Management device, planning method, and program

Abstract

This system aims to prevent significant losses to the profits of renewable energy generators, even if their power generation forecasts or electricity sales price forecasts are inaccurate. [Solution] The management device of the embodiment includes, for each power generation system having a renewable energy source and a storage battery charged from the renewable energy source, a control unit that creates a battery charge / discharge plan that maximizes the sum of predicted values ​​of electricity sales revenue and imbalance settlement, based on upper and lower limits of the predicted power generation value created from at least a first predicted power generation value, a second predicted power generation value, and a third predicted power generation value, a predicted electricity sales price pattern created from at least a first predicted electricity sales price value, a second predicted electricity sales price value, and a third predicted electricity sales price pattern created from at least a first predicted imbalance price value, a second predicted imbalance price value, and a third predicted imbalance price pattern.

Description

[Technical Field]

[0001] Embodiments of the present invention relate to a management device, a plan creation method, and a program. [Background technology]

[0002] In recent years, there has been progress in the development of virtual power plants (VPPs), which remotely and centrally control distributed energy resources (DERs) and use multiple power generators as a power generation balancing group.

[0003] Furthermore, renewable energy generators such as solar power generation equipment and wind turbines, as well as power generation systems equipped with renewable energy generators and storage batteries, are sometimes used as DERs (Data Energy Regeneration Systems). For example, there are systems that predict the amount of electricity output by a power generation system equipped with storage batteries and solar power generation equipment, and notify administrators such as power companies that manage the entire power grid in advance. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Patent No. 7173896 [Overview of the project] [Problems that the invention aims to solve]

[0005] Furthermore, regarding operational support devices configured to propose responses from power plants to output curtailment instructions issued to maintain the balance of power supply and demand in the power grid, there is conventional technology for creating operational plans for power plants that maximize the profits of renewable energy power generators.

[0006] However, this conventional technology creates operational plans based on the assumption that the calculated power generation forecasts and electricity sales price forecasts are correct. If the forecasts are wrong, the profits of renewable energy power generators could be significantly harmed.

[0007] Furthermore, there are conventional techniques that take into account the uncertainty of renewable energy generation forecasts, create multiple generation patterns that account for the possibility of incorrect forecasts, and then create battery operation plans and generation plans that maximize expected profits by adding imbalance charges to electricity sales revenue.

[0008] However, this conventional technology does not take into account the uncertainty in predicting electricity sales prices and imbalance prices, so if the predictions are wrong, renewable energy power generators' profits could be significantly harmed.

[0009] Therefore, the present invention has been made in view of the above circumstances, and aims to provide a management device, a planning method, and a program that can prevent a situation in which the profits of renewable energy power generators are greatly harmed, even if the power generation forecast or the electricity sales price forecast for renewable energy generators is inaccurate. [Means for solving the problem]

[0010] The management device of the embodiment provides, for each power generation system having a renewable energy power source and a storage battery charged by the renewable energy power source, a first power generation forecast value which is a predicted value of the amount of power generated by the renewable energy power source, a second power generation forecast value which is a predicted value higher than the first power generation forecast value, a third power generation forecast value which is a predicted value lower than the first power generation forecast value, a first electricity sales price forecast value which is a predicted value of the electricity sales price in the wholesale electricity market for which the power generated by the power generation system is sold, a second electricity sales price forecast value which is higher than the first electricity sales price forecast value, a third electricity sales price forecast value which is lower than the first electricity sales price forecast value, a first imbalance price forecast value which is a predicted value of the unit price of the imbalance charge that arises from the difference between the power generation plan value and the power generation actual value of the power generation system, and a forecast value higher than the first imbalance price forecast value. The system includes a battery charge / discharge plan and a power generation plan for the power generation system, which take a second imbalance unit price forecast value and a third imbalance unit price forecast value which is lower than the first imbalance unit price forecast value as inputs, and based on upper and lower limits of the power generation forecast value created from at least the first power generation forecast value, the second power generation forecast value and the third power generation forecast value, a power sales price forecast value pattern created from at least the first power generation forecast value, the second power sales price forecast value and the third power sales price forecast value, and an imbalance unit price forecast value pattern created from at least the first imbalance unit price forecast value, the second imbalance unit price forecast value and the third imbalance unit price forecast value as inputs. [Brief explanation of the drawing]

[0011] [Figure 1] Figure 1 is an overall diagram of the power system according to the first embodiment. [Figure 2] Figure 2 shows the functional configuration of the control device and power generation system according to the first embodiment. [Figure 3] Figure 3 is an explanatory diagram illustrating an example of calculating a predicted electricity sales price pattern. [Figure 4] Figure 4 is an explanatory diagram illustrating an example of calculating the upper and lower limits for wind power generation forecasts. [Figure 5] FIG. 5 is a diagram showing an example of a power plan (battery charge / discharge plan and power generation plan) and operation results in the prior art. [Figure 6] FIG. 6 is a diagram showing an example of a power plan (battery charge / discharge plan and power generation plan) and operation results in the first embodiment. [Figure 7] FIG. 7 is a flowchart showing the processing flow by the management device of the first embodiment. [Figure 8] FIG. 8 is a diagram showing the functional configuration of the management device and the power generation system of the second embodiment. [Figure 9] FIG. 9 is a flowchart showing the processing flow by the management device of the second embodiment. MODE FOR CARRYING OUT THE INVENTION

[0012] Hereinafter, embodiments (first embodiment, second embodiment) of the management device, the power plan creation method, and the program of the present invention will be described with reference to the drawings. In the following, the time refers to two types: a case where it indicates an instant of time and a case where it indicates the time width of a unit time. The unit time is, for example, 30 minutes.

[0013] (First Embodiment) FIG. 27 is an overall configuration diagram of the power system S of the first embodiment. The power system S includes an aggregator 1, a consumer 4, a wholesale power market 5, a wide-area organization 6, and a plurality of power plants 8.

[0014] The wide-area organization 6 is a power wide-area operation promotion organization. To ensure a stable power supply, it monitors the national power supply and demand situation and the operation status of the power grid 2 for 24 hours, and manages the power supply and demand of grid operators. All grid users (power generation companies, retail electricity companies, etc.) using the power grid 2 are required to submit power generation and demand plans (annual, monthly, weekly, previous day) to the wide-area organization 6. The wide-area organization 6 checks the consistency of these plans and transfers them to grid operators.

[0015] Aggregator 1 is a business operator that manages multiple power plants 8 together. In other words, the multiple power plants 8 function as a power generation BG (balancing group) 3.

[0016] Aggregator 1 predicts the power generation of each of the multiple power plants 8 and sells the electricity on the wholesale electricity market 5, as well as directly trading electricity with consumers 4. Aggregator 1 also contributes to load leveling of the power grid 2 by controlling the power generation of each of the power plants 8 based on their power generation schedules.

[0017] Aggregator 1 is equipped with a control device 10. The control device 10 creates a power generation plan for the power plant 8 and reports it to the regional organization 6 in advance by a predetermined date. For example, the control device 10 reports the power generation plan for the following day to the regional organization 6 the day before, and also reports the revised power generation plan to the regional organization 6 on the day of, if necessary.

[0018] Power plant 8 includes, for example, a power generation system 20, a power plant 21, a storage battery 22, etc. The power generation system 20 includes a battery 25 and either a PV (photovoltaics) system 26 or a wind turbine 31. The PV 26 is a photovoltaic power generation device and is an example of a renewable energy source. The wind turbine 31 is an example of a renewable energy source.

[0019] The power generation system 20 can be configured to include renewable energy sources. Therefore, instead of the PV26 or wind turbine 31, the power generation system 20 may be configured to include power generation devices that utilize naturally occurring environmental resources such as geothermal energy and water. In the following explanation, we will mainly use the case where the power generation system 20 is equipped with a wind turbine 31 as an example.

[0020] The battery 25 of the power generation system 20 is charged by the wind turbine 31. In other words, the amount of power generated by the power generation system 20 is controlled by controlling the charging and discharging of the battery 25. Here, power generation is a concept that also includes the discharge of the battery 25.

[0021] In this embodiment, we will describe as an example a configuration in which the management device 10 creates charge and discharge plans for each of the batteries 25 of one or more power generation systems 20 included in the power generation BG3 (hereinafter also referred to as "battery charge and discharge plans," "battery plans," etc.) and power generation plans for the power generation systems 20.

[0022] The management device 10 is a PC (Personal Computer) or the like, and its hardware configuration utilizes a standard computer equipped with a CPU (Central Processing Unit), memory, an HDD (Hard Disk Drive), a communication interface (I / F), a display device such as a display, and an input device such as a keyboard or mouse.

[0023] Figure 2 shows the functional configuration of the control device 10 and power generation system 20 according to the first embodiment.

[0024] First, let me describe the power generation system 20. The power generation system 20 includes an electricity meter 23 and a power generation unit 24.

[0025] First, let me describe the power generation system 20. The power generation system 20 includes an electricity meter 23 and a power generation unit 24.

[0026] The energy meter 23 measures the amount of electricity output from the power generation unit 24. In other words, the energy meter 23 measures the power generation system output, which is the amount of electricity generated by the power generation system 20. The power generation system output is the amount of electricity supplied from the power generation system 20 to the power grid 2. The energy meter 23 measures the power generation system output every unit of time.

[0027] The unit time is the management time managed by the power system S. In this embodiment, a configuration in which the unit time is 30 minutes will be described as an example.

[0028] The power generation unit 24 comprises a storage battery 25, a wind turbine generator 31, and a control unit 27. As described above, in this embodiment, the storage battery 25 is charged by the wind turbine generator 31. The control unit 27 controls the charging and discharging of the storage battery 25 based on the charge / discharge plan and power generation plan received from the management device 10. Details of this control will be described later. The control unit 27 also outputs the amount of wind power generated by the wind turbine generator 31, the amount of charge / discharge power of the storage battery 25, and the battery SoC (State of Charge), which is the charge rate of the storage battery 25, to the management device 10 at regular intervals. Note that this information may also be output to the management device 10 from the energy meter 23, the storage battery 25, and the wind turbine generator 31, respectively.

[0029] The management device 10 will now be described. The management device 10 comprises a communication unit 11, an input unit 12, a display unit 13, a storage unit 14, and a control unit 15. Each of the units 11 to 15 is connected to the others via a bus or the like, as shown in the figure.

[0030] The communication unit 11 communicates with the wholesale electricity market 5, the wide-area organization 6, the forecasting system 7, and the power generation system 20. The input unit 12 is an input device such as a keyboard that accepts user input. The display unit 13 is a display that shows various kinds of information. The storage unit 14 stores various kinds of information. The storage unit 14 is, for example, an HDD or memory.

[0031] The control unit 15 is an arithmetic unit that performs information processing. The control unit 15 includes a prediction distribution creation unit 15A, a unit price pattern creation unit 15B, a plan creation unit 15C, a power generation system control unit 15D, a display control unit 15E, and a power generation forecast upper and lower limit calculation unit 15F.

[0032] At least one of the parts 15A to 15F is implemented by, for example, one or more processors. For example, each part 15A to 15F may be implemented by having a processor such as a CPU execute a program, i.e., by software. Alternatively, each part 15A to 15F may be implemented by a dedicated IC (Integrated Circuit) or other processor, i.e., by hardware. Furthermore, each part 15A to 15F may be implemented by using a combination of software and hardware. When multiple processors are used, each processor may implement one of the parts 15A to 15F, or it may implement two or more of the parts 15A to 15F.

[0033] The prediction distribution creation unit 15A takes at least a first predicted power generation value, which is a predicted value of the amount of power generated by the wind turbine 31, a second predicted power generation value, which is a predicted value higher than the first predicted power generation value, and a third predicted power generation value, which is a predicted value lower than the first predicted power generation value, as input and creates a probability distribution of the predicted power generation value for each unit of time by estimating the parameters of a predetermined probability distribution for each unit of time.

[0034] Furthermore, the prediction distribution creation unit 15A takes as input at least a first predicted electricity price, which is a predicted value of the electricity price at which the electricity generated by the power generation system 20 is sold in the wholesale electricity market, a second predicted electricity price that is higher than the first predicted electricity price, and a third predicted electricity price that is lower than the first predicted electricity price, and creates a probability distribution of the predicted electricity price for each unit of time by estimating the parameters of a predetermined probability distribution for each unit of time.

[0035] Furthermore, the prediction distribution creation unit 15A takes as input at least a first imbalance unit price prediction value, which is a predicted value of the unit price of the imbalance charge arising from the difference between the power generation plan value and the power generation actual value of the power generation system 20, a second imbalance unit price prediction value that is a predicted value higher than the first imbalance unit price prediction value, and a third imbalance unit price prediction value that is a predicted value lower than the first imbalance unit price prediction value, and creates a probability distribution of the imbalance unit price prediction value for each unit time by estimating the parameters of a predetermined probability distribution for each unit time.

[0036] For example, the prediction distribution creation unit 15A creates the following distributions from the predicted power generation values ​​of the wind turbine 31, the predicted electricity sales price (predicted electricity sales price), and the predicted imbalance price (predicted imbalance price): Dpow(t) (where t is time identification information), Djepx(t) (a distribution of the predicted electricity sales price), and Dimbalance(t) (a distribution of the predicted imbalance price) for each unit of time.

[0037] The predicted power generation values ​​for the wind turbine 31 refer to a representative predicted value Ppow(t) (first predicted power generation value) of the amount of electricity generated by the wind turbine 31 per unit time, and two or more other predicted values ​​Ppow,k(t) (where k is the prediction index) (second predicted power generation value, third predicted power generation value, etc.). In the following, the predicted power generation values ​​for the wind turbine 31 may be referred to as wind power generation prediction values.

[0038] One method for obtaining predicted power generation values ​​for the wind turbine 31 is to obtain three or more predicted power generation values ​​for the wind turbine 31 from a prediction system 7 or the like. Alternatively, three or more predicted wind power generation values ​​can be obtained from the input unit 12 or the memory unit 14 or the like. Furthermore, predicted wind power generation values ​​for each unit of time may be derived based on measured values ​​of wind power generation from the wind turbine 31 obtained from the power generation system 20, as well as meteorological information such as sunshine amount.

[0039] The predicted electricity sales price is a representative predicted value Cjepx(t) (first predicted electricity sales price) and two or more other predicted values ​​Cjepx,k(t) (where k is the prediction index) for the electricity sales price per unit time of the electricity output from the power generation system 20 (e.g., second predicted electricity sales price, third predicted electricity sales price).

[0040] The predicted electricity selling price may be any of the following: a predicted transaction price in the wholesale electricity market 5, or a transaction price predetermined by contract. In this embodiment, the form in which the electricity selling price is a predicted transaction price for each unit of time will be explained as an example. One method for obtaining the predicted electricity selling price is to obtain three or more predicted electricity selling price values ​​from a prediction system 7 or the like.

[0041] The imbalance unit price forecast values ​​are a representative forecast value Cimbalace(t) (first imbalance unit price forecast value) for multiple costs arising from the difference between the forecast and actual values ​​of electricity output per unit time from the power generation system 20, and two or more other forecast values ​​Cimbalace,k(t) (where k is the forecast index) (second imbalance unit price forecast value, third imbalance unit price forecast value, etc.).

[0042] One method for obtaining predicted imbalance unit prices is to obtain three or more predicted imbalance unit prices from a prediction system 7, for example. The predicted imbalance unit prices may be any of the following: predicted transaction prices in the wholesale electricity market 5, or transaction prices predetermined by contract. In this embodiment, the form in which the imbalance unit price is a predicted value for each unit time will be explained as an example.

[0043] The predictive distribution generation unit 15A is equivalent to estimating the parameters a(t) and b(t) (t=1..T) of the Johnson SB distribution given by the following equation (1). The Johnson SB distribution is a distribution that can be used in an asymmetric form while setting upper and lower limits, based on the normal distribution.

[0044] The parameters a(t) and b(t) are determined for each unit time for the wind power generation forecast, the electricity sales price forecast, and the imbalance price forecast, respectively. For example, for wind power generation forecasts, using a representative wind power generation forecast Ppow(t) and other wind power generation forecasts Ppow,k(t) (k=1..m) as input x, the parameters apow(t) and bpow(t) can be determined by maximum likelihood estimation, and the distribution Dpow(t) of power generation forecasts for each unit time can be derived.

[0045] Similarly, the parameters ajepx(t), bjepx(t) and the probability distribution Djepx(t) for a typical predicted electricity sales price Cjepx(t) per unit time t, and for other predicted electricity sales price Cjepx,k(t) (k=1..m), can also be determined. Furthermore, the parameters aimbalance(t), biimbalance(t) and the probability distribution Dimbalance(t) for a typical predicted imbalance price Cimbalance(t) and for other predicted imbalance price Cimbalance,k(t) (k=1..m) can also be determined.

[0046]

number

[0047] Here, Θ(·) represents the cumulative distribution function of the standard normal distribution, and ln represents the natural logarithm.

[0048] The unit price pattern creation unit 15B creates multiple electricity sales price prediction value patterns by extracting electricity sales price prediction values ​​from a probability distribution of electricity sales price prediction values ​​for each unit of time, by repeating trials according to the probability values ​​of the said probability distribution for each unit of time.

[0049] Furthermore, the unit price pattern creation unit 15B creates multiple imbalance unit price prediction value patterns by extracting imbalance unit price prediction values ​​from the probability distribution of imbalance unit price prediction values ​​for each unit of time, by repeatedly conducting trials according to the probability values ​​of the said probability distribution for each unit of time.

[0050] For example, the unit price pattern creation unit 15B creates at least two or more electricity sales price prediction patterns 40 based on the probability distribution Djepx(t)(t=1..T) of the electricity sales price prediction values ​​created by the prediction distribution creation unit 15A, and at least two or more imbalance price prediction patterns 50 based on the probability distribution Dimbalance(t)(t=1..T) of the imbalance price prediction values.

[0051] A unit price pattern is a pattern that represents the time progression of the predicted electricity sales price and the predicted imbalance price on the day the power generation plan is executed. In other words, a unit price pattern is the predicted electricity sales price Cjepx(n,t) and the predicted imbalance price Cimbalance(n,t) for each time t on the day the power generation plan is executed. n is the sequential number of the pattern (it is desirable for n to be 100 or more).

[0052] Figure 3 is an explanatory diagram of 40 examples of electricity sales price prediction patterns. In Figure 3, the horizontal axis represents time t, and the vertical axis represents the electricity sales price.

[0053] The unit price pattern creation unit 15B, for example, derives a random percentile value according to the distribution Djepx(t) (symbols G1, G2, etc. in Figure 3(a)) having ajepx(t) and bjepx(t) calculated by the prediction distribution creation unit 15A, and inputs this percentile value into the percentile function of the probability density function given by equation (1) to obtain the predicted electricity sales price Cjepx(n,t) (Figure 3(b)).

[0054] By performing this multiple times for all time units, a power sales price prediction pattern 40 can be created. An imbalance price prediction pattern 50 is created in the same way. The more power sales price prediction patterns 40 and imbalance price prediction patterns 50 are created, the more accurate the battery storage plan and power generation plan can be created in the planning unit 15C (details will be described later).

[0055] The power generation forecast upper and lower limit calculation unit 15F calculates the upper and lower limits of the power generation forecast for each unit time based on a predetermined confidence interval setting value, using the probability distribution of the power generation forecast value for each unit time.

[0056] Specifically, the power generation forecast upper and lower limit calculation unit 15F calculates the upper and lower limits for each unit time based on the probability distribution Dpow(t)(t=1..T) of the power generation forecast for the wind turbine generator 31 created by the forecast distribution creation unit 15A.

[0057] Here, Figure 4 is an explanatory diagram illustrating an example of calculating the upper and lower limits of wind power generation forecasts. For example, as shown in Figure 4, by inputting the 90th percentile value and the 10th percentile value as the 90% confidence interval of the forecast into the percentile function of the probability density function given by equation (1), the upper and lower limits of the wind turbine generator 31's power generation forecast, Ppow,max(t) (Figure 4(b)), and the lower limit of the wind turbine generator 31's power generation forecast, Ppow,min(t) (Figure 4(b)), can be calculated.

[0058] Returning to Figure 2, the explanation continues. The planning unit 15C takes the upper and lower limits of the predicted power generation value of the wind turbine 31, the predicted power sales price pattern, and the predicted imbalance price pattern as inputs to create a charge / discharge plan for the battery 25 and a power generation plan for the power generation system 20 that maximize the sum of the predicted power sales revenue and imbalance settlement values.

[0059] For example, the planning unit 15C creates a charge / discharge plan and a power generation plan that maximizes the expected revenue of the power generation system 20, based on the predicted power generation value and electricity sales price of the wind turbine 31, and multiple electricity sales price prediction patterns 40 and multiple imbalance price prediction patterns 50 created by the price pattern creation unit 15C. In this embodiment, the case in which the planning unit 15C creates a charge / discharge plan and a power generation plan for 24 hours (T=48) of the following day will be described as an example.

[0060] In detail, the planning unit 15C calculates the optimal solution to the optimization problem that maximizes the objective function F in equation (2) under the constraints shown in equations (3) to (17). By calculating this optimal solution that maximizes the objective function F, the planning unit 15C calculates the charge / discharge plan and power generation plan for each unit time for each power generation system 20.

[0061] The optimization problems given by equations (1) to (18) are known as mixed linear programming problems. The programming unit 15C calculates the optimal solution that maximizes the objective function F in equation (2) using methods such as the simplex method or the interior point method.

[0062]

number

[0063] Here, Cprem(t) is the premium unit price per unit time. Psell(t) is the electricity sales volume. Pover(t) is the excess power amount. Plack(t) is the power shortage amount. Cpenalty is the penalty (charge). Pbuy(t) is the electricity purchase volume.

[0064]

Number

[0065] Here, Ppln(t) is the planned power amount. Ppow'(t) is the correction value of the power generation prediction value of the wind turbine 31. Pout BT (t) is the discharge amount of the storage battery 25. Pin BT (t) is the charge amount of the storage battery 25.

[0066]

Number

[0067] Here, Z JPX (t) is a parameter that takes a value of "1" during power sales and "0" during non-power sales. Pmin JPX is the minimum unit of power sales in the market. Pmax TPO is the maximum power generation amount. Prate TPO (t) is the output suppression rate.

[0068]

Number

[0069] Here, Pout_min BT is the minimum output (discharge) of the storage battery 25. Zout BT (t) is a parameter that takes a value of "1" during discharge and "0" during non-discharge. Pout BT (t) is the discharge amount of the storage battery 25. Pout_max BTThis is the maximum output of battery 25. dt is a parameter for converting the unit from kW to kWh, and is specifically "0.5". Pin_min BT This is the minimum input (charging) of the battery 25. BT (t) is a parameter that takes the value "1" when charging and "0" when not charging. BT (t) is the charge level of battery 25. Pin_max BT This is the maximum input of the battery 25. Wmin BT This is the minimum charge capacity of battery 25 (for example, about 10% of the maximum charge capacity). BT (t) is the remaining charge of battery 25. Wmax BT This is the maximum storage capacity of battery 25. BT This is the charging efficiency. 1 / η BT This is the discharge rate. Pcsp BT This is the power consumed when operating the storage battery 25.

[0070]

number

[0071] Here, Kw(t) is the power generation correction coefficient. Ka is a predetermined constant, for example, around 0.9. Kb is a predetermined constant, for example, around 1.1.

[0072] The distinctive features of this embodiment will now be described. First, the expected value in equation (2) is calculated using sample mean approximation. Specifically, as shown in equation (18), the average of the N predicted electricity sales price values ​​Cjepx(n,t) and the predicted imbalance price value Cimbalance(n,t) obtained by the unit price pattern generation unit 15B is calculated, and this average value is taken as the expected value. It is known that equation (18) coincides with the expected value optimization problem in equation (2) when N is sufficiently large.

[0073]

number

[0074] Next, we will explain power generation forecast correction and imbalance prediction calculation. In this embodiment, equations (13) to (15) allow us to determine the battery charge / discharge plan that maximizes the expected revenue of the power generation system 20, as determined by equation (2) (equation (18)), and simultaneously determine the power generation forecast Ppow'(t) that maximizes the expected profit. Note that equation (15) is a constraint equation to prevent the correction coefficients for adjacent unit times from diverging too much.

[0075] Equations (16) and (17) are predicted values ​​of the imbalance amount at a unit time t. Equation (16) is the predicted value of the maximum surplus imbalance amount, and equation (17) is the predicted value of the maximum deficit imbalance amount. The estimated surplus imbalance amount Pover(t) at a unit time t is calculated by adding the power generation plan value Ppln(t) and the power generation plan (Ppow, max(t) + Pout) which takes into account the charge and discharge amount of the battery 25 in addition to the power generation limit of the wind turbine 31. BT (t)-Pin BT It is the difference between (t) and (t). Also, the estimated imbalance Plack(t) is calculated by adding the charge / discharge amount of the battery 25 to the power generation plan value Ppln(t) and the power generation plan (Ppow,min(t)+Pout) which is calculated by adding the charge / discharge amount of the battery 25 to the lower limit of the power generation amount of the wind turbine 31. BT (t)-Pin BT This is the difference from (t).

[0076] Refer to Figures 5 and 6 to explain examples of calculation results for the conventional technology and this embodiment. Figure 5 shows an example of power planning (battery charge / discharge plan and power generation plan) and operational results in the conventional technology. Figure 6 shows an example of power planning (battery charge / discharge plan and power generation plan) and operational results in the first embodiment.

[0077] In Figure 5(a), the symbol PGP represents the power generation plan. The symbol PP represents the predicted buying and selling price. The sum of the symbols WPF and CP represents the predicted wind power generation value from the wind turbine 31. Of these, the symbol CP is the power allocated to charging the battery 25 in the charge-discharge plan. The symbol DP is the power discharged from the battery 25 in the charge-discharge plan.

[0078] Furthermore, in Figure 6(a), the symbol WP is a typical power generation forecast Ppow(t) for the wind turbine 31, and in this example, it is assumed to be equal to the wind power generation forecast value (the sum of the symbols WPF and CP) for the wind turbine 31 in Figure 5(a). Also, the sum of the symbols WPF' and CP is the optimal power generation forecast P'pow(t) for the symbol WP, which is obtained by modifying the symbol WP so that the expected return of the power generation system 20 is maximized.

[0079] Figures 5(b) and 6(b) show the operational performance. Code PR represents the trading price. Codes WPR and CR are added together to represent the wind power generation performance by the wind turbine 31. Of these, code CR is the power allocated to charging the battery 25. Code DR is the power discharged from the battery 25. Code PGR is the power generation performance.

[0080] Comparing Figure 5(b) and Figure 6(b), Figure 6(b) generally shows that the occurrence of under-imbalances, where the actual power generation (code PGR) is smaller than the power generation plan (code PGP), is suppressed, and the losses of the power generation system 20 are reduced.

[0081] Next, the processing performed by the control device 10 will be explained with reference to Figure 7. Figure 7 is a flowchart showing the processing flow by the control device 10 in the first embodiment. Here, the process of creating a charge / discharge plan and a power generation plan that maximize the objective function F (Equations (2), (18)) will be explained.

[0082] The prediction distribution creation unit 15A obtains at least three predicted power generation values ​​for the wind turbine 31 from the prediction system 7 (S101). Similarly, the prediction distribution creation unit 15A obtains at least three predicted electricity sales prices and at least three predicted imbalance prices from the prediction system 7 (S102, S103).

[0083] Next, the prediction distribution creation unit 15A creates a probability distribution for each prediction based on the representative power generation prediction value Ppow(t) for the wind turbine 31, multiple other power generation prediction values ​​Ppow,k(t) for the wind turbine 31, a representative electricity sales price prediction value Cjepx(t), multiple other electricity sales price prediction values ​​Cjepx,k(t), a representative imbalance price prediction value Cimbalance(t), and multiple other imbalance price prediction values ​​Cimbalance(t) obtained in S101 to S103 (S104).

[0084] Specifically, for each prediction and each time point, we determine the parameters apow(t) and bpow(t) of the probability distribution Dpow(t) for the wind turbine generator 31 power generation prediction which is most likely to match the probability density function of equation (1), the parameters ajepx(t) and bjepx(t) of the probability distribution Djepx(t) for the electricity sales price prediction, and the parameters aimbalance(t) and bimbalance(t) of the probability distribution Dimbalance(t) for the imbalance price prediction.

[0085] Next, in the unit price pattern creation unit 15B, N unit price patterns Cjepx(n,t) and Cimbalance(n,t) (n=1..N) are created for the electricity sales price forecast and the imbalance price forecast, respectively (S105).

[0086] Furthermore, the power generation forecast upper and lower limit calculation unit 15F calculates the upper and lower limit values ​​Ppow,min(t) and Ppow,max(t) of the power generation forecast for the wind turbine 31 from the power generation forecast distribution Dpow(t) of the wind turbine 31, for percentile values ​​that are predetermined upper and lower limits of the confidence interval (for example, maximum value 90%, minimum value 10%) (S106).

[0087] Next, the planning unit 15C takes the predicted power generation value Ppow(k,t) of the wind turbine 31, the predicted electricity sales price pattern Cjepx(n,t), the predicted imbalance price pattern Cimbalance(n,t), the initial remaining charge of the battery 25, the charge / discharge efficiency of the battery 25, and the rated output of the power generation system 20 as inputs and calculates the optimal solution to the optimization problem that maximizes the objective function F (equations (2), (18)) under the constraints shown in equations (3) to (17) (S107). By calculating the optimal solution that maximizes the objective function F, the planning unit 15A calculates the charge / discharge plan and power generation plan for each unit time for each power generation system 20.

[0088] Next, the display control unit 15E displays the charge / discharge plan and power generation plan created in S107 on the display unit 13 (S108).

[0089] Next, the power generation system control unit 15D executes the charge / discharge plan and power generation plan created in S107 (S109). For example, the power generation system control unit 15D controls the battery 25 of the power generation system 20 based on the charge / discharge plan and power generation plan created in S107. Then, this routine ends.

[0090] In this way, the management device 10 of the first embodiment makes it possible to avoid situations in which the profits of renewable energy generators, such as wind turbines 31, are significantly harmed even if the power generation forecast or the electricity sales price forecast for renewable energy generators are incorrect. Specifically, this is as follows.

[0091] The planning unit 15C takes the predicted power generation value of the wind turbine 31 obtained from the prediction system 7, the upper and lower limits of the predicted power generation value of the wind turbine 31 created by the power generation upper and lower limit prediction calculation unit 15F, and the unit price pattern created by the unit price pattern creation unit 15B as input to obtain the optimal solution that maximizes the objective function F. This makes it possible to create a battery storage plan and a power generation plan that take prediction errors into account without performing sequential optimization, for example, as in the technology of Japanese Patent Publication No. 7048797 (Patent Document 2). As a result, the processing time to obtain the battery storage plan and power generation plan is significantly reduced. In contrast, conventional technology had the problem of long processing times because, for example, multiple power generation patterns were created, and an operation plan was created for each to select the optimal one.

[0092] Furthermore, unlike the technology described in Patent Document 2, it is possible to create battery storage plans and power generation plans that take into account prediction errors in the electricity selling price.

[0093] (Second Embodiment) Next, a second embodiment will be described. Matters similar to those in the first embodiment will be omitted from the explanation as appropriate. Figure 8 shows the functional configuration of the control device 20 and power generation system 20 of the second embodiment.

[0094] In the second embodiment, multiple confidence intervals can be set for calculating the upper and lower limits of the power generation forecast for the wind turbine 31, which were previously fixed as one in the first embodiment. This allows for the selection of a confidence interval that maximizes the expected return of the power generation system 20.

[0095] The control unit 15 in the management device 10 further includes an optimal plan selection unit 15G. Based on a pre-provided list of multiple confidence interval settings (length = L), the optimal plan selection unit 15G sequentially provides confidence intervals as input to the power generation forecast upper and lower limit calculation unit 15F, compares the value of the calculation result Fs (s=1..L) of the plan creation unit 15C in each confidence interval, and outputs a battery charge / discharge plan and a power generation plan that maximize Fs.

[0096] Figure 9 is a flowchart showing the processing flow by the control device 10 of the second embodiment.

[0097] The prediction distribution creation unit 15A obtains at least three predicted power generation values ​​for the wind turbine 31 from the prediction system 7 (S201). Similarly, the prediction distribution creation unit 15A obtains at least three predicted electricity sales prices and at least three predicted imbalance prices from the prediction system 7 (S202, S203).

[0098] Next, the prediction distribution creation unit 15A creates a probability distribution for each prediction based on the representative power generation prediction value Ppow(t) for the wind turbine 31, multiple other power generation prediction values ​​Ppow,k(t) for the wind turbine 31, a representative electricity sales price prediction value Cjepx(t), multiple other electricity sales price prediction values ​​Cjepx,k(t), a representative imbalance price prediction value Cimbalance(t), and multiple other imbalance price prediction values ​​Cimbalance(t) obtained in S201 to S203 (S104).

[0099] Specifically, for each prediction and each time point, we determine the parameters apow(t) and bpow(t) of the probability distribution Dpow(t) for the wind turbine generator 31 power generation prediction which is most likely to match the probability density function of equation (1), the parameters ajepx(t) and bjepx(t) of the probability distribution Djepx(t) for the electricity sales price prediction, and the parameters aimbalance(t) and bimbalance(t) of the probability distribution Dimbalance(t) for the imbalance price prediction.

[0100] Next, in the unit price pattern creation unit 15B, N unit price patterns Cjepx(n,t) and Cimbalance(n,t) (n=1..N) are created for the electricity sales price forecast and the imbalance price forecast, respectively (S205).

[0101] Next, the control unit 15 sets the confidence interval list to the optimal plan selection unit 15G (S206), and pops the first confidence interval from the confidence interval list (S207).

[0102] Next, for the confidence interval popped in S207, the power generation forecast upper and lower limit calculation unit 15F calculates the upper and lower limit values ​​Ppow,min(t) and Ppow,max(t) of the power generation forecast for the wind turbine 31 from the power generation forecast distribution Dpow(t) of the power generation forecast values ​​for the wind turbine 31, corresponding to the upper and lower limit percentile values ​​(S208).

[0103] Next, the planning unit 15C takes the predicted power generation value Ppow(k,t) of the wind turbine 31, the predicted electricity sales price pattern Cjepx(n,t), the predicted imbalance price pattern Cimbalance(n,t), the initial remaining charge of the battery 25, the charge / discharge efficiency of the battery 25, and the rated output of the power generation system 20 as inputs, and calculates the optimal solution Fs of the optimization problem that maximizes the objective function F (equations (2), (18)) under the constraints shown in equations (3) to (17), and the optimal plan selection unit 15G sets the confidence interval value F S Remember them as a pair (S209).

[0104] Next, the optimal plan selection unit 15G performs a determination process that returns to S207 if there is a confidence interval list remaining (Yes in S210), and proceeds to S211 if there is no list remaining (No in S210).

[0105] In S211, the optimal plan selection unit 15G selects the largest Fs from the stored Fs and calculates it as the optimal solution.

[0106] Next, the display control unit 15E displays the charge / discharge plan and power generation plan created in S211 on the display unit 13 (S212).

[0107] Next, the power generation system control unit 15D executes the charge / discharge plan and power generation plan created in S211 (S213). For example, the power generation system control unit 15D controls the battery 25 of the power generation system 20 based on the charge / discharge plan and power generation plan selected in S211. Then, this routine ends.

[0108] Thus, according to the second embodiment, multiple confidence intervals can be set for calculating the upper and lower limits of the power generation forecast for the wind turbine 31, and the confidence interval that maximizes the expected return of the power generation system 20 can be selected.

[0109] For example, a wider confidence interval is not always better. A wider confidence interval can lead to larger surplus or deficit imbalances, for instance.

[0110] In the embodiment described above, the program for executing the information processing is configured as a module that includes each of the multiple functional units. In actual hardware, for example, the CPU reads the information processing program from ROM (Read Only Memory) or HDD and executes it, thereby loading each of the multiple functional units onto RAM (Random Access Memory), and each of the multiple functional units is generated on RAM (main memory). It is also possible to implement some or all of the multiple functional units using dedicated hardware such as ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array).

[0111] While embodiments of this disclosure have been described above, these embodiments are provided as examples only and are not intended to limit the scope of this disclosure. The novel embodiments described above can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The embodiments described above are included within the scope or spirit of this disclosure and within the scope of the invention and its equivalents as described in the claims.

[0112] For example, some or all of the functions of the management device 10 may be implemented by a cloud computing system. [Explanation of symbols]

[0113] 1...Aggregator, 2...Power grid, 3...Power generation BG, 4...Consumer, 5...Wholesale electricity market, 6...Wide-area organization, 7...Forecasting system, 8...Power plant, 10...Management device, 11...Communication unit, 12...Input unit, 13...Display unit, 14...Storage unit, 15A...Forecast distribution creation unit, 15B...Unit price pattern creation unit, 15C...Plan creation unit, 15D...Power generation system control unit, 15E...Display control unit, 15F...Power generation forecast upper and lower limit calculation unit, 20...Power generation system, 21...Power plant, 22...Storage battery, 23...Energy meter, 24...Power generation unit, 25...Storage battery, 26...PV, 27...Control unit, 28...Power generation unit, 29...Control unit, 31...Wind turbine, S...Power system

Claims

1. For each power generation system having a renewable energy source and a storage battery charged by the renewable energy source, A first predicted power generation value is a predicted value of the amount of electricity generated by the renewable energy source, a second predicted power generation value is a predicted value higher than the first predicted power generation value, and a third predicted power generation value is a predicted value lower than the first predicted power generation value. A first predicted electricity price is a predicted value of the electricity price at which the electricity generated by the power generation system is sold in the wholesale electricity market, a second predicted electricity price is higher than the first predicted electricity price, and a third predicted electricity price is lower than the first predicted electricity price. A first imbalance unit price forecast value is a predicted value of the unit price of the imbalance charge that arises from the difference between the planned power generation value and the actual power generation value of the power generation system, a second imbalance unit price forecast value is a predicted value higher than the first imbalance unit price forecast value, and a third imbalance unit price forecast value is a predicted value lower than the first imbalance unit price forecast value. Using as input, An upper and lower limit of the power generation forecast, created from at least the first power generation forecast, the second power generation forecast, and the third power generation forecast, A pattern of predicted electricity sales prices created from at least a first predicted electricity sales price, the second predicted electricity sales price, and the third predicted electricity sales price, Based on an imbalance unit price forecast pattern created from at least the first imbalance unit price forecast, the second imbalance unit price forecast, and the third imbalance unit price forecast, A management device comprising a battery charge / discharge plan that maximizes the sum of electricity sales revenue and predicted imbalance settlement values, and a control unit that creates a power generation plan for the power generation system.

2. A prediction distribution creation unit that takes at least the first predicted power generation value, the second predicted power generation value, and the third predicted power generation value as input and performs parameter estimation of a predetermined probability distribution for each unit of time to create a probability distribution of the predicted power generation value for each unit of time, A power generation forecast upper and lower limit calculation unit calculates the upper and lower limits of the power generation forecast for each unit time based on a predetermined confidence interval setting value from the probability distribution of the power generation forecast value for each unit time, The control device according to claim 1, comprising:

3. A prediction distribution creation unit creates a probability distribution of the predicted electricity price for each unit of time by taking at least a first predicted electricity price, the second predicted electricity price, and the third predicted electricity price as input and performing parameter estimation of a predetermined probability distribution for each unit of time. A prediction distribution creation unit creates a probability distribution of the predicted imbalance unit price for each unit time by taking at least the first predicted imbalance unit price, the second predicted imbalance unit price, and the third predicted imbalance unit price as input and performing parameter estimation of a predetermined probability distribution for each unit time. From the probability distribution of the predicted electricity selling price for each unit of time, a plurality of patterns of predicted electricity selling prices are created by repeatedly performing trials according to the probability values ​​of the said probability distribution for each unit of time, thereby extracting predicted electricity selling price values. A unit price pattern creation unit creates multiple imbalance unit price prediction value patterns by extracting imbalance unit price prediction values ​​from the probability distribution of the imbalance unit price prediction values ​​for each unit of time, by repeating trials according to the probability values ​​of the said probability distribution for each unit of time, The control device according to claim 1, comprising:

4. The management device according to claim 1, comprising: a planning unit that takes as input the upper and lower limits of the predicted power generation value of the renewable energy power source, the predicted power sales price pattern, and the predicted imbalance price pattern, a charge / discharge plan for the storage battery that maximizes the sum of the predicted values ​​of power sales revenue and imbalance settlement, and a planning unit that creates the power generation plan for the power generation system.

5. A method for creating a plan using a management device equipped with a control unit, The control unit, For each power generation system having a renewable energy source and a storage battery charged by the renewable energy source, A first predicted power generation value is a predicted value of the amount of electricity generated by the renewable energy source, a second predicted power generation value is a predicted value higher than the first predicted power generation value, and a third predicted power generation value is a predicted value lower than the first predicted power generation value. A first predicted electricity price is a predicted value of the electricity price at which the electricity generated by the power generation system is sold in the wholesale electricity market, a second predicted electricity price is higher than the first predicted electricity price, and a third predicted electricity price is lower than the first predicted electricity price. A first imbalance unit price forecast value is a predicted value of the unit price of the imbalance charge that arises from the difference between the planned power generation value and the actual power generation value of the power generation system, a second imbalance unit price forecast value is a predicted value higher than the first imbalance unit price forecast value, and a third imbalance unit price forecast value is a predicted value lower than the first imbalance unit price forecast value. Using as input, An upper and lower limit of the power generation forecast, created from at least the first power generation forecast, the second power generation forecast, and the third power generation forecast, A pattern of predicted electricity sales prices created from at least a first predicted electricity sales price, the second predicted electricity sales price, and the third predicted electricity sales price, Based on an imbalance unit price forecast pattern created from at least the first imbalance unit price forecast, the second imbalance unit price forecast, and the third imbalance unit price forecast, A method for creating a charging and discharging plan for a battery and a power generation plan for a power generation system, which maximize the sum of the predicted values ​​of electricity sales revenue and imbalance settlement.

6. The computer, which is the management device, For each power generation system having a renewable energy source and a storage battery charged by the renewable energy source, A first predicted power generation value is a predicted value of the amount of electricity generated by the renewable energy source, a second predicted power generation value is a predicted value higher than the first predicted power generation value, and a third predicted power generation value is a predicted value lower than the first predicted power generation value. A first predicted electricity price is a predicted value of the electricity price at which the electricity generated by the power generation system is sold in the wholesale electricity market, a second predicted electricity price is higher than the first predicted electricity price, and a third predicted electricity price is lower than the first predicted electricity price. A first imbalance unit price forecast value is a predicted value of the unit price of the imbalance charge that arises from the difference between the planned power generation value and the actual power generation value of the power generation system, a second imbalance unit price forecast value is a predicted value higher than the first imbalance unit price forecast value, and a third imbalance unit price forecast value is a predicted value lower than the first imbalance unit price forecast value. Using as input, An upper and lower limit of the power generation forecast, created from at least the first power generation forecast, the second power generation forecast, and the third power generation forecast, A pattern of predicted electricity sales prices created from at least a first predicted electricity sales price, the second predicted electricity sales price, and the third predicted electricity sales price, Based on an imbalance unit price forecast pattern created from at least the first imbalance unit price forecast, the second imbalance unit price forecast, and the third imbalance unit price forecast, A program for causing a battery charge / discharge plan and a power generation plan for the power generation system to function as a control unit that maximizes the sum of the predicted values ​​of electricity sales revenue and imbalance settlement.

Images