JP2010286190A - Method of predicting load and load predicting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of predicting load, and a load predicting device capable of performing calculation with high accuracy by preventing dispersion in the result of the prediction of a load depending on conditions. <P>SOLUTION: The load predicting device calculates a load of each air conditioner at a prescribed time by using a total-load predicting calculating means 3 on the basis of a rated load value of each air conditioner stored in an air conditioner rated load storing means 1 in advance, and an operation schedule stored in an air conditioner operation schedule storing means 2 in advance, and determines the total-load prediction value by accumulating the loads of all of the air conditioners to apply it as the total load of heat source equipment. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a load prediction method and apparatus for predicting a heat load of a heat source facility.

To reduce CO ₂ , further optimization and energy saving of operation control of heat source facilities such as a heat storage tank and a refrigerator associated with air conditioning control are required while maintaining comfort of air conditioning control.
Conventionally, as a method for optimizing the operation of a heat source facility, there is a method of performing load prediction using ARIMA (Auto-Regressive Integrated Moving Average Model) and performing operation control based on the obtained predicted load (for example, see Patent Document 1). ).

  In addition, there is a method of creating a schedule for controlling the number of operating heat source facilities by combining a predicted load according to time and an actual load (for example, see Patent Document 2). At the start of operation, the number of operating units of the heat source facility is controlled based on the predicted load at the start of operation, and when the operation shifts to the steady operation, the operation number control is performed based on the predicted load at the time of shifting to the steady operation until the effect of the control of the operation number appears. If the operation is steady, the number of operating units is controlled based on the actually measured load. If it is time to shift to overtime operation, the number of operating units is controlled based on the predicted load when shifting to overtime operation until the end of operation.

JP-A-8-75219 JP 2000-18682 A

In the load prediction using the ARIMA model as in the load prediction method disclosed in Patent Document 1, there is a problem in the accuracy of parameters such as weather conditions and facility operation conditions, so there is a large difference between the prediction and the actual. Operation control could not be performed accurately even using the predicted load.
In addition, when performing operation control according to a schedule as in Patent Document 2, there may be a difference between actual load due to variations in device characteristics, etc., regarding the initial load and the load at the time of transition to steady operation, If the difference between the actual size and the actual size increases, useless driving will be performed.

  The present invention has been made to solve the above-described problems, and provides a load prediction method and a load prediction apparatus that can prevent a prediction result from varying depending on conditions and can calculate with high accuracy. Objective.

  The load prediction method according to the present invention includes a predicted load value calculation step of calculating a load for each air conditioner at a predetermined time using a pre-stored rated load value for each air conditioner and an operation schedule, and a predetermined time A total load prediction value is calculated by accumulating the load for each air conditioner for all air conditioners, and this is used as a total load prediction value calculation step using this as a load of the heat source equipment.

  The load prediction method according to the present invention uses an actual load value on an arbitrary day before a predetermined date for each air conditioner and an operation schedule for each air conditioner stored in advance, and performs air conditioning at a predetermined time on a predetermined date. A predicted load value calculation step for calculating the load for each unit, and the total load prediction value is calculated by accumulating the load for each air conditioner for all the air conditioners at a predetermined time and using this as the load of the heat source equipment. And a value calculation step.

  In the load prediction method according to the present invention, the predicted maximum temperature on a predetermined day and the maximum temperature on any day before the predetermined date are calculated for the load of each air conditioner at a predetermined time on the predetermined date calculated in the predicted load value calculation step. A load prediction correction process that adds a correction value obtained by converting the difference into the amount of heat is provided. In the overall load value prediction process, the load for each air conditioner at the predetermined time corrected in the load prediction correction process is accumulated for all air conditioners. It is.

  The load prediction method according to the present invention uses a predicted outside air temperature on a predetermined day to calculate a load for each air conditioner at a predetermined time on a predetermined day from a function that defines a relationship between the load of each air conditioner and the outside air temperature. A value calculation step, and a total load prediction value is calculated by accumulating the load of each air conditioner at a predetermined time for all the air conditioners, and this is used as a load of the heat source equipment. Is.

  The load predicting apparatus according to the present invention calculates a load for each air conditioner at a predetermined time using a pre-stored rated load value for each air conditioner and an operation schedule, and accumulates all the air conditioners for a total. A load prediction value is obtained, and an overall load prediction calculation means that uses this as a load of the heat source equipment is provided.

  The load predicting apparatus according to the present invention uses an actual load value of an arbitrary day before a predetermined date for each air conditioner and an operation schedule for each air conditioner stored in advance to perform air conditioning at a predetermined time on a predetermined day. The load for each unit is calculated, and the total load prediction value is obtained by accumulating all the air conditioners, and the total load prediction calculation means for using this as the load of the heat source equipment is provided.

  The load prediction device according to the present invention includes a load prediction correction unit that calculates a correction value obtained by converting a calorific value of a difference between a predicted maximum temperature on a predetermined day and a maximum temperature on an arbitrary day before the predetermined date, and an overall load prediction calculation unit Is to correct the load for each air conditioner at a predetermined time on a predetermined day using the correction value calculated by the load prediction correction means, and the corrected load is accumulated for all air conditioners.

  The load predicting apparatus according to the present invention calculates a load for each air conditioner at a predetermined time on a predetermined day from a function that defines a relationship between the load of each air conditioner and the outside air temperature, using the predicted outside air temperature on a predetermined day, The total load predicted value is obtained by accumulating the air conditioner, and the total load predicting calculation means for using this as the load of the heat source equipment is provided.

  According to this invention, the load for each air conditioner at a predetermined time is calculated using the pre-stored rated load value for each air conditioner and the operation schedule, and the total load predicted value is accumulated for all air conditioners. Therefore, by calculating the load at each time based on the driving schedule, it is possible to calculate the accuracy with preventing the prediction result from varying depending on the conditions.

  According to this invention, using the actual load value of an arbitrary day before a predetermined day for each air conditioner and the operation schedule for each air conditioner stored in advance, the air conditioner for each air conditioner at a predetermined time on a predetermined day. Since the load is calculated and the total load predicted value is obtained for all air conditioners, the variation in the prediction results can be further suppressed by feeding back the actual load of each air conditioner to the predicted value calculation. .

  According to this invention, for the load of each air conditioner at a predetermined time on a predetermined day, a correction value obtained by converting the difference between the predicted maximum temperature on the predetermined day and the maximum temperature on an arbitrary day before the predetermined date into a calorie value is corrected. Since the load for each air conditioner at the specified time is accumulated for all the air conditioners, even when there is a large difference in outside air temperature between the predetermined date and the day before the predetermined date, the load is predicted accurately. be able to.

  According to this invention, using the predicted outside air temperature on a predetermined day, the load for each air conditioner at a predetermined time on a predetermined day is calculated from a function that defines the relationship between the load of each air conditioner and the outside air temperature, and all the air conditioners Since the total load prediction value is obtained by accumulating, the load prediction can be performed in consideration of the device characteristics of the air conditioner.

It is a block diagram which shows the structure of the load prediction apparatus which concerns on Embodiment 1 of this invention. 3 is a flowchart showing an operation of the load prediction apparatus according to the first embodiment. It is explanatory drawing which shows an example of the operation schedule of each air conditioner which the air conditioner operation schedule memory | storage means shown in FIG. 1 has memorize | stored. 10 is a flowchart showing an operation of the load prediction apparatus according to the second embodiment. It is a graph showing the correlation between the load of the air conditioner i and the outside air temperature.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing the configuration of the load prediction apparatus according to Embodiment 1 of the present invention. The load predicting apparatus according to the first embodiment includes an air conditioner rated load storage unit 1 that stores the rated value of an air conditioner installed in each facility, and an air conditioner operation that stores an operation schedule of each air conditioner. The schedule storage means 2 and the total load prediction calculating means 3 for predicting the load of each air conditioner and calculating the cumulative value of all the air conditioners are provided. In addition, in the load prediction apparatus of this Embodiment 1, since the previous day performance load memory | storage means 4, the external temperature acquisition means 5, and the load prediction correction | amendment means 6 shown in FIG. 1 are unnecessary, description is abbreviate | omitted. These will be described later in Embodiments 2 and 3.

FIG. 2 is a flowchart showing the operation of the load prediction apparatus according to the first embodiment. Here, it is assumed that the heat load prediction of the heat source facility on that day is performed at the time of the day change (0:00).
First, in step ST1, the overall load prediction calculation means 3 reads the rated value of each air conditioner from the air conditioner rated load storage means 1. The rated value is a value of the processing capability specific to the device under a predetermined condition described in a catalog or the like.

  Subsequently, in step ST <b> 2, the overall load prediction calculation unit 3 reads the operation schedule of each air conditioner on the current day from the air conditioner operation schedule storage unit 2. FIG. 3 is an explanatory diagram showing an example of an operation schedule of each air conditioner. The operation schedule of FIG. 3 shows the operation state of each air conditioner (1 to i) at each time on April 27, which is the load prediction day. Note that this operation schedule is determined in advance on a weekday, holiday basis, or the like.

Subsequently, in step ST3 (load predicted value calculation step), the overall load prediction calculation means 3 reads the read rated value Lp (i) of the air conditioner i and the operating state S (i, t of the air conditioner i at time t. ) To calculate the predicted load value {Lp (i) × S (i, t)} at time t of the air conditioner i.
In subsequent step ST4 (total load predicted value calculating step), the total load prediction calculating means 3 accumulates the load predicted values of all the air conditioners at time t in accordance with the following equation (1), and the total load predicted value Hp of the heat source equipment. (T).
Hp (t) = Σ {Lp (i) × S (i, t)} (1)

  The overall load prediction value calculated by the load prediction device is used, for example, for operation control on the day of the heat source facility.

  As described above, according to the first embodiment, the rated load value for each air conditioner stored in advance in the air conditioner rated load storage unit 1 and the operation schedule stored in advance in the air conditioner operation schedule storage unit 2 are obtained. The total load prediction calculation means 3 calculates the load for each air conditioner at a predetermined time, and accumulates all the air conditioners to obtain the total load predicted value, which is used as the load of the heat source facility. . For this reason, by calculating the load for each time based on the operation schedule, it is possible to calculate with high accuracy while preventing the prediction result from varying depending on the conditions.

  In the first embodiment, the load prediction target date is the day of operation control of the heat source facility, but it goes without saying that the load prediction can be performed on a day other than the day on which the operation schedule exists.

Embodiment 2. FIG.
Although the load prediction device according to the first embodiment calculates the total load predicted value using the rated load value of the air conditioner, in the second embodiment, instead of the rated load value, the entire load of the day is calculated from the actual load value of the previous day. Calculate the predicted load value. Since the configuration of the load prediction device according to the second embodiment is the same as that of the load prediction device according to the first embodiment (FIG. 1) in the drawings, the following description is made with reference to FIG. To do. In the second embodiment, the load prediction device stores the actual load value of the previous day of each air conditioner in addition to the air conditioner rated load storage means 1, the air conditioner operation schedule storage means 2 and the overall load prediction calculation means 3. The previous day performance load storage means 4 is provided.

FIG. 4 is a flowchart showing the operation of the load prediction apparatus according to the second embodiment. As in the first embodiment, the heat load prediction of the heat source facility on that day is performed at the time of the day change (0:00).
First, in step ST <b> 11, the overall load prediction calculation unit 3 reads the previous day actual load value from the previous day actual load storage unit 4. The actual load value is, for example, a load heat quantity Q per hour, and the load heat quantity Q is a water supply temperature T1 sent from the heat source equipment to the air conditioner and a return water temperature T2 returned from the air conditioner to the heat source equipment. And the value calculated by the following equation (2) using the return water flow rate F.
Q = (T2−T1) × F (2)

  Subsequently, in step ST12, the overall load prediction calculation unit 3 reads out the operation schedule of each air conditioner on the day as shown in FIG. 3 from the air conditioner operation schedule storage unit 2.

Subsequently, in step ST13, the overall load prediction calculation means 3 reads the actual load value Lm (i, d-1, t) at the d-1 time t on the previous day of the air conditioner i from the previous day actual load storage means 4. Then, the overall load prediction calculation means 3 uses the actual load value Lm (i, d-1, t) and the load predicted value Lp (i, d-1, t) at the time d-1 on the previous day to perform air conditioning. The predicted load value Lp (i, d, t) at the time d at the current day of the machine i is calculated from the following equation (3).
Lp (i, d, t) = {α × Lm (i, d−1, t)
+ (1−α) × Lp (i, d−1, t)} (3)
Here, α is a weighting factor for adjustment for determining the ratio between the actual load value of the previous day and the predicted load value.

In subsequent step ST14, the total load prediction calculation means 3 uses the operating state S (i, t) of the air conditioner i at the time t to accumulate the predicted load values of all the air conditioners at the time t according to the following equation (4). The total load predicted value Hp (t) of the heat source equipment is used.
Hp (t) = Σ [{α × Lm (i, d−1, t)
+ (1-α) × Lp (i, d−1, t)}
× S (i, t)] (4)

  The overall load prediction value for the day calculated by the load prediction device is used for, for example, operation control on the day of the heat source facility.

  As described above, according to the second embodiment, the actual load value of the previous day for each air conditioner stored in the previous day actual load storage unit 4 and the individual air conditioner stored in advance in the air conditioner operation schedule storage unit 2. Using the operation schedule, the total load prediction calculation means 3 calculates the load for each air conditioner at a predetermined time of the day, and accumulates all the air conditioners to obtain the total load predicted value, which is calculated as the load of the heat source facility. Configured to do. For this reason, it is possible to further suppress variations in the prediction results by feeding back the actual load of each air conditioner to the prediction value calculation.

In the second embodiment, the load prediction target date is the day of operation control of the heat source facility, but it goes without saying that the load prediction can be performed on a day other than the day on which the operation schedule exists.
Moreover, although the actual load value of the day before a load prediction object day was used, it cannot be overemphasized that load prediction can be implemented even if it uses the actual load value of days other than the previous day.

Embodiment 3 FIG.
The calculation method of the predicted load value Lp (i, t) at time t of the air conditioner i can be calculated by a method other than the method according to the first and second embodiments, and an example will be described in the third embodiment. In the load predicting apparatus according to the third embodiment, a predicted value of the outside air temperature correlated with the load is used. The configuration of the load predicting apparatus according to the third embodiment is the same as that of the load predicting apparatus according to the first embodiment (FIG. 1) on the drawing, and therefore FIG. 1 is used in the following description. I will explain. In the third embodiment, the load prediction device predicts the outside air temperature of the day in addition to the air conditioner rated load storage means 1, the air conditioner operation schedule storage means 2, the overall load prediction calculation means 3 and the previous day actual load storage means 4. An outside air temperature acquisition unit 5 that acquires and stores a value via a communication line or the like, and a load prediction correction unit 6 that obtains a correction value by converting the temperature difference between the previous day and the highest temperature on the previous day into a calorific value.

First, a method for calculating the predicted load value Lp (i, t) using the predicted outside air temperature value held by the outside air temperature acquisition means 5 instead of the rated load value in the first embodiment will be described.
The overall load prediction calculation means 3 reads the predicted outside air temperature value To (t) at the time t of the day from the outside air temperature acquisition means 5 and correlates the predicted outside air temperature value To (t) with the load of each air conditioner i. The load prediction value Lp (i, t) is calculated by the following prediction equation (5) expressed by a linear linear equation. In addition, in FIG. 5, the graph showing the correlation of the load of the air conditioner i and external temperature is shown.
Lp (i, t) = a (i) × To (t) + b (i) (5)
Here, To (t) is the predicted outside air temperature at time t, and a (i) and b (i) are the regression coefficients of the outside air temperature and load of the air conditioner i.
The total load predicted value Hp (i, t) may be calculated by the above formula (1) using the load predicted value Lp (i, t) calculated by the above formula (5).

Next, in the calculation method according to the second embodiment, a method for calculating the total load predicted value Hp (t) using the predicted value of the outside air temperature held by the outside air temperature acquisition unit 5 will be described. Here, the load predicted value is corrected using the difference between the predicted maximum temperature on the current day and the maximum temperature on the previous day, in consideration of the case where the difference between the previous day and the outside temperature on the current day is large.
The load prediction correction means 6 first calculates the difference ΔTo between the highest outdoor temperature on the previous day and the predicted highest outdoor temperature on the current day from the outside air temperature acquisition means 5, and further calculates the correction value (c × ΔTo) by converting the amount of heat (load). Prediction correction process). Subsequently, the total load prediction calculation means 3 adds the correction value (c × ΔTo) to the load prediction value Lp (i, d, t) as shown in the following formula (6), and calculates the total load prediction value Hp (t). Correct.
Hp (t) = Σ [{α × Lm (i, d−1, t)
+ (1−α) × Lp (i, d−1, t) + c × ΔTo}
× S (i, t)] (6)
Here, c is a calorie conversion coefficient.

  As described above, according to the third embodiment, using the predicted outside air temperature acquired by the outside air temperature acquiring unit 5 on the day, the overall load prediction calculating unit 3 determines the relationship between the load of each air conditioner and the outside air temperature. The load for each air conditioner at a predetermined time of the day was calculated, and the total load predicted value was obtained by accumulating all the air conditioners, and this was used as the load of the heat source equipment. For this reason, it is possible to perform load prediction in consideration of the equipment characteristics of the air conditioner.

  Further, according to the third embodiment, the load prediction correction unit 6 calculates a correction value obtained by converting the difference between the predicted maximum temperature on the current day and the maximum temperature on the previous day into a calorific value, and the overall load prediction calculation unit 3 calculates the correction value. Is used to correct the load for each air conditioner at a predetermined time of the day, and the corrected load is accumulated for all air conditioners. For this reason, even when the difference between the outside air temperatures on the current day and the previous day is large, the load can be predicted with high accuracy.

In the third embodiment, the load prediction target day is the day of the operation control of the heat source equipment. However, the load prediction is performed on a day other than the day when the operation schedule exists and the predicted outside air temperature can be obtained. Needless to say, it can be implemented.
Moreover, although the outside temperature and the actual load value on the day before the load prediction target day are used, it goes without saying that the load prediction can be performed even if the outside temperature and the actual load value on the day other than the previous day are used.

DESCRIPTION OF SYMBOLS 1 Air conditioner rated load memory | storage means 2 Air conditioner operation schedule memory | storage means 3 Whole load prediction calculation means 4 Previous day result load memory means 5 Outside temperature acquisition means 6 Load prediction correction means

Claims (8)

A load prediction method for predicting a load of a heat source facility,
A predicted load value calculation step of calculating a load for each air conditioner at a predetermined time using a pre-stored rated load value for each air conditioner and an operation schedule;
A load comprising a total load predicted value calculation step of calculating a total load predicted value by accumulating the load for each air conditioner at the predetermined time for all the air conditioners, and setting this as a load of the heat source facility. Prediction method. A load prediction method for predicting a load of a heat source facility,
The load for each air conditioner at the predetermined time on the predetermined day is calculated using the actual load value on any day before the predetermined day for each air conditioner and the operation schedule for each air conditioner stored in advance. A predicted load value calculation step,
A load comprising a total load predicted value calculation step of calculating a total load predicted value by accumulating the load for each air conditioner at the predetermined time for all the air conditioners, and setting this as a load of the heat source facility. Prediction method. The load prediction method according to claim 2,
A correction value obtained by converting the difference between the predicted maximum temperature on the predetermined day and the maximum temperature on any day before the predetermined date to the calorific value for the load for each air conditioner calculated at the predetermined time on the predetermined date calculated in the load predicted value calculation step. A load prediction correction process for adding
In the overall load value prediction step, the load prediction method is characterized by accumulating the load for each air conditioner at the predetermined time corrected in the load prediction correction step for all the air conditioners. A load prediction method for predicting a load of a heat source facility,
A predicted load value calculation step of calculating a load for each air conditioner at a predetermined time on the predetermined date from a function that defines a relationship between the load of each air conditioner and the external air temperature using a predicted external temperature on a predetermined day;
A load comprising a total load predicted value calculation step of calculating a total load predicted value by accumulating the load for each air conditioner at the predetermined time for all the air conditioners, and setting this as a load of the heat source facility. Prediction method. A load prediction device for predicting a load of a heat source facility,
Using the rated load value and operation schedule for each air conditioner stored in advance, the load for each air conditioner at a predetermined time is calculated, and the total load predicted value is obtained by accumulating all air conditioners. A load predicting apparatus comprising an overall load predicting calculating unit as a load of the heat source facility. A load prediction device for predicting a load of a heat source facility,
The load for each air conditioner at the predetermined time on the predetermined day is calculated using the actual load value on any day before the predetermined day for each air conditioner and the operation schedule for each air conditioner stored in advance. And a load predicting device comprising a total load predicting calculation means for obtaining a total load predicted value by accumulating all air conditioners and using this as a load of the heat source equipment. The load prediction device according to claim 6,
Load prediction correction means for calculating a correction value obtained by converting the difference between the predicted maximum temperature on a predetermined day and the maximum temperature on any day before the predetermined date into a calorific value,
The total load prediction calculation means corrects the load for each air conditioner at a predetermined time on the predetermined day using the correction value calculated by the load prediction correction means, and accumulates the corrected loads for all the air conditioners. A characteristic load prediction device. A load prediction device for predicting a load of a heat source facility,
Using the predicted outside air temperature on a given day, calculate the load for each air conditioner at a given time on the given day from a function that defines the relationship between the load on each air conditioner and the outside air temperature. A load prediction apparatus comprising an overall load prediction calculation unit that obtains an overall load prediction value and uses this as a load of the heat source facility.

Images