JP7572911B2 - Power system operation plan generating device and power system operation plan generating method - Google Patents

Description

The present invention relates to a power system operation plan generating device and a power system operation plan generating method.

In power grids, renewable energy sources that do not emit greenhouse gases are increasingly being introduced. The output of solar and wind power, which are typical renewable energy sources, fluctuates depending on weather conditions. In addition, with the spread of electric vehicles, the number of quick charging stations is increasing and their capacity is increasing, and this demand for electricity may affect the power grid. There is a need for system operation that takes into account uncertain parameters (hereinafter referred to as uncertain parameters), such as the output of renewable energy sources and the charging demand for electric vehicles.

On the other hand, there are two types of variables in system operation: variables that must be determined in advance, such as the day before (hereinafter referred to as operation planning variables), and variables that can be determined in real time (hereinafter referred to as same-day control variables). For example, operation planning variables in a power distribution system include the tap position of tap adjustment equipment (e.g., LRT: Load Ratio control Transformer, SVR: Step Voltage Regulator), and the reserved amount of adjustment margin of distributed energy resources (DER: Derived Energy Resource) owned by consumers, such as private generators, storage batteries, and electric vehicles. In addition, examples of control variables for the day include SVC (Static Var Compensator), distributed energy resources DER such as grid storage batteries owned by grid operators, and the amount of the command to activate the adjustment margin of distributed energy resources DER owned by reserved consumers.

The first method that comes to mind for determining operational planning variables is to assume that predictions of uncertain parameters will be correct and select a combination of values that will optimize system operation KPIs (e.g., violations of voltage tolerances and line capacity, and response costs). This method does not take into account cases where predictions are incorrect, and depending on the fluctuation pattern of the uncertain parameters, there is a possibility that system operation KPIs may deteriorate drastically (e.g., violations of constraints such as voltage tolerances and line capacity, and extreme costs due to additional responses, etc.).

Therefore, it is necessary to determine operational planning variables so that system operation KPIs (e.g., voltage tolerance range, line capacity violations, response costs) do not deteriorate excessively (hereafter referred to as "robust") even if the predictions of uncertain parameters fluctuate.

As one technique for determining operational planning variables aimed at robust grid operation, Patent Literature 1 describes an "operational planning device, operation planning method, and operation planning program." In Patent Literature 1, in order to handle the uncertainty of the output of renewable energy sources in an optimization problem for solving generator start-stop plans, a method is used to express this uncertainty using multiple scenarios.

JP 2016-63609 A

In general, increasing the number of scenarios increases the calculation time for the optimization problem. For this reason, for example, when there is a limit to the calculation time, it is not possible to use a sufficient number of scenarios, which can lead to a problem that the reliability of the obtained solution decreases.

In order to solve the above problem, Patent Document 1 assumes that the optimization problem can be simplified to one that is independent for each time period, with the aim of expressing uncertainty with a small number of scenarios. However, tap adjustment equipment used in distribution systems has constraints that span time periods, such as an upper limit on the number of tap position adjustments per day. For this reason, it is not possible to simplify the optimization problem to one that is independent for each time period.

The present invention has been made in consideration of the above, and aims to provide a power system operation plan generation device and a power system operation plan generation technology that can generate an operation plan that satisfies three requirements: short calculation time, robustness against mispredictions, and consideration of constraints across time periods.

In view of the above, the present invention provides a power system operation plan generating device that takes into account the uncertainty parameters in the power system, determines operation plan variables that should be determined in advance and on-day control variables that can be determined in real time, and performs operation planning for the power system, the power system operation plan generating device being characterized by comprising an operation plan generating unit that obtains information on the operation plan variables and on-day control variables in the power system, and creates a proposed operation plan for the objective function value in the most severe scenario that assumes a severe state of the power system, a most severe scenario generating unit that determines the most severe scenario and the objective function value at that time from fluctuation information on the uncertainty parameters, the proposed operation plan, and the objective function value of the on-day control variables, and a day control amount generating unit that determines the operation plan using system information, demand information, and the most severe scenario of the power system.

The present invention is also a power system operation plan generation method that takes into account uncertain parameters in a power system, determines operation plan variables that should be determined in advance and on-day control variables that can be determined in real time, and performs operation planning for the power system, in which information on the operation plan variables and on-day control variables in the power system are obtained, and a proposed operation plan is created for the objective function value in the most severe scenario that assumes a severe state of the power system. The power system operation planning problem is divided into a main problem of determining an operation plan and a subordinate problem, and the subordinate problem is further divided into a main problem of determining the most severe scenario and a subordinate problem of determining the control amount for the day, thereby performing operation planning for the power system as a solution to the three-layered power system operation planning problem.

The present invention has the advantage of being able to derive an operational plan that satisfies three requirements: short calculation time, robustness to mispredictions, and consideration of constraints across time periods.

FIG. 2 is a diagram showing the processing contents in a calculation unit of the power system operation plan generating device as representative processing function units. 13 is a diagram showing how the operation plan xp is obtained as a solution to the operation plan determination problem. FIG. 11 is a diagram showing an example of uncertain parameter variation information. A diagram showing an example of the most severe scenario. 1 is a process flow of a power system operation plan generation method. FIG. 13 is a diagram showing the difference in effect when the present invention is compared with a conventional method.

Below, an embodiment of the power system operation plan generating device and power system operation plan generating technology according to the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to these embodiments. Moreover, each embodiment can be appropriately combined as long as the processing contents are not contradictory.

In the following description, the power system operation plan generating device will be described in Example 1, the concept of generating a power system operation plan will be described in Example 2, and the method of generating a power system operation plan will be described in Example 3.

In the first embodiment of the present invention, a power system operation plan generating device is described. As an application example, a power system operation plan generating device that formulates an operation plan that is robust against fluctuations in uncertain parameters is described.

First, the configuration of the power system operation plan generating device will be described. Figure 1 shows the processing contents of the calculation unit of the power system operation plan generating device, which is realized using a computer device, as its representative processing function unit.

As shown in FIG. 1, the robust operation plan generating device 10 has an operation plan generating unit 20, a most severe scenario generating unit 30, and a daily control amount generating unit 40. However, in addition to the functional units shown in FIG. 1, the operation plan formulation device 10 may also have, for example, a display unit that displays various information, an input unit that inputs input data, and a communication interface unit that communicates with other terminals.

To carry out its processing, the robust operation plan generation device 10 receives tap adjustment equipment information D1, customer-owned DER information D2, uncertain parameter fluctuation information D3, demand information D4, and system information D5 as external inputs.

The operation plan generation unit 20 in the robust operation plan generation device 10 obtains tap adjustment equipment information D1 and customer-owned DER information D2 as external input data, and also inputs the objective function value ηp and sensitivity information b1 calculated by the most severe scenario generation unit 30. Based on this input data, the operation plan decision problem is generated, and the solved operation plan proposal xp is output to the most severe scenario generation unit 30.

The tap adjustment equipment information D1, which is one of the external inputs used in this case, includes, for example, the equipment type of each tap adjustment equipment (e.g., LRT, SVR, etc.), the installation bus ID, the number of tap stages, the tap width for each stage, the tap reference position, the response speed to control commands, the limit on the number of tap adjustments per day, the equipment cost, and the installation cost.

The consumer-owned DER information D2 from the same external input may include, for example, the type of equipment (e.g., self-generating equipment, electric vehicle, etc.), installed bus ID, facility capacity, and response speed to control commands. The consumer-owned DER information 110 may be input individually for each distributed energy source DER, or the combined value of multiple distributed energy source DERs may be input.

Here, the objective function value η represents the objective function when the most severe scenario determination problem is solved in the most severe scenario generation unit 30, that is, the value of equation (14) described below. Also, the sensitivity information b1 is the amount of change in the objective function value ηp when the proposed operation plan xp is slightly changed.

The proposed operation plan xp is found as a solution to the operation plan determination problem. An example of this will be explained using FIG. 2. Each row in FIG. 2 represents a time step 221, and each column represents a tap adjustment device ID 222 and a customer-owned DER ID 223. For each tap adjustment device ID 222, the tap position 224 at each time step 221 is stored. Furthermore, for each customer-owned DER ID 223, the adjustment margin reservation amount 225 for each time step is stored. Since the adjustment margin reservation amount 225 can take values for both an increase and decrease in demand, it can also take negative values as shown in FIG. 2.

Returning to Figure 1, the most severe scenario generation unit 30 in the power system operation plan generation device 10 inputs uncertain parameter fluctuation information D3 as input data from outside, and also inputs the operation plan proposal x determined by the operation plan generation unit 20, and the objective function value φp ^KPI and sensitivity information b2 determined by the daily control quantity generation unit 40, generates a most severe scenario decision problem based on this input data, outputs the solved most severe scenario up to the daily control quantity generation unit 40, and outputs the objective function value φp ^KPI and the sensitivity information b2 to the most severe scenario generation unit 30.

An example of uncertain parameter fluctuation information D3, which is an external input used in this case, is shown in the graph in Figure 3. In the graph, solar power generation output is considered as an example of an uncertain parameter. The horizontal axis represents the time of day/24 hours, and the vertical axis represents solar power generation output. The portion represented by the gray range is fluctuation information 121 of the solar power generation output. It is assumed that the solar power generation output may fluctuate within this gray range. The range of the above fluctuation information 121 may be determined based on actual values, or may be determined based on a weather forecast, etc.

The objective function value φp ^KPI represents the objective function when the most severe scenario determination problem is solved, that is, the value of equation (15) described later. The sensitivity information b2 is the amount of change in the objective function value φp ^KPI when the most severe scenario u is slightly changed.

An example of the most severe scenario up is shown in the graph in Figure 4. In the graph, solar power generation output is considered as an example of an uncertain parameter. The horizontal axis represents the time of day/24 hours, and the vertical axis represents the solar power generation output. The portion represented in grey is the fluctuation information 121 of the solar power generation output. The thick line 241 represents the most severe scenario up. The most severe scenario up is determined by solving the most severe scenario determination problem. Furthermore, the thick line 241 falls within the grey range 121 at any time step.

Returning to FIG. 1 again, the daily control quantity generation unit 40 in the power system operation plan generation device 10 obtains customer-owned DER information D2, demand information D4, and system information D5 as input data from outside, and also inputs the most severe scenario up determined by the most severe scenario generation unit 30. An daily control quantity determination problem is generated based on these input data, and the solved objective function value φp ^KPI and sensitivity information b2 are output to the most severe scenario generation unit 30.

In addition, the proposed operation plan xp and the objective function value ηp (= theoretical worst value of the system operation KPI) are output from the power system operation plan generating device 10 as the final solution obtained as a result of repeated processing in the operation plan generating unit 20, the most severe scenario generating unit 30, and the daily control amount generating unit 40.

In Example 2, the concept of generating a power system operation plan is explained in detail using mathematical expressions, and how the device shown in FIG. 1 can generate a robust operation plan. Note that in order to clarify the characteristics of the operation planning method according to the present invention, it is best to compare it with a normal operation planning method, so the normal operation planning method will be explained first.

A typical operational planning method is characterized in that the user inputs a set of scenarios that express uncertainty in advance, and an operational plan is obtained by solving a mathematical programming problem whose objective function is to optimize the system operation KPIs. To be more specific, using a mathematical example, an example of the mathematical programming problem is expressed in the following equations (1) and (2).

In formulas (1) and (2), lowercase s represents a scenario, and uppercase S represents a set of input scenarios. Lowercase t represents time, uppercase T represents a set of times t included in the planning period, and φ ^KPI represents a system operation KPI. The subscripts of min represent decision variables, x represents a vector of operation planning variables (e.g., tap position control amount, customer-owned DER adjustment margin reservation amount), and y represents a vector of the day's control variables (e.g., reserved adjustment margin activation command amount). Note that these are vectors, and formula (2) is a constraint formula showing the system constraint conditions for determining formula (1).

Here, the operational plan variable vector x needs to be determined at a stage where there is still a risk that the uncertain parameters will vary from the predicted values, so the same operational plan x is determined for all scenarios S entered in advance by the user. In contrast, the current day control variable vector y is a variable that can be adjusted in real time after the uncertain parameters have become clear (for example, after the actual photovoltaic power generation output has been measured), so a different current day control y is determined for each scenario s. In normal operational planning methods, the above decision variables x and y are optimized simultaneously within a single mathematical programming problem.

In addition, the system constraint of equation (2) is set as a constraint equation for the mathematical programming problem shown in equation (1). Examples of the system constraints include the energy balance equation for each bus, the calculation equation for the voltage of each bus, the calculation equation for the distribution loss of each line, the upper limit of the number of tap position adjustments per day for tap adjustment devices, and the upper and lower limits of the adjustable amount of consumer DER devices. Note that A1, A2, and C in equation (2) are matrices or vectors that represent the system constraints, and the constraint equation in equation (2) is a matrix representation of the system constraints.

This mathematical programming problem can be solved as mixed integer programming (MILP) or mixed integer second-order cone programming (MISOCP) using a mathematical programming solver such as CPLEX or Gurobi. As a result of the solution, an operation plan x that minimizes (=optimizes) the system operation KPI during the planning period T for the input scenario s is obtained. This operation plan x is expected to prevent extreme deterioration of the system operation KPI due to violations of voltage tolerances, etc., for the input scenario s.

However, if the input scenario set S does not include scenario u that causes the worst deterioration of the system operation KPIs (hereinafter referred to as the most severe scenario), there is a possibility that an extreme deterioration of the system operation KPIs may occur. Therefore, in order to obtain an operation plan x that is robust against any fluctuation pattern, it is necessary to include the most severe scenario u in the input scenario set S. However, it is difficult to select an appropriate most severe scenario u in advance for the following two reasons.

The first reason is that there are a huge number of candidates for the most severe scenario u. In particular, if the number of uncertain parameters increases due to the spread of renewable energy sources and electric vehicles, the number of candidates will increase explosively.

The second reason is that the most severe scenario u varies depending on the operational plan x, so it is difficult to appropriately select the most severe scenario u before deciding on the operational plan x.

On the other hand, it is possible to input all possible candidates for the most severe scenario u, but this would require a huge number of input scenarios, which would increase the calculation time and may not complete within the specified time limit.

Due to the above problems with conventional operational planning methods, the present invention proposes the following operational planning method. The proposed operational planning method (hereinafter, the proposed method) has the following two advantages over conventional operational planning methods:

The first point is that, whereas conventional operational planning methods require the preparation of a set of scenarios that express uncertainty, the proposed method requires only the input of the range of variation of the uncertain parameters, and does not require the preparation of a set of scenarios.

The first point is that while conventional operational planning methods can only guarantee robustness for the set of input scenarios, the proposed method can guarantee robustness for any fluctuation pattern within the fluctuation range.

The mathematical programming problem (hereinafter, the robust operation planning problem) to be solved by the proposed method according to the present invention is shown in the following equations (3) and (4). Note that some points that have already been explained may be omitted.

In equation (3), u is a vector of uncertain parameters, and like x and y, is a decision variable of the power system operation planning problem. As the objective function and constraint equation, φ ^KPI , which is the system operation KPI, and equation (4) are set as the system constraint, as in the case of a normal operation planning method.

The feature of this power system operation planning problem is that the objective function has a three-layer structure of min and max. Starting from the outside of equation (3), there is a part that determines the operation plan x that minimizes (=optimizes) the system operation KPI, a part that determines the value of the uncertain parameter u (=most severe scenario) that maximizes (=worsens) the system operation KPI, and a part that determines the day's control y that minimizes the system operation KPI.

By solving this, the most severe scenario u, which is the fluctuation pattern within the fluctuation range that will cause the system operation KPI to deteriorate the most, is determined. At the same time, an operation plan x and an on-day control y that optimize the system operation KPI in the most severe scenario u are determined. Furthermore, the value of the system operation KPI at this time becomes the theoretical worst value, and even if any other fluctuation pattern occurs, it will not deteriorate beyond this theoretical worst value. Therefore, by solving this power system operation planning problem, an operation plan x that is robust against any fluctuation pattern can be obtained. Furthermore, since the most severe scenario u is generated while solving the power system operation planning problem, there is no need to prepare a candidate scenario s in advance.

Moreover, the system operation KPI (φ ^KPI ), which is an objective function, is defined, for example, by the following equation (5).

In this formula (5), φ ^cost , φ ^loss , and φ ^penalty respectively represent the operation cost, the transmission loss, and the constraint violation penalty. ^{W c} , W ^l , and W ^p respectively represent the weighting coefficients of the operation cost, the transmission loss, and the constraint violation penalty. The ratio of the weighting coefficients is generally set as in formula (6). Note that terms other than these may be added to the system operation KPI.

Moreover, the transmission loss φ ^loss is expressed by the following formula (7), for example: In formula (7), the time t in lowercase and the time T in uppercase, which is a set of times t included in the planning period, are in a relationship of tεT.

In this formula (7), j represents the bus number, N _tap represents a set of bus numbers with tap adjustment devices such as SVRs, and N _der represents a set of bus numbers with customer-owned DERs. C _j ^tap represents the cost per tap adjustment at the tap adjustment device. The value of C _j ^tap is calculated, for example, by dividing the sum of the tap adjustment device price and installation cost by the number of tap adjustments until the device reaches its end of life. x _j,t ^tap represents the number of tap adjustments at the tap adjustment device. Also, c _j,t ^der_r and c _j,t ^der_c represent the incentive unit price for reservation of adjustment margin of customer-owned DER and the incentive unit price for day-of control. _{x j,t} ^der and y _j,t ^der represent the reservation amount of adjustment margin of customer-owned DER and the activation command amount for the day.

As a result, the first term on the right side of equation (7) corresponds to the adjustment cost of the tap adjustment device, the second term corresponds to the incentive cost for the adjustment margin reservation, and the third term corresponds to the incentive cost for the actual control command. Note that xj _,ttap ^and yj _,tder ^are included in the above-mentioned operation plan variable vector x, and xj _,tder ^is included in the current day control variable vector y.

Moreover, the transmission loss φ ^loss in equation (5) is expressed by the following equation (8), for example: In equation (8), the time t in lowercase and the time T in uppercase, which is a set of times t included in the planning period, are in a relationship of tεT.

In equation (8), B represents a set of lines. r _ij represents the resistance value of the line from bus (bus line) i to bus j, and l ij,s,t represent the absolute value of the current amount in the line from bus i to bus j at scenario s and time t.

Moreover, the constraint violation penalty φ ^penalty in equation (5) is expressed, for example, in the following equation (9): In equation (9), the lowercase time t and the uppercase time T, which is a set of times t included in the planning period, are in a relationship of tεT.

In equation (9), ζ _ij,t represents the excess of the line constraint on the line connecting bus i to bus j, and η _j,t represents the deviation of the voltage at bus j from the allowable range. B represents a set of buses.

The operation plan generation unit 20 in the power system operation plan generation device 10 shown in FIG. 1 solves the power system operation plan problem shown in the above equations (1) to (9). However, since the power system operation plan problem has a three-layer structure as described above, mathematical programming solvers such as CPLEX and Gurobi do not support multi-layered mathematical programming problems, and cannot solve the problem as is.

Below, we explain an example of how to solve the problem using Benders decomposition, a typical decomposition method. In Benders decomposition, a mathematical programming problem is decomposed into two problems: a primary problem and a subordinate problem. First, the primary problem is solved, and the solution is used as a constraint to solve the subordinate problem. The subordinate problem passes a constraint equation called a Benders cut to the primary problem. This constraint equation of the primary problem is added to limit the search area for the solution, and then the problem is solved. By repeating the above process, a globally optimal solution is found.

Since the power system operation planning problem has a three-layer structure, the Benders decomposition is applied twice. First, the outermost min part is separated into the main problem (hereafter, the operation plan decision problem) and the rest into dependent problems. Since this dependent problem has a dual structure of max and min, the Benders decomposition is applied again, and it is again separated into the main problem and dependent problem (hereafter, the most severe scenario decision problem and the daily control amount decision problem, respectively). In this way, the three-layered power system operation planning problem can be decomposed into the following three mathematical programming problems.

These are the operation plan determination problem, the most severe scenario determination problem, and the daily control amount determination problem. The operation plan determination problem is processed by the operation plan generation unit 20 in Figure 1, the most severe scenario determination problem is processed by the most severe scenario generation unit 30, and the daily control amount determination problem is processed by the daily control amount generation unit 40.

In this relationship, the operation plan decision problem is the main problem, and its subordinate problem is the most severe scenario decision problem. At the same time, the most severe scenario decision problem is the main problem for the daily control amount decision problem, and its subordinate problem is the most severe scenario decision problem.

The above operation plan determination problem processed by the operation plan generation unit 20 is expressed by the following equations (10) and (11).

In these equations, θ, the objective function of the operation plan decision problem, represents the estimated value of the objective function value ηp of the most severe scenario decision problem (= the estimated value of the system operation KPI). The decision variables are θ and x. The proposed operation plan xp obtained by solving this is determined so as to minimize the estimated value of the system operation KPI. Note that variables with the symbol p indicate constant values obtained as a calculation result.

A Benders cut is set as a constraint in equation (11) and is added for each iteration. b1 is sensitivity information defined as the amount of change in ηp when the proposed operation plan xp is changed slightly, and is also called the shadow price. The sensitivity information is found as a solution to the dual problem of the dependent problem. The Benders cut equation constructed above can be regarded as a linear approximation equation for η with respect to x. By sequentially adding this, the estimation accuracy of θ is improved and the proposed operation plan xp is updated.

The most severe scenario determination problem processed by the most severe scenario generation unit 30 is expressed by the following equation (9).

In these equations, θ of the objective function of the most severe scenario determination problem represents the estimated value of the objective function value ^φpKPI (=system operation KPI) of the most severe scenario determination problem. The decision variables are θ and x. The proposed operation plan xp obtained by solving this is determined so as to minimize the estimated value of the system operation KPI. A vendor's cut is set as the constraint condition in equation (13) and is added for each iterative calculation. b2 is sensitivity information defined as the amount of change in ^φKPI when the scenario up is slightly changed. By adding this sequentially, the estimation accuracy of is improved and the most severe scenario up is updated.

The problem of determining the current day control amount processed by the current day control amount generation unit 40 is expressed by the following equations (14) and (15). The decision variable is the current day control y, and solving this is determined so as to minimize the estimated value of the system operation KPI.

In these equations, φ ^KPI of the objective function of the problem of determining the control amount of the day represents the system operation KPI. The constraint condition of equation (15) sets the system constraint. Therefore, in the problem of determining the control amount of the day, the system operation KPI is directly calculated based on the system constraint.

The three decomposed mathematical programming problems are solved alternately through iterative calculations as described above. The condition for determining when to end the iterative calculations is expressed by the following equation (16).

In this formula, ε is the allowable error rate, and is set to, for example, 0.0001. UB and LB represent the upper and lower limits of the objective function value, respectively. In the case of repeated calculations between the operation plan decision problem and the most severe scenario decision problem, UB and LB are calculated using the following formula (17).

In this formula, θp is the value of θ determined by the operation plan determination problem. In other words, it can be interpreted as repeating the calculation until θ, the estimated value of η, becomes sufficiently close to the true value η. In addition, when repeating the calculation between the most severe scenario determination problem and the day's control plan, UB and LB are calculated using the following formula (18).

In this formula, ηp is the value of η determined by the uncertain parameter determination problem. This can also be interpreted as repeating the calculation until the estimated value of φ ^KPI , η, becomes sufficiently close to the true value of φ ^KPI .

In Example 3, the method for generating a power system operation plan is explained using the process flow in Figure 5.

In the flow of FIG. 5, when processing is started, in processing step S2, the operation plan generation unit 20 functions to find a solution to the operation plan determination problem, equations (10) and (11). However, in this case, the operation plan determination problem is treated as the main problem, and the other parts are treated as dependent problems. In order to do this, the vendor's cut is generated and added in processing step S1 using equations (10) and (11), and processing step S2 is carried out under this condition. Note that the input of the operation plan variables used at this time (tap adjustment device input 100, customer-owned DER information 110) and the conditions for the most severe scenario (objective function value η, sensitivity information b1) are as shown in FIG. 1.

Next, in the process of the most severe scenario generation unit 30, in processing step S4, a solution to the most severe scenario determination problem, equations (12) and (13), is found. However, in this case, in order to make the most severe scenario determination problem the main problem and the daily control amount determination problem the dependent problem, in processing step S3, a vendor's cut is generated and added using equations (12) and (13), and processing step S4 is carried out under this condition. Note that the uncertain parameter fluctuation information (output of renewable energy sources and charging demand for electric vehicles), the daily control amount conditions (objective function value η, sensitivity information b2), and the proposed operation plan xp used at this time are as shown in FIG. 1.

Next, in the process of the current day control amount generating unit 40, in process step S5, a solution to the current day control amount determination problem, equations (14) and (15), is found. However, the current day control amount determination process in process step S5 is repeatedly executed until the calculation end determination is made using equations (16) and (18) that the most severe scenario determination problem in process step S6 has converged, and the calculation end determination is made using equations (16) and (17) that the operation plan determination problem in process step S7 has converged. The demand information and system information used at this time are as shown in FIG. 1.

Figure 6 is a diagram showing the difference in effect when comparing the present invention with the conventional method. The final output of the power system operation plan generation device in Figure 1 is displayed on a monitor screen or the like, and is shown as an image of what that might look like.

According to this notation, in conventional cases, the system operation KPI for one pre-assumed scenario would be calculated as point information, but in the case of the present invention, the most severe scenario is generated internally and the system operation KPI value of 270 (theoretical worst value) for that is calculated. Therefore, no matter what conceivable scenario occurs, the system operation KPI value is output in a format that falls within a range. In other words, compared to conventional cases, it is calculated as a range rather than a point.

10: Power system operation plan generating device 20: Operation plan generating unit 30: Most severe scenario generating unit 40: In-day control amount generating unit

Claims (4)

A power system operation plan generating device that takes into account uncertainty parameters in a power system, determines operation plan variables that should be determined in advance and on-day control variables that can be determined in real time, and performs an operation plan for the power system,
1. A power system operation plan generation device comprising: an operation plan generation unit that obtains information on operation plan variables and information on current day control variables in a power system, and creates a proposed operation plan for an objective function value in a most severe scenario that assumes a severe state of the power system; a most severe scenario generation unit that obtains the most severe scenario and its objective function value from information on fluctuations in uncertainty parameters, the proposed operation plan, and objective function values of current day control variables; and a current day control variable generation unit that determines an operation plan using system information and demand information of the power system, and the most severe scenario. The power system operation plan generating apparatus according to claim 1,
The power system operation plan generating device is characterized in that the operation plan variables include one or more of a tap position of a tap adjustment device and a regulation margin of a distributed energy resource owned by a consumer. The power system operation plan generating apparatus according to claim 1,
The power system operation plan generating device is characterized in that the control variables for the day include one or more of the activation command amounts of the adjustment margin of distributed energy resources owned by a power system operator and distributed energy resources owned by a reserved consumer. A power system operation plan generation method for determining operation plan variables to be determined in advance and on-day control variables that can be determined in real time, taking into account uncertainty parameters in the power system, and performing an operation plan for the power system, comprising:
A power system operation plan generation method for a power system operation planning problem in which information on operation plan variables and information on on-day control variables in a power system are obtained, and a proposed operation plan is created for an objective function value in a most severe scenario that assumes a severe state of the power system, the power system operation planning problem is divided into a main problem of an operation plan determination problem and a subordinate problem, and the subordinate problem is further divided into a main problem of a most severe scenario determination problem and a subordinate problem of on-day control amount determination problem, thereby performing an operation plan for the power system as a solution to a three-tiered power system operation planning problem.

Images