JP2007014179A - Demand control unit, predicting method and program of consumed electric power - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a demand control unit which smoothly controls equipment not exceeding contracted electric power, by accurately predicting an amount of consumed electric power of whole load equipment, and to provide the predicting method of the consumed electric power used for it. <P>SOLUTION: A history record of a relationship or the like between operation states of the main load equipment and the actual amount of the consumed electric power at that time is stored in an information table, wherein prediction of the relation by theoretical calculation is difficult, after which the information table is used when the amount of the consumed electric power is predicted. Consequently, the amount of the consumed electric power is predicted with high accuracy. Specifically, a history position corresponding to the operation states of the load equipment at the current moment is specified on the information table, then the predicted amount of the electric power is obtained by adding the actually consumed electric power stored in the information table thereafter. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a demand control device, a power consumption prediction method suitable for use in the demand control device, and a program related thereto.

  As a contract form with an electric power company, a consumer can apply a demand contract. In this demand contract, the contract power amount is set based on the demand value. Here, the demand value is a value of the maximum demand power amount calculated every 30 minutes by a watt hour meter (demand meter) out of the demand power amount for a certain time unit (for example, one month). . The watt-hour meter can sequentially store the maximum electricity demand value.

  The contract power amount that is the basis for calculating the basic charge is determined based on the past demand value. The contract power amount is set as an upper limit value of the power amount that can be consumed every 30 minutes. If the demand power amount exceeds the contract power amount even once in a certain month, the power demand will be increased in the following month. Even if the amount does not exceed the contracted power amount, the contracted power amount is immediately reset from that month, which changes the basic charge over the next 12 months.

  A demand control device is used as a control device for smoothly operating such a power system. The demand control device acquires the current power consumption of the entire load device (for example, air conditioner, showcase, lighting, etc.) from the watt hour meter within the 30 minutes that is the set time limit of the contract power consumption. Calculate the amount of power that can be used within the remaining time limit for the contract power amount. On the other hand, the predicted value of the power consumption of the entire load device within the remaining time period is calculated each time. Then, the calculated usable power amount is compared with the predicted power consumption, and if there is a possibility that the predicted power consumption exceeds the usable power amount, for example, an operation stop command is output to the air conditioner, and the air conditioning The power consumption is reduced by stopping the cooling or heating operation or the air blowing operation of the device. Thereafter, the demand control device sequentially calculates the amount of power that can be used within the remaining time period and the calculation of the predicted power consumption for each set time described above, so that the power demand amount does not exceed the contracted power amount. In addition, the operation of each load device is limited or controlled.

  As an example of such a demand control apparatus, there is an apparatus described in Patent Document 1 or the like.

  As a method of calculating the predicted power consumption of the entire load device within the remaining time period, a method of linearly predicting based on the fluctuation value of the power consumption consumed during the current minute time is known. This will be described below with reference to the drawings.

FIG. 1 shows a linear prediction method of power consumption in a certain 30-minute section. Referring to FIG. 1, the horizontal axis represents the time axis, the vertical axis represents the accumulated energy axis, power consumption until a timing T ₂ is represented by a solid line in the graph. The accumulated power consumption is reset to zero every 30 minutes. At timing T ₂ , for each 30-minute contracted power amount, while calculating the available power amount within the remaining time limit and predicting the power consumption amount, Restrict and control operations.

In the prediction of power consumption, the power consumption W ₂ at the timing T ₂ (points on the graph X ₂₎
Then, the deviation ΔW of the electric energy from the electric power consumption W ₁ (point X ₁ on the graph) at the timing T ₁ that is traced back by the minute time ΔT from the timing T ₂ is used. That is, the power consumption amount at T = 30 minutes is predicted from this power amount deviation ΔW. Specifically, the predicted power consumption W ′ at T = 30 minutes is calculated using a straight line l ′ (broken line graph) connecting the points X ₁ and X ₂ . Note that the slope of the straight line l ′ is (ΔW / ΔT).

At this time, if the predicted power consumption W ′ exceeds the contract power consumption, the operation of the load device is restricted to reduce the power consumption, and the power consumption after timing T ₂ exceeds the contract power consumption. Control not to. In FIG. 1, a straight line l (dotted line graph) indicates the predicted power consumption when the power consumption is limited. That is, control is performed so that the predicted power consumption W at T = 30 minutes does not exceed the contract power consumption.
JP-A-7-143670

  However, according to such a conventional technique, as a method of calculating the predicted power consumption of the entire load device within the remaining time period, the prediction is based only on the fluctuation value of the power consumed during the current minute time. Therefore, the prediction may be wrong due to the operation of the load device that is difficult to predict, and as a result, the usable electric energy may be tight within the remaining time period. In such a case, it may be necessary to perform control to stop the operation of almost all of the load devices. For example, if the showcase is stopped, the quality of the displayed food is deteriorated. .

  In addition, the demand control device predicts that the power consumption amount consumed within a minute time at a certain timing is large, and as a result, the power consumption amount exceeds the contract power amount within the time limit, and the operation of the corresponding load device is Even if the operation of the load device is stopped without actually stopping it, the subsequent power consumption will be lower than the predicted value, and the contracted power amount will be exceeded within the time limit without stopping the operation of the load device. It may happen if you didn't. In this case, there is a problem that the operation of the load device is stopped unnecessarily even for a short time.

  Accordingly, the present invention provides a demand control device capable of accurately predicting the power consumption amount within each time period so as to avoid the above-described problems, thereby smoothly controlling the equipment without exceeding the contract power. And a power consumption prediction method used therein and a program related to them.

  The present invention mainly uses a history of the relationship between the operating state of a load device and the power consumption at that time, and thereby a demand control device capable of accurately predicting the power consumption within each time period, and The present invention provides a power consumption prediction method used there and a program related to them.

  The demand control device according to the invention of claim 1 is a demand control device that controls a load device based on a predicted power consumption that is predicted to be consumed by the end timing of a time unit. It has a table production | generation means which produces | generates the information table regarding the log | history of electric energy, and the prediction calculation means which calculates the said predicted power consumption using the information table produced | generated by this table production | generation means, It is characterized by the above-mentioned.

  According to a second aspect of the present invention, in the demand control device according to the first aspect, the table generating means includes a preset operating state of the load device among the load devices in the target unit and an actual power consumption at that time. Storage means for storing the relationship as a history in the information table, wherein the prediction calculation means specifies a history position corresponding to the current operating state of the preset load device on the information table, and specifies And calculating means for calculating the predicted power consumption based on the actual power consumption acquired from the information table following the history position.

  If the operation status of the load device at a certain past timing is the same as the current operation status, the behavior of the load device after that will be the same as after the past timing, and the power consumption may follow the same change. Is expensive. In other words, if the operating state of the load device matches at a certain past timing, the operating state after the current time is likely to change in the same way as the subsequent history, and therefore the actual power consumption consumed in the subsequent history The accuracy of the predicted power consumption can be improved by predicting the power consumption after the current time based on the above. Therefore, as in the second aspect of the invention, the history is stored in the information table, and the predicted power consumption is calculated based on the power consumption acquired from the information table. Predict with high accuracy.

  According to a third aspect of the present invention, in the demand control device according to the second aspect, the storage means associates time information as information for specifying the history position in association with the operating state and actual power consumption at that time. When the plurality of history positions corresponding to the current operating state of the preset load device are specified on the information table, the calculating means stores the information at the current time. The history position at the nearest time is selected as the history position to be specified in the calculation of the predicted power consumption.

  In a load device that requires setting, the setting contents are likely to be the same every day. That is, there is a high possibility that the operating state of the load device is the same at approximately the same time every day. Therefore, there is a high possibility that there is a history position where there is a location that coincides at almost the same time every day in the history, and that the subsequent changes in the operating state are exactly the same. Therefore, as in the invention of claim 3, when the time information is stored together with the information table and a plurality of history positions corresponding to the current operating state of the load device are specified, the time of the time closest to the current time is determined. By selecting the history position, it is possible to further improve the prediction accuracy of the power consumption of the entire load device.

According to a fourth aspect of the present invention, in the demand control device according to the second or third aspect, the calculation means includes a case where a history position corresponding to a current operation state of the preset load device does not exist on the information table. Calculating the predicted power consumption by linear prediction from the deviation between the actual power consumption acquired immediately before and the actual power consumption at the present time;
It is characterized by that.

  In the invention of claim 4, when there is no operation state that matches the operation state of the current load device in the history, it is calculated using linear prediction from deviation information of the previous power consumption and the current power consumption. A predicted power consumption is calculated based on the predicted value. That is, it has a function of complementing the function of the demand control device according to claim 2 or 3 when there is no operation state in the history that matches the operation state of the current load device. Further, since the demand control device according to claim 2 or 3 does not have sufficient history immediately after the start of use, in that case, while calculating the predicted power consumption mainly by the invention of claim 4, Accumulate the history and use the history with time to improve the accuracy of the prediction. The invention of claim 4 is useful in such a process.

  According to a fifth aspect of the present invention, in the demand control apparatus according to any one of the second to fourth aspects, the storage means has the actual power consumption within a predetermined range from an average value of the power consumption in the information table. Only in this case, the actual power consumption is stored in the information table in association with the operation state.

  The demand control device according to any one of claims 2 to 4 uses the history to predict the power consumption with high accuracy. Therefore, if exceptional information is mixed in the history, there is a possibility that an error having a high influence degree is caused when the position is referred to in the prediction of the power consumption amount at a later date. Therefore, according to the invention of claim 5, only when the power consumption falls within a predetermined range from the average value of the power consumption in the information table, the power consumption is associated with the operating state of the device in the information table. It is very beneficial to eliminate exceptional information by storing.

  According to a sixth aspect of the present invention, in the demand control device according to any one of the second to fifth aspects, when a showcase is included as the preset load device, the storage means is configured to defrost and cool the showcase. And a process of storing a relationship between an operation state and the actual power consumption amount as a history in the information table.

  According to a seventh aspect of the present invention, in the demand control device according to any one of the second to sixth aspects, when an air conditioner is included as the preset load device, the storage means is configured to turn the air conditioner on and off. And a process of storing a relationship between the state of the / OFF operation and the actual power consumption amount as a history in the information table.

  According to an eighth aspect of the present invention, in the demand control device according to the sixth aspect, the storage means stores, in the information table, the difference between the showcase internal temperature and the set temperature at that time, in addition to the relationship. It is characterized by that.

  Load devices that can consume a large amount of power are most likely to affect the amount of power consumed. Therefore, in the invention according to any one of claims 6 to 8, the operation state that is most likely to affect the power consumption of such a load device is stored in the information table as a history.

  As for the showcase, there are many examples in which a large number of showcases and one refrigerator are connected, and as a result, when predicting changes in the power consumption of the showcase by simple theoretical calculations, The accuracy is extremely low, and a large calculation is required to increase the accuracy. Therefore, it is very useful to use an information table in which the relationship between the operating state of the showcase and the power consumption is stored as a history. In addition, if the difference between the internal temperature and the set temperature is different even in the same showcase state, there is a high possibility that the power consumption and behavior will change, so this is also stored as a history.

  A ninth aspect of the present invention is the demand control device according to the first aspect, wherein the information table includes an operation state of a preset load device among the load devices in the target unit and an actual state in the target unit at that time. A first reference value that is an average value of the increase in power consumption and / or a second reference value that is larger than the first reference value are stored in association with each other, and the prediction calculation unit is configured to set the preset value in advance. It is determined whether or not the same operational state as the current operational state of the loaded device is present in the information table, and if the same operational state exists, the first reference value or the corresponding operational state The predicted power consumption is calculated based on the second reference value.

  Among the load devices in the target unit, even if the load device set in advance, that is, the load device whose operation state (operation state) is monitored has the same operation state, the influence of the operation state of other load devices As a result, there may be variations in the amount of increase in the actual power consumption in the target unit every predetermined time. For this reason, the first reference value that is an average value of the increase amount may also vary. Therefore, the predicted power consumption calculated based on the first reference value is highly accurate but may vary.

According to the invention of claim 9, for example, when the second reference value is set to a value larger than the first reference value and the predicted power consumption calculated based on the second reference value always exceeds the actual power consumption, When calculating the predicted power consumption, select whether to use the first reference value or the second reference value depending on the situation such as the remaining time until the end timing of the time unit, for example. As a result, an appropriate predicted power consumption can be calculated according to the situation, and based on this, demand control that does not exceed the contract power can be performed.
The first reference value described above includes, for example, the average value, median value, mode value, variance value, and approximate value of the increase amount of the actual power consumption in the target unit. is there.

  The invention according to claim 10 is the demand control device according to claim 1, wherein the information table includes one or more time zones, an operation state of the preset load device in each time zone, and the time at that time The first reference value that is an average value of the increase amount of the actual power consumption in the target unit and / or the second reference value that is larger than the first reference value are stored in association with each other, and the prediction calculation means When the combination of the time zone to which the current time belongs and the preset operating state of the load device at the time exists in the information table, the combination corresponding to the combination of the time zone and the operating state The predicted power consumption is calculated based on the first reference value or the second reference value.

  The influence of the time zone can be considered as one of the environmental operation conditions that vary the relationship between the preset operating state of the load device among the load devices in the target unit and the actual power consumption in the target unit. For example, if the load equipment includes a showcase or a refrigerator that cools it, go to the store of the showcase or the store / facility where the showcase or refrigerator is installed. The number of customers visiting is influenced by the time of day. Accordingly, the amount of electric power consumed to cool the interior of the showcase, which has been heated due to the delivery of food items, the entrance / exit of store visitors, body temperature, and the like, is also affected by the time zone.

  According to the invention of claim 10, as described above, the predicted power consumption is calculated based on the first reference value or the second reference value at this time in consideration of the time zone and the preset operating state of the load device in the time zone. Therefore, the predicted power consumption can be calculated with high accuracy.

  According to an eleventh aspect of the present invention, in the demand control device according to the ninth or tenth aspect, the prediction calculation means further includes a remaining time calculation means for calculating a remaining time from a current time to the end timing, and the prediction calculation means. Calculates the predicted power consumption based on the first reference value if the remaining time is a predetermined time or more, and calculates the predicted power consumption based on the second reference value if the remaining time is less than the predetermined time. The predicted power consumption is calculated.

  As described above, the predicted power consumption calculated based on the first reference value may vary, and the actual power consumption may exceed the predicted power consumption. As a result of performing demand control based on such predicted power consumption, the actual power consumption in the target unit may exceed the predicted power consumption, but even in such a case, the remaining time If there is enough, it can be recovered by the subsequent control of the load device.

  On the other hand, when the demand control is performed based on the predicted power consumption calculated based on the second reference value, the demand control can be performed with emphasis on the fact that the actual power consumption does not exceed the predetermined value.

  According to the invention of claim 11, when the remaining time is sufficient, the predicted power consumption is calculated based on the first reference value, and based on this, a margin is given with emphasis on maintaining the comfort of the store / facility. On the contrary, when the remaining time is short, the predicted power consumption is calculated based on the second reference value, and the actual power consumption does not exceed the predetermined value based on this. It is possible to perform demand control that emphasizes this.

  According to a twelfth aspect of the present invention, in the demand control device according to the ninth or eleventh aspect, the preset load device includes a plurality of showcases, and each showcase has an operation state of the preset load device. It includes a state where a pull-down operation is performed.

    If multiple showcases perform pull-down operation at the same time after defrosting operation, the power consumption of the refrigerator connected to these showcases will increase sharply and will have a significant effect on the actual power consumption of the entire load equipment. There is a case.

  According to claim 12, when the preset load device includes a plurality of showcases, including a state in which each showcase is performing a pull-down operation as an operation state of the preset load device, Since the sudden increase in power consumption due to the pull-down operation of the showcase is taken into account, the predicted power consumption can be calculated with higher accuracy.

  A thirteenth aspect of the present invention is the demand control device according to the ninth or twelfth aspect, wherein the second reference value is a maximum value of an increase in actual power consumption in the target unit.

  According to the invention of claim 13, since the maximum value of the increase amount of the actual power consumption in the target unit is adopted as the second reference value, the predicted power consumption calculated based on the second reference value is ensured. The value can be larger than the actual power consumption.

  A demand control device according to a fourteenth aspect of the invention is a demand control device that controls a load device based on a predicted power consumption that is predicted to be consumed by the end timing of a time unit. Representative pattern generation means for generating a representative pattern table describing a representative pattern of a time-series change in the operating state of the device, and the predicted power consumption using the representative pattern table generated by the representative pattern generation means And a prediction calculation means for calculating the quantity.

  According to the invention of claim 14, the present invention relates to a load device stored and tabulated by using a table describing a typical pattern (hereinafter referred to as a modeled pattern) of time-series changes in the operating state of the load device. Variations in operating states can be reduced, and as a result, the amount of information in the table can be reduced, while the time required for search and the required amount of memory can be reduced.

  There are individual differences between the load devices, and the environment in which the load devices are placed can exist with a high probability even if compared to the same time zone of the previous day, for example, so the operating conditions related to the load devices Is stored in a table, the range of variation in the operation state data is enormous.

  Therefore, in order to eliminate the influence of the variation as described above, for example, the operation state of each load device with respect to time is substantially the same or approximately as compared with the case where its characteristic main part is actual. It is modeled by analogy, simplification, abstraction, formalization, etc. so that the state of the load device is in the state shown in the model, and by referring to the information, The amount of information can be reduced, the time required for searching and the amount of memory required can be reduced. For example, by using an operation command schedule for a load device in the model, operation state data to be referred to can be easily created.

  According to the invention of claim 14, when using the operation state of the load device in the target unit, instead of receiving the operation state directly from the load device by communicating with the load device, the By referring to the generated representative pattern table, the operating state of the load device can be obtained and the power consumption can be predicted. Thereby, for example, communication overhead can be avoided.

  According to a fifteenth aspect of the present invention, in the demand control device according to the fourteenth aspect, a linear prediction power consumption for calculating a predicted power consumption by linear prediction from a deviation between the actual power consumption acquired immediately before and the actual power consumption at the present time. An amount calculating unit, and using the linear predicted power consumption calculated by the linear predicted power consumption calculating unit and the predicted power consumption calculated by the prediction calculating unit, a corrected predicted power consumption is calculated. It is characterized by calculating.

  The invention according to claim 16 is the demand control device according to claim 15, wherein the corrected predicted power consumption is calculated by weighted averaging the linear predicted power consumption and the predicted power consumption. To do.

  As in the inventions of claims 15 and 16, a linear prediction power consumption is calculated by linear prediction from a deviation between the actual power consumption acquired immediately before and the actual power consumption at the present time, and the linear prediction power consumption, By using the predicted power consumption calculated by the prediction calculation unit and calculating the corrected predicted power consumption by taking a weighted average of these, for example, the accuracy of prediction of the power consumption can be improved. .

  The invention according to claim 17 is the demand control device according to any one of claims 14 to 16, further comprising table generating means for generating an information table relating to a history of actual power consumption in a target unit of demand control, wherein the information table Includes a first reference value that is an average value of an operating state of a preset load device among the load devices in the target unit and an increase amount of the actual power consumption in the target unit at that time, and / or Or the second reference value that is larger than the first reference value is stored in association with each other, and the prediction calculation means is the same as the current operation state of the preset load device obtained from the representative pattern table. It is determined whether or not an operation state exists in the information table. If the same operation state exists, the first reference corresponding to the operation state is determined. Or calculates said predicted consumed electric power amount based on said second reference value, characterized in that.

  The invention according to claim 18 is the demand control device according to any one of claims 14 to 16, further comprising table generating means for generating an information table relating to a history of actual power consumption in a target unit of demand control, wherein the information table 1 is an average value of one or two or more time zones, the preset operating state of the load device in each time zone, and the increase amount of the actual power consumption in the target unit at that time A reference value and / or a second reference value that is larger than the first reference value are stored in association with each other, and the prediction calculation means is configured to store the time zone to which the current time belongs and the time obtained from the representative pattern table. When a combination with the preset operation state of the load device exists in the information table, the combination of the time zone and the operation state Calculates said predicted consumed electric power amount based on said first reference value or said second reference value corresponding to Align, characterized in that.

  According to a nineteenth aspect of the present invention, in the demand control device according to the seventeenth or eighteenth aspect, the prediction calculation means further includes a remaining time calculation means for calculating a remaining time from a current time to the end timing, and the prediction calculation means. Calculates the predicted power consumption based on the first reference value if the remaining time is a predetermined time or more, and calculates the predicted power consumption based on the second reference value if the remaining time is less than the predetermined time. The predicted power consumption is calculated.

  The invention according to claim 20 is the demand control device according to any one of claims 17 to 19, wherein the preset load device includes a plurality of showcases, and the operation state of the preset load device is: Each showcase includes a state in which a pull-down operation is performed.

  A twenty-first aspect of the invention is the demand control device according to any one of the seventeenth to twentieth aspects, wherein the second reference value is a maximum value of an increase in actual power consumption in the target unit. To do.

  A power consumption prediction method according to a twenty-second aspect of the present invention is the power consumption prediction method for calculating and acquiring the amount of power predicted to be consumed in the power monitoring unit by the end timing of the time unit. Among them, information about the preset operation pattern of the load device and the history of the actual power consumption consumed in the unit at that time is updated and stored, and the history information is used until the end timing of the time unit. The amount of power consumed in the unit is predicted and calculated.

  The invention of claim 23 is the power consumption prediction method according to claim 22, wherein an operation pattern corresponding to the preset operation pattern of the load device is specified from the history information, and the actual power consumption in the history thereafter Based on the amount, the amount of power consumed in the unit before the end timing of the time unit is calculated.

  The power consumption prediction method according to the invention of claim 24 is a power consumption prediction method for calculating and acquiring a predicted power consumption amount predicted to be consumed in the power monitoring unit by the end timing of the time unit. The first reference value and / or the first reference value which is an average value of the operation state of the load device set in advance among the load devices and the actual power consumption increase amount in the power monitoring unit at that time The second reference value that is a larger value is associated and stored in the information table, and it is determined whether or not the same operation state as the current operation state of the preset load device exists in the information table. When the operating state is present, the predicted power consumption is calculated based on the first reference value or the second reference value corresponding to the operating state.

The power monitoring unit described above corresponds to a target unit for demand control when, for example, demand control is performed, and the same applies to the following.
Further, the first reference value described above includes, for example, the average value of the increase amount of the actual power consumption in the power monitoring unit, the variance value, the approximate value thereof, and the like.

  A power consumption prediction method according to a twenty-fifth aspect of the invention is a power consumption prediction method for calculating and acquiring a predicted power consumption amount that is predicted to be consumed in the power monitoring unit by the end timing of the time unit. A first reference value which is an average value of a time zone, an operating state of the preset load device in each time zone, and an increase in actual power consumption in the power monitoring unit at that time, and / or the first The second reference value, which is a value greater than one reference value, is stored in association with each other, and the same time zone and operation as the combination of the time zone to which the current time belongs and the preset operating state of the load device at that time It is determined whether or not a combination with a state exists in the information table. If a combination with the same time zone and an operation state exists, the combination of the time zone and the operation state is determined. Calculates said predicted consumed electric power amount based on said first reference value or said second reference value corresponding to the mating saw, characterized in that.

  According to a twenty-sixth aspect of the present invention, in the power consumption prediction method according to the twenty-fourth or twenty-fifth aspect, the remaining time from the current time to the end timing is calculated, and if the remaining time is equal to or longer than a predetermined time, the first The predicted power consumption is calculated based on a reference value, and if the remaining time is less than a predetermined time, the predicted power consumption is calculated based on the second reference value.

  According to a twenty-seventh aspect of the present invention, in the power consumption prediction method according to the twenty-fourth or twenty-sixth aspect, the preset load device includes a plurality of showcases, and the operation state of the preset load device is each It includes a state in which the showcase is in a pull-down operation.

  The invention of claim 28 is the power consumption prediction method according to claim 24 or claim 27, wherein the second reference value is a maximum value of an increase in actual power consumption in the power monitoring unit. .

  A power consumption prediction method according to a twenty-ninth aspect of the present invention is a power consumption prediction method for calculating and acquiring a predicted power consumption amount that is predicted to be consumed in the power monitoring unit before the end timing of the time unit. Generating a representative pattern table describing a representative pattern of a time-series change in the operating state of the load device, predicting the predicted power consumption using the representative pattern table, and within the target unit for demand control An information table related to the history of actual power consumption in the power monitoring unit is generated, the preset operating state of the load device among the load devices in the power monitoring unit, and the increase amount of the actual power consumption in the power monitoring unit at that time Information table in association with a first reference value that is an average value of the first reference value and / or a second reference value that is a value greater than the first reference value It is stored and it is determined whether the same operational state as the current operational state of the preset load device exists in the information table, and if the same operational state exists, the operational state corresponds to the operational state The predicted power consumption is calculated based on the first reference value or the second reference value.

  The invention of claim 30 is the power consumption prediction method according to claim 29, wherein the predicted power consumption is calculated by linear prediction from the deviation between the actual power consumption acquired immediately before and the actual power consumption at the present time, and the linear prediction is performed. Using the linear predicted power amount calculated by the power consumption calculating unit and the predicted power consumption calculated by the prediction calculating unit, or weighting the predicted power consumption calculated by the prediction calculating unit The corrected predicted power consumption is calculated using the average value.

  A power consumption prediction method according to a thirty-first aspect of the present invention is the power consumption prediction method for calculating and acquiring the predicted power consumption that is predicted to be consumed in the power monitoring unit by the end timing of the time unit. Generating a representative pattern table describing a representative pattern of a time-series change in the operating state of the load device, predicting the predicted power consumption using the representative pattern table, and within the target unit for demand control The information table regarding the history of the actual power consumption at the time is generated, one or more time zones, the preset operating state of the load device in each time zone, and the actual power consumption at the power monitoring unit at that time The first reference value that is an average value of the increase amount of the first and / or the second reference value that is larger than the first reference value is associated with the case. Determining whether or not the information table has a combination of the same time zone and operating state as the combination of the time zone to which the current time belongs and the preset operating state of the load device at the time, When there is a combination of the same time zone and operating state, the predicted power consumption is calculated based on the first reference value or the second reference value corresponding to the combination of the time zone and the operating state. It is characterized by calculating.

  According to a thirty-second aspect of the present invention, in the power consumption prediction method according to the thirty-first aspect, a predicted power consumption is calculated by linear prediction from a deviation between the actual power consumption acquired immediately before and the actual power consumption at the present time, and the linear prediction is performed. The corrected predicted power consumption is calculated by using the linear prediction power consumption calculated by the power consumption calculation means and the predicted power consumption calculated by the prediction calculation means or by weighted averaging. And

  The invention of claim 33 is the power consumption prediction method according to any one of claims 29 to 32, wherein the remaining time from the current time to the end timing is calculated, and if the remaining time is equal to or longer than a predetermined time, the first The predicted power consumption is calculated based on one reference value, and if the remaining time is less than a predetermined time, the predicted power consumption is calculated based on the second reference value.

  The invention of claim 34 is the power consumption prediction method according to any one of claims 29 to 33, wherein the preset load device includes a plurality of showcases, and the operation state of the preset load device is: Each showcase includes a state in which pull-down operation is performed.

  The invention of claim 35 is the power consumption prediction method according to any one of claims 29 to 34, wherein the second reference value is a maximum value of an increase amount of the actual power consumption in the power monitoring unit. To do.

  In the inventions of claims 17 to 21 or 29 to 35, the inventions of claims 14 to 16 have the effects of the inventions of claims 9 to 13 or 24 to 28 of the invention. In addition, effects such as reduction in the amount of information in the table, reduction in the time and memory required for the search, avoidance of communication overhead, and the like can be achieved, and the prediction accuracy of the power consumption can be improved.

  The second reference value is a value obtained by multiplying the first reference value by a constant, a median value or a mode value if the second reference value is a value larger than the first reference value in addition to the maximum increase in actual power consumption. It may be the average value of the top 30% of the measurement data. Further, it may be one that uses the dispersion or standard deviation of measurement data. When a value obtained by multiplying the first reference value by a constant is stored, the constant may be stored instead, and the second reference value may be calculated each time. This saves memory.

  As the second reference value, it is preferable to adopt the maximum value from the viewpoint of simple and reliable risk management.

  A thirty-sixth aspect of the present invention is a program for giving a computer the function of each means of the demand control apparatus according to any one of the first to twenty-first aspects.

  A thirty-seventh aspect of the invention is a program for giving a computer a function of executing the power consumption prediction method according to any of the twenty-second to thirty-fifth aspects.

  The significance or effect of the present invention will become more apparent from the following description of embodiments.

  However, the following embodiment is merely one embodiment of the present invention, and the meaning of the term of the present invention or each constituent element is not limited to that described in the following embodiment. Absent.

(First embodiment)
Hereinafter, a first embodiment of the present invention (hereinafter referred to as the first embodiment) will be described with reference to the drawings.

  In the first embodiment, in order to predict the power consumption of the entire load device within a certain time period, unlike the conventional case, the operation state of the main load device, the power consumption of the entire load device at that time, and the time at that time An information table storing a history associated with information or the like is used. The main load devices are, for example, showcases and air conditioners with large power consumption. If the predicted power consumption exceeds the amount of power that can be used within the remaining time limit, for example, by controlling the outdoor unit of any air conditioner, further stopping the operation of the entire air conditioner, etc. The power consumption is reduced. This is because even if the outdoor unit of the air conditioner is stopped, indoor comfort can be sufficiently ensured only by the air blowing operation. Moreover, even if the operation of the entire air conditioner is stopped, it is difficult for a person to feel the lack of indoor comfort for a certain amount of time, and the influence is small.

  FIG. 2 shows a connection configuration diagram of the demand control device and the load device according to the first embodiment.

  Referring to FIG. 2, 11 is a power meter that measures the amount of power of the entire load device connected to the power system, 12 is a store controller that is a demand control device, 13 is lighting, 14 is any other load device, 15 is A refrigerator for cooling the showcase, 16 or more showcases, 17 an air conditioner, 18 an outdoor unit, and 19 a refrigerant pipe.

  In the figure, the power meter 11 or the outdoor unit 18, the power line and the communication network constitute one power control system, and the store controller performs demand control. This power control system corresponds to a “target unit” or a “power monitoring unit” in the claims.

  As shown in the drawing, each load device is connected to a controller on the load device side connected to the store controller 12 via a communication network, and each controller sets and controls the connected device. These settings and control are performed in accordance with commands from the store controller 12, but can be manually set and controlled on each device side. The controller also collects operating states of connected devices, and the store controller 12 reads the data as necessary. In addition to the above-described operation, the store controller 12 also performs an operation of reading the amount of power at an arbitrary timing from the power meter. The store controller 12 can be connected to the Internet network, and an administrator can remotely set, control, monitor, measure, and manage the store controller and devices.

  Note that one refrigerator 15 and one or a plurality of showcases 16 form a set by connecting a cooling system. Although only one set is shown in FIG. 2, there may be a plurality of sets. In the first embodiment, there are a plurality of sets. Only one set of the air conditioner 17 and the outdoor unit 18 is shown in FIG. 2, but there are a plurality of sets.

  The cooling system connection means that one refrigerator 15 and one or a plurality of showcases 16 are connected by the same refrigerant pipe 19, and each showcase has an electromagnetic valve. The amount of flowing refrigerant can be controlled at an arbitrary opening. As a result, the amount of cooling refrigerant introduced is adjusted by the control on the showcase side, and the cooling condition is adjusted.

  Since the refrigerator 15 is used for the cooling operation in the showcase 16, the refrigerator 15 operates in conjunction with the cooling operation in the showcase 16, but does not receive ON / OFF control directly from the showcase 16. . The refrigerator 15 is turned off when the solenoid valves for all the showcases are closed. That is, when all the showcases 16 close the solenoid valves, the refrigerator 15 stops its operation, and if one showcase 16 is operating, the refrigerator 15 continues to operate.

  The showcase 16 performs an operation called defrosting during the cooling operation. That is, it is known that frost grows on the cooling pipe (heat exchanger, evaporator) piped to the display shelf, and the cooling performance is lowered by this frost. Therefore, since it is necessary to remove this frost, defrosting operation is performed. In this case, for defrosting, the showcase 16 closes the electromagnetic valve and turns on the defrosting heater. In addition, after performing the defrosting operation only for the specified time considered that the frost has been removed, the defrosting heater is turned off, the solenoid valve is opened again to perform rapid cooling, and the cooling operation is resumed.

  Note that this rapid cooling is one of the causes of peak power consumption. The defrosting operation in each showcase is scheduled in advance, and the defrosting operation is performed at almost the same time every day, but defrosting time is taken into consideration so that the power consumption peaks of many showcases do not overlap. Yes. Scheduling is set from the store controller 12 or the controller of the showcase 16.

  Note that the behavior of each showcase 16 is not constant due to individual differences, the environment around the showcase, and the difference in cooling objects. When a large number of showcases 16 are used at the same time, the characteristics of the power consumption of the entire showcase 16 are grasped. Is extremely difficult. That is, in order to predict the power consumption of the plurality of showcases 16 with high accuracy by theoretical calculation, various factors such as the individual differences described above must be taken into account, and extremely complicated calculation is required. Note that the above-mentioned various factors include the effects of air conditioning, and the effects of air conditioning have a considerable impact on the cooling of products in showcases and cannot be handled by uniform calculations.

  Therefore, in the first embodiment, a dynamic history of the operating state of main load equipment such as a showcase or an air conditioner with large power consumption is stored without using theoretical calculation values for power consumption prediction. Use an information table. In this case, theoretical calculation is unnecessary and prediction according to a specific situation can be performed individually.

  Below, the method of power consumption prediction in the store controller 12 in the first embodiment and the method of controlling the load device will be described.

  FIG. 3 shows a functional block diagram of the store controller 12. Referring to FIG. 3, reference numeral 21 denotes an electric energy acquisition unit that acquires electric energy data from the electric power meter 11, 22 denotes a timer for performing time management such as a clock, and 23 denotes control and processing of each part of the store controller 12. CPU, 24 is a storage unit such as a memory and storage, 25 is a communication unit that communicates with the controller of each load device, 26 is a device control unit that controls each load device, collects data, etc. through the communication unit 25, 27 is a store An input unit 28 for setting the controller 12 or the like, and a display unit 28 such as a liquid crystal screen used for setting the store controller 12 or the like.

  The store controller 12 performs the following operation during the prediction operation.

  The store controller 12 obtains power amount data from the power meter 11 through the power amount acquisition unit 21 in order to calculate the current usable power amount. The available electric energy is calculated from this value and the contract electric energy.

  Next, the store controller 12 predicts future power consumption. For this purpose, communication is performed with the controller of each load device through the device control unit 26 and the communication unit 25, and the current setting and operation state of each device are obtained. In addition, although the setting of each device is made from the store controller 12, there is a possibility that the setting has been manually changed directly in each device, so the latest setting is grasped by obtaining the operating state. To do.

  Based on the current setting and status of each device, the device control unit 26 determines whether there is a matching history part in the history on the information table stored in the storage unit 24 (details will be described later). A search is performed and a matching position (hereinafter referred to as “history position”) is specified. If a plurality of history positions are specified, the history position at the time closest to the current time is selected as the history position to be specified. If the history position corresponding to the current operation state of the load device does not exist on the information table, the search operation is terminated.

  Note that the actual power consumption value in the information table is used to calculate the predicted power consumption value. That is, the predicted power consumption is calculated based on the actual power consumption acquired in the future direction from the identified history position in the information table. If the history position corresponding to the current operating state of the load device does not exist on the information table, it indicates that there is no actual power consumption that can be used in the information table. Further, the predicted power consumption is calculated by linear prediction from the deviation between the accumulated actual power consumption and the current actual power consumption. This method is the same as the conventional prediction method described with reference to FIG. With the above method, the CPU 23 mainly calculates a predicted value of power consumption.

  Finally, the store controller 12 determines whether the used power amount does not exceed the contracted power amount based on the predicted value of the consumed power amount, and if the predicted consumed power amount can be used within the remaining time limit. If it exceeds the maximum value, the operation of a specific device is restricted in order to reduce power consumption thereafter. That is, for example, the device control unit 26 performs control such as stopping an outdoor unit of an arbitrary air conditioner considered to have relatively little influence on others, stopping operation of the entire air conditioner, and the like through the communication unit 25. This reduces power consumption. If it is determined that the reduction in power consumption cannot be achieved, the store controller 12 performs an operation such as sounding a buzzer to inform the person that the reduction in power consumption cannot be achieved, and the person's judgment Leave it to. In other words, it is possible to respond to humans, make decisions, and manually limit the operation of each device to reduce power consumption.

  Next, the aforementioned information table will be described.

  The information table stores a history in which the operation state of the main load device, the power consumption of the entire load device at that time, and time information at that time are associated. The main load devices are, for example, showcases and air conditioners with large power consumption. FIG. 4 shows a schematic diagram of an information table in the first embodiment.

  Referring to FIG. 4, the leftmost column is an item stored in the table, and the right adjacent column is the content of each item stored first, and the operation state and power of each device. Such as quantity. In the first embodiment, a history is added toward the right side of the table every minute. In the above-described history position specification, a search is performed in the time axis direction, so the history position represents the position of any time column. Note that the information table has a history of 10 days, for example.

  The top row is the time in the store controller 12 and is a value obtained from the timer 22.

  The second line from the top represents the operating state of the showcase 1 group. The group referred to here is an aggregate that is synchronously set and controlled among a large number of showcases. Naturally, the operating state is also in a synchronous relationship. That is, this aggregate is different from a set of aggregates composed of a single refrigerator 15 and a plurality of showcases 16 connected to the cooling system. Of the numerical values representing the operating states in the table, zero indicates switching of the operation, that is, switching from the defrosting operation to the cooling operation, or switching from the cooling operation to the defrosting operation, and a positive value after the defrosting operation is performed. Represents elapsed time. A negative value represents an elapsed time since the cooling operation was performed.

  The next line represents a level determined from the value of the internal temperature deviation in one representative showcase from the showcase 1 group, that is, the value of “set temperature-internal temperature”. The method of selecting one representative showcase is arbitrary, and the level is “a” if the absolute value of “set temperature−internal temperature” is within a certain range from zero. If it falls within the range, b and c are further determined to exceed the range but fall within another next range. For example, if the absolute value is greater than or equal to zero and less than or equal to 1, it is defined as a.

The 4th to 9th lines from the top relate to the showcase 2 group or the showcase 4 group, and the content is the same as in the case of the showcase 1.
The tenth and eleventh lines from the top relate to the outdoor units of the air conditioners 1 and 2 and indicate whether the outdoor unit is operating, that is, whether the compressor is operating. In the first embodiment, history is stored for two air conditioners. Note that the history content may include the strong, medium, and weak operation states of the air conditioner in addition to the ON / OFF operation of the outdoor unit.

  The next line is the total actual power consumption at that time. In the first embodiment, a difference value from the actual power consumption one minute ago is stored. In addition, the actual power consumption at that time may be stored as it is.

  The bottom line of the information table is a flag indicating whether or not the actual power consumption is a value that can be used when the actual power consumption is later used as history information. A value of 1 in the table indicates that it is valid (usable).

  The reason why the information on whether or not the power consumption is an effective value is necessary is because, for example, the following reasons can be considered. In other words, when there is a time zone during which demonstration sales etc. were performed for a certain day, power consumption increases different from normal, but this is an exceptional case different from daily sequential operation, If this value is used as learning data at a later date, it is unlikely that highly accurate prediction can be performed, that is, it is highly likely that a large error will occur in the predicted value.

  Here, the case where the power consumption amount W is valid means that M−3σ ≦ W ≦ M + 3σ with respect to the average value M and the standard deviation σ of the power consumption amount at that time in the past 10 days in the information table. A case that falls within the range. This range is a range in which 99% of the samples are included around the average value in the normal distribution. If the power amount is not an effective value, it is not necessary to store it in the table, but in the first embodiment, all data is stored in the table in time series.

  Note that the values of various items in the information table of FIG. 4 are values for convenience, and thus do not represent actual operations.

  Finally, with reference to FIG. 5, the operation | movement in the store controller 12 is demonstrated using a flowchart. In this operation, every time an arbitrary time (for example, 1 minute) elapses, the prediction of the power consumption, the comparison between the predicted value and the power consumption at that time, and the contract power consumption, and the corresponding Load device control and information table update processing are performed.

  In step S101, the power amount acquisition unit 21 acquires the power amount at the current time from the power meter 11, and obtains a difference value from the power amount one minute ago.

  In step S102, the difference value of the electric energy acquired in step S101 is added to the integrated electric energy one minute ago, and the integrated electric energy from the start time of the set time limit (30 minutes) at the present time is calculated. The integrated power amount is reset every set time limit.

  In step S103, information regarding the time is obtained from the timer 22, and the current time is obtained.

  In step S104, the operation state (defrosting operation / cooling operation) of the current showcase and the operation state of each device such as the internal temperature and the outdoor unit operation state of the air conditioner are obtained through the communication unit 25.

  In step S105, based on the current operation state of each device obtained in step S104, the information table stored in the storage unit 24 is referred to search for the corresponding history position, and the history position is determined. Identify. At that time, a search is performed with reference to the above-mentioned valid flag.

  In step S106, it is determined whether the history position has been identified. That is, if the corresponding history position can be identified on the information table in step S105, the process proceeds to step S108, and if not, the process proceeds to step S107.

  In step S107, the history position is not specified on the information table, that is, since there is no value on the history of the matching situation, the deviation power amount between the previous integrated power amount and the current integrated power amount is used. The amount of power consumption consumed between the current time and the end of the set time period is predicted by linear prediction. This method is the same as the conventional prediction method described with reference to FIG.

  In step S108, when the history position is searched, it is determined whether or not there are corresponding places in a plurality of places. If multiple locations are applicable, the process proceeds to step S110. If only one location is applicable, the process proceeds to step S109.

  In step S109, using the actual power consumption stored in the information table, the power consumption consumed between the present time and the end of the set time limit is predicted. Specifically, the value of the electric energy for the time interval from the current time to the end of the set time period after the history position specified in the table is added to obtain the predicted value.

  In step S110, since the history position candidates correspond to a plurality of locations, the history position closest to the current time in the past 10 days is selected and specified regardless of the date. If there are a plurality of history positions closest to that time at the same time, the history position on the current day or the day close to the current day is selected and specified.

  In step S111, as in step S109, the power consumption consumed between the current time and the end of the set time limit is predicted using the actual power consumption stored in the information table. Specifically, the value of the electric energy for the time interval from the current time to the end of the set time period after the history position specified in the table is added to obtain the predicted value.

  In step S112, a comparison is made as to whether or not the sum of the accumulated power consumption and the predicted value of the power consumption exceeds the contract power consumption. If it is found that the limit is exceeded, the operation of each device is restricted and controlled. In the first embodiment, specifically, operation restriction such as stopping an appropriate number of outdoor units of air conditioners is performed in consideration of priority. In addition, control is performed such as stopping the operation of the entire air conditioner indoor / outdoor and further stopping all the air conditioners.

  In step S113, the operation state and time information of each device at the current time, the difference value between the power amount acquired in step S101, and the like are added to the information table as history information. Specifically, the showcase defrosting / cooling elapsed time and the internal temperature deviation are determined, and the difference value between the result and the electric energy is added to the information table. At this time, the effective flag is also set based on the difference value of the electric energy.

In addition, when the operation state of each device obtained from the controller of each load device through the device control unit 26 and the communication unit 25 described above is different from the operation state set by the store controller 12, the device is manually manually operated on the device. Since the setting has been changed and is considered to be an exceptional case different from the daily sequential operation, it is unlikely that accurate prediction will be possible by using this value at a later date, that is, the predicted value is large. There is a high possibility that errors will occur. Therefore, information to that effect may be stored as an identification bit, for example. In this case, when the history position is specified in step S105, the search is performed in consideration of the identification bit together with the above-described valid flag.
(Second Embodiment)
Hereinafter, a second embodiment of the present invention (hereinafter referred to as a second embodiment) will be described with reference to the drawings.

The connection configuration of the demand control device and the load device according to the second embodiment is basically as follows:
This is the same as the connection configuration of the demand control device and the load device of the first embodiment shown in FIG. Therefore, in the description of the demand control device, the load device connection configuration, and the load device according to the second embodiment, the description of the portions common to the first embodiment is omitted.

  In the demand control device according to the second embodiment, the store controller 12 predicts the power consumption of the entire load device within a predetermined time period for performing demand control for each predetermined prediction cycle. When predicting the power consumption, the store controller 12 uses the average value of the increase in the operation state of the main load device in the past and the power consumption of the entire load device at that time (hereinafter referred to as the average power consumption). Or an information table in which the maximum value of power consumption for each prediction period (hereinafter referred to as maximum power consumption increase) and time information at that time are associated. When the remaining time until the end of the predetermined time period is large, the store controller 12 predicts the power consumption of the entire load device based on the average power consumption increase amount. When the remaining time is small, the store controller 12 is based on the maximum power consumption increase amount. Predict the power consumption of the entire load device. The main load equipment is, for example, a showcase with a large amount of power consumption, an air conditioner, or an outdoor unit of the air conditioner.

  In the showcase 16 shown in FIG. 2, when the store is closed, the interior lighting is turned off and a cover is attached so as not to let cool air escape. Thereby, the heat load of the showcase 16 is reduced. Further, the set temperature in the target cabinet of the showcase 16 is also changed. Such an operating state of the showcase 16 during the closing of the store is called NSB (Light Set Back).

Even if the showcase 16 is open, the illumination intensity of the interior lighting may be lowered and the interior set temperature may be increased for the purpose of saving energy. Such an operating state of the showcase during the opening of the store is called DSB (Demand Set Back).

When the operation state of the showcase 16 is NSB or DSB, the power consumption by the showcase 16 is small.

  Whether the operation state of the showcase 16 is NSB or DSB has a considerable influence on the prediction of the power consumption of the entire load device. Therefore, in the second embodiment, the power consumption of the entire load device is accurately determined. In order to predict, it is considered whether the operation state of the showcase 16 described above is NSB or DSB.

  The controller of the showcase 16 determines whether the showcase 16 is NSB or DSB and notifies the store controller 12 of it.

  When there are a plurality of showcases 16, the timing at which the operation state of the plurality of showcases 16 becomes NSB and the timing at which the DSB becomes DSB is normally scheduled to be substantially the same for the entire showcase 16. Yes. However, it may be scheduled separately for each showcase 16.

  Moreover, the showcase 16 performs rapid cooling until the inside temperature which is heated and becomes high temperature reaches a predetermined set value after the defrosting operation.

  Such rapid cooling after the defrosting operation of the showcase 16 is referred to as pull-down operation. When there are a plurality of showcases 16, when most of the plurality of showcases 16 are simultaneously pulled down, the power consumption of the refrigerator 15 that cools all the showcases 16 is drastically and greatly increased. To increase.

  Whether or not the operation state of the showcase 16 is a pull-down operation has a considerable influence on the prediction of the power consumption of the entire load device. In the second embodiment, the power consumption of the entire load device is accurately predicted. Therefore, pull-down operation is considered as one of the operation states of the showcase 16.

  It should be noted that how long the individual showcases 16 start after the start of the defrosting operation before the pull-down operation is performed depends on the individual operating conditions of the showcases 16 and cannot be grasped in advance. The controller of the showcase 16 determines whether each showcase 16 is in a pull-down operation while monitoring the operation pattern of each showcase 16 and notifies the store controller 12 of it.

  In addition to the three operating states of heating, cooling, and drying, the air conditioner 17 can take a fourth operating state (hereinafter referred to as “thermo”) after the temperature in the store reaches the target temperature. it can. The controller of the air conditioner 17 determines which of the four operating states the air conditioner 17 is in and notifies the store controller 12 of it.

  The controller of the air conditioner 17 also determines whether the outdoor unit 18 is on or off, and notifies the store controller 12 of it.

  Here, the outdoor unit 18 being off is a case where the air conditioner 17 is completely stopped, and it is determined that the other units are on.

  In FIG. 2, only one air conditioner 17 is connected to the outdoor unit 18, but a plurality of air conditioners may be connected.

  FIG. 6 shows a schematic diagram of an information table according to the second embodiment.

  The information table according to the second embodiment is constructed for each prediction cycle by the store controller 12.

  The information table includes a “time zone” column 30 in which a time zone for each predetermined time is stored, an “entire showcase” column 31 in which the operating state of the showcase 16 is stored, a “showcase No. 1” column 32, “ Showcase No. 2 ”column 33,“ air conditioner ”column 34 in which the operating state of the air conditioner 17 is stored,“ outdoor unit ”column 35 in which the operating state of the outdoor unit 18 is stored, and the power consumption of the entire load device The “power consumption amount” column 36 for storing information related to.

  In the “time zone” column 30, a time zone obtained by dividing 24 hours every predetermined time, for example, in the information table of FIG. 6, “24 hours are divided every 2 hours” “2 o'clock”, “2 o'clock to 4 o'clock”... Are stored.

  The “showcase whole” column 31 includes an “NSB” column 37 and a “DSB” column 38 for storing identification information indicating whether the entire showcase 16 is NSB or DSB. When the entire showcase 16 is NSB, “ON” is stored in the “NSB” column 37 and “OFF” is stored in the “DSB” column 38.

  In the second embodiment, even when the showcase 16 is composed of a plurality of showcases, the switching between NSB and DSB is scheduled to be the same timing for the entire showcase 16. Yes.

  When scheduling is performed at different timings for each showcase, an “NSB” field and a “DSB” field are provided for each showcase.

  In the “Showcase No. 1” column 32 and the “Showcase No. 2” column 33, a plurality of showcases 16, for example, two showcases No. 1 (not shown) and the showcase No. 1 are displayed. 2 (not shown), showcase No. 1 and showcase No. 1 2 Each operation state is stored.

  Specifically, the “Showcase No. 1” column 32 shows the Showcase No. 1 column. 1 includes a “defrost” column 39 and a “pull-down” column 40 for storing identification information indicating whether the operation state 1 is a defrosting state or a pull-down state.

  Showcase No. When the operation state 1 is the defrosting state, “ON” is stored in the “defrosting” column 39 and “OFF” is stored in the “pull-down” column 40.

  Showcase No. When the operation state 1 is a pull-down operation state, “OFF” is stored in the “frosting” column 39 and “ON” is stored in the “pull-down” column 40.

Showcase No. When the operation state 1 is the normal cooling operation, “OFF” is stored in the “frost removal” column 39 and “OFF” is also stored in the “pull-down” column 40.
The “Showcase No. 2” column 33 includes a “Defrosting” column 41 and a “Pulldown” column 42, and is similar to the “Defrosting” column 39 and the “Pulldown” column 40 of the “Showcase No.1” column 32. “ON” and “OFF” are stored according to the rule.

  In the “air conditioner” column 34, “cooling” column 43, “heating” column 44 for storing identification information indicating whether the air conditioner 17 is in an operation state of cooling, heating, dry, or thermo, It consists of a “dry” column 45 and a “thermo” column 46.

  When the air conditioner 17 is in any one of the four operation states, “ON” is stored in the column corresponding to the state, and “OFF” is stored in the other columns. When the air conditioner 17 is stopped, “OFF” is stored in all the columns constituting the “air conditioner” column 34.

The “outdoor unit” column 35 stores identification information indicating whether or not the outdoor unit 18 has stopped.
When the outdoor unit 18 is stopped, “OFF” is stored in the “outdoor unit” column 35, and when it is not stopped, “ON” is stored.

  The “power consumption” column 36 has an “average value” column 47 in which the average power consumption increase amount of the entire load device is stored, and an “average frequency” column 48 in which the number of times the average power consumption increase has been calculated is stored. It consists of a “maximum value” column 49 in which the maximum amount of increase in power consumption of the entire load device is stored, and a “recording date” column 50 in which the date and time when the maximum power consumption increase is recorded are stored.

In the information table shown in FIG. 6, for example, in the row with the number 51, the main load device is
“Time zone” column 30 = “0 o'clock to 2 o'clock”
“NSB” column 37 of “the entire showcase” 31 = “ON”
“DSB” field 38 of “the entire showcase” 31 = “OFF”
“Showcase No. 1” column 32 “Defrosting” column 39 = “OFF”
“Showcase No. 1” column 32 “pull-down” column 40 = “OFF”
“Showcase No. 2” column 33 “Defrost” column 41 = “OFF”
“Show pull-down” column 42 of “Showcase No. 2” column 33 = “OFF”
“Cooling” column 43 of “Air conditioner” column 34 = “ON”
“Heating” column 44 of the “air conditioner” column 34 = “OFF”
“Dry” field 45 = “OFF” in “Air conditioner” field 34
“Thermo” column 46 of the “air conditioner” column 34 = “OFF”
“Outdoor unit” column 35 = “ON”
It shows that an operation state pattern is taken.

On the line numbered 51,
The “average value” field 47 of the “power consumption” field 36 = 1500 kW
“Power consumption” column 36 “Average number” column 48 = 10 times “Power consumption” column 36 “Maximum value” column 49 = 1750 kW
"Recording date" field 50 of "Power consumption" field 36 = 2005.3.1
Further, in the load equipment which has been main in the past, the same operation state pattern as the above-described operation state pattern is generated 10 times (the “average number of times” column 48 = 10 times in the “power consumption” column 36), The average power consumption increase amount is 1500 kW (“average value” field 47 = 1500 kW in the “power consumption” field 36), and the maximum power consumption increase is 1750 kW (“maximum value” field in the “power consumption” field 36. 49 = 1750kw), indicating that the date when the maximum power consumption increase was recorded was March 1, 2005 (the “recording date” column 50 = 2005.3.1 in the “power consumption” column 36). Yes.

  As described later, in the second embodiment, the upper limit value of the value stored in the “average number of times” column 48 of the “power consumption” column 36 is determined, and the upper limit value is 100 times.

  In the demand control apparatus according to the second embodiment, the store controller 12 learns the operation state of the main load device, thereby constructing an information table, and based on the information table, the power consumption of the entire load device is determined. Predict.

  First, a method for constructing the information table by the store controller 12 will be described.

  The store controller 12 displays the showcase number of the showcase 16 constituting the showcase 16 at a predetermined timing. 1. Showcase no. 2 showcase No. 1 and showcase No. The identification information indicating the operation state of the outdoor unit 18 is acquired from the controller of the air conditioner 17, the identification information indicating the operation state of the air conditioner 17 is acquired from the controller of the air conditioner 17. Furthermore, the power consumption of the entire load device is acquired from the power meter 11 at the same timing as the predetermined timing. The power meter 11 integrates the power consumption of the entire load device for each predetermined time period during which demand control is performed. In the second embodiment, the predetermined time period for performing the demand control is a 30-minute time period, and the power meter 11 newly starts to accumulate power consumption for 30 minutes from the 0-minute time point of the 30-minute time period. It should be noted that the timing at the time of 0 minute needs to be synchronized with the timer of the watt hour meter used by the electric power company for calculating the electricity charge, and in the second embodiment, such adjustment is made in advance.

  The store controller 12 acquires the acquired showcase number. 1. Showcase no. 2. Whether the information table has the same pattern as the operation state pattern of the main load equipment consisting of the identification information indicating the operation state of the air conditioner 17 and the outdoor unit 18 and the time zone to which the time at which the identification information was acquired belongs. Search for no. As a result of the search, when the operation state pattern exists in the information table, an “average value” column 47, an “average number” column 48, and a “maximum value” column 49 corresponding to the operation state pattern in the information table. The “record date” column 50 is updated based on the power consumption acquired from the power meter 11.

The store controller 12 calculates the average power consumption increase amount by the average value processing of the following equation (1), and stores it in the “average value” column 47.
Average power consumption increase = (average power consumption increase recorded in the past x number of recordings + increase)
/ (Recording count + 1) ... (1)
Here, the average power consumption increase amount recorded in the past is the average power consumption increase amount previously stored in the “average value” column 47, and the number of times of recording is the calculation of the average power consumption increase amount. The amount of increase is the difference between the power consumption of the entire load device acquired from the current power meter 11 and the power consumption of the entire load device acquired by the previous power meter 11.

  When the number of times of recording reaches 100, the store controller 12 uses the learning result to predict the power consumption of the entire load device, assuming that the learning of the operation state of the main load device has been temporarily completed. To do.

  Further, the store controller 12 stores the number of times of recording in its built-in storage device (not shown) and stores it in the “average number of times” column 48 in the information table. However, after the number of recordings reaches 100, the updating of the “average number” column 48 is stopped.

  In the maximum value processing, the maximum power consumption increase amount previously stored in the “maximum value” column 48 is compared with the increase amount calculated from the power consumption amount acquired from the power meter 11 this time. Is larger than the maximum power consumption increase amount stored in the storage area, the increase amount is stored in the “maximum value” column 49 as the maximum power consumption increase amount, and the acquired date is stored in the “recording date” column 50 Say.

  Further, in the maximum value processing, even if the increase amount is smaller, the past maximum power consumption occurrence date stored in the “recording date” column 50 is compared with the date acquired this time, and the past maximum If one month or more has elapsed since the date of occurrence of the increase in power consumption, the “maximum value” column 49 and the “recording date” column 50 are updated with the increase amount. In this update, the newly calculated average power consumption increase amount is compared with the increase amount, and updated with the larger value.

  If the acquired operation state pattern does not exist in the information table, a storage area is newly added to the information table, and the operation state pattern is stored in the corresponding column of the added storage area. At this time, the “average value” column 47 and the “maximum value” column 49 store the amount of increase calculated from the power consumption acquired from the power meter 11, and the “average number” column 48 stores “1” and “record”. The “day” column 50 stores the date on which the power consumption is acquired.

  In this way, the store controller 12 learns the operation state of the main load device every predetermined period, and constructs the information table.

  The store controller 12 stores the operation state pattern of the main load device acquired last time in its internal storage device (not shown), and stores the previous operation state pattern stored this time and the operation state acquired this time. If there is a difference with the pattern other than the “time zone” column 30, it is determined that the operating state of the main load device has changed during the prediction period. In this case, since the influence of the different operation state has arisen on the acquired power consumption, the acquired operation state pattern and the power consumption are not reflected in the information table.

  Further, the store controller 12 does not prepare an information table storage area corresponding to all of the main operation state patterns of the load device in advance, but creates an information table storage area each time a new operation state pattern is generated. Expand. There are many operation state patterns that do not occur even if there is a possibility that the operation state pattern generated varies depending on the environmental conditions of the store. Therefore, by using the method of expanding the storage area of the information table each time a new operation state pattern is generated as described above, storage for operation state patterns that do not actually occur is possible. The area can be reduced.

  Next, a method of predicting the power consumption of the entire load device for demand control by the store controller 12 will be described.

  The store controller 12 updates the information table based on the operation state pattern of the main load device acquired every prediction cycle, and the operation state pattern continues until the end of the 30-minute time limit for performing demand control. Predict the power consumption of the entire load equipment. The prediction cycle is generally from 10 seconds to 1 minute, and in the second embodiment, it is 30 seconds.

  Specifically, it is determined whether or not the value in the “average number of times” column 48 corresponding to the same operation state pattern as the acquired operation state pattern is 100 in the information table. When the value of the “average number of times” column 48 does not reach 100, it is determined that the operation state pattern is not sufficiently learned, and is calculated from the power consumption of the entire load device acquired from the power meter 11. The power consumption amount at the end of the 30-minute time period is predicted using only the increased amount. That is, the store controller 12 calculates the predicted power consumption of the entire load device at the end of the 30-minute time period based only on the power consumption of the entire load device acquired from the power meter 11 based on the following equation (2).

Predicted power consumption = (Increase amount / Prediction cycle) x Remaining time + Power consumption to date
... (2)
Here, the amount of increase is the same as the amount of increase in Equation (1). The remaining time is the remaining time from the current time until the end of the 30-minute time limit. The power consumption up to now is the cumulative power consumption from the start of demand control to the present.

  For example, when 4 minutes and 30 seconds have elapsed since the start of demand control, the amount of increase is the amount of increase in power consumption when 4 minutes and 30 seconds have elapsed since the start of demand control from the amount of power consumed when 4 minutes have elapsed since the start of demand control. The remaining time is 1530 seconds, and the power consumption up to the present time is the cumulative power consumption from the start of demand control to the point of 4 minutes 30 seconds.

  When the number of times in the “average number of times” column 48 corresponding to the operation state pattern has reached 100, based on the value in the “average value” column 47 or “maximum value” column 49 corresponding to the operation state pattern. To predict the power consumption at the end of the 30-minute time limit.

  Specifically, when the remaining time until the end of the 30-minute time period is large, for example, when the remaining time is 10 minutes or more, the average power consumption increase amount stored in the “average value” column 47 is read, When the remaining time is short, for example, when it is less than 10 minutes, the maximum power consumption increase amount stored in the “maximum value” column 49 is read, and the entire load equipment at the end of the 30-minute time limit is read based on the following equation (3). Calculate the predicted power consumption.

Predicted power consumption = (Increase amount / Prediction cycle) x Remaining time + Power consumption to date
... (3)
Here, the increase amount is the average power consumption increase amount when the remaining time is 10 minutes or more, and the maximum power consumption increase amount when the remaining time is less than 10 minutes.

  When the remaining time is 10 minutes or more, the predicted power consumption is calculated from Equation (3) based on the average power consumption increase.

  When the remaining time is less than 10 minutes, the predicted power consumption is calculated based on the maximum power consumption increase.

  In the demand control device according to the second embodiment, the store controller 12 performs, for example, the following control on the main load device managed by the store controller 12 based on the predicted power consumption calculated by the method described above. Do.

  When the predicted power consumption is larger than a predetermined value set in advance, it is assumed that the contract power consumption may be exceeded, and the operation control for stopping the air conditioner is performed according to a predetermined rule. Although this rule is determined in advance, the air conditioner to be stopped may be determined according to the degree exceeding the predetermined value, the priority order of the air conditioners to be stopped may be determined, or the air conditioner to be stopped Alternatively, a method may be used in which the rotation order of the air conditioners that are stopped is determined in advance so as not to be biased. Although a method of stopping the operation of the showcase 16 or the refrigerator 15 is also conceivable, the air conditioner 17 in the second embodiment may adversely affect the storage state of the food stored in the storage of the showcase 16. Is supposed to be shut down.

  In addition, when the remaining time until the end of the 30-minute time limit is more than 10 minutes, there is a margin until the end, so the demand control is performed with emphasis on maintaining the comfort of the store / facility . Conversely, when the remaining time is less than 10 minutes, emphasis is placed on the fact that the power consumption does not exceed a predetermined value, and control is performed to reduce the power consumption early.

  FIG. 7 shows a flowchart of the demand control operation by the store controller 12.

  In step S200, the store controller 12 learns the main operating state of the load device and the power consumption of the entire load device at this time. That is, the store controller 12 acquires the operation state pattern of the load device and the power consumption of the entire load device, which are main for each prediction cycle, and constructs an information table shown in FIG.

  In step S300, the store controller 12 predicts the entire load device based on the acquired operation state pattern and the power consumption amount or the average power consumption increase amount or the maximum power consumption increase amount corresponding to the acquired operation state pattern in the information table. Calculate power consumption.

  Next, in step S400, the store controller 12 performs operation control of the load device managed by the store controller 12 so as not to exceed the contracted power amount within a predetermined time period based on the predicted power consumption.

  The store controller 12 repeats steps S200, S300, and S400 for each prediction cycle.

  FIG. 8 shows a flowchart of detailed operations of the store controller 12 in the learning process of step S200 in the flowchart shown in FIG.

  In step S201, the store controller 12 acquires the operating state identification information from the main load device, acquires the power consumption of the entire load device from the power meter 11, and proceeds to step S202.

  In step S202, the store controller 12 acquires the time at which the identification information and the power consumption are acquired from its own clock function, and proceeds to step S203.

  In step S203, the time zone to which the acquired time belongs is calculated, and the process proceeds to step S204.

  In step S204, the information table is searched, and it is determined whether or not there is an operation state pattern in the information table that is the same as the operation state pattern consisting of the acquired operation state of the main load device and the calculated time zone. If the same operation state pattern exists, the process proceeds to step S205, and if not, the process proceeds to step S206.

  In step S205, a row in which the same operation state pattern as the acquired operation state pattern exists in the information table is selected, and the process proceeds to step S207.

  In step S206, a new line is added to the information table as an area for storing the value calculated from the acquired operation state pattern and power consumption, the operation state pattern is stored in the corresponding column, and the process proceeds to step S207.

  In step S207, the average power consumption increase amount, the average number of times, the maximum power consumption increase amount, and the maximum power consumption increase amount of the entire load device corresponding to the operation state pattern of the selected row or the added row in the information table have occurred. Update.

  FIG. 9 shows a flowchart of detailed operation of the store controller 12 in the power consumption amount prediction process of step S300 in the flowchart shown in FIG.

  In step S301, the store controller 12 acquires the operating state identification information from the main load device, acquires the power consumption of the entire load device from the power meter 11, and proceeds to step S302.

  In step S302, the store controller 12 acquires the time when the identification information and the power consumption are acquired from its own clock function, and the process proceeds to step S303.

  In step S303, a time zone to which the acquired time belongs is calculated, and the process proceeds to step S304.

  In step S304, the information table is searched, and it is determined whether or not there is an operation state pattern in the information table that is the same as the acquired operation state pattern of the load device and the calculated operation state time zone. If the same operation state pattern exists, the process proceeds to step S305, and if not, the process proceeds to step S306.

  In step S305, it is determined whether or not the remaining time until the end of the predetermined time limit is within 10 minutes. If it is within 10 minutes, the process proceeds to step S307, and if not, the process proceeds to step S308.

  In step S306, the amount of increase in power consumption in the prediction cycle is calculated from the acquired power consumption of the entire load device, and the process proceeds to step S309.

  In step S307, the maximum power consumption increase amount corresponding to the same operation state pattern as the acquired operation state pattern in the information table is updated, and this is set as the increase amount, and the process proceeds to step S309.

  In step S308, the average power consumption increase amount corresponding to the same operation state pattern as the acquired operation state pattern in the information table is updated, and this is set as the increase amount, and the process proceeds to step S309.

  In step S309, the predicted power consumption of the entire load device at the end of the predetermined time period is calculated based on the above formula (2) or (3).

  FIG. 10 shows a flowchart of detailed operations of the store controller 12 in the demand control process of step S400 in the flowchart shown in FIG.

  In step S401, it is determined whether the predicted power consumption exceeds the contracted power amount contracted with the power company. If the contracted power amount is exceeded, the process proceeds to step S402, and if not, the process ends.

  In step S402, when there are a plurality of air conditioners 17, it is determined whether all of the air conditioners 17 are stopped. If they are stopped, the process proceeds to step S403. Otherwise, the process proceeds to step S404. move on.

  In step S403, an alarm buzzer (not shown) provided in the store controller 12 is sounded to alert the store manager or the like.

  In step S404, all the air conditioners 17 are stopped.

  As described above, in the demand control device according to the second embodiment, the store controller 12 includes the operation state pattern of the main load device managed by the store controller 12, the average power consumption increase amount of the entire load device at that time, The maximum power consumption increase amount and the power consumption increase amount are learned, and an information table is constructed.

  Further, the store controller 12 calculates based on the average power consumption increase or maximum power consumption in the information table corresponding to the acquired operation state pattern of the main load device, or from the acquired power consumption of the entire load device. Based on the increased power consumption, the predicted power consumption at the end of the predetermined time period is calculated.

  Furthermore, when the predicted power consumption exceeds the contract power consumption, the store controller 12 stops the main load device managed by the store controller 12, such as an air conditioner.

  According to the second embodiment, the store management controller 12 does not manage and control all load devices in the store. Therefore, even if the operation state of the main load device managed by the store controller 12 is the same, it is affected by the power consumption of the load device not managed by the store controller 12, and the power consumption of the entire load device varies. May occur. In such a case, the power consumption at the end of the 30-minute time period is predicted based only on the average power consumption increase amount of the entire load device based on the operation state of the main load device, and if demand control is performed based on this, Power consumption may exceed the predicted power consumption. Even in such a case, if the remaining time until the end of the 30-minute time limit is large, the contracted power amount is not exceeded at the end of the 30-minute time period by calculating the predicted power consumption again and correcting the demand control. However, if the remaining time until the end of the 30-minute time limit is small, even if the predicted power consumption is calculated again and the demand control is corrected, the contracted power consumption at the end of the 30-minute time limit There is also a risk of exceeding.

  In the second embodiment, when the remaining time until the end of the 30-minute time period is small as described above, the predicted power consumption is calculated by adopting the maximum power consumption increase instead of the average power consumption increase. Even if the actual power consumption increases slightly, it is possible to avoid exceeding the predicted power consumption.

  In addition, the influence of the time zone can be considered as one of the causes of the variation in the relationship between the main load device operating state and the power consumption of the entire load device. For example, the operating state and power consumption of the showcase 16 or the refrigerator 15 are affected by the amount of food items brought into the storehouse of the showcase 16 and the number of visitors. The number depends on the time zone. Further, there is a high possibility that other load devices not managed by the store controller 12 also depend on the time zone. Therefore, the predicted power consumption can be calculated with high accuracy by classifying the operation states of the main load devices as in the second embodiment for each time zone.

  Also, showcase no. 1 and showcase no. When 2 is in the defrosting operation, even if the predicted power consumption is calculated based on the operation state pattern, the pull-down operation starts at an unexpected time. The amount may be exceeded, and there is a risk that the response in demand control will be delayed.

  Therefore, in the second embodiment, the showcase No. 1 and showcase no. When 2 is defrosting operation, compared to the predicted power consumption calculated based on the operation state pattern when these are changed to pull-down operation, the predicted power consumption in the pull-down operation is larger In this case, the predicted power consumption may be adopted. In this process as well, whether to use the maximum power consumption increase amount or the average power consumption increase amount follows the rule of the remaining time.

  In the second embodiment, the operating conditions of the showcase 16, the air conditioner 17, and the outdoor unit 18 are adopted as the main load devices. Large lighting 13 and other equipment 14 may be employed.

In the second embodiment, the operation state pattern of the main load device indicates the binary value of each operation pattern by ON and OFF except for the time zone. However, in the actual software, “1” and “ 0 ”. Since all can be represented by binary values of on / off, one bit may be assigned to each item on the software memory. According to this, the memory capacity is naturally smaller than allocating about 1 byte to each item. In addition, learning data can be easily handled when searching for an operation state pattern in the information table. In FIG. 6, the time zone stored in the “time zone” column 30 is also expressed as 0 to 2 o'clock, but since there are 12 time zones in 24 hours, it is expressed by 4 bits from 0 to 11. It is good to do.
(Third embodiment)
Hereinafter, a third embodiment of the present invention (hereinafter referred to as a third embodiment) will be described with reference to the drawings.

  The connection configuration between the demand control device and the load device according to the third embodiment is basically the same as the connection configuration between the demand control device and the load device according to the first embodiment shown in FIG. Therefore, in the description of the demand control device, the load device connection configuration, and the load device according to the third embodiment, the description of the portions common to the first embodiment is omitted. The learning process, the prediction process, and the demand control process according to the third embodiment are basically the same as the processes shown in the second embodiment. Therefore, among the above three processes according to the third embodiment, the description of the parts common to the second embodiment is omitted.

  As described above, the showcase 16 performs the pull-down operation after the defrosting operation until the heated interior temperature reaches a predetermined set value. This pull-down operation is compared with the heating operation after the defrosting operation. However, the power consumption is a large value. When there are a plurality of showcases 16 and when most of the plurality of showcases 16 simultaneously perform pull-down operation, the power consumption of the refrigerator 15 increases rapidly. As described above, whether or not the operation state of the showcase 16 is the pull-down operation has a considerable influence on the prediction of the power consumption of the entire load device.

  On the other hand, how long each individual showcase 16 starts after the start of the defrosting operation before the pull-down operation is performed and how long the pull-down operation is performed depends on individual differences of each showcase 16 and each showcase. Since it depends on the environment where 16 is placed, it cannot be grasped in advance and the range of variation is enormous. Considering the case of creating the information table in FIG. 6 from this, for example, the operation state pattern acquired in step S201 in FIG. 8 may exist with a high probability even if it is slightly different from the same time zone of the previous day. . As a result, the main case may be that the process branches to step S206 in step S204, that is, a new row is added to the information table. As a result, the amount of information in the information table becomes enormous, and the time required for search and the required memory amount increase.

  Therefore, in the third embodiment, in order to eliminate the influence of the variation as described above, a modeling pattern is created by modeling the operation state in the time series direction of the showcase. In consideration of the actual power consumption effect, the model is focused on pull-down operation, showcase defrosting, and draining operation that have a considerable impact on the prediction of the power consumption of the entire load equipment. (Hereinafter referred to as defrost model). Specifically, instead of acquiring the operating state of the load device that was performed in step 201, based on the schedule of the defrosting operation, the operating state of each showcase is created as a model by applying the above model (hereinafter referred to as the model) This is referred to as an operation state model), and each showcase group is considered to be in the state shown in the operation state model, and the information is referred to. Therefore, it is not necessary to use information such as whether or not the showcase 16 is in the pull-down operation, which is notified from the controller of the showcase 16 to the store controller 12 as in the first and second embodiments.

  Specifically, in the defrost model, each time required for defrosting, draining, and pull-down is set in advance. For each set time, an optimal value is determined in advance by experiment, measurement, or the like. In the third embodiment, they are 5 minutes, 10 minutes, and 30 minutes, respectively. This setting may be changed for each showcase. Using this defrosting model and defrosting operation schedule, an operating state model of each showcase at each time is created. With this operating state model, it is possible to determine which showcase can be considered in what state when what time.

  An operating state model is shown in FIG. This table corresponds to the schematic diagram of the information table in the first embodiment shown in FIG. The vertical axis direction represents the state of each showcase. “1” in the figure indicates that the showcase is in that state. For example, taking 8: 2 as an example, the showcase 1 is in the defrosting state, and the showcases 2 and 3 are in the pull-down state.

  FIG. 12 shows a second table in the third embodiment, which corresponds to FIG. 6 in the second embodiment. The average value and the maximum value in the figure are values in one minute. Although not shown in the third embodiment, this table may include operating states obtained through the communication unit 25 in FIG. 3 such as load devices other than the showcase, such as air conditioners and outdoor units. .

  In the second table in the third embodiment, the case where NSB and DSB are used is omitted from the description. The defrosting model also considers draining operation of water generated by melting ice after defrosting in the showcase. The ON / OFF information for each showcase defrosting, draining, and pull-down operation in the information table, as described above, is created by applying the defrosting model to the defrosting operation schedule. Depends on the operating state model. This is included in the description of the flow that will be described in detail later.

  FIG. 13 shows a flowchart of the demand control operation by the store controller 12 in the third embodiment. In the third embodiment, the loop (steps S200 to S400) in the flow is executed every minute. In the third embodiment, an example in which demand control is performed in units of 30 minutes will be described. Therefore, hereinafter, 30 minutes after the start time is called the final time period, and the timing of the current time within the 30 minutes is called the current time period (within a range of 0 to 29 minutes). Note that step S500 is added as compared to the case of FIG. 7 of the second embodiment.

  In step S500, the defrosting operation schedule set in the store controller 12 is read. The operating state model described above is created by applying a defrosting model to this.

  FIG. 14 shows a flowchart of the detailed operation of the store controller 12 in the learning process of step S200 in the flowchart shown in FIG. Compared with the case of FIG. 8 of the second embodiment, step S201 is not performed. This is because it is not necessary to acquire the operating state of the load device in the third embodiment as described above. Further, steps S210 to S212 are added compared to the case of FIG. These are described in detail below.

  In step S210, the current operating state of the showcase is extracted from the operating state model created in step S500.

  In step S211, the current power consumption of the entire load device is acquired from the power meter 11.

  In step S212, the information obtained in steps S210 and S211 is stored in a memory (this data is referred to as a first table, not shown). Although not used in the third embodiment, the first table may include operating states obtained through the communication unit 25 in FIG. 3 such as load devices other than the showcase, such as air conditioners and outdoor units. .

  In steps S213 to S215, in accordance with steps S204 to S206 of the second embodiment, it is confirmed whether the same data exists in the second table, and a value calculated from the operation state pattern and the power consumption is recorded in the second table. Decide whether to add a new row as an area for storage, or select a row (existing row) that has the same operation state pattern as the acquired operation state pattern in the second table as an update target To do.

  In step S216, in accordance with step S207 of the second embodiment, the operation state pattern and the average power consumption increase amount, the average number of times, the maximum power consumption increase amount, and the maximum power consumption increase amount corresponding to the entire load device are displayed. The second table stored for each time zone is updated. Specifically, the date and time when the average power consumption increase amount, the average number of times, the maximum power consumption increase amount and the maximum power consumption increase amount of the entire corresponding load device occur in the above-described additional row or selection row in the information table, Store. The second table corresponds to the information table created in step S207 of FIG. 8, but its constituent elements are different from those of the second embodiment (see FIG. 11).

  The step S216 is executed every two hours, and the second table for the two hours is updated. The update timing is, for example, 8 o'clock just, 10 o'clock just, or the like.

  FIG. 15 shows a flowchart of detailed operations of the store controller 12 in the power consumption amount prediction process of step S300 in the flowchart shown in FIG. Compared to the case of FIG. 9 of the second embodiment, step S301 is not performed. This is because it is not necessary to acquire the operating state of the load device in the third embodiment as described above. Further, compared to the case of FIG. 8, steps S310 to S312 and steps S317 and S318 are added. Note that steps S313 to S316 are similar to steps S306 to S309 of the second embodiment. These are described in detail below.

  In step S310, the power consumption amount at the time is read from the first table.

  In step S311, the predicted power consumption consumed from the current time limit to 1 minute later is calculated by linear prediction using the deviation power amount between the previous one minute and the current accumulated power amount (this The predicted power amount is called a linear predicted value P). The loop timer i described later is set to 29, and then the process proceeds to step S312.

  In step S312, the operating state of the showcase at a predetermined time is read from the first table. The loop timer i represents the future time from the present, but since the operating state model is created by applying the defrost model, the operating state model describes the operating state of each load device in the future. Therefore, it can be read out.

  In step S304, according to the second embodiment, it is confirmed whether the same data exists in the second table. If the same data does not exist, the process proceeds to step S305. If the same data exists, the process proceeds to step S306. Transition.

  In step S305, it is determined whether the current time period is within 10 minutes from the end time period.

  In steps S313 to S315, each increase amount or the linear prediction value P is set as the table prediction value Q in accordance with the second embodiment.

In step S316, a predicted value POW _i for one minute is calculated (loop timer i = 0 to 29). A prediction value is obtained by adding a value obtained by weighting the difference between the table prediction value Q and the linear prediction value P by α times to the linear prediction value P (POW _i = P + α (Q−P)). The value of α can be customized to an optimum value according to the environment to be controlled.

When obtaining the predicted value POW _i for one minute, a past average value or maximum value (P ′) can be used. As a result, POW _i = P + α (Q−P ′)) can be used as a calculation formula.

  As described above, in this embodiment, the linear prediction value P is mainly used to predict the electric energy, and the prediction value (table prediction value Q) using the second table corrects the linear prediction value P. Use for the purpose.

In step S317, the loop timer i is decremented each time to calculate POW _i for each one minute from the current time period to the end time period, and the loop of steps S312 to S316 is executed for the remaining time. That is, in order to obtain data after the current time limit for each loop, the first table is read in step S312, the second table is searched in step S304, and the POW _i for each minute is calculated in step S316.

In step S318, POW _i from the current time period to the end time period is integrated to calculate an increase amount of the predicted power consumption amount at the end time period with respect to the power consumption amount at the current time period.

  Of the flowchart shown in FIG. 13, the flowchart of the detailed operation of the store controller 12 in the demand control process of step S400 is the same as FIG.

  As described above, it is possible to reduce the variation in the operation state related to the load device on the information table, and as a result, it is possible to reduce the information amount of the information table, while reducing the time required for the search and the required memory amount. Thus, the effect of avoiding the overhead of communication between the demand controller and the load device can be achieved. Also, using the predicted power consumption by linear prediction and the predicted power consumption using the second table, for example, as described above, taking these weighted averages to calculate the predicted power consumption in a corrective manner. Thus, it can be expected that the drawbacks in both the prediction method using the linear prediction and the prediction method using the second table are compensated, and the accuracy of the power consumption prediction can be improved.

  The demand control apparatus and the power consumption prediction method used therefor in the present embodiment can be realized in hardware by a CPU, memory, or other LSI of an arbitrary computer. In terms of software, it is realized by a program having a demand control function and a power consumption prediction function loaded in a memory. FIG. 3 shows a functional block diagram of a demand control apparatus realized by hardware and software. However, it goes without saying that these functional blocks can be realized in various forms such as hardware only, software only, or a combination thereof.

  The embodiments of the present invention can be appropriately modified in various ways within the scope of the technical idea shown in the claims.

It is a figure explaining the method of linear prediction of the power consumption which concerns on embodiment. It is a figure which shows the connection structure of the demand control apparatus and load apparatus which concern on embodiment. It is a functional block diagram of the demand control apparatus which concerns on embodiment. It is a schematic diagram of the information table which concerns on embodiment. It is a flowchart explaining the operation | movement in the demand control apparatus which concerns on embodiment. It is a schematic diagram of the information table which concerns on 2nd Embodiment. It is a flowchart explaining the operation | movement in the demand control apparatus which concerns on 2nd Embodiment. It is a flowchart explaining the operation | movement in the demand control apparatus which concerns on 2nd Embodiment. It is a flowchart explaining the operation | movement in the demand control apparatus which concerns on 2nd Embodiment. It is a flowchart explaining the operation | movement in the demand control apparatus which concerns on 2nd Embodiment. It is a schematic diagram of the operation state model which concerns on 3rd Embodiment. It is a schematic diagram of the 2nd table which concerns on 3rd Embodiment. It is a flowchart explaining the operation | movement in the demand control apparatus which concerns on 3rd Embodiment. It is a flowchart explaining the operation | movement in the demand control apparatus which concerns on 3rd Embodiment. It is a flowchart explaining the operation | movement in the demand control apparatus which concerns on 3rd Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Electric power meter 12 Store controller 16 Showcase 17 Air conditioner 18 Outdoor unit 21 Electric energy acquisition part 23 CPU
24 storage unit 25 device control unit 26 communication unit

Claims (37)

In the demand control device that controls the load device based on the predicted power consumption that is predicted to be consumed by the end timing of the time unit,
Table generating means for generating an information table related to the history of actual power consumption in the target unit of demand control;
Prediction calculation means for calculating the predicted power consumption using the information table generated by the table generation means;
A demand control device characterized by that. In claim 1,
The table generation means has storage means for storing in the information table as a history the relationship between the operation state of the load equipment set in advance among the load equipment in the target unit and the actual power consumption at that time,
The prediction calculation unit specifies a history position corresponding to a current operation state of the preset load device on the information table, and the actual consumption acquired from the information table following the specified history position Having a calculation means for calculating the predicted power consumption based on the electric energy;
A demand control device characterized by that. In claim 2,
The storage means stores the time information as information for specifying the history position in association with the operation state and the actual power consumption at that time together with the information table,
When a plurality of history positions corresponding to the current operating state of the preset load device are specified on the information table, the calculation means determines the history position of the time closest to the current time among the history positions. Select as the history position to be specified in the calculation of the predicted power consumption,
A demand control device characterized by that. In claim 2 or 3,
When the history position corresponding to the current operation state of the preset load device does not exist on the information table, the calculation means calculates the actual power consumption acquired immediately before and the actual power consumption at the current time. Calculate the predicted power consumption by linear prediction from the deviation of the amount,
A demand control device characterized by that. In any of claims 2 to 4,
The storage means stores the actual power consumption in the information table in association with the operation state only when the actual power consumption is within a predetermined range from the average power consumption in the information table. Demand control apparatus characterized by that. In any of claims 2 to 5,
When a showcase is included as the preset load device,
The storage means includes a process of storing a relationship between the state of the defrosting and cooling operation of the showcase and the actual power consumption in the information table as a history.
A demand control device characterized by that. In any of claims 2 to 6,
When an air conditioner is included as the preset load device,
The storage means includes a process of storing, in the information table, the relationship between the strength of the air conditioner and the state of the ON / OFF operation and the actual power consumption as a history.
A demand control device characterized by that. In claim 6,
The storage means stores, in the information table, a difference between the temperature in the showcase at that time and a set temperature in addition to the relationship.
A demand control device characterized by that. In claim 1,
The information table includes a first reference that is an average value of an operation state of a preset load device among the load devices in the target unit and an increase amount of actual power consumption in the target unit at that time. Stored in association with a value and / or a second reference value that is greater than the first reference value,
The prediction calculation means determines whether or not an operation state identical to the current operation state of the preset load device exists in the information table. If the same operation state exists, the operation state Calculating the predicted power consumption based on the first reference value or the second reference value corresponding to
A demand control device characterized by that. In claim 1,
The information table includes an average value of one or two or more time zones, an operation state of the preset load device in each time zone, and an increase amount of actual power consumption in the target unit at that time. A first reference value and / or a second reference value that is greater than the first reference value are stored in association with each other;
When the combination of the time zone to which the current time belongs and the preset operation state of the load device at the time is present in the information table, the prediction calculation unit calculates the time zone and the operation state. Calculating the predicted power consumption based on the first reference value or the second reference value corresponding to a combination;
A demand control device characterized by that. The prediction calculation means further comprises a remaining time calculation means for calculating a remaining time from the current time to the end timing,
If the remaining time is equal to or longer than a predetermined time, the prediction calculation means calculates the predicted power consumption based on the first reference value,
If the remaining time is less than a predetermined time, the predicted power consumption is calculated based on the second reference value.
The demand control apparatus according to claim 9 or 10, characterized by the above.   10. The preset load device includes a plurality of showcases, and an operation state of the preset load device includes a state in which each showcase is performing a pull-down operation. The demand control apparatus according to any one of 11.   The demand control apparatus according to any one of claims 9 to 12, wherein the second reference value is a maximum value of an increase amount of actual power consumption in the target unit. In the demand control device that controls the load device based on the predicted power consumption that is predicted to be consumed by the end timing of the time unit,
Representative pattern generation means for generating a representative pattern table describing a representative pattern of a time-series change in the operation state of the load device in the target unit of demand control;
Prediction calculation means for calculating the predicted power consumption using the representative pattern table generated by the representative pattern generation means;
A demand control device characterized by that. In claim 14,
From the deviation between the actual power consumption acquired immediately before and the actual power consumption at the present time, linear prediction power consumption calculation means for calculating the predicted power consumption by linear prediction,
A corrected predicted power consumption is calculated using the linear predicted power consumption calculated by the linear predicted power consumption calculation means and the predicted power consumption calculated by the prediction calculation means.
A demand control device characterized by that. In claim 15,
A weighted average of the linear predicted power consumption and the predicted power consumption to calculate the corrected predicted power consumption;
A demand control device characterized by that. In any of claims 14 to 16,
Table generating means for generating an information table related to the history of actual power consumption in the target unit of demand control;
The information table includes a first reference that is an average value of an operation state of a preset load device among the load devices in the target unit and an increase amount of actual power consumption in the target unit at that time. Stored in association with a value and / or a second reference value that is greater than the first reference value,
The prediction calculation means determines whether or not an operation state identical to the current operation state of the preset load device obtained from the representative pattern table exists in the information table, and the same operation state exists. In this case, the predicted power consumption is calculated based on the first reference value or the second reference value corresponding to the operating state.
A demand control device characterized by that. In any of claims 14 to 16,
Table generating means for generating an information table related to the history of actual power consumption in the target unit of demand control;
The information table includes an average value of one or two or more time zones, an operation state of the preset load device in each time zone, and an increase amount of actual power consumption in the target unit at that time. A first reference value and / or a second reference value that is greater than the first reference value are stored in association with each other;
When the combination of the time zone to which the current time belongs and the preset operating state of the load device at the time obtained from the representative pattern table exists in the information table, the prediction calculation means Calculating the predicted power consumption based on the first reference value or the second reference value corresponding to a combination of a band and the operating state;
A demand control device characterized by that. The prediction calculation means further comprises a remaining time calculation means for calculating a remaining time from the current time to the end timing,
If the remaining time is equal to or longer than a predetermined time, the prediction calculation means calculates the predicted power consumption based on the first reference value,
If the remaining time is less than a predetermined time, the predicted power consumption is calculated based on the second reference value.
The demand control apparatus according to claim 17 or 18, characterized by the above.   18. The preset load device includes a plurality of showcases, and the operation state of the preset load device includes a state in which each showcase is performing a pull-down operation. The demand control device according to any one of 19.   The demand control device according to any one of claims 17 to 20, wherein the second reference value is a maximum value of an increase amount of actual power consumption in the target unit. In the power consumption prediction method for calculating and acquiring the amount of power expected to be consumed in the power monitoring unit by the end timing of the time unit,
Of the load devices in the unit, update and store information on the history of the operation pattern of the preset load device and the actual power consumption consumed in the unit at that time, using the history information, Predicting and calculating the amount of power consumed in the unit by the end timing of the time unit,
The power consumption prediction method characterized by the above-mentioned. In claim 22,
From the history information, an operation pattern corresponding to the preset operation pattern of the load device is specified, and based on the actual power consumption in the subsequent history, the time unit ends the timing. To calculate the amount of power consumed
The power consumption prediction method characterized by the above-mentioned. In the power consumption prediction method for calculating and obtaining the predicted power consumption that is predicted to be consumed in the power monitoring unit by the end timing of the time unit,
A first reference value which is an average value of an operating state of a preset load device among the load devices in the power monitoring unit and an increase amount of actual power consumption in the power monitoring unit at that time, and / or Storing the second reference value, which is larger than the first reference value, in the information table in association with the second reference value;
It is determined whether or not the same operating state as the current operating state of the preset load device exists in the information table. If the same operating state exists, the first corresponding to the operating state is determined. Calculating the predicted power consumption based on a reference value or the second reference value;
The power consumption prediction method characterized by the above-mentioned. In the power consumption prediction method for calculating and obtaining the predicted power consumption that is predicted to be consumed in the power monitoring unit by the end timing of the time unit,
A first reference value that is an average value of one or two or more time zones, an operating state of the preset load device in each time zone, and an increase in actual power consumption in the power monitoring unit at that time And / or a second reference value that is greater than the first reference value in association with the second reference value;
It is determined whether the same combination of time zone and operation state as the combination of the time zone to which the current time belongs and the preset operation state of the load device at the time exists in the information table, and the same When a combination of a time zone and an operating state exists, the predicted power consumption is calculated based on the first reference value or the second reference value corresponding to the combination of the time zone and the operating state. ,
The power consumption prediction method characterized by the above-mentioned. In claim 24 or claim 25,
Calculate the remaining time from the current time to the end timing,
If the remaining time is equal to or longer than a predetermined time, the predicted power consumption is calculated based on the first reference value,
If the remaining time is less than a predetermined time, the predicted power consumption is calculated based on the second reference value.
The power consumption prediction method characterized by the above-mentioned.   25. The device according to claim 24, wherein the preset load device includes a plurality of showcases, and the operation state of the preset load device includes a state in which each showcase is performing a pull-down operation. 27. The power consumption prediction method according to any one of 26.   The power consumption prediction method according to any one of claims 24 to 27, wherein the second reference value is a maximum value of an increase in actual power consumption in the power monitoring unit. In the power consumption prediction method for calculating and obtaining the predicted power consumption that is predicted to be consumed in the power monitoring unit by the end timing of the time unit,
Generate a representative pattern table describing a representative pattern of a time-series change in the operating state of the load device in the power monitoring unit, and predict the predicted power consumption using the representative pattern table,
Generate an information table about the history of actual power consumption in the target unit of demand control,
A first reference value which is an average value of an operating state of a preset load device among the load devices in the power monitoring unit and an increase amount of actual power consumption in the power monitoring unit at that time, and / or Storing the second reference value, which is larger than the first reference value, in the information table in association with the second reference value;
It is determined whether or not the same operating state as the current operating state of the preset load device exists in the information table. If the same operating state exists, the first corresponding to the operating state is determined. Calculating the predicted power consumption based on a reference value or the second reference value;
The power consumption prediction method characterized by the above-mentioned. In claim 29,
Calculate the predicted power consumption by linear prediction from the deviation between the actual power consumption acquired immediately before and the actual power consumption at the present time,
The predicted power consumption calculated using the linear predicted power consumption calculated by the linear predicted power consumption calculation means and the predicted power consumption calculated by the prediction calculation means, or calculated by the prediction calculation means Calculate the corrected predicted power consumption using the weighted average value with
The power consumption prediction method characterized by the above-mentioned. In the power consumption prediction method for calculating and obtaining the predicted power consumption that is predicted to be consumed in the power monitoring unit by the end timing of the time unit,
Generate a representative pattern table describing a representative pattern of a time-series change in the operating state of the load device in the power monitoring unit, and predict the predicted power consumption using the representative pattern table,
Generate an information table about the history of actual power consumption in the target unit of demand control,
A first reference value that is an average value of one or two or more time zones, an operating state of the preset load device in each time zone, and an increase in actual power consumption in the power monitoring unit at that time And / or a second reference value that is greater than the first reference value in association with the second reference value;
It is determined whether the same combination of time zone and operation state as the combination of the time zone to which the current time belongs and the preset operation state of the load device at the time exists in the information table, and the same When a combination of a time zone and an operating state exists, the predicted power consumption is calculated based on the first reference value or the second reference value corresponding to the combination of the time zone and the operating state. ,
The power consumption prediction method characterized by the above-mentioned. In claim 31,
Calculate the predicted power consumption by linear prediction from the deviation between the actual power consumption acquired immediately before and the actual power consumption at the present time,
Using the linear predicted power consumption calculated by the linear predicted power consumption calculating means and the predicted power consumption calculated by the prediction calculating means or by calculating a weighted average to calculate a corrected predicted power consumption;
The power consumption prediction method characterized by the above-mentioned. In any of claims 29 to 32,
Calculate the remaining time from the current time to the end timing,
If the remaining time is equal to or longer than a predetermined time, the predicted power consumption is calculated based on the first reference value,
If the remaining time is less than a predetermined time, the predicted power consumption is calculated based on the second reference value.
The power consumption prediction method characterized by the above-mentioned.   30. The preset load device includes a plurality of showcases, and the operation state of the preset load device includes a state in which each showcase is performing a pull-down operation. 34. The power consumption prediction method according to any one of 33.   The power consumption prediction method according to any one of claims 29 to 34, wherein the second reference value is a maximum value of an increase in actual power consumption in the power monitoring unit.   A program for giving a computer the function of each means of the demand control device according to any one of claims 1 to 21.   A program for giving a computer a function of executing the power consumption prediction method according to any one of claims 22 to 35.