JP2013066330A - Power usage schedule adjusting device, power usage schedule adjusting method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten time required for determining a schedule of power usage under restrictions of power supply.SOLUTION: The power usage schedule adjusting device calculates a schedule in which a plurality of works using power under restrictions of power supply to which a maximum limit is provided to peak power and power consumption amount. The power usage schedule adjusting device has: an input part to which a request to perform a work and information indicative of the maximum power consumption and duration of the work are input; and a schedule calculation part which sequentially determines whether or not the plurality of works inputted in the input part satisfy the restrictions of power supply and can be performed on the basis of the maximum power consumption and duration of the work, and applies a branch and bound method to reduce combinations of the work to be performed to calculate the schedule in which the work is performed.

Description

  The present invention relates to a power usage schedule adjustment device, a power usage schedule adjustment method, and a program.

Research institutes and universities have schedules for conducting experiments based on the number of people working and the presence or absence of materials. Furthermore, when power supply restrictions are added, it is necessary to determine a schedule such as the date and time for conducting the experiment in consideration of the amount of power used in the experiment. In this way, creating a schedule under constraints is known as a scheduling problem that finds an optimal schedule that can be executed under various constraints in the management of production, delivery, project, factorial work, etc. (patents) References 1, 2).
The scheduling problem is a combination optimization problem, and an optimal solution can be obtained by examining all combinations.

JP 2010-257383 A JP 2011-066541 A

  However, when the total number of combinations of experiments (operations) increases, there is a problem that the time required for examining all combinations of experiments increases exponentially. At this time, if a time limit is provided in determining the schedule of the work to be performed, there is a possibility that the work execution schedule cannot be determined and the work cannot be performed.

  The present invention has been made to solve the above problems, and its object is to suppress an exponential increase in the time required to determine the schedule of work using power under the constraints of power supply, and to improve the schedule. An object of the present invention is to provide a power usage schedule adjustment device, a power usage schedule adjustment method, and a program that can shorten the time required for determination.

  In order to solve the above problem, the power usage schedule adjustment apparatus according to the present invention has a schedule for performing a plurality of operations using power under the restriction of power supply in which upper limits are set for peak power and power consumption. A power usage schedule adjustment device to calculate, an input unit for inputting information indicating a request to perform the operation, duration time and maximum power consumption of the operation, and a plurality of operations input to the input unit In contrast, the task is determined by sequentially determining whether it can be executed by satisfying the power supply constraints from the duration and maximum power consumption of the task, and applying the branch and bound method to reduce the combination of tasks to be performed. And a schedule calculation unit for calculating a schedule for performing the above.

  Further, in the present invention described above, in the above-described invention, when a time zone in which the work is to be performed is provided as a restriction condition in the request to perform the work, the schedule calculation unit Then, it is determined in order whether the power supply constraint and the constraint condition are satisfied or not.

  Also, the power usage schedule adjustment method of the present invention is a power usage schedule adjustment method for calculating a schedule for performing a plurality of operations using power under the restriction of power supply in which an upper limit is provided for peak power and power consumption. A method for adjusting a power usage schedule in an apparatus, wherein an input step for inputting information indicating a request for performing the work, a duration of the work and a maximum power consumption, and a plurality of work input in the input step In contrast, the task is determined by sequentially determining whether it can be executed by satisfying the power supply constraints from the duration and maximum power consumption of the task, and applying the branch and bound method to reduce the combination of tasks to be performed. And a schedule calculating step for calculating a schedule for performing the above.

  Further, the program of the present invention is provided in a power usage schedule adjustment device that calculates a schedule for performing a plurality of operations using power under the restriction of power supply in which upper limits are set for peak power and power consumption. An input step in which information indicating the request to execute the work, the duration of the work and the maximum power consumption is input to the computer, and a plurality of works input in the input step, From the duration and the maximum power consumption, it is sequentially determined whether or not the power supply can be satisfied and executed, and a branch and bound method that reduces the combination of work to be performed is applied to calculate a schedule for performing the work. This is a program for executing a schedule calculation step.

  According to this invention, it is determined in order whether or not each work using power satisfies the power supply constraints from the duration and maximum power consumption of the work, and a search area that is a combination of work to be performed is determined. Reduce according to the judgment result. Thereby, it is possible to determine the schedule of the work to be performed without checking all the combinations of the works, and it is possible to reduce the time required for determining the schedule of the work.

It is a schematic block diagram which shows the structure of the electric power utilization schedule adjustment apparatus 1 in this embodiment. It is an example of the screen which inputs the experiment item in this embodiment. It is a figure which shows an example of the experimental power profile in this embodiment. It is a conceptual diagram which shows the branch and bound method which the schedule calculation part 13 of this embodiment performs. It is a figure which shows the example of an output of the experiment schedule in this embodiment. It is a flowchart which shows the process in which the schedule calculation part 13 calculates an experiment schedule. It is a figure which shows the experiment information which the schedule calculation part 13 produced | generated from the application information memorize | stored in the memory | storage part 12, and an experiment power profile.

  Hereinafter, a power usage schedule adjustment device, a power usage schedule adjustment method, and a program according to an embodiment of the present invention will be described with reference to the drawings. Here, an experiment is taken as an example of work that uses power, and an experiment schedule is planned as a power usage schedule under the constraints of power supply with an upper limit for both peak power (maximum power consumption) and power consumption. The apparatus etc. which perform are demonstrated.

FIG. 1 is a schematic block diagram illustrating a configuration of a power usage schedule adjustment device 1 according to the present embodiment. As illustrated in FIG. 1, the power usage schedule adjustment device 1 includes an input unit 11, a storage unit 12, a schedule calculation unit 13, and a display unit 14.
Application information including information on each experiment item related to the experiment application is input to the input unit 11 in accordance with an operation performed by a person in charge of schedule planning who has received an application for performing an experiment from a user. Here, the experiment items are the experiment name (operating equipment during the experiment), the duration, and the number of experiments. FIG. 2 is an example of a screen for inputting an experiment item in the present embodiment. As shown in the figure, the experiment name, duration, and number of experiments may be selected from predetermined ones.

Returning to FIG. 1, the description of the power usage schedule adjustment device 1 will be continued.
The storage unit 12 stores application information input to the input unit 11. The storage unit 12 stores in advance an experimental power profile in which the peak power in each experiment is associated with the experiment name.
FIG. 3 is a diagram illustrating an example of an experimental power profile in the present embodiment. As shown in the figure, the peak power is associated with the experiment name. For example, a peak power of 150 [kW] is associated with an experiment whose experiment name is a clean room.

Returning to FIG. 1, the description of the power usage schedule adjustment device 1 will be continued.
The schedule calculation unit 13 generates experiment information corresponding to each experiment based on the application information stored in the storage unit 12 and the experiment power profile. For example, the power consumption of an experiment that consumes power A [kW] between the experiment start time t0 and the experiment end time t1 is expressed as the following equation (1).

  In addition, the power consumption of the experiment in which the power consumed during the implementation changes once is expressed as the following equation (2). At this time, time t1 is the time when the power consumption changes, and time t2 is the experiment end time.

In the experiment in which the electric power consumed during the operation changes a plurality of times, the case division increases according to the number of changes.
Further, when there is a restriction on a time zone to be performed for an experiment in which power consumption is expressed using Expression (1), the following expressions (3-1) and (3-2) are established.

The experiment in which the power consumption is expressed by the formulas (3-1) and (3-2) is restricted to be performed in the morning.
Further, the schedule calculation unit 13 formulates a scheduling problem with constraints as the following equations (4) to (6) based on the power consumption Pexpk (t) of each experiment.

Here, Ptotal (t) in equation (4) is the total power of all experiments at time t, and Pbase (t) is the base power (air conditioning equipment and other loads) at time t. As the base power at each time, a statistical value based on past power consumption or the like may be used. In this case, the storage unit 12 may store the base power at each time in advance.
N is the total number of experiments. Pwmax is a power constraint value as an upper limit value of the total peak power, and Pwhmax is a power amount constraint value as an upper limit value of the power consumption. The power constraint value Pwmax and the power amount constraint value Pwhmax are determined in advance as power supply constraints. Tst is the start time of the adjustment period of the experiment schedule, and tfin is the end time of the adjustment period of the experiment schedule.

  At this time, the schedule calculation unit 13 sets the experiment schedule as a problem of maximizing z (the number of experiments to be performed) represented by the following equation (7).

In Expression (7), δk is a variable indicating whether or not the experiment k can be performed (0-1), δk = 0 indicates that the experiment k is not performed, and δk = 1 indicates that the experiment k is performed.
In order to plan an experiment schedule for performing as many experiments as possible, the schedule calculation unit 13 satisfies Formula (4) to (6), and formula (7) indicating the sum total of δk indicating whether each experiment is possible or not. Calculate the combination of experiments to maximize.

The above problem is classified as a mixed integer programming problem. When a search is performed on all 0-1 variables (δk: k = 1,..., N), the search area becomes 2 ⁿ . As the number of experiments to be scheduled increases, the search area increases exponentially, making it difficult to obtain a solution.
Therefore, the schedule calculation unit 13 uses the branch and bound method to plan the experiment schedule while reducing the search area.

FIG. 4 is a conceptual diagram showing a branch and bound method performed by the schedule calculation unit 13 of the present embodiment. The schedule calculation unit 13 determines whether or not each constraint can be satisfied when the experiment 1 corresponding to the 0-1 variable (δ1) that is one of the 0-1 variables (δk) is executed (nodes). N _11). Here, the state at the node N ₁₁ is a state that does not specify whether or not to perform each experiment.

When all the constraints can be satisfied, the schedule calculation unit 13 is the upper bound of the objective function (formula (7)) when the experiment 1 is performed (node N ₂₁ ) and when the experiment 1 is not performed (node N ₂₂ ). Is calculated. For example, when the upper bound of the node N ₂₂ is smaller than the upper bound of the node N ₁₁ , the experiment 1 is selected, and the search for the search region (the node N ₃₃ , the node N ₃₄ ,...) Below the node N ₂₂ is performed. It will not be done.
On the other hand, when there is a constraint that cannot be satisfied, the schedule calculation unit 13 selects not to perform the experiment 1 and continues the search. In this case, the search for the search region (node N ₃₁ , node N ₃₂ ,...) Below node N ₂₁ is not performed.

Also, in FIG. 4, node N ₂₁ is in a state where it is determined that experiment 1 is to be performed and whether or not another experiment is to be performed. The node N ₂₂ is a state where it is determined that the experiment 1 is not performed, and whether or not another experiment is performed is not determined.
The node N ₃₁ is in a state where it is determined that the experiment 1 and the experiment 2 are performed, and whether or not another experiment is performed is not determined. The node N ₃₂ is a state in which the experiment 1 is performed, the experiment 2 is not performed, and whether or not another experiment is performed is not defined. Node N ₃₃ is a state in which Experiment 1 is not performed, Experiment 2 is determined to be performed, and whether or not another experiment is performed is not determined. The node N ₃₄ is in a state where it is determined that the experiment 1 and the experiment 2 are not performed, and whether or not another experiment is performed is not determined.

As described above, the schedule calculation unit 13 determines the experiment schedule while limiting the search area. The schedule calculation unit 13 outputs and displays the calculated experiment schedule on the display unit 14.
The schedule calculation unit 13 stores the calculated experiment schedule in the storage unit 12. The power usage schedule adjustment device 1 may output the experiment schedule stored in the storage unit 12 to the outside in accordance with the operation of the person in charge of schedule planning input to the input unit 11.

  FIG. 5 is a diagram illustrating an output example of an experiment schedule in the present embodiment. In the figure, the horizontal axis indicates time, and the vertical axis indicates power consumption at each time. Moreover, the time when each experiment is carried out is shown by a graph. In the output example shown here, an experiment schedule is made under the restriction of power supply in which the sum of peak powers is 1200 [kW] or less. FIG. 5A is an output example of an experimental schedule that satisfies the power supply constraints calculated by the schedule calculation unit 13. FIG. 5B is an example of an experiment schedule that does not satisfy the constraints of power supply. In the experiment schedule shown in the figure, it is set as a restriction to perform each experiment between 7 o'clock and 21 o'clock.

Hereinafter, the process of calculating the experiment schedule by the power usage schedule adjustment apparatus 1 will be described with reference to FIGS. 6 and 7. FIG. 6 is a flowchart illustrating a process in which the schedule calculation unit 13 calculates an experiment schedule. FIG. 7 is a diagram illustrating the application information stored in the storage unit 12 by the schedule calculation unit 13 and the experimental information generated from the experimental power profile.
In the experimental information shown in FIG. 7, a desired time zone that is a desired time zone for performing each experiment 1 to 7 is added. The desired time zone is input from the outside via the input unit 11 and becomes one of the constraint conditions when calculating the experiment schedule. Each experiment is associated with the experiment name, duration, peak power, and desired time zone using the experiment number as a key. For example, for the experiment with the experiment number of Experiment 3, the shaking table (100%), 2 [h], 800 [kW], and AM (7: 0 to 12:00) are the experiment name, duration, peak The power and the desired time zone are associated with each other. In addition, the experiment numbers are assigned in the order in which they are input to the input unit 11, for example.
Hereinafter, a description will be given of a case where the experiment starts on the hour from 7 o'clock to 20 o'clock.

When the process for calculating the experiment schedule is started, application information is input to the input unit 11 and the application information is stored in the storage unit 12 as shown in FIG. 6 (step S101).
The schedule calculation unit 13 generates experimental information from the application information and the experimental power profile stored in the storage unit 12 (step S102).
The schedule calculation unit 13 initializes a state in which it is not determined whether or not to perform each experiment as an experiment schedule, and stores the experiment schedule in the storage unit 12 (step S103). At this time, the base power is set as the power at each time.

  The schedule calculation unit 13 repeatedly performs the processing of step S104 to step S107 in order from the experiment with the smallest experiment number. The schedule calculation unit 13 reads the experiment schedule stored in the storage unit 12 (step S104), and determines whether or not the read experiment schedule has a start time at which the experiment to be determined can be performed (step S105). ). When a plurality of experiment schedules are stored in the storage unit 12, the schedule calculation unit 13 executes the processes after step S <b> 105 for each experiment schedule.

When the experiment 1 shown in FIG. 7 is a determination target, the schedule calculation unit 13 determines whether or not there is a feasible start time in the desired time zone of the experiment 1. Specifically, it is determined whether or not the experiment 1 having a duration of 4 hours can be started from 7 o'clock or 8 o'clock so as to end in the morning. At this time, it is determined whether or not the constraint condition is satisfied when Experiment 1 is performed. In the schedule calculation unit 13, the peak power obtained by adding 500 [kW] to the power (base power and power used in other experiments) used at 7, 8, 9, 10, and 11 o'clock is a power constraint value. That Pwmax has not been exceeded, and that the amount of power used, including the amount of power (4 [h] × 500 [kW]) used when Experiment 1 was implemented, has not exceeded the power amount constraint value Pwhmax. judge.
When the power constraint value Pwmax and the power amount constraint value Pwhmax are not exceeded, that is, when the power supply constraint is satisfied, there is a start time at which Experiment 1 can be performed. On the other hand, when either or both of the power constraint value Pwmax and the power amount constraint value Pwhmax are exceeded, that is, when the power supply constraint cannot be satisfied, there is no start time at which Experiment 1 can be performed.

  In step S105, when there is a start time that can be performed for the determination target experiment (step S105: YES), the schedule calculation unit 13 calculates the objective function (formula (7)) when the determination target experiment is performed. The upper bound and the upper bound of the objective function when the determination target experiment is not performed are calculated. The schedule calculation unit 13 performs selection to increase the upper bound. That is, the schedule calculation unit 13 performs selection to perform the experiment to be determined, and reduces the search area without searching for an experiment schedule including a combination that does not perform the experiment to be determined.

  The schedule calculation unit 13 generates an experiment schedule for performing the experiment from the start time determined to be feasible, stores the generated experiment schedule in the storage unit 12, and updates the experiment schedule (step S106). At this time, if there are a plurality of start times determined to be feasible, the experiment schedule is updated from each start time to the experiment schedule. That is, a plurality of experiment schedules are generated. When the experiment 1 in FIG. 7 is a determination target, the schedule calculation unit 13 generates an experiment schedule for performing the experiment 1 from 7:00 and an experiment schedule for performing the experiment 1 from 8:00, and stores the generated experiment schedule. The experiment schedule is updated by storing in the unit 12.

On the other hand, in step S105, when there is no start time that can be performed for the determination target experiment (step S105: NO), the schedule calculation unit 13 selects not to perform the determination target experiment and performs the experiment. Do not generate experiment schedules. That is, the schedule calculation unit 13 reduces the search area without searching for an experiment schedule including a combination for performing an experiment to be determined.
The schedule calculation unit 13 stores the generated experiment schedule in the storage unit 12 and updates the experiment schedule (step S107).

The schedule calculation unit 13 repeatedly performs steps S104 to S107 for each experiment, and calculates an experiment schedule that maximizes the objective function (formula (7)). For example, when the experiment information is the experiment information shown in FIG. 7, the experiment schedule for starting Experiment 1, Experiment 3, Experiment 4, and Experiment 6 in the morning is a combination of experiments that maximizes the objective function.
The schedule calculation unit 13 reads from the storage unit 12 an experiment schedule indicating whether or not to perform the experiment, and outputs the read experiment schedule to the display unit 14 (step S108). As described above, the experiment schedule includes information indicating the start time of the experiment to be performed.

As described above, in the power usage schedule adjustment device 1, the schedule calculation unit 13 generates experimental information based on the application information and the experimental power profile. The schedule calculation unit 13 calculates an experiment schedule that satisfies a predetermined constraint by using a branch and bound method, stores it in the storage unit 12, and outputs it to the display unit 14. At this time, the schedule calculation unit 13 determines whether or not the experiment included in the experiment information can be performed by satisfying the power supply constraints in order, thereby reducing the search area and applying for the experiment. The experiment schedule can be calculated without examining all the combinations.
As a result, the power usage schedule adjustment device 1 suppresses an exponential increase in the time required to determine the schedule of work using power under the constraints of power supply, and shortens the time required to determine the schedule. Can do.

  In the present embodiment, the configuration in which the peak power of each experiment is stored in advance in the storage unit 12 as the experiment power profile has been described. However, the application information including the peak power may be input to the input unit 11 without being limited thereto.

  Note that the above-described power usage schedule adjustment apparatus 1 may have a computer system therein. In that case, the process performed by the functional units of the input unit 11, the storage unit 12, the schedule calculation unit 13, and the display unit 14 described above is stored in a computer-readable recording medium in the form of a program. When the computer reads and executes the program, a process for calculating an experiment schedule is performed. Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Alternatively, the computer program may be distributed to the computer via a communication line, and the computer that has received the distribution may execute the program.

  DESCRIPTION OF SYMBOLS 1 ... Power utilization schedule adjustment apparatus, 11 ... Input part, 12 ... Memory | storage part, 13 ... Schedule calculation part, 14 ... Display part

Claims (4)

A power use schedule adjustment device that calculates a schedule for performing a plurality of operations using power under the restriction of power supply in which an upper limit is provided for peak power and power consumption,
An input unit for inputting information indicating a request for performing the work, and a duration and maximum power consumption of the work;
For a plurality of operations input to the input unit, it is sequentially determined whether or not it can be executed by satisfying the power supply constraints from the duration and maximum power consumption of the operation, and the combination of operations to be performed is reduced. A power usage schedule adjustment apparatus comprising: a schedule calculation unit that calculates a schedule for performing the work by applying a branch and bound method. When a time zone for performing the work is provided as a constraint in the request to perform the work,
The power usage schedule according to claim 1, wherein the schedule calculation unit sequentially determines whether or not the plurality of tasks can be executed by satisfying the power supply constraint and the constraint condition. Adjustment device. A power usage schedule adjustment method in a power usage schedule adjustment device that calculates a schedule for performing a plurality of operations that use power under the constraints of power supply in which an upper limit is set for peak power and power consumption,
An input step in which information indicating the request to perform the work, and the duration and maximum power consumption of the work are input;
For a plurality of operations input in the input step, it is determined in order from the duration and maximum power consumption of the operation whether the power supply can be satisfied and executed, and the combination of operations to be performed is reduced. And a schedule calculation step of calculating a schedule for performing the work by applying a branch and bound method. A computer provided in a power usage schedule adjustment device that calculates a schedule for performing a plurality of operations using power under the restriction of power supply in which an upper limit is set for peak power and power consumption,
An input step in which information indicating the request to perform the work, and the duration and maximum power consumption of the work are input;
For a plurality of operations input in the input step, it is determined in order from the duration and maximum power consumption of the operation whether the power supply can be satisfied and executed, and the combination of operations to be performed is reduced. A program for executing a schedule calculation step of calculating a schedule for performing the work by applying a branch and bound method.

Images