Demand and power supply assumption system

Abstract

To provide a demand and power supply assumption system capable of accurately estimating future demand and power supply transitions in response to changes or the like in geographical elements in land.SOLUTION: A demand and power supply assumption system 1 includes: a first model generation unit 5 that generates a first model by using an image of a power station's supply area as first input data and using feature elements of features included in the image as first output data; a second model generation unit 6 that generates a second model using, as second input data, feature elements of an image at a first point in time, power performance at the first point in time, and an elapsed time from the first point in time to a second point in time, and using, as second output data, feature elements of the image at the second point in time, and potential at the second point in time; and an estimation unit 7 that estimates feature elements at a future point in time and potential at a future point in time from a latest image, latest performance, and a predetermined elapsed time from the present to the predetermined future point in time on the basis of the first and second models.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a demand and power supply forecasting system that forecasts future demand for electricity supplied by a power station and future power sources that will supply electricity received by the power station, and in particular to a demand and power supply forecasting system that takes into account the geographical factors of the supply area to which the power station supplies electricity. [Background technology]

[0002] Traditionally, future electricity demand is estimated by taking into account the most recent demand record of electric power stations (power plants and substations), the results of surveys of consumers who have contracted for more than a certain amount of electricity, as well as information on business locations and urban development. Furthermore, power sources are estimated based on the most recent applications and grid connection records of generators with renewable energy power generation facilities. Thus, future demand and power sources are assumed primarily based on recent trends. However, in addition to these trends, factors that affect the assumptions of demand and power sources include geographical elements such as land use and the area of ​​land where power generation facilities are installed, as well as the business forms that are characteristic of each consumer, such as weather, business days, factory production status, etc. Therefore, if demand and power sources are assumed mainly based on recent trends, the accuracy of the results of the assumptions may not be good. Therefore, in recent years, techniques have been developed to improve the accuracy of the predicted results, and inventions relating to these techniques have already been disclosed.

[0003] Patent Document 1, entitled "Power Demand Forecasting System and Power Demand Forecasting Method," discloses an invention relating to a power demand forecasting system that improves the accuracy of power demand forecasts. The following describes the invention disclosed in Patent Document 1. The invention disclosed in Patent Document 1 is characterized by comprising a data storage unit that stores, for each of a plurality of customer groups of an electricity seller, training data including actual data on fluctuation factors of electricity demand and actual data on electricity demand in a database, a model generation unit that generates, for each of the plurality of customer groups, a demand forecast model that represents the relationship between the fluctuation factor data and electricity demand data based on the plurality of training data stored in the database, an input information acquisition unit that acquires predicted data of fluctuation factors for each of the plurality of customer groups, an individual demand forecast unit that calculates predicted data of electricity demand for each of the plurality of customer groups based on the predicted data of fluctuation factors and the demand forecast model, and a total demand forecast unit that calculates predicted data of total electricity demand for the electricity seller based on the predicted data of electricity demand calculated for each of the plurality of customer groups. According to the invention configured as above, it is possible to group customers according to the differences in factors that affect their power demand, predict the power demand for each group with high accuracy, and aggregate the prediction results to predict the total power demand with high accuracy, which is effective in improving the accuracy of power demand prediction. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2022-84328 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the invention disclosed in Patent Document 1, multiple customer groups are divided based on the characteristic business operations of each consumer, such as a first group that consumes electricity in offices and a second group that consumes electricity in production factories. In other words, the invention disclosed in Patent Document 1 does not take into consideration how future demand will change in accordance with changes in the geographical elements of the land in the power station's supply area, or changes in the equipment of consumers and generators. Therefore, the invention disclosed in Patent Document 1 may not be able to fully solve the problem of improving the accuracy of the forecast results regarding demand. Furthermore, no consideration is given to the problem of improving the accuracy of the forecast results regarding power sources.

[0006] The present invention has been made in response to such conventional circumstances, and aims to provide a demand and power supply forecasting system that can accurately forecast future trends in demand and power supply due to changes in the geographical elements of the land and changes in the equipment of consumers and generators. [Means for solving the problem]

[0007] In order to achieve the above object, a first invention includes a first model generation unit that uses, as first input data, an image of a service area to which a power station supplies power, taken from the air, and uses, as first output data, feature elements extracted from features included in the image, to generate a first model; a second model generation unit that uses, as second input data, the feature elements in the image at the first time point, the actual power at the first time point, and the elapsed time from the first time point to a second time point after the first time point, and uses, as second output data, the feature elements in the image at the second time point and the potential at the second time point; and an estimation unit that estimates, based on the first and second models, third output data including feature elements at a future time point and a potential at a future time point that correspond to third input data including the latest image, the latest actual power, and a predetermined elapsed time from the present time to a predetermined future time point, wherein the actual power data is at least one of actual demand and actual power source data, and the potential is at least one of actual demand and actual power source data.

[0008] In the invention configured as described above, the image taken from the sky is a satellite image or an aerial image taken from an aircraft such as an airplane, helicopter, or drone. Furthermore, a feature refers to any natural or man-made object on the ground. The extracted feature elements can be classified by use, the location of each feature type, and the area of ​​each feature type. Examples of types include fields, vacant lots, factories, commercial facilities, buildings, houses, solar panels, and forests.

[0009] Furthermore, demand potential and power supply potential are concepts that represent the possibility of electricity demand and the possibility of power supply, respectively. Demand potential may increase in the future if the number of factories, commercial facilities, and houses increases, and may decrease if the area of ​​farmland and vacant land increases. Furthermore, power supply potential may increase in the future if the area of ​​solar panel installation increases, for example. Therefore, it is believed that both the demand potential and the power source potential are closely related to the characteristic elements of the features.

[0010] In the invention having the above configuration, after the first model extracts feature elements from the image, the second model outputs the feature elements in the image extracted by the first model and the potential after an elapsed time based on the actual power consumption. Therefore, by combining the first model and the second model, the potential corresponding to the image is output via the feature elements of the features. Therefore, the estimation unit estimates characteristic elements at a future time and at least one of the demand potential and the power supply potential at a future time from the latest image, the latest performance, and a predetermined elapsed time based on the first and second models.

[0011] The second invention is characterized in that, in the first invention, the actual results are demand results and power source results, and the potential is demand potential and power source potential, and the second invention is characterized in that it includes a calculation unit that calculates a demand forecast result by adding the demand potential at a future time point and the forecasted demand for each power station, and that calculates a power source forecast result by adding the power source potential at a future time point and the maximum power that can be received for each power station.

[0012] In the invention configured as described above, the estimated demand is the demand for electric power that is estimated to be required in the future by consumers at each electric power station. The maximum received power is the maximum amount of power that each power station receives from a power generator such as a renewable energy power generation facility.

[0013] Furthermore, both the demand potential and the power source potential are values ​​that take into account the characteristics of features, i.e., geographical factors. Furthermore, the expected demand and the maximum available power are values ​​that indicate the future intentions of the consumer and the generator, respectively.

[0014] Therefore, in the invention having the above configuration, in addition to the function of the first invention, the calculation unit calculates the demand forecast result and the power source forecast result that take into account geographical factors and the intentions of the consumers.

[0015] A third invention is characterized in that, in the second invention, the calculation unit calculates a demand trend based on at least a demand potential and a predetermined elapsed time, and calculates a power supply trend based on at least a power supply potential and a predetermined elapsed time. In the invention having such a configuration, the demand trend is, for example, the rate of change in the demand potential over a predetermined elapsed time, and the power supply trend is, for example, the rate of change in the power supply potential over a predetermined elapsed time.

[0016] The predetermined elapsed time may also be divided into desired periods. In this case, the demand trend at the end of the first divided period is calculated based on the latest demand record, the demand potential at the end of the first period, and the elapsed time corresponding to the first period. The demand trend for each subsequent divided period is calculated based on the demand potential for each divided period and the elapsed time corresponding to each divided period. The same applies to the power supply trend.

[0017] In the invention having the above configuration, in addition to the function of the second invention, the demand trend and power supply trend are obtained for each predetermined elapsed time or for each desired period. Therefore, from these demand trends and power supply trends, the degree of change in the demand potential and the power supply potential are output in detail for each predetermined elapsed time, etc.

[0018] A fourth invention is characterized in that, in the first or second invention, the characteristic elements include the type, position, and area of ​​the feature. In such an invention, the features of the features are recognized, for example, by using well-known deep learning image recognition techniques.

[0019] In the invention having the above configuration, in addition to the effects of the first or second invention, by matching the position at the first point in time with the position at the second point in time, the type and area at the second point in time are compared with the type and area at the first point in time.

[0020] As for types, as described above, a variety of natural and artificial objects such as fields, vacant lots, factories, commercial facilities, buildings, houses, solar panels, etc. are recognized, and their areas are also recognized. This allows the estimation unit to estimate in detail changes in the geographical elements of the land, such as the area of ​​solar panels installed when farmland is converted into solar panels, or the area of ​​vacant land when a house is converted into vacant land. [Effects of the Invention]

[0021] According to the first invention, by combining the first model and the second model, the potential corresponding to the image is output through the characteristic elements of the features, and since the potential is considered to be closely related to the characteristic elements, it is possible to accurately predict future changes in the potential due to changes in the geographical elements of the land.

[0022] According to the second invention, in addition to the effects of the first invention, the calculation unit calculates the demand forecast results and power source forecast results taking into account the geographical elements of the land and the intentions of the consumers, making it possible to realistically and accurately estimate future demand and power sources.

[0023] According to the third invention, in addition to the effect of the second invention, the calculation unit outputs detailed changes in demand potential and changes in power supply potential from the demand trend and power supply trend at predetermined intervals, etc., so that these changes over time can be intuitively grasped.

[0024] According to the fourth invention, in addition to the effects of the first or second invention, the characteristic elements allow the estimation unit to estimate in detail changes in the geographical elements of the land, making it possible to more accurately predict future trends in demand and power sources due to changes in the geographical elements of the land. [Brief explanation of the drawings]

[0025] [Figure 1] 1 is a configuration diagram of a demand and power supply forecasting system according to an embodiment. [Figure 2] FIG. 2 is a process diagram of a model generation method executed by the demand and power supply forecasting system according to the embodiment. [Figure 3] 1 is a list of first training data used in a model generation method executed by a demand and power supply estimation system according to an embodiment. [Figure 4] 10 is a list of second training data used in a model generation method executed by a demand and power supply estimation system according to an embodiment. [Figure 5] FIG. 2 is a process diagram of an estimation method executed by a demand and power source estimation system according to an embodiment. [Figure 6] 10 is an example of a demand trend survey used by the demand and power supply forecasting system according to the embodiment. [Figure 7] 10 is an example of a power supply application status used by the demand and power supply forecasting system according to the embodiment. [Figure 8]10 is a list of output results output by the demand and power supply forecasting system according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION [Example]

[0026] A demand and power supply forecasting system according to an embodiment of the present invention will be described in detail with reference to Figures 1 to 8. Figure 1 is a configuration diagram of a demand and power supply forecasting system according to an embodiment. As shown in FIG. 1, a demand and power supply estimation system 1 according to the embodiment includes an input unit 2, an output unit 3, a control unit 4, and a storage unit 9. Of these, the control unit 4 is a central control device that controls all operations from the input unit 2 to the memory unit 9, and is equipped with a first model generation unit 5, a second model generation unit 6, an estimation unit 7, and a calculation unit 8.

[0027] Next, the memory unit 9 is a memory means for storing various data that are input, generated, output, etc. in association with the multiple processes (see Figures 2 and 5) executed by the demand and power source estimation system 1, and includes an input data memory unit 10, an output data memory unit 11, a first teacher data memory unit 12, a first model memory unit 13, a second teacher data memory unit 14, a second model memory unit 15, a first estimated data memory unit 16, and a second estimated data memory unit 17. Each component will be described below in order.

[0028] The input unit 2 receives first training data for generating a first model, second training data for generating a second model, third input data to be input to the estimation unit 7, and the like. The first training data is a combination of first input data and first output data corresponding to the first input data, and the second training data is a combination of second input data and second output data corresponding to the second input data. The contents of the first input data and first output data will be described with reference to Fig. 3. The contents of the second input data and second output data will be described with reference to Fig. 4. The contents of the third input data will be described with reference to Fig. 8.

[0029] The output unit 3 outputs the demand forecast results, power supply forecast results, demand trends, power supply trends, etc. calculated by the calculation unit 8. The output results from the output unit 3 can be viewed on a display screen (not shown) provided in the demand and power supply forecasting system 1.

[0030] The first model generation unit 5 generates a first model using the first training data, and the second model generation unit 6 generates a second model using the second training data.

[0031] The estimation unit 7 estimates third output data corresponding to the third input data based on the first model generated by the first model generation unit 5 and the second model generated by the second model generation unit 6. The content of this third output data will be described with reference to FIG. 8.

[0032] The calculation unit 8 calculates a demand estimation result by adding the demand potential included in the third output data estimated by the estimation unit 7 to the estimated demand for each power station, and also calculates a demand estimation result by adding the power source potential included in the third output data to the maximum power receivable for each power station. The estimated demand and maximum power receivable are input via the input unit 2. The details of these will be described using Figs. 6 and 7, respectively.

[0033] The calculation unit 8 also calculates a demand trend based on at least the demand potential and the predetermined elapsed time included in the third input data, and calculates a power trend based on at least the power potential and the predetermined elapsed time. The contents of the demand trend and power trend will be explained using FIG. 8.

[0034] The input data storage unit 10 stores the third input data, the expected demand, the maximum receivable power, and the like input to the input unit 2. Furthermore, the output data storage unit 11 stores the demand estimation results, power estimation results, demand trends, and power trends calculated by the calculation unit 8 using the demand potential, power potential, etc. estimated by the estimation unit 7.

[0035] The first teacher data storage unit 12 stores the first teacher data input to the input unit 2. Furthermore, the first model storage unit 13 stores the first model generated by the first model generation unit 5.

[0036] The second teacher data storage unit 14 stores the second teacher data input to the input unit 2. Furthermore, the second model storage unit 15 stores the second model generated by the second model generation unit 6.

[0037] The first estimated data storage unit 16 stores the feature elements estimated by the estimation unit 7 based on the first model. The second estimated data storage unit 17 stores the feature elements, demand potential, and power supply potential estimated by the estimation unit 7 based on the first and second models. The content of the feature elements, demand potential, etc. will be described with reference to FIG. 8.

[0038] Next, a model generation method executed by the demand and power supply forecasting system will be described with reference to Fig. 2. Fig. 2 is a process diagram of the model generation method executed by the demand and power supply forecasting system according to the embodiment. 2, the model generation method 20 executed by the demand and power supply estimation system 1 includes a first teacher data acquisition step of step S21, a first model generation step of step S22, a second teacher data acquisition step of step S23, and a second model generation step of step S24. Each step will be described below.

[0039] The first teacher data acquisition step of step S21 is a step in which the first model generation unit 5 acquires the first teacher data input via the input unit 2. This first training data is obtained by taking an image of part or all of the supply area to which the power station supplies electricity taken from above as the first input data, and extracting feature elements that represent the characteristics of features contained in the image as the first output data.

[0040] The first model generation step of step S22 is a step of generating a first model by machine learning using the first training data acquired by the first model generation unit 5. This machine learning is performed by a known learning method.

[0041] The second training data acquisition process of step S23 is a process in which the second model generation unit 6 acquires the first output data output by the first model generation unit 5, a portion of the second input data input via the input unit 2, and the second output data as second training data.

[0042] In this process, the first output data becomes the feature elements of the feature at the first time point. Furthermore, part of the second input data is the actual power consumption at a first time point and the time elapsed from the first time point to a second time point after the first time point. Therefore, the second input data is the feature elements at the first time point, the actual power consumption at the first time point, and the elapsed time.

[0043] The second output data is the feature element at the second time point and the potential at the second time point. Here, actual power consumption refers to actual demand and actual power supply. Potential refers to demand potential and power supply potential. The first point in time and the second point in time are both arbitrary points in time. Therefore, when the estimation unit 7 estimates the future demand potential and power supply potential, the first point in time can be set as the current point in time, and the second point in time can be set as a future point in time after the current point in time.

[0044] The second model generation step of step S24 is a step in which the second model generation unit 6 generates a second model by machine learning using the acquired second training data. This machine learning is performed by a known learning method, as in the case of the first model generation unit 5.

[0045] Next, the contents of the first training data will be described with reference to Fig. 3. Fig. 3 is a list of the first training data used in the model generation method executed by the demand and power supply estimation system according to the embodiment. As shown in FIG. 3, the first training data is a combination of an image, which is the first input data, and feature elements of features, which is the first output data. Of these, the images are aerial image data of part or all of the service area to which the power station supplies electricity, photographed from the sky by a camera mounted on an aircraft, and are identified by the image numbers (1 to 5) listed in the first column. Note that the images identified by each image number may be photographed from the same or different locations and ranges. The first output data, which is the feature elements, is the type (second column), position (third column), and area (fourth column) recognized from the image. Of these, the type and area are recognized by well-known deep learning image recognition techniques such as instance segmentation.

[0046] The types of feature elements are classified by the purpose of the feature, such as rice paddies, solar panels, factories, etc. The locations are the center coordinates (longitudes X11 to X52, latitudes Y11 to Y52) of each recognized type, obtained from a two-dimensional topographical map obtained by aerial surveying using an aircraft. The areas are the areas (S11 to S53) of each recognized type.

[0047] Next, the contents of the second training data will be described with reference to Fig. 4. Fig. 4 is a list of the second training data used in the model generation method executed by the demand and power supply estimation system according to the embodiment. As shown in FIG. 4, the second training data is a combination of the second input data and the second output data. The second input data includes characteristic elements at a first time point (first column), actual demand at the power station at the first time point (second column), actual power supply at the power station at the first time point (third column), and the elapsed time from the first time point to a second time point after the first time point (fourth column).

[0048] The second output data is the characteristic element at the second time point (fifth column), the demand potential at the second time point (sixth column), and the power supply potential at the second time point (seventh column). The position included in the feature element at the first time point is the same as the position included in the feature element at the second time point. In other words, the second training data is data used by the second model generation unit 6 to learn changes in type and area extracted from two or more images taken at different times so as to include the same position, and changes in demand and power supply related to these changes.

[0049] More specifically, characteristic element E11, demand performance D1, and power supply performance P1 at a first time point change to characteristic element E12, demand potential DP12, and power supply potential PP12 at a second time point after November 15th, 1 year. The feature element E11 and the feature element E12 have the same position, but the type and area are the same or different.

[0050] Furthermore, the characteristic element E31, actual demand D3, and actual power supply P3 at the first time point change to the characteristic element E32, demand potential DP32, and power supply potential PP32 at the second time point after February 3, 2 years. Furthermore, the characteristic element E31, demand record D3, and power supply record P3 at the first time point change to the characteristic element E33, demand potential DP33, and power supply potential PP33 at the second time point after July 3, 7 years. The feature element E31, the feature element E32, and the feature element E33 all include the same position. However, the types and areas may be the same or different. In this way, there may be multiple second points in time. The positions included in the characteristic elements E11 to E71 are different from each other.

[0051] Furthermore, the estimation method executed by the demand and power supply estimation system will be described with reference to Fig. 5. Fig. 5 is a process diagram of the estimation method executed by the demand and power supply estimation system according to the embodiment. 5, the estimation method 30 includes a third input data acquisition step in step S31, a third output data estimation step in step S32, an estimated demand etc. acquisition step in step S33, an estimated result calculation step in step S34, and a trend calculation step in step S35. Each step will be described below.

[0052] The third input data acquisition step of step S31 is a step in which the estimation unit 7 acquires third input data input via the input unit 2. The third input data includes the latest image of the supply area of ​​the electric power station taken from the air, the latest actual demand data of the electric power station, the latest actual power supply data of the electric power station, and a predetermined elapsed time from the present time to a predetermined future time point. The future time is the time at which the demand and power supply are expected to change, specifically, the year, month, and day. The predetermined elapsed time is expressed as either the year, month, and day, or the year, month, and day.

[0053] The third output data estimation step of step S32 is a step in which the estimation unit 7 estimates third output data corresponding to the third input data based on the first model and the second model. The third output data includes feature elements extracted from features of features included in the latest image at a future time, a demand potential at a future time, and a power supply potential at a future time. With regard to the third input data, in detail, the estimation unit 7 extracts a current feature element corresponding to the latest acquired image based on the first model, and uses this current feature element as one piece of the third input data.

[0054] The estimated demand and other information acquisition step of step S33 is a step in which the calculation unit 8 acquires the estimated demand and the maximum power that can be received for each power station. This estimated demand is the electricity demand that consumers at each power station are expected to need in the future, and is a value obtained from a demand trend survey in which a questionnaire is administered to multiple consumers. The maximum received power is the maximum amount of power that each power station receives from a power generator such as a renewable energy power generation facility, and is a value obtained based on the power source application status obtained by conducting a survey of multiple power generators. The demand trend survey and the power supply application status are each stored in the input data storage unit 10.

[0055] The calculation unit 8 also has a character recognition function that can read characters. Therefore, the calculation unit 8 acquires the estimated demand and the maximum receivable power by referring to the demand trend survey and the power source application status stored in the input data storage unit 10, and automatically reads the corresponding estimated demand and maximum receivable power.

[0056] Here, the expected demand for each electric power station will be described with reference to Fig. 6. Fig. 6 shows an example of a demand trend survey used by the demand and power supply expectation system according to the embodiment. FIG. 6 shows the results of a demand trend survey conducted by questionnaire surveying multiple consumers, which were input via the input unit 2. In the table shown in FIG. 6, 4.3 to 0.4 [MVA] in the "2023 Actual Results" in the top right column is the latest actual demand results for each of substations A to D, which is one of the third input data.

[0057] The second to seventh columns in the bottom row are the expected demand for each year in the future. For example, if the future point in time calculated from the present time and the specified elapsed time is 2025 for substation A, then 4.5 MVA for "2025 expected" is selected as the expected demand. For substations B to D, the expected demand for each corresponding year is also selected according to the future point in time.

[0058] Next, the maximum power received for each power station will be described with reference to Fig. 7. Fig. 7 shows an example of a power supply application status used by the demand and power supply estimation system according to the embodiment. FIG. 7 shows the status of power supply applications based on a questionnaire sent to a plurality of power generators, and is input via the input unit 2.

[0059] In the table shown in Fig. 7, the value shown in the rightmost column at the bottom is the maximum power that can be received. Therefore, for example, for substation A, if the future point in time calculated from the present time and the predetermined elapsed time is after the power receiving start date shown in the third column at the bottom, 5000 [kVA] is selected as the maximum power that can be received. On the other hand, if the future point in time is before the power receiving start date, the maximum power that can be received is zero. For substations B to D, the maximum power to be received is also selected depending on whether the future time is after the start date of receiving power or before the start date of receiving power.

[0060] Returning to Fig. 5, the step of calculating the estimated results in step S34 is a step in which the calculation unit 8 calculates the demand estimation result and the power source estimation result. These demand estimation result and power source estimation result are the final estimated results. Therefore, the demand estimation result is a value obtained by adding the demand potential at a future time point estimated by the estimation unit 7 and the estimated demand (see FIG. 6) selected according to the future time point. The power supply estimation result is a value obtained by adding the power supply potential at a future time point estimated by the estimation unit 7 to the maximum power supply available at this future time point (see FIG. 7).

[0061] The trend calculation process of step S35 is a process in which the calculation unit 8 calculates a demand trend based on at least the demand potential estimated by the estimation unit 7 and a predetermined elapsed time, and calculates a power supply trend based on at least the power supply potential estimated by the estimation unit 7 and a predetermined elapsed time.

[0062] The demand trend is the difference obtained by subtracting the latest actual demand from the future demand potential, and dividing the difference by the product of the latest actual demand and a predetermined elapsed time. The power trend is a value obtained by subtracting the latest power performance from the power potential at a future point in time, and dividing the difference by the product of the latest power performance and a predetermined elapsed time. The predetermined elapsed time may be set to a desired period, such as every year. In this case, it is necessary to estimate the demand potential and the power supply potential every year from the present time.

[0063] The output results output by the demand and power supply forecasting system will be described with reference to Fig. 8. Fig. 8 is a list of output results output by the demand and power supply forecasting system according to the embodiment. 8, the first to fourth columns are the third input data acquired by the estimation unit 7. Among these, the latest image number in the first column is a number for identifying the latest image captured of the supply area. The predetermined elapsed time is set to one year. Therefore, the fifth to seventh columns are the third output data estimated by the estimation unit 7 for each year.

[0064] The eighth and ninth columns are the annual demand forecast results and power supply forecast results, respectively, calculated by the calculation unit 8. As mentioned above, the demand forecast result is the sum of the demand potential at a future point in time and the forecast demand selected for that future point in time, and the power supply forecast result is the sum of the power supply potential at a future point in time and the maximum power that can be received for that future point in time. The future point in time is also calculated from the present time and a predetermined elapsed time.

[0065] Next, the tenth and eleventh columns are the demand trend and power supply trend for each year calculated by the calculation unit 8, respectively. Specifically, in the 10th column, demand trend DT(1) is calculated using the formula [D(1)-DL] / [DL·(1 year)]. Demand trend DT(6) is calculated using the formula [D(6)-D(5)] / [D(5)·(1 year)]. Similarly, demand trend DT(10) is calculated using the formula [D(10)-D(9)] / [D(9)·(1 year)]. That is, the demand trend is calculated based on a predetermined elapsed time using the latest demand DL and the subsequent estimated demand potential D(1), or two types of estimated demand potentials D(6), D(5), etc. The power generation trend in column 11 is calculated annually in the same manner as above.

[0066] As described above, in the demand and power supply forecasting system 1, as shown in FIG. 8, the demand potential and power supply potential corresponding to the latest image are output using the latest image of the supply area, the latest demand record, and the latest power supply record, via feature elements extracted from the features of the features contained in the latest image. Therefore, the demand and power supply forecasting system 1 can accurately forecast future demand potential and power supply potential changes that accompany changes in the geographical elements of land in a supply area.

[0067] Furthermore, according to the demand and power supply estimation system 1, the estimation unit 7 automatically estimates future demand potential from the latest images and the latest performance data, eliminating the need for a person in charge to manually calculate these values ​​as in the past. Therefore, the demand and power supply estimation system 1 can save the person in charge time and effort, and can estimate future potential easily and efficiently.

[0068] Furthermore, according to the demand and power supply forecasting system 1, the calculation unit 8 calculates the demand forecast results and power supply forecast results, so that future trends in demand and power supply can be accurately predicted due to changes in the geographical elements of the land as well as changes in the equipment of consumers and power generation equipment.

[0069] In addition, according to the demand and power supply forecasting system 1, the calculation unit 8 outputs detailed changes in demand potential and changes in power supply potential from the demand trend and power supply trend at predetermined intervals, allowing these changes over time to be intuitively grasped. Therefore, if the demand trend and power supply trend are used as materials for managing facilities at a power station, for example, it can be expected that the facilities management will be carried out appropriately.

[0070] The demand and power supply forecasting system according to the present invention is not limited to the one shown in the embodiment. For example, either the demand forecast result or the power supply forecast result may be forecast. In addition, in the forecasting method 30, the trend calculation step of step S35 may be omitted. [Industrial Applicability]

[0071] The present invention can be used as a demand and power supply forecasting system that forecasts future demand for power and future power supplies. [Explanation of symbols]

[0072] 1...Demand and power supply estimation system 2...Input unit 3...Output unit 4...Control unit 5...First model generation unit 6...Second model generation unit 7...Estimation unit 8...Calculation unit 9...Memory unit 10...Input data memory unit 11...Output data memory unit 12...First teacher data memory unit 13...First model memory unit 14...Second teacher data memory unit 15...Second model memory unit 16...First estimated data memory unit 17...Second estimated data memory unit 20...Model generation method 30...Assumption method

Claims

1. a first model generation unit that generates a first model by using an image of a service area to which the electric power station supplies power taken from above as first input data and feature elements obtained by extracting features of features included in the image as first output data; a second model generation unit that generates a second model using the feature element in the image at a first time point, the actual power at the first time point, and the elapsed time from the first time point to a second time point after the first time point as second input data, and uses the feature element in the image at the second time point and the potential at the second time point as second output data; an estimation unit that estimates third output data including the feature element at a future time point and the potential at a future time point, the third output data corresponding to third input data including the latest image, the latest performance, and a predetermined elapsed time from the present time point to a predetermined future time point, based on the first and second models; The actual results are at least one of a demand actual result and a power source actual result, The demand and power supply forecasting system is characterized in that the potential is at least one of a demand potential and a power supply potential.

2. The actual results are the demand actual results and the power source actual results, the potential is the demand potential and the power supply potential, 2. The demand and power supply estimation system according to claim 1, further comprising a calculation unit that calculates a demand estimation result by adding together the demand potential at the future time point and the estimated demand for each of the electric power stations, and that calculates a power supply estimation result by adding together the power supply potential at the future time point and the maximum power that can be received for each of the electric power stations.

3. 3. The demand and power supply forecasting system according to claim 2, wherein the calculation unit calculates a demand trend based on at least the demand potential and the predetermined elapsed time, and calculates a power supply trend based on at least the power supply potential and the predetermined elapsed time.

4. 3. The demand and power supply forecasting system according to claim 1, wherein the characteristic elements include the type, location, and area of ​​the feature.

Images