JP2013134550A - Linkage control system and linkage control method for storage battery and it apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a linkage control system which estimates the characteristics (time, frequency, interval, and the like) of a future power failure in advance, and controls a no-break power supply having an IT apparatus and a storage battery in an interconnected manner for the supposed power failure, achieving a high charge-discharge efficiency of the storage battery.SOLUTION: A linkage control system comprises a commercial power supply 10, a voltage logger 20 which measures the voltage of the commercial power supply, a no-break power supply 30 to which a plurality of storage batteries 32 are connected, a power supply tap 40 having a power measuring mechanism, and a plurality of IT apparatuses 50 including a server 51 connected to the power supply tap. The voltage logger, the no-break power supply, the power supply tap having a power measuring mechanism, the server, and the IT apparatuses are connected to a network. A control part 55 in the server estimates the duration time of a future power failure, obtains the power limit value of the IT apparatus from the residual quantity of the storage battery and the estimated duration time of the power failure, so as to control the power of the IT apparatus within the power limit value. Furthermore, the control part 55 in the server estimates the interval of power failures so as to control the charge mode of the storage battery.

Description

  The present invention relates to a system including an IT device and a large-capacity storage battery, and more particularly to an IT device-storage battery link control system suitable for efficiently using the storage battery and operating the IT device stably for a long time. .

  In areas where power grid infrastructure is insufficient, power outages are frequent. Therefore, there is a method in which a generator is placed in the building, the generator is operated during a power failure, and the power in the building is supplied. The server room is further provided with a lead-acid battery array for several hours necessary for repair of the generator in preparation for a failure of the generator, and supplies power to the server and the client PC. The storage battery array generally takes the form of expanding the capacity of a lead storage battery of an uninterruptible power supply (UPS). The storage battery array is positioned as a backup for the generator, and since the storage battery array is not frequently used, an inexpensive lead storage battery is used, but its volume and weight are large.

  On the other hand, although the power grid infrastructure is in place, there are areas where the demand for power increases rapidly due to the use of air conditioners in the summer, and there is a shortage of power. In this case, there is less damage than a sudden power outage, and a planned power outage is carried out with a rotating power supply at a fixed time every day. Although the generator is operated at the place where the generator is placed in the building, the IT equipment must be stopped at the place where the generator is not provided. On the other hand, there are places where lithium ion storage battery arrays are placed in order to continue business services. Lithium-ion batteries can be charged quickly and have a large number of charge / discharge cycles (cycle performance), so they can handle planned power outages. As described above, the storage battery array is used as a means for supplying power at the time of a power failure, although it takes different forms depending on the characteristics of the power failure and the facility of the building.

  By the way, in the case of a planned power outage, the power company defines the time and the demand side can grasp the power outage. However, it is desirable that the demand side can grasp the power outage even in the case of an unplanned power outage. An example of a method for predicting the occurrence of an unplanned power outage is disclosed in Patent Document 1. This document discloses a technique for predicting the risk of a power failure, starting a power generator in advance of the power failure, and switching from a commercial power source, thereby reducing the capacity of the storage battery connected to the UPS or eliminating the need for a UPS. The prediction here is to determine that the commercial power source is “high risk of power failure within a predetermined time”, and is calculated from weather information obtained by connecting to the Internet and a predetermined condition. . That is, the prediction in Patent Document 1 predicts the time when a power failure occurs.

JP 2004-355219 A

  In patent document 1, since the purpose was to start the generator before the power failure, it was sufficient to predict the time when the power failure occurred. As a result of a global survey of server rooms and offices where IT devices (routers, switches, servers, storages, etc.) are placed, it has been found that there are many locations with storage battery arrays attached to IT devices. Therefore, in the present invention, attention has been paid to the efficient use of the storage battery array. High-efficiency use includes, for example, extending the discharge time of the storage battery, reducing the discharge capacity margin, and reducing unnecessary rapid charging. In the case of a lead storage battery array, it is possible to reduce battery cost and weight by reducing the amount of battery. In the case of a lithium ion storage battery array, it can cope with frequent power outages and can be used as an alternative to the generator, thereby reducing the cost of the diesel fuel cost of the generator several times the electricity cost. Thus, there are various advantages by using the storage battery array with high efficiency.

In order to increase the efficiency of the storage battery array, it is useful to predict power outages in advance. For example, if the interval between power failures is known, if charging is performed at that interval, useless rapid charging can be prevented. At this time, the item to be predicted by the power failure is different from the power failure occurrence time in Patent Document 1. Also,
The efficiency of the storage battery is expected to be higher by controlling the IT device and the UPS in cooperation with each other than the UPS having the storage battery.

  Accordingly, an object of the present invention is to estimate the characteristics (time, frequency, interval, etc.) of a power failure that should come in advance, and to control the storage battery by linking an IT device and a UPS having a storage battery in association with the assumed power failure. It is providing the system which implement | achieves the highly efficient charging / discharging. Efficiently controlling the discharge of the storage battery increases the apparent capacity and improves the IT performance.

  The above objects and novel features of the present invention will become apparent from the description of the present specification and the accompanying drawings.

  In order to solve the above-described problems, the present invention employs, for example, the configurations described in the claims.

  If an example of the linkage control system of the IT equipment and storage battery of the present invention is given, an uninterruptible power supply connected to the commercial power supply, a voltage logger for measuring the voltage of the commercial power supply, and connected to a plurality of storage batteries The voltage logger, the uninterruptible power supply device, comprising: a power supply device; a power strip with a power measurement mechanism connected to the uninterruptible power supply device; and an IT device including a server connected to the power tap with the power measurement mechanism The power measuring mechanism-equipped power strip and the server are connected to the network, and the control unit inside the server predicts the duration of the power outage, and determines the IT equipment's capacity based on the predicted power outage duration and the remaining capacity of the storage battery. A power limit value is obtained, and the power of the IT device is controlled so as to satisfy the power limit value.

  Moreover, if an example of the linkage control method of the IT device and the storage battery according to the present invention is given, a commercial power source, a voltage logger for measuring the voltage of the commercial power source, and the commercial power source are connected to a plurality of storage batteries. The voltage logger, the uninterruptible power supply comprises an uninterruptible power supply, an IT device including a power tap with a power measurement mechanism connected to the uninterruptible power supply, and a server connected to the power tap with the power measurement mechanism. The power supply device, the power measuring mechanism-equipped power strip, and the server are linked control methods for a system connected to a network, the step of predicting the duration of a power outage, the predicted duration of power outage and the remaining storage battery A step of obtaining a power limit value of the IT device from the amount, and a step of controlling the power of the IT device so as to satisfy the power limit value.

  Estimate the characteristics (time, frequency, interval, etc.) of power outages for frequent power outages, and perform linkage control of IT equipment and storage batteries according to the power outage characteristics, so that IT equipment services can be continued and performance can be reduced as much as possible. While suppressing, the storage battery array can be charged and discharged efficiently. Thereby, the battery capacity of a storage battery can be reduced. From another perspective, the performance of IT equipment can be improved with the same storage battery capacity. Furthermore, by controlling the charging method of the storage battery array, it contributes to extending the life of the storage battery.

It is a figure which shows an example of a structure of the system for performing linkage control of IT apparatus and a storage battery. It is a figure which shows an example of the network structure of multiple bases containing the system of this invention. It is a figure which shows the operation example of the charge scheduling by power failure interval prediction. It is a figure which shows the operation example of the discharge scheduling with a large performance limitation by power failure time prediction. It is a figure which shows the operation example of the discharge scheduling with a small performance limitation by power failure time prediction. It is a figure which shows the example of an operation | movement by the power failure time at the time of power failure several times, and space | interval prediction. It is a figure which shows an example of the table of the prediction item of a power failure. It is a figure which shows the example of coping operation | movement when a power failure prediction comes off. It is a figure which shows an example of the control part which carries out linkage control of IT apparatus and a storage battery. It is a figure which shows an example of the power failure prediction part in a control part. It is a figure which shows an example of the IT electric power budget part in a control part. It is a figure which shows an example of the discharge characteristic of a storage battery. It is a figure which shows an example of the IT electric power control part in a control part. It is a figure which shows an example of the operation | movement of the low power control of IT apparatus. It is a figure which shows an example of the flowchart which carries out linkage control of IT apparatus and a storage battery.

  DESCRIPTION OF EMBODIMENTS Embodiments for carrying out the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same names and reference symbols throughout the drawings for describing the embodiments for carrying out the invention, and the repetitive description thereof will be omitted.

  FIG. 1 shows a main part of a system for performing linkage control between an IT device and a storage battery according to an example of the present invention. The connection indicates the power supply system. Electric power is supplied from a power plant by a commercial power supply CPWR10. A voltage logger VLTLOG20 is placed at a location where the voltage of the commercial power supply can be measured. The voltage logger 20 can record the voltage in time series or when an event occurs, and can measure a power failure in addition to swell, dip, instantaneous interruption, and the like according to the voltage value. An uninterruptible power supply UPS 30 is connected to the commercial power supply 10. There are various types of UPS 30 such as a constant commercial method, a line interactive method, and a constant inverter method, but there is no particular limitation. Here, a constant inverter system is described. The UPS 30 includes a rectifier, a converter CNV31 that converts AC to DC, a storage battery array BAT32, a storage battery control unit CHGCTL33 that controls the storage battery array, an inverter INV34 that converts DC to AC, and a control unit UCTL35 of UPS. The UPS control unit 35 has functions such as monitoring of the remaining capacity of the storage battery, network support, a shutdown instruction to the IT device, and charge control according to the normal or rapid charge mode. A power supply tap PDU 40 with a power measurement mechanism is connected to the UPS 30. The power tap PDU 40 with a power measuring mechanism can measure a current value and a voltage value at each outlet, and has a mechanism for remotely monitoring power and controlling on / off of power supply through a network connection. Each IT device 50 is connected to the PDU 40. For example, the servers SVR1 (51), SVR2 (52), and the router RTR53 are used. However, a modem, a firewall, a switch, and a storage may be connected. The server SVR1 (51) has a built-in control unit CTL55 that controls the IT device of the present invention and the storage battery in association with each other.

  FIG. 2 shows a network configuration diagram of a plurality of bases including the system according to the first embodiment of the present invention. Connections indicate network signals. The IT device / storage battery linkage control system (Sys1) in FIG. 1 is connected to other systems (Sys2, Sys3) via a WAN (Wide Area Network). In the system (Sys1), a router (RTR) 53 connects an external WAN and an internal LAN (Local Area Network) 60, a plurality of servers (SVR1, SVR2), a UPS 30, and a power tap with power measurement (PDU) 40. A LAN 60 is formed by being connected to a router (RTR) 53. Information on the UPS, PDU, router, voltage logger (VLTLOG), and temperature / humidity logger (TMPLOG) can be communicated with the servers (SVR1, SVR2) via the LAN 60. This means that the server can monitor and control the internal state of other devices using the LAN. For example, voltage for a voltage logger, temperature, humidity for a temperature / humidity sensor, UPS remaining battery level, storage battery terminal voltage, UPS mode setting, power value of each IT device with power measurement function power tap, server, router And low power control of IT equipment.

  The feature of the present invention is that the linkage control between the IT device and the storage battery is performed in accordance with the predicted characteristics (time, frequency, interval, etc.) of the power failure. Therefore, first, an example will be given to show what effect and utility can be obtained by predicting a power failure. It also serves as an explanation of operation examples of IT equipment and storage batteries. The characteristics of power outages include power outage intervals / frequency and duration of power outages.

  It shows that charge scheduling of a storage battery array is attained by predicting a power failure interval. FIG. 3 shows an example of charge scheduling based on prediction of power failure intervals. When an IT device is operated using a storage battery during a power failure (time t1-t2), the storage battery is discharged, and the battery capacity becomes substantially zero at time t2. If the prediction is not performed, it is not possible to know when the next power failure will occur, so it is necessary to quickly charge and prepare for the next power failure, and the charging is completed at time t3. On the other hand, when making a prediction, the next power failure starts at time t4, and there is sufficient time between power failures, so it can be seen that low-speed charging may be performed. In the case of no prediction, excessively rapid charging will unduly shorten the life of the lead battery in particular. In addition, in the next power failure, when prediction is not performed after the power failure ends at time t5, if low-speed charging is performed, if a power failure occurs at time t6, charging becomes insufficient, and the storage battery is depleted at time t7. , IT equipment services cannot be continued. When the prediction is performed, the charging of the battery can be completed by time t6 by rapid charging. In this way, by predicting the power failure interval, it is possible to schedule charging, and it is possible to avoid insufficient charging due to useless rapid charging or low-speed charging.

  Next, the effect of predicting the duration of a power outage will be shown. In IT services, it is desirable to continue the service even during a power outage and to have as high a performance as possible. However, high performance means that a large amount of storage battery is consumed, and the capacity of the storage battery can be depleted. There is a trade-off between performance and battery capacity. Therefore, it will be shown how the performance is improved by predicting the duration of the power failure according to the present invention. By predicting the power outage time, it is possible to estimate the time for which the storage battery is used, so that it is possible to schedule discharge. If you don't predict it, you don't know how long the power outage will last. It is possible to take. FIG. 4 shows an example when the battery usage limit is large, and FIG. 5 shows an example when the battery usage limit is small. In FIG. 4, a power failure occurs between times t1 and t2. When there was no prediction of the power outage time, the battery usage limit was large, so the IT service could be continued during the power outage, but the performance was limited to prf1. When the power failure time is predicted, since the battery usage time is known in advance, it is possible to improve the performance up to prf2 by judging including the remaining amount of the storage battery. This is because the margin of the remaining amount of the storage battery can be reduced because the power failure time is grasped. In FIG. 5, a power failure occurs between times t3 and t5. When the power failure time was not predicted, although the performance was as high as prf4, the battery was used up, so the battery was depleted at time t4 during the power failure. Therefore, the power of the IT device is zero between time t4 and t5, and the IT service has to be stopped. The result was undesirable for a server that was supposed to operate 24 hours a day. When the power failure time is predicted, the battery usage time is known in advance, so the performance was suppressed by prf3, but the storage battery could be used during the power failure time. Thus, by predicting the power failure time, it is possible to schedule battery discharge and improve the performance of the IT equipment.

  Furthermore, a new effect when predicting the duration and interval of a power failure is shown. 4 and 5 are examples when a power outage occurs once a day. It is important to predict both power outage time and power outage interval when planning how to use batteries when power outages occur multiple times a day. You can plan how to use the battery multiple times and when to charge it. FIG. 6 shows an example when multiple power outages occur. There have been two power outages. When prediction was not performed, the performance of prf7 was limited in the first power failure at time t6-t7, and although the battery capacity decreased from C3 to C1, the IT service could be continued. In the second power outage from time t8 to t10, if the performance restriction at prf7 is the same as that at the first time, the battery is depleted at time t9 and the IT service cannot be continued. When predicting the power outage time and the power outage interval, there are two methods depending on the type of battery. One is a method of estimating the total time of a plurality of power outages and determining the storage battery usage for each power outage so that the IT service for that time can be continued. This method is suitable for lead-acid batteries that require time for charging. The other is a method of estimating IT power outage time and power outage intervals, performing quick charging between power outages, and recovering battery capacity to improve IT service continuity and performance. This is a method suitable for a lithium-ion battery with little life deterioration during rapid charging. In the first method, by limiting the performance to prf6, the storage battery can be used even during two power outages, and the service can be continued. The battery capacity decreases from C3 to C2 at the first power failure, and becomes zero from C2 at the next power failure. In the second method, the performance is set to prf7 higher than prf6, and the battery capacity is recovered by performing rapid charging between power outages, so that both continuity of service and high performance can be achieved. The battery capacity is reduced from C3 to C1 due to a power failure at time t6-t7, but rapid charging is performed by time t8 of the next power failure, and the battery capacity is restored to C3 to prepare for a power failure from time t8.

  In this way, by predicting and estimating the duration of a power outage and the interval between power outages, it is possible to perform optimal scheduling of battery charging / discharging, and accordingly, service continuity and performance improvement of IT equipment There is an advantage that can be achieved. Based on this, in Example 1 of the present invention, an example of items predicted by a power failure is shown in FIG. Although the management unit is one day, it is not particularly limited. Here, the table is divided into summary and details. The summary includes the presence or absence of power outages, the number of power outages per day, the longest power outage time, the total power outage time, and the shortest and longest time between power outages. As detailed data, the start time and end time of the power outage are listed as to when the power outage occurs in 24 hours. A method may be used in which a table is provided for each minute and the presence or absence of a power failure is described. The power outage interval needs to be calculated across management units (1 day). On March 1, it is predicted that a power outage of 9: 00-12: 00, 16: 00-18: 00, 20: 00-21: 00: 00 will occur, the number of power outages will be 3 times, and the maximum power outage time will be 3 The total power outage time is 6 hours, the power outage interval is a minimum of 2 hours, and a maximum of 13 hours. The longest time between power outages is 13 hours from 21:00 on March 1 to 10:00 on March 2.

  The characteristics of power outages (time, frequency, interval, etc.) are expected to change depending on the location and day of the week.

  Regarding the prediction of power outage, the prediction may be out of order due to the estimation. There are cases where the power failure does not occur and there are cases where the power failure time and interval are different from the actual ones. For this reason, this patent describes how to deal with the case where the prediction is wrong. The linkage control between the IT device and the storage battery based on the prediction of the power failure is feedforward control for preventing the battery remaining amount from running out. When the prediction is lost, the control is switched to feedback control that is determined from the remaining battery level. An example is shown in FIG. In order to make the figure easy to see, the performance between prf1 and prf2 and between the battery capacities C1 and C2 are shown compressed. Although the time t1-t2 is estimated as the power failure time, it is actually a case where a power failure occurs at time t1-t5 which is longer than t1-t2. If it is longer than the predicted time, feedback is made based on the actual measurement of the remaining battery level. If there is no measure against time extension, the battery runs out at time t3, and the IT device cannot be operated during time t3-t5. Therefore, if the predicted planning time is exceeded, the performance limit of the IT device is tightened based on the actual measurement of the remaining battery level and the power value of the IT device. The frequency of actual measurement is increased as the remaining amount decreases. The intervals between times t2-t3, t3-t4, and t4-t5 are shortened in order. In this way, a measure is taken to cope with a case where prediction is lost by a method of feeding back the measured value of the remaining battery level to the performance limit so that the battery capacity has during the power failure.

  As mentioned above, it showed about the advantage of predicting a power failure, the operation of IT apparatus and a storage battery, the example of linkage control, and the control example of the countermeasure when a power failure prediction comes off. Next, the detailed structure of the linkage control unit between the IT device and the storage battery will be described based on the operation example of the cooperative control between the IT device and the storage battery.

  FIG. 9 shows the configuration of a control unit CTL 55 that controls the linkage of the IT device and the storage battery according to the first embodiment of the present invention. The control unit CTL 55 operates on the server SVR. The control unit 55 may be configured by using hardware such as a microcomputer or may be configured by software. It has an interface with a voltage monitor, predicts the time and frequency of a power outage, makes a power outage plan, and actually has an interface with a power outage prediction unit POE (Power Outages Estimation) 70 and UPS 30 showing power outage information about whether or not a power outage is present. It has an interface with the IT power budget unit (IPB) 80 and the power tap PDU 40 for determining the power limit value of the IT device from the predicted power outage time and the remaining battery capacity of the UPS. An IT power control unit IPC (IT low-Power Control) 90 that performs low power control so as to satisfy the power limit value is configured. An outline of the operation of the control unit CTL 55 is as follows. The power limit value of the IT device is determined from the remaining battery power so that the power failure time and interval are predicted and the IT device can be operated for the estimated power failure time. Then, low power control of the IT device is performed so that the power limit value of the IT device is set.

  Next, the configuration of each functional block will be described. FIG. 10 shows the configuration of the power failure prediction unit POE70. The power outage prediction unit POE outputs the actual power outage information indicating whether or not a power outage has occurred and ends using the measured value of the voltage logger 20, and the data extraction unit stores the voltage logger voltage data. Statistical analysis of the measured voltage data makes a power outage plan. The data extraction unit 71 performs data shaping and extraction on the voltage data of the voltage logger VLTLOG20 and date / time data acquired from the OS. You may extract in a time series, and you may extract according to the event of a voltage fluctuation. The data extracted by the data extraction unit 71 is stored in the data storage unit 72 in a time series format with the date and time added. The statistical analysis unit 73 statistically analyzes the time series data stored in the data storage unit 72 to generate a time series model. Time series models include, but are not limited to, moving average (MA), autoregressive model (AR), moving average autoregressive model (ARMA), autoregressive conditional variance heterogeneous model (GAARCH), and the like. The statistical analysis unit 73 updates the time series model with daily data. For time series data, general statistical processing such as simply taking the average of every Sunday as a prediction of Sunday may be performed. Also, accuracy may be improved by combining machine learning. The power outage calculation unit 74 sets up a power outage plan that predicts information on the frequency, time, and interval of the current outage at a fixed time set by a timer, for example, 0:00, saves the plan data, and the IT power budget unit 80 is notified.

  FIG. 11 shows the configuration of the IT power budget unit IPB 80 of the first embodiment of the present invention. The IT power budget unit 80 determines the IT power limit value, which is the limit power value of the IT equipment, based on the power outage plan of the power outage prediction unit 70 and the storage battery remaining amount of the UPS 30, and the interval between power outages predicted in the power outage plan. It has a function of determining the charging mode. The battery usage planning unit 82 is used to charge the battery in which mode and in what mode by using the remaining power and voltage of the storage battery from the UPS 30, the power failure plan from the power failure prediction unit POE 70, the actual power failure information and the specifications 81 of the storage battery array as inputs. The IT power limit value is calculated from the battery charging plan that determines the above, the remaining battery level, and the scheduled power outage time. The charging mode is notified to the UPS 30 according to the battery charging plan. When the power failure occurs, the remaining battery capacity of the UPS 30 is measured regularly or irregularly with a timer, and the IT power limit value is calculated from the power failure duration and the remaining battery capacity indicated in the power failure plan each time.

  Here, a method for calculating the IT power limit value will be described. In order to continuously operate the IT device during the power failure time estimated by the power failure prediction unit POE 70, it is necessary to determine how much current is supplied to the IT device 50 based on the battery remaining amount and the terminal voltage of the UPS 30. For this purpose, it is necessary to calculate from the discharge characteristics of the storage battery. FIG. 12 shows an example of the discharge characteristics of the storage battery. For simplicity, a single 12V storage battery is used. The storage battery array may be considered in parallel or in series as appropriate. When the rated capacity is C, the discharge time of the storage battery varies depending on the magnitude of the discharge current CA. It is 40 minutes at 1 CA, but 20 hours at 0.05 CA. Therefore, the discharge current may be determined within a range that satisfies the estimated power failure time. This is the IT power limit value. For example, in order to have a power failure time of 2 hours, the discharge current may be set to 0.25 CA or less.

  Next, FIG. 13 shows the configuration of the IT power control unit IPC90. The IT power control unit IPC 90 receives the power data of each IT device acquired by the power tap with power measurement PDU 40 and the IT power limit value calculated by the IT power budget unit IPB 80 as input, and performs low power control of each IT device.

  FIG. 14 shows an operation example in which low power control is performed on the IT device with respect to the IT power limit value during the power failure. Low power control is performed so as to satisfy the IT power limit value calculated from the remaining amount of storage battery and the estimated power failure time. The IT power limit value is reviewed periodically so that the remaining battery power is exhausted at the remaining power failure prediction time. The IT power limit value is set using the actual measured power value.

  The above is the configuration of the control unit CTL 55 that links and controls the IT device and the storage battery. Now, how to perform overall control using the control software constituting the control unit will be described with reference to a flowchart. FIG. 15 shows a flowchart. At a fixed time (S1), for example, at 0:00, the power outage time of the day, the frequency of power outage, and the interval are predicted (S2). When a power failure occurs (S3), matching with the power failure estimated on a regular basis is performed, and the duration and interval of the power failure are read out. The battery remaining amount is measured (S4), and the limit value of IT power is set from the duration of power failure (S5). The IT power limit value is compared (S10) with the minimum value (complete power limit) in which the IT device consumes the least power in a state where the power is completely limited. If the IT power limit value is large, IT low power control is performed. I do. The IT power limit value is compared with the IT power actual measurement value (S11). If the IT power limit value is large, the IT low power control does not limit the power because the budget is achieved. When the IT power limit value is small, the budget is not achieved, and thus low power control is performed so that the IT power becomes equal to or less than the IT power limit value (S12). As an exception process, the minimum power value of the IT device is compared with the IT power limit value. If the minimum value of the IT device is large (S10), the battery capacity is insufficient and the IT service cannot be continued. S13), shutdown is performed (S14). The timer interrupt (S6) periodically measures the remaining battery power (S7), and resets the IT power limit value (S5), thereby improving the IT performance while using the battery to the maximum during a power failure. Can be improved. When power is restored (S8), the low power control of IT during a power failure is canceled (S9), and according to the predicted power failure interval, a charging mode is set for rapid charging when the power failure interval is short and normal charging when it is long (S15). ) And continue processing. This is the operation in the flowchart.

  If the IT power limit value is set while periodically measuring the remaining battery power and low power control is required, the loop of S6 → S7 → S5 → S8 → S10 → S11 → S12 → S6 When unnecessary, a loop of S6 → S7 → S5 → S8 → S10 → S11 → S6 is formed. The loop trigger is a periodic measurement of the remaining battery level. As described above, it is possible to predict the power failure time, and to perform low power control of the IT device so as to satisfy the remaining battery power so that the performance degradation of the IT device is as small as possible and the service can be continued.

10 Commercial power supply CPWR
20 Voltage logger VLTLOG
30 Uninterruptible power supply UPS
31 Converter CNV
32 Battery array BAT
33 Storage battery controller CHGCTL
34 Inverter INV
35 UPS control unit UCTL
40 Power tap PDU with power measurement mechanism
50 IT equipment 51 Server SVR1
52 Server SVR2
53 router RTR
55 Server Control Unit CTL
60 LAN
70 Power Failure Prediction Department POE
71 Data Extraction Unit 72 Data Accumulation Unit 73 Statistical Analysis Unit 74 Power Outage Calculation Unit 80 IT Power Budget Unit IPB
81 Storage battery specification 82 Battery usage planning unit 90 IT power control unit IPC
91 Low power controller

Claims (12)

Commercial power,
A voltage logger for measuring the voltage of the commercial power supply;
An uninterruptible power supply connected to the commercial power supply and connected to a plurality of storage batteries;
A power strip with a power measuring mechanism connected to the uninterruptible power supply;
It is composed of IT equipment including a server connected to the power measuring mechanism-equipped power strip,
The voltage logger, the uninterruptible power supply, the power measuring mechanism-equipped power strip, and the server are connected to a network,
The control unit inside the server predicts the duration of the power outage, obtains the power limit value of the IT device from the predicted duration of the power outage and the remaining amount of the storage battery, and sets the power of the IT device to satisfy the power limit value. An IT device and storage battery linkage control system characterized by control. In the linkage control system of the IT device and the storage battery according to claim 1,
The control unit measures the remaining amount of the storage battery regularly or irregularly, obtains the power limit value of the IT device from the predicted power outage duration and the measured remaining amount of the storage battery each time, and sets the power limit value An IT device and storage battery linkage control system, wherein power of the IT device is controlled to satisfy In the linkage control system of the IT device and the storage battery according to claim 1,
The controller is
A power failure prediction unit that accumulates voltage data of the voltage logger and predicts a duration of a power failure based on the accumulated voltage data;
An IT power budget unit for obtaining a power limit value of an IT device from the predicted duration of power outage and the measured remaining battery capacity;
An IT device and a storage battery comprising: an IT power control unit that performs low power control based on power data of the IT device acquired by the power tap with the power measuring mechanism and a power limit value of the IT device. Linkage control system. In the linkage control system of the IT device and the storage battery according to claim 3,
The power failure prediction unit statistically analyzes the accumulated time series data to generate a time series model, and the IT device and storage battery linkage control system. In the linkage control system of the IT device and the storage battery according to claim 1,
When the power failure time is longer than the predicted power failure time, the control unit performs feedback control of the performance limitation of the IT device based on the measured value of the remaining capacity of the storage battery. Control system. In the linkage control system of the IT device and the storage battery according to claim 1,
The control unit further predicts an interval between power outages, and controls a charging mode of the storage battery based on the predicted power outage interval. In the linkage control system of the IT device and the storage battery according to claim 6,
The control unit accumulates voltage data of the voltage logger, and predicts a power failure interval based on the accumulated voltage data. A commercial power supply, a voltage logger for measuring the voltage of the commercial power supply, an uninterruptible power supply connected to the commercial power supply and connected to a plurality of storage batteries, and a power supply with a power measuring mechanism connected to the uninterruptible power supply It comprises an IT device including a tap and a server connected to the power measuring mechanism-equipped power strip, and the voltage logger, the uninterruptible power supply, the power measuring mechanism-equipped power tap, and the server are connected to a network. A system linkage control method comprising:
Predicting the duration of a power outage;
Obtaining a power limit value of the IT device from the predicted power outage duration and the remaining amount of storage battery;
And a method for controlling the power of the IT device so as to satisfy a power limit value. In the linkage control method of the IT device and the storage battery according to claim 8,
Measuring the remaining amount of the storage battery regularly or irregularly;
A step of obtaining a power limit value of the IT device from the predicted duration of the power outage and the measured remaining amount of the storage battery each time;
And a method for controlling the power of the IT device so as to satisfy a power limit value. In the linkage control method of the IT device and the storage battery according to claim 8,
The step of predicting the duration of the power outage stores the voltage data of the voltage logger, and predicts the duration of the power outage based on the accumulated voltage data. In the IT device and storage battery linkage control method according to claim 8, further comprising:
Predicting power outage intervals;
And a step of controlling a charging mode of the storage battery based on the predicted power outage interval. The IT device and storage battery link control method according to claim 11,
The step of predicting the interval between power outages accumulates voltage data of the voltage logger, and predicts the interval between power outages based on the accumulated voltage data.

Images