JPH06276679A - Demand controller - Google Patents

Abstract

PURPOSE:To provide a demand controller, in which a deterioration in control performance is prevented when controlled time-series data of working energy is used for forecasting control. CONSTITUTION:A demand controller forecasts working power energy by using a database of prior-term working power energy for a load group and limits the power energy under control for the load group below a given set value. In addition, the data controlled by a forecasted database selective means 12 is excluded from the time-series power energy data for the controlled load (A), and on the basis of these data a forecasting means 16 forecasts the working power energy and a control means 18 controls the power.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a demand control device for controlling a total power consumption value of a load group to be equal to or less than a set demand value, and more particularly to a demand control device for more appropriately controlling a load. .

[0002]

2. Description of the Related Art In recent years, there has been an active movement among power consumers to set a maximum power consumption value, that is, a demand value of power, and to suppress the total power consumption value within the range of the demand value to save energy. Here, it is conceivable to predict the used power amount from the time series data of the past power amount and control the load group to be equal to or less than the total used power value based on the predicted value. Regarding the prediction method, for example, the correlation coefficient between the time series data of the past electric energy and the current time series data is obtained, and the correlation coefficient is the closest to 1, that is, the most similar to the current time series data. Current time series data is obtained to predict future power consumption.

[0003]

However, when performing prediction from past time series data, if the prediction is performed using the time series data after control, the original time series data without control cannot be reproduced. In other words, it should be predicted based on the original time series data that is not controlled originally, but since the prediction is performed using the time series data after control, it is impossible to reproduce the uncontrolled original time series data. Become. Therefore, if the prediction is performed using the time-series data after the control, the control performance is rather deteriorated.

More specifically, in the peak cut control from the power consumption data of the previous day, that is, when a part of the load is cut off so that the total power consumption value of the load group becomes equal to or less than the set power value, If you predict the next day based on the time series data that is controlled so that it does not exceed the set power value, you should originally predict that the predicted value will exceed the set value, but it will not exceed the set value. You will predict.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a demand control device which does not use control time series data and is not affected by the control and which predicts the power consumption value more accurately. is there.

[0006]

A demand control device according to the present invention is, firstly, connected to a load group and predicts the used power amount based on the past used power amount data of the load group as a prediction base. In the demand control device that controls the total power consumption value of the load group to be equal to or less than the set value based on the predicted value, only the control data deleting unit that deletes the controlled data from the prediction base, and the data without the control. And a predicting unit that predicts the amount of electric power used by using the non-control data as a prediction base, and performs control based on the prediction from the non-control data.

Secondly, the used power amount is predicted from the integrating means for integrating the used power amount of the past used power amount data which is not controlled, and the integrated value integrated by the integrating means. And a prediction means for performing the prediction.

The third feature is that the power consumption is predicted based on only the past power consumption data of the load to which no control is applied as a prediction base.

A fourth feature is that power consumption is predicted based on past power consumption data of a load that is not controlled and data of external conditions such as outside temperature. Is.

Fifth, in the case where the load to which the above control is applied is not controlled based on the past power consumption data of the load to be controlled and the control amount applied to the load to be controlled. It is characterized in that past power consumption data is estimated, power consumption is predicted at least based on the estimated past power consumption data, and control is performed according to the prediction.

Further, a sixth feature is that the control amount to be applied to the load to be controlled is predicted from the control amount data added to the load to be controlled, and the load control is performed based on those predicted values. .

[0012]

In the first demand control device according to the present invention,
The control data deleting means deletes the controlled data from the prediction base, and the predicting means predicts the power consumption from the non-controlled data consisting only of the data not subjected to the control. Since the data to which control has been added is not deleted and put in the prediction base, prediction can be performed without being affected by past control.

In the second configuration, the accumulating unit integrates past power consumption data without control to the power consumption for each predetermined time, and the predicting unit predicts the power consumption based on the accumulated value. Then, the controlled object is controlled according to this prediction. Since the integrated value does not change regardless of what peak shift is performed within the predetermined time, prediction can be performed without being affected by past control.

Further, in the third configuration, since only the past power consumption data of the load to which no control is applied is used as the prediction base, it is possible to eliminate the influence of the control on the prediction.

Further, in the fourth configuration, the power consumption is predicted based on not only the past power consumption data of the load to which no control is applied but also the data of the external condition such as the outside air temperature as a prediction base. More accurate control can be performed.

Further, in the fifth configuration, the load subjected to the above control is not controlled based on the past power consumption data of the load to be controlled and the control amount added to the load to be controlled. In the case where the load is not controlled more accurately, the past power consumption data is estimated and the power consumption is predicted based on at least the estimated past power consumption data. Can be predicted. In the sixth configuration, the control amount to be applied to the load to be controlled is predicted from the control amount data applied to the load to be controlled, and the load control is performed based on those predicted values. Therefore, the waveform data when control is not applied is performed. There is no need to estimate.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings.

Demand control device 1 shown as the first embodiment
As shown in FIG. 1, 0 has a prediction base selecting unit 12 as a control data deleting unit, a time series storage unit 14, a predicting unit 16 and a control unit 18, and the prediction base selecting unit 1
The input end of 2 and the output end of the control means 18 are connected to the load A to be controlled.
It is connected to the.

Here, when the prediction base selecting means 12 receives the past power consumption time series data from the controlled load A, it deletes only the cycle data to which the control is added from this power consumption data. For example, when the past power consumption data shown in (a) of FIG. 3 is given, the data of cycle 1 and the data of cycle 3 are not controlled, and the data of cycle 2 are subjected to peak cut control. Therefore, the data of cycle 2 is deleted. The break of the cycle here is the minimum value in the midnight time of the day. This is to reduce the fluctuation of data at a value close to the demand value.

Next, the time-series storage means 14 stores the time-series data in which the data output by the prediction base selection means 12 under the control is deleted. For example, when the power consumption data of FIG. 3A is obtained, the prediction-based selection unit 1
2 deletes the data of cycle 2, so that the data shown in FIG.
The waveform data shown in is created. In addition, the prediction means 16
The power consumption is predicted from the time-series data obtained by deleting the controlled data stored in the time-series storage means 14. Further, the control unit 18 controls the load A to be controlled according to the power consumption amount predicted by the prediction unit 16.

The operation of the demand control apparatus 10 having the above configuration will be described with reference to the flowchart shown in FIG. 2 and the like. First, the time series data of the power consumption actually used by the controlled load A to the controlled load A will be described. Prediction base selection means 12
Is output to. The prediction base selection means 12 counts each cycle (step 010), determines whether or not control is applied to the cycle data (step 011),
If control is applied to the cycle data, the cycle data is deleted (step 012). The period data which is not controlled is as it is, the time series storage means 1
Sent to 4. The “joint complement” shown in step 013 of FIG. 2 is a waveform having continuity as shown in FIG. 3 because the waveform data of electric energy loses continuity when data of a certain cycle is deleted and connected. This is performed in order to obtain data, and the process of connecting the data in each cycle is performed. Next, the prediction unit 16 predicts the amount of electric power used from the time series data stored in the time series storage unit 14, and the control unit 18 controls the control target based on the value predicted by the prediction unit 16. By doing so, the cycle data that has been subjected to the peak cut control is cut, and therefore, the predicted cycle data does not affect the prediction of the power consumption, and the prediction of the power usage becomes more realistic. can do.

In the above-mentioned embodiment, the prediction base selecting means 12 sets the break of the cycle to the minimum value in the midnight time zone of the day, but the invention is not limited to this.
It is also possible to set a minimum point below a certain threshold.

Demand control device 20 shown as a second embodiment
As shown in FIG. 4, has an original waveform integration means 22, a time series storage means 24, a prediction means 26, and a control means 28, and the input end of the original waveform integration means 22 and the output end of the control means 28 are
It is connected to the controlled load A. Here, the original waveform integrating means 22 performs a process of integrating the used power amount every 3 hours from the original waveform data of the used power amount in the past. Further, the time-series storage means 24 configures the time-series data of the integrated data for every three hours integrated by the original waveform integration means 22. Also,
The predicting unit 26 predicts the amount of electric power used from the integrated data for which the time series storage unit 24 is configured in time series for every three hours. The control unit 28 controls the load A to be controlled according to the prediction of the prediction unit 26.

The operating state of the second embodiment will be described.
First, assuming that the original waveform data as shown in FIG. 6A is obtained, the original waveform integrating means 22 integrates the used electric energy every 3 hours. That is, at time 0, −1 time point and −
The value obtained by integrating the power consumptions at the 2nd time point and the -3rd time point is the value at the 0th time point. At the 3rd time point, the power consumptions at the 1st time point, the 2nd time point and the 3rd time point are integrated to obtain a value at the 3rd time point. The demand target value here (this is referred to as “first demand target value”) is 2
It is 00 kW.

Next, based on the value accumulated every 3 hours,
The prediction means 26 predicts the amount of power used. In this embodiment,
A waveform obtained by connecting the integrated values for every 3 hours with a straight line as shown in FIG. 6B is used as the predicted value. The demand target value here (this is referred to as “second demand target value”) is shown in FIG.
In FIG. 6A, the value is 200 KW, but in FIG. 6B, the value obtained by multiplying the demand target value in FIG. 6A by 3 is multiplied by a safety factor between 0 and 1. This is to set the target value to a lower value in advance in order to ensure demand control.

Next, the control means 28 controls based on the waveform data predicted by the prediction means 26. That is,
If at least one of the three accumulated time points exceeds the second demand target value, a process of averaging the three demand values is performed. In the case of the waveform data shown in FIG. 6 (b),
In the set consisting of one time point, two time points, and three time points, the value at the three time points exceeds the demand target value, so the value obtained by dividing the sum of the one time point, the two time points, and the three time points by three is the value at each time point. Even in the set of four time points, five time points, and six time points, since each time point exceeds the second demand target value, the averaging process is similarly performed. -2 time point, -1 time point, 0 time point set and 7
In the set of time points, 8 time points, and 9 time points, the averaging process is not performed because the value at each time point does not exceed the second demand target value. The waveform data shown in FIG. 6C is obtained as described above, and the control means 28 controls the load A to be controlled based on this waveform data. Demand control is performed, for example, every day based on this waveform data. That is, to explain according to the flowchart of FIG. 5, similarly to the above, as shown in FIG. 6B, the power consumption after 3 hours is predicted (step 020) and the threshold value (second demand target) is calculated. Value) is determined (step 021),
If it exceeds, the excess is calculated (step 02
2) The peak shift control is performed by performing the averaging process (step 023). Since the waveform data that is the basis of the prediction may change, the 3-hour data is integrated again and stored in the time series storage means 24 as a prediction base. With the above configuration, even if the added data for every three hours is drawn from the peak-shifted waveform data, it does not differ from the added data of the original waveform data, so the same control as the control based on the data that has not been applied It can be performed.

Demand control device 3 shown as a third embodiment
As shown in FIG. 7, 0 has a time series storage means 34, a prediction means 36, and a control means 38 for taking in past power consumption data of loads other than the control target, and this time series storage means 34. Takes in past power consumption data of loads other than the control target, and the prediction unit 36 predicts the power consumption based on the data of the loads other than the control target. Control means 38
Controls the load A to be controlled according to the prediction. Here, for example, it is assumed that the load to be controlled is the water heater and the load not to be controlled is a load other than the water heater.

The operation of the demand control device according to the third embodiment will be described with reference to the flowchart of FIG.
The predicting means 36 predicts the amount of electric power used according to certain data (step 030). Here, initially, the prediction is made according to the actual power consumption data, but from the next time, the prediction is performed using the power consumption data excluding the control target data from step 033. Then, according to the prediction, the control means 38 controls the controlled object by, for example, peak shift distribution (step 031). Next, it is determined whether one day has passed while performing the above control (step 032). When one day has passed, the data for the controlled load A is input to the time series storage means 34 from the data for one day. Instead, it is removed (step 033), and the data of the power consumption other than the control target is stored in the time series storage means 34 and used as the prediction base (step 034). According to the present embodiment, since the power consumption of the load to be controlled is not used as the prediction base, the time series data subjected to the control is not used and the prediction error due to the control can be eliminated.

Here, the demand value when performing control such as peak shift is the maximum amount of electric power used by loads other than the control target. This is because, for example, in the case of peak shift control, the operation time of a load that performs peak shift may be shifted to a time zone when there is a margin in the use of electric power, so the demand value is the amount of power used by only the load that is not controlled To judge. To determine the maximum amount of power, for example, the maximum amount of power used for one year is used as the demand value. This is a system of actual quantity contract that monitors the average value of electric power consumption for a certain month every 30 minutes and uses the largest value of the maximum value and the maximum electric power consumption of the 11 months before that month as the contract electric power. However, if the maximum amount of electric power used for one year is monitored, the amount of electric power can be suppressed within the amount of electric power according to the actual amount contract, so that the electric power charge can be suppressed. The demand value in the case of peak cut control is also the maximum power consumption of loads other than the control target.

When controlling the water heater by peak shift control, for example, it is desirable that the time of shifting is as close as possible to the reserved use time of the water heater so as not to exceed the demand value. This is because when the operating time of the water heater is shifted, the hot water must be kept warm until the reserved time, which contributes to power saving by shortening the warming time.

Further, when the peak shift is performed, for example, when a plurality of air conditioners are installed in a dormitory or the like, when the air conditioners are switched on at the same time, the peak power becomes extremely large. Therefore, when a predetermined number or more of switches are turned on within a predetermined time, it is conceivable to provide a detection means for detecting this and shift the operation time. This is because, for example, when an air conditioner operation is reserved, it is realistic to set the operation time in advance depending on the number of reservations, but when a plurality of switches are turned on at the same time, the operation time may be shifted later. Conceivable.

Next, the demand control device 40 shown as the fourth embodiment, as shown in FIG.
It is similar to that of the third embodiment in that the power consumption amount is used as a prediction base, but is characterized by having an estimating means 42 for estimating the power consumption amount of the controlled load A from other data. In the present embodiment, the controlled load A is, for example, an air conditioner. Here, the other data is time-series data such as the outside temperature and the set temperature, and the power consumption of the control target is estimated according to the flowchart shown in FIG. That is, from the indoor conditions, the indoor temperature, and the outside air temperature, heat acquisition (internal heat generation in the room such as the human body, lighting, and OA equipment) and heat loss (heat that escapes naturally due to drafts, heat conduction, etc.) are calculated. The room heat load (sum of heat acquisition and heat loss) is calculated from heat acquisition and heat loss. Further, the outside air load (the amount of heat that enters and leaves when air is taken in from the outside) is calculated from the indoor temperature and the outside air temperature, and the air conditioner load power is calculated from the room heat load and the outside air load. That is, the amount of heat is converted into electric power. With the configuration according to the present embodiment, the amount of electric power used that is not controlled by the control target is estimated, and the estimated amount of electric power used is also the basis of the prediction, so it is suitable for a load with a severe behavior such as an air conditioner. That is, in comparison with the third embodiment, the third embodiment uses the water heater as a load and therefore has a relatively simple behavior, whereas the fourth embodiment uses the air conditioner as a load to be controlled. Since the value fluctuates significantly, the prediction method also needs to consider many conditions.

Next, the demand control device 50 shown as the fifth embodiment is substantially the same as the third embodiment as shown in FIG. 11, except that the estimating means 52 uses the power consumption data from the controlled load A. The control amount of the control performed by the control unit 58 is fetched, and the power consumption of the control target load A is estimated from the power consumption data from the control target load A and the control amount.
Specifically, when the control amount is controlled to be 80%, the power consumption amount actually used by the controlled load A is multiplied by the reciprocal of the control amount, that is, 1 / 0.8. Then, the amount of electric power used before the control is calculated, and this value is added to the amount of electric power used by the load B other than the control target to be stored in the time series storage means 54. The prediction unit 56 makes a prediction based on the data stored in the time-series storage unit 54, and the control unit 58 controls the load A to be controlled according to the prediction. In this embodiment, since the predicted power consumption before control is estimated from the actually controlled amount, more reliable control can be performed, and if the control ratio is known, a waveform that is not controlled easily can be obtained. It has the advantage of being predictable. In addition,
Since the estimating means 52 estimates the power consumption of the controlled load A, the data received by the estimating means 52 from the control means 58 is only the controlled load A data.

Next, the demand control device 60 shown as the sixth embodiment has a time series storage means 64, a first prediction means 66, a control time series storage means 63, a second prediction means 67 and a control means 68. The input end of the time series storage means 64 and the output end of the control means 68 are connected to the load A to be controlled. Here, the time-series storage means 64 takes in the data of the amount of power used from the controlled load A, estimates the amount of used power of the controlled load (controlled), and stores it as time-series data. The first predicting unit 66 predicts the power consumption of the load to be controlled based on the power consumption data stored in the time-series storage unit 64, and outputs it to the control unit 68. On the other hand, the control time series storage means 63 fetches and stores the data of the control amount added by the control means 68 from the control means 68. The second prediction means 67 is the control time series storage means 63.
The control amount is predicted from the time series data of, and the predicted control amount is output to the control means 68. The control unit 68 controls the load A to be controlled based on the predicted value of the amount of electric power used by the load A to be controlled and the predicted value of the control amount.

FIG. 13 is a flow chart showing the operation of the loop of the control time series storage means 63, the second predicting means 67 and the control means 68. First, the second predicting means 67 predicts the control amount (step 060). The control means 68 performs control according to the predicted value (step 061). here,
The control means 68 determines whether or not the threshold value, that is, the target value, is exceeded, and if it exceeds the threshold value, the control amount is adjusted to control the load to be controlled (step 063). It is stored in the sequence storage means 63 (step 064). The predicted value from the first predicting means 66 is used to modify the controlled variable applied by the controlling means 68. In the present embodiment, it is not necessary to estimate the original waveform that has not been controlled, and there is a base control amount in advance. Therefore, there is an advantage that the change of the control amount can be small.

In FIG. 12, the output of the control time series storage means 63 is directly input to the first prediction means 66, and the first prediction means 66 performs both functions of the first prediction means 66 and the second prediction means 67. If this is done, the configuration can be simplified. In addition, as shown in FIG. 14, the time-series data of the controlled variable takes the load to be controlled into the number of digits, that is, 1
One kind of load is taken for each digit, and the ratio of the load energization time within the unit time (30 minutes) is expressed by the numbers 0-9.

[0037]

In the demand control device according to the present invention, the predictive control is performed without using the time-series data of the controlled portion, so that the control performance of the predictive control using the controlled time-series data is improved. It can prevent the deterioration.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a configuration of a demand control device according to a first embodiment.

FIG. 2 is a flowchart showing the operation of the demand control device according to the first embodiment.

FIG. 3 is an explanatory diagram showing an operation of the demand control device according to the first embodiment.

FIG. 4 is a configuration diagram showing a configuration of a demand control device according to a second embodiment.

FIG. 5 is a flowchart showing the operation of the demand control device according to the second embodiment.

FIG. 6 is an explanatory diagram showing an operation of the demand control device according to the second embodiment.

FIG. 7 is a configuration diagram showing a configuration of a demand control device according to a third embodiment.

FIG. 8 is a flowchart showing the operation of the demand control device according to the third embodiment.

FIG. 9 is a configuration diagram showing a configuration of a demand control device according to a fourth embodiment.

FIG. 10 is a flowchart showing the operation of the estimating means.

FIG. 11 is a configuration diagram showing a configuration of a demand control device according to a fifth embodiment.

FIG. 12 is a configuration diagram showing a configuration of a demand control device according to a sixth embodiment.

FIG. 13 is a flowchart showing the operation of the demand control device according to the sixth embodiment.

FIG. 14 is a diagram showing time series data of a controlled variable.

[Explanation of symbols]

10, 20, 30, 40, 50, 60 Demand control device 12 Prediction base selection means 14, 24, 34, 44, 54, 64 Time series storage means 16 26, 36, 46, 56 Prediction means 18, 28, 38, 48, 58, 68 Control means 22 Original waveform integrating means

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yosuke Otsuka 2-14-40 Ofuna, Kamakura City, Kanagawa Prefecture Mitsubishi Electric Corporation Living Systems Research Center (72) Inventor Shigeki Suzuki 2-14, Ofuna, Kamakura, Kanagawa Prefecture No. 40 Mitsubishi Electric Corporation Life Systems Research Institute (72) Inventor Kazuhiko Nanjo 2-14-40 Ofuna, Kamakura-shi, Kanagawa Mitsubishi Electric Corporation Life Systems Research Institute

Claims (6)

[Claims] 1. A value that is connected to a load group, predicts the amount of power used by using the past power consumption data of the load group as a prediction base, and sets the total amount of power used by the load group based on the predicted value. In the demand control device to be controlled below, control data deletion means for deleting data to which control has been applied from the past power consumption data, and outputting the result as a prediction base, and control output from the control data deletion means A demand control device, comprising: a prediction unit that predicts the amount of electric power used based on the prediction base that includes only data that is not added. 2. A value that is connected to a load group, predicts the amount of power used based on the past power consumption data of the load group as a prediction base, and sets the total amount of power used by the load group based on the predicted value. In the demand control device to be controlled below, an integrating unit that integrates the amount of power used for each predetermined time of the amount of power used in the past without control, and the amount of power used based on the integrated value integrated by the integrating unit as a prediction base. A demand control device, comprising: a prediction unit that predicts. 3. A value that is connected to a load group, predicts the power usage amount based on the past power usage data of the load group as a prediction base, and sets the total power usage value of the load group based on the predicted value. In the demand control device to be controlled below, the demand control device is characterized in that the power consumption is predicted based only on the past power consumption data of the load without control. 4. A value which is connected to a load group, predicts the power usage amount based on the past power usage data of the load group as a prediction base, and sets the total power usage value of the load group based on the predicted value. In the demand control device controlled as follows, the demand control device is characterized in that the power consumption is predicted based on the past power consumption data of the load to which no control is applied and the external environment condition data. 5. A value that is connected to a load group, predicts the power usage amount based on the past power usage data of the load group as a prediction base, and sets the total power usage value of the load group based on the predicted value. In the demand control device to be controlled below, the past when the load to which the above control is added is not controlled from the past power consumption data of the load to be controlled and the control amount added to the load to be controlled The demand control device, comprising: an estimation unit that estimates the used power amount data of the above, and that predicts the used power amount based on at least the past used power amount data estimated by the estimation unit. 6. A value that is connected to a load group, predicts the power usage amount based on the past power usage data of the load group as a prediction base, and sets the total power usage value of the load group based on the predicted value. In the demand control device to be controlled below, the demand control device is characterized in that the control amount to be applied to the load to be controlled is predicted from the control amount data added to the load to be controlled, and load control is performed by the predicted control amount.