WO2015087470A1 - 需要予測装置、プログラム - Google Patents
需要予測装置、プログラム Download PDFInfo
- Publication number
- WO2015087470A1 WO2015087470A1 PCT/JP2014/004800 JP2014004800W WO2015087470A1 WO 2015087470 A1 WO2015087470 A1 WO 2015087470A1 JP 2014004800 W JP2014004800 W JP 2014004800W WO 2015087470 A1 WO2015087470 A1 WO 2015087470A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- prediction
- power
- model
- unit
- period
- 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.)
- Ceased
Links
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for AC mains or AC distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B15/00—Systems controlled by a computer
- G05B15/02—Systems controlled by a computer electric
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for AC mains or AC distribution networks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/26—Pc applications
- G05B2219/2639—Energy management, use maximum of cheap power, keep peak load low
-
- H02J2103/30—
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02B90/20—Smart grids as enabling technology in buildings sector
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S20/00—Management or operation of end-user stationary applications or the last stages of power distribution; Controlling, monitoring or operating thereof
Definitions
- the present invention generally relates to a demand prediction apparatus and a program, and more particularly relates to a demand prediction apparatus that calculates an amount of electric power for which demand is predicted in the future, and a program that causes a computer to function as the demand prediction apparatus.
- the technology described in Document 1 stores future energy consumption values corresponding to past weather conditions, and analyzes future user behavior patterns based on the past energy consumption values, thereby responding to future behavior patterns. It is a technology that predicts the predicted energy consumption.
- the technology described in Document 1 predicts a plurality of future behavior patterns of a user corresponding to future weather information based on the analyzed past behavior patterns, thereby providing energy corresponding to each of the plurality of behavior patterns.
- the consumption forecast value is obtained.
- As an action pattern in Document 1 a distinction between presence and absence for each time zone is shown.
- the technique described in Document 1 uses a distinction between presence and absence as an action pattern, and associates weather information with this action pattern. It is possible to obtain a predicted energy consumption value.
- An object of the present invention is to provide a demand forecasting device that makes it possible to accurately determine the amount of power for which demand is predicted in a future forecast period, and further provides a program that causes a computer to function as this demand forecasting device. The purpose is to do.
- the demand prediction device includes an acquisition unit that acquires a power value consumed for each branch circuit branched by a distribution board provided in a power consumer from the measurement device, and the acquisition unit acquires the power value
- a storage unit that stores power information in which a power value for each branch circuit is associated with a date and time, and an electrical load that uses power for each branch circuit based on the power information stored in the storage unit 1
- the extraction unit that extracts the usage period of the first time, the order in which the usage period occurred, and the amount of power consumed in each of the usage period
- the demand in the forecast period defined as the period after the determination time is characterized by comprising a prediction unit for calculating a result predicts the amount of power expected.
- the demand prediction apparatus includes an acquisition unit, a storage unit, an extraction unit, a model generation unit, and a prediction unit.
- the said acquisition part acquires the consumed electric power value from each measuring device about each of the branch circuit of N system (N is an integer greater than or equal to 2) branched by the distribution board provided in the electric power consumer.
- the storage unit stores power information in which the power value and date / time of each of the N-system branch circuits acquired by the acquisition unit are associated with each other.
- the extraction unit extracts a usage period in which the electric load uses power once for each of the N branch circuits based on the power information stored in the storage unit.
- the model generation unit is a combination of at least one branch circuit that is related to the order in which the use periods extracted by the extraction unit among the N system branch circuits are generated based on the power information of a predetermined period. Is a model group.
- the model generation unit generates a plurality of models so as to correspond one-to-one to a plurality of model groups.
- Each of the plurality of models includes, for at least one branch circuit belonging to the corresponding model group, the order in which the usage periods occur, the amount of power consumed in each of the usage periods, and each of the usage periods. Start time and duration.
- the model generation unit further obtains a transition probability to another model among the plurality of models for each of the plurality of generated models, and stores the plurality of models and the transition probability in a model storage unit. .
- the prediction unit collates the power information immediately before the determination time with the plurality of models using the specified determination time as a reference, and applies the power information immediately before the determination time from the plurality of models.
- a full first model is extracted.
- the prediction unit selects a second model based on the transition probability from the first model (for example, the transition probability is maximized) among the plurality of models. Select from.
- the prediction unit calculates, as a prediction result, an amount of power for which demand is predicted in a prediction period determined as a period after the determination time, based on the selected second model.
- the program according to the present invention causes a computer to function as the demand prediction apparatus described above.
- the demand prediction apparatus 10 described below includes an acquisition unit 11, a storage unit 12, an extraction unit 13, a model generation unit 14, and a prediction unit 15.
- the acquisition unit 11 acquires, from the measurement device 23, the power value consumed for each branch circuit 22 branched by a distribution board 21 provided in a power consumer 20.
- the storage unit 12 stores power information in which the power value for each branch circuit 22 acquired by the acquisition unit 11 is associated with the date and time.
- the extraction unit 13 extracts one usage period in which the electrical load 24 uses power for each branch circuit 22 based on the power information stored in the storage unit 12. Based on the power information for a predetermined period stored in the storage unit 12, the model generation unit 14 determines the order in which the usage periods occur, the amount of power consumed in each of the usage periods, the start time and the continuation of the usage period. A model including time is generated. Further, the model generation unit 14 obtains a transition probability to another model for each generated model.
- the prediction unit 15 performs a prediction period T1 that is determined as a period after the determination time tx based on the past model and the transition probability with respect to the determination time tx at the specified determination time tx.
- the amount of power for which demand is predicted is calculated as a prediction result.
- the demand prediction device 10 determines the start time and duration of the order of use of power for each branch circuit 22, the amount of power consumed in one use of the electrical load 24, and the usage period of the electrical load 24.
- transition probabilities are obtained among a plurality of models.
- the model before the determination time tx is estimated based on the order of use of power for each branch circuit 22 in the past at the determination time tx, and the model generated in the prediction period T1 is determined as the transition probability based on the estimated model. Estimated based on.
- the prediction unit 15 generates a plurality of prediction results and obtains a prediction probability for each prediction result.
- the demand prediction device 10 includes an output unit 16 that presents a prediction result by the prediction unit 15 to the presentation device 30.
- the demand prediction apparatus 10 further preferably includes an input unit 17 and a correction unit 18.
- the input unit 17 receives the actual amount of power consumed in the prediction period T1.
- the correction unit 18 corrects the transition probability between models in the model generation unit 14 using an error between the power amount of the prediction result predicted by the prediction unit 15 and the actual power amount input to the input unit 17.
- the correction unit 18 corrects the power amount for each model in the model generation unit 14 using an error between the power amount of the prediction result predicted by the prediction unit 15 and the actual power amount input to the input unit 17. It may be a configuration.
- the model generation unit 14 generates a model including consumption information regarding consumption of at least one of the gas and tap water of the consumer 20.
- consumption of the gas (or tap water) of the consumer 20 means consumption of gas (or tap water) in the equipment etc. with which the consumer 20 is equipped.
- the model generation unit 14 generates a model including the external environment of the customer 20, and the transition probability is set according to the external environment.
- the prediction unit 15 takes into account the external environment in the prediction period T1, and the prediction result It is desirable to calculate
- the model generation unit 14 finds at least a relationship in the order in which the use periods occur among the branch circuits 22 of the plurality of systems.
- a combination of one branch circuit 22 is desirably used as a model group.
- the model generation unit 14 determines the order in which the usage periods occur, the amount of power consumed in each usage period, the start time of each usage period, and A model including the duration is generated.
- the demand prediction apparatus 10 includes a computer that realizes the above-described functions by executing a program as a main hardware configuration. That is, this program is a program that causes a computer to function as the demand prediction apparatus 10.
- This type of computer includes portable terminals such as smartphones and tablet terminals in addition to personal computers.
- the computer may have a configuration in which a processor and a memory are integrally provided, such as a microcomputer.
- the program may be written in a ROM (Read Only Memory) in advance or provided through an electric communication line such as the Internet.
- the program may be provided by a computer-readable recording medium.
- the consumer 20 includes a distribution board 21 that receives commercial power supplied by an electric power company.
- the distribution board 21 branches the received power to a plurality of branch circuits 22 and distributes the power to the electrical load 24 used by the customer 20.
- the measuring device 23 measures the power value consumed for each branch circuit 22. That is, the measuring device 23 measures the consumed power value for each of the branch circuits 22 of the plurality of systems.
- the measuring device 23 employs either a configuration built in the distribution board 21 or a configuration arranged outside the distribution board 21.
- the branch circuit 22 and the electrical load 24 correspond one-to-one or one-to-many. That is, the branch circuit 22 may correspond to the electrical load 24 on a one-to-one basis for an electrical load 24 that consumes a relatively large amount of power, such as an air conditioner, an IH cooking heater (IH), and a microwave oven.
- the branch circuit 22 is often assigned in units (places) in the customer 20 as a unit.
- the measuring device 23 monitors the passing current for each branch circuit 22 with a Rogowski coil or a clamp-type current sensor, and uses the integrated value of the product of the monitored current value and the voltage value between the lines of the branch circuit 22 as a power value. calculate. That is, the power value measured by the measuring device 23 is not actually instantaneous power, but is selected in a predetermined unit time (for example, in the range of about 30 seconds to 10 minutes, and preferably 30 seconds or 1 minute). ). In general, the instantaneous power for each branch circuit 22 fluctuates with time even within a unit time, but in this embodiment, the fluctuation of the instantaneous power within the unit time is not taken into account, and the integrated power in the unit time is taken into account. Use quantity as power value. This power value can be regarded as equivalent to the average power value (instantaneous power) in unit time.
- the consumer 20 generally means the customer who receives electric power from an electric power company, the consumer 20 shall mean the space which the said customer occupies here. That is, the consumer 20 includes a detached house, a dwelling unit housing, a tenant building tenant, and the like. However, these are merely examples and are not intended to limit the customer 20.
- the demand prediction device 10 includes an acquisition unit 11 that acquires a power value for each branch circuit 22 measured by the measurement device 23.
- the power value acquired by the acquisition unit 11 from the measurement device 23 is associated with the date and time and stored in the storage unit 12 as power information.
- the date and time is measured by a built-in clock 19 such as a real-time clock built in the demand prediction apparatus 10. That is, the power information includes the power value for each unit time and the date and time when the power value was acquired.
- the storage unit 12 stores power information for each of a plurality of branch circuits 22 branched by the distribution board 21. That is, the storage unit 12 stores a history of power value transitions for each of the multiple systems of branch circuits 22.
- the storage unit 12 has a capacity that can store power information for a period selected from one week, one month, six months, one year, two years, and the like. The longer the period for storing power information in the storage unit 12, the greater the amount of information, and it is considered that a model with high prediction accuracy can be generated. However, if long-term power information was collected, the start of use was delayed, In addition, the model may change due to changes in user lifestyle. Therefore, it is desirable to generate a model using power information for a relatively short period and modify the model as necessary.
- This embodiment is based on the expectation that regularity may be found in the order in which the user uses the electric load 24 in the consumer 20.
- This order includes the case where the electric load 24 is used independently from the other electric loads 24. That is, not only the case where a plurality of electric loads 24 are used in order but also the case where only one electric load 24 is used is included.
- the usage state of each electric load 24 is estimated using the change in the power value for each of the branch circuits 22 of the plurality of systems, and further, the use of the plurality of electric loads 24 is used.
- the power consumption is estimated by estimating the relationship with respect to the state. In order to find the relationship between the usage states of the plurality of electrical loads 24, it is first necessary to extract one usage period in which each of the plurality of electrical loads 24 uses power.
- the extraction unit 13 uses the power information stored in the storage unit 12 to obtain a standby power value for each of the plurality of branch circuits 22 and calculates a power value for the standby power value for each of the plurality of branch circuits 22. Based on the change, the period of operation and non-operation of the electrical load 24 is extracted.
- a change in the power value in a predetermined period such as one day is used.
- the power value history the power value stored in the storage unit 12 is used.
- the comparison value is changed so as to be gradually decreased and compared with the power value, and the comparison value satisfying the condition that the duration of the state in which the power value is smaller than the comparison value is equal to or longer than a predetermined reference time.
- the minimum value may be used as the standby power value.
- the reference time to be compared with the duration is determined according to the frequency and time zone in which the electric load 24 is used. That is, the reference time is set according to the time during which the target electric load 24 is in a non-operating state.
- the model generation unit 14 is in the order (order) in which the power usage periods of the branch circuits 22 in the plurality of systems have occurred. Evaluate the relationship.
- the model generation unit 14 uses, as a model group, a combination of at least one branch circuit 22 in which the relationship is found in the order in which the usage periods occur among the branch circuits 22 of the plurality of systems. Further, the model generation unit 14 includes, for each model group, a model including, in attributes, the order (order) in which the usage periods have occurred, the amount of power consumed in each usage period, and the start time and duration of each usage period. Is generated.
- the model generation unit 14 determines the order in which the usage periods occur and the amount of power consumed in each usage period for at least one branch circuit 22 in which the relationship is found in the order in which the usage periods occur. Then, a model including the start time and duration of each use period is generated.
- the order in which the usage periods occur means the order in which the usage periods start, that is, the order in which the usage periods start in the at least one branch circuit 22 belonging to the model group.
- the occurrence frequency for each time zone is obtained for each power usage period of the branch circuits 22 of the multiple systems, and when the occurrence frequency is equal to or higher than a reference value, the correlation of the power usage period is calculated. What is necessary is just to evaluate strength.
- FIG. 2 shows power information from around 6:00 to 15:00 on a holiday at home during the day.
- C1 represents the living room
- C2 represents the kitchen
- C3 represents the air conditioner
- C4 represents the power value of the branch circuit 22 corresponding to the washing machine.
- the electric load 24 or the room name corresponding to each of the branch circuits 22 of a plurality of systems is used.
- the branch circuit 22 of the washing machine forms a model group independent of the other three branch circuits 22.
- whether or not to form a model group is determined based on the occurrence frequency of the power usage period and the evaluation of the correlation coefficient between the plurality of branch circuits 22.
- three branch circuits 22 of a living room, a kitchen, and an air conditioner form a model group.
- the model generation unit 14 determines the order (order) in which the usage periods occur, the amount of power consumed in each usage period, and the amount of power consumed in each usage period for the branch circuits 22 belonging to the model group.
- a model including the start time and the duration as attributes is generated. In the case of the case of FIG. 2, for example, a model as shown in Table 1 is generated.
- ⁇ 11 to ⁇ 13, ⁇ 21 to ⁇ 23, and ⁇ 31 to ⁇ 33 each represent an allowable error. That is, the allowable errors ⁇ 11 to ⁇ 13 represent the degree of variation regarding the start time, the allowable errors ⁇ 21 to ⁇ 23 represent the degree of variation regarding the duration, and the allowable errors ⁇ 31 to ⁇ 33 represent the degree of variation regarding the electric energy.
- the power information from around 6:00 to around 9:00 will be as shown in FIG. 3, for example, and a model as shown in Table 2 is generated.
- Table 2 the allowable errors ⁇ 11 to ⁇ 13, ⁇ 21 to ⁇ 23, and ⁇ 31 to ⁇ 33 have the same values as holidays, but the allowable errors actually vary depending on the power information used for generating the model.
- model generation unit 14 extracts model groups for other time zones, and generates a model for each model group.
- Tables 3 and 4 are models generated for the branch circuit 22 corresponding to the washing machine, where Table 3 corresponds to holidays and Table 4 corresponds to weekdays.
- Table 4 shows a model in which power is not used. This model confirms that no power is used in the branch circuit 22 corresponding to the washing machine from 13:00 to 18:00. Represents.
- the model generation unit 14 obtains a transition probability between a plurality of models.
- the transition probability here refers to the probability of transition from the model of interest to another model of the plurality of models when attention is paid to one of the models, that is, the model of the model of interest. Next, it means the probability that another model will come.
- the transition probability indicates the probability that the transition of the power value that occurs next applies to the model other than the model of interest among the multiple models. This is the value obtained for each model. For example, the transition probability is high between a plurality of models on holidays, and the transition probability is low between a model on holidays and a model on weekdays.
- transition probabilities are determined as shown in Table 5, for example.
- Table 5 shows the respective transition probabilities from the model shown in the left column to the model shown in the upper column.
- the model and the transition probability generated by the model generation unit 14 as described above are stored in the model storage unit 41.
- the model and the transition probability generated each time the model generation unit 14 generates a new model is stored in the model storage unit 41, and the data (model and transition probability) stored in the model storage unit 41 is updated as needed.
- the Further, the model and the transition probability stored in the model storage unit 41 are appropriately corrected as will be described later.
- the model and the transition probability are stored in the model storage unit 41, it is possible to calculate the amount of power for which demand is predicted in the prediction period T1 after the specified determination time tx as follows.
- the prediction unit 15 determines to which model the transition of the past power value applies from the determination time tx. For example, as shown in FIG. 4, if the determination time tx is past 9:00 and the power information before the determination time tx is known, the power information before the determination time tx is checked against the model, It is determined whether it fits the model. In the case of the illustrated example, it can be seen that this applies to model 1 out of models 1 to 4 described above. In addition, when it applies to both the model 1 and the model 2, the similarity is evaluated and the one that is similar is selected. For the evaluation of the degree of similarity, a distance related to the start time (such as the square root of the sum of squared differences) may be used.
- the prediction unit 15 obtains a transition probability of the model, and an event corresponding to the model having the maximum transition probability occurs after the determination time tx. Predict. That is, in the example shown in FIG. 4, it is determined from the power information up to the determination time tx that the model 1 is applicable, and models 3 and 4 are extracted as models representing events that occur after the model 1. In this case, model 3 is selected because model 3 has a higher transition probability than model 4 with respect to model 1.
- the prediction unit 15 determines that the model applicable to the power information immediately before the determination time tx is the model 1, the prediction unit 15 predicts that the probability of the model 3 occurring after the determination time tx is high.
- the amount of power for which demand is predicted in the prediction period T1 (for example, the period until the end time of the day) determined as the period after the determination time tx includes the power amount of the model 3.
- the prediction period T1 may be set by the end time of a period when power saving is requested when there is a power saving request (so-called demand response) from an electric power supplier that supplies power. Further, the prediction period T1 may be appropriately set by the user regardless of the power saving request. When a plurality of models occurring in the prediction period T1 are predicted, the amount of power whose demand is predicted in the prediction period T1 after the determination time tx is added up for the plurality of predicted models.
- the power information immediately before the determination time tx is applied to the model 1.
- the model 1 since the model 1 has not ended yet at the determination time tx, the power amount of the remaining time of the model can be added up. desirable.
- the model does not end at the determination time tx.
- the prediction unit 15 obtains only one prediction result (the amount of power whose demand is predicted in the prediction period T1).
- the prediction unit 15 may generate a plurality of prediction results and obtain a prediction probability for each prediction result.
- the prediction unit 15 predicts the model 1 estimated from the power information immediately before the determination time tx and includes the power amount of the model 3.
- a result and a prediction result including the amount of power of the model 4 are generated.
- the prediction probability that the model 1 to the model 3 is generated is 80% as a transition probability
- the prediction probability that the model 1 to the model 4 is generated is 20% that is a transition probability.
- the prediction probability of the prediction result including the power amount of the model 3 is 80%
- the prediction probability of the prediction result including the power amount of the model 4 is 20%.
- the prediction unit 15 obtains a prediction probability for each combination using each transition probability.
- the prediction unit 15 uses each prediction probability as a weighting coefficient.
- a weighted average is obtained by weighting the model power.
- the prediction unit 15 desirably uses this weighted average for the amount of power for which demand is predicted.
- the total amount of power for each model may be the amount of power for which demand is predicted in the prediction period T1 with a combination of models that maximizes the prediction probability.
- the electric energy in the prediction period T1 predicted by the prediction unit 15 as described above is presented to the presentation device 30 through the output unit 16. Therefore, the user can check the content presented on the presentation device 30.
- the prediction unit 15 obtains the amount of power for which demand is predicted in the prediction period T1, it is possible to configure a device that provides support or advice regarding the use of power in the consumer 20 by using this function. .
- the transition probability between models is automatically set based on the power information stored in the storage unit 12, but the transition probability may change depending on various conditions. Therefore, it is desirable that the demand prediction apparatus 10 includes an input unit 17 to which a condition for determining the transition probability is input and a correction unit 18 that corrects the transition probability.
- the actual amount of power consumed in the prediction period T1 is input. That is, by giving the actual power amount from the input unit 17 after the end of the prediction period T1, it is possible to compare with the power amount for which the demand is predicted before the start of the prediction period T1 at the determination time tx.
- the correction unit 18 does not correct the transition probability if the error of the predicted power amount (the power amount of the prediction result) with respect to the actual power amount is assumed, and exceeds the range where the error is assumed. If so, the transition probability is corrected.
- the correction unit 18 corrects the transition probability using a plurality of prediction results, rather than determining whether or not to correct the transition probability using a single prediction result.
- the predicted power amount can be adjusted to the actual power amount by correcting the transition probability based on a plurality of prediction results.
- the correction part 18 may be comprised so that the electric energy for every model may be corrected instead of correcting a transition probability.
- the operation device that inputs conditions to the input unit 17 and the display device that becomes the presentation device 30 can use an operation display device that is configured exclusively for the demand prediction device 10.
- the operation display unit employs a configuration in which a display unit that is a flat panel display such as a liquid crystal display unit and an operation unit such as a touch panel or a push button switch are integrally provided.
- the interface part (not shown) for communicating with a terminal device may be provided in the demand prediction apparatus 10, and a terminal device may be used as an operating device and a display.
- a personal computer, a smartphone, a tablet terminal, or the like is used as this type of terminal device.
- the model generation unit 14 generates a model based only on the power information, but generates a model including consumption information regarding consumption of at least one of the gas and tap water of the consumer 20. Also good.
- the model includes at least one of the consumption of gas and the consumption of tap water in addition to the amount of electric power.
- the model generation unit 14 may generate a model including the external environment of the customer 20.
- the transition probability is set according to the external environment in the transition model. That is, the model generation unit 14 obtains a transition probability according to the external environment.
- weather unsunny weather, cloudy weather, rainy weather, etc.
- outdoor temperature outdoor humidity, and the like are used.
- These external environments are preferably acquired through a telecommunication line such as the Internet. That is, it is desirable that the demand prediction apparatus 10 includes a communication unit (not shown) that enables communication through an electric communication line.
- the prediction unit 15 calculates a prediction result in consideration of the external environment in the prediction period T1. For example, it is assumed that the prediction result determined by the prediction unit 15 using the transition probability predicts the use of the washing machine in the prediction period T1. At this time, the prediction unit 15 may use the electric energy for washing only as a prediction result if the weather, which is the external environment, is fine, and may use the electric energy obtained by summing washing and drying as the prediction result in the case of rainy weather. In addition, the prediction unit 15 may obtain the prediction result by adjusting the amount of power predicted to be consumed by the air conditioner according to the outside air temperature that is the external environment. If the external environment is taken into consideration in this way, the prediction accuracy of the electric energy that the prediction unit 15 predicts as the demand in the prediction period T1 can be improved.
- the demand prediction device 10 can be installed for each customer 20.
- the power value from the measuring device 23 may be acquired from the HEMS controller through an electric communication line such as the Internet.
- the demand prediction device 10 is provided in a server that communicates through an electric communication line.
- the server collects the power values consumed for each branch circuit from the plurality of consumers 20 and generates a model for each consumer 20.
- the server includes the demand prediction device 10 as described above, when similarities are found in models between different consumers 20, the customers 20 having high similarity may be grouped together. In this case, it is possible to apply the same model to a plurality of consumers 20 grouped together.
- the purpose is to obtain the amount of power for which demand is expected in the prediction period T1 after the determination time tx.
- the demand prediction apparatus 10 of this embodiment it is also possible to utilize the demand prediction apparatus 10 of this embodiment as follows.
- the power storage device stores the power generated by the solar power generation device before the start of the power saving request target period. It is possible to advise you to do so.
- the washing machine when focusing on the washing machine, if the target period of the power saving request is included in the time zone where the washing machine is scheduled to be used, the washing machine should be used before or after the target period. Advice can be given. Similarly, it is possible to give advice to change the time zone in which the washing machine is used when rainy weather is predicted in the time zone in which the washing machine is scheduled to be used. In addition, when the use of the washing machine is predicted, if the washing machine is not used, a configuration may be added in which it is determined that there is a possibility that the user 20 has an abnormality in the consumer 20.
- the demand prediction device 10 can determine the amount of power for which demand is predicted for the appropriately set prediction period T1, and thus various applications are possible using the predicted amount of power.
- the above-described embodiment is an example of the present invention.
- the present invention is not limited to the above-described embodiment, and various modifications can be made according to design and the like as long as the technical idea according to the present invention is not deviated from this embodiment. Of course, it can be changed.
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Power Engineering (AREA)
- Water Supply & Treatment (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- General Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
Description
表1において、σ11~σ13、σ21~σ23、σ31~σ33は、それぞれ許容誤差を表す。つまり、許容誤差σ11~σ13は開始時刻に関するばらつきの程度を表し、許容誤差σ21~σ23は継続時間に関するばらつきの程度を表し、許容誤差σ31~σ33は電力量に関するばらつきの程度を表している。
モデル生成部14は、同様にして他の時間帯についてもモデル群を抽出し、モデル群ごとにモデルを生成する。表3、表4は洗濯機に対応した分岐回路22について生成したモデルであって、表3は休日に対応し、表4は平日に対応している。
以上のようにしてモデル生成部14が生成したモデルと遷移確率とは、モデル記憶部41に保存される。モデル生成部14が新たなモデルを生成する度に生成されたモデルと遷移確率とはモデル記憶部41に保存され、モデル記憶部41に保存されているデータ(モデルおよび遷移確率)は随時更新される。また、モデル記憶部41に保存されたモデルおよび遷移確率は、後述するように適宜に修正される。モデル記憶部41にモデルおよび遷移確率が保存されていると、以下のようにして、指定された判定時刻tx後の予測期間T1において需要が予測される電力量を算出することが可能になる。
Claims (9)
- 電力の需要家に設けられた分電盤で分岐された分岐回路ごとに消費された電力値を計測装置から取得する取得部と、
前記取得部が取得した前記分岐回路ごとの電力値と日時とを対応付けた電力情報を記憶する記憶部と、
前記記憶部が記憶している前記電力情報に基づいて、前記分岐回路ごとに電気負荷が電力を使用する1回の使用期間を抽出する抽出部と、
前記記憶部が記憶している所定期間の前記電力情報に基づいて、前記使用期間が生じた順番と、前記使用期間のそれぞれで消費された電力量と、前記使用期間の開始時刻および継続時間とを含むモデルを生成し、さらに生成した前記モデルごとに他のモデルへの遷移確率を求めるモデル生成部と、
指定された判定時刻において、当該判定時刻に対する過去の前記モデルと前記遷移確率とに基づいて、当該判定時刻後の期間として定められる予測期間において需要が予測される電力量を予測結果として算出する予測部とを備える
ことを特徴とする需要予測装置。 - 前記予測部は、前記予測結果を複数生成し、かつ前記予測結果ごとに予測確率を求める
請求項1記載の需要予測装置。 - 前記予測部による前記予測結果を提示装置に提示する出力部をさらに備える
請求項1又は2記載の需要予測装置。 - 前記予測期間において消費された実際の電力量が入力される入力部と、
前記予測部が予測した前記予測結果の電力量と前記入力部に入力された実際の電力量との誤差を用いて、前記モデル生成部における前記モデル間の遷移確率を修正する修正部とをさらに備える
請求項1~3のいずれか1項に記載の需要予測装置。 - 前記予測期間において消費された実際の電力量が入力される入力部と、
前記予測部が予測した前記予測結果の電力量と前記入力部に入力された実際の電力量との誤差を用いて、前記モデル生成部における前記モデルごとの電力量を修正する修正部とをさらに備える
請求項1~3のいずれか1項に記載の需要予測装置。 - 前記モデル生成部は、
前記需要家のガスと水道水との少なくとも一方の消費に関する消費情報を含むモデルを生成する
請求項1~5のいずれか1項に記載の需要予測装置。 - 前記モデル生成部は、
前記需要家の外部環境を含むモデルを生成し、かつ前記遷移確率は前記外部環境に応じて設定され、
前記予測部は、
前記予測期間における前記外部環境を考慮して前記予測結果を算出する
請求項1~6のいずれか1項に記載の需要予測装置。 - 前記モデル生成部は、
前記抽出部が複数系統の前記分岐回路の各々について前記使用期間を抽出すると、複数系統の前記分岐回路のうち使用期間の生じた順番に関係性が見出された少なくとも1系統の前記分岐回路の組み合わせをモデル群とし、
前記モデル群に属する少なくとも1系統の前記分岐回路について、前記使用期間の生じた順番と、各々の前記使用期間で消費された電力量と、各々の前記使用期間の開始時刻および継続時間とを含むモデルを生成する
請求項1~7のいずれか1項に記載の需要予測装置。 - コンピュータを、請求項1~8のいずれか1項に記載の需要予測装置として機能させるプログラム。
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US15/100,150 US20170017215A1 (en) | 2013-12-10 | 2014-09-18 | Demand prediction system and program |
| JP2015552290A JPWO2015087470A1 (ja) | 2013-12-10 | 2014-09-18 | 需要予測装置、プログラム |
| EP14870498.4A EP3082096A4 (en) | 2013-12-10 | 2014-09-18 | DEMAND PRESENCE DEVICE AND PROGRAM |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2013255395 | 2013-12-10 | ||
| JP2013-255395 | 2013-12-10 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2015087470A1 true WO2015087470A1 (ja) | 2015-06-18 |
Family
ID=53370808
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2014/004800 Ceased WO2015087470A1 (ja) | 2013-12-10 | 2014-09-18 | 需要予測装置、プログラム |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US20170017215A1 (ja) |
| EP (1) | EP3082096A4 (ja) |
| JP (1) | JPWO2015087470A1 (ja) |
| WO (1) | WO2015087470A1 (ja) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105976072A (zh) * | 2016-05-31 | 2016-09-28 | 中国地质科学院矿产资源研究所 | 基于“s”形模型的电力需求预测方法 |
| CN106126769A (zh) * | 2016-05-31 | 2016-11-16 | 中国地质科学院矿产资源研究所 | 基于s形模型的铅需求预测方法 |
| CN108572631A (zh) * | 2018-03-08 | 2018-09-25 | 华南理工大学 | 一种基于二型模糊认知图的智能控制系统及方法 |
| EP3467984A4 (en) * | 2016-05-30 | 2019-11-13 | Hitachi, Ltd. | SYSTEM MONITOR AND PROGRAM |
| JP2021036752A (ja) * | 2019-08-30 | 2021-03-04 | 株式会社日立製作所 | 分散リソースの調整力管理装置および調整力管理方法 |
| JP2022029292A (ja) * | 2020-08-04 | 2022-02-17 | 東京瓦斯株式会社 | ガスの需要予測方法、システム、プログラム、記録媒体およびサーバー |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017093193A (ja) * | 2015-11-12 | 2017-05-25 | 富士通株式会社 | 電力調達調整プログラム、電力調達調整装置および電力調達調整方法 |
| WO2017183337A1 (ja) * | 2016-04-22 | 2017-10-26 | 住友電気工業株式会社 | デマンド監視装置、デマンド監視システム、デマンド監視方法およびデマンド監視プログラム |
| JP6735219B2 (ja) * | 2016-12-05 | 2020-08-05 | 株式会社日立製作所 | 予測システム及び予測方法 |
| US10819116B2 (en) * | 2017-02-28 | 2020-10-27 | International Business Machines Corporation | Forecasting solar power generation using real-time power data |
| US10447077B2 (en) | 2017-04-26 | 2019-10-15 | Kohler Co. | Predictive generator events |
| CN110245771B (zh) * | 2018-03-09 | 2021-08-20 | 亿可能源科技(上海)有限公司 | 需量预测方法、需量控制方法及系统 |
| AU2022284507B2 (en) * | 2021-06-02 | 2025-01-16 | Ihi Corporation | Energy consumption prediction device, energy consumption prediction method, and energy consumption prediction program |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011232903A (ja) | 2010-04-27 | 2011-11-17 | Sekisui Chem Co Ltd | 予測表示サーバ及び予測表示システム |
| WO2014126131A1 (ja) * | 2013-02-13 | 2014-08-21 | 株式会社構造計画研究所 | 生活行動推定システム、生活行動推定装置、生活行動推定プログラム、及び記録媒体 |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4181950A (en) * | 1977-09-30 | 1980-01-01 | Westinghouse Electric Corp. | Adaptive priority determination power demand control method |
| JP5618501B2 (ja) * | 2009-07-14 | 2014-11-05 | 株式会社東芝 | 需要予測装置、プログラムおよび記録媒体 |
| US9251479B2 (en) * | 2010-07-02 | 2016-02-02 | General Electric Technology Gmbh | Multi-interval dispatch method for enabling dispatchers in power grid control centers to manage changes |
| US8930037B2 (en) * | 2010-10-01 | 2015-01-06 | General Electric Company | Energy manager with minimum use energy profile |
| JP6277537B2 (ja) * | 2012-04-09 | 2018-02-14 | パナソニックIpマネジメント株式会社 | エネルギー管理装置、エネルギー管理システム、エネルギー管理方法、およびプログラム |
| SG11201500172TA (en) * | 2012-07-13 | 2015-07-30 | Nitto Denko Corp | On-demand multiple power source management system, on-demand multiple power source management system program and computer-readable recording medium on which said program is recorded |
-
2014
- 2014-09-18 EP EP14870498.4A patent/EP3082096A4/en not_active Withdrawn
- 2014-09-18 US US15/100,150 patent/US20170017215A1/en not_active Abandoned
- 2014-09-18 JP JP2015552290A patent/JPWO2015087470A1/ja not_active Withdrawn
- 2014-09-18 WO PCT/JP2014/004800 patent/WO2015087470A1/ja not_active Ceased
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011232903A (ja) | 2010-04-27 | 2011-11-17 | Sekisui Chem Co Ltd | 予測表示サーバ及び予測表示システム |
| WO2014126131A1 (ja) * | 2013-02-13 | 2014-08-21 | 株式会社構造計画研究所 | 生活行動推定システム、生活行動推定装置、生活行動推定プログラム、及び記録媒体 |
Non-Patent Citations (1)
| Title |
|---|
| See also references of EP3082096A4 |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3467984A4 (en) * | 2016-05-30 | 2019-11-13 | Hitachi, Ltd. | SYSTEM MONITOR AND PROGRAM |
| CN105976072A (zh) * | 2016-05-31 | 2016-09-28 | 中国地质科学院矿产资源研究所 | 基于“s”形模型的电力需求预测方法 |
| CN106126769A (zh) * | 2016-05-31 | 2016-11-16 | 中国地质科学院矿产资源研究所 | 基于s形模型的铅需求预测方法 |
| CN108572631A (zh) * | 2018-03-08 | 2018-09-25 | 华南理工大学 | 一种基于二型模糊认知图的智能控制系统及方法 |
| JP2021036752A (ja) * | 2019-08-30 | 2021-03-04 | 株式会社日立製作所 | 分散リソースの調整力管理装置および調整力管理方法 |
| JP7184716B2 (ja) | 2019-08-30 | 2022-12-06 | 株式会社日立製作所 | 分散リソースの調整力管理装置および調整力管理方法 |
| JP2022029292A (ja) * | 2020-08-04 | 2022-02-17 | 東京瓦斯株式会社 | ガスの需要予測方法、システム、プログラム、記録媒体およびサーバー |
| JP7461244B2 (ja) | 2020-08-04 | 2024-04-03 | 東京瓦斯株式会社 | ガスの需要予測方法、システム、プログラム、記録媒体およびサーバー |
Also Published As
| Publication number | Publication date |
|---|---|
| EP3082096A1 (en) | 2016-10-19 |
| EP3082096A4 (en) | 2016-10-26 |
| JPWO2015087470A1 (ja) | 2017-03-16 |
| US20170017215A1 (en) | 2017-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2015087470A1 (ja) | 需要予測装置、プログラム | |
| JP4648482B1 (ja) | 分析装置及び計測管理システム | |
| JP5929575B2 (ja) | 消費電力管理装置及び消費電力管理システム | |
| JP6610059B2 (ja) | 電力需要誘導装置 | |
| US10380706B2 (en) | Equipment management apparatus | |
| US11329602B2 (en) | Device and method for simulating power consumption by power load | |
| EP3958195B1 (en) | Energy prediction system, energy prediction method, program, recording medium, and management system | |
| CN110312637B (zh) | 改进的充电装置,尤其适用于电动汽车 | |
| EP2730003B1 (en) | Distributed energy grid management | |
| Huang et al. | Real-time data and energy management in microgrids | |
| JP2017169289A (ja) | 電力予測システム、電力予測方法及びプログラム | |
| JP2016077051A (ja) | デマンドレスポンス量配分システム | |
| KR101866564B1 (ko) | 원격검침 로드 프로파일의 부하기기별 전력 사용 패턴 분리 장치 및 방법 | |
| JP6264695B2 (ja) | 電力管理装置、及び情報提示方法 | |
| JPWO2018042521A1 (ja) | 給湯制御システム、サーバ、給湯制御方法及びプログラム | |
| JP6437139B2 (ja) | 電力管理装置、サーバ、電力管理システム、電力管理方法、及び、プログラム | |
| JPWO2017104237A1 (ja) | 情報処理装置、その情報処理方法、およびプログラム | |
| JP6292515B2 (ja) | 需要予測装置、プログラム | |
| JP2016127622A (ja) | 電力需要予測システム | |
| US10601225B2 (en) | Management device, management system, control method for management device, and control program | |
| JP5984095B2 (ja) | 行動管理装置、行動管理方法およびプログラム | |
| JP6641235B2 (ja) | 負荷推定装置及び負荷推定方法 | |
| JP2016218681A (ja) | インセンティブ予測量提示装置 | |
| AU2023288807B2 (en) | Hot water supply system and hot water supply method | |
| EP4546247A1 (en) | Water heater system and water heater method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 14870498 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 2015552290 Country of ref document: JP Kind code of ref document: A |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 15100150 Country of ref document: US |
|
| REEP | Request for entry into the european phase |
Ref document number: 2014870498 Country of ref document: EP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2014870498 Country of ref document: EP |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |