Congestion probability prediction device and congestion probability prediction method

Abstract

To provide a congestion probability prediction device capable of facilitating power generation control of power generation equipment with non-firm type connection while especially considering the uncertainties.SOLUTION: A disclosed congestion probability prediction device for predicting the congestion of a power system includes: a weather forecast data input unit that acquires weather forecast data regarding the weather a predetermined number of days after the current date; a power system data input unit that acquires demand data regarding electricity demand a specified number of days after the current date and that acquires system data regarding the power system after the predetermined number of days after the current date; a system congestion probability prediction unit that predicts congestion points and congestion probability data of the power system the specified number of days after the current date.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a congestion probability prediction device and a congestion probability prediction method.

Background Art

[0002] In a power system, it is necessary to balance demand and supply within a certain area. For example, Patent Document 1 describes a technology in which an energy operation system acquires information including current and future weather conditions and social environment situation patterns, predicts future energy demand and power generation amounts based on these, and controls the supply and demand within the management area based on the prediction results.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In recent years, with the large-scale introduction of renewable energy such as solar power generation and wind power generation, when the power generation amount of renewable energy increases, the amount of power passing through power transmission and transformation equipment such as power transmission lines and substation equipment in the power system exceeds the operating capacity of the power transmission and transformation equipment, resulting in system congestion. As a means of promoting the introduction of renewable energy while avoiding this system congestion, a system connection method for power generation equipment called non-firm connection has been introduced.

[0005] Non-firm connection means connecting power generation equipment to the power system on the premise of suppressing power generation output when there is no room left in the operating capacity of the power transmission and transformation equipment. In non-firm connection, when congestion occurs in the power transmission and transformation equipment, the operator of the power system instructs the power generation equipment to suppress the generated power, and the operator of the power generation equipment avoids system congestion by suppressing the generated power. This instruction is called generation control.

[0006] The timing and extent of power generation control depend on the power generated by all power generation facilities connected to the power grid, the power consumption of the load (or consumer), and the physical configuration of the power grid's transmission and distribution equipment. Therefore, it is difficult for operators or businesses of non-firmly connected power generation facilities to predict when and how much power generation control will occur. As a result, there has been a problem of the amount of power generation control (hereinafter referred to as "power generation control amount") being large as a safety precaution. Furthermore, the amount of power generation controlled is subject to uncertainty due to the influence of renewable energy, which fluctuates depending on weather conditions. This further complicates the operational planning of power generation facilities.

[0007] However, Patent Document 1 only considers the balance of power supply and demand, and does not focus on the relationship between the supply and demand balance and the constraints such as the operating capacity of power transmission and distribution equipment in the power grid, which changes moment by moment, and on controlling power generation in response to these changes. For operators or businesses of non-firm-connected power generation facilities, the technology in Patent Document 1 is not easy to use. Therefore, the present invention aims to facilitate power generation control for non-firm connection power generation equipment, taking uncertainty into particular. [Means for solving the problem]

[0008] The congestion probability prediction device of the present invention is a congestion probability prediction device for predicting congestion in a power grid, comprising: a weather forecast data input unit that acquires weather forecast data relating to the weather a predetermined number of days from the present; a power grid data input unit that acquires demand data relating to electricity demand a predetermined number of days from the present, and also acquires grid data relating to the power grid a predetermined number of days from the present; and a grid congestion probability prediction unit that predicts the congested locations and congestion probability data of the power grid a predetermined number of days from the present based on the weather forecast data, the demand data, and the grid data. The congestion probability data includes the congestion probability of the power system, the amount of power generation controlled in the power system, and the probability distribution of the amount of power generation controlled. The system congestion probability prediction unit outputs the predicted congestion locations and the predicted congestion probability data, and for each time period, outputs the highest probability predicted output after power generation control, and the ratio of the highest probability predicted output after power generation control to the predicted output before power generation control. Furthermore, for each time period where this ratio is not 100%, it outputs the maximum and minimum possible predicted outputs after power generation control. It is characterized by the following. Other means will be described within the descriptions of embodiments for carrying out the invention. [Effects of the Invention]

[0009] According to the present invention, power generation control for non-firm connection power generation equipment can be facilitated, especially while taking uncertainty into consideration. [Brief explanation of the drawing]

[0010] [Figure 1] This is a diagram showing the configuration of the first embodiment. [Figure 2] This is a flowchart of the processing procedure in the first embodiment. [Figure 3] This is a flowchart detailing step S131. [Figure 4] This is a detailed flowchart of step S132. [Figure 5] This is a detailed flowchart of step S1324. [Figure 6] This is a detailed flowchart of step S1325. [Figure 7] This is an example of congestion probability data. [Figure 8] This is an example of a power grid diagram showing congested areas. [Figure 9] This is an example of a time-series graph showing the changes in power generation control amounts. [Figure 10] This diagram shows the configuration of a power system and the flow of signals. [Figure 11] This is a flowchart of the processing procedure in the second embodiment. [Figure 12] This is a diagram showing an example of a train schedule. [Figure 13A] This figure shows the power output of the power generation equipment, the power consumption of the load, and the charge level of the energy storage equipment in a time-series manner when the present invention is not applied. [Figure 13B] This figure shows the power output of the power generation equipment, the power consumption of the load, and the charge level of the energy storage equipment in time series when the operation plan of the present invention is applied. [Figure 14] This is a flowchart of the processing procedure in the third embodiment. [Figure 15]It is a diagram showing the configuration of the fourth embodiment.

Mode for Carrying Out the Invention

[0011] Hereinafter, four embodiments will be described with reference to the drawings. The first embodiment is an example in which the congestion probability prediction device distributes the prediction result of line congestion. The second embodiment is an example in which the congestion probability prediction device creates an operation plan for electrical equipment. The third embodiment is an example in which the congestion probability prediction device calculates and outputs an electricity price. The fourth embodiment is an example in which the congestion probability prediction device acquires a power generation plan from a power generation company or the like.

[0012] 〈First Embodiment〉 FIG. 1 is a diagram showing the configuration of the first embodiment. The congestion probability prediction device 1 is connected to the operator 3 of the power grid, the weather prediction company 4, the power generation company 51, and the aggregator 52 via the network 2 (connected to each computer). The aggregator 52 is a main body that bundles and integrally manages and operates power generation facilities, loads, energy storage facilities, etc. connected to the power grid, and is a main body that transmits an operation plan to the power generation facilities, etc. that it manages and operates (details will be described later).

[0013] The congestion probability prediction device 1 is a general computer and includes a central control device, an input device, an output device, a main memory device, an auxiliary memory device, a communication device, etc. (not shown). Then, the central control device realizes the functions of the program by reading the program from the auxiliary memory device to the main memory device. The power grid data input unit 11, the weather prediction data input unit 12, and the line congestion probability prediction unit 13 in FIG. 1 are programs.

[0014] The power grid data input unit 11 of the congestion probability prediction device 1 acquires generation data 31, demand data 32, and grid data 33 from the operator 3 of the power grid via the network 2. The generation data 31 stores the location of the power generation facilities and the rated capacity of the power generation facilities capable of output. The rated capacity may be stored in a future time series. Demand data 32 stores the location and power consumption of consumers who consume electricity. Power consumption may be stored chronologically as a future electricity consumption plan for consumers. Demand data 32 may be an electricity consumption plan that aggregates many consumers by region, or it may be an electricity consumption plan for several specific large consumers.

[0015] Furthermore, the demand data 32 may be a pinpoint value whose value is uniquely determined, a predicted value that includes a confidence interval or variance, or a probability prediction of a probability distribution. For example, the probability distribution of the demand data 32 can be predicted using statistical methods based on date and time, weather data, past demand performance data, etc. System data 33 stores location information, topology information, and impedance information of transmission lines and transformers that constitute the power system.

[0016] It should be noted that these data are merely examples, and the data acquired by the power system data input unit 11 is not limited to these. The processing performed by the power system data input unit 11 may be fully automated, manually performed by a user using a keyboard or the like, or a combination of both.

[0017] The weather forecast data input unit 12 of the congestion probability prediction device 1 acquires weather forecast data 41 from a weather forecasting company 4 via the network 2. Weather forecast data 41 stores predicted values ​​for future solar radiation, wind speed, wind direction, temperature, humidity, precipitation, etc. Weather forecast data 41 is calculated using numerical prediction models (weather models), machine learning, statistical methods, or a combination thereof. Weather forecast data 41 is data with geographical resolution; for example, it is defined on each mesh or grid point after dividing a map into a mesh. Weather forecast data 41 is also data with temporal resolution, storing predicted values ​​for a predetermined number of days (from several days to several months) from the present as future prediction values.

[0018] Furthermore, the weather forecast data 41 may be a pinpoint value whose value is uniquely determined, a forecast value that includes a confidence interval or variance, or a probability forecast value of a probability distribution. The shape of the probability distribution is not particularly limited and may be any shape such as a normal distribution, asymmetric distribution, or multimodal distribution.

[0019] The grid congestion probability prediction unit 13 of the congestion probability prediction device 1 transmits (distributes) the probabilistically predicted congestion locations and times (see Figure 7) to the power generator 51 and the aggregator 52 via the network 2. Furthermore, the grid congestion probability prediction unit 13 transmits the probabilistically predicted power generation control amount (see Figure 8) to the power generator 51 and the aggregator 52.

[0020] Figure 2 is a flowchart of the processing procedure in the first embodiment. In step S130, the congestion probability prediction unit 13 of the congestion probability prediction device 1 accepts input (selection) from the user via an input device of the area (prediction target area) and the period (prediction target period) for which congestion will be predicted. The prediction target area and prediction target period may be predetermined, or they may be input externally each time congestion probability prediction is needed. The prediction target period is usually a period that starts from the present and ends at a predetermined number of days later.

[0021] In step S131, the power system data input unit 11, the weather forecast data input unit 12, and the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 prepare the input data. Details of step S131 will be shown later in Figure 3. In step S132, the power grid data input unit 11 and the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 calculate the power generation control amount. Details of step S132 will be described later in Figures 4, 5, and 6.

[0022] In step S133, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 outputs (transmits) the probability of congestion locations, congestion times, and power generation control amounts. Specifically, the grid congestion probability prediction unit 13 outputs the congestion probability data shown in Figure 7, the power grid diagram shown in Figure 8, and the time-series transition diagram shown in Figure 9 to the power generation company 51 and the aggregator 52. After that, the processing procedure ends.

[0023] Figure 3 is a flowchart detailing step S131. In step S1311, the power grid data input unit 11 of the congestion probability prediction device 1 extracts coordinate data (location data) of renewable energy sources, such as wind power generation facilities and solar power generation facilities, whose output is determined by weather conditions, from the grid data 33. After that, the process branches into two paths.

[0024] In step S1312, the weather forecast data input unit 12 of the congestion probability prediction device 1 acquires wind speed probability data for a location or a nearby location that matches the coordinate data of the wind power generation facility from the weather forecast data 41 which has geographical resolution. The wind speed probability data acquired here is probability data relating to the weather a predetermined number of days later, based on the present.

[0025] In step S1313, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 first obtains a power curve representing the relationship between the output and wind speed of a wind power generation facility from an arbitrary facility. The grid congestion probability prediction unit 13 may create the power curve using methods such as regression analysis based on past output and wind speed data. Secondly, the grid congestion probability prediction unit 13 uses the power curve to calculate the probability prediction value of the amount of power generated by the wind power generation equipment related to the power curve obtained from the wind speed probability data (predicted value).

[0026] In step S1314, the weather forecast data input unit 12 of the congestion probability prediction device 1 acquires solar radiation probability data (predicted values) for a location or nearby location that matches the coordinate data of the solar power generation facility from the weather forecast data 41 which has geographical resolution. The solar radiation probability data acquired here is also probability data relating to the weather a predetermined number of days later, based on the present.

[0027] In step S1315, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 first obtains a power curve representing the relationship between the output and solar radiation of a photovoltaic power generation facility from an arbitrary facility. The grid congestion probability prediction unit 13 may create the power curve using methods such as regression analysis based on past data of output and solar radiation. Secondly, the grid congestion probability prediction unit 13 uses the power curve to calculate a probability prediction value of the amount of power generated by the solar power generation equipment related to the power curve obtained from the solar radiation probability data (predicted value).

[0028] After steps S1313 and S1315 are completed, the grid congestion probability prediction unit 13 will have a time-series of probability prediction values ​​for the amount of power generated in the prediction area for the prediction period. Then, the process proceeds to step S132.

[0029] Figure 4 is a flowchart detailing step S132. In step S1320, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 extracts a sample using pseudorandom numbers from the probability prediction values ​​of renewable energy generation and the probability prediction values ​​of demand data. In step S1321, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 calculates the initial power flow cross-section of the power grid within the prediction target area during the prediction target period, based on the extracted samples and conventional power sources such as thermal and nuclear power plants. The method for calculating the initial power flow cross-section is the power flow calculation method commonly used in grid analysis.

[0030] In step S1322, the grid congestion probability prediction unit 13 assigns an initial value of "1" to time T. Time T then changes to 2, 3, 4, ... In the process described later, for each time T within the prediction period, the amount of power generation control is calculated as its probability prediction value. The calculation of the amount of power generation control is repeated, for example, every hour.

[0031] In step S1323, the power system data input unit 11 of the congestion probability prediction device 1 acquires power generation data 31 and demand data 32 at time T. Here, time T is a predetermined number of days after the present. In step S1324, the grid congestion probability prediction unit 13 calculates the grid constraint quantity at time T. Details of step S1324 will be shown later in Figure 5. In step S1325, the grid congestion probability prediction unit 13 calculates the supply and demand constraint amount at time T. Details of step S1325 will be shown later in Figure 6.

[0032] In step S1326, the system congestion probability prediction unit 13 determines whether time T has reached the last time of the prediction period. Specifically, if time T has reached the last time of the prediction period (step S1326 "Yes"), the system congestion probability prediction unit 13 proceeds to step S1328; otherwise (step S1326 "No"), it proceeds to step S1327.

[0033] In step S1327, the system congestion probability prediction unit 13 adds "1" to time T and then returns to step S1323. In step S1328, the system congestion probability prediction unit 13 performs a termination determination. Specifically, if the system congestion probability prediction unit 13 has not exceeded a specified number of repetitions or a specified repetition time (step S1328 "No"), it returns to step S1320; otherwise (step S1328 "Yes"), it proceeds to step S1329.

[0034] In step S1329, the system congestion probability prediction unit 13 performs aggregation processing to create congestion probability data (Figure 7). The details of the aggregation processing will be described later in the explanation of Figure 7. The process then moves to step S133. As described above, in step S133, the grid congestion probability prediction unit 13 outputs congestion probability data to the outside of the congestion probability prediction device 1. As will be described later, the congestion probability data includes the location of grid congestion, the predicted time of grid congestion, the amount of power generation control, the probability distribution of the amount of power generation control, the probability of grid congestion occurring, or a combination thereof.

[0035] Figure 5 is a flowchart detailing step S1324 (calculation of systematic constraints at time T). In step S1324a, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 performs power flow calculations for the power grid in the area to be predicted and identifies the presence or absence of congestion points. Congestion points are locations where the transmitted power exceeds the operating capacity of transmission and distribution equipment such as transmission lines and substations. Grid congestion occurs in congested points.

[0036] The operating capacity of power transmission and distribution equipment is determined not only by the thermal capacity of the equipment but also by the constraints necessary for the stable operation of the power system. For example, when an accident such as a lightning strike occurs in the power system, the grid congestion probability prediction unit 13 defines the upper limit of the transmitted power necessary for the stable operation of the power system from the viewpoint of voltage stability, frequency stability, etc., as the operating capacity for each power transmission and distribution facility. The grid data 33 (Figure 1) also contains data on the constraints of the power system, including the operating capacity.

[0037] In step S1324b, the system congestion probability prediction unit 13 determines whether or not there are congested areas. Specifically, if there are congested areas (step S1324b "Yes"), the system congestion probability prediction unit 13 proceeds to step S1324c; otherwise (step S1324b "No"), it proceeds to step S1325.

[0038] In step S1324c, the system congestion probability prediction unit 13 extracts congested areas. In step S1324d, the grid congestion probability prediction unit 13 calculates a grid constraint amount ΔPa, which is a positive value obtained by subtracting the operating capacity of the transmission and distribution equipment from the predicted power generation value. The grid constraint amount ΔPa can also be described as the amount of excessive load (overload) on the transmission and distribution equipment. In step S1324e, the grid congestion probability prediction unit 13 decides to reduce the output of the power generation equipment using non-firm connection by ΔPa at time T. Then, the process returns to step S1324a.

[0039] The method for handling predicted grid congestion varies depending on the region and the timing of the system's application. Figure 5 shows the power generation control method for power generation facilities to which non-firm connection is applied. Note that the method for handling grid congestion is not limited to this method; the effects of the present invention can be obtained even if other systems are followed. As described above, non-firm connection is the connection of power generation facilities to the power grid on the premise that power output will be suppressed (power generation control) when the operational capacity of the transmission and distribution facilities runs out.

[0040] The grid congestion probability prediction unit 13 identifies power generation facilities to which non-firm connections are applied that are likely to cause grid congestion based on the direction of the power flow in the power flow calculation results, and proportionally allocates the grid constraint amount ΔPa to all identified power generation facilities according to their rated capacity, etc. Each identified power generation facility is expected to have its output suppressed (power generation controlled) according to the allocated grid constraint amount ΔPa. The grid congestion probability prediction unit 13 repeats this power generation control process until there are no more congested areas.

[0041] Figure 6 is a flowchart detailing step S1325 (calculation of supply and demand constraints at time T). In step S1325a, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 determines whether the predicted power generation value for the prediction target area exceeds the predicted demand value. Specifically, firstly, the grid congestion probability prediction unit 13 compares the sum of the predicted power generation values ​​for the prediction target area with the sum of the predicted demand amounts. Secondly, the grid congestion probability prediction unit 13 proceeds to step S1325b if the sum of the predicted power generation values ​​is greater (step S1325a "Yes"), and to step S1326 otherwise (step S1325a "No").

[0042] In step S1325b, the grid congestion probability prediction unit 13 calculates the supply and demand constraint ΔPb, which is the result of subtracting the sum of the predicted demand from the sum of the predicted power generation values ​​in the target area. The supply and demand constraint ΔPb can also be said to be the amount of power generation exceeding the demand. The specific method for calculating the supply and demand constraint ΔPb varies depending on the region and the timing of the application of the system.

[0043] In step S1325c, the grid congestion probability prediction unit 13 calculates the power generation control amount ΔPab for each power generation facility. Specifically, firstly, the grid congestion probability prediction unit 13 proportionally allocates the supply and demand constraint amount ΔPb to all the identified power generation facilities according to their rated capacity, etc. Secondly, the grid congestion probability prediction unit 13 calculates a power generation control amount ΔPab for each power generation facility. The power generation control amount ΔPab is the sum of a proportionally allocated grid constraint amount ΔPa and a proportionally allocated supply and demand constraint amount ΔPb.

[0044] The grid congestion probability prediction unit 13, for example, controls power generation to thermal power generation facilities and the like in accordance with priority power supply rules, and finally controls power generation to renewable energy sources. The grid congestion probability prediction unit 13 repeatedly performs power generation control processing in accordance with priority power supply rules until the sum of the predicted power generation values ​​in the target area matches the sum of the predicted demand. Note that the processing method during supply and demand constraints is not limited to this priority power supply rule, and the effects of the present invention can be obtained even if a method that complies with other systems is used.

[0045] Figure 7 shows an example of the congestion probability data results. In step S1329, the grid congestion probability prediction unit 13 creates congestion probability data (Figure 7) as a result of the aggregation process in step S1329 and outputs it externally. In the congestion probability data, the time is stored in the time column 72, the presence or absence of congestion is stored in the congestion occurrence column 73, the power generation control amount is stored in the power generation control amount column 74, the frequency is stored in the frequency column 75, the probability distribution of the power generation control amount is stored in the probability distribution of the power generation control amount column 76, and the congestion probability is stored in the congestion probability column 77.

[0046] The system congestion location in column 71 is an identifier (a, b, ..., z) that uniquely identifies the location of a power transmission line, etc., where system congestion may or may not occur. The times in column 72 indicate the times when system congestion is expected to occur or not occur. The congestion occurrence status in column 73 is either "1" indicating that congestion occurs at that location at that time, or "0" indicating that no congestion occurs.

[0047] The power generation control amount in column 74 is the power generation control amount for the power generation equipment. If the presence or absence of congestion is "0", the power generation control amount is also "0". The power generation control amount may be the analyzed value itself, or it may be the bin average of the analyzed values ​​with an arbitrary range. The frequency in column 75 is the number of times the power generation control amount occurs at that grid congestion location at that time, or the number of times the power generation control amount does not occur at that grid congestion location at that time, in a sufficiently large number of repeated simulations. The probability distribution of the power generation control amount in column 76 represents the probability that the power generation control amount occurs at that time in that location of grid congestion. The congestion probability in column 77 represents the probability that congestion will occur at that location at that time.

[0048] Figure 7 shows the following: The system congestion probability prediction unit 13 performed 1000 simulations to simulate whether or not system congestion would occur at the system congestion location "a" at the time "20xx / yy / zz / 11:00". Of these, 200 times there was no congestion, and 800 times congestion occurred. Furthermore, of these 800 instances, the amount of power generation control was 50 MW in 500 instances, and the amount of power generation control was 100 MW in 300 instances.

[0049] Therefore, in the probability distribution of the amount of power generation controlled, the probability that the amount of power generation controlled is "50" MW is 500 / 1000 = 50%. The probability that the amount of power generation controlled is "100" MW is 300 / 1000 = 30%. The probability that the amount of power generation controlled is "0" MW is 200 / 1000 = 20%. The probability of system congestion occurring is 50% + 30% = 80%. Incidentally, the system congestion probability prediction unit 13 performed 1000 simulations to determine whether or not system congestion would occur at the system congestion location "z" at the time "20xx / yy / zz / 23:00". As a result, system congestion did not occur.

[0050] Figure 8 is an example of a power system diagram showing congested areas. In step S133, the system congestion probability prediction unit 13 of the congestion probability prediction device 1 outputs the location of the congestion (congested area), the time of the congestion (congestion time), and the congestion probability to the outside. In the example of Figure 7, the system congestion probability prediction unit 13 highlights the congested area at the congestion time (transmission line a as the system congestion location) with a thick line on the system diagram, which consists of substations (○) and transmission lines (straight lines) in the prediction target area. Furthermore, the system congestion probability prediction unit 13 displays a congestion probability of "80%" in association with the congested area. The method of outputting the congested area and congestion time is not limited to the example in Figure 8, and may be, for example, outputting the congested area as text data along with a transmission line identification number and congestion time, or by other means.

[0051] Figure 9 shows an example of a time-series graph of the power generation control amount. In step S133, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 outputs a time-series graph of the power generation control amount to the outside. In the example in Figure 9, the grid congestion probability prediction unit 13 illustrates the power generation control amount for a certain power generation facility in a time series as the difference (height of the white portion of the bar graph) between the predicted output before power generation control and the predicted output after power generation control which has the highest probability. As mentioned above, the power generation control amount is caused by both grid congestion and the balance of power supply and demand (power generation control amount ΔPab = ΔPa + ΔPb).

[0052] The grid congestion probability prediction unit 13 displays, for each time period, the most probable predicted output after power generation control (shaded bar graph), and its ratio (%) to the predicted output before power generation control. Furthermore, for each time period where this ratio is not "100%", the grid congestion probability prediction unit 13 displays the maximum and minimum values ​​(MW) of the possible predicted output after power generation control in an "E-shaped chart".

[0053] For example, focusing on the 9 o'clock time slot, the most probable predicted output after power generation control is 240 MW, and the ratio of this 240 MW to the predicted output of 300 MW before power generation control is "80%". This "80%" is a ratio, not a probability. During the same time slot, the probability distribution of predicted output after power generation control is distributed within a range of approximately 30 MW, including the output level of 240 MW that corresponds to that "80%".

[0054] The method for outputting the power generation control amount is not limited to the example in Figure 9. For example, it may be a method of outputting the power generation control amount as text data along with a number that identifies the power generation equipment and the time of congestion, or by other methods.

[0055] As shown in Figure 1, the congestion probability prediction device 1 predicts the grid congestion probability and transmits the prediction results to the power generator 51 and the aggregator 52. The power generator 51 and the aggregator 52 reflect the received grid congestion probability prediction results in the operation plans of the power generation facilities, loads, energy storage facilities, etc. that they manage. Based on these grid congestion probability prediction results, the power generator 51 and the aggregator 52 can suppress the amount of power generation control of power generation facilities during grid congestion by creating operation plans that increase the amount of power consumed by loads or charge energy storage facilities, taking uncertainty into consideration. As a result of the effects of the present invention, owners and operators of power generation facilities can create operation plans that reduce the amount of power generation control, taking uncertainty into consideration, and realize the effective use of renewable energy generated power.

[0056] <Second Example> In the second embodiment, the congestion probability prediction device 1 in the first embodiment creates an operation plan for the electrical equipment (described immediately below) based on the predicted congestion locations and power generation control amounts.

[0057] Figure 10 is a diagram showing the configuration of the power system and the flow of signals. In Figure 2, the renewable energy generation equipment (also simply called "renewable energy") 62, load 63, and energy storage equipment 64 constitute the electrical equipment managed by the aggregator 52. The aggregator 52 receives an operation plan from the congestion probability prediction device 1. The operation plan received here pertains to the entire electrical equipment, including the renewable energy 62, load 63, and energy storage equipment 64 (see Figure 12). The aggregator 52 transmits the operation plan to each of these electrical pieces of equipment. The operation plans transmitted here pertain to the renewable energy 62, load 63, and energy storage equipment 64, respectively (individual commands). Note that reference numeral 61 indicates a transformer.

[0058] The electrical equipment operates according to the received operating plan. Figure 10 shows an example in which the aggregator 52 manages renewable energy 62, loads 63, and energy storage equipment 64. However, the electrical equipment managed by the aggregator 52 is not limited to these; the effects of the present invention can be obtained with any combination of equipment that generates electricity, equipment that consumes electricity, and equipment that stores and discharges electricity.

[0059] Figure 11 is a flowchart of the processing procedure in the second embodiment. Steps S130 to S132 in Figure 11 are the same as steps S130 to S132 in Figure 1. In step S134, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 creates an operating plan for the electrical equipment. Specifically, the grid congestion probability prediction unit 13 creates an operating plan for each electrical equipment under its jurisdiction to maximize the profitability of the electrical equipment under its jurisdiction, based on the predicted congestion locations and power generation control amounts. The grid congestion probability prediction unit 13 may apply an optimization algorithm such as linear programming to create the operating plan, or it may apply techniques that are common in this field.

[0060] In step S135, the grid congestion probability prediction unit 13 outputs the operation plan for the electrical equipment (Figure 12) to the power company 51 and the aggregator 52.

[0061] The grid congestion probability prediction unit 13 may create multiple operating plans. Furthermore, the grid congestion probability prediction unit 13 can switch operating plans using, for example, the difference between the planned operating value and the actual value of the power generation control amount as a threshold. The grid congestion probability prediction unit 13 may predetermine the threshold through simulation or the like, or it may accept the threshold from an external source and accept the modified threshold as needed. In addition, the grid congestion probability prediction unit 13 may use the difference between the planned operating value and the actual value of the power generation amount of renewable energy 62 as the threshold.

[0062] Figure 12 shows an example of an operation plan. The grid congestion probability prediction unit 13 transmits an operation plan to each of the renewable energy sources 62, loads 63, and energy storage equipment 64 via the aggregator 52 prior to the time when power generation control is predicted to occur. The operation plan for load 63 is, for example, an instruction to perform a start-up operation in advance in order to consume power in accordance with the occurrence of power generation control. The operation plan for energy storage equipment 64 is an instruction to perform a discharge operation in advance in order to store power in accordance with the occurrence of power generation control. The operation plan for renewable energy sources 62 is, for example, an instruction to reduce the amount of power generation control (minimize the reduction amount) when performing power generation operation.

[0063] Ultimately, if the renewable energy source 62 is forced to reduce its power output, the grid congestion probability prediction unit 13 transmits an operating plan to the load 62 and the energy storage equipment 64 that absorbs as much power output as possible, after giving them sufficient preparation time, thereby minimizing power generation control for the renewable energy source 62.

[0064] Figure 12 shows the following: (1) The congestion probability prediction device 1 predicts in advance that grid congestion will occur during the congestion time, that is, that power generation control for renewable energy 62 will be necessary. Generally, the congestion time is a predetermined number of days (several days or several months) after the present. (2) The congestion probability prediction device 1 predicts this two days before the time of congestion. In other words, in the example in Figure 12, the time of congestion is two days from the present.

[0065] (3) The congestion probability prediction device 1 transmits the following operating plan to the load 63 via the aggregator 52. (3-1) Load 63 will be shut down until one day before the peak hours. (3-2) 24 hours before or immediately before peak hours, load 63 will be started (prepared). (3-3) During peak hours, load 63 will operate with increased power consumption.

[0066] (4) The congestion probability prediction device 1 transmits the following operating plan to the energy storage equipment 64 via the aggregator 52. (4-1) The energy storage equipment 64 will discharge until two hours before the peak hours. (4-2) The energy storage equipment 64 will be charged one hour before or immediately before the peak hours. (4-3) During peak hours, the energy storage equipment 64 will be charged.

[0067] (5) The congestion probability prediction device 1 transmits the following operating plan to the renewable energy 62 via the aggregator 52. (5-1) Renewable Energy 62 will operate under normal power generation conditions until just before peak hours. (5-2) During peak hours, renewable energy 62 will operate with reduced (minimized) power generation control levels.

[0068] As is clear from Figure 12, the operation plan reflects the congestion locations, congestion times, and power generation control amounts predicted by the grid congestion probability prediction unit 13.

[0069] Figure 13 illustrates the effects of the present invention. Figure 13A shows the power output of the power generation equipment, the power consumption of the load, and the charge level of the energy storage equipment 64 in time series when the present invention is not applied. Figure 13B shows the power output of the power generation equipment, the power consumption of the load, and the charge level of the energy storage equipment 64 in time series when the operation plan of the present invention is applied.

[0070] In Figures 13A and 13B, the use of the electricity generated by renewable energy 62 is one of the following: consumption by load 63 (hatched area), storage by energy storage equipment 64 (hatched area), or output to the power grid (shaded area). Here, "output to the power grid" means consumption by loads other than those covered by the operation plan, or outflow to other power grids.

[0071] Looking at Figure 13A, we can see the following: Load 63 continues to operate at a constant rate, and its power consumption remains constant. • The energy storage equipment 64 remains shut down (neither discharging nor storing energy) until the time when the renewable energy generation power 62 increases ("1 hour before"). • At the time when the power generated by renewable energy 62 increases ("1 hour before"), the energy storage equipment 64 starts charging and reduces the power generated and output to the power grid, thereby alleviating congestion in the power grid. However, there is a limit to the amount of electricity that the energy storage equipment 64 can charge, and charging must be stopped when the charge rate approaches 100%. Therefore, if power cannot be transmitted due to grid congestion, it is necessary to suppress (control) the power generated by renewable energy 62 after the time of "power generation control occurrence".

[0072] Looking at Figure 12B, we can see the following: • To prepare for future grid congestion, load 63 shifts its operating time. In other words, it consumes power intensively only after the time when renewable energy 62's power generation is likely to increase and grid congestion may occur ("1 hour prior"), thereby avoiding the control of renewable energy 62's power generation. • In preparation for future grid congestion during charging, the energy storage equipment 64 is discharged in advance ("1 day to 1 hour before") to reduce its charge level. • At times when the power generated by renewable energy 62 increases and grid congestion may occur (from "1 hour prior"), the energy storage equipment 64 starts charging to avoid controlling the power generation of renewable energy 62.

[0073] As is clear from Figures 12 and 13, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 creates an operation plan for the load or energy storage equipment so that the amount of power generation control for the power generation equipment is small, that is, so that the amount of power generation suppression is small.

[0074] As an effect of the second embodiment, the amount of renewable energy 62 power generation control is reduced, enabling the effective use of renewable energy with low carbon dioxide emissions. In addition, consumers with power-consuming loads 63 may find that purchasing electricity generated by renewable energy 62 is cheaper than purchasing electricity through retailers, thus minimizing electricity purchase costs. Furthermore, businesses operating energy storage facilities 64 can maximize their profits by efficiently operating the energy storage facilities 64.

[0075] <Third Embodiment> In the third embodiment, the congestion probability prediction device 1 in the first embodiment calculates and outputs the electricity price from weather forecast data 41, demand data 32, and grid data 33.

[0076] Market-driven methods such as zone systems and nodal systems exist for managing congestion in the power grid. These methods alleviate congestion by differentiating electricity prices by region or location. Specifically, in areas or locations where congestion occurs, lowering electricity prices prevents power generation facilities with high costs from securing contracts in the electricity market, resulting in reduced power generation. On the other hand, raising electricity prices in areas or locations where congestion does not occur encourages increased power output from power generation facilities, leading to increased contracts in the market and alternative consumption of electricity that is being suppressed in congested areas or locations. Through such market-mediated price signals, grid congestion can be resolved.

[0077] Figure 14 is a flowchart of the processing procedure for the third embodiment. In the third embodiment, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 calculates the electricity price in step S136, replacing steps S132 and S133 in the first embodiment, in a form corresponding to a market-driven congestion management method. Then, in step S137, the grid congestion probability prediction unit 13 outputs the electricity price to the power company 51 and the aggregator 52. In other words, the grid congestion probability prediction unit 13 predicts the electricity price for each region or location. For example, the grid congestion probability prediction unit 13 predicts a lower electricity price for a region or location if the predicted amount of power generation control is large. The grid congestion probability prediction unit 13 may use a zone-based or nodal-based prediction method, which is commonly used as a method for predicting electricity prices.

[0078] In step S131 (step S1311) of Figure 14, the power grid data input unit 11 also acquires demand data 32. The grid congestion probability prediction unit 13 can also calculate the electricity price by comparing the predicted generation value with the demand data 32.

[0079] The power generator 51 and the aggregator 52 obtain predicted electricity prices from the congestion probability prediction device 1 and reflect these predicted electricity prices in the operation plans of the power generation facilities, loads, and energy storage facilities they manage. As an effect of the third embodiment, even in a market-driven congestion management system, owners and operators of electrical facilities can create operation plans that reduce the amount of power generation control, thereby enabling the effective use of renewable energy generation.

[0080] <Fourth Embodiment> In the fourth embodiment, the congestion probability prediction device 1 in the first embodiment obtains power generation plans from the power generation company 51 and the aggregator 52 and uses them to calculate the amount of power generation to be controlled.

[0081] Figure 15 shows the configuration of the fourth embodiment. In the fourth embodiment, similar to the first embodiment, the power generator 51a and the aggregator 52a obtain predicted values ​​of the power generation control amount from the congestion probability prediction device 1 via the network 2. In addition, in the fourth embodiment, the congestion probability prediction device 1 includes a power generation plan data input unit 14 as a program. The power generation plan data input unit 14 obtains power generation plans from other power generators 51b and other aggregators 52b.

[0082] In step S132 of Figure 2, the power generation plan data input unit 14 of the congestion probability prediction device 1 uses power generation plans obtained from power generators 51b and aggregators 52b when extracting congested areas by power flow calculation and calculating the amount of power generation to be controlled. By using the power generation plans created by power generators 51b and aggregators 52b, the accuracy of power flow calculations is improved and the accuracy of power generation control amount predictions is improved compared to using only the power generation data 31 published by the power system operator 3.

[0083] In Figure 15, the power generator 51a and aggregator 52a that acquire (receive) predicted values ​​of power generation control, and the power generator 51b and aggregator 52b that transmit the power generation plan, exist separately. However, even if one power generator 51 or one aggregator 52 receives the predicted values ​​of power generation control and transmits the power generation plan, the effects of the present invention can be realized. As a result of the effects of the present invention, owners and operators of electrical equipment can create operation plans that reduce the amount of power generation control, thereby realizing the effective use of renewable energy generated.

[0084] If multiple aggregators 52 exist within the same power grid, the grid congestion probability prediction unit 13 of the congestion probability prediction device 1 may distribute a single created operation plan among the multiple aggregators according to a predetermined rule. For example, if the power generation control amount in the operation plan is x, the grid congestion probability prediction unit 13 may distribute x to each aggregator according to the rated capacity of the renewable energy 62 under the jurisdiction of each aggregator. The grid congestion probability prediction unit 13 may also distribute the number of loads or the amount of consumption that should be prepared for startup to each aggregator 52 according to the number or scale of loads 63 under the jurisdiction of each aggregator. Furthermore, the grid congestion probability prediction unit 13 may distribute the number of energy storage facilities or the amount of discharge and storage to each aggregator 52 according to the number or scale of energy storage facilities 64 under the jurisdiction of each aggregator. More generally, the grid congestion probability prediction unit 13 may distribute the operation plan among multiple aggregators.

[0085] The congestion probability prediction device of this embodiment provides the following effects. (1) The congestion probability prediction device can predict congestion locations and congestion probability data for power systems to which non-firm connections are applied, taking uncertainty into account. (2) The congestion probability prediction device may use pinpoint or probabilistic weather forecast data and demand data. (3) The congestion probability prediction device can predict the congestion probability of the power grid, the amount of power generation controlled by the power grid, and the probability distribution of the amount of power generation controlled. (4) The congestion probability prediction device can predict congested locations and congestion probability data in the power grid based on the constraints of the power grid, including the operational capacity. (5) The congestion probability prediction device can output predicted congestion locations and congestion probability data to an aggregator or the like. (6) The congestion probability prediction device can create an operation plan.

[0086] (7) The congestion probability prediction device can predict electricity prices. (8) The congestion probability prediction device can acquire demand data as a power consumption plan for large consumers. (9) The congestion probability prediction device can create an operating plan for power generation equipment, loads, and energy storage equipment. (10) The congestion probability prediction device can distribute the operation plan among multiple aggregators. (11) The congestion probability prediction device can obtain power generation plans from aggregators, etc. [Explanation of Symbols]

[0087] 1. Congestion probability prediction device 2 Network 3. Operators of the power grid 4 Weather forecasting companies 11 Power System Data Input Section 12 Weather forecast data input section 13. System congestion probability prediction unit 14. Power Generation Plan Data Input Section 51 Power Generation Companies 52 Aggregators 62. Renewable energy 63 load 64 Energy Storage Systems

Claims

1. A congestion probability prediction device for predicting congestion in the power grid, A weather forecast data input unit that acquires weather forecast data for a predetermined number of days in the future, based on the current weather. A power system data input unit that acquires demand data for electricity demand a predetermined number of days later, based on the present, and also acquires system data for the aforementioned power system a predetermined number of days later, based on the present. A grid congestion probability prediction unit predicts the locations and congestion probability data of the power grid a predetermined number of days after the present, based on the weather forecast data, the demand data and the grid data, Equipped with, The aforementioned congestion probability data is, This includes the congestion probability of the power system, the amount of power generation controlled by the power system, and the probability distribution of the amount of power generation controlled. The aforementioned system congestion probability prediction unit is: The predicted congestion locations and the predicted congestion probability data are output, For each time period, the most probable predicted output after power generation control, and the ratio of the most probable predicted output after power generation control to the predicted output before power generation control are output. Furthermore, for each time period when the ratio is not 100%, the maximum and minimum values ​​of the predicted output after power generation control will be output. A congestion probability prediction device characterized by the following.

2. The weather forecast data and the demand data are The predicted value must include the confidence interval or variance, or it must be a probability prediction of a probability distribution. The congestion probability prediction device according to claim 1, characterized by the following:

3. The aforementioned system data is The data concerns the constraints of the power system, including its operating capacity. The congestion probability prediction device according to claim 1, characterized by the following:

4. The aforementioned system congestion probability prediction unit is: Based on the predicted congestion locations or the predicted congestion probability data, create an operation plan for the equipment within the power system. The congestion probability prediction device according to claim 1, characterized by the following:

5. The aforementioned system congestion probability prediction unit is: Based on the aforementioned weather forecast data, demand data, and grid data, predict the electricity price for each region or location within the power grid. The congestion probability prediction device according to claim 1, characterized by the following:

6. The aforementioned power system data input unit is To acquire the aforementioned demand data as a power consumption plan for multiple specific large consumers, The congestion probability prediction device according to claim 1, characterized by the following:

7. The equipment within the aforementioned power system is It includes at least one of the following: power generation equipment, loads, and energy storage equipment. The aforementioned system congestion probability prediction unit is: To create an operating plan for the load or the energy storage equipment such that the amount of power generation control for the power generation equipment is reduced, The congestion probability prediction device according to claim 4, characterized by the following:

8. The aforementioned system congestion probability prediction unit is: If there are multiple aggregators operating power generation equipment, loads, or energy storage equipment together within the same power system, the operation plan is to be distributed among the multiple aggregators. The congestion probability prediction device according to claim 4, characterized by the following:

9. It includes a power generation plan data input unit that acquires power generation plans from power generation operators or the aggregators, The aforementioned system congestion probability prediction unit is: Based on the power generation plan obtained, predict the congested areas and congestion probability data of the power grid. The congestion probability prediction device according to claim 8, characterized by the following:

10. A method for predicting the probability of congestion in a power grid, using a congestion probability prediction device, The weather forecast data input unit of the aforementioned congestion probability prediction device is: Obtain weather forecast data for a predetermined number of days from the present, The power system data input unit of the aforementioned congestion probability prediction device is: Demand data for electricity demand a predetermined number of days later, based on the present, is acquired, and system data for the aforementioned power grid a predetermined number of days later, based on the present, The system congestion probability prediction unit of the congestion probability prediction device is Based on the weather forecast data, demand data, and grid data, the congestion locations and congestion probability data of the power grid a predetermined number of days later, with the present as the reference point, The aforementioned congestion probability data is, This includes the congestion probability of the power system, the amount of power generation controlled by the power system, and the probability distribution of the amount of power generation controlled. The aforementioned system congestion probability prediction unit is: The predicted congestion locations and the predicted congestion probability data are output, For each time period, the most probable predicted output after power generation control, and the ratio of the most probable predicted output after power generation control to the predicted output before power generation control are output. Furthermore, for each time period when the ratio is not 100%, the maximum and minimum values ​​of the predicted output after power generation control will be output. A congestion probability prediction method characterized by the following.

Images