CN100594648C - Demand control apparatus and power consumption prediction method - Google Patents

Demand control apparatus and power consumption prediction method Download PDF

Info

Publication number
CN100594648C
CN100594648C CN200610004225A CN200610004225A CN100594648C CN 100594648 C CN100594648 C CN 100594648C CN 200610004225 A CN200610004225 A CN 200610004225A CN 200610004225 A CN200610004225 A CN 200610004225A CN 100594648 C CN100594648 C CN 100594648C
Authority
CN
China
Prior art keywords
power
mentioned
load machine
operate condition
time
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN200610004225A
Other languages
Chinese (zh)
Other versions
CN1874105A (en
Inventor
中岛英树
川合毅
伊藤和雄
藤原正人
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sanyo Electric Co Ltd
Original Assignee
Sanyo Electric Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from JP2005197048A external-priority patent/JP4346584B2/en
Application filed by Sanyo Electric Co Ltd filed Critical Sanyo Electric Co Ltd
Publication of CN1874105A publication Critical patent/CN1874105A/en
Application granted granted Critical
Publication of CN100594648C publication Critical patent/CN100594648C/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

A history record of, for example, the relationship between the operation state of a main load appliance and the amount of electric power consumed at that time is stored in an information table, and later, when the amount of electric power expected to be consumed is predicted, by using that information table, it is predicted with high accuracy. Specifically, the history position corresponding to the operation state of the load appliance at the current moment is identified in the information table, and, by adding the amount of electric power actually consumed as stored in the information table thereafter, the predicted amount of electric power expected to be consumed is calculated.

Description

Request control device and consumed power Forecasting Methodology
Technical field
The present invention relates to a kind of control device and suitable consumed power Forecasting Methodology that wherein adopts and program relevant of asking with them.
Background technology
As with the contract form of Utilities Electric Co., demander can the application request contract.In this request contract, constitute the structure of setting the contract quantity of power according to value request.Here, so-called value request, the value of the maximum demanded power amount that be meant in the demand power amount of certain constant chronomere (for example 1 month) amount, quantity of power meter (demand meter, demand meter) calculated in per 30 minutes.The quantity of power meter can be stored one by one to this maximum demand power value.
As should carrying out based on the contract quantity of power to the calculating of the basic charge of the part of the expense of Utilities Electric Co. payment, the contract quantity of power is that the value request according to the past determines.The contract quantity of power, be set at the higher limit of per 30 minutes consumable quantity of power, even if if producing a demand power amount certain moon above this contract quantity of power, even then the demand power amount of its next month does not surpass the contract quantity of power, also can reset the contract quantity of power immediately from this month beginning, like this, basic charge can be changed (can increase) afterwards during 12 months.
So, in the request contract, the required horsepower amount of observations in per 30 minutes during 30 minutes, because based on the maximum (being value request) in the chronomere (for example 1 month) of the demand power amount of being observed, calculating becomes the contract quantity of power on the basis of basic charge calculating, therefore requires to make the required power amount during per 30 minutes to be accommodated in the contract quantity of power all the time with interior control (request control).
As the control device corresponding, adopt the request control device with such expectation.The request control device, as in above-mentioned 30 minutes of setting time limit of contract quantity of power, obtain load machine integral body (constituting) in the consumed power amount of current time according to the quantity of power meter, calculate the interior quantity of power used of residue time limit (for example 10 minutes) of each contract quantity of power correspondence by for example air-conditioning machine, show cabinet, illumination etc.It calculates the predicted value of the consumed power amount of the load machine integral body in each residue time limit on the other hand.And, the quantity of power of calculating used is compared with the forecast consumption quantity of power, exist the forecast consumption quantity of power might surpass under the situation of spendable quantity of power, for example to the halt instruction of air-conditioning machine output action, cold-room or the greenhouse action or the air-supply action etc. of air-conditioning machine are stopped, and the power use amount is cut down.Thereafter, the request control device, in each above-mentioned setting time limit, according to the calculating that remains the quantity of power used in the time limit on one side one by one, calculating with the forecast consumption quantity of power, be no more than the mode of contract quantity of power on one side according to the demand power amount, carry out the restriction or the control of the action of each load machine.
As an example of such request control device, the device described in Japanese kokai publication hei 7-143670 communique etc. is arranged.
In addition, as the method for the forecast consumption quantity of power of calculating the load machine integral body in the remaining time limit, known a kind of based on the change value in the quantity of power of current small time internal consumption, linear forecast method.Below, employing figure describes this method,
Fig. 1 represents the linear Forecasting Methodology of certain interval interior consumed power of 30 minutes.With reference to Fig. 1, transverse axis is a time shaft, and the longitudinal axis is for accumulative total consumed power amount axle, to certain moment T 2Till the consumed power amount represented by block curve figure.Accumulative total consumed power amount was reset to zero in per 30 minutes.The elapsed time that order began from the moment that resets is T.At moment T 2, for this contract quantity of power of 30 minutes, Yi Bian remain the computing of the quantity of power used in the time limit, and the prediction of consumed power amount, Yi Bian, carry out the restriction or the control of the action of each load machine according to the mode that is no more than the contract quantity of power.
In the prediction of consumed power amount, adopt at moment T 2Accumulative total consumed power amount W 2(the some X on the curve chart 2), with from moment T 2Only retreat the moment T of small time Δ T 1Accumulative total consumed power amount W 1(the some X on the curve chart 1) between deviation delta the W (=W of quantity of power 2-W 1).That is, according to the accumulative total consumed power amount in this quantity of power deviation delta W prediction T=30 minute.Specifically, use tie point X 1With an X 2Straight line 1 ' (curve chart of dotted line), calculate T=30 minute forecast consumption quantity of power (prediction accumulative total consumed power amount) W '.That is, will moment T 2Consumed power amount during each later Δ T is regarded Δ W as, and the accumulative total consumed power amount W ' of finish time in 30 minutes intervals is predicted.In addition, the slope of straight line 1 ' is (Δ W/ Δ T).
At this moment, when forecast consumption quantity of power W ' surpasses the contract quantity of power, carry out restriction, the consumption of reduction quantity of power of the action of load machine, control is so that the accumulative total consumed power amount of 30 minutes interval finish times is no more than the contract quantity of power.Forecast consumption quantity of power when restriction that this quantity of power consumes is carried out in Fig. 1 cathetus 1 (curve chart of dotted line) expression.That is, control so that T=30 minute forecast consumption quantity of power (prediction accumulative total consumed power amount) W is no more than the contract quantity of power.
Yet, in the prior art, method as the forecast consumption quantity of power of calculating the load machine integral body in the remaining time limit, because only adopting with the change value in the quantity of power of current small time internal consumption is that forecast method is carried out on the basis, therefore in some cases can be because of the action of the load machine of predicting difficulty etc. etc., cause producing the situation of error prediction, its result is, be created in the remaining time limit the few such situation of spendable quantity of power.Under such situation, the problem that can produce is, also have must make the load machine almost all action stop such control, if for example the show cabinet action stops, then Chen Lie food can go bad.
In addition, also have following situation, sometime, it is big that the change value of the quantity of power of small time internal consumption becomes, and its result is that the consumed power amount of request control device prediction in this time limit surpasses the contract quantity of power.In this case, though request control device, the control that the action of corresponding load machine is stopped, even keep former state but the action that does not also in fact make the load machine in some cases stops, thereafter consumed power amount also can be lower than predicted value, stop even without the action that makes the load machine, in this time limit, also can not surpass the contract quantity of power.That is, in the prior art, even if the problem that existence also needn't make the action of load machine stop at short notice.
Summary of the invention
Therefore, the present invention is for fear of problem as described above, its purpose is to provide a kind of prediction of carrying out the consumed power amount in each time limit (during the unit) more accurately, can be no more than contract power thus and successfully carries out the request control device of machine control and consumed power Forecasting Methodology and the relative program that adopts thus.
It is a kind of that one of purpose of the present invention is to provide, the resume of the operate condition by adopting the load machine and the relation of the consumed power amount of this moment, thus can carry out the request control device of prediction of the consumed power amount in each time limit and consumed power Forecasting Methodology and the relative program that therefore adopts accurately.
To achieve these goals, the 1st request control device of the present invention, it is based on the predicted forecast consumption quantity of power of the consumption of load machine before the finish time during the unit, carry out the control of load machine, it is characterized in that, have: form generates mechanism, and operate condition, the actual power consumption amount of this moment and the time information of this moment of the load machine that is set in advance in the load machine in its object unit that will ask to control carry out association and generate information form as resume; With prediction and calculation mechanism, it adopts by this form and generates the information form that mechanism generated, calculate the forecast consumption quantity of power, form generates mechanism and has preservation mechanism, it is with the operate condition of the load machine that is set in advance in the load machine in the object unit, carry out relatedly being kept in the information form with actual power consumption amount and the time information of this moment of this moment as resume, prediction and calculation mechanism has calculation mechanism, its on information form, determine with the load machine that is set in advance in the corresponding resume position of the operate condition of current time, follow determined resume position and based in the time interval till from the current time to the finish time from the actual power consumption amount of the load machine consumption that information form was obtained, calculate the forecast consumption quantity of power; When on information form, not having corresponding resume position, according to the deviation of the actual power consumption amount that is obtained before recently with the actual power consumption amount of current time, by linear prediction, calculating forecast consumption quantity of power.
In addition for example, in above-mentioned the 1st request control device, above table generates mechanism, has preservation mechanism, it is with the operate condition of the load machine that is set in advance in the load machine in the above-mentioned object unit, be kept in the above-mentioned information form as resume with the relation of actual power consumption amount at this moment, above-mentioned prediction and calculation mechanism, has calculation mechanism, its on above-mentioned information form to carrying out specific in the corresponding resume position of the operate condition of current time with the above-mentioned load machine that is set in advance, then specific resume position based on from the above-mentioned actual power consumption amount that above-mentioned information form was obtained, calculate above-mentioned forecast consumption quantity of power.
When the operate condition of in the past certain load machine constantly is identical with current operate condition, the then movement of Zhi Hou load machine, identical with movement that certain moment in this past is later, same variation is also followed in the consumption of very big quantity of power possibly.Promptly, if the operate condition of current load machine is consistent with the operate condition of in the past certain load machine constantly, then same with resume afterwards, operate condition after the very big current time possibly changes, therefore, after resume in the actual power consumption that consumed be the basis, by the later consumed power of current time is predicted, thereby the precision of forecast consumption power is improved.Thereby, as mentioned above, resume are kept in the information form, based on from the consumed power amount that information form was obtained, by calculating the forecast consumption quantity of power, thereby can predict the consumed power amount of load machine integral body accurately.
In addition for example, in above-mentioned the 1st request control device, above-mentioned preservation mechanism, time information is related with above-mentioned operate condition and the actual power consumption amount of this moment, it is carried out information specific as being used for to above-mentioned resume position, be kept at simultaneously in the above-mentioned information form, aforementioned calculation mechanism, on above-mentioned information form, at specific a plurality of and above-mentioned predefined load machine during in the corresponding resume position of the operate condition of current time point, by the calculating of above-mentioned forecast consumption quantity of power select will with the resume position in nearest moment in the moment of current time point wherein as should specific resume position.
Need with the load machine of relevant setting such as action in, very big this setting content possibly is all identical every day.That is, the operate condition of load machine becomes identical in the essentially identical moment very big every day possibly.Therefore, in resume, exist every day possibly very greatly consistent the locating of basic identical moment operate condition, and, the resume position that the variation of operate condition afterwards is in full accord, such.Therefore, as above-mentioned, be kept at time information in the information form simultaneously, during the corresponding resume position of the operate condition of the load machine of specific a plurality of and current time, by selection and the current time resume position in the nearest moment, thereby the precision of prediction of the consumed power amount of load machine integral body is improved.
In addition for example, in above-mentioned the 1st request control device, aforementioned calculation mechanism, when not existing with above-mentioned predefined load machine on the above-mentioned information form in the corresponding resume position of the operate condition of current time point, according to the deviation of the above-mentioned actual power consumption amount of above-mentioned actual power consumption amount that is obtained before and current time, by the above-mentioned forecast consumption quantity of power of linear prediction and calculation.
So, when not having the operate condition consistent in the resume with the operate condition of current load machine, with according to before the deviation information of consumed power amount of consumed power amount and current time point, the predicted value that adopts linear forecasting institute to calculate is the basis, calculating forecast consumption quantity of power.That is, when not having the operate condition consistent in the resume, the function of request control device is replenished by linear prediction with the operate condition of current load machine.And, after using beginning, fully do not possess under the situation of resume, on one side by linear prediction and calculation forecast consumption quantity of power, accumulate resume on the other hand.And, along with effluxion, use resume simultaneously, accuracy of predicting is improved.Above-mentioned linear prediction, useful in such process.
In addition for example, in above-mentioned the 1st request control device, above-mentioned preservation mechanism, its only in above-mentioned actual power consumption amount under the mean value of the consumed power amount from the above-mentioned information form situation to the given scope, be kept in the above-mentioned information form this actual power consumption amount is related with above-mentioned operate condition.
Above-mentioned the 1st request control device, it adopts resume to carry out the prediction of the high consumed power amount of precision.Therefore, in resume, in case sneak into the information of exception, then in the prediction of in the future consumed power amount, probably can the high error of influential degree under with reference to the situation of its position.Therefore, as above-mentioned, when the consumed power amount, only be accommodated in from the mean value of the consumed power amount information form to given scope in the time, just this consumed power amount is kept in the information form with the operate condition of above-mentioned load machine is related.Like this, be expected to get rid of the information of exception, very useful.
In addition for example, in above-mentioned the 1st request control device, as above-mentioned predefined load machine, comprising show cabinet under interior situation, in the processing that above-mentioned preservation mechanism carries out, comprise the relation of the state of the defrosting of above-mentioned show cabinet and cooling action and above-mentioned actual power consumption amount is kept at processing in the above-mentioned information form as resume.
In addition for example, in above-mentioned the 1st request control device, comprising under the situation of air conditioner as above-mentioned predefined load machine, in the processing that above-mentioned preservation mechanism carries out, comprise the relation of the state of the power of above-mentioned air conditioner and ON/OFF action and above-mentioned actual power consumption amount is kept at processing in the above-mentioned information form as resume.
In addition for example, in above-mentioned the 1st request control device, above-mentioned preservation mechanism with the storehouse temperature of the above-mentioned show cabinet of this moment and the difference of design temperature, is kept in the above-mentioned information form with above-mentioned relation.
Can consume the load machine of high-power amount, most possibly the consumed power volume production be given birth to influence.Therefore, the highest as above-mentioned formation to the possibility of the influence of consumed power amount, the operate condition of such load machine is kept in the information form as resume.
And, relevant show cabinet, many show cabinets are more with the example that 1 refrigerator is connected, its result is, undertaken by easy Theoretical Calculation under the situation of prediction of variation of power consumption of show cabinet, and its precision of prediction is very low, for precision is improved, need large-scale computing.Therefore, use the information form that the relation of the operate condition of show cabinet and consumed power amount is preserved as resume, be highly profitable.In addition, in show cabinet,,, then, therefore also it can be preserved as resume because very possible consumed power amount or movement change if the difference of storehouse temperature and design temperature is different even other state is identical.
In addition for example, in above-mentioned the 1st request control device, in the above-mentioned information form, operate condition with predefined load machine in the load machine in the above-mentioned object unit, with as the 1st fiducial value of the mean value of the recruitment of the actual power consumption amount in the above-mentioned object unit of this moment and/or as the value 2nd fiducial value related preservation bigger than the 1st fiducial value, above-mentioned prediction and calculation mechanism, judge whether being present in the above-mentioned information form at the identical operate condition of the operate condition of current time point with above-mentioned predefined load machine, when having identical operate condition, based on above-mentioned 1st fiducial value corresponding or above-mentioned the 2nd fiducial value, calculate above-mentioned forecast consumption quantity of power with this operate condition.
Even predefined load machine in the load machine in the object unit, promptly monitor that the operate condition of load machine of operate condition (operating condition) is identical, because of the influence of the operate condition of the load machine beyond it, the actual power consumption amount in the object unit can produce deviation in the recruitment of preset time in some cases.Therefore, also can produce deviation in the 1st fiducial value as the mean value of this recruitment in some cases.Therefore, in some cases,, but also can produce deviation even if the precision of the forecast consumption quantity of power of calculating based on the 1st fiducial value is very high.
For example set under the situation of the 2nd fiducial value in the mode that must surpass the actual power consumption amount according to forecast consumption performance number based on aforementioned calculation, when calculating the forecast consumption quantity of power, according to the situation of remaining time before the finish time during the unit for example etc., just can select to be to use the 1st fiducial value also to be to use the 2nd fiducial value.Its result is according to the suitable forecast consumption quantity of power of situation calculating, just can carry out surpassing the request control of contract power based on the forecast consumption quantity of power.In addition, above-mentioned the 1st fiducial value comprises mean value or median, the most frequent value, dispersion value and their approximation etc. of the recruitment of the actual power consumption amount in the object unit for example, below too.
In addition for example, in above-mentioned the 1st request control device, in above-mentioned information form, with one or the plural time period, in the load machine in the above-mentioned object unit by the operate condition of predefined load machine in each time period, with as the 1st fiducial value of the mean value of the recruitment of actual power consumption amount in the above-mentioned object unit of this moment and/or as the related preservation of the 2nd fiducial value of the value bigger than the 1st fiducial value, above-mentioned prediction and calculation mechanism, in above-mentioned information form, exist under the current time time period and when the combination of the operate condition of the above-mentioned predefined load machine in this moment, based on corresponding above-mentioned the 1st fiducial value or above-mentioned the 2nd fiducial value of combination of this time period and this operate condition, calculate above-mentioned forecast consumption quantity of power.
Consider the influence of time period, as being had one of environment operating condition of deviation by the relation of actual power consumption amount in the operate condition of predefined load machine and the object unit in the load machine that makes in the object unit.For example, in the load machine, comprise show cabinet or to its refrigerator that cools off when interior, the person's of patronizing the quantity of moving into the situation of moving into of food etc. or the shop facility of this show cabinet or refrigerator is set in the storehouse of show cabinet is subjected to the influence of time period.Therefore and since to because of food etc. move into or the person's of patronizing turnover or body temperature etc. cause cooling off in the storehouse of the show cabinet that temperature rises, thereby the quantity of power that is consumed also can be subjected to the influence of time period.
In view of like this, as mentioned above, consider the time period and in this time period by the operate condition of predefined load machine, if calculate the forecast consumption quantity of power, then can calculate the forecast consumption quantity of power accurately based on the 1st fiducial value corresponding or the 2nd fiducial value with it.
In addition for example, in above-mentioned the 1st request control device, above-mentioned prediction and calculation mechanism also possesses calculation mechanism remaining time, and it is to calculating the remaining time till the moment to the above-mentioned finish time of current time point, above-mentioned prediction and calculation mechanism, if in above-mentioned remaining time is more than preset time, then calculate above-mentioned forecast consumption quantity of power, if not enough preset time of above-mentioned remaining time based on above-mentioned the 1st fiducial value, then, calculate above-mentioned forecast consumption quantity of power based on above-mentioned the 2nd fiducial value.
To consider that also the actual power consumption amount surpasses the situation of forecast consumption quantity of power as producing deviation in the above-mentioned forecast consumption quantity of power based on the calculating of the 1st fiducial value.Ask the result who controls to be based on such forecast consumption quantity of power, though the actual power consumption amount in the object unit of also will considering surpasses the situation of forecast consumption quantity of power, even but under such situation, under remaining time enough situation, by after the control of load machine just can recover.
On the other hand, can ask to control and pay attention to the request control that the actual power consumption amount does not surpass set-point based on the forecast consumption quantity of power of calculating according to the 2nd fiducial value.
According to as above-mentioned formation, in remaining time when enough, calculate the forecast consumption quantity of power based on the 1st fiducial value, based on the forecast consumption quantity of power, can pay attention to the keeping of comfort level of shop facility, the request control of surplus is arranged, on the contrary, in remaining time seldom the time, calculate the forecast consumption quantity of power based on the 2nd fiducial value, can pay attention to the request control that the actual power consumption amount surpasses set-point based on the forecast consumption quantity of power.
In addition for example, in above-mentioned the 1st request control device, above-mentioned predefined load machine comprises many show cabinets, and the operate condition of above-mentioned predefined load machine comprises the executing state of the quick-frozen running of each show cabinet.
If many show cabinets carry out the quick-frozen running simultaneously after the defrosting running, the consumed power amount that then connects the refrigerator of these show cabinets sharply increases, and gives birth to very big influence in some cases the actual power consumption volume production of load machine integral body.
Consider these, when predefined load machine comprises many show cabinets,, comprise the executing state of the quick-frozen running of each show cabinet as the operate condition of this predefined load machine.The increase of the rapid consumed power amount that causes because of the quick-frozen running of considering by show cabinet, thus the forecast consumption quantity of power can be calculated more accurately.
In addition for example, in above-mentioned the 1st request control device, above-mentioned the 2nd fiducial value is the maximum of the recruitment of actual power consumption amount in the above-mentioned object unit.
So, if the maximum that adopts the recruitment of actual power consumption amount in the object unit as the 2nd fiducial value, then can make the forecast consumption quantity of power of calculating based on the 2nd fiducial value become the value bigger than actual power consumption amount definitely.
In addition, to achieve these goals, the 2nd request control device of the present invention, be a kind of control of carrying out the load machine based on the forecast consumption quantity of power of consumption of prediction before the finish time during the unit, the request control device, it is characterized in that, have: the representative pattern generates mechanism, and schema table is represented in its generation, has described the representational pattern of timing variations of the operate condition of the object unit internal burden machine of asking control; With prediction and calculation mechanism, it adopts by this representative pattern and generates the above-mentioned representative schema table that mechanism generated, and calculates above-mentioned forecast consumption quantity of power.
The form of the representational pattern of timing variations (below be called the modelling pattern) of the operate condition of above-mentioned load machine has been described by employing, thus can reduce preserve and formization after the deviation of the relevant operate condition of load machine.Its result is can cut down the amount of information of form, on the other hand the amount of memory of cutting down the needed time of retrieval or needing.
After between the load machine, having individual difference, even the placement environment to the load machine compared with the identical time period of for example the previous day, then owing to can have some differences with very high probability, therefore when the operate condition that the load machine is relevant was kept in the form, the scope of the deviation of these operate condition data increased.
Therefore, in order to get rid of the influence of deviation as described above, thus according to will be for example be that real situation compares and carries out modellings such as analogyization, simplification, abstract or formalization for identical or approximate in fact mode with the operate condition of constantly corresponding each load machine and its feature major part.And, the state of load machine is considered as by the state shown in the model, by with reference to this information, thereby the needed time is cut down and retrieved to the amount of information that realizes above table or the reduction of the memory space that needs etc.And in this model, for example by adopting action command plan to the load machine, thereby can make easily should reference operate condition data etc.
In addition, under the situation of the operate condition of the load machine in utilizing above-mentioned object unit, by communicating etc. with above-mentioned load machine, thereby replace directly accepting operate condition from above-mentioned load machine, by the above-mentioned representative schema table that generates of reference, thereby obtain the operate condition of above-mentioned load machine, can carry out the prediction of above-mentioned consumed power amount.Like this, just, can avoid the expense of for example communicating by letter.
In addition for example, in above-mentioned the 2nd request control device, has linear forecast consumption quantity of power calculation mechanism, it is according to the deviation of the actual power consumption amount of actual power consumption amount that is obtained before and current time point, thereby calculate linear prediction consumed power amount by linear prediction, the above-mentioned linear forecast consumption quantity of power that employing is calculated by this linear forecast consumption quantity of power calculation mechanism, with the above-mentioned forecast consumption quantity of power of being calculated by above-mentioned prediction and calculation mechanism, calculate revisal forecast consumption quantity of power, carry out the control of above-mentioned load machine based on this revisal forecast consumption quantity of power.
In addition for example, in above-mentioned the 2nd request control device, be weighted on average, calculate above-mentioned revisal forecast consumption quantity of power to above-mentioned linear forecast consumption quantity of power, with above-mentioned forecast consumption quantity of power.
Deviation according to the actual power consumption amount of actual power consumption amount that is obtained before and current time point, by the linear forecast consumption quantity of power of linear prediction and calculation, the above-mentioned forecast consumption quantity of power that adopts this linear forecast consumption quantity of power and calculated by above-mentioned prediction and calculation mechanism is for example got their weighted average and is calculated revisal forecast consumption quantity of power.The precision of prediction of consumed power amount is improved.
In addition for example, in above-mentioned the 2nd request control device, have form and generate mechanism, it generates the relevant information form of resume of actual power consumption amount in the object unit of controlling with request, in above-mentioned information form, operate condition with predefined load machine in the load machine in the above-mentioned object unit, with as the 1st fiducial value of the mean value of the recruitment of actual power consumption amount in the above-mentioned object unit of this moment and/or as the related preservation of the 2nd fiducial value of the value bigger than the 1st fiducial value, above-mentioned prediction and calculation mechanism, to judging with whether being present in the above-mentioned information form at the identical operate condition of the operate condition of current time from the resulting above-mentioned predefined load machine of above-mentioned representative schema table, when having identical operate condition, based on above-mentioned 1st fiducial value corresponding or above-mentioned the 2nd fiducial value, calculate above-mentioned forecast consumption quantity of power with this operate condition.
In addition for example, in above-mentioned the 2nd request control device, have form and generate mechanism, it generates the relevant information form of resume of actual power consumption amount in the object unit of controlling with request, in above-mentioned information form, with one or the plural time period, in the load machine in the above-mentioned object unit by the operate condition of predefined load machine in each time period, with as the 1st fiducial value of the mean value of the recruitment of actual power consumption amount in the above-mentioned object unit of this moment and/or as the related preservation of the 2nd fiducial value of the value bigger than the 1st fiducial value, above-mentioned prediction and calculation mechanism, it exists time period under moment of current time point with resulting from above-mentioned representative schema table in above-mentioned information form, when the combination of the operate condition of this above-mentioned predefined load machine constantly, based on corresponding above-mentioned the 1st fiducial value or above-mentioned the 2nd fiducial value of combination of this time period and this operate condition, calculate above-mentioned forecast consumption quantity of power.
In addition for example, in above-mentioned the 2nd request control device, above-mentioned prediction and calculation mechanism also possesses calculation mechanism remaining time, and it is to calculating the remaining time till from current time to the above-mentioned finish time, above-mentioned prediction and calculation mechanism, if be more than preset time above-mentioned remaining time, then calculate above-mentioned forecast consumption quantity of power, if not enough preset time of above-mentioned remaining time based on above-mentioned the 1st fiducial value, then, calculate above-mentioned forecast consumption quantity of power based on above-mentioned the 2nd fiducial value.
In addition for example, in above-mentioned the 2nd request control device, above-mentionedly comprised a plurality of show cabinets by predefined load machine, above-mentioned by the operate condition of predefined load machine, comprise the executing state of the quick-frozen running of each show cabinet.
In addition for example, in above-mentioned the 2nd request control device, above-mentioned the 2nd fiducial value is the maximum of the recruitment of the actual power consumption amount in the above-mentioned object unit.
In addition, to achieve these goals, the 1st consumed power Forecasting Methodology of the present invention, calculate and obtained before the finish time during the unit in the predicted forecast consumption quantity of power of object unit internal consumption, it is characterized in that, operate condition with the load machine that is set in advance in the load machine in the object unit, the time information of the actual power consumption amount of this moment and this moment is associated and is kept in the information form as resume, on information form, determine with the load machine that is set in advance in the corresponding resume position of the operate condition of current time, follow determined resume position and based in the time interval till from the current time to the finish time from the actual power consumption amount of the load machine consumption that information form was obtained, calculate the forecast consumption quantity of power; When on information form, not having corresponding resume position, according to the deviation of the actual power consumption amount that is obtained before recently with the actual power consumption amount of current time, by linear prediction, calculating forecast consumption quantity of power.
Specifically for example, in above-mentioned the 1st consumed power Forecasting Methodology, will with above-mentioned electric power monitoring unit internal burden machine in by the resume of actual power consumption amount are relevant in the pattern of predefined load machine and the above-mentioned electric power monitoring unit of this moment information as above-mentioned record information, it is carried out more new record, adopt this record information, calculate above-mentioned forecast consumption quantity of power.
In addition for example, in above-mentioned the 1st consumed power Forecasting Methodology,, carry out specific to the pattern corresponding with the pattern of above-mentioned predefined load machine according to above-mentioned record information, based on the above-mentioned consumed power amount in the resume after it, calculate above-mentioned forecast consumption quantity of power.
In addition for example, in above-mentioned the 1st consumed power Forecasting Methodology, by in information form, operate condition with predefined load machine in the load machine in the above-mentioned electric power monitoring unit, with as the 1st fiducial value of the mean value of the recruitment of the actual power consumption amount in the above-mentioned electric power monitoring unit of this moment and/or as the related preservation of the 2nd fiducial value of the value bigger than the 1st fiducial value, thereby construct above-mentioned record information, judge whether existing in the above-mentioned information form at the identical operate condition of the operate condition of current time with above-mentioned predefined load machine, when having identical operate condition, based on above-mentioned 1st fiducial value corresponding or above-mentioned the 2nd fiducial value, calculate above-mentioned forecast consumption quantity of power with this operate condition.
In addition, so-called above-mentioned electric power monitoring unit, with for example when asking to control, the object unit of request control is suitable, below too.
In addition, above-mentioned the 1st fiducial value comprises for example mean value, dispersion value and their approximation etc. of the recruitment of the actual power consumption amount in the electric power monitoring unit, below too.
In addition for example, in above-mentioned the 1st consumed power Forecasting Methodology, by with the time period more than 1 or 2, in the load machine in the above-mentioned electric power monitoring unit by the operate condition of predefined load machine in each time period, with as this moment above-mentioned electric power monitoring unit in the actual power consumption amount recruitment mean value the 1st fiducial value and/or be kept in the information table as the 2nd fiducial value of the value bigger than the 1st fiducial value is related, thereby construct above-mentioned record information, to whether exist in the above-mentioned information form with current time under time period and above-mentionedly judged in this moment by the identical time period of the combination of the operate condition of predefined load machine and the combination of operate condition, when having the combination of identical time period and operate condition, based on corresponding above-mentioned the 1st fiducial value or above-mentioned the 2nd fiducial value of combination of this time period and this operate condition, calculate above-mentioned forecast consumption quantity of power.
In addition for example, in above-mentioned the 1st consumed power Forecasting Methodology, the remaining time of calculating till from moment of current time point to the above-mentioned finish time, if above-mentioned remaining time is more than preset time, then calculate above-mentioned forecast consumption quantity of power based on above-mentioned the 1st fiducial value, if above-mentioned remaining time, then based on above-mentioned the 2nd fiducial value, calculated above-mentioned forecast consumption quantity of power not enough preset time.
In addition for example, in above-mentioned the 1st consumed power Forecasting Methodology, above-mentioned predefined load machine comprises many show cabinets, and the operate condition of above-mentioned predefined load machine comprises the executing state of the quick-frozen running of each show cabinet.
In addition for example, in above-mentioned the 1st consumed power Forecasting Methodology, above-mentioned the 2nd fiducial value is the maximum of the recruitment of the actual power consumption amount in the above-mentioned electric power monitoring unit.
In addition, to achieve these goals, the 2nd consumed power Forecasting Methodology of the present invention, calculate to obtain prediction the consumption in the electric power monitoring unit before the finish time during the unit, the forecast consumption quantity of power, it is characterized in that, generate and represent schema table, the representational pattern of timing variations that it has described the operate condition of above-mentioned electric power monitoring unit internal burden machine adopts this representative schema table to calculate above-mentioned forecast consumption quantity of power.
Specifically for example, in above-mentioned the 2nd consumed power Forecasting Methodology, generate with above-mentioned electric power monitoring unit in the relevant information form of resume of actual power consumption amount, with in the load machine in the above-mentioned electric power monitoring unit by the pattern of predefined load machine, with as this moment above-mentioned electric power monitoring unit in the actual power consumption amount recruitment mean value the 1st fiducial value and/or be kept in the above-mentioned information form as the 2nd fiducial value of the value bigger than the 1st fiducial value is related, to judging with whether being present in the above-mentioned information form at the identical operate condition of the operate condition of current time from the resulting above-mentioned predefined load machine of above-mentioned representative schema table, when having identical operate condition, based on above-mentioned 1st fiducial value corresponding or above-mentioned the 2nd fiducial value, calculate above-mentioned forecast consumption quantity of power with this operate condition.
In addition for example, in above-mentioned the 2nd consumed power Forecasting Methodology, generate with above-mentioned electric power monitoring unit in the relevant information form of resume of actual power consumption amount, with one or the plural time period, in the load machine in the above-mentioned electric power monitoring unit by the operate condition of predefined load machine in each time period, with as this moment above-mentioned electric power monitoring unit in the actual power consumption amount recruitment mean value the 1st fiducial value and/or be kept in the above-mentioned information form as the 2nd fiducial value of the value bigger than the 1st fiducial value is related, whether it is to existing time period under the moment with current time point and identical according to the combination of the operate condition of the resulting above-mentioned predefined load machine in this moment of above-mentioned representative schema table in above-mentioned information form, the combination of time period and operate condition is judged, when having the combination of identical time period and operate condition, based on corresponding above-mentioned the 1st fiducial value or above-mentioned the 2nd fiducial value of combination of this time period and this operate condition, calculate above-mentioned forecast consumption quantity of power.
In addition for example, in above-mentioned the 2nd consumed power Forecasting Methodology, according to the deviation of the actual power consumption amount of resulting actual power consumption amount and current time before, by linear prediction, calculate linear forecast consumption quantity of power, adopt above-mentioned linear forecast consumption quantity of power of being calculated and the above-mentioned forecast consumption quantity of power of being calculated, perhaps adopt the average weighted value of above-mentioned linear forecast consumption quantity of power of being calculated and the above-mentioned forecast consumption quantity of power of being calculated, calculate revisal forecast consumption quantity of power, the forecast consumption quantity of power that this revisal forecast consumption quantity of power finally should be calculated as this consumed power Forecasting Methodology is handled.
In addition for example, in above-mentioned the 2nd consumed power Forecasting Methodology, calculate the remaining time till the moment to the above-mentioned finish time of current time point,, then calculate above-mentioned forecast consumption performance number based on above-mentioned the 1st fiducial value if be more than preset time remaining time; If remaining time, then calculated above-mentioned forecast consumption quantity of power based on above-mentioned the 2nd fiducial value not enough preset time.
In addition for example, in above-mentioned the 2nd consumed power Forecasting Methodology, above-mentioned predefined load machine comprises many show cabinets, and the operate condition of above-mentioned predefined load machine comprises the executing state of the quick-frozen running of each show cabinet.
In addition for example, in above-mentioned the 2nd consumed power Forecasting Methodology, above-mentioned the 2nd fiducial value is the maximum of the recruitment of the actual power consumption amount in the above-mentioned electric power monitoring unit.
Utilize the formation of above-mentioned the 1st fiducial value etc. in the control device in above-mentioned the 2nd request, perhaps in above-mentioned the 2nd consumed power Forecasting Methodology, utilize in the method for above-mentioned the 1st fiducial value etc., remove the formation of in above-mentioned the 1st request control device, utilizing above-mentioned the 1st fiducial value etc., perhaps in above-mentioned the 1st consumed power Forecasting Methodology, utilize outside the effect that the method for above-mentioned the 1st fiducial value etc. has, the amount of information that also possesses above-mentioned the 2nd form that has of request control device is cut down, retrieve the reduction of needed time or needed memory space, the effect that the avoiding of the expense of communication waited, and can also realize the raising of the precision of prediction of above-mentioned consumed power amount.
In addition, above-mentioned the 2nd fiducial value is not limited to the maximum of the recruitment of actual power consumption amount.That is, if, then can adopt various values as above-mentioned the 2nd fiducial value than the big value of the 1st fiducial value.For example, except that the value of the constant multiple of the 1st fiducial value, 30% the mean value etc. of last place value that also can adopt the median of the 1st fiducial value or mode, determination data is as the 2nd fiducial value.Further, the 2nd fiducial value also can be the dispersion of determination data or standard deviation etc.Under the situation of value as the 2nd fiducial value of the constant several times that adopt above-mentioned the 1st fiducial value, not to preserve the 2nd fiducial value, but preserve each the 2nd fiducial value of calculating of this constant number.Like this, can the conserve memory amount.
But because this point of risk management simple and easy and that reliability is high, therefore preferred employing maximum is as the 2nd fiducial value.
And to achieve these goals, the 1st program of the present invention is given the above-mentioned the 1st or the 2nd function of asking each mechanism of control device to computer.
In addition, to achieve these goals, the 2nd program of the present invention is given to computer and to be carried out the above-mentioned the 1st or the function of the 2nd consumed power Forecasting Methodology.
According to the explanation of execution mode shown below, can further understand meaning of the present invention even effect.
But, following execution mode, nothing but one embodiment of the present invention, the meaning of the nomenclature of the present invention and even each constitutive requirements, it is described to be not limited to following execution mode.
Description of drawings
Fig. 1 is the figure of the linear forecast method of the consumed power of explanation the 1st execution mode
Fig. 2 represents the request control device of the 1st execution mode and the figure that connects and composes of load machine.
Fig. 3 is the functional block diagram of the request control device of the 1st execution mode.
Fig. 4 is the schematic diagram of the information form of the 1st execution mode.
Fig. 5 is the flow chart of the action of the request control device of explanation the 1st execution mode.
Fig. 6 is the schematic diagram of the information form of the 2nd execution mode.
Fig. 7 is the flow chart of the action of the request control device of explanation the 2nd execution mode.
Fig. 8 is the flow chart of the action of the request control device of explanation the 2nd execution mode.
Fig. 9 is the flow chart of the action of the request control device of explanation the 2nd execution mode.
Figure 10 is the flow chart of the action of the request control device of explanation the 2nd execution mode.
Figure 11 is the schematic diagram of the operate condition model of expression the 3rd execution mode.
Figure 12 is the schematic diagram of the 2nd form of the 3rd execution mode.
Figure 13 is the flow chart of the action of the request control device of explanation the 3rd execution mode.
Figure 14 is the flow chart of the action of the request control device of explanation the 3rd execution mode.
Figure 15 is the flow chart of the action of the request control device of explanation the 3rd execution mode.
Embodiment
(the 1st execution mode)
Below, with reference to accompanying drawing, the 1st execution mode of the present invention (below be called present embodiment one) is described.
Present embodiment one, with different in the past, for the consumed power amount of the load machine integral body in certain regulation time limit (30 minutes wait unit during) is predicted, adopt the information form of will the operate condition of main load machine related resume preservation with the consumed power amount of at that time load machine integral body and temporal information at that time etc.So-called main load machine is the load machine of the part in the load machine integral body relevant with electric power system, for example show cabinet that power consumption is big or air conditioner etc.If in the remaining time limit, surpass under the situation of spendable quantity of power in the forecast consumption quantity of power, then stop, and then make the action of air conditioner integral body stop the control that waits by the off-premises station that for example makes any air conditioner, thus reduction consumed power amount.Even if this is the comfort level that also can fully guarantee within doors because of the off-premises station of air conditioner being stopped, only moving by blowing.In addition, even if because the action of air-conditioning machine integral body is stopped, the people is difficult to also feel that comfort level within doors is not good enough in some times, influences little.
Fig. 2 represents the figure that connects and composes of the request control device of relevant present embodiment one and load machine.
With reference to Fig. 2,11 is the power meter of the quantity of power of the instrumentation load machine integral body relevant with electric power system.12 is the shop controller as the request control device, and 13 are illumination, and 14 are other load machine arbitrarily, and 15 for show cabinet cools off the refrigerators of usefulness, and 16 is one or many show cabinets, and 17 is air conditioner, and 18 is its off-premises station, and 19 is the refrigerant pipe arrangement.13～18 is to become the load machine of being asked the object controlled by shop controller 12 all.
In addition, in figure, constitute 1 electric control system, ask control by shop controller 12 by power meter 11 and even off-premises station 13 and power line and communication network.This electric control system be equivalent in the claim scope " object unit " perhaps " electric power monitoring unit ", and, also 13～18 (perhaps parts wherein) can be interpreted as " object unit " or " electric power monitoring unit ".
In each load machine, as shown in the figure, be connected with the controller of the load machine side that is connected with shop controller 12 by communication network, each controller carries out the setting and the control of the machine that connected.Though these settings and control are according to carrying out from the instruction of shop controller 12, though in each machine side by manually also setting and control.In addition, controller is also collected the operate condition of the machine that is connected, and shop controller 12 reads in those data as required.Shop controller 12 except that above-mentioned action, is also read in the action of the quantity of power in this moment at any time by power meter.In addition, shop controller 12, can with Internet connection, the manager can carry out setting, control, supervision, instrumentation and the management etc. of shop controller or machine by remote operation.
In addition, 1 refrigerator 15 and one or more show cabinet 16 are by be connected to become 1 group of refrigeration system.Though only illustrate 1 group among Fig. 2, also can there be many groups, there are many groups in the present embodiment one.Though only illustrate 1 group in Fig. 2, there are many groups in the group of air conditioner 17 and outdoor machine 18.
Also have, the connection of so-called cooling system is meant 1 refrigerator 15 and one or many show cabinets 16, connects by same refrigerant pipe arrangement 19.In each show cabinet, electromagnetically operated valve is installed, the amount of the flowing refrigerant of aperture control arbitrarily.So just, regulate the intake of cooling by the control of show cabinet one side, regulate the cooling situation with refrigerant.
Though because refrigerator 15, be used for the action of the cooling in the show cabinet 16, therefore with show cabinet 16 in cooling action interlock and work, be not the control of directly accepting ON/OFF from show cabinet 16.When refrigerator 15 cut out electromagnetically operated valve for all show cabinets, refrigerator 15 was in the OFF state.That is, close under the situation of electromagnetically operated valve at all show cabinets 16, refrigerator 15 quits work, if even if a show cabinet 16 is being worked, refrigerator 15 also can work on so.
In addition, show cabinet 16 is called the action of defrosting (removing frost) during the cooling action.Be known in the middle frost of can growing of cooling water pipe (heat exchanger, evaporator) of pipe arrangement in the showcase, reduce because of this frost causes cooling performance.This frost need be removed, and removal action that should frost is the defrosting action.When defrosting action, show cabinet 16 is closed electromagnetically operated valve, and Defrost heater is changed to ON.In addition, only, Defrost heater is changed to OFF, opens electromagnetically operated valve once more, carry out quench cooled, restart the cooling action thinking that the stipulated time that frost can be removed defrost after the action.
In addition, this quench cooled becomes one of the essential factor on the peak that produces quantity of power consumption.Defrosting action in each show cabinet is planned in advance.Though almost every day is in the action that can defrost of identical time period, consider implementation time decentralized of defrosting action so that the peak value of the consumed power of many show cabinets does not have overlappingly, and implement above-mentioned plan.Plan is set in the controller of shop controller 12 or show cabinet 16.
In addition, the change of each show cabinet 16, non-constant because of the difference of the environment of its individual difference, show cabinet periphery and refrigerant, when using a plurality of show cabinet 16 at the same time, be difficult to hold the characteristic of the consumed power of their integral body.That is,, must add various factorss such as above-mentioned individual difference, need extremely complicated calculating in order to predict the consumed power amount of many show cabinets 16 accurately by Theoretical Calculation.In addition, also comprise the influence of air-conditioning in the above-mentioned various factors, air-conditioning can bring no small influence for the cooling of commodity in the show cabinet, is unable to cope with in unified calculation.
Therefore, in the present embodiment one, in the prediction of consumed power, do not adopt calculated value, and adopt the saved information form of dynamic resume of the operate condition of basic load machines such as general's show cabinet that for example power consumption is big or air-conditioning machine.In this case, do not need Theoretical Calculation, and can carry out separately predicting accordingly with concrete situation.
Below, narrate at the method for the control of the consumed power amount forecast method in the shop controller 12 in the present embodiment one, load machine.
Fig. 3 represents the functional block diagram of shop controller 12.With reference to Fig. 3,21 for obtaining the quantity of power acquisition portion of quantity of power data from power meter 11,22 timers that are used to carry out time management for clock and watch etc., 23 CPU (CentralProcessing Unit) for the control of various piece of carrying out shop controller 12 and processing etc., 24 is storage parts such as memory or storage device, 25 Department of Communication Forces for communicating with the controller of each load machine, 26 for to carry out the control of each load machine by Department of Communication Force 25, the machine control part of data collection etc., 27 input parts for the setting that is used to carry out shop controller 12 etc., 28 be the display part of the liquid crystal panel of use when the setting of shop controller 12 etc. etc.
Shop controller 12 carries out such action shown below when predicting action.
Shop controller 12 in order to calculate current spendable quantity of power, and obtains the quantity of power data by quantity of power acquisition portion 21 from power meter 11.Calculate spendable quantity of power according to this value and contract quantity of power.
Then, shop controller 12 is predicted consumed power amount from now on.Therefore, communicate with the controller of each load machine, obtain the setting or the operate condition of current each machine by machine control part 26, Department of Communication Force 25.In addition, though the setting of each machine carry out by shop controller 12 since also can be directly in each machine by manually setting change, thereby therefore grasp up-to-date setting by the acquisition operate condition.
Machine control part 26, according to setting, state at this resulting current each machine, (be described in detail later) in the resume on the information form in being stored in storage part 24, whether there is the retrieval of consistent resume part, carries out specific consistent location (below be called " resume position ").If under the situation of specific a plurality of resume position, then select with the resume position in nearest moment in the moment of wherein current time point as should specific resume position.And, when on above-mentioned information form, not having the corresponding resume position of operate condition with the current time point of load machine, finish retrieval actions.
In addition, about the calculating of the predicted value of consumed power amount, adopt the value of the actual power consumption in the information form.That is, begin then to calculate above-mentioned forecast consumption quantity of power to the actual power consumption amount that following direction was obtained based on resume position in above-mentioned information form, specific.In addition, when not having the corresponding resume position of operate condition with the current time point of load machine on the above-mentioned information form, owing to do not have operable actual power consumption amount in the above-mentioned information form of expression, so the above-mentioned above-mentioned deviation that adds up the actual power consumption amount that adds up real consumption quantity of power and current time point that basis is obtained is before this just calculated above-mentioned forecast consumption quantity of power by linear prediction.This method, the existing Forecasting Methodology that describes with reference Fig. 1 is identical.By above method, CPU23 mainly carries out the calculating of the predicted value of consumed power amount.
At last, shop controller 12, predicted value according to the consumed power amount, use quantity of power whether to surpass the judgement of contract quantity of power, if when the forecast consumption quantity of power surpasses in the remaining time limit spendable quantity of power, in order to cut down consumed power amount after this, and carry out the motion limits of particular machines.That is, machine control part 26 is for example considered the control that action to the stopping of the off-premises station of the fewer air conditioner arbitrarily of the influence of others, air conditioner integral body stops etc. by Department of Communication Force 25, cuts down the consumed power amount thus.In addition, however under the situation of judging the reduction that can't realize the consumed power amount, the action that shop controller 12 makes alarm buzzer send the sound etc. can't realize the meaning of the reduction of consumed power amount to people's report, and people's judgement is allowed in control thereafter.That is, will deal with, judge and entrust people,, carry out the reduction of consumed power amount by manually carrying out the motion limits of each machine.
Then, above-mentioned information form is described.
In the information form, preserve the operate condition of main load machine and the consumed power amount of load machine integral body at that time and the resume that are mutually related such as temporal information at that time.Main load machine is the big show cabinet of for example power consumption or air-conditioning machine etc.Fig. 4 represents the schematic diagram of the information form in the present embodiment one.
With reference to Fig. 4, the first from left row are the projects that are kept in this information form, and its adjacent the right one row are contents of the projects of preserving the earliest, are the operate condition of each machine or quantity of power etc.In the present embodiment one, every the right-hand resume that append of 1 fen clockwise table.Owing to retrieve at time-axis direction, therefore the position that certain time one is listed as is represented in so-called above-mentioned resume position in above-mentioned resume position is specific.In addition, information form has for example resume of 10 feed rations.
Top line is the moment in the shop controller 12, is the value that obtains from timer 22.
From top beginning the 2nd row, the operate condition that the expression show cabinet is 1 group.So-called at this said group, in existing a plurality of show cabinets, accept the situation of aggregate of setting, the control of action synchronously, the operate condition also synchronous relation certainly in same group is arranged.That is, this aggregate, different with 1 group of aggregate that connection by above-mentioned cooling system is made up of 1 refrigerator 15 and many show cabinets 16.In addition, in the numerical value of the operate condition in the expression group, the switching of 0 expression action, promptly from the switching of defrosting action to the cooling action, perhaps from the switching of cooling action to the defrosting action, positive value representation is from the elapsed time of the action beginning that defrosts.Negative value representation is from cooling off the elapsed time of action beginning.
Delegation then, expression according to from the group of show cabinet 1 as temperature deviation in the storehouse in the show cabinet of representative, the i.e. standard that determines of the value of " " design temperature "-" temperature in the storehouse " " (temperature is poor in design temperature and the storehouse).The method for selecting of a show cabinet of representative is arbitrarily.And above-mentioned standard if be that absolute value according to " temperature in design temperature-storehouse " is incorporated in certain scope of 0, then is a; Being incorporated in the next scope if surpass this scope, then is b, when further surpassing this scope and being incorporated in the other next scope, then is the mode of c, carries out specified standard.For example, if above-mentioned absolute value then is defined as a more than 0 below 1; If, then be defined as b, if, then be defined as c above 2 and below 3 above 1 and below 2; If in the time of more than it, then be defined as d.
Walk to the 9th behavior from the top the several the 4th and end, be 4 groups of 2 groups of relevant show cabinets and even show cabinets, content is identical during with show cabinet 1.
From top several the 10th row, the 11st row, be the off-premises station of relevant air conditioner 1,2, expression is that off-premises station is being worked in work or compressor reducer.In the present embodiment one, preserve the resume of relevant 2 air-conditioning machines.In addition, as the content of resume, except that the ON/OFF of the action of off-premises station, can also add the air-conditioning machine strong, in, weak operating state.
Its ensuing delegation is illustrated in the actual power consumption amount of that integral body constantly.In the present embodiment one, the difference of preservation and the actual power consumption amount before 1 minute.In addition, also can preserve keeping former state in that actual power consumption amount constantly.
Whether information form the most descending adopted thereafter under the situation of this actual power consumption amount as record information for being illustrated in, be the significant notation of spendable value.Being to be expressed as effectively (can use) at 1 o'clock in the information form, is to be expressed as and must effectively (to use) under 0 the situation.
Needing whether expression consumed power amount (actual power consumption amount) is the information of effective value, is because consider for example following reason.Promptly, under the situation of the time period that only holds a sales exhibition some day etc., quantity of power consumes and usual different can the increase, but this is the situation of the exception different with the everyday actions of every day, in the future if will be worth as learning data use and then be difficult to consider to carry out the high prediction of precision, think that promptly the possibility of the error that generation is bigger in the predicted value uprises.
The moment that is having, order is W from the consumed power amount that power meter 11 obtains, when the mean value of the consumed power amount of order during the past 10 days that with in the information form this is benchmark constantly and standard deviation are respectively M and σ, consumed power amount W is included into by in the represented scope of inequality " M-3 σ≤W≤M+3 σ " time, consumed power amount W is effectively, is invalid when not including in.This scope is for mean value being the scope that 99% sample is contained at the center in regular distribution.In addition, when quantity of power is not effective value, can not be kept in the information form, but chronologically all data are kept in the information form in the present embodiment one.
In addition, the value of the projects in the information form of Fig. 4 adopts value easily, therefore is not the action under the expression actual conditions.
At last, with reference to Fig. 5, adopt flow chart that the action in the shop controller 12 is described.In addition, in this action, whenever process random time (for example 1 minute), carry out the consumed power amount prediction and with this predicted value with the consumed power amount in this moment and, compare with the contract quantity of power, carry out and the corresponding load machine control of this comparative result, also have the renewal processing of information form.
In step S101, quantity of power acquisition portion 21 obtains the difference with 1 minute quantity of power before from the quantity of power of power meter 11 acquisition current times.
In step S102, the difference of the quantity of power that will be obtained in step S101 and the integral power amount before 1 minute are carried out addition, calculate the integral power amount that begins from the zero hour in setting time limit (30 minutes) of current time.In addition, the cumulative power amount was reset by each setting time limit.That is, whenever set the time limit (during the unit) through conduct 30 minutes, the cumulative power amount was returned as zero of initial value.
In step S103, obtain constantly relevant information from timer 22, obtain current time.
In step S104, obtain the operate condition of each machine such as off-premises station operate condition of the operate condition (defrosting action/cooling is moved) of the show cabinet of current times and storehouse temperature, air conditioner by Department of Communication Force 25.
In step S105, according to the operate condition of each machine of the current time that is obtained at step S104, whether the information form with reference to being kept in the storage part 24 exists the retrieval of this resume position, carries out specific to the resume position.That is, to the operate condition of each machine of the current time that in step S104, obtained, whether be kept at information form in information (information of the 2nd～11 row of Fig. 6) certain operate condition of each machine in specific past is consistent judges.And, when having the operate condition of consistent each machine of past, then carry out specific as the resume position position of the operate condition of in store this each machine of past in the information form.At this moment, carry out the specific of resume position with reference to above-mentioned significant notation.
In step S106, whether carry out can specific resume position judgement.That is, if can be on information form in step S105 specific this resume position, then be transferred to step S108, if can not be specific, then be transferred to step S107.
In step S107, since can't be on above-mentioned information form specific resume position, it is the value that does not have consistent situation on the resume, therefore the deviation quantity of power that adopts integral power amount and current cumulative power amount before this is by linear prediction, and the consumed power amount of consumption during till the end from current time to the setting time limit is predicted.This method is identical with the existing Forecasting Methodology that describes with reference to Fig. 1.
In step S108, when the resume position is retrieved, carry out whether existing in a plurality of positions the judgement of this position.Under the situation about all meeting in a plurality of positions, be transferred to step S110; In that only a position sign is fashionable, be transferred to step S109.
In step S109, adopt the actual power consumption amount be kept in the information form, the consumed power amount that consumes during till from current time to the end of setting the time limit is predicted.Specifically, with in the information form by the value after the whole additions of value of the quantity of power of specific resume position time interval amount later, till from current time to the end of setting the time limit as predicted value.
For example, when " being limited to 30 minutes during setting; be 2 o'clock sharps on January 10 zero hour of setting the time limit; current time point is 2: 10 of January 10 ", by specific resume position 8: 4 of January 9, and, if the consumed power amount of assigning to from 8: 4 of January 9 of calculating according to the save value of information form till 8: 24 of January 9 is Wa, then will be from current time to the end of setting the time limit till during the consumed power amount that consumed be predicted as Wa.
In step S110 because candidate and a plurality of position of resume position meet, therefore regardless of the date how, can select during 10 days in the past to carry out specific with the nearest resume position of current time.If further exist a plurality of and this moment during approaching resume position at synchronization, then select the same day or with the same day approaching one day resume position carry out specific.
In step S111, S109 is same with step, adopts the actual power consumption amount be kept in the information form, and the consumed power amount that consumes during till from current time to the end of setting the time limit is predicted.Specifically, will be in table by specific resume position later, begin the value of the time interval whole additions of value amount, quantity of power till the end of setting the time limit as predicted value from current time.
Among the step S112 that after the prediction that finishes to be undertaken, divides a word with a hyphen at the end of a line by step S107, S109 or S111, the cumulative power amount that will calculate by step S102, with calculating, by step S107, S109 or S111 from current time point to the predicted value of setting the consumed power amount that consumes between tailend in time limit and, and the contract quantity of power compare.And, differentiating both and surpassing under the situation of contract quantity of power, carry out the restriction control of the action of each machine.In the present embodiment one, specifically, consider that priority makes the off-premises station of the air conditioner of right quantity stop the motion limits that waits.And the action of carrying out the indoor and outdoor integral body of air conditioner stops, and then the control that whole air conditioners are stopped etc.
After step S112, in step S113, with the operate condition and the time information of each machine of current time, also have and be appended in the information form as record information by the difference of the quantity of power that step S101 obtained etc.Specifically, carry out the defrosting/cooling elapsed time about show cabinet and the judgement of storehouse temperature deviation, this result of determination is appended in information form, the difference of quantity of power that will be corresponding with the 12nd row of Fig. 6 be appended in information form simultaneously.At this moment, above-mentioned significant notation of while also is kept in the information form and (for example carries out the setting of significant notation according to the difference of quantity of power).
In addition, as above-mentioned, have each the load machine that obtains from the controller of each load machine by machine control part 26 and Department of Communication Force 25 operate condition, with the different situation of operate condition of shop controller 12 settings.This difference is directly because of causing by manually carrying out change setting in load machine side.Thinking to produce so different situation, is the situation of the exception different with the everyday actions of every day.Therefore, be difficult to consider in the future to carry out the high prediction of precision, cause predicted value to produce very mistake possibly because of adopting resume value in this case by adopting in the resume value that produces under this different situation.Therefore, can in information form, preserve the information of this meaning as for example discrimination bit.At this moment, in step S105, the resume position is carried out when specific, also have discrimination bit to retrieve simultaneously with reference to above-mentioned significant notation.
(the 2nd execution mode)
Below, with reference to the accompanying drawings, the 2nd execution mode of the present invention (below be called present embodiment two) is described.
Connecting and composing of request control device that present embodiment two is relevant and load machine, identical with connecting and composing of the request control device of present embodiment one shown in Figure 2 and load machine basically.Therefore connecting and composing of present embodiment two is relevant request control device and load machine also has in the explanation of load machine, the explanation omission of the part identical with present embodiment one.
In the request control device of relevant present embodiment two, shop controller 12 is predicted the consumed power amount of the load machine integral body in the setting time limit of asking to control (give and definite time-lag: for example 30 minutes) by each given predetermined period (for example per 30 seconds).When the forecast consumption quantity of power, shop controller 12, adopt:
The past actions state of main load machine;
The mean value (below be called mean consumption power recruitment) of the recruitment of each predetermined period of the consumed power amount of the load machine integral body of this moment, the perhaps maximum of the recruitment of each predetermined period of the consumed power amount of Ci Shi load machine integral body (below be called maximum consumption power recruitment);
The temporal information of this moment,
Information form is mutually related.
Shop controller 12, remaining time before setting the time limit end more for a long time, based on mean consumption power recruitment the consumed power amount of load machine integral body is predicted, remaining time more after a little while, based on maximum consumption power recruitment the consumed power amount of load machine integral body is predicted.Main load machine is off-premises station of show cabinet, air conditioner or this air conditioner that for example the power consumption quantity of power is big etc.
Show cabinet 16 as shown in Figure 2, install when the shop at closing time, its storehouse intraoral illumination is switched off, and allows cold air such overcover of running away.Alleviate the heat load of show cabinet 16 like this.And then also with design temperature change in the object library of show cabinet 16.The operate condition of show cabinet at closing time 16 like this is known as NSB (Night Set Back).
Show cabinet 16 even also can be that purpose reduces the illumination temperature in the storehouse with the effect of saving the energy when running a shop in some cases, improves design temperature in the storehouse simultaneously.The operate condition of the show cabinet during running a shop like this is known as DSB (Demand Set Back).
When the operate condition of show cabinet 16 was NSB or DSB, the consumed power quantitative change that consumes because of show cabinet 16 got smaller.
Because the operate condition of show cabinet 16 is NSB or DSB, can give the no small influence of prediction generating of the consumed power amount of load machine integral body, therefore in the present embodiment two, in order to predict the consumed power amount of load machine integral body accurately, and consider that the operate condition of above-mentioned show cabinet 16 is NSB or DSB.
The controller of show cabinet 16 is that NSB or DSB judge to show cabinet 16, and to shop controller 12 notices.
In addition, under the situation that has many show cabinets 16, it is NSB the moment, DSB the moment, according to being planned in the show cabinet 16 integral body essentially identical mode that becomes usually at the operate condition of these many show cabinets 16.But also each show cabinet 16 is planned respectively in some cases.
Show cabinet 16 in addition, after the defrosting action, heat, carry out quench cooled the storehouse temperature that becomes high temperature is cooled to till the given set point.
Quench cooled after the defrosting action of such show cabinet 16 is known as quick-frozen (pull down) running.Under the situation that has many show cabinets 16, when these many show cabinets 16 major parts are in quick-frozen when running together, with all show cabinets 16 all the consumed power amount of the refrigerators 15 of cooling sharply and significantly increase.
Because whether the operate condition of show cabinet 16 is the quick-frozen running, can give the no small influence of prediction generating of the consumed power amount of load machine integral body, therefore in the present embodiment two, in order accurately the consumed power amount of load machine integral body to be predicted, consider the quick-frozen running, as one of operate condition of show cabinet 16.
In addition since after defrosting action beginning through how long each show cabinet 16 of back is in the quick-frozen running, depend on show cabinet 16 operating condition separately, therefore can't grasp in advance.The controller of show cabinet 16, the operation mode of each show cabinet 16 monitored on one side judge whether each show cabinet 16 become quick-frozen running on one side, and to shop controller 12 notices.
Air conditioner 17, except that greenhouse, cold-room, dry 3 kinds of operating states, the temperature in the shop arrives after the target temperature, can also obtain the 4th operating state (below be called " constant temperature ").In above-mentioned 4 kinds of operating states which kind of controller of air conditioner 17 be in to air conditioner 17 and judge, and to shop controller 12 notices.
In addition, the controller of air conditioner 17 also leaves or closes off-premises station 18 and judges, and to shop controller 12 notices.
At this, so-called off-premises station 18 is meant that for closing air conditioner 17 is in situation about stopping fully, except that air conditioner 17 is in situation about stopping fully, judges that all off-premises station 18 is for opening.
In addition, among Fig. 2,, also can connect many air-conditioning machines in some cases though only connect 1 air conditioner 17 in the off-premises station 18.
Fig. 6 represents the schematic diagram of the information form of relevant present embodiment two.
Construct the information form of present embodiment two by shop controller 12 by each predetermined period (for example per 30 seconds).
Information form is by " time period " hurdle 30 of the time period of preserving each preset time, " show cabinet integral body " hurdle 31, " show cabinet No.1 " hurdle 32 and " show cabinet No.2 " hurdle 33 of preserving the operate condition of show cabinet 16, " air conditioner " hurdle 34 of preserving the operate condition of air conditioner 17, " off-premises station " hurdle 35 of operate condition of preserving off-premises station 18 and " consumed power amount " hurdle 36 that the information relevant with the consumed power amount of load machine integral body is preserved, formation.
In " time period " hurdle 30, preserve the time period that 24 hours were split to form by each given time.For example in the information form of Fig. 6, preserve 24 hours by " 0 point～2 point ", " 2 point～4 point " that were split to form in per two hours ...
" show cabinet integral body " hurdle 31 is made up of " NSB " hurdle 37 and " DSB " hurdle 38.In " NSB " hurdle 37 and " DSB " hurdle 38, preserve identifying information, it is to show cabinet 16 integral body (being that many show cabinets are whole), be NSB, or DSB or NSB and DSB in which carry out specific.When show cabinet 16 integral body are NSB, in " NSB " hurdle 37, preserve " ON ", in " DSB " hurdle 38, preserve " OFF ".
Present embodiment two supposes that show cabinet 16 is made of many show cabinets, and the switching instant of NSB and DSB is according to being in the mode in the identical moment in show cabinet 16 integral body and being planned.
In addition, at the switching instant of NSB and DSB, under the situation about can different modes being planned according to each each show cabinet, " NSB " hurdle and " DSB " hurdle are set by each each show cabinet.
About " show cabinet No.1 " hurdle 32 and " show cabinet No.2 " hurdle 33, by many show cabinets 16, under the situation that for example 2 show cabinet No.1 (not shown) and show cabinet No.2 (not shown) constitute, preserve show cabinet No.1 and show cabinet No.2 operate condition separately.
Specifically, " show cabinet No.1 " hurdle 32, by the operate condition that is used to preserve expression show cabinet No.1 be defrost state (state of the action that defrosts), still quick-frozen state (carrying out the state that quick-frozen is turned round) identifying information, " defrosting " hurdle 39 and " quick-frozen " hurdle 40 form.
When the operate condition of show cabinet No.1 is defrost state, in " defrosting " hurdle 39, preserve " ON ", in " quick-frozen " hurdle 40, preserve " OFF ".
When the operate condition of show cabinet No.1 is in the quick-frozen operating condition, in " defrosting " hurdle 39, preserve " OFF ", in " quick-frozen " hurdle 40, preserve " ON ".
When the operate condition of show cabinet No.1 is common cooling running, in " defrosting " hurdle 39, preserve " OFF " also preservation " OFF " in " quick-frozen " hurdle 40.
" show cabinet No.2 " hurdle 33 is made up of " defrosting " hurdle 41 and " quick-frozen " hurdle 42, according to the rule same with " defrosting " hurdle 39 on " show cabinet No.1 " hurdle 32 and " quick-frozen " hurdle 40, preservation " ON ", " OFF ".
" air conditioner " hurdle 34, by be used for preserving expression air conditioner 17 be in cold-room, greenhouse, drying and constant temperature a certain operate condition identifying information, " cold-room " hurdle 43, " greenhouse " hurdle 44, " drying " hurdle 45 and " constant temperature " hurdle 46 form.
When air conditioner 17, a certain in being in above-mentioned 4 kinds of operate conditions, in the hurdle corresponding, preserve " ON ", preserve in the hurdle beyond this " OFF " with this state.In addition, when air conditioner 17 stops, in all hurdles that constitute " air conditioner " hurdle 34, preserving " OFF ".
In " off-premises station " hurdle 35, preserve the identifying information whether expression stops off-premises station 18.When off-premises station 18 stops, in " off-premises station " hurdle 35, preserving " OFF ", when off-premises station 18 does not stop, preserving " ON ".
" consumed power amount " hurdle 36, by: preserve the mean consumption power recruitment of load machine integral body " mean value " hurdle 47, preserve " average time " hurdle 48 of calculating the number of times after this mean consumption power recruitment, " maximum " hurdle 49 of the maximum consumption power recruitment of preserving load machine integral body and keeping records " record day " hurdle 50 of the date of this maximum consumption power recruitment, time etc.
In information form shown in Figure 6, the row of additional instance such as number 51, expression basic load machine become following operate condition pattern (following this operate condition pattern is called " operate condition Mode A "):
" time period " hurdle 30=" 0 point～2 point "
" NSB " hurdle 37=" ON " of " show cabinet integral body " 31
" DSB " hurdle 38=" OFF " of " show cabinet integral body " 31
" defrosting " the hurdle 39=" OFF " on " show cabinet No.1 " hurdle 32
" quick-frozen " the hurdle 40=" OFF " on " show cabinet No.1 " hurdle 32
" defrosting " the hurdle 41=" OFF " on " show cabinet No.2 " hurdle 33
" quick-frozen " the hurdle 42=" OFF " on " show cabinet No.2 " hurdle 33
" cold-room " the hurdle 43=" ON " on " air conditioner " hurdle 34
" greenhouse " the hurdle 44=" OFF " on " air conditioner " hurdle 34
" drying " the hurdle 45=" OFF " on " air conditioner " hurdle 34
" constant temperature " the hurdle 46=" OFF " on " air conditioner " hurdle 34
" off-premises station " hurdle 35=" ON "
" 0 point～2 point " on " time period " hurdle 30 in the operate condition Mode A is meant the operate condition of each the load machine shown in the operate condition Mode A, and the time period in 0 point～2 is implemented.
In the row that has added number 51, become:
" mean value " the hurdle 47=1500kw on " consumed power amount " hurdle 36
" average time " hurdle on " consumed power amount " hurdle 36 48=10 time
" maximum " the hurdle 49=1750kw on " consumed power amount " hurdle 36
" record day " the hurdle 50=2005.3.1 on " consumed power amount " hurdle 36
These expressions, in the main in the past load machine, the operate condition mode producing identical 10 times (" average time " hurdle on " consumed power amount " hurdle 36 48=10 time) with above-mentioned operate condition A, mean consumption power recruitment and maximum consumption power recruitment in these 10 times that produce, be respectively 1500kw (" mean value " the hurdle 47=1500kw on " consumed power amount " hurdle 36) and 1750kw (" maximum " the hurdle 49=1750kw on " consumed power amount " hurdle 36), the date of having write down this maximum consumption power recruitment is on March 1st, 2005 (" record day " the hurdle 50=2005.3.1 on " consumed power amount " hurdle 36).
In addition, as described later, in the present embodiment two, stipulated the higher limit of the value of preservation in " average time " hurdle 48 on " consumed power amount " hurdle 36, this higher limit is 100 times.
The request control device of relevant present embodiment two is, shop controller 12 by learning the operate condition of main load machine, thereby is constructed information form, based on this information form, the consumed power amount of load machine integral body is predicted.
At first, the method for constructing at the above-mentioned information form that is undertaken by shop controller 12 describes.
Shop controller 12, in the given moment, from each controller of the show cabinet No.1 that constitutes show cabinet 16 and show cabinet No.2, obtain the identifying information of the operate condition of expression show cabinet No.1 and show cabinet No.2, obtain the identifying information of the operate condition of expression air conditioner 17 from the controller of air conditioner 17, obtain the identifying information of the operate condition of expression off-premises station 18.And then, with the above-mentioned given identical moment in the moment, obtain the consumed power amount of load machine integral body from power meter 11.The above-mentioned given moment, synchronous with predetermined period, in present embodiment two, with 30 seconds interval visit.
Power meter 11 is asked setting time limit of controlling by each, carries out the accumulative total of the consumed power amount of load machine integral body.In the present embodiment two, be limited to 30 minute time limit during the setting of asking to control, power meter 11 restarts the accumulative total of the consumed power amount during 30 minutes from 0 fen time point in 30 minute time limit.In addition, need make the timer in the moment of this 0 fen time point, synchronous with the timer that Utilities Electric Co. uses tricks to count in the power meter that is utilized at electricity expense, in the present embodiment two, this adjustment is carried out in advance.
Whether shop controller 12 is present in the above-mentioned information form the operate condition pattern identical with the operate condition pattern of the basic load machine of being made up of " identifying information (corresponding with hurdle 37～46) of the operate condition of expression show cabinet No.1, the show cabinet No.2, air conditioner 17 and the off-premises station 18 that are obtained, in addition obtain time period (corresponding with hurdle 30) under moment of those identifying informations " and retrieves.The result of retrieval is, when above-mentioned operate condition pattern is present in the above-mentioned information form, based on the consumed power amount that is obtained from power meter 11, " mean value " hurdle 47 corresponding with the above-mentioned operate condition pattern in above-mentioned information form, " average time " hurdle 48, " maximum " hurdle 49 and " record day " hurdle 50 are upgraded.
Now, for convenience of description, be conceived to above-mentioned operate condition Mode A, the more new element of the save value on " consumed power amount " hurdle 36 corresponding with the operate condition Mode A is described.When the operate condition pattern identical with the operate condition Mode A is present in the information form, shop controller 12, handle calculating mean consumption power recruitment by the mean value of following formula (1), will calculate resulting value by this and be kept in " mean value " hurdle 47 corresponding with the operate condition Mode A.
Mean consumption power recruitment=(the mean consumption power recruitment * record number of times+recruitment that is write down in the past)/(record number of times+1) ... (1)
At this, so-called " the mean consumption power recruitment that is write down in the past ", the mean consumption power recruitment of being preserved before being meant in " mean value " hurdle 47, so-called " record number of times ", be meant the number of times of the calculating that averages the consumed power recruitment, so-called " recruitment " is meant the consumed power amount of the load machine integral body that is obtained from this power meter 11 and by the difference between the consumed power amount of the load machine integral body that (before promptly 30 seconds) power meter 11 is obtained last time.
In addition, shop controller 12, when the record number of times reached 100 times, the study that becomes the operate condition of main load machine temporarily was through with, and utilized the result of this study in the consumed power amount prediction of load machine integral body.
And shop controller 12 is put in the storage device (not shown) within it, and the record number of times in the memory-type (1) will write down number of times simultaneously and be kept in " average time " hurdle 48 corresponding with operate condition Mode A in the information form.But, reach after 100 times at the record number of times, the renewal on " average time " hurdle 48 is stopped.
Capable corresponding mean consumption power recruitment 1500kW with the number 51 that has added Fig. 6, when being meant under each load machine works by the operate condition shown in the above-mentioned operate condition Mode A, the average out to 1500kW of the consumed power amount of load machine total consumption during 30 seconds.
For example, constructing " the mean consumption power recruitment that in the past write down " and " the record number of times " corresponding with the operate condition Mode A, be respectively under the state of information form of " 1499kW (=13491kW) " and " 9 times ", again observe the operate condition Mode A 1 o'clock sharp constantly, when its recruitment (i.e. the consumed power amount of the load machine integral body till 0: 59: 30 constantly to 1: 0: 0 constantly) when being 1509kW, " mean value " hurdle 47 that will be corresponding with the operate condition Mode A and the preservation content on " average time " hurdle 48 are updated to as shown in Figure 6 " 1500kW (=(13491kW+1509kW)/10) " and " 10 times " respectively.
In addition, when the operate condition pattern identical with the operate condition Mode A is present in the information form, shop controller 12 is handled calculating maximum consumption power recruitment by maximum, will calculate resulting value by this and be kept in " maximum " hurdle 49 corresponding with the operate condition Mode A.
So-called maximum is handled, be called the maximum consumption power recruitment that will before be kept in " maximum " hurdle 49, compare with the recruitment of being calculated according to the consumed power amount of this and the last time that are obtained from power meter 11 (identical) with the recruitment the formula (1), when this recruitment is bigger than previous in store maximum consumption power recruitment, this recruitment is kept in " maximum " hurdle 49 as new maximum consumption power recruitment, and the date (carrying out the date of this preservation) that will obtain this recruitment simultaneously is kept at the processing in " record day " hurdle 50.
Promptly, in " maximum " hurdle 49 corresponding with the operate condition Mode A, when carrying out work, preserve the maximum (based on the maximum of resume) of the consumed power amount that 30 minutes internal burden machine integral body consumed at each load machine by the operate condition shown in the above-mentioned operate condition Mode A.
In addition, in maximum is handled, even when a side of above-mentioned recruitment is very little, will be in " record day " hurdle 50 the maximum consumption power in store past produce and day compare with this resulting date, produce day through more than 1 month the time in the maximum consumption power recruitment in this resulting date from past, regard above-mentioned recruitment as new maximum consumption power recruitment, " maximum " hurdle 49 and " record day " hurdle 50 are upgraded.But, in this upgrades, mean consumption power recruitment and the above-mentioned recruitment that recomputates being compared, value that will a bigger side is regarded maximum consumption power recruitment as, and it is upgraded.
When the operate condition pattern that is obtained is not present in the information form, in this information form, set up new storage area, and with this operate condition pattern to this hurdle of the storage area set up preserve.At this moment, in " mean value " hurdle 47 and " maximum " hurdle 49,, the recruitment of being calculated according to the consumed power amount that is obtained from power meter 11 is preserved in the moment that has obtained this operate condition pattern, in " average time " hurdle 48, preserve " 1 "; In " record day " hurdle 50, preserve the date that has obtained this consumed power amount.
So, shop controller 12 every the operate condition of the main load machine of given period study, is constructed above-mentioned information form.
Shop controller 12, put within it in the storage device (not shown), the operate condition pattern of the storage basic load machine that obtained last time, stored last time the operate condition pattern and this operate condition pattern that is obtained between, except that " time period " hurdle 30, exist when not existing together, judge that then the operate condition of main load machine has changed during predetermined period (30 seconds) process.At this moment, owing to vividly do the influence that state changes for the consumed power volume production that is obtained, therefore operate condition pattern and the consumed power amount that is obtained can not be reflected in the information form.
In addition, shop controller 12 is not the storage area of preparing the information form corresponding with the everything state model of basic load machine in advance, but when producing new operate condition pattern, the storage area of information form is amplified at every turn.The operate condition pattern that produces because of the environmental condition in shop can be different, even if possible, the operate condition pattern of Chan Shenging is not a lot of yet.Thereby by adopting as above-mentioned when producing new operate condition pattern, the method that the storage area of information form is amplified then, thus even if possible, also the storage area that is used for the operate condition pattern that reality do not produce can be cut down.
Then, the Forecasting Methodology at request consumed power amount control, load machine integral body that is used for being undertaken by shop controller 12 describes.
Shop controller 12, based on operate condition pattern by the basic load machine that each predetermined period obtained, information form is upgraded, suppose that simultaneously this operate condition pattern continued always, predicted the consumed power amount of load machine integral body before 30 minute finish time in time limit of asking to control.In addition, the general forecast cycle is till 10 seconds to 1 minute, in the present embodiment two, as above-mentioned be 30 seconds.
Specifically, in information form, whether judgement is 100 corresponding to the value on " average time " hurdle 48 of the operate condition pattern identical with the operate condition pattern that is obtained.When the value on " average time " hurdle 48 does not reach 100, judgement is for the not fully study of this operate condition pattern, only adopt the recruitment calculated according to the consumed power amount of the current load machine integral body that is obtained from power meter 11 (promptly, suppose and continue same consumed power state), the consumed power amount when 30 minute time limit finished is predicted.That is, shop controller 12 only according to the consumed power amount of the load machine integral body that is obtained by power meter 11, calculates the forecast consumption quantity of power of the load machine integral body of 30 minute time limit when finishing based on formula (2).
Forecast consumption quantity of power=(recruitment/predetermined period) * remaining time+consumed power amount till current ... (2)
At this, " recruitment " in the formula (2) is identical with recruitment in the formula (1)." remaining time " is the remaining time till when finishing from current time point to 30 minute time limit." the consumed power amount till current " is from the accumulation consumed power amount of request control beginning time point till current.
For example, when request control begins through 4 minutes and 30 seconds, in formula (2), " recruitment ", being request control beginning back begins after the request control beginning recruitment through the consumed power amount in moment of 4 minutes 30 seconds from the consumed power amount through 4 minutes time point, " remaining time " is 1530 seconds (=30 minutes and-4 minutes and 30 seconds), and " the consumed power amount till current " is to through the accumulation consumed power amount till 4 minutes 30 seconds the time point from request control beginning.
When " average time " hurdle 48 corresponding to the operate condition pattern (be called operate condition Mode B) identical with the operate condition pattern that is obtained reaches 100 times, based on " mean value " hurdle 47 corresponding with this operate condition Mode B or the value on " maximum " hurdle 49, the consumed power amount when 30 minute time limit finished is predicted.
Specifically, remaining time before the end in 30 minute time limit more for a long time, for example be more than 10 minutes the time remaining time, read out in mean consumption power recruitment in store in " mean value " hurdle 47 corresponding with the operate condition Mode B, when remaining time more after a little while, for example be that less than is in the time of 10 minutes, read out in maximum consumption power recruitment in store in " maximum " hurdle 49 corresponding, calculate the forecast consumption quantity of power of the load machine integral body when 30 minute time limit finished based on following formula (3) with the operate condition Mode B.
Forecast consumption quantity of power=(recruitment/predetermined period) * remaining time+consumed power amount till current ... (3)
As the recruitment in the formula (3), be more than 10 minutes the time when remaining time, adopt mean consumption power recruitment, when less than in the time of 10 minutes, adopt maximum consumption power recruitment.
When remaining time is more than 10 minutes the time, and based on the mean consumption power recruitment corresponding with the operate condition Mode B, through type (3) calculates the forecast consumption quantity of power.
When the remaining time of less than in the time of 10 minutes, based on the average maximum consumption power recruitment corresponding with the operate condition Mode B, through type (3) calculates the forecast consumption quantity of power.
In the request control device of relevant present embodiment two, shop controller 12 based on by the forecast consumption quantity of power that said method calculated, carries out for example following such control for the basic load machine of shop controller 12 management.
When the forecast consumption quantity of power is bigger than predefined set-point,, carry out the running control that air conditioner stops by given rule as surpassing the contract quantity of power.Though this rule is predesignated, but the also air conditioner that is stopped according to the degree decision that surpasses set-point, also can determine the priority of the air-conditioning machine that stopped, also can be predetermined according to the mode of the air-conditioning machine equilibrium that is stopped the air-conditioning machine to be stopped in turn, method.Though also can consider method that 15 runnings of show cabinet 16 or refrigerator are stopped, yet owing to produce baneful influence also may for the preservation state of food in the storehouse that is kept at show cabinet 16, so in the present embodiment two air conditioner 17 turned round and stop.
In addition, when longer than 10 minutes,, therefore keep the comfort level of shop facility emphatically, the request control of over head time is arranged owing to till finishing, there is the unnecessary time than the remaining time till the end in 30 minute time limit.On the contrary, when less than 10 minutes and remaining time more after a little while, the consumed power amount is no more than set-point emphatically, becomes the control that the consumed power amount is cut down in advance.
Fig. 7 represents the flow chart of the request control action that undertaken by shop controller 12.
In step S200, shop controller 12 carries out the operate condition of basic load machine and the study of the consumed power amount of the load machine integral body of this moment.That is, shop controller 12 obtains the operate condition pattern of basic load machine and the consumed power amount of load machine integral body by each predetermined period, constructs information form as shown in Figure 6.
In step S300, shop controller 12, based on the consumed power amount of operate condition pattern that is obtained by step S200 and load machine integral body (average time less than 100 times time), perhaps based on mean consumption power recruitment corresponding or maximum consumption power recruitment in operate condition pattern that is obtained by step S200 and the information form, the forecast consumption quantity of power of computational load machine integral body with this operate condition pattern.
Then, in step S400, shop controller 12 based on the forecast consumption quantity of power, according to setting in the time limit, is no more than the mode of contract quantity of power, carries out load the operation of a machine control that shop controller 12 is managed.
Shop controller 12 is by each predetermined period repeating step S200, S300 and S400.
Fig. 8 in as shown in Figure 7 the flow chart of expression, the flow chart of the detailed action of the study of the step S200 shop controller 12 in handling.
In step S201, shop controller 12, the identifying information of acquisition operate condition from the basic load machine, the consumed power amount of acquisition load machine integral body from power meter 11 is transferred to step S202 simultaneously.
In step S202, shop controller 12 by the time clock feature of self, obtains to have obtained the moment of above-mentioned identifying information and above-mentioned consumed power amount, is transferred to step S203.
In step S203, carry out judgement that the time period under the moment that is obtained is classified, be transferred to step S204.
In step S204, information form is retrieved, whether the operate condition pattern identical with the operate condition pattern of being made up of the operate condition of the basic load machine that is obtained and above-mentioned time period of being calculated is present in the information form judges.In addition, in this is judged, also can adopt the method for the homogeny of ignoring the time period.When having identical operate condition pattern, be transferred to step S205, otherwise, be transferred to step S206.
In step S205, there is the row (row shown in the row 51 of Fig. 6) with the operate condition pattern that the existing operate condition pattern that obtains is identical in information form in selection, is transferred to step S207.
In step S206, in information form, append new row as being used for, preserve the new operate condition pattern that is obtained, be transferred to step S207 at this new row to new operate condition pattern that is obtained and the zone of preserving according to the value that the consumed power amount is calculated.
In step S207, to upgrading (perhaps preserving again) with operate condition pattern mean consumption power recruitment corresponding, load machine integral body, average time, the maximum consumption power recruitment of the row that is appended by the selected row of step S205 or by step S206 and date of producing maximum consumption power recruitment.
Fig. 9 represents in as shown in Figure 7 the flow chart, the flow chart of the detailed action of the shop controller 12 in the consumed power amount prediction processing of step S300.
In step S301, shop controller 12 from the identifying information of basic load machine acquisition operate condition, obtains the consumed power amount of load machine integral body simultaneously from power meter 11, be transferred to step S302.
In step S302, shop controller 12 by the time clock feature of self, obtains to have obtained the moment of above-mentioned identifying information and above-mentioned consumed power amount, is transferred to step S303.
In step S303, carry out judgement that the time period under the moment that is obtained is classified, be transferred to step S304.
In step S304, information form is retrieved, whether the operate condition pattern identical with the operate condition pattern of being made up of the operate condition of the basic load machine that is obtained and above-mentioned time period of being calculated is present in the information form judges.In addition, in this is judged, also can adopt the method for the homogeny of ignoring the time period.When having identical operate condition pattern, be transferred to step S305, otherwise, be transferred to step S306.But, even there is identical operate condition pattern, when " average time " less than corresponding 100 times with its operate condition pattern, as fully study be transferred to step S306.
In step S305, judge up to setting the remaining time of time limit till when finishing whether in 10 minutes, if in 10 minutes, then be transferred to step S307, otherwise be transferred to step S308.
In step S306, the recruitment according to the consumed power amount in consumed power amount calculating predetermined period of the load machine integral body that is obtained is transferred to step S309.
In step S307, with in the information form, the corresponding maximum consumption power recruitment of the operate condition pattern identical with the operate condition pattern that is obtained is as recruitment, is transferred to step S309.
In step S308, with in the information form, the corresponding mean consumption power recruitment of the operate condition pattern identical with the operate condition pattern that is obtained is as recruitment, is transferred to step S309.
In step S309,, calculate the forecast consumption quantity of power of the load machine integral body when the setting time limit finishes based on above-mentioned formula (2) or (3).
Figure 10 represents in as shown in Figure 7 the flow chart, the flow chart of the detailed action of the shop controller 12 in the request control and treatment of step S400.
In step S401, judge whether the forecast consumption quantity of power that step S309 calculated by Fig. 9 surpasses the signatory contract quantity of power with Utilities Electric Co..When surpassing the contract quantity of power, be transferred to step S402, otherwise end process.
In step S402, when having many air conditioners 17, judge whether air conditioner 17 all stops, and when stopping, then being transferred to step S403, otherwise be transferred to step S404.
In step S403, the alarm buzzer (not shown) that shop controller 12 is possessed sends the sound, arouses shop manager's etc. attention.
In step 404, stop all air conditioners 17.
As previously discussed, in the present embodiment two, shop controller 12 as the request control device, the operate condition pattern of the basic load machine of study shop controller 12 management and mean consumption power recruitment, maximum consumption power recruitment and the consumed power recruitment of the load machine integral body of this moment are constructed information form.
In addition, shop controller 12, based on operate condition pattern mean consumption power corresponding, in information form the recruitment or the maximum consumption power of the basic load machine that is obtained, perhaps, calculate the forecast consumption quantity of power when the setting time limit finishes based on the consumed power recruitment of being calculated according to the consumed power amount of the load machine integral body that is obtained.
Further, shop controller 12, when the forecast consumption quantity of power surpassed the contract quantity of power, basic load machine, for example air-conditioning machine that shop controller 12 is managed stopped.
In present embodiment two, shop controller 12 is not that all the load machines in the shop are managed and control.Therefore even the operate condition of the basic load machine managed of shop controller 12 is identical, the influence of the consumed power of the load machine of not managing because of shop controller 12, the consumed power amount of load machine integral body can produce deviation in some cases.In this case, based on the operate condition of basic load machine only according to the mean consumption power recruitment of load machine integral body, consumed power amount when 30 minute time limit finished is predicted, ask control as if predicting the outcome based on this, then Shi Ji consumed power amount can surpass the forecast consumption quantity of power.Even in this case, if a lot of situation remaining time before when 30 minute time limit finished, then after calculating the forecast consumption quantity of power once more, request control is revised, thereby can when 30 minute time limit finished, be carried out the feasible contract quantity of power that is no more than of control.Yet even recomputate the forecast consumption quantity of power, revise request control the remaining time before when 30 minute time limit finished more after a little while, also has to surpass the possible of contract quantity of power when 30 minute time limit finished.
Consider this factor, in the present embodiment two, as above-mentioned remaining time before when 30 minute time limit finished more after a little while, be not to adopt mean consumption power recruitment but adopt maximum consumption power recruitment, calculate the forecast consumption quantity of power.Like this, a little even actual consumed power amount increases, also can avoid the cumulative actual consumed power amount when 30 minute time limit finished to surpass the forecast consumption quantity of power.
In addition, can think the different influence of time period, one of reason of disperseing as the relation of the consumed power amount of operate condition that makes the basic load machine and load machine integral body.For example, the operate condition of show cabinet 16 or refrigerator 15 or consumed power amount, the influence that is subjected in the storehouse of show cabinet 16, to move into food or comes the shop personnel amount, and in the storehouse, move into food or come the shop personnel amount to depend on the time period.In addition, other load machines of management possibility of yet depending on the time period is not higher for 12 of shop controllers.Therefore, as the 2nd execution mode, by the operate condition of basic load machine being classified, thereby can calculate the forecast consumption quantity of power accurately by each time period.
In addition, when show cabinet No.1 or show cabinet No.2 are in the defrosting running, even based on its operate condition mode computation forecast consumption quantity of power, also can be because of beginning quick-frozen running in during unexpected cause that actual consumed power amount surpasses the forecast consumption quantity of power, the correspondence that has in the request control produces the possibility that postpones.
Therefore, in the present embodiment two, when show cabinet No.1 or show cabinet No.2 are in defrosting running (defrosting action), the forecast consumption quantity of power of the operate condition mode computation of the forecast consumption quantity of power of the operate condition mode computation when becoming the quick-frozen running to changing based on their when being in the defrosting running based on them compares, when the former (the forecast consumption quantity of power during the quick-frozen running) when a side is big, can adopt the former as the forecast consumption quantity of power that should finally predict.In addition,, also be to use mean consumption power recruitment, also according to the rule of remaining time even if in should handling, be to use maximum consumption power recruitment.
In addition, in the present embodiment two,,, except that in addition, also can adopt bigger illumination 13 or other machine 14 of proportion in the consumed power amount of shop facility though adopt show cabinet 16, air conditioner 17 and off-premises station 18 as the basic load machine.
In addition, in the present embodiment two, though the operate condition pattern of basic load machine was removed outside the time period, the ON and the OFF that show with the two-value as each project (operation mode) represent that in the software of reality, they are represented with " 1 " and " 0 ".Because each project all with ON and OFF two value representations, therefore can be given 1 of each allocation of items on the memory of software.So, to compare with giving about 1 byte of each allocation of items, memory span certainly can be littler.And, even the retrieval of the operate condition pattern in carrying out information form also can be handled learning data easily.In addition, the time period that will preserve to " time period " hurdle 30 among Fig. 6, show according to the mode of 0 point～2.At this moment,, therefore the time period is carried out information specific, the numerical value performance of 12 kinds till can 0～11 owing to constituted by 24 hours in 1 day.Consider this situation, just can be by the existing time period of 4 bit tables.
(the 3rd execution mode)
Below, with reference to accompanying drawing, the 3rd execution mode of the present invention (below be called present embodiment three) is described.
The request control device of present embodiment three and connecting and composing of load device, identical with connecting and composing of the request control device of present embodiment one shown in Figure 2 and load machine basically.Therefore with the request control device of relevant present embodiment three and load machine connect and compose and the explanation of load machine in, with the explanation omission of the common part of present embodiment one.In addition, the study processing of relevant present embodiment three, prediction processing and request control and treatment are identical with the processing shown in the present embodiment two basically.Therefore with relevant in above-mentioned 3 kinds of processing of present embodiment three and the explanation omission of the common part of present embodiment two.
As above-mentioned show cabinet 16, though after the defrosting action, heating, make the storehouse temperature that becomes high temperature be cooled to quick-frozen running till the given set point, but add heat run relatively after this quick-frozen running and the defrosting action, its consumed power measurer has bigger value.When having many show cabinets 16, be in together in more than half part of these many show cabinets 16 under the situation of quick-frozen running, the consumed power amount of refrigerator 15 sharply increases.So whether the operate condition of show cabinet 16 can be given the prediction of the consumed power amount of load machine integral body for the quick-frozen running and bring no small influence.
On the other hand, because each show cabinet 16 is becoming the quick-frozen running after through how long after the beginning of defrosting action, carry out quick-frozen how long running, depend on show cabinet 16 individual difference or the environment placed of each show cabinet 16 etc. separately, therefore can't grasp in advance, so the expanded range of its deviation.Therefore if observe the situation of the information form of construction drawing 6, then by the resulting operate condition patterns such as step S201 of for example Fig. 8, compare with the operate condition pattern of this time period of the day before yesterday, can there be some differences with higher probability.Its result is, mainly become by step S204 to step 3206 branch, promptly in information form, append the situation of new row.Its result is that the amount of information of information form enlarges, the memory space increase of retrieving the needed time or needing.
Therefore, in the present embodiment three, in order to get rid of the influence of the above such deviation, by show cabinet is carried out modelling at the operate condition (operating condition) of sequential direction, thus the analogue formation pattern.During this modelling, consider influence to the consumed power amount of reality, be conceived to quick-frozen running and the defrosting of show cabinet and each running that dewaters to the no small influence of prediction generating of the consumed power amount of load machine integral body.Below, will be conceived to the model of these runnings, be called " defrosting model ".In addition, the what is called running that dewaters is meant by the running of the defrosting in the show cabinet ice to be dissolved and with the running of generation water discharge.
Specifically, the operate condition of the acquisition load machine that replacement is undertaken by the step 201 of Fig. 8, with the plan of defrosting running (below be called " defrosting operating plan ") be benchmark, makes the operate condition (following model with this made is called " operate condition model ") of each show cabinet of the above-mentioned defrosting model of application on model.And, the operate condition of each show cabinet is considered as being present in the operate condition shown in this operate condition model, with reference to the information of operate condition model.And, in the present embodiment three, do not need as present embodiment one, two, use by the controller of show cabinet 16 to shop controller 12 notices, whether this show cabinet 16 be the information of quick-frozen running etc.
Specifically, in the defrosting model, preestablish defrosting running, the running that dewaters, quick-frozen needed separately time of running.This needed separately time, by experiment or decision optimum value in advance such as instrumentation.In the present embodiment three, they were set to respectively 5 minutes, 10 minutes, 30 minutes.In addition, this setting can be changed by each show cabinet.Use this defrosting model and defrosting operating plan, make the operate condition model of each each plan constantly.According to this operate condition model, can determine which kind of state when which show cabinet be considered as being in.In addition, " defrosting running " and " defrosting action " is equivalent.
Figure 11 represents the table of the operate condition model of made.This table is equivalent to the schematic diagram of the information form in the present embodiment shown in Figure 4, represents the representational pattern of timing variations of the operate condition of each show cabinet.In the present embodiment three, suppose that show cabinet 16 is many, for example situation about constituting by 3 show cabinet No.1, show cabinet No.2, show cabinet No.3 (all diagram).Y direction is represented each state of each show cabinet." 1 " among the figure, the expression show cabinet is in this state.For example in 8: 2 timesharing, expression show cabinet No.1 is in defrost state (state of the running that defrosts), show cabinet No.2 and No 3 and is in quick-frozen state (carrying out the state of quick-frozen running).
Figure 12 represents the 2nd form of present embodiment three, is equivalent to the Fig. 6 in the present embodiment two.Mean value among the figure or maximum are the value in 1 minute.In addition, though not shown in the present embodiment three, in the 2nd form, comprise Department of Communication Force 25 each load machine that obtained, except that show cabinet the operate condition of air conditioner or off-premises station etc. for example by Fig. 3.
In the 2nd form in present embodiment three, for convenience of description, omit the situation of using NSB or DSB.And, in the defrosting model, as mentioned above, also consider to dewater running.The defrosting of each show cabinet in the 2nd form (information form), dewater, ON/OFF information of each running of quick-frozen, as mentioned above, depend on defrosting model be applied to the to defrost operate condition model of each show cabinet of operating plan made.About these, can be clear and definite according to the explanation of back detailed process.
Figure 13 represents the flow chart of request control action in the present embodiment three, that undertaken by shop controller 12.In the present embodiment three, after the processing of step S500, carried out circulation in this flow process (step S200～S400) every 1 minute.In the present embodiment three, just describe with the example that was unit asks to control in 30 minutes.Therefore, below, will be called the finish time from the moment after zero hour to 30 of request control minute, the current time in these 30 minutes is called current time (0～29 minute scope).In addition, in the present embodiment three, compare, appended step S500 with the situation of Fig. 7 of present embodiment two.
In step S500, read out in the defrosting operating plan of setting in the shop controller 12.And the model that will defrost is applied to this in the works, makes the such operate condition model of Figure 11.In each show cabinet, with defrosting, dewater and each running of quick-frozen is the plan carried out when, the i.e. operating plan that defrosts, be set in advance in the shop controller 12.Consider to realize making the decentralized of can be not overlapping not such each running of execution of the peak value of the consumed power of many show cabinets, the plan of decision defrosting running.In addition, though the defrosting operating plan is set by shop controller 12, also can set by the controller of show cabinet 16.
Figure 14 represents in as shown in figure 13 the flow chart, the flow chart of the detailed action of the shop controller 12 of the study of step S200 in handling.Compare with the situation of Fig. 8 of present embodiment two, not steps performed S201.This be because as above-mentioned, do not need to obtain the reason of the actual act state of load machine in the present embodiment three.In addition, compare, append step S210～S212 with the situation of Fig. 8.Especially emphatically these differences below are elaborated.
In step S202, shop controller 12 obtains current time, then in step S203, carries out judgement that the time period under the current time that is obtained is classified, is transferred to step S210.
In step S210,, from operate condition model, extract the operate condition (operating condition) of each current show cabinet on one side by step S500 manufacturing on one side with reference to the current time that is obtained by step S202.For example, be 8: 2 timesharing at current time, extract the operate condition of " the show cabinet No.1 running that defrosts, and show cabinet No.2 and No.3 carry out quick-frozen and turn round ".
Then step S210 is in step S211, and the consumed power amount from the current load machine integral body of power meter 11 acquisitions is transferred to step S212.
In step S212, will be stored in the memory simultaneously by the information that step S202, S203, S210 and S211 are obtained is interrelated.To be called the 1st form by the constructed table of these data of preserving (table data).The 1st form adopts the extraction data made from the form of Fig. 1.Therefore, the 1st form, same with the table of Figure 11, represent the representational pattern of timing variations of the operate condition of each show cabinet.
In addition, the 1st form is preserved the data in 1 week.On the other hand, by the operate condition model of step S500 made, do not have relevant key element every day.That is, this operate condition model is for carrying out the set of specific data to the operating condition of each show cabinet in 1 day.Therefore, though in the 1st form, preserve the data in 1 week, yet in the 1st form with the operate condition of related each show cabinet of the identical moment (8: 2 grade), have nothing to do with every day, all identical.But the data of the relevant consumed power amount that is obtained by step S211 are kept at the amount in 1 week in the 1st form.The data that are kept at the consumed power amount in the 1st form reach after the amount in 1 week, from former place data are covered.
And, though do not adopt in the present embodiment three, also can in the 1st form, further preserve communications status by each load machine, for example air conditioner or off-premises station etc. beyond the show cabinet that Department of Communication Force 25 obtained of Fig. 3.
If step S212 finishes, then be transferred to step S213.In step S213～S215, carry out the processing that takes place among the step S204～S206 of present embodiment two.
In step S213, judge in the 2nd form (information form), whether there is identical with the operate condition pattern of forming by the operate condition that is kept at each show cabinet in the 1st form with related time period (promptly being kept at the operate condition and the time period of each show cabinet in the 1st form at the moment) of this operate condition, operate condition pattern.In addition, in this is judged, also can adopt the method for the homogeny of ignoring the time period.And, when having identical operate condition pattern, the row in the 2nd form (information form) of in store this operate condition pattern is carried out specific, be transferred to step S214, otherwise be transferred to step S215.
In step S214, select the row (original row) in the 2nd form (information form) of in store same action state model, be transferred to step S216.In step S215, in the 2nd form, append new row as the zone that is used for preserving, in this new row, preserve the new operate condition pattern that is obtained, be transferred to step S216 according to the value of being calculated by new operate condition pattern that is obtained and consumed power amount.
In step S216, carry out the processing that step S207 took place of present embodiment two.In step S216, mean consumption power recruitment, average time, the maximum consumption power recruitment of operate condition pattern and the load machine integral body corresponding with it and time on date of producing maximum consumption power recruitment are classified with the time period, be kept at simultaneously in the 2nd form.
Specifically, employing is kept at the data of the consumed power amount in the 1st form, to upgrading (perhaps preserving again) with operate condition pattern mean consumption power recruitment corresponding, load machine integral body, average time, the maximum consumption power recruitment of the row that is appended by the selected row of the step S214 in the 2nd form or by step S215 and date of producing maximum consumption power recruitment.Mean consumption power recruitment, average time, maximum consumption power recruitment and the date that produces maximum consumption power recruitment, same with the content that present embodiment two is illustrated, decide according to present embodiment two illustrated regulations.
And, though the 2nd form is equivalent to the information form by step S207 (Fig. 8) made of present embodiment two, compare its inscape difference (with reference to Figure 12) with the 2nd execution mode.And, though the diagram difference of Figure 13 and Figure 14, but in fact, every two hours execution in step S216, carry out once these two hours the amount the 2nd form renewal (therefore, in the present embodiment three, adopt the data of the consumed power amount that will be obtained by step S211 to preserve the once method in the 1st form) every 1 minute.In the moment of this renewal, be for example 8 o'clock sharps, 10 o'clock sharps etc.
Figure 15 represents in as shown in figure 13 the flow chart, the flow chart of the detailed action of the shop controller 12 in the consumed power amount prediction processing of step S300.Compare with the situation of Fig. 9 of present embodiment two, do not carry out step S301.This be because as above-mentioned, in present embodiment three, do not need to obtain the reason of the actual act state of load machine.And, compare with the situation of Fig. 8, append step S310～S312 and step S317, S318.In addition, step S313～S316, for present embodiment two step S306～S309 took place.About being described in detail below these.
At first, in step S302, shop controller 12 obtains current time, then in step S303, carries out judgement that the time period under the current time that is obtained is classified, is transferred to step S310.In step S310, from the 1st form, read the consumed power amount of the load machine integral body in this moment (current time), be transferred to step S311.
In step S311, adopt the integral power amount before current 1 minute and the deviation (deviation quantity of power) of current integral power amount, by linear prediction, the forecast consumption quantity of power (this forecast consumption quantity of power is called linear predicted value P) that is consumed till calculating after current time to 1 minute.Linear predicted value P equates with above-mentioned deviation (deviation quantity of power).And i is set to after 29 from cycle tiemr (looptimer), is transferred to step S312.The value of cycle tiemr i, the elapsed time (that is above-mentioned current time) that begin the zero hour that the imagination expression is controlled from the request that is with 30 minutes unit carries out.
In step S312, from the 1st form, read the operate condition etc. of the show cabinet of given time.For example, in the zero hour that with 30 minutes was the request control carried out of unit is that 8: 00 and current time are under 8: 10 the situation, when i=29, from the 1st form, read (except the data of consumed power amount) such as operate conditions with 8 o'clock 29 fens each corresponding show cabinets; When i=28, from the 1st form, read (except the data of consumed power amount) such as operate conditions with 8 o'clock 28 fens each corresponding show cabinets.When i≤27 too.Though cycle tiemr i shows the time than current more past future, but owing in the 1st form, record the operate condition model that is suitable for defrosting model made and based on the operate condition of each following load machine of operate condition model record etc., therefore can carry out above-mentioned reading.
In addition, as above-mentioned, though in the 1st form, preserve data in 1 week, the operate condition of each show cabinet related with synchronization (8: 2 grade), with have nothing to do every day all identical.Therefore, in above-mentioned numerical example, when i=29, from the 1st form, read and any (except that the data of consumed power amount) such as operate conditions of 8 o'clock 29 fens each corresponding show cabinets of every day.
If step S312 finishes then to be transferred to step S304.In step S304, carry out the processing that present embodiment two takes place.That is, the 2nd form is retrieved, judged whether identical with the operate condition pattern of forming by the operate condition of each show cabinet of reading with related time period of this operate condition, operate condition pattern is present in the 2nd form in step S312.In addition, in this is judged, also can adopt the method for the homogeny of ignoring the time period.When having identical operate condition pattern, be transferred to step S305; Otherwise, be transferred to step S313.But,, when " average time " corresponding with this operate condition pattern is no more than given number of times (for example 10 times),, be transferred to step S313 as not fully study even there is identical operate condition pattern.In addition, below will be judged as the operate condition pattern in the 2nd form of same action state model, be called operate condition pattern C.
In step S305, judge whether current time is in (finish time in 30 minute time limit) the 10 minutes finish time, when being in 10 minutes the time, be transferred to step S314, when not in 10 minutes, be transferred to step S315.
Step S313～S315 carries out the processing that present embodiment two takes place.In step S313, the linear predicted value P that will be calculated by step S311 is transferred to step S316 as form predicted value Q (substitution Q).
In step S314, maximum consumption power recruitment that will be corresponding with the operate condition pattern C in the 2nd form is transferred to step S316 as form predicted value Q (substitution Q).The assumed load machine with operate condition pattern C corresponding operation state under when working, by the form predicted value Q that step S314 is determined, the maximum consumption quantity of power of expression load machine integral body in 1 minute.
In step S315, mean consumption power recruitment that will be corresponding with the operate condition pattern C in the 2nd form is transferred to step S316 as form predicted value Q (substitution Q).The assumed load machine with operate condition pattern C corresponding operation state under when working, by the table predicted value Q that step S315 is determined, the mean consumption quantity of power of expression load machine integral body in 1 minute.
In step S316, the predicted value POW of the consumed power amount of computational load machine integral body in 1 minute iAt predicted value POW iMiddle " i " that adds, the value of expression cycle tiemr i, the value of i desirable 0～29.Specifically, the value of the weighting after only taking advantage of α doubly the difference of form predicted value Q and linear predicted value P, with linear predicted value P with carry out the resulting value of addition, as predicted value POW i(that is POW, i=P+ α (Q-P)).The value of this α can be customized for optimum value according to the environment of controlling object.In principle, set α, but α promptly can be 0 also can be 1 according to the mode that set up " 0＜α＜1 ".
In addition, the predicted value POW during obtaining above-mentioned 1 minute iWhen, can adopt the mean value of linear predicted value P of given period amount in the past or maximum (with it as P ').Its result is to adopt POW i=P+ α (Q-P ').
As mentioned above, in the present embodiment, mainly adopt linear predicted value P to carry out the prediction of following consumed power amount, use the predicted value (form predicted value Q) that has adopted the 2nd form because of the purpose of the linear predicted value P of revisal.
Processing at step S316 finishes among the step S317 of back transfer cycle tiemr i only to be reduced after 1, and i and above-mentioned current time (unit is for dividing) are compared.And,, return step S312 when i is current time when above.When i is during less than current time, then be transferred to step S318.That is the POW during each 1 minute till calculating, from current time to the finish time iBefore, repeat the circulation of step S312～S316 after, be transferred to step S318.
In step S318, to the POW till from current time to the finish time iAdd up, calculate the recruitment of the finish time relative of forecast consumption power before with the consumed power amount of current time.By with this recruitment, add up to the accumulative total consumed power amount before the current time that is calculated according to the acquisition value among the step S211 (Figure 14), just can calculate at order request control beginning time point is the forecast consumption quantity of power (accumulation consumed power amount) of finish time of zero.
Identical among the flow chart of the detailed action of the shop controller 12 in the flow chart as shown in figure 13, in the request control and treatment of step S400 and Figure 10.That is, shop controller 12 carries out based on the suitable control in the forecast consumption quantity of power of the above-mentioned finish time (accumulation consumed power amount) for the load machine.
In present embodiment three, with reference to according to being not the actual act state of load machine but the 1st form of operate condition model made, thereby construct the 2nd form that should become information form as above-mentioned.Therefore, can reduce the deviation of the operate condition relevant with the load machine on the information form, its result is can cut down the amount of information of information form, the memory space that can cut down the needed time of retrieval simultaneously or need.In addition, realization can avoid asking the effect of the communication cost etc. between control device and the load machine.In addition, for example as mentioned above, employing by linear prediction the forecast consumption quantity of power and adopt the forecast consumption quantity of power of the 2nd form, by getting their weighted average, revisal ground calculates the forecast consumption quantity of power, thereby can be expected to fill up the prediction by linear prediction and adopt shortcoming in prediction both sides' the predicted method of the 2nd form, consumed power amount accuracy of predicting is improved.
The request control device of above-mentioned present embodiment with and the consumed power Forecasting Methodology that adopts, adopt CPU, memory, other LSI hardware such as (Large Scale Integration) of any computer to realize, in addition, can realize by the software that is loaded on realization request controlled function in the memory and program of consumed power forecast function etc.In addition, under the situation of the use of considering software, among above-mentioned Fig. 3, the functional block diagram of the request control device that expression realizes by hardware and software.But these functional blocks can only be used hardware certainly, only realize with software or their various forms such as combination.
Embodiments of the present invention in the technological thought scope shown in the scope of patent request, also can suitably be carried out various changes.

Claims (6)

1, a kind of request control device, it carries out the control of load machine based on the predicted forecast consumption quantity of power of the consumption of load machine before the finish time during the unit, it is characterized in that having:
Form generates mechanism, and operate condition, the actual power consumption amount of this moment and the time information of this moment of the load machine that is set in advance in the load machine in its object unit that will ask to control carry out association and generate information form as resume; With
Prediction and calculation mechanism, it adopts by this form and generates the information form that mechanism generated, and calculates described forecast consumption quantity of power,
Described form generates mechanism and has preservation mechanism, its operate condition and actual power consumption amount of this moment and the time information of this moment with the load machine that is set in advance in the load machine in the described object unit carries out relatedly being kept in the described information form as resume
Described prediction and calculation mechanism has calculation mechanism, its on described information form, determine with the load machine that is set in advance in the corresponding resume position of the operate condition of current time, follow determined resume position and based in the time interval till from the current time to the finish time from the actual power consumption amount of the load machine consumption that information form was obtained, calculate described forecast consumption quantity of power; When on information form, not having corresponding resume position,,, calculate described forecast consumption quantity of power by linear prediction according to the deviation of the described actual power consumption amount of described actual power consumption amount that is obtained before recently and current time.
2, request control device according to claim 1 is characterized in that,
Described preservation mechanism only under the situation in the mean value of the actual power consumption amount of the described information form of described actual power consumption span is in the scope of regulation, is kept in the described information form this actual power consumption amount is related with described operate condition.
3, request control device according to claim 1 is characterized in that,
As described predefined load machine, comprising under the situation of show cabinet, in the processing that described preservation mechanism carries out, comprise the relation of the actual power consumption amount of the defrosting of described show cabinet and the state of cooling action, described show cabinet and the time information of this moment is kept at processing in the described information form as resume.
4, request control device according to claim 1 is characterized in that,
As described predefined load machine, comprising under the situation of air conditioner, in the processing that described preservation mechanism carries out, comprise the relation of the actual power consumption amount of cold-room, greenhouse, drying and the constant temperature of described air conditioner and the state of open and close action, described air conditioner and the time information of this moment is kept at processing in the described information form as resume.
5, request control device according to claim 3 is characterized in that,
Described preservation mechanism is kept at the storehouse temperature and the difference between the design temperature of the described show cabinet of this moment in the described information form.
6, a kind of consumed power Forecasting Methodology calculate to obtain before the finish time during the unit to it is characterized in that in the predicted forecast consumption quantity of power of object unit internal consumption,
Operate condition, the actual power consumption amount of this moment and the time information of this moment of the load machine that is set in advance in the load machine in the object unit be associated to be kept in the information form as resume,
On described information form, determine with the load machine that is set in advance in the corresponding resume position of the operate condition of current time, follow determined resume position and based in the time interval till from the current time to the finish time from the actual power consumption amount of the load machine consumption that information form was obtained, calculate described forecast consumption quantity of power; When on information form, not having corresponding resume position,,, calculate described forecast consumption quantity of power by linear prediction according to the deviation of the described actual power consumption amount of described actual power consumption amount that is obtained before recently and current time.
CN200610004225A 2005-06-01 2006-01-27 Demand control apparatus and power consumption prediction method Expired - Fee Related CN100594648C (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
JP2005161352 2005-06-01
JP2005161352 2005-06-01
JP2005197048 2005-07-06
JP2005197048A JP4346584B2 (en) 2004-08-12 2005-07-06 Demand control apparatus and program

Related Child Applications (1)

Application Number Title Priority Date Filing Date
CN2008100989776A Division CN101345420B (en) 2005-06-01 2006-01-27 Demand control unit, predicting method and program of consumed electric power

Publications (2)

Publication Number Publication Date
CN1874105A CN1874105A (en) 2006-12-06
CN100594648C true CN100594648C (en) 2010-03-17

Family

ID=37484432

Family Applications (2)

Application Number Title Priority Date Filing Date
CN2008100989776A Expired - Fee Related CN101345420B (en) 2005-06-01 2006-01-27 Demand control unit, predicting method and program of consumed electric power
CN200610004225A Expired - Fee Related CN100594648C (en) 2005-06-01 2006-01-27 Demand control apparatus and power consumption prediction method

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN2008100989776A Expired - Fee Related CN101345420B (en) 2005-06-01 2006-01-27 Demand control unit, predicting method and program of consumed electric power

Country Status (1)

Country Link
CN (2) CN101345420B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107949967A (en) * 2015-10-05 2018-04-20 宝马股份公司 The operation reserve of local storage determines

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4680287B2 (en) * 2008-09-17 2011-05-11 三菱電機株式会社 Air conditioner
JP4703736B2 (en) * 2009-03-02 2011-06-15 株式会社東芝 Energy management system and method
JP2011052952A (en) * 2009-08-07 2011-03-17 Sanyo Electric Co Ltd Operation management device for refrigerating machine
JP6047894B2 (en) * 2012-01-12 2016-12-21 オムロン株式会社 Simulation device, simulation method, program, and recording medium
JP6430106B2 (en) * 2013-08-26 2018-11-28 東芝ライフスタイル株式会社 Power consumption output device
CN110245771B (en) * 2018-03-09 2021-08-20 亿可能源科技（上海）有限公司 Demand prediction method, demand control method and system
CN110107998B (en) * 2019-05-28 2021-01-05 珠海格力电器股份有限公司 Energy-saving control method, equipment and medium for multi-connected cold and hot water unit

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TW342478B (en) * 1996-01-08 1998-10-11 Matsushita Electric Ind Co Ltd Time series signal predicting apparatus
US8243803B2 (en) * 2001-11-30 2012-08-14 Ntt Docomo, Inc. Moving picture coding apparatus, moving picture decoding apparatus, moving picture coding method, moving picture decoding method, program, and computer-readable recording medium containing the program
JP2004257276A (en) * 2003-02-25 2004-09-16 Osaka Gas Co Ltd Cogeneration system

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107949967A (en) * 2015-10-05 2018-04-20 宝马股份公司 The operation reserve of local storage determines
CN107949967B (en) * 2015-10-05 2021-09-03 宝马股份公司 Determination of operating policy of local storage

Also Published As

Publication number Publication date
CN101345420B (en) 2011-03-02
CN1874105A (en) 2006-12-06
CN101345420A (en) 2009-01-14

Similar Documents

Publication Publication Date Title
CN100594648C (en) Demand control apparatus and power consumption prediction method
EP1729223A2 (en) Demand control apparatus, electric power consumption prediction method, and program therefor
Huang et al. Simulation-based performance evaluation of model predictive control for building energy systems
Chen et al. MPC-based appliance scheduling for residential building energy management controller
US10371405B2 (en) Building power management systems
CN100572956C (en) Environmental control system
CN104969133B (en) Cloud enables building automation system
JP5691655B2 (en) Air conditioning system and its control device
JP2019522290A (en) Variable refrigerant flow system using multilevel model predictive control
EP1758225A2 (en) Apparatus, method and program for electric power amount control
Staino et al. Cooperative optimization of building energy systems in an economic model predictive control framework
CN109711614A (en) A kind of the dynamic optimization progress control method and system of distributed busbar protection
Soares et al. Using reinforcement learning for maximizing residential self-consumption–Results from a field test
TW202045870A (en) Store air-conditioning system
Gholizadeh et al. Benefit assessment of water-heater management on residential demand response: An event driven approach
Ra et al. Real-time model predictive cooling control for an HVAC system in a factory building
Dai et al. A reinforcement learning-enabled iterative learning control strategy of air-conditioning systems for building energy saving by shortening the morning start period
Lee et al. Smart-valve-assisted model-free predictive control system for chiller plants
JP6389599B2 (en) Operation plan creation device and operation plan creation method
US11519628B2 (en) System and method for optimizing energy use of a structure using a clustering-based rule-mining approach
JP2016023880A (en) Control apparatus of equipment and control method of equipment
CN114543303B (en) Operation optimization method and system for central air-conditioning refrigeration station based on operation big data
JP5763955B2 (en) Air conditioning load prediction device and air conditioning load prediction method
Máša et al. Using a utility system grey-box model as a support tool for progressive energy management and automation of buildings
Gayeski et al. Predictive pre-cooling of thermo-active building systems with low-lift chillers. Part I: control algorithm

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20100317

Termination date: 20140127