JP2017211763A - Information processor, information processing system and information processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processor which corrects a power demand prediction at a more suitable timing with a more suitable correction amount by preparing a setting which is a reference to correct the power demand prediction and changing the setting.SOLUTION: The information processor has a prediction part, a determination part, a calculation part and a correction part. The prediction part predicts a power demand prediction value which is a future power demand value. The determination part determines in a preset deviation monitoring time zone whether to correct the power demand prediction value on the basis of a deviation of the power demand prediction value from an actual power demand value which is a recorded actual power demand value. The calculation part calculates a correction amount of the power demand prediction value when the determination part determines a correction. The correction part corrects the power demand prediction value on the basis of the correction amount.SELECTED DRAWING: Figure 1

Description

Embodiments described herein relate generally to an information processing apparatus, an information processing system, and an information processing program.

In recent years, with the introduction of storage batteries, peak cuts in received power have become possible. In other words, the peak cut means that the storage battery is charged during a time period when the power consumption is low, and the storage battery is discharged during a time period when the power usage reaches a peak, thereby reducing the peak of the power usage. . In addition, since the storage battery is charged during the inexpensive nighttime electricity bill period and discharged during the daytime when the electricity bill is high, the total electricity bill can be reduced.

In storage battery control, power demand is predicted every day, a peak cut line is calculated based on the result, and charge / discharge control of the storage battery is performed. However, if the power demand is deviated, appropriate charge / discharge control cannot be performed.

For example, when the predicted value of the power demand is predicted to be lower than the actual value, sufficient discharge is not performed and the amount of received power is increased despite the remaining power storage amount at the time of power peak. Conversely,
When predicted to be higher than the actual value, the storage battery is discharged ahead of schedule, and the necessary peak cut cannot be performed.

In order to avoid the above problems, correction of power demand prediction reflecting the actual value of the day is performed. However, when using the data of all time zones up to the current time of the day as the method of obtaining the actual value to be reflected in the correction, it is not possible to follow a rapid fluctuation in power demand. In addition, when using only the most recent actual value data for the day, the predicted value will be updated more frequently than necessary even if there is a slight fluctuation in power demand. Is unable to follow.

Japanese Patent Laid-Open No. 2006-19450

The problem to be solved by the embodiments of the present invention is to provide an information processing apparatus capable of appropriately correcting power demand prediction in order to improve the problems of the conventional power demand prediction method.

In order to solve the above problems, the information processing apparatus according to the embodiment includes a prediction unit, a determination unit, a calculation unit, and a correction unit. The prediction unit predicts a power demand prediction value that is a future power demand value. The determination unit determines whether to correct the predicted power demand value based on a difference between the actual power demand value and the predicted power demand value in a preset deviation monitoring time zone. Determine whether. A calculation part calculates the correction amount of the said electric power demand predicted value, when it is judged that the said judgment part correct | amends. The correction unit corrects the power demand prediction value based on the correction amount.

It is a block diagram of the information processing apparatus which concerns on embodiment of this invention. It is a flowchart which shows operation | movement of the information processing apparatus which concerns on embodiment of this invention. It is a flowchart which shows the method of correct | amending the electric power demand prediction of the information processing apparatus which concerns on embodiment of this invention. It is a graph which shows an example of the method of correct | amending the electric power demand prediction of the information processing apparatus which concerns on embodiment of this invention. It is a graph which shows the example which changes with seasons etc. in the method of correct | amending the electric power demand forecast of the information processing apparatus which concerns on embodiment of this invention. It is a graph which shows the example which differs in the method of correct | amending the correction determination retroactive period in the method of correct | amending the electric power demand prediction of the information processing apparatus which concerns on embodiment of this invention. It is a graph which shows the example from which the method of correcting the electric power demand prediction of the information processing apparatus which concerns on embodiment of this invention differs in the method of calculating correction amount. It is a graph which shows the example which correct | amends the electric power demand prediction of the terminal time of correction | amendment target time slot | zones by the method of correct | amending the electric power demand prediction of the information processing apparatus which concerns on embodiment of this invention.

  Hereinafter, embodiments for carrying out the invention will be described with reference to the drawings.

(Embodiment)
(Constitution)
FIG. 1 shows an information processing apparatus according to the embodiment. The information processing apparatus 1 includes an acquisition unit 11, a storage unit 12, an output unit 13, a prediction unit 14, a determination unit 15, a calculation unit 16, a correction unit 17, and a creation unit 18.

The acquisition unit 11 acquires information that affects the power demand. The memory | storage part 12 memorize | stores the past data which is the pattern of the time change of the past power demand, and the weather information of the past day.

The prediction unit 14 includes weather information on the day when the power demand prediction acquired by the acquisition unit 11 is performed, and the storage unit 1.
Based on the past data stored in step 2, the time change of the future power demand is predicted.

The determination unit 15 determines whether or not to correct a power demand prediction that is a temporal change in the predicted future power demand based on the actual value of the day.

Specifically, it is determined whether or not the time for determining the correction of the power demand prediction is a deviation monitoring time zone that is a time zone suitable for correcting the power demand prediction. Further, it is determined whether the error rate of the power demand record with respect to the power demand prediction is larger than a preset error rate threshold. The error rate here is the ratio of the difference between the total value (integrated value) of the power demand forecast and the power demand record.

However, when determining whether or not to correct the power demand prediction, since the determination is based on the divergence between the power demand prediction and the actual power demand, the determination criterion is not the error rate [%], but the divergence power amount [kWh]
] May be used. The calculation unit 16 calculates an error rate of the actual power demand value with respect to the predicted power demand value during a predetermined correction determination retroactive period. Further, when correcting the power demand prediction, a correction amount is calculated. The correction amount is an amount for correcting the power demand prediction in a predetermined time zone. The correction unit 17 corrects the power demand prediction based on the correction amount calculated by the calculation unit 16.

The creation unit 18 creates a charge / discharge plan based on the temporal change in the power demand predicted by the prediction unit 14. A charging / discharging plan is the time and quantity which charge or discharge of the storage battery planned beforehand.

The output unit 13 outputs a power demand prediction predicted by the prediction unit 14, a power demand prediction corrected by the correction unit 17, a charge / discharge plan generated or corrected by the generation unit 18, and the like.

(Function)
FIG. 2 shows a sequence diagram regarding the operation of the information processing apparatus 1 according to the embodiment. First, the prediction unit 14 of the information processing apparatus 1 predicts a temporal change in power demand on the current day or the next day (S1).
1). At that time, the weather information on the day when the power demand prediction is performed, which is acquired by the acquisition unit 11, and the storage unit 1
2. Predict based on past data stored in step 2.

Next, the preparation part 18 produces a charging / discharging plan based on the time change of the electric power demand amount estimated by the prediction part 14 (S12). The charge / discharge plan is a plan for setting in advance the time and amount for charging or discharging the storage battery.

Then, the determination unit 15 determines whether or not to correct the power demand prediction that is a temporal change in the predicted future power demand based on the actual value of the day (S13). Detailed steps for determination will be described with reference to the flowchart of FIG.

When the power demand prediction is not corrected (NO in S13), the storage battery is charged / discharged by the charge / discharge plan created in S12.

On the other hand, when the power demand prediction is corrected (YES in S13), the correction unit 17 corrects the power demand prediction (S14). A detailed method of correction will be described with reference to the flowchart of FIG. And the preparation part 18 correct | amends the charging / discharging plan of a storage battery based on the electric power demand prediction corrected (
S15).

Thereafter, the output unit 13 outputs the power demand prediction corrected by the correction unit 17 or the charge / discharge plan created by the creation unit 18 (S16). The output power demand prediction may be a power demand prediction predicted by the prediction unit 14. The output method may be displayed on a display unit (not shown), or may be output and controlled on a control unit (not shown).

FIG. 3 shows a flowchart of the step of determining whether or not to correct the power demand prediction and the method of correcting the power demand prediction.

First, the determination unit 15 determines whether the time for determining the correction of the power demand prediction is the deviation monitoring time zone (S21). The divergence monitoring time zone is a time zone for monitoring the divergence between the power demand prediction and the power demand performance set in advance. In other words, during this divergence monitoring time zone, it is monitored whether the power demand prediction and the power demand record are divergence.

This is because the time period to be monitored is limited to the set time period, thereby eliminating unnecessary correction of power demand prediction. The start time of the divergence monitoring time zone is set in a time zone after the time when the load (actual power demand actual value) starts rising in order to monitor the divergence for the purpose of correcting the power demand prediction. For the purpose of correcting the power demand forecast, for example, the time zone suitable for the deviation monitoring time zone is a time zone in which the storage battery has been charged / discharged in the past, or the past power demand actual value is relatively high, This is a time zone that is relevant for creating a storage battery charge / discharge plan.

When the time for judging the correction of the power demand prediction is not the deviation monitoring time zone (NO in S21),
The flowchart ends and the power demand forecast is not corrected.

On the other hand, when the time for determining the correction of the power demand forecast is the deviation monitoring time zone (Y in S21)
ES), the calculation unit 16 calculates the error rate of the actual power demand value with respect to the predicted power demand value during the correction determination retroactive period (S22). The correction retroactive retroactive period refers to the number of hours or minutes in the past from the current data required to determine whether or not to correct the power demand forecast value when monitoring the deviation between the actual power demand and the power demand forecast. It is a period that determines whether to use the minute. This period is preset.

The correction judgment is set up and the power demand results and power demand forecasts for the retroactive period are accumulated, and the error rate of the power demand results with respect to the power demand forecasts, that is, the ratio of the difference between the total values of the power demand forecasts and the power demand results. calculate.

The purpose of setting the correction determination retroactive period is to prevent overcorrection of the power demand prediction when the error rate is large in a time zone with a small amount of power demand.

If the correction determination retroactive period is too short, the power demand prediction is frequently corrected with a slight fluctuation in the actual power demand from the calculation of the error rate and the determination of the threshold value. For example, the medium period is calculated based on the past power demand actual value, and the time from the rise of the power demand amount to 70 to 90% of the peak of the power demand amount is appropriate.
For example, it is 2 hours to 4 hours. However, the retroactive determination retroactive period is not limited to 2 hours to 4 hours.

Thereafter, the determination unit 15 determines whether the error rate is larger than the error rate threshold (S23). The error rate is a percentage indicating the deviation of ± 1% from the accumulated total of power demand actual values when the accumulated total of predicted power demand in the corresponding time zone is “1” (100%). It is. The error rate threshold is an error rate threshold for determining whether to correct power demand prediction. The error rate threshold may be changed depending on the season, date, and day of the week.

The time period to be integrated when calculating the integrated total of the predicted power demand value and the actual power demand value uses the time of the retroactive determination retroactive period with reference to the current time. The correction determination retroactive period is set in advance with a setting suitable for the consumer. For example, the correction determination retroactive period is 2
In the case of time, the total sum of the predicted power demand value and the actual power demand value is integrated in the time zone from two hours ago to the present time.

A specific error rate calculation method will be described. First, the total accumulated power demand forecast is calculated using the formula (
In 1), the total sum of the actual power demand values is obtained by equation (2).

Accumulated total power demand forecast value = Σ correction judgment retrospective forecast value (1)
Accumulated total of power demand actual values = Σ correction determination retroactive period actual value (2)

Then, as the divergence value, a difference between the integrated total of the power demand predicted value and the integrated total of the power demand actual value is calculated. In addition, as shown in Formula (3), the difference is obtained with an absolute value.
Deviation value = | "Total accumulated power demand value"-"Total accumulated power demand forecast" |
)

Finally, when the error rate is calculated as a percentage (%), the equation (4) is obtained.
Error rate = (deviation value ÷ “total sum of power demand forecast value”) × 100 (%) (4)

When the calculated error rate is smaller than a preset error rate threshold (NO in S23)
), The flowchart ends and the power demand forecast is not corrected. That is, in the case of equation (5),
Electricity demand forecast is not corrected.
Error rate> Error rate threshold value (5)

On the other hand, when the error rate is larger than the error rate threshold (YES in S23), the calculation unit 16 calculates a correction amount to correct the power demand prediction (S24). The correction amount is an amount for correcting the power demand prediction value in a predetermined correction target time zone.

The correction target time zone is a time for correcting the power demand prediction that is determined appropriately for each consumer. This is because if the forecast is corrected at the same rate in both the time zone where the power demand is high and the time zone where the power demand is low, it may be overcorrected in the time zone where the power demand is low. Set. For example, the correction start time is from the current time, but the correction end time is set to 19:30 or the like based on the power demand.

When the error rate is larger than the error rate threshold, the equation (6) is shown. An equation for calculating the correction amount is shown in Equation (7). That is, in the case of equation (6), the calculation unit 1 is used to correct the power demand prediction.
6 calculates the correction amount as shown in Equation (7).

In Expression (7), the predicted value of the correction determination retroactive period is a power demand predicted value of the correction determination retroactive retroactive period set in advance. For example, when the correction determination retroactive retroactive period is 2 hours, it means a power demand prediction value from 2 hours before to the current time.

Explaining the meaning of equation (7), from the actual power demand value and the predicted power demand value,
"Total accumulated power demand value" / "Total accumulated power demand value")-1) is obtained, multiplied by the predicted power demand value before correction, integrated in the correction target time zone, and corrected target time zone Divide by number.

The number of correction target time zones is a number when the correction target time zones are divided every certain time. For example, when the correction target time zone is 2 hours and divided every 30 minutes, the number of correction target time zones is four.
Error rate <Error rate threshold (6)
Correction amount = (Σ (predicted power demand value in correction target time zone) × ((“power demand actual value” / “predicted power demand value”) − 1)) ÷ number of correction target time zones (7)

The actual power demand value here and the power demand forecast value used the total sum of their respective values.
An instantaneous value at a certain time may be used.

Finally, the correction unit 17 corrects the power demand prediction (S25). Specifically, the power demand prediction value is corrected by adding the correction amount calculated by the calculation unit 16 to the original power demand prediction value before correction. This is expressed by the formula (8).
Corrected predicted value = predicted power demand predicted value + correction amount (8)

The target power demand prediction value before correction is a target power demand prediction value in the correction target time zone. For example, when the correction target time zone is 19:30 from the current time and currently 11:30, 1
The power demand prediction value until 9:30 is the power demand prediction value before correction. That is, from 11:30
The power demand forecast value until 9:30 is corrected.

However, when the correction amount is a negative value and the calculation result of Equation (8) is 0 or less, the corrected predicted value is 0.

FIG. 4 shows an image diagram relating to a method for correcting the predicted power demand value. The horizontal axis represents time, and the vertical axis represents demand, which represents a change in the daily power demand of a certain consumer over time. A solid line is a power demand actual value, a dotted line is a power demand prediction value before correction, and a broken line is a power demand prediction value after correction. The period (a) is the deviation monitoring time period, the period (b) is the retroactive determination retroactive period, and the period (c) is the correction target time period. The correction amount of (d) is the correction amount for power demand prediction obtained by Expression (8). FIG. 4 shows that the time when the amount of power of the actual power demand value rises is 8:00 and the current time is 12:00, and (c) the power demand prediction value in the correction target time period is corrected from the current time. .

In FIG. 4, (b) the correction determination retroactive period is 2 hours, so the correction determination retroactive period is from 10:00 to 12:00. Also, (a) the divergence monitoring time zone is from 10:00 to 15:00,
c) The correction target time zone is set to 15:00. However, FIG. 4 is merely an example, and the correction determination retroactive period, the deviation monitoring time zone, and the correction target time zone are changed to a time more suitable for the customer.

The corrected power demand prediction value indicated by a broken line is (b) an error rate that is a ratio of an integrated value of a deviation between the power demand actual value and the power demand prediction value with respect to the power demand prediction value during the correction determination retroactive period. (C) On the premise that the correction target time period is continued, (d) the correction amount is calculated as constant.

As for the correction method, (d) the correction amount is calculated based on Equation (7), and (c) the power demand forecast value for the period of the correction target time zone is a uniform amount as shown in Equation (8). (D) It is obtained by adding the correction amount.

(effect)
By setting a standard for correcting the power demand forecast and changing the standard to an appropriate value for the consumer, it is possible to prevent overcorrection and to correct the power demand forecast at a more appropriate timing and with a correction amount. it can. Thereby, an appropriate charging / discharging plan can be created.

Correcting at a more appropriate correction timing means that the difference between the power demand forecast and the power demand record, which is the basis for determining whether to correct the power demand forecast, Rather than using the data after the rise time of the amount of power in the actual power demand, it is not necessary to make corrections, and it is necessary to make corrections in the time zone involved in creating a storage battery charge / discharge plan. .

A more appropriate correction amount is a period for performing a preset correction based on the ratio of deviation between the power demand record and the power demand forecast after the rise time of the power demand record used at the correction timing. By calculating the correction amount as deviating at a constant rate within (correction target time zone), more accurate correction is made in the time zone relating to the charge / discharge plan creation of the storage battery.

In addition, by changing parameters such as deviation monitoring time zone, correction target time zone, correction judgment retroactive period, error rate threshold, etc. according to different power demand patterns for each power consumer, more power consumers Expansion of application to can be expected.

The deviation monitoring time zone, the correction target time zone, the correction determination retroactive period, and the error rate threshold described as values set in advance in the present embodiment should be adjusted so as to be suitable for each power consumer. Thus, the effect of the present embodiment can be realized. The deviation monitoring time zone, the correction target time zone, the correction determination retroactive period, and the error rate threshold can be changed at any time.

In addition, regarding the rise time of the power demand actual value that is the start time of the correction determination retroactive period shown in FIG. 4, the rise time determination is based on the slope that is the time change of the power amount of the power demand actual. The rise time may be detected, or the time may be determined in advance with reference to past power demand results. When determining the time, the rising time may be changed depending on the season, day of the week, or the like. Along with that, (a) Deviation monitoring time zone and (b
) The start time of the retroactive determination retroactive period may be advanced. For example, in summer, the actual power demand is 1
In the case of starting earlier, as shown in FIG. 5, the setting may be advanced by one hour, the start time of the retroactive determination retroactive period may be set from 7 o'clock, and the start time of the deviation monitoring time zone may be set from 9 o'clock.

Further, (b) the correction determination retroactive period does not necessarily have to be a fixed time. That is, only the start time of the retroactive determination retroactive period may be set, and from that time to the current time. For example, in FIG. 4, when the start time of the correction determination retroactive period is set as the rise time of the actual power demand value, the correction determination retroactive period is from 8:00 to 12:00 as shown in FIG.

Further, in the present embodiment, it has been described that the power demand prediction is uniformly corrected in the correction target time zone, but the amount to be corrected may be changed depending on the time zone. That is, the magnitude of the correction amount may be changed depending on the time zone. For example, in addition to (d) correcting the power demand prediction value as shown in FIG. 4 and distributing the correction amount with respect to time at a constant rate, as shown in FIG. Can also be distributed.

Moreover, although the power demand prediction value at the end time of the correction target time zone in FIG. 4 is discontinuous, the power demand prediction value may be corrected to be continuous as shown in FIG.

Further, in this embodiment, when determining whether to correct the power demand prediction value, the error rate is used and the formula (
As in 4), the ratio of the divergence value to the total accumulated power demand forecast value was calculated.
Like, you may calculate the ratio of the divergence value with respect to the integration total of an electric power demand performance value. Further, when determining whether to correct the power demand prediction value, the determination may be based on the amount of power instead of the error rate. That is, it is also possible to predetermine a threshold value for the amount of power that deviates between the predicted power demand value and the actual power demand value, and to correct the predicted power demand value when the power amount deviates beyond the threshold value.
Error rate = (deviation value / "total accumulated power demand value") x 100 (%) (9)

Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 1 Information processing apparatus 11 Acquisition part 12 Storage part 13 Output part 14 Prediction part 15 Judgment part 16 Calculation part 17 Correction part 18 Creation part

Claims (9)

A forecasting unit for forecasting a power demand forecast value that is a future power demand value;
It is determined whether or not to correct the predicted power demand value based on a difference between the actual power demand value and the predicted power demand value in a preset deviation monitoring time zone. A determination unit;
A calculation unit that calculates a correction amount of the power demand prediction value when it is determined to be corrected by the determination unit;
A correction unit that corrects the power demand prediction value based on the correction amount;
An information processing apparatus comprising: The determination unit, as a divergence between the actual power demand value and the predicted power demand value, the difference between the integrated values in the correction determination retroactive period set in advance, the predicted power demand value in the correction determination retroactive period The information processing apparatus according to claim 1, wherein the power demand prediction value is determined to be corrected when a ratio to the value is greater than a preset threshold value. In the correction target time zone, which is a period for performing preset correction,
The information processing apparatus according to claim 1, wherein the power demand prediction value is corrected based on the correction amount. The calculation unit calculates the correction amount in the correction target time zone to be constant based on a ratio of the actual power demand value to the predicted power demand value in a preset correction determination retroactive period. The information processing apparatus described. The calculation unit, based on a ratio of the power demand actual value to the power demand predicted value in a preset correction determination retroactive period, the correction amount of the correction target time zone is a ratio of the power amount of the power demand predicted value The information processing device according to claim 3, wherein the information processing device is calculated so that Based on the power demand prediction value, a creation unit for creating a charge / discharge plan for planning the time and amount for charging or discharging a storage battery,
The information processing apparatus according to claim 1, further comprising: An output unit for outputting the power demand forecast value or the charge / discharge plan;
An information processing apparatus according to any one of claims 1 to 6. A forecasting unit for forecasting a power demand forecast value that is a future power demand value;
It is determined whether or not to correct the predicted power demand value based on the difference between the actual power demand value and the predicted power demand value in the retroactive period set in advance. A determination unit;
A calculation unit that calculates a correction amount of the power demand prediction value when it is determined to be corrected by the determination unit;
A correction unit that corrects the power demand prediction value based on the correction amount;
An information processing system comprising: Information processing device
A forecasting means for forecasting a power demand forecast value that is a future power demand value;
It is determined whether or not to correct the predicted power demand value based on the difference between the actual power demand value and the predicted power demand value in the retroactive period set in advance. Judgment means,
A calculation unit that calculates a correction amount of the power demand prediction value when it is determined to be corrected by the determination unit;
Correction means for correcting the power demand prediction value based on the correction amount;
Information processing program to make it function.

Images