conditioning operation. Thus, comfort and energy saving are in a trade-off relationship.

  The present invention predicts a peak point of energy usage accompanying a change in setting value when a setting change that increases the amount of energy usage is required for a device, and suppresses an increase in energy usage at the peak point. Objective.

  When the energy consumption monitoring device according to the present invention requires a setting change to increase the energy consumption to a device that consumes energy, the set value after the change and a prediction model for the preset energy usage are used. Based on a prediction means for predicting a peak time point of energy usage within a predetermined period when the set value is changed, and a time length from the time point when the setting change is requested to the peak time point predicted by the prediction means, and Generating means for generating a set value model indicating a set value for the device for reducing an increase in energy usage accompanying the change of the set value based on the set value before and after the setting change. And

  The generation unit generates a set value model for calculating a set value for the device between a time point when a setting change is requested and the peak time point.

  In addition, when a set return rate indicating a degree of reduction of the increase amount is set in advance, the generation unit calculates a set value at the peak time point based on the set value before and after the setting change and the set return rate. The set value model is generated so as to return to the set value to be set.

  Further, the generating means generates the set value model so that the set value is changed stepwise from the time when the setting change is requested to the peak time.

  In addition, when a setting change in which the energy usage is increased before the set value model is generated, there is a suppression unit that suppresses generation of the set value model based on the request.

  The device management system according to the present invention includes an operation control device that controls operation of a device that consumes energy, an energy usage monitoring device that monitors the energy usage of the device, and a storage that stores a prediction model of the energy usage. And the energy usage monitoring device changes the setting value based on the changed setting value and the prediction model when the device is requested to change the setting to increase the energy usage amount. Prediction means for predicting the peak time point of energy usage within a predetermined period, and the time length from the time point when the setting change is requested to the peak time point predicted by the prediction means and the set value before and after the setting change. Based on this, a generation method for generating a set value model indicating a set value for the device for reducing an increase in the amount of energy usage accompanying a change in the set value When has the operation control device, and controls the operation of the device after changing the set value for the said device, in accordance with the set value model generated by the energy usage monitor.

  The program according to the present invention is based on a set value after change and a prediction model for a preset energy usage when a computer is requested to change a setting that increases the amount of energy used by a device that consumes energy. Predicting means for predicting the peak time of energy usage within a predetermined period when the set value is changed, the time length from when the setting change is requested to the peak time predicted by the predicting means and the setting change Based on the setting values before and after, the functioning unit functions as a generation unit that generates a setting value model indicating a setting value for the device for reducing an increase in the amount of energy used due to a change in the setting value.

  According to the present invention, when a setting change that increases the amount of energy used is required for a device, the peak point of energy usage associated with the change in the set value is predicted, and an increase in energy usage at the peak point is suppressed. be able to.

It is the whole air-conditioning system lineblock diagram in this embodiment. It is a hardware block diagram of the computer which forms the used electric energy monitoring apparatus in this Embodiment. It is a block block diagram of the used electric energy monitoring apparatus in this Embodiment. In this Embodiment, it is the figure which showed typically an example of the prediction model produced | generated with the prediction model of the used electric energy produced | generated previously, and a setting change. It is the flowchart which showed the setting value model production | generation process in this Embodiment. In this Embodiment, it is the conceptual diagram which showed typically the transition of the setting value to an air conditioner.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is an overall configuration diagram of an air conditioning system in the present embodiment. In the present embodiment, an air conditioner that consumes electricity will be described as an example of a device that operates by consuming energy. FIG. 1 shows an air conditioning system 1, an air conditioner 3 installed in a living room 2, a remote controller 4, and a PC 5. The air conditioning system 1 is a form of a device management system, and is a system that manages and monitors the air conditioner 3. Although the air conditioning system 1 manages and monitors a plurality of air conditioners 3, only one set of the air conditioner 3 and the remote controller 4 are shown in FIG. 1 for convenience.

  The resident changes the set value of the room temperature by performing a predetermined operation on the remote controller 4. The resident can also change the set value of the room temperature from a predetermined application operating on the PC 5. Although it is assumed that the PC 5 is installed in the occupant's seat, each PC 5 has a function for changing the setting of the room temperature by the above application as well. Only one is shown. The air conditioning system 1 controls the operation of the air conditioner 3 so that the room temperature is specified by the resident in response to a request for changing the set value.

  The air conditioning system 1 includes an air conditioning control device 11 that controls the operation of the air conditioner 3, a power consumption monitoring device 12 that monitors the power consumption of the air conditioner 3 as an energy consumption, and a prediction model storage unit 13. Note that apparatus configurations not used in the description of the present embodiment are omitted from FIG.

  By the way, in recent years, measured data such as room temperature and humidity, setting change requests such as room temperature, and data of time-series changes in power consumption are accumulated, and this accumulated data (big data) is deep learning (machine learning). The driving model is generated by analysis. The air conditioning controller 11 normally controls the operation of the air conditioner 3 according to the operation schedule. In addition, model data indicating a fluctuation state of the power consumption when the air conditioner 3 is operated according to the operation schedule is generated by machine learning, and the model data is stored in the prediction model storage unit 13 to predict the power consumption. It is preset and registered as a model. For example, when actual measurement data such as room temperature and humidity for the latest 30 minutes are input to this prediction model, it is possible to predict the power consumption after 10 minutes. Further, when the set room temperature of the air conditioner 3 is changed, it can be seen how much the power consumption changes after the setting is changed. Specifically, when the changed setting value is input to the program that outputs the prediction model, a prediction model of the power consumption that reflects the changed setting value is generated. In the present embodiment, the peak time prediction unit 122 has a function of generating a prediction model that accompanies the change of the set value.

  FIG. 2 is a hardware configuration diagram of a computer forming the power consumption monitoring device 12 in the present embodiment. The power consumption monitoring apparatus 12 in the present embodiment can be realized with a general-purpose hardware configuration that has existed in the past. That is, the power consumption monitoring device 12 is configured by connecting the CPU 21, ROM 22, RAM 23, hard disk drive (HDD) 24, and network controller 25 provided as communication means to the internal bus 26 as shown in FIG. . In this embodiment, it is omitted from the drawing because it is not particularly used. However, a mouse or keyboard may be provided as input means, and a display may be provided as a display device.

  FIG. 3 is a block configuration diagram of the power consumption monitoring device 12 in the present embodiment. The power consumption monitoring device 12 includes a setting value receiving unit 121, a peak time prediction unit 122, a setting value model generation unit 123, a setting value storage unit 124, and a setting return rate storage unit 125. Note that apparatus configurations that are not used in the description of the present embodiment are omitted from FIG. When the resident operates the remote controller 4 or the PC 5 to request a change in the room temperature value set in the air conditioner 3 (hereinafter referred to as “set value”), the set value receiving unit 121 displays the requested content. The set value received and changed is stored in the set value storage unit 124. The peak time prediction unit 122 is provided as a prediction unit, and when the air conditioner 3 is requested to change the setting to increase the amount of power used, the changed set value and the power used in the prediction model storage unit 13 are set. Based on the prediction model of the amount, the peak time point of the power consumption within a predetermined period when the set value is changed is predicted. In the present embodiment, the description will be made assuming one day as the predetermined period. The “change in which the amount of power used is increased” is a change in the set value that lowers the room temperature when the air conditioner 3 is in cooling operation (for example, if the current set value is 28 degrees, the change to 26 degrees Change). Also, it means a change in the set value that raises the room temperature when the air conditioner 3 is in a heating operation (for example, a change to 23 degrees when the current set value is 20 degrees). The set value model generation unit 123 is provided as a generation unit, and sets the time length from the time when the setting change is requested to the peak time predicted by the peak time prediction unit 122 and the set value before and after the setting change to the air conditioner 3. Based on this, a set value model is generated that indicates a set value for the air conditioner 3 for reducing the amount of increase in the amount of power used due to the change in the set value. Further, the setting value model generation unit 123 serves as a suppression unit that suppresses the generation of the setting value model based on the request when the setting change in which the power consumption is increased before the generation of the setting value model is requested. Function.

  Although the power consumption in the air conditioner 3 is increased by changing the set value so that the power consumption increases, in the present embodiment, as will be described in detail later, the power consumption after the setting value is changed. It is characterized by predicting the peak point of time and reducing the amount of power used at the peak point. In this embodiment, a set return rate indicating the degree of increase reduction is set as an index value indicating how much to reduce. The setting return rate storage unit 125 stores the setting return rate.

  The constituent elements 121 to 123 in the power consumption monitoring device 12 are realized by a cooperative operation of a computer that forms the power consumption monitoring device 12 and a program that operates on the CPU 21 mounted on the computer. The storage units 124 and 125 are realized by the HDD 24 mounted in the power consumption monitoring device 12. Alternatively, the RAM 23 or an external storage means may be used via a network.

  Further, the program used in this embodiment can be provided not only by communication means but also by storing it in a computer-readable recording medium such as a CD-ROM or USB memory. The program provided from the communication means or the recording medium is installed in the computer, and various processes are realized by the CPU of the computer sequentially executing the program.

  The operation in this embodiment will be described. Here, the air conditioner 3 is described as being in a cooling operation.

  As described above, the air conditioning control device 11 controls the operation of the air conditioner 3 according to the operation schedule created in advance, but the transition of the power consumption that changes with the passage of time is predicted by referring to the prediction model. it can. FIG. 4 shows an example of the prediction model. In FIG. 4, the horizontal axis represents time, and the vertical axis represents the amount of power used, and shows a certain day's transition. In FIG. 4, the prediction model registered in the prediction model storage unit 13 is indicated by a one-dot chain line. Here, it is assumed that the resident operates the remote controller 4 or the PC 5 and requests to lower the room temperature of the room 2. Hereinafter, the set value model generation process performed by the power consumption monitoring device 12 in response to this request will be described with reference to the flowchart shown in FIG.

  When the set value receiving unit 121 in the power consumption monitoring device 12 receives the changed set value in response to a request from the resident, the set value receiving unit 121 records it in the set value storage unit 124 (step 101). In general, the resident designates the temperature to be lowered by performing a predetermined number of button operations on the current set value, so the set value receiving unit 121 interprets the operation content and converts it to an absolute temperature. And recorded in the set value storage unit 124. Thereby, the temperature before and after the setting change can be grasped by referring to the set value storage unit 124. When recording the temperature, the operation date and time may also be recorded. In the present embodiment, the set value storage unit 124 is provided in the power consumption monitoring device 12, but information recorded by the air conditioning system 1 may be referred to.

  By the way, the setting change of the air conditioner 3 can be performed by each resident from the personal computer 5 and a setting value model to be described later is generated in response to the request for the setting change. However, another resident may request a setting change before the setting value model generation is completed, that is, during the setting value model generation. In this case, processing becomes complicated, and there is a possibility that an appropriate prediction model is not generated. Therefore, in the present embodiment, if the present process being executed in response to the request before this request has not been completed, that is, if the setting value model is being generated (Y in step 102), the administrator or the like If the generation is not permitted from (No in step 103), this process is terminated. If the generation is permitted (Y in step 103), this processing is continued according to conditions such as waiting until the generation of the setting value model is completed or forcibly overwriting the setting value model being generated. In the present embodiment, the set value designated by the resident may be recorded in the set value storage unit 124 even when the set value model is not generated. At this time, flag information for determining whether or not the setting change request has caused the generation of the setting value model may be registered in association with the setting value.

  As described above, when the setting change does not increase the power consumption (N in Step 104), the power consumption after the setting change does not exceed the prediction model, so the process is terminated without newly generating a prediction model. To do. On the other hand, when the request by the resident lowers the set temperature (Y in step 104), the peak time prediction unit 122 acquires a prediction model of the energy usage from the prediction model storage unit 13, and the prediction model is within a predetermined period. The peak point of energy usage on the current day when the setting value is changed based on the actual information up to the point when the setting value change is requested (hereinafter “current time”) and the setting value after the change. Is predicted (step 105). At this time, the peak time prediction unit 122 generates a prediction model after the setting change. In FIG. 4, the prediction model after the setting change is indicated by a solid line. As is clear from FIG. 4, at the peak time, an increase in power consumption of the increase amount D is predicted as the setting is changed.

  When the peak time is predicted, the set value model generation unit 123 calculates the time (time length) from the current time to the peak time (step 106). Subsequently, the set value model generation unit 123 generates a set value model (step 107). Before generating the set value model, the set temperature of the air conditioner 3 at the peak time will be described for reference.

  In the present embodiment, the set temperature of the air conditioner 3 at the peak time is calculated by the following equation (1).

Peak point set temperature P = set value after change + (set return rate × ABS (set value before change−set value after change)) (1)
However, ABS means an absolute value. For example, if the pre-change set value is 25 degrees, the post-change set value is 20 degrees, and 50% is set as the set return rate, the peak time set temperature P is calculated as follows.
P = 20 + (0.5 × ABS (25−20)) = 22.5

  That is, since the occupant changed the setting to 20 degrees when it was currently set to 25 degrees, the air-conditioning control is performed so that the room temperature becomes 20 degrees unless the setting is changed until the peak time. However, in the present embodiment, it means that the air conditioning is controlled so as to be 22.5 degrees at the peak time of the amount of power used. In the present embodiment, a set value model for realizing such air conditioning control is generated. This set value model can be expressed by Expression (2).

T = setting value after change + (setting return rate × ABS (setting value before change−setting value after change)) / L × t (2)
Here, L is the time length between the current time and the peak time, and t is the time for obtaining the set value (elapsed time from the current time). For example, when the time length L between the current time point and the peak time point is 10 minutes (L = 600 seconds), the set value T after 3 minutes (t = 180 seconds) from the current time point is calculated as follows. .

  T = 20 + ((0.5 × ABS (25-20)) / 600) × 180 = 20.75

In this way, the set value for the air conditioner 3 can be calculated for each elapsed time from the present time. For example, if the unit time for obtaining the set value is 1 minute,
T = 20 + ((0.5 × ABS (25-20)) / 600) × 60 = 20.25
And two minutes from now,
T = 20 + ((0.5 × ABS (25−20)) / 600) × 120 = 20.5
It becomes.

  When the set value model generation unit 123 generates the set value model as described above, the air conditioning control device 11 controls the operation of the air conditioner 3 according to the set value model. By controlling the set temperature step by step for each unit time until the peak time, the resident is less likely to feel an increase in room temperature, and the comfort in the living room 2 is less likely to be impaired.

  The setting value according to the setting value model generated by the above processing can be represented by a solid line T from the current time point to the peak time point in FIG. In FIG. 4, the set value that increases stepwise is shown by a straight line. However, if the unit time is shortened, the change of the set value becomes linear.

  When such air conditioning control is carried out, first, the set temperature at the peak time is corrected upward, so that an effect of reducing the amount of power used can be obtained as compared with the case of air conditioning control according to the set value by the resident. Based on FIG. 4, a part of the used electric energy in the area portion surrounded by the prediction model and the post-change prediction model is reduced. This reduction effect depends on the value of the set return rate. As the setting return rate increases, the setting value before the setting change is restored, and the reduction effect increases. On the contrary, the smaller the setting return rate is, the closer to the set value after the setting change, the smaller the reduction effect. The setting return rate may be set as appropriate by an administrator or the like in consideration of the state of the amount of power used based on the operation of the air conditioner 3. Alternatively, the set return rate may be automatically adjusted with reference to the resident's request recorded in the set value storage unit 124. For example, when the set temperature is continuously requested to be set downward, it is determined that there are many residents who feel that the interior of the living room 2 is hot, and the setting return rate is adjusted to be small.

  According to the present embodiment, the peak time of power consumption after setting change is predicted, and the power consumption at the peak time is suppressed. Thereby, the electric power consumption on the day can be reduced as a whole. By the way, if the amount of power used is reduced based on a time point earlier than the peak time point, the set temperature of the room 2 is raised earlier than in the case of the peak time point, and comfort is impaired. On the other hand, if the power consumption is reduced based on a time point later than the peak time point, an energy saving effect cannot be expected as compared with the peak time point. In consideration of the trade-off between the comfort and the energy saving effect, in the present embodiment, the power consumption is reduced based on the peak time of the power consumption after the setting is changed.

  In the present embodiment, the air conditioner 3 has been described as an example, but the present invention can also be applied to other electric devices. In this embodiment, electricity is taken as an example of energy, but it can be applied to other energy such as water and gas.

  DESCRIPTION OF SYMBOLS 1 Air conditioning system, 2 Living room, 3 Air conditioner, 4 Remote control, 5 Personal computer (PC), 11 Air-conditioning control apparatus, 12 Power consumption monitoring apparatus, 13 Prediction model memory | storage part, 14 Setting value memory | storage part, 21 CPU, 22 ROM , 23 RAM, 24 hard disk drive (HDD), 25 network controller, 26 internal bus, 121 set value reception unit, 122 peak time prediction unit, 123 set value model generation unit, 124 set value storage unit, 125 set return rate storage unit .

Claims (7)

When a setting change that increases the amount of energy used is requested to a device that consumes energy, the setting value is changed based on the changed setting value and the preset model for predicting energy usage. A prediction means for predicting the peak point of energy usage within a predetermined period;
Based on the time length from the time point when the setting change is requested to the peak time point predicted by the predicting means and the setting value before and after the setting change, the amount for increasing the amount of energy use accompanying the change in the setting value is reduced. Generating means for generating a setting value model indicating a setting value for the device;
An energy usage monitoring device comprising:   2. The energy usage monitoring according to claim 1, wherein the generation unit generates a setting value model for calculating a setting value for the device between a time point when a setting change is requested and the peak time point. apparatus.   When a set return rate indicating a degree of reduction of the increase amount is set in advance, the generation unit calculates a set value at the peak time point based on the set value before and after the setting change and the set return rate. The energy usage monitoring apparatus according to claim 2, wherein the set value model is generated so as to return to a set value.   The energy generation amount monitoring according to claim 2, wherein the generation unit generates the setting value model so that the setting value is changed in a stepwise manner from a time point when a setting change is requested to the peak time point. apparatus.   2. The apparatus according to claim 1, further comprising: a suppression unit that suppresses generation of the setting value model based on the request when a setting change in which an energy consumption is increased before the setting value model is generated is requested. The energy usage monitoring device described in 1. An operation control device that controls the operation of energy consuming equipment;
An energy usage monitoring device for monitoring the energy usage of the device;
Storage means for storing a prediction model of energy consumption;
Have
The energy usage monitoring device is:
When the device is requested to change the setting to increase the energy consumption, based on the changed setting value and the prediction model, the peak time point of the energy usage within a predetermined period when the setting value is changed. Prediction means to predict;
Based on the time length from the time point when the setting change is requested to the peak time point predicted by the predicting means and the setting value before and after the setting change, the amount for increasing the amount of energy use accompanying the change in the setting value is reduced. Generating means for generating a setting value model indicating a setting value for the device;
Have
The operation control device controls the operation of the device according to the set value model generated by the energy usage monitoring device after the change of the set value to the device.
A device management system characterized by that. Computer
When a setting change that increases the amount of energy used is requested to a device that consumes energy, the setting value is changed based on the changed setting value and the preset model for predicting energy usage. A prediction means for predicting the peak point of energy usage within a predetermined period,
Based on the time length from the time point when the setting change is requested to the peak time point predicted by the predicting means and the setting value before and after the setting change, the amount for increasing the amount of energy use accompanying the change in the setting value is reduced. Generating means for generating a setting value model indicating a setting value for the device;
Program to function as.

Images