Power level control device

Abstract

A power level control device (10) comprises a premise parameter determination unit (110) that determines, as a parameter serving as a premise for calculating power consumption cost efficiency, at least one of: the length of one turn; the number of room divisions when a range in which air conditioning of an air conditioner affects a server is defined as a room; an in-room initial temperature level indicating a setting stage of an in-room initial temperature; an air conditioning control level indicating a control stage of air conditioning control of the air conditioner; and an air conditioning power efficiency group indicating a group of air conditioners that execute control of the same air conditioning control level.

Description

Power consumption control device

[0001] The present invention relates to a power control device for controlling the power consumption of servers and air conditioners in each room of a data center (hereinafter sometimes referred to as "DC").

[0002] With the goal of reducing greenhouse gas emissions, efforts are being made to expand the use of renewable energy sources such as solar and wind power, promoting the adoption of facilities, equipment, and technologies. However, because the supply of renewable energy is affected by changes in the natural environment and it cannot be stored, it is necessary to maximize the efficiency of renewable energy utilization by increasing demand in line with peak electricity supply. Consequently, it is anticipated that electricity prices will tend to fluctuate throughout the day, for example, by making daytime electricity, which is expected to be generated from solar power, cheaper than nighttime electricity.

[0003] Furthermore, the load within the data center (DC) fluctuates throughout the day, and the system controlling the DC needs to consider both the load arriving at the DC and fluctuations in electricity prices. In addition, the amount of data processed (load) in the DC is increasing year by year, and in this respect as well, it is necessary to improve the overall power consumption efficiency of the DC (the amount of power consumed by the DC as a whole for a certain amount of data processing). In a DC, in addition to the power consumption of servers, the power consumption of air conditioning accounts for a large proportion, and a reduction in overall power consumption is required for the DC as a whole.

[0004] Non-Patent Document 1 describes a technology that optimizes the overall power consumption of a data center by considering the power consumption of air conditioning and servers (IT equipment). The data center air conditioning-linked IT load allocation optimization method described in Non-Patent Document 1 predicts the future load trends of IT equipment by collecting operational and monitoring information of IT equipment in the data center, and calculates the power increase of the air conditioning system in accordance with the power increase of the IT equipment. Then, it solves an optimization problem that minimizes the objective function, which is the power consumption of the data center, so that the load concentration rate on IT equipment increases over time, that is, so that the number of operating IT equipment is reduced. This calculates the allocation of IT load (virtual machines) to IT equipment that minimizes the power consumption of the data center.

[0005] However, the technology described in Non-Patent Document 1 employs a general rule-based standard in the air conditioning power model used to calculate the power consumption of air conditioning equipment, which is independent of the different equipment conditions for each DC. Therefore, it was difficult to optimize the power consumption of the DC by taking into account individual equipment conditions such as the placement of air conditioning equipment, airflow, server placement configuration within the DC, and thermal cooling efficiency.

[0006] Furthermore, the technology described in Non-Patent Document 1 did not take into account the daily fluctuations in electricity rates, the fluctuations in the load flowing into the DC, the load arrangement within the DC, or the air conditioning control, in order to optimize power cost efficiency. In relation to these points, the present inventors have developed the technology described in Patent Document 1 based on the following technical concept.

[0007] For a given load arriving in a data center (DC), factors affecting the power consumption of air conditioning include the air conditioning control level of the air conditioner (control parameter <1>) and the load distribution within the DC. Load distribution within the DC involves how much load is distributed to each room in the DC (load distribution pattern between DC rooms (distribution pattern): control parameter <2>) and how the distributed load is further allocated to the servers within each room (load distribution pattern within the DC room (distribution pattern): control parameter <3>).

[0008] Here, we consider a "room-specific load distribution amount / load placement pattern" (hereinafter referred to as the "distribution placement pattern") in which a predetermined control time for updating the air conditioning control and load placement within each room is defined as one turn (for example, one hour), the load distribution amount to each room for the total DC load amount in each turn is changed, and furthermore, the load distribution amount to each room is changed according to the distribution ratio for each area within the room (the "placement control area 30" in Figure 23 described later).

[0009] After fixing one of these distribution patterns, that is, fixing the load distribution amount (load distribution pattern) for each room in each turn and the load distribution pattern within the DC room, we determine the power consumption cost efficiency (E) that maximizes the total air conditioning power cost efficiency for the DC and the air conditioning control pattern for each room at that time. Here, if the number of control stages of the air conditioning control is "G", the number of rooms is "R", and the number of turns is "tn", then the number of patterns of the air conditioning control levels is G tn・R This will be the street. tn・R The pattern that maximizes power consumption cost efficiency (E) is determined from the street patterns. As a result, the power consumption control device described in Patent Document 1 can maximize the power consumption cost efficiency of the entire DC by optimizing the air conditioning control level of each room, the load distribution amount between rooms, and the load arrangement pattern within each room when the amount of load flowing into the DC and the electricity charges fluctuate during a certain continuous period of time.

[0010] Jun Okitsu et al. (4 authors), "Optimization Method for IT Load Allocation Linked to Air Conditioning for Environmentally Friendly Data Centers," FIT (Forum on Information Technology) 2010, 9th Forum on Information Science and Technology, RC-009.

[0011] International Publication No. 2024 / 171365

[0012] However, the server environment in a data center room varies greatly from room to room due to environmental factors affecting cooling, such as the number of servers installed, the amount of heat generated in the room, the number of air conditioners installed, and the arrangement of racks and air conditioners. Furthermore, in order to begin learning and operating optimal air conditioning control, it was necessary to predetermine the parameters that would serve as the basis for the control. However, conventional technology does not have defined rules for determining the optimal parameters, which can lead to longer learning times because unnecessary learning patterns cannot be eliminated, or the accuracy of optimization decreases and the power saving effect is reduced, leaving room for further improvement.

[0013] The present invention was made in consideration of these points, and aims to improve the accuracy of optimization and enhance power saving effects by eliminating unnecessary learning patterns and shortening the learning time through measurement to understand the cooling characteristics of the DC room, and by determining the optimal assumption values ​​(parameters).

[0014] The power consumption control device according to the present invention is a power consumption control device that is communicated to a room-specific control device that controls a plurality of servers and a plurality of air conditioners in a room in a data center, and the power consumption control device generates learning data by controlling the air conditioning control level of the air conditioners, the load distribution pattern between the rooms, and the distribution pattern of the load distributed to each room to the servers via the room-specific control device, for each turn representing a predetermined control time, thereby generating a distribution and placement pattern that represents the distribution pattern and placement pattern that maximizes the power consumption cost efficiency of the entire data center. The power consumption control device determines the air conditioning control level of the air conditioner, and is characterized in that it includes a prerequisite parameter determination unit that determines at least one of the following parameters as prerequisites for calculating the power consumption cost efficiency: (1) the length of one turn, (2) the number of room divisions when the area affected by the air conditioning of the air conditioner to the server is defined as a room, (3) the initial room temperature level indicating the setting stage of the initial temperature inside the room, (4) the air conditioning control level indicating the control stage of the air conditioning control of the air conditioner, and (5) the air conditioning power efficiency group indicating a group of air conditioners that perform control at the same air conditioning control level.

[0015] According to the present invention, by measuring the cooling characteristics of the DC room, it is possible to eliminate unnecessary learning patterns and shorten the learning time, and by determining the optimal assumption values ​​(parameters), optimization accuracy can be improved and power saving effects can be enhanced.

[0016] This is a functional block diagram showing an example configuration of the power consumption control device according to this embodiment. This is a diagram showing an example of default control of an air conditioner according to this embodiment. This is a diagram showing the initial settings of an air conditioner in the turn length determination process according to this embodiment. This is a diagram showing the measurement results in the turn length determination process according to this embodiment. This is a diagram showing the measurement results in the room division setting process according to this embodiment. This is a diagram showing the measurement results of upper limit determination in the room initial temperature level determination process according to this embodiment. This is a diagram showing the measurement results of upper limit determination in the room initial temperature level determination process according to this embodiment. This is a diagram showing the measurement results of lower limit determination in the room initial temperature level determination process according to this embodiment. This is a diagram showing an example of control stages of the air conditioning control level according to this embodiment. This is a diagram showing the measurement results of the default power consumption of an air conditioner in the power efficiency group determination process according to this embodiment. This is a diagram showing the measurement results of an air conditioner in the power efficiency group determination process according to this embodiment. This is a diagram showing the power consumption difference of an air conditioner in the power efficiency group determination process according to this embodiment. This is a diagram showing the power efficiency of each air conditioner in the power efficiency group determination process according to this embodiment. This is a diagram showing the power efficiency classification in the power efficiency group determination process according to this embodiment. This is a diagram for explaining the learning pattern for obtaining the learning history according to this embodiment. This is a diagram illustrating the air conditioning power function according to this embodiment. This is a diagram illustrating the temperature function according to this embodiment. This is a functional block diagram showing an example configuration of the in-room control device according to this embodiment. This is a hardware configuration diagram showing an example of a computer that realizes the functions of the power consumption control device and the in-room control device according to this embodiment. This is a diagram showing the overall configuration of a power consumption control system including the power consumption control device according to the prior art and this embodiment. This is a diagram for explaining the logical area in a room according to the prior art. This is a diagram for explaining the area within a room according to the prior art and this embodiment. This is a diagram for explaining the learning pattern for acquiring the learning history according to the prior art. This is a diagram showing an example of the control stages of the air conditioning control level according to the prior art.

[0017] <Prior Art and its Problems> First, the prior art that forms the basis of this invention and its problems will be explained in detail.

[0018] The power consumption control device described in Patent Document 1 calculates the power consumption cost efficiency (E) of the entire DC for N turns, after changing the pattern of the air conditioning control level for each room for each room, based on the load distribution amount and load placement pattern (details of the "distribution placement pattern" are described later). Specifically, in calculating this power consumption cost efficiency (E), one pattern from each distribution placement pattern is sequentially fixed (selected), and the power consumption cost efficiency (E) is calculated. This power consumption cost efficiency (E) is defined as the value obtained by dividing the sum of the electricity charges based on the air conditioning power consumption amount for each consecutive turn and the electricity charges based on the server load amount for each consecutive turn by the total load amount for each consecutive turn, as shown in the following formula (1).

[0019]

[0020] Here, the coefficients and parameters for the j-th turn (tj turn) are as follows: "m" is the number of rooms, and "n" is the number of turns. "c[tj]" represents the power consumption cost efficiency coefficient for the turn, specifically the electricity rate coefficient. For example, (7 yen / kWh), it is a coefficient that converts power consumption into electricity rates. "k" is a coefficient that converts server load into server power consumption (server load amount / power coefficient). The parameters of equation (1) are as follows.

[0021]

[0022] Here, the load amount and arrangement pattern within the room refer to the "distribution arrangement pattern". Also, the total load amount flowing into DC1000 from turn "1" to turn "n", indicated by X[tj] in the lower part of equation (1), is given by the following equation (2).

[0023]

[0024] In calculating this power consumption cost efficiency (E), the number of patterns for the air conditioning control level is, as described above, G, where G is the number of control stages, R is the number of rooms, and tn is the number of turns. tn・R The power consumption control device then fixes one of the distribution arrangement patterns, calculates the power consumption cost efficiency (E) for the DC total, and determines the distribution arrangement pattern that maximizes the power consumption cost efficiency (E) for the DC total, as well as the air conditioning control pattern for each room at that time.

[0025] Figure 21 is a diagram showing the overall configuration of a power consumption control system 1A, including a conventional power consumption control device 10A. As shown in Figure 21, the power consumption control system 1A is configured to include a DC (data center) 1000 having multiple rooms 500 (rooms "A", "B", and "C" in Figure 21) each equipped with multiple servers 3 and one or more air conditioners 4, an in-room control device 20A that is connected to the multiple servers 3 and one or more air conditioners 4 in each room 500 and provided corresponding to each room 500, and a power consumption control device 10A that is connected to each in-room control device 20A and controls the total power consumption of the DC 1000.

[0026] In the conventional technology, Room 500 is defined as a cohesive, logical area where a specific group of air conditioners exerts a cooling effect on a group of servers. Normally, Room 500 within DC 1000 can be considered a single area surrounded by walls on all four sides, but in DC 1000, a group of servers may be arranged in one very large, cohesive space. In such cases, depending on the location of the air conditioners and the arrangement of the servers within Room 500, even if physically located within the same Room 500, some servers may not be affected by the cooling effect of the air conditioner 4.

[0027] Therefore, for example, as shown by reference numeral 501 in Figure 22, if air conditioning unit group "A" and air conditioning unit group "B" are physically located in separate rooms, before executing control within room 500, an area <1> is identified and set as a room in which only the air conditioning of air conditioning unit group "A" can be controlled to lower a predetermined temperature (e.g., 2°C). Similarly, an area <2> is identified and set as a room in which only the air conditioning of air conditioning unit group "B" can be controlled to lower a predetermined temperature (e.g., 2°C). Note that, as shown by reference numeral 502 in Figure 22, if air conditioning unit group "A" affects all servers and air conditioning unit group "B" affects all servers, then area <1> and area <2> are set as the same room.

[0028] Returning to Figure 21, in DC1000, a room control device 20A is provided in association with each of the rooms 500 defined as logical areas as described above. The room control device 20A corresponding to each room 500 may acquire status information (such as air conditioning power consumption) of the air conditioners 4 installed in the room 500 or transmit air conditioning control information (such as set temperature and airflow) via an air conditioning management device (not shown), or it may be directly connected to each air conditioner 4 without going through the air conditioning management device.

[0029] Furthermore, the in-room control device 20A is connected to each server 3 in room 500. The in-room control device 20A may be connected to the server 3 via a server management device (not shown) to acquire status information (server load, etc.) and transmit control information (load allocation, etc.) from the servers 3 located in room 500, or it may be directly connected to the servers 3.

[0030] In this description, it is assumed that a virtualization infrastructure is built and operated on each server 3 within DC1000. Known open-source virtualization infrastructures include OpenStack®, software for building cloud environments, and Kubernetes®, software for managing and operating containerized workloads and services. OpenStack is primarily used for managing and operating physical machines and virtual machines (VMs). Kubernetes is primarily used for managing and operating containers. In this specification, a virtualized application (consisting of one or more containers, or one or more VMs, etc.) on a virtualization infrastructure is referred to as a virtual resource. In Kubernetes, the smallest execution unit of an application is a Pod, which consists of one or more containers.

[0031] In the conventional DC1000, the entire server 3 to be accommodated is divided into areas where multiple servers 3 (server groups) are located, as shown in Figure 23, and controlled as a "location control area". This location control area 30 is an area that accommodates a group of servers where virtual resources are located, and represents a consolidated area for processing load. Figure 23 shows an example in which location control areas "1" to "4" are provided.

[0032] Furthermore, corresponding to the server group placement control area 30 in the conventional technology, an "air conditioning control area" is provided as shown in Figure 23. The air conditioning control area 40 is a consolidated area for measuring the room temperature effect of air conditioning control, and it faces either the intake side or the exhaust side of the server 3. The air supplied from the air conditioner 4 is blown out from the intake side air conditioning control area 40 (air conditioning control areas "1", "2", "5", and "6" in Figure 23) via, for example, piping installed under the floor of the DC1000. Then, air heated by the heat from each server 3 is taken in from the intake port of the piping installed in the exhaust side air conditioning control area 40 (air conditioning control areas "3" and "4" in Figure 23), creating an airflow that returns to the air conditioner 4.

[0033] Each of these air conditioning control zones 40 is equipped with multiple sensors (temperature sensors, etc.). In addition, each of the placement control zones 30 is equipped with temperature sensors at the intake and exhaust ports of each server 3 and the servers 3 selected within the area. Information obtained from these sensors (sensor information) can be acquired by the room control device 20A and the power consumption control device 10A associated with that room via a communication line or the like.

[0034] Furthermore, in conventional technology, in each room within the DC, the room control device 20A controls a total l × m × n × o pattern (hereinafter referred to as the "predetermined learning pattern") by changing the initial room temperature (l), server load (m), server load arrangement pattern (n), and air conditioning control level (o), and acquires information on the air conditioning power consumption, the room temperature at the end of the turn, and the room temperature reward pass / fail judgment as learning history.

[0035] The in-room control device 20A sets the initial temperature (initial room temperature) to three levels: "18°C", "24°C", and "30°C", and sets the server load to three levels: "30kW", "60kW", and "90kW", as shown in Figure 24. Regarding the load distribution pattern, the example shown in Figure 24 indicates that the server load of the room 500 is distributed as "25%-25%-25%-25%", "50%-50%-0%-0%", and "0%-0%-50%-50%" in each of the four distribution control areas "1", "2", "3", and "4" (see Figure 23).

[0036] Regarding air conditioning control, the output value of the air conditioning control, which indicates the air conditioning control capacity, is divided into multiple stages according to its intensity, and these are defined as air conditioning control levels. For example, as shown in Figure 25, air conditioning control level "1" corresponds to a set temperature of 32°C. Air conditioning control level "2" corresponds to a set temperature of 28°C. Then, up to air conditioning control level "5", the higher the intensity of the air conditioning control level, the higher the cooling capacity. However, this tends to increase the amount of electricity consumed by the air conditioning system.

[0037] In the example shown in FIG. 24, it is set in three steps as "set temperature 32°C" (air conditioning control level "1" in FIG. 25), "set temperature 24°C" (air conditioning control level "3" in FIG. 25), and "set temperature 16°C" (air conditioning control level "5" in FIG. 25).

[0038] As a result of executing in a predetermined learning pattern, the in-room control device 20A acquires information on the air conditioning power consumption amount in that turn. Also, the in-room control device 20A acquires information on the in-room temperature at the end of the turn from the temperature sensor. Note that the in-room temperature reward pass / fail determination is information indicating the determination result (pass / fail) as to whether the temperature after the turn exceeds a predetermined in-room temperature threshold value (in-room temperature threshold). The in-room temperature threshold is a threshold value (limit temperature) set by the system management side for each room 500 of DC1000, and is set on the premise that control in a state exceeding the temperature threshold is not performed. For patterns that result in "fail" in this in-room temperature reward pass / fail determination, since they are not executed in actual operation, they are not stored as learning data. Thereby, the calculation load of the power amount control device 10A can be reduced.

[0039] Note that in the in-room control device 20A according to the prior art, information on each parameter set in this predetermined learning pattern was set in advance by an administrator or the like of the power amount control system 1A.

[0040] However, as described above, the server environment of the DC room varies greatly from room to room depending on environmental factors affecting cooling, such as the number of servers installed, the amount of heat generated in the room, the number of air conditioners installed, and the layout relationship between the number of racks and the air conditioners. Therefore, in starting the learning and operation of optimal air conditioning control, it is necessary to optimally set the parameters that are the premise of control, taking into account the cooling characteristics of the actual DC environment and the limitations of the learning time.

[0041] <This Embodiment> Next, the power consumption control system 1 including the power consumption control device 10 according to this embodiment will be described in detail. The overall configuration of the power consumption control system 1 according to this embodiment is the same as that of the power consumption control system 1A according to the prior art shown in FIG. 21. As shown in FIG. 21, the power consumption control device 10 includes a DC (data center) 1000 having a plurality of rooms 500 (rooms "A", "B", and "C" in FIG. 21) equipped with a plurality of servers 3 and one or more air conditioners 4, a room internal control device 20 communicatively connected to the plurality of servers 3 and one or more air conditioners 4 in the room 500 and provided corresponding to each room 500, and a power consumption control device 10 communicatively connected to each room internal control device 20 and controlling the total power consumption of the DC 1000.

[0042] When starting learning and control for calculating the power consumption cost efficiency (E) represented by formula (1), the power consumption control device 10 according to this embodiment optimally determines the premise parameters through measurement for grasping the cooling characteristics of the DC room. In this embodiment, the power consumption control device 10 is characterized in that (1) the length of one cycle, (2) the number of room divisions, (3) the initial temperature level in the room, (4) the air conditioning control level, and (5) the air conditioning power efficiency group are determined in advance as the premise parameters. In the power consumption control system 1 according to this embodiment, by appropriately determining these premise parameters, when calculating the power consumption cost efficiency (E), unnecessary learning patterns are excluded to shorten the learning time, the accuracy of optimization is improved, and the power saving effect can also be improved. Note that the power consumption control device 10 may set not only when determining all the parameters (1) to (5) as the premise parameters, but also at least any one of the premise parameters (1) to (5).

[0043] Next, the power consumption control device 10 and the room internal control device 20 constituting the power consumption control system 1 according to this embodiment will be specifically described.

[0044] <Power Consumption Control Device> Figure 1 is a functional block diagram showing an example of the configuration of the power consumption control device 10 according to this embodiment. The power consumption control device 10 collects learning history (learning data) from each of the in-room control devices 20 corresponding to each room 500 and generates learning history information 550 for each room 500. Then, based on the learning history, the power consumption control device 10 calculates the overall power cost efficiency of the DC when the air conditioning control level in each room is changed for each distribution arrangement pattern, and extracts the air conditioning control level for each room that maximizes the overall power cost efficiency of the DC in the relevant pattern. The power consumption control device 10 determines the distribution arrangement pattern and the air conditioning control level in the relevant pattern that maximize the overall power consumption cost efficiency of the DC from all distribution arrangement patterns. This process is referred to as the "cost efficiency maximum pattern determination process". The power consumption control device 10 is characterized by having a prerequisite parameter determination unit 110 (Figure 1) that determines various parameters (prerequisite parameters) that are assumed when calculating this power consumption cost efficiency (E). This power consumption control device 10 is composed of a computer that includes a control unit 100, an input / output unit 160, and a storage unit 170.

[0045] The input / output unit 160 performs input and output of information between the in-room control devices 20 and external system management devices (not shown), etc. This input / output unit 160 consists of a communication interface that sends and receives information via a communication line and an input / output interface that performs input and output of information between the input device such as a keyboard and an output device such as a monitor (not shown).

[0046] The storage unit 170 is composed of a hard disk, flash memory, RAM (Random Access Memory), etc. Programs for executing each function of the control unit 100 and information necessary for processing by the control unit 100 are temporarily stored in this storage unit 170. For example, various prerequisite parameters determined by the prerequisite parameter determination unit 110, which will be described later, the power consumption cost efficiency for the relevant turn "c[tj]" (power consumption cost efficiency coefficient (electricity rate coefficient) at turn j) (i.e., information on the change in the electricity rate coefficient), and various coefficients such as the load-server power proportionality coefficient "k" are stored in the storage unit 180. In addition, the storage unit 170 stores the total load amount in DC 1000 for N turns (information on the load amount flowing into DC in each turn) from the management device of the power consumption control system 1 (Figure 21), etc.

[0047] The control unit 100 oversees the overall processing performed by the power consumption control device 10, and as shown in Figure 1, it includes a prerequisite parameter determination unit 110, a room-by-room learning history information generation unit 120, an air conditioning power / temperature function approximation unit 130, a distribution arrangement pattern generation unit 140, and an optimal control calculation unit 150 (power cost efficiency calculation unit 151, control pattern determination unit 152).

[0048] The prerequisite parameter determination unit 110 determines various parameters (prerequisite parameters) that are used as prerequisites when calculating the power consumption cost efficiency (E). This prerequisite parameter determination unit 110 includes an initial room temperature determination unit 111, a turn length determination unit 112, a room division setting unit 113, an initial room temperature level determination unit 114, an air conditioning control level determination unit 115, and a power efficiency group determination unit 116.

[0049] The initial room temperature determination unit 111 determines the default temperature (initial room temperature) in the DC room at the present time (before the room division described later). The initial room temperature determination unit 111 determines the average intake temperature of the server 3 in the DC room, which can be stably obtained when the default control is executed on each air conditioner 4 in the DC room, as the default temperature (initial room temperature).

[0050] Figure 2 shows an example of the default control of the air conditioner 4 according to this embodiment. This default control means that each air conditioner 4 installed in the room is controlled in a way that allows it to operate for a predetermined time (for example, 1 hour) without changing the average temperature of the intake port of the server 3. For example, as shown in Figure 2, for the air conditioners 4 (No. 10, 11, 12) of Company A, the indoor set temperature is set to "26°C", the outlet set temperature to "18°C", control method #1 to "intake priority", control method #2 to "variable airflow", and airflow to "23 (Hz)". In this case, the temperature difference between the indoor set temperature and the outlet set temperature is "8°C". Furthermore, for Company B's air conditioners (No. 20, 21, 22, 23), the indoor set temperature is set to "26°C", the outlet set temperature to "19°C", control method #1 "Intake priority", control method #2 "Variable airflow", and airflow to "38 (Hz)". In this case, the temperature difference between the indoor set temperature and the outlet set temperature is "7°C".

[0051] The initial room temperature determination unit 111 measures the average temperature of the air intake of the server 3 when the air conditioning control and room temperature are stable, and determines the default temperature (initial room temperature). Since the DC room temperature is stable, in the default control, the average temperature of the air intake and the room set temperature will be approximately the same.

[0052] Returning to Figure 1, the turn length determination unit 112 determines the length of one turn, which is the periodic period for updating the air conditioning control value (air conditioning control level). Specifically, the turn length determination unit 112 operates each air conditioner 4 at the lowest temperature it can set, starting from the default temperature, and measures the difference between the intake temperature of the server 3 measured previously and the intake temperature of the server 3 at predetermined intervals (for example, every hour). The operation of the air conditioner 4 at the lowest temperature it can set will be referred to as "maximum control" below. The turn length determination unit 112 considers that convergence has occurred when, at the nth time, the average difference from the previous intake temperature (average intake temperature difference) falls within -m degrees (below a predetermined temperature). The turn length determination unit 112 assumes a time lag until the first batch of air conditioning starts to take effect and determines the length of time up to the (n-1)th time as one turn.

[0053] Specifically, as shown in Figure 3, all air conditioners 4 are set to the default temperature (initial room temperature), and the air outlet temperature setting for air conditioner 4 is set to "X". This X is the lowest temperature that air conditioner 4 can set. The indoor set temperature is set to X + temperature difference (8°C). Control method #1 is set to "intake priority", and control method #2 is set to "variable airflow". The airflow is set to "variable" (automatic control). In this state, the turn length determination unit 112 monitors the temperature of the intake port temperature sensor of each server 3 every hour from the measurement start time, and checks whether the average difference from the previous (1 hour ago) intake port temperature (average intake port temperature difference) is less than or equal to -m degrees (here, a predetermined temperature of 1 degree).

[0054] Figure 4 shows the measurement results when the start time is 12:00. At 13:00, one hour after the start of measurement, the difference from the previous intake port temperature is 1 degree or more for all intake ports 1 to 5. Therefore, the average intake port temperature difference is greater than 1 degree, and the termination flag is "FALSE". At 14:00, one hour later, the difference from the previous intake port temperature is less than 1 degree for all intake ports 1 to 5. Therefore, the average intake port temperature difference is 1 degree or less, and the termination flag is "TRUE", ending the measurement.

[0055] The turn length determination unit 112 determines the length of one turn as "(measurement end time) - (measurement start time) - 1". In this case, (14:00 - 12:00) - 1 = 1. In other words, the turn length determination unit 112 determines the turn length to be "1 hour".

[0056] Returning to Figure 1, the room division setting unit 113 determines the logical number of room divisions necessary for optimal control. The room division setting unit 113 also sets up the rooms 500 within the DC 1000 as a unified logical area within the determined number of divisions, where a specific group of air conditioners will have a cooling effect on the server group.

[0057] The room division setting unit 113 determines the total number of items N required for acquiring the learning history (learning data) from the following formula: Total number of items N = Initial temperature level count × (Air conditioning control level count ^ Power efficiency group count) × Load level count × Outside temperature level count ... Formula (3) Then, it determines the upper limit of the number of room divisions D from the following formula: Upper limit of the number of room divisions D = Initial learning time (M) / N × T (length of one turn) ... Formula (4)

[0058] Here, it is assumed that the initial number of temperature levels "3", the number of air conditioning control levels "3", the number of power efficiency groups "2", the number of load levels "3", and the number of outside temperature levels "3" were predetermined by the administrator of the power consumption control system 1. Therefore, the total number of items N = 3 × (3^2) × 3 × 3 = 243. If the user-specified learning data measurement time is 30 days (720 hours), the maximum number of room divisions D will be as follows: Maximum number of room divisions D = 720 / 243 × 1 = 2.963. Note that the decimal part is truncated and the maximum number of room divisions D is set to "2".

[0059] Furthermore, the room division setting unit 113 identifies the average intake temperature of the area (intake) that has changed the most from the default temperature (initial room temperature) when each air conditioner 4 is operated sequentially at maximum power control (other air conditioners 4 are at default control) for one turn starting from the default temperature (initial room temperature). Then, the room division setting unit 113 extracts areas (intake) where the temperature has changed by more than m times the average intake temperature of the identified area, and checks which of the following patterns it falls into: (Pattern 1) The area affected by each air conditioner 4 overlaps with that of at least one other air conditioner 4. In this case, no room division is performed. (Pattern 2) The area affected by each air conditioner 4 does not overlap with that of any other air conditioner 4. In this case, room division is performed in the areas that do not overlap. If it falls under (Pattern 2), it is determined whether the number of room divisions exceeds the upper limit number of room divisions D, and if it does, the server area (the "placement control area" in Figure 23) is divided within the range of the upper limit number of room divisions D. If the number of room divisions exceeds the maximum number of room divisions D, a predetermined division rule is established, for example, by dividing the rooms in order from the server area with the lowest identification number within the range of the maximum number of room divisions D.

[0060] Specifically, the room division setting unit 113 selects one target air conditioner 4 (referred to as the "target air conditioner") and sets it as follows: Initial temperature: default temperature (initial room temperature) Measurement time: length of one turn (1 hour) X is set as the lowest temperature (discharge setting temperature) that the air conditioner 4 can set, and the other settings are controlled as shown in Figure 3. Note that the lowest temperature X that the air conditioner 4 can set varies depending on the air conditioner 4. For air conditioners 4 other than the one selected (non-target air conditioners), the room division setting unit 113 sets them as follows: Initial temperature: default temperature (initial room temperature) Measurement time: length of one turn (1 hour) The non-target air conditioners are then operated with default control (control at the default temperature (initial room temperature)).

[0061] The measurement results are shown in Figure 5. Here, there are three air conditioners 4 (No. 1, 2, 3), and the number of zones is "10". Of these zones, the zones indicated by even numbers are the intake zones that are the subject of the measurement results, and the zones indicated by odd numbers are the exhaust zones. The room division setting unit 113 compares the average intake temperature of each zone (intake) at the start and end of the measurement and identifies the intake zone where the temperature has dropped the most. Then, the room division setting unit 113 checks the zones where the temperature has dropped by m times (in this case, 0.5 times) or more of the average intake temperature of the zone where the temperature has dropped the most.

[0062] In the air conditioner 4 (No. 1) indicated by reference numeral 51 in Figure 5, the lowest average intake temperature was in area 2 at "-1.10 ℃" (reference numeral α). １ ) And the value obtained by multiplying that value by m (0.5) is "-0.55" and the temperature has dropped by more than that is "-0.55" in area 4 (symbol α ２ ) In the air conditioner 4 (No. 2) shown by reference numeral 52 in Figure 5, the lowest average intake temperature was in area 4 at "-1.37 °C" (reference numeral β). １ ) And the value obtained by multiplying that value by m (0.5) is "-0.685", and the temperature has dropped by more than that is "-1.26" (symbol β) in area 2. ２ ) and the "-1.01" (symbol β) of area 6 ３). In the air conditioner 4 (No.3) indicated by reference numeral 53 in FIG. 5, the average intake air temperature dropped the most in area 8, which was "-1.31 °C" (reference numeral γ １ ). And the temperature that has dropped below the value obtained by multiplying that value by m (0.5 times), which is "-0.655", is "-0.93" in area 10 (reference numeral γ ２ ).

[0063] From the above measurement results, based on the above (Pattern 1), the room division setting unit 113 does not divide the rooms for air conditioner No.1 and air conditioner No.2 because there are areas 2 and 4 where the influence ranges of air conditioner No.1 overlap. Also, based on (Pattern 2), the room division setting unit 113 divides the rooms in the non-overlapping areas because the influence range of air conditioner No.3 does not overlap with those of the other air conditioners No.1 and No.2. As a result, the room division setting unit 113 divides room 500 into two rooms, with the areas of air conditioners No.1 and 2 being room "1" and the area of air conditioner No.3 being room "2". Note that the room division setting unit 113 confirms that the number of rooms after division is within the range of the upper limit number D of room division (here, "2") and performs the setting of room division.

[0064] Returning to FIG. 1, the initial room temperature level determination unit 114 determines the initial room temperature level based on the upper limit temperature, which is the limit temperature that does not exceed the room temperature threshold condition, and the lower limit temperature when each air conditioner 4 is forcibly controlled. This initial room temperature level determination unit 114 sets the average intake air temperature at the end of the cycle when controlling at the limit set temperature that does not exceed the room temperature threshold condition by increasing the set temperature of the default control of each air conditioner 4 by 1 °C from the default temperature (initial room temperature) as the upper limit temperature. This upper limit temperature is hereinafter defined as "default temperature <high temperature Ver>" (high temperature default temperature). Also, the initial room temperature level determination unit 114 sets the average intake air temperature at the end of the cycle when each air conditioner 4 is forcibly controlled from the default temperature (initial room temperature) as the lower limit temperature. Then, the initial room temperature level determination unit 114 divides the upper limit temperature and the lower limit temperature by the specified number of divisions (for example, "3") to determine the initial room temperature level.

[0065] First, the process of determining the upper limit temperature by the room initial temperature level determination unit 114 will be explained. The room initial temperature level determination unit 114 performs default control for all air conditioners 4. Initial temperature: Default temperature (initial room temperature) Measurement time: Length of one turn (1 hour) Here, only the first temperature setting is under default control, and from the second time onward, the temperature (instantaneous value) of the temperature sensor and the average temperature of the air intake are measured and it is determined whether or not they exceed the threshold. [Determination result 1]: If the temperature (instantaneous value) of the temperature sensor is below the threshold (temperature sensor threshold) and the average temperature of the air intake is below the threshold (average temperature threshold of the air intake), the set temperature is set to "+1℃". [Determination result 2]: If the temperature sensor threshold (instantaneous value) or the average temperature threshold of the air intake is exceeded, after forced cooling (forced control), the average temperature of the air intake at the end of the measurement when the room set temperature was "-1℃" (previous measurement) of the measurement that exceeded the threshold is determined as the upper limit value (upper limit temperature).

[0066] Figures 6, labeled 61 and 62, show the measurement results for determining the upper limit when the temperature sensor threshold (instantaneous value) is 31.5°C and the average intake temperature threshold is 28°C. Figure 6, labeled 61, shows an example where the initial set temperature is the default temperature (initial room temperature) of 26°C, measurement started at 11:00, and was checked at 11:30 during the first turn. At 11:30, the temperature in each intake area (areas 2, 4, 6, 8, 10) does not exceed the temperature sensor threshold (instantaneous value). Also, the average intake temperature is "22.402°C", which does not exceed the average intake temperature threshold of "28°C", and is normal. Therefore, measurement is continued.

[0067] In Figure 6, reference numeral 62 indicates the measurement result at 12:00, the end of the measurement. At the end of the measurement, the temperature in each intake port area (areas 2, 4, 6, 8, 10) did not exceed the temperature sensor threshold (instantaneous value). Also, the average intake port temperature was "26.334 °C," which did not exceed the average intake port temperature threshold of "28 °C." Therefore, based on the above [Judgment Result 1], the set temperature is set to "+1 °C."

[0068] In Figure 7, reference numeral 71 indicates an example where the set temperature was increased by 1°C to 27°C, measurement started at 11:00, and was checked at 11:30, midway through one turn. At 11:30, the temperature in each intake port area (areas 2, 4, 6, 8, 10) did not exceed the temperature sensor threshold (instantaneous value). Also, the average intake port temperature was "24.402°C," which did not exceed the intake port average temperature threshold of "28°C," and was normal. Therefore, measurement was continued.

[0069] In Figure 7, reference numeral 72 indicates the measurement result at 12:00, the end time of the measurement. Here, the temperature of the intake port area 10 is "32.1°C" (reference numeral a), which exceeds the temperature sensor threshold (instantaneous value) of "31.5°C". The average intake port temperature is "27.686°C", which does not exceed the intake port average temperature threshold of "28°C". In this case, since area 10 exceeds the temperature sensor threshold (instantaneous value), forced cooling is performed based on the above [Judgment Result 2]. The room initial temperature level determination unit 114 then sets the intake port average temperature at the end of the previous measurement (set temperature 26°C), which is "26.334°C", as the upper limit (reference numeral b in Figure 6).

[0070] The room's initial temperature level determination unit 114 has an upper limit of 26.334°C, but to allow for temperature adjustment, a range of ±X°C is provided, so the upper limit of the initial temperature level is 26.334°C ± X°C. Furthermore, the upper limit of the initial temperature level, 26.334°C ± X°C, is set as the default temperature <High Temperature Version>.

[0071] Next, the process for determining the lower limit temperature by the room initial temperature level determination unit 114 will be explained. The room initial temperature level determination unit 114 performs default control for all air conditioners 4. Initial temperature: Initial room temperature (default temperature <high ​​temperature version>) Measurement time: Length of one turn (1 hour) Here, similar to the control shown in Figure 3, the outlet set temperature is set to the lowest temperature X that the air conditioner 4 can set, and the indoor set temperature is set to "X + (temperature difference) (8°C)".

[0072] Figure 8 shows the measurement results for determining the lower limit. As indicated by reference numeral 81 in Figure 8, the average intake temperature was "26.334 °C" at 11:00 when the measurement started. Then, as indicated by reference numeral 82 in Figure 8, the average intake temperature was "21.428 °C" (reference numeral c) at 12:00 when the measurement ended. This average intake temperature of "21.428 °C" at the end of the measurement is set as the lower limit. The room initial temperature level determination unit 114 sets the lower limit to 21.428 °C, but to allow for temperature adjustment, a range of ±X °C is provided, so the lower limit of the room initial temperature level is 21.428 ± X °C.

[0073] The room's initial temperature level determination unit 114 then divides the upper and lower temperature limits into a specified number of divisions (for example, "3") and determines the room's initial temperature level.

[0074] Returning to Figure 1, the air conditioning control level determination unit 115, in the process of determining the upper limit value by the room initial temperature level determination unit 114, sets the set temperature (indoor set temperature) when the default temperature <High Temperature Ver> state is created as the upper limit control level for air conditioning control. Also, in determining the lower limit value, the air conditioning control level determination unit 115 sets the set temperature (indoor set temperature) when the strongest control is executed, that is, the lowest value of the discharge set temperature that the air conditioner 4 can set + (temperature difference) (8°C) as the lower limit control level for air conditioning control. Then, the air conditioning control level determination unit 115 divides the upper limit control level and the lower limit control level into a predetermined number of divisions (for example, "3") and determines the air conditioning control level. For example, as shown in Figure 9, the air conditioning control level determination unit 115 determines three air conditioning control levels.

[0075] Returning to Figure 1, the power efficiency group determination unit 116 determines power efficiency groups, which are divided according to the level of power efficiency, such as a group of power-efficient air conditioners 4, a group of power-inefficient air conditioners 4 (and an intermediate group). The same air conditioning control is applied to air conditioners 4 belonging to the same power efficiency group. Therefore, by the power efficiency group determination unit 116 grouping based on power efficiency, the amount of computation related to the control patterns of the air conditioners 4 can be reduced compared to calculating individual control patterns for all of the air conditioners 4.

[0076] The power efficiency group determination unit 116 performs the following power efficiency group determination process: <S1> Determine the default temperature <High Temperature Ver> (default temperature for high temperature). <S2> From the default temperature <High Temperature Ver>, measure the power consumption of each air conditioner 4 when the strongest control is executed for one turn for all air conditioners 4. This power consumption is taken as the default power consumption. <S3> Set only one of the air conditioners to the strongest control and the others to the default temperature <High Temperature Ver> state, and calculate the difference (power consumption difference) between the power consumption of the selected air conditioner (target air conditioner) and the default power consumption of that target air conditioner. <S4> Determine the power efficiency from the power consumption difference and the temperature difference in the intake area due to the strongest control, and determine the power efficiency ranking of each air conditioner 4. Then, determine the power efficiency group of the air conditioner 4 based on the determined power efficiency ranking and the number of power efficiency groups set in advance. A detailed explanation follows below.

[0077] <S1> The power efficiency group determination unit 116 first determines the default temperature <High Temperature Ver.>. Here, it is assumed that the default temperature <High Temperature Ver.> is determined to be 26°C using the same method as the room initial temperature level determination unit 114.

[0078] <S2> Next, the power efficiency group determination unit 116 measures the power consumption of each air conditioner 4 when the strongest control is executed for one turn for all air conditioners 4, starting from the default temperature <High Temperature Ver>. Here, the settings for each air conditioner 4 are as follows: Initial temperature: Initial room temperature (default temperature <High Temperature Ver>) Measurement time: Length of one turn (1 hour) Let X be the lowest temperature that the air conditioner 4 can set (discharge setting temperature), and let the room setting temperature be X + (temperature difference) (8°C).

[0079] As a result of this maximum control measurement, as shown in Figure 10, when there are three air conditioners 4 (No. 1, No. 2, No. 3), the power efficiency group determination unit 116 measures the default power consumption of each air conditioner 4 and calculates the total power consumption.

[0080] <S3> Next, the power efficiency group determination unit 116 selects one air conditioner 4 from all the air conditioners 4 as the target air conditioner 4 and sets it as follows: Initial temperature: Initial room temperature (default temperature <high ​​temperature version>) Measurement time: Length of one turn (1 hour) X is set as the lowest temperature that the air conditioner 4 can set (discharge setting temperature), and the room setting temperature is set to X + (temperature difference) (8°C). The other settings are controlled as shown in Figure 3. The power efficiency group determination unit 116 sets the other air conditioners 4 (non-target air conditioners) as follows: Initial temperature: Initial room temperature (default temperature <high ​​temperature version>) Measurement time: Length of one turn (1 hour) The non-target air conditioners are then controlled with default control <high ​​temperature version>, where the initial room temperature is the default temperature <high ​​temperature version>.

[0081] Through this measurement, the power efficiency group determination unit 116 obtains, for example, the measurement results of the average intake temperature of each area at the end of the cycle as shown in Figure 11 for air conditioner 4 "No. 1". Here, the power efficiency group determination unit 116 calculates the temperature difference between the average intake temperature at the start and the end of one cycle for each intake area (areas 2, 4, 6, 8, 10). Furthermore, the power efficiency group determination unit 116 calculates the sum of the temperature differences for each intake area (here, "0.92") (symbol d). The power efficiency group determination unit 116 calculates this sum of temperature differences for all air conditioners 4 "No. 1", "No. 2", and "No. 3".

[0082] Furthermore, the power efficiency group determination unit 116 measures the power consumption of the target air conditioner and calculates the difference from the default power consumption as the power consumption difference, as shown in Figure 12. The power efficiency group determination unit 116 performs this process for each air conditioner 4 as the target air conditioner.

[0083] <S4> Next, the power efficiency group determination unit 116 calculates the power efficiency for each air conditioner 4 and determines the power efficiency ranking of each air conditioner 4. Here, power efficiency is defined as shown in the following equation (5): Power efficiency = temperature difference (°C) / power consumption difference (kW) ...Equation (5) Note that the temperature difference in this equation (5) is the sum of the temperature differences in each air intake area shown in Figure 11. Furthermore, the smaller the value of this power efficiency, the better the efficiency.

[0084] Figure 13 shows the results of calculating the power efficiency for each air conditioner 4 using equation (5), based on the temperature difference (total of intake ports) and the power consumption difference. Based on these power efficiency calculations, the air conditioners 4 are ranked in the order of "No. 1", "No. 2", and "No. 3" in descending order of power efficiency value.

[0085] Furthermore, the power efficiency group determination unit 116 groups the air conditioners 4 according to a predetermined power efficiency group classification using the temperature difference and power consumption difference of each air conditioner 4. Figure 14 is a table showing the classification of power efficiency. The air conditioners are classified into three ranks, A, B, and C, depending on whether the temperature difference is positive (increased) or negative (decreased), and whether the power consumption difference is positive (increased) or negative (decreased). Rank A corresponds to cases where both the temperature and power consumption difference decrease. Rank B corresponds to cases where the temperature decreases and the power consumption difference increases, and cases where the temperature increases and the power consumption difference decreases. Rank C corresponds to cases where both the temperature increases and the power consumption difference increases.

[0086] As shown in Figure 15, the power efficiency group determination unit 116 determines a power efficiency ranking in which air conditioner "No. 1" belongs to rank B and air conditioners "No. 2" and "No. 3" belong to rank C. In this case, although three power efficiency groups (three ranks) were determined, there were no air conditioners 4 that belonged to rank A, so two power efficiency groups were determined: air conditioner "No. 1" which is more power efficient and rank B, and air conditioners "No. 2" and "No. 3" which are less power efficient and rank C. In this way, the power efficiency group determination unit 116 divides each air conditioner 4 into a predetermined number of power efficiency groups and enables the same air conditioning control to be performed within each power efficiency group.

[0087] As explained above, by having the prerequisite parameter determination unit 110 appropriately determine various parameters, unnecessary learning patterns can be eliminated when calculating the power consumption cost efficiency (E), shortening the learning time, improving the accuracy of optimization, and enhancing the power saving effect.

[0088] The following describes the "cost-efficiency maximum pattern determination process," in which the power consumption control device 10 determines, from among all distribution arrangement patterns, the distribution arrangement pattern that maximizes the overall power consumption cost efficiency of the DC, and the air conditioning control level for that pattern, using the various prerequisite parameters determined by the prerequisite parameter determination unit 110. This process is performed by the room-by-room learning history information generation unit 120, the air conditioning power / temperature function approximation unit 130, the distribution arrangement pattern generation unit 140, and the optimal control calculation unit 150, as shown in Figure 1.

[0089] The room-specific learning history information generation unit 120 acquires the learning history executed according to a predetermined learning pattern from each of the in-room control devices 20 corresponding to each room 500, as set by the room division setting unit 113. Here, the predetermined learning pattern is, for example, the total l × m × n × o pattern, which is obtained by changing the initial room temperature (l), server load (m), server load arrangement pattern (n), and air conditioning control level (o), as shown in Figure 16. At this time, the initial room temperature (l) is set to each of the initial room temperature levels determined by the room-specific initial temperature level determination unit 114. The air conditioning control level (o) is set to the air conditioning control level determined by the air conditioning control level determination unit 115. If the outside temperature is included as a parameter of the predetermined learning pattern, the room-specific learning history information generation unit 120 may add the outside temperature classified into predetermined ranges to the learning history information 550. The room-specific learning history information generation unit 120 acquires information from each room control device 20 as a learning history, including the amount of air conditioning power consumption, the room temperature at the end of the turn, and the room temperature reward pass / fail judgment for that learning pattern, as control results for that learning pattern.

[0090] The room-specific learning history information generation unit 120 then classifies the learning history executed with a predetermined learning pattern according to the initial temperature range, generates learning history information 550 for each room 500, and stores it in the storage unit 260. For example, as the initial temperature range (initial temperature level inside the room), the room-specific initial temperature level determination unit 114 measures an upper limit temperature of "26.334°C" (set to 26°C) and a lower limit temperature of "21.428°C" (set to 21°C). For each range (initial temperatures of 22°C, 24°C, and 26°C in Figure 16) divided by a specified number of divisions (for example, "3"), information on server load, load distribution pattern, and air conditioning control level is stored as a learning dataset, associated with the air conditioning power consumption at that time, the room temperature at the end of the turn, and the room temperature reward pass / fail judgment.

[0091] The room-specific learning history information generation unit 120 transmits, for example, learning history collection instruction information, which includes information on a predetermined learning pattern, to each room-specific control device 20. As a result, the room-specific learning history information generation unit 120 acquires the learning history in each room 500 and generates learning history information 550.

[0092] Returning to Figure 1, the air conditioning power and temperature function approximation unit 130 approximates the air conditioning power function and temperature function for each room 500, for each load arrangement pattern and air conditioning control level (each combination of load arrangement pattern and air conditioning control level). These approximated air conditioning power function and temperature function equations are used in the calculation of power cost efficiency, which will be described later.

[0093] The air conditioning power and temperature function approximation unit 130 defines the air conditioning power function as the amount of air conditioning power consumed relative to the load in room 500 for each room "ri" (where i is a positive integer) set by the room division setting unit 113. The air conditioning power and temperature function approximation unit 130 also defines the temperature function as the temperature at the end of the turn relative to the load in room 500.

[0094] Here, let "A[ri][tj]" be the air conditioning control level for the j-th turn "tj" in room "ri" (where j is a positive integer). Note that the air conditioning control level for turn "tj" in room "ri" is uniquely determined, for example, "A(1)" (air conditioning control level "1" as shown in Figure 9). Also, let "p[ri][tj]" be the load distribution pattern for turn "tj" in room "ri".

[0095] The air conditioning power / temperature function approximation unit 130 defines the air conditioning power function as the air conditioning power consumption for the load (server load) allocated to room 500, and approximates it using the learning history information 550 of room 500. Figure 17 shows an example where the load allocation pattern is "p[r1][t1]", and in air conditioning control when a certain initial temperature in a room and a certain air conditioning control level are selected, the corresponding air conditioning power consumption is obtained by acquiring learning data for loads of "60kW", "80kW", and "100kW" from the learning history information of the range that includes the initial temperature in the room. As a result, the equation indicated by the symbol e in Figure 17 can be obtained as an approximated equation of the air conditioning power function.

[0096] The air conditioning power / temperature function approximation unit 130 defines the temperature function as the temperature after the end of a turn for the load amount distributed to the room 500, and approximates the temperature function using the learning history information 550 of the room 500. Figure 18 shows an example where the load distribution pattern is "p[r1][t1]", and in air conditioning control when a certain initial temperature and air conditioning control level are selected, the learning data for load amounts of "60kW", "80kW", and "100kW" is obtained from the learning history information of the range that includes the initial temperature in the room, thereby obtaining the corresponding temperature after the end of a turn. As a result, the equation indicated by the sign f in Figure 18 can be obtained as an approximated equation of the temperature function.

[0097] As described above, the distribution and placement pattern generation unit 140 defines a predetermined time interval for updating the air conditioning control and load placement within the room 500 as one turn, and sets the time determined by the turn length determination unit 112 (for example, one hour) as this one turn. The distribution and placement pattern generation unit 140 then changes the load distribution amount to each room 500 for the total load amount of DC 1000 in each turn, and further changes the load distribution amount to each room 500 according to the placement ratio for each area (placement control area 30) within the room 500, thereby generating a "room-specific load distribution amount / load placement pattern" (distribution and placement pattern).

[0098] Here, let "X[tj]" be the amount of load flowing into DC1000 in each turn. At this time, the constraint is Σx[ri][tj] = X[tj]. Note that "x[ri][tj]" is the amount of load in room "ri" when it is turn "tj". For example, when given "X[t1] = 50" (total load amount in turn "1": 50), the distribution pattern generation unit 140 generates load distribution patterns for turn "1", such as a pattern that distributes 0 kW to room "1" and 50 kW to room "2" (x[r1][t1] = 0, x[r2][t1] = 50), or a pattern that distributes 10 kW to room "1" and 40 kW to room "2" (x[r1][t1] = 10, x[r2][t1] = 40).

[0099] Furthermore, the distribution pattern generation unit 140 generates load distribution patterns for each area (distribution control area 30) within the room 500, such as "25%-25%-25%-25%", "50%-50%-0%-0%", and "0%-0%-50%-50%" (see Figure 16).

[0100] Returning to Figure 1, the optimal control calculation unit 150 calculates the power consumption cost efficiency (E) of the entire DC1000 over N turns by changing the pattern of the air conditioning control level in each distribution arrangement pattern. The optimal control calculation unit 150 then calculates the distribution arrangement pattern and the air conditioning control level for each room 500 as the optimal solution that maximizes the power consumption cost efficiency (E) of the DC1000 as a whole. This optimal control calculation unit 150 includes a power cost efficiency calculation unit 151 and a control pattern determination unit 152.

[0101] The power cost efficiency calculation unit 151 calculates the power consumption cost efficiency (E) of the entire DC 1000 over N turns when the air conditioning control level pattern is changed for each room 500 in each distribution arrangement pattern. The power cost efficiency calculation unit 151 may, when calculating the power consumption cost efficiency (E) of the entire DC 1000, exclude data from the learning dataset stored in the learning history information 550 where the room temperature reward pass / fail judgment was "failed". By doing so, the power cost efficiency calculation unit 151 can speed up the calculation of power consumption cost efficiency (E).

[0102] Specifically, the power cost efficiency calculation unit 151 uses the above-described formula (1) to calculate the power consumption cost efficiency (E) for each air conditioning control level pattern. In calculating this power consumption cost efficiency (E), the power cost efficiency calculation unit 151 fixes (selects) one of the generated distribution patterns. In this way, the amount of distributed load and the load distribution pattern to each room for each turn can be uniquely determined. Then, the power cost efficiency calculation unit 151 calculates the power consumption cost efficiency (E) when the air conditioning control level pattern is changed for each room 500 in the fixed (selected) load distribution pattern. The power cost efficiency calculation unit 151 sequentially fixes (selects) each load distribution pattern and calculates the power consumption cost efficiency (E).

[0103] The control pattern determination unit 152 determines Eall = min(E{p}) which maximizes the power consumption cost efficiency (E) for the entire DC1000 among the total load distribution patterns calculated by the power cost efficiency calculation unit 151. In other words, the control pattern determination unit 152 further determines the optimal distribution pattern and air conditioning control level for each room 500 which maximizes the power consumption cost efficiency (E) for the entire DC1000 from among the power consumption cost efficiency (E) that maximizes the cost efficiency calculated for each selected distribution pattern. Then, the control pattern determination unit 152 sets the load distribution in each room 500 (load distribution pattern between rooms, load distribution pattern within rooms) based on the determined optimal distribution pattern and air conditioning control level, and executes air conditioning control at the determined air conditioning control level for each room 500.

[0104] <In-Room Control Device> Next, the in-room control device 20 will be described. This in-room control device 20 is provided in correspondence with each room 500 and has the same configuration.

[0105] Figure 19 is a functional block diagram showing an example configuration of the in-room control device 20 according to this embodiment. When the in-room control device 20 receives a learning history collection instruction from the power consumption control device 10, it changes the initial temperature, load amount (server load amount allocated to the room 500), load distribution pattern, and air conditioning control level to control the server 3 and air conditioner 4, collects predetermined air conditioning cooling learning data (room temperature, air conditioning power consumption, etc.) as learning history and transmits it to the power consumption control device 10. The in-room control device 20 also causes the server 3 and air conditioner 4 in the room 500 to perform control based on the load amount and load distribution pattern determined by the power consumption control device 10 that maximizes power consumption cost efficiency, and control based on the determined air conditioning control level. The in-room control device 20 also controls the air conditioner 4 based on instructions from the prerequisite parameter determination unit 110 of the power consumption control device 10, and transmits the power consumption of the air conditioner 4 as a control result, the temperature measured by the temperature sensor, etc., to the power consumption control device 10.

[0106] This in-room control device 20 is configured with a computer comprising a control unit 200, an input / output unit 250, and a storage unit 260.

[0107] The input / output unit 250 performs input and output of information between the power consumption control device 10 and various devices within the room 500 (each server 3 and each air conditioner 4). This input / output unit 250 consists of a communication interface that transmits and receives information via a communication line, and an input / output interface that performs input and output of information between it and input devices such as a keyboard (not shown) and output devices such as a monitor.

[0108] The storage unit 260 is composed of a hard disk, flash memory, RAM (Random Access Memory), etc. This storage unit 260 temporarily stores programs for executing the various functions of the control unit 200, as well as information necessary for the processing of the control unit 200.

[0109] The control unit 200 oversees all processes performed by the in-room control device 20 and, as shown in Figure 19, includes a temperature measurement unit 210, a learning history collection unit 220, an in-room server control unit 230, and an in-room air conditioning control unit 240.

[0110] The temperature measurement unit 210 measures the room temperature using temperature sensors installed in the room 500. The room temperature is pre-set as the "average intake temperature," which is the average temperature of the area on the intake side of the room 500 (for example, the sensor temperature of the intake port of the server 3 in the arrangement control area 30 in Figure 23, or the sensor temperature of the air conditioning control area <intake port side> 40).

[0111] The temperature measurement unit 210 measures a predetermined room temperature (for example, the average temperature of the air intake) as the "initial room temperature" when determining the prerequisite parameters and when starting to collect the learning history. It also measures the room temperature (average temperature of the air intake) at the end of the collection of the learning history for a predetermined time (a predetermined turn) as the "room temperature at the end of the turn".

[0112] When the learning history collection unit 220 receives learning history collection instruction information from the power consumption control device 10, it collects information on the amount of air conditioning power consumed per turn when controlled based on a learning pattern that changes the initial temperature in the room, the server load (the amount of server load allocated to its own room), the load placement pattern, and the air conditioning control level, as well as the room temperature at the end of the turn (average temperature of the air intake) and the room temperature reward pass / fail judgment.

[0113] Here, "turn" is set to the length of one turn (for example, one hour) determined by the turn length determination unit 112 of the power consumption control device 10. Also, the air conditioning control level is set to one of three air conditioning levels, for example, as shown in Figure 9, determined by the air conditioning control level determination unit 115 of the power consumption control device 10.

[0114] The learning history collection unit 220 controls a total of l × m × n × o patterns (a predetermined learning pattern) in its own room 500 by changing the initial room temperature (l), server load (m), server load arrangement pattern (n), and air conditioning control level (o), and acquires information on the air conditioning power consumption, the room temperature at the end of the turn, and the room temperature reward pass / fail judgment as learning history (see Figure 16).

[0115] The learning history collection unit 220 acquires information on the amount of air conditioning power consumed during a turn, as a result of executing a predetermined learning pattern, from the air conditioning power measurement unit 242, which will be described later. The learning history collection unit 220 also acquires information on the room temperature at the end of the turn (average temperature of the air intake) from the temperature measurement unit 210.

[0116] The learning history collection unit 220 transmits the learning history, which was executed according to a predetermined learning pattern, to the power consumption control device 10, along with the identification information of the room control device 20 itself. When generating this predetermined learning history, the learning history collection unit 220 also instructs the room server control unit 230 and the room air conditioning control unit 240 to perform load placement and air conditioning control within the room 500.

[0117] Returning to Figure 19, the in-room server control unit 230, during the collection of learning history (learning phase) and operation (operation phase), causes the server 3 to execute control based on the load placement pattern for the placement control area 30 (Figure 23) set up in the room 500, and also measures the server power consumption for each turn. This in-room server control unit 230 includes a load placement pattern setting unit 231 and a server power measurement unit 232.

[0118] The load placement pattern setting unit 231 sets a placement pattern in which the server load amount set by the learning history collection unit 220 is placed in each placement control area 30 (Figure 23) within the room 500. Then, according to the set placement pattern, the load placement pattern setting unit 231 generates and places virtual resources (VMs, containers, etc.) in each placement control area 30 and executes the load processing set during the collection of the learning history. Furthermore, during operation (operational phase), the load placement pattern setting unit 231 sets a placement pattern for the server groups in each placement control area 30 (Figure 23) within the room 500 that is shown as a pattern that maximizes the power consumption cost efficiency of the entire DC, as determined by the power consumption control device 10.

[0119] The server power measurement unit 232 measures the amount of server power (server power consumption) when each server 3 performs processing in the load placement pattern setting unit 231.

[0120] The in-room air conditioning control unit 240 causes the air conditioner 4 to perform air conditioning control at a preset air conditioning control level (set temperature) and measures the amount of air conditioning power consumed. This in-room air conditioning control unit 240 includes an air conditioning control execution unit 241 and an air conditioning power measurement unit 242.

[0121] The air conditioning control execution unit 241 controls each air conditioner 4 in the room 500 at the air conditioning control level set by the prerequisite parameter determination unit 110 and the learning history collection unit 220 of the power consumption control device 10. Furthermore, during operation (operational phase), the air conditioning control execution unit 241 performs air conditioning control of each air conditioner 4 in the room 500 at the air conditioning control level determined by the power consumption control device 10, which is a pattern that maximizes the power consumption cost efficiency of the entire DC.

[0122] The air conditioning power measurement unit 242 measures the amount of air conditioning power consumed by the air conditioner 4 in the room 500 at each turn during the determination of prerequisite parameters, collection of learning history, and operation.

[0123] <Hardware Configuration> The power consumption control device 10 and the in-room control device 20 according to this embodiment are implemented by a computer 900 having a configuration such as that shown in Figure 20. Figure 20 is a hardware configuration diagram showing an example of a computer 900 that implements the functions of the power consumption control device 10 and the in-room control device 20 according to this embodiment. The computer 900 has a CPU (Central Processing Unit) 901, ROM (Read Only Memory) 902, RAM 903, HDD (Hard Disk Drive) 904, input / output I / F (Interface) 905, communication I / F 906, and media I / F 907.

[0124] The CPU 901 operates based on programs stored in the ROM 902 or HDD 904 and is controlled by the control unit. The ROM 902 stores boot programs executed by the CPU 901 when the computer 900 starts up, as well as programs related to the computer 900's hardware.

[0125] The CPU 901 controls input devices 910, such as a mouse or keyboard, and output devices 911, such as a display or printer, via the input / output interface 905. The CPU 901 acquires data from the input devices 910 and outputs the generated data to the output devices 911 via the input / output interface 905. In addition to the CPU 901, a GPU (Graphics Processing Unit) or the like may also be used as a processor.

[0126] The HDD 904 stores programs executed by the CPU 901 and data used by those programs. The communication I / F 906 receives data from other devices via a communication network (e.g., NW (Network) 920) and outputs it to the CPU 901, and also transmits data generated by the CPU 901 to other devices via the communication network.

[0127] The media interface 907 reads a program or data stored in the recording medium 912 and outputs it to the CPU 901 via the RAM 903. The CPU 901 loads the program related to the desired processing from the recording medium 912 onto the RAM 903 via the media interface 907 and executes the loaded program. The recording medium 912 can be an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase Change Rewritable Disk), a magneto-optical recording medium such as an MO (Magneto Optical Disk), a magnetic recording medium, or a semiconductor memory.

[0128] For example, when computer 900 functions as the power consumption control device 10 and the in-room control device 20 of the present invention, the CPU 901 of computer 900 realizes the functions of the power consumption control device 10 and the in-room control device 20 by executing a program loaded on RAM 903. The HDD 904 stores the data in RAM 903. The CPU 901 reads and executes a program related to the desired processing from the recording medium 912. Alternatively, the CPU 901 may read a program related to the desired processing from another device via a communication network (NW 920).

[0129] <Effects> The effects of the power consumption control device 10, etc. according to the present invention will be described below. The power consumption control device according to the present invention is a power consumption control device 10 that is communicated to a room control device 20 for each room 500 that controls a plurality of servers 3 and a plurality of air conditioners 4 in a room 500 in a data center 1000. The power consumption control device 10 generates learning data by controlling the air conditioning control level of the air conditioners 4, the load distribution pattern between rooms 500, and the placement pattern of the load distributed to each room 500 to the servers 3 via the room control device 20 for each room 500, for each turn indicating a predetermined control time, thereby maximizing the power consumption cost efficiency of the entire data center 1000. The air conditioning control level of the air conditioner 4 is determined in a distribution arrangement pattern that indicates turns, and the power consumption control device 10 is characterized by having a prerequisite parameter determination unit 110 that determines at least one of the following parameters as prerequisites for calculating power consumption cost efficiency: (1) the length of one turn, (2) the number of room divisions when the area affected by the air conditioning of the air conditioner 4 to the server 3 is considered a room, (3) the initial room temperature level indicating the setting stage of the initial temperature inside the room, (4) the air conditioning control level indicating the control stage of the air conditioning control of the air conditioner 4, and (5) the air conditioning power efficiency group indicating a group of air conditioners that perform control at the same air conditioning control level.

[0130] In this way, the prerequisite parameter determination unit 110 of the power consumption control device 10 can determine the parameters (prerequisite parameters) that are assumed when calculating power consumption cost efficiency. Specifically, the power consumption control device 10 sets each parameter, such as the length of one turn, the number of room divisions, the initial temperature level inside the room, the air conditioning control level, and the air conditioning power efficiency group, to the optimal prerequisite values ​​according to the applied environment. As a result, the power consumption control device 10 can eliminate learning patterns that do not affect the derivation of the optimal solution for power consumption cost efficiency, thereby shortening the learning time and improving the optimization accuracy to enhance the power saving effect.

[0131] Furthermore, the prerequisite parameter determination unit 110 of the power consumption control device 10 is characterized by comprising: an initial room temperature determination unit 111 that determines the average intake temperature of multiple servers 3 when default control is performed to indicate a predetermined state of air conditioning control for the air conditioner 4 as the default temperature; and a turn length determination unit 112 that, when the prerequisite parameter determination unit 110 determines the length of one turn, operates each air conditioner 4 in the strongest control, which indicates control at the lowest temperature that each air conditioner 4 can set from the default temperature state, and determines the length of one turn based on the time until the average of the difference in each intake area falls below a predetermined temperature, using the difference between the intake temperature measured at predetermined time intervals and the intake temperature measured previously.

[0132] In this way, the power consumption control device 10 can determine the optimal length of one turn according to the applicable environment. Therefore, the power consumption control device 10 can improve the optimization accuracy and enhance the power saving effect.

[0133] Furthermore, the prerequisite parameter determination unit 110 of the power consumption control device 10 determines the maximum number of room divisions by dividing the learning time set for generating learning data to calculate power consumption cost efficiency by a value obtained by multiplying the number of items required for acquiring learning data by the length of one turn, and also selects one unit for each air conditioner 4 in order, performs default control at the default temperature for the air conditioners 4 other than the selected unit for one turn, and performs maximum control from the default temperature to the lowest temperature for the selected air conditioner 4 for one turn, and determines whether or not to divide the room based on whether or not the area affected by the air conditioning of each air conditioner 4 overlaps with the area affected by the other air conditioners 4, and if it is determined that to divide the room, the room division setting unit 113 is provided to set the number of room divisions to be less than or equal to the maximum number of room divisions.

[0134] In this way, the power consumption control device 10 can determine the upper limit of the number of room divisions and then divide the rooms 500 of the data center 1000 into an optimal number of rooms that match the applicable environment. Therefore, the power consumption control device 10 can suppress the increase in the number of learning patterns and improve the power saving effect by improving the optimization accuracy.

[0135] Furthermore, the prerequisite parameter determination unit 110 of the power consumption control device 10 is characterized by having a power efficiency group determination unit 116 which, when determining the air conditioning power efficiency group, sets the default control setting temperature of each air conditioner 4 to a limit setting temperature that does not exceed the room temperature threshold condition by raising the default control setting temperature by 1 degree from the default temperature state and setting the average intake port temperature at the end of the turn as the high default temperature, measures the power consumption when the strongest control at the lowest temperature is executed for one turn from the high default temperature and sets this as the default power consumption of each air conditioner 4, selects one of the air conditioners 4, performs default control at the high default temperature for the air conditioners 4 other than the selected air conditioner 4, measures the power consumption when the strongest control at the lowest temperature is executed for the selected air conditioner 4 and measures the temperature difference in the intake port area of ​​each server 3 for one turn, calculates the power efficiency using the power consumption difference obtained by subtracting the measured power consumption from the default power consumption and the temperature difference, and groups each air conditioner 4 according to a predetermined power efficiency group classification.

[0136] In this way, the power consumption control device 10 can classify each air conditioner 4 into a predetermined power efficiency group and set the same air conditioning control for the same power efficiency group. Therefore, the power consumption control device 10 can suppress an increase in the number of learned patterns.

[0137] It should be noted that the present invention is not limited to the embodiments described above, and many modifications are possible within the technical concept of the present invention by those with ordinary skill in the art.

[0138] 1 Power Consumption Control System 3 Server 4 Air Conditioner 10 Power Consumption Control Device 20 In-Room Control Device 30 Placement Control Area 40 Air Conditioning Control Area 100, 200 Control Unit 110 Prerequisite Parameter Determination Unit 111 Initial Room Temperature Determination Unit 112 Turn Length Determination Unit 113 Room Division Setting Unit 114 In-Room Initial Temperature Level Determination Unit 115 Air Conditioning Control Level Determination Unit 116 Power Efficiency Group Determination Unit 120 Room-Specific Learning History Information Generation Unit 130 Air Conditioning Power / Temperature Function Approximation Unit 140 Distribution Placement Pattern Generation Unit 150 Optimal Control Calculation Unit 151 Power Cost Efficiency Calculation Unit 152 Control Pattern Determination Unit 160, 250 Input / Output Unit 170, 260 Storage Unit 210 Temperature Measurement Unit 220 Learning History Collection Unit 230 In-Room Server Control Unit 231 Load placement pattern setting unit 232 Server power measurement unit 240 Room air conditioning control unit 241 Air conditioning control execution unit 242 Air conditioning power measurement unit 500 Room 550 Learning history information 1000 Data center (DC)

Claims

1. A power consumption control device that is communicated to a room-specific control device that controls multiple servers and multiple air conditioners in a room within a data center, wherein the power consumption control device generates learning data by controlling the air conditioning control level of the air conditioners, the load distribution pattern between the rooms, and the distribution pattern of the loads distributed to each room to the servers via the room-specific control device, for each turn representing a predetermined control time, and determines the air conditioning control level of the air conditioners in a distribution and distribution pattern that represents the distribution and distribution pattern that maximizes the power consumption cost efficiency of the entire data center, wherein the power consumption control device uses the following parameters as prerequisites for calculating the power consumption cost efficiency: A power consumption control device characterized by comprising a prerequisite parameter determination unit that determines at least one of the following: (1) the length of one turn; (2) the number of room divisions when the area affected by the air conditioning of the air conditioner is defined as a room; (3) the initial room temperature level indicating the setting stage of the initial temperature inside the room; (4) the air conditioning control level indicating the control stage of the air conditioning control of the air conditioner; and (5) the air conditioning power efficiency group indicating a group of air conditioners that perform the same air conditioning control level.

2. The power control device according to claim 1, wherein the premise parameter determination unit comprises: an initial room temperature determination unit that determines the average intake temperature of a plurality of servers when default control is performed to indicate a predetermined state of air conditioning control for the air conditioner as the default temperature; and a turn length determination unit that, when the premise parameter determination unit determines the length of one turn, operates each air conditioner in the strongest control, which indicates control at the lowest temperature that each air conditioner can set, from the default temperature state, and determines the length of one turn based on the time until the average of the differences in the area of ​​each intake is below a predetermined temperature, using the difference between the intake temperature measured at predetermined time intervals and the intake temperature measured previously.

3. The power consumption control device according to claim 2, wherein the prerequisite parameter determination unit, when determining the number of room divisions, determines the upper limit of room divisions by dividing the learning time set for generating the learning data for calculating the power consumption cost efficiency by a value obtained by multiplying the number of items required for acquiring the learning data by the length of one turn, and sequentially selects one air conditioner for each air conditioner, performs default control at the default temperature for one turn for the air conditioners other than the selected air conditioner, and performs maximum control from the default temperature to the minimum temperature for one turn for the selected air conditioner, and determines whether or not to divide the room based on whether or not the area affected by the air conditioning of each air conditioner overlaps with the area affected by other air conditioners, and if it is determined that to divide the room, the room division setting unit sets the number of room divisions to be less than or equal to the upper limit of room divisions.

4. The power control device according to claim 3, wherein the premise parameter determination unit, when determining the air conditioning power efficiency group, sets the default control setting temperature of each air conditioner to a limit setting temperature that does not exceed the room temperature threshold condition by raising the default control setting temperature of each air conditioner by 1 degree from the default temperature state, and sets the average intake temperature at the end of the turn as the high default temperature; for each air conditioner, measures the power consumption when the strongest control at the lowest temperature is executed for one turn from the high default temperature, and sets this as the default power consumption of each air conditioner; selects one of the air conditioners, performs default control at the high default temperature for the air conditioners other than the selected air conditioner, measures the power consumption when the strongest control at the lowest temperature is executed for the selected air conditioner and measures the temperature difference in the intake area of ​​each server for one turn; calculates the power efficiency using the power consumption difference obtained by subtracting the measured power consumption from the default power consumption and the temperature difference, and groups each air conditioner according to a predetermined power efficiency group classification.

Images