JP2011188590A - Generator operation plan determination device, method for determining generator operation plan, and generator operation plan determination program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To make a starting plan of a plurality of generators satisfying a demand forecast of power and various kinds of restrictions suitably and at low processing load. <P>SOLUTION: The demand forecast is obtained and an initial plan is obtained that contains an initial value of a starting state value of each of the generators determined according to a time-series demand forecast. Based on the initial plan, a generator subject to correction of the starting state value is determined. Based on a power generation cost formula including the starting state value of each of the generators, a starting state value of each generator is obtained so that the demand forecast can be satisfied and a calculated cost becomes the minimum. In that case, the calculation is made by changing a coefficient of the power generation cost formula so as to make it easy for the starting state values of subject generators to converge on 0 and for the starting state values of non-subject generators, which are generators other than the subject generators, converge on 1. When the starting state values of the subject generators are calculated as values other than 0, it is treated as 1. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a generator operation plan determination device, a generator operation plan determination method, and a generator operation plan determination program, particularly a thermal power plant start / stop plan, a thermal power plant, a pumped storage power plant, a hydroelectric power plant, etc. A power plant operation plan that satisfies the constraints such as supply and demand balance, reserve capacity, fuel consumption constraints, tidal current constraints, dam water level constraints, etc., and minimizes the power generation costs of thermal power plants About.

  In order to maintain the reliability of the power system, create a generator start / stop plan that ensures reserve power to maintain the reliability even with demand forecast errors while matching the power demand with the power supplied by the generator. There is a need to. In order to operate economically at the same time as reliability, it is necessary to create a plan to minimize the power generation cost by creating a shutdown of the generator so that the power generation cost is minimized.

  In particular, competition between operators will become intense due to the liberalization of electric power, so it is possible to combine generator outputs from thermal power plants, pumped-storage power plants, and hydroelectric power plants that can be handled as supply power that can adjust the output based on the directive. It is important to minimize power generation costs. Especially in the power supply plan of a thermal power plant, it becomes difficult to optimize the whole supply power because it becomes a combination problem of start and stop.

  As a conventional technique for generating an operation plan for a thermal power plant, all start and stop variables are relaxed to real numbers in the start and stop plan and the load distribution plan, and an optimal solution is obtained as a quadratic programming problem as shown in FIG. There is a method of determining a start / stop variable as 1 or 0 using a genetic algorithm (see, for example, Patent Document 1). The starting / stopping variable of the thermal power can take only either “1” indicating start or “0” indicating stop. However, in the calculation result, there is a generator having a value other than this as shown in FIG. 17, and the plan cannot be determined. For this reason, in the method disclosed in Patent Document 1, as shown by “*” in FIG. 18, this undecided value is determined by using a genetic algorithm for only the start / stop variables larger than 0 and smaller than 1 in FIG. 17. The start / stop variable is determined.

JP 2007-166855 A

  In the case of the technique disclosed in Patent Document 1, if an individual (start / stop plan) created by a genetic algorithm and the number of generations, for example, 50 generations and 100 generations, load distribution calculation is required 5000 times, and the calculation time becomes long. . Moreover, since the plan which considered system reliability, such as fuel consumption restrictions and tidal current restrictions, is not made into the object, the result according to the actual condition cannot be obtained. Furthermore, since only the thermal power generator is targeted for planning, it does not include pumped-storage power plants or hydroelectric power plants, so it is not possible to minimize power generation costs using variable supply power.

  Since the thermal power plant startup / shutdown plan is a combination optimization problem, for example, the total number of startup / shutdowns for 24 hours with 10 thermal power generators is 10 to the 72nd power. Even if one plan can be evaluated in 1 microsecond, it takes 10 59 years to evaluate all plans, and it is virtually impossible to check the total number to find the optimal solution. .

  For this reason, it is practically impossible to determine the truly optimal generator output of the thermal power plant, the pumped-storage power plant, and the hydroelectric power plant at the same time that the start-up / stop plan for the thermal power plant is created. This is to handle the combination problem. Therefore, by relaxing the start / stop planning problem of the thermal power generator from the combination problem to the problem of real variables, the optimal solution can be obtained by the quadratic programming method and the linear programming method in a practical time.

  In particular, constraints that can be expressed by linear constraints such as fuel consumption constraints, power flow constraints, or dam water level constraints can be taken into account at the same time when obtaining an optimal solution. However, there is a problem that there is a variable in which the start / stop variable of the thermal power generator is not fixed to 0 or 1, and there is an undefined start / stop variable.

  The present invention has been made in response to the above-described points, and an object of the present invention is to generate a power generation forecast and a plurality of generator startup plans that satisfy various restrictions with a suitable and low processing load.

  In order to solve the above-described problem, one aspect of the present invention is a generator operation plan determination apparatus that determines a startup state of a plurality of generators based on a time-series predicted value of power demand, the power demand time-series A demand forecast acquisition unit that obtains a demand forecast that is information of a forecast value, and a startup state value that is information indicating either the start or stop state of each of the plurality of generators by a numerical value of 1 or 0 An initial plan acquisition unit that acquires a generator start plan whose initial value is information determined according to the time series of the demand speed, and a generator to be corrected that corrects the start state value based on the generator start plan A correction target determination unit that determines a certain target generator, and an expression for calculating a power generation cost for calculating a power generation cost for an output for each of the plurality of generators, and based on a power generation cost calculation formula that includes the activation state value ,in front An activation state calculation unit that obtains activation state values of the plurality of generators in time series so that the demand prediction is satisfied in a time series and the calculated cost is minimized, and the activation state calculation unit includes: When obtaining the activation state value in time series, the activation state value of the target generator easily converges to 0, and the activation state value of the non-target generator that is a generator other than the object generator converges to 1. The power generation cost calculation formula is calculated by changing the coefficient of the power generation cost calculation formula so that the constraint of the startup state value is a real number from 0 to 1, so that the startup state value of the target generator is a value other than 0. Is calculated as 1, it is characterized by being set to 1.

  According to another aspect of the present invention, there is provided a generator operation plan determination method for determining a startup state of a plurality of generators based on a time-series predicted value of power demand, wherein a time-series predicted value of power demand is determined. It is information in which an initial value of a starting state value, which is information indicating a demand prediction that is information, and that indicates either the starting or stopping state of each of the plurality of generators by a numerical value of 1 or 0, is defined. Obtain a generator startup plan, determine a target generator that is a correction target generator to correct the startup state value based on the generator startup plan, and calculate a power generation cost for output for each of the plurality of generators Based on a power generation cost calculation formula that includes the start-up state value, a plurality of the plurality of power generation costs are calculated so that the demand forecast is satisfied and the calculated cost is minimized in the time series. Generator When obtaining the dynamic state value in time series and obtaining the activation state value in time series, the activation state value of the target generator is likely to converge to 0, and the non-target generator that is a generator other than the target generator The power generation cost calculation formula is calculated by changing the coefficient of the power generation cost calculation formula so that the startup status value of the current value easily converges to 1, and the constraint of the startup status value is a real number from 0 to 1, and the target generator When the activation state value is calculated as a value other than 0, it is set to 1.

  According to still another aspect of the present invention, there is provided a generator operation plan determination program for determining a startup state of a plurality of generators based on a time-series predicted value of power demand, the time-series predicted value of power demand. A step of obtaining a demand forecast that is information of the above, and an initial value of a starting state value that is information indicating either a starting or stopping state of each of the plurality of generators by a numerical value of 1 or 0 is defined Obtaining a generator start plan that is information; determining a target generator that is a correction target generator to correct the start state value based on the generator start plan; and each of the plurality of generators A formula for calculating a power generation cost for calculating a power generation cost with respect to an output, and based on a power generation cost calculation formula including the activation state value, the demand forecast is satisfied and calculated in the time series In order to minimize the cost, when the activation state values of the plurality of generators are obtained in time series, and when the activation state values are obtained in time series, the activation state values of the target generator are likely to converge to 0. Thus, the coefficient of the power generation cost calculation formula is changed so that the starting state value of the non-target generator that is a generator other than the target generator easily converges to 1, thereby limiting the starting state value from 0 to 1. And causing the generator operation plan determination device to execute a step of calculating the power generation cost calculation formula as a real number of the target generator and a step of setting the value to 1 when the activation state value of the target generator is calculated as a value other than 0. Features.

  ADVANTAGE OF THE INVENTION According to this invention, the starting plan of the several generator which satisfy | fills the demand prediction of electric power and various restrictions can be produced | generated with a suitable and low processing load.

It is a flowchart which shows the calculation process by the power plant operation plan preparation apparatus which concerns on embodiment of this invention. It is a figure which shows the hardware constitutions of the power plant operation plan preparation apparatus which concerns on embodiment of this invention. It is a figure which shows the detailed structure of the power plant operation plan preparation apparatus which concerns on embodiment of this invention. It is a figure which shows the storage area comprised in the main memory of the power plant operation plan creation apparatus which concerns on embodiment of this invention. It is a figure which shows the example of the demand prediction which concerns on embodiment of this invention. It is a figure which shows the example of the initial plan which concerns on embodiment of this invention. It is a figure which shows the setting screen of correction values, such as a coefficient, concerning embodiment of this invention. It is a figure which shows the example of grouping of the generator which concerns on embodiment of this invention. It is a figure which shows the example of the group selected as object in the process which concerns on embodiment of this invention. It is a figure which shows real number relaxation of the generator of the process target which concerns on embodiment of this invention. It is a figure which shows the example of the result calculated in the generator of the process target which concerns on embodiment of this invention. It is a figure which shows the example of the result as which the starting state variable was determined in the generator of the process target which concerns on embodiment of this invention. It is a figure which shows the calculation result in the group of the process target which concerns on embodiment of this invention. It is a figure which shows the result of having repeated the update process of the starting variable which concerns on embodiment of this invention. It is a figure which shows the example of the screen which confirms the constraint condition which concerns on embodiment of this invention. It is a figure which shows the other example of the determination aspect of the process target generator which concerns on embodiment of this invention. It is a figure which shows the result of having calculated | required the optimal solution of the secondary programming method problem in formulation of a power plant operation plan. It is a figure which shows the case where the part which became uncertain in the optimal solution of a quadratic programming problem is specified. In the calculation process which concerns on embodiment of this invention, it is a figure which shows the example in the case of calculating considering a some non-target generator as a single generator. In the calculation process which concerns on embodiment of this invention, it is a figure which shows the example in the case of calculating considering a some non-target generator as a single generator. In the calculation process which concerns on embodiment of this invention, it is a figure which shows the example in the case of calculating considering a some non-target generator as a single generator. In the calculation process which concerns on embodiment of this invention, it is a figure which shows the example in the case of calculating considering a some non-target generator as a single generator.

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the drawings.
In the present embodiment, the undefined start / stop variable is limited to one generator at one time, and the fuel function or start-up cost of the generator at the undefined time is larger than the original function. By reducing the fuel cost function or start-up cost of the generator to be fixed or lower to the original function and applying quadratic programming or linear programming to solve the problem, The start / stop plan is determined and resolved by classifying the generators into the generators that require the stop 0 and other start values.

  FIG. 1 is a calculation processing flow by a power plant operation plan creation device to which the present invention is applied, FIG. 2 is a schematic overall configuration diagram of the power plant operation plan creation device 1, and FIG. 3 is a detailed configuration of the power plant operation plan creation device 4 and 4 show storage areas configured in a main storage device included in the power plant operation plan creation device.

  As shown in FIG. 2, the power plant operation plan creation device 1 includes a central processing unit (CPU) 2, a main storage device 3, an input / output device 4, and an external storage device 5. . FIG. 3 is a detailed functional configuration diagram of the power plant operation plan creation apparatus 1 according to the present invention. The power plant operation plan creation device 1 includes an input device 6, a display device 7, a database 9, a reading device 10, and a calculation processing unit 20. The calculation processing unit 20 is realized by the CPU 2 executing a program that is stored in advance in the storage medium of the external storage device 5 and is read into the main storage device 3 via the reading device 10. Is not limited to such a programmed general purpose processor. For example, the calculation processing unit 20 can be configured by a combination with a specific hardware device including a wired logic that executes each processing of the present invention.

  The input / output device 4 includes an input device 6 having a keyboard and a mouse shown in FIG. 3 and a display device 7 as an output device. The input / output device 4 may be provided with an input device such as a pointing device or a touch sensor, or an output device such as a liquid crystal display device, a printer, or a speaker instead of or in combination with these devices. The external storage device 5 includes a hard disk device, a floppy disk device, a CD-ROM (compact disc-read only memory) device, a DAT (digital video memory) device, a RAM (random access memory) device, a DVD (digital video disc) device, and the like. A nonvolatile memory or the like can be used.

  The external storage device 5 includes a mass storage device for holding the database 9 shown in FIG. 3, a storage medium for holding a processing program, and a reading device 10 for reading information held in the storage medium. Although used, both the database and the processing program can be held in one external storage device. As the storage medium, a floppy disk, CD-ROM, magnetic tape, optical disk, magneto-optical disk, DAT, RAM, DVD, nonvolatile memory, or the like can be used.

  The input device 6 accepts selection of options displayed on the display device 7, input of data, and the like, and transmits them to the calculation processing unit 20. The display device 7 displays the data sent from the input device 6. The calculation processing unit 20 operates the supply power based on the data transmitted from the input device 6, the data read from the database 9, the processing program read from the reading device 10, and the data transmitted from the power system 11. Create a plan. The power system 11 includes a database (not shown). The power system 11 monitors and controls the power system 12.

  The processing result of the calculation processing unit 20 is sent to the display device 7 for display and stored in the database 9. Further, when the power plant operation plan creation device 1 is requested to create an operation plan under the conditions set by the operation / planning system and control system of the power system 11, the plan created by the calculation processing unit 20 and The evaluation result is also notified to the power system 11. Upon receiving this notification, the power system 11 outputs an output control signal to the generator that is one element of the power system 12 based on the notified plan to control the generator, and generates power from the power system 11. The machine output data is fetched, and the fetched data and data created in the planning system (not shown) are stored in an internal database (not shown).

  The calculation processing unit 20 includes a control unit 21, a plan creation data reading unit 22, a plan creation strategy setting unit 23, an initial plan reading unit 24, a target generator determination unit 25, a real number relaxation unit 26, an optimization processing unit 27, and start / stop. An optimumity check unit 29 including a determination unit 28, a best solution check unit 30 and an optimum solution check unit 31, and a calculation result output unit 32 are provided. Further, the calculation processing unit 20 is connected to an external power system 11 via a transmission line 33, and the power system 11 is connected to a power system 12 that is an actual transmission and transmission substation facility. Thus, in this embodiment, the power plant operation plan creation device 1 including the calculation processing unit 20 functions as a generator operation plan determination device that determines the operation plan of the generator included in the power system 12.

  The power system 11 includes a system for planning / operation and control of the power system 12, a power system 12 and a database (not shown) for holding information indicating the state of the power system 12, and the like. The state of the power system 12 is detected by a relay, a sensor, or the like, notified to the power system 11 through the transmission line 34, and stored in a database (not shown) of the power system 11.

  The control unit 21 processes and processes data for smoothly transmitting and receiving data and processing programs between the power system 11 and each of the processing units 22 to 32, and controls the transmission and reception of the entire system. Make the process work properly.

  The plan creation data reading unit 22 reads the demand forecast value, generator characteristics, constraint conditions, etc. for creating the operation plan held in the storage medium of the database 9 or the reading device 10 via the control unit 21 and displays them. The information is displayed on the device 7 and the conditions are changed as necessary. Further, the changed new condition is stored in the database 9 and the generator characteristic storage area M101 and the constraint condition storage area M102 of the main storage device 3 through the control unit 21, and each process is performed through the control unit 21. These data are transmitted to the units 23-32.

  In this embodiment, unless otherwise specified, reading of data from the database 9 is performed via the data reading unit 22, and the plan creation data reading unit 22, the plan creation strategy setting unit 23, the initial plan reading unit 24, The target generator determination unit 25, the real number relaxation unit 26, the optimization processing unit 27, the start / stop determination unit 28, the best solution check unit 30, the optimality check unit 29 including the optimal solution check unit 31, and the calculation result output unit 32 are Data is exchanged via the control unit 21, and data and calculation results input from the input device 6 are output to the display device 7 or stored in the database 9 as necessary.

  The plan creation strategy setting unit 23 determines conditions for determining a stopped generator group to be used when creating an operation plan held in the storage medium of the database 9 or the reader 10 via the control unit 21, a fuel cost function, The order and the like such as the correction method of the start-up cost are read and displayed on the display device 7, and the conditions are changed or set as necessary. The changed new condition is stored in the database 9 and the strategy storage area M103 of the main storage device 3 through the control unit 21, and these data are transmitted to the processing units 24-32 through the control unit 21. To do.

  The initial plan reading unit 24 selects an initial plan stored in the storage medium of the database 9 or the reading device 10 via the control unit 21, displays it on the display device 7, and changes or sets conditions as necessary. And stored in the database 9 and the current solution storage area M104, the best solution storage area M105, the optimum solution storage area M106, and the like of the main storage device 3. The best solution storage area M105 and the optimum solution storage area M106 are not overwritten and deleted from the previous operation plan each time the best operation plan and the optimum operation plan are updated, but the previous operation plan is stacked. It is structured to be stored and retrieved.

  In addition to the various data and conditions set by the plan creation data reading unit 22, the plan creation strategy setting unit 23, and the initial plan reading unit 24 via the control unit 21, the target generator determination unit 25 includes an optimization check unit 29. Based on the results, the generator group, the priority of the group, and the thermal power generator included in the group are determined. Further, the target generator determining unit 25 determines a generator group to be used by the real number relaxation unit 26 and the optimization processing unit 27, a priority order of the group, and a thermal power generator as necessary. To the processing unit 27. These pieces of information are stored in the database 9 and the target generator storage area M107 and the generator group storage area M108 of the main storage device 3 by the target generator determination unit 25.

  The real number relaxation unit 26 is based on various data and conditions set by the plan creation data reading unit 22, the plan creation strategy setting unit 23, the initial plan reading unit 24, and the target generator determination unit 25, and the time and power generation for real number relaxation. The machine is determined and displayed on the display device 7. In addition, the real number relaxation unit 26 acquires a correction value for correcting a coefficient included in the calculation formula of the power generation cost of the generator by operating the input device by the operator, and determines whether the real number is to be relaxed or other generator Is corrected and stored in the database 9 or the temporary fuel cost function storage area M109 of the main storage device 3 or the like.

  The optimization processing unit 27 receives various data and conditions set by the plan creation data reading unit 22, the plan creation strategy setting unit 23, the initial plan reading unit 24, the target generator determination unit 25, and the real number relaxation unit 26 via the control unit 21. Based on the above, the constraint conditions and the objective function are formulated, and a calculation process for deriving an optimal solution, that is, an optimal power plant operation plan, is performed using mathematical programming methods such as linear programming and quadratic programming. The calculation result is stored in the current solution storage area M104, the best solution storage area M105, the optimum solution storage area M106, etc. of the main storage area 3.

  The start / stop determination unit 28 determines the start / stop of all thermal power generators based on the value of the start / stop variable created by the optimization processing unit 27 via the control unit 21, and this start / stop plan is stored in the main storage area 3. The current solution storage area M104, the best solution storage area M105, the optimum solution storage area M106, etc. are stored.

  The optimality check unit 29 includes a best solution check unit 30 and an optimal solution check unit 31. The best solution check unit 30 generates the power generation cost of the power plant operation plan stored in the current solution storage region M104 of the main storage region 3 created by the optimization processing unit 27 and the power plant operation stored in the best solution storage region M105. The power generation costs of the plans are compared, and the power plant operation plan in the current solution storage area M104 is stored in the best solution storage area M105 as necessary.

  The optimal solution check unit 31 compares the power generation cost of the power plant operation plan stored in the best solution storage region M105 of the main storage region 3 with the power generation cost of the power plant operation plan stored in the optimal solution storage region M106. If necessary, the power plant operation plan in the best solution storage area M105 is stored in the optimum solution storage area M106.

  The calculation result output unit 32 is information for supporting the data input from the input device 6 and the plan creation conditions and data set and calculated by each of the processing units 22 to 31, the generated power plant operation plan, the transition of the objective function, etc. Are displayed on the display device 7.

  Next, the operation of the power plant operation planning apparatus of the present invention will be described with reference to the processing flow diagram shown in FIG. The plan creation data reading unit 22 reads the plan creation data in step S101. Here, the input information of the power plant operation plan creation target date displayed on the display device 7 is input by the input device 6 and confirmed. As a result, the plan creation data reading unit 22 reads the total demand in the target period, the fixed supply power by nuclear power and flowing hydropower, the fixed supply power with uncertain supply power such as wind power and solar power, etc. from the database 9, and these are read from the total demand. Calculate the deduction demand to be supplied by thermal power, pumping, and hydraulic power minus the fixed supply capacity. That is, the plan creation data reading unit 22 functions as a demand prediction acquisition unit.

  FIG. 5 shows an example of the deduction demand displayed on the display device 7. In the following, the deduction demand is expressed as demand unless otherwise specified. This demand is used as a demand forecast. As information read in S101, the upper and lower limits of the output of the thermal power generator, the fuel cost function (secondary function, linear function, etc. for the generator output), start-up cost, fuel type to be used, minimum continuous stop time, minimum continuous operation I have time. The pumped storage power plant has output upper and lower limits, initial output water level, ultimate water level, water level upper and lower limits, and pumping efficiency during power generation and pump power. In hydroelectric power plants, there are output upper and lower limits or upper and lower limits of water consumption, electric water ratio, power generation efficiency curve, flow time delay until the discharged water reaches the downstream dam, etc.

  These generator characteristics of thermal power, pumping water and hydraulic power are read from the database 9. Operation reserve capacity, lower reserve capacity that is the lower operation reserve capacity so that a balance between supply and demand can be achieved even when demand is less than the assumed demand, the area where each generator is connected, and the power flow restrictions on the transmission lines connecting the areas System constraints such as fuel consumption upper and lower limit constraints such as LNG are read from the database 9. These read data are displayed on the display device 7, and the conditions are changed via the input device 6 and stored in the main storage area 3. The generator characteristic is stored in the generator characteristic storage area M101, and the constraint condition is stored in the constraint condition storage area M102.

  The process S102 is a process for setting a plan creation strategy, which is processed by the plan creation strategy setting unit 23. This uses the information in the database 9 to select which initial plan from a plurality of initial plans, how to determine the generator for changing the start / stop variable, and when changing the generator group The classification method and the order of the group to be changed, the fuel cost function, the startup cost correction method, the range of the startup / shutdown plan to be corrected, and the optimization method selection strategy are determined. The determined various information is stored in the database 9 and also in the main storage device 3.

  These strategies are also stored in the strategy storage area M103. The initial plan is selected, for example, in the order in which the unit price at maximum output is cheap, that is, in order of good power generation efficiency, and when the total rated output of the selected generators satisfies the supply reserve. It is created by the priority method that is created by planning to start the thermal power generator. In addition, in the plan with all thermal power generators in operation, the plan of calculation results created by real number relaxation of all start and stop variables of thermal power with quadratic programming or linear programming, or the plan of the calculation results, It is possible to use a plan in which all starting and stopping variables are larger than 0, a power plant operation plan having a past maximum demand or a minimum demand on the same level as the planning target date, and the like. Further, these plan candidates are displayed on the display device 7, and the start / stop plan of the thermal power generator is manually corrected using the input device 6 or the start / stop plan of the thermal power generator is manually created from the beginning. Also included.

  Here, the classification strategy when dividing into generator groups is to determine a group for correcting the start / stop variable of the thermal power generator. In this embodiment, the classification strategy is grouped according to the length of the operation time in the initial plan. . Specifically, it is grouped according to the length of the number of operating hours at the minimum demand time and the maximum demand time. For example, assume that the initial plan selected or created is FIG. The first row of the table shows the number of thermal power generators, 10 generators from 1 to 10, and the first column shows the time, from 1 o'clock to 24 o'clock. The numerical values in the table are start / stop variables indicating the start / stop state of each device at each time. 1 indicates start and 0 indicates stop.

  In the example of FIG. 5, since the minimum demand time is 5 o'clock and the maximum demand time is 15 o'clock in the case of the example of FIG. 5, the operation is performed even at the least important time. Group 1 from generators 1 to 5, group 2 from generators 6 to 8 excluding the generators included in group 1 from the generators operating at the maximum important time, both minimum and maximum important times Is divided into three groups, group 0 of generators 9 and 10 stopped at. FIG. 8 shows this grouped display on the display device 7.

  Thermal power generators with lower efficiency, i.e. higher power generation costs, have shorter operating times. In general, the power generation cost is reduced when the number of generators operating at each time is reduced with respect to the initial plan or the plan created in the middle. Generally, a thermal power generator is designed to have maximum efficiency at maximum output, and the maximum output is determined as a rated output. For this reason, the smaller the number of generators operating at the same time, the more the distributed output of each generator increases, so that the operating efficiency improves and the unit price of power generation decreases. As a result, the overall power generation cost is reduced.

  In the present embodiment, the plan for stopping the generators as much as possible in order for each group is corrected. Since group 0 in FIG. 8 is stopped at all times, it is excluded from the targets to be revised. The generator of group 2 has a shorter operating time than group 1. This indicates that the power generation unit price of group 2 is higher than the power generation unit cost of group 1. This is because, as described above, the initial plan is formulated so that the generator with the lower unit price is operated preferentially by considering the unit price of each generator when the initial plan is formulated. by.

  Therefore, the group 2 is first set as the group for the plan correction, and after the power generation cost cannot be reduced in the group 2, the plan correction for the generator of the group 1 is performed next. The reason for grouping in this way is that in the process of correcting the operation plan by applying the technique of the present embodiment to the initial plan, a generator with a low power generation cost, that is, a generator to be preferentially operated is generated. This is because there is a possibility that the generator is stopped before the generator with high cost, that is, the generator to be stopped with priority. Such a problem is particularly likely to occur in an initial plan or a current plan in which supply reserve capacity is excessive such that all generators are operated.

  Similarly, in the case of grouping according to the length of the operation time, it is possible to include a plurality of generators in the group by giving a width to the operation time. Or, instead of grouping at the initial planning stage, it is determined that there are many hours that can be stopped for all operating generators, or that there are many hours that can be stopped at the minimum demand time and the maximum demand time. After that, there is also a method of grouping.

  As will be described later from processing S111 to processing 113, the plan in which the power generation cost is reduced when each of them is set as a stop candidate in the same group is adopted as the best plan, so there is no priority order of generators in the group. Further, as shown in FIG. 16, it is possible to set a strategy for displaying an initial plan or a current plan on the display device 7 and manually specifying a place to be undetermined by the input device 6.

  As a strategy for determining the range of the startup / shutdown plan to be corrected, FIG. 10 shows the uncertain location by the symbol “*”, but the initial plan or the current plan is shown in FIG. 6 and the generator 7 is selected as the undefined generator. In this case, as a strategy, there are a method in which all 24 hours are unconfirmed and a method in which the symbol “*” is used only for the time at which the vehicle was originally operated as in the symbol “*” in FIG. Since the former is uncertain at all times, processing is simpler because the number of cases in the calculation process is reduced, and the latter can reduce the number of undefined variables, thereby reducing the scale of the problem and the amount of calculation processing. Can be reduced.

  In the startup / shutdown plan correction method according to the present embodiment, the coefficient of the fuel cost function of the generator whose startup / shutdown variable is undecided, that is, the generator to be corrected with the startup / shutdown variable set to “*” as described above. And the start-up cost is made larger than the original fuel cost function and start-up cost. Alternatively, the calculation is performed with the coefficient of the fuel cost function and the start-up cost of the generator for which the start / stop variable is determined smaller than the original fuel cost function and start-up cost. That is, the calculation is performed by changing the coefficient of the fuel cost function and the start-up cost so that the power generation cost of the generator to be corrected is calculated higher than the power generation cost of the other generators. As a result, when the optimum solution is obtained, the start / stop variable of the generator to be corrected easily converges to “0”, and the start / stop variables of other generators easily converge to “1”. Incidentally, since the fuel cost of the generator is generated only when it is in operation, even if the fuel cost of the generator that has been stopped is removed from the correction target, the problem still remains in the optimization processing unit 27 in steps S108 and S111. This does not affect the calculation results.

  Here, the power generation cost per unit time of the generator is calculated by an expression such as “aP2 + bP + c” where the output of the generator per unit time is “P”. Then, when obtaining the optimum solution of the start / stop variable in consideration of the demand forecast and the fuel consumption constraint described above, the cost is minimized by using the formula “aP2 + bP + cu” obtained by adding the start / stop variable “u” to the above formula. To find a solution. FIG. 7 shows on the display device 7 the quadratic and linear coefficients a and b of the fuel cost function, the constant term c and the start-up cost, and these values can be corrected using the input device 6. Here, a toggle switch for correcting either the indeterminate generator or the established generator is used. As already described, since the operation of the confirmed generator cannot be newly stopped, the object of the present embodiment is to stop the unestablished generator as much as possible.

  Among the undetermined generators, the generator whose start / stop variable is to be corrected has its start / stop variable calculated when calculating the above-described formula in the calculation of the process S108 of the optimization processing unit 27 to obtain the optimum solution. The coefficient of the fuel cost function and the start-up cost (hereinafter referred to as a coefficient, etc.) are corrected so as to easily converge to “0”. This correction is performed by the real number relaxation unit 26 in S107 based on a value preset by the operator on the screen as shown in FIG.

  For example, in an extreme setting, a or b which is a quadratic or linear coefficient of the fuel cost function is set to a very large value such as infinity. Or, the above-mentioned coefficient and the like are set to zero so that the output of the generator that is not subject to the correction of the start / stop variable, such as the generator whose operation is defined, becomes large, that is, the start variable converges to “1”. . That is, in the start / stop variable correcting method in the present embodiment, the start / stop variable of the generator to be corrected converges to “0”, and the start / stop variable of the generator not to be corrected converges to “1”. In addition, there is a large difference between the coefficient for the generator to be corrected and the coefficient for the generator not to be corrected.

  These only have to be realized on the screen shown in FIG. 7, and it is not necessary to stick to the format of FIG. In addition, a numerical value N is input in order to make each coefficient of the indeterminate generator N times or each coefficient of the definite generator 1 / N times the fuel cost function and start-up cost shown in FIG. It may be a screen. The value set in this way is stored in the strategy storage area M103.

  The optimization method selection strategy is to select a solver that is an optimization program for solving a mathematical programming problem stored in the external storage device 5. Examples of solvers include quadratic programming, linear programming, and equal lambda. The selected solver is stored in the solver storage area M111 of the main storage device 3.

  In the process S103, the initial plan reading unit 24 selects and determines an initial plan from the initial plan candidates stored in the database 9 based on the strategy storage area M103 of the main storage device 3. This initial plan includes a thermal power plant startup / shutdown plan, thermal power, pumping, hydroelectric generator output and power generation costs. This initial plan is selected from candidates in the database, displayed on the display device 7, and corrected as necessary by the input device 6, and the current solution storage area M103, the best solution storage area M104 and the optimum solution storage area of the main storage device 3 are selected. It is stored in the solution storage area M105. That is, the initial plan reading unit 24 functions as an initial plan acquisition unit. This is because there is only one plan at the stage when the initial plan is read, so it is the current solution, the best solution, and the optimal solution.

  Here, the current solution, the best solution, and the optimum solution will be described. The current solution is a start / stop plan that is updated by calculation by the optimization processing unit 27 for the generator that is the target generator by the target generator determination unit 25. The best solution is the startup / shutdown plan determined within the above-mentioned group, and if the updated current solution is superior to the current best solution, that is, the cost is lower, the current solution The best solution is updated. The optimal solution is a start / stop plan that is determined after the best solution in each group is determined. If the updated best solution is less expensive than the optimal solution at that time, the best solution The optimal solution is updated.

  For example, in the case of the grouping shown in FIG. 8, first, the generators 6, 7, and 8 included in the group 2 are sequentially set as target generators, and the calculation in each case is executed by the optimization processing unit 27. As a result, the current solution is obtained and the best solution is updated. Then, after the best solution in group 2 is determined as the optimum solution, the same calculation is further performed for group 1 on the assumption of the optimum solution, and the optimum solution is updated.

  In the determination process of the generator group in the process S104, the target generator determination unit 25 uses the strategy storage area M103 as a classification strategy for dividing the current plan thermal power start / stop plan as the initial plan and the generator group from the current solution storage area M104. Read, determine the class of the generator group, and store it in the generator group storage area M108. FIG. 8 shows the determined example on the display device 7. As shown in FIG. 8, group 2 is first selected.

  The process S105 is a process of selecting the next group, and the target generator determination unit 25 reads the generator number of the currently selected group from the generator group storage area M108. Further, the group check end flag in FIG. 8 is a flag indicating whether or not the group has been processed. “1” indicates that the group has not been processed yet, and 0 indicates that the group has been processed. However, in the group 0, since the generators that have stopped at all times from the set strategy are removed from the stop targets, the flag is set to 0 from the beginning.

  The group priority is advanced from the upper line where the group check end flag is 1. That is, the priority order of the upper row group is high. As described above, since it is the group 2 that should be preferentially corrected, the group 2 is displayed in the upper row. The group check end flag is determined to have ended in the process S116 for determining whether or not the process of the target group has ended, and the group check end flag of the group that has been the target of the generator group storage area M108 is 0. Set to

  In process S106, the generator of the group currently being checked by the target generator determination unit 25 is read from the generator group storage area M108, and the generator whose update flag is 1 and whose generator check end flag is 1 is started. Select as a stop change target. The generators of the read group are in the format as shown in FIG. 9, and when the process proceeds to the Yes side in step S113, the update flag of the generator that becomes the best solution by stopping in the same group is 0. Set to The generator whose update flag is 0 is excluded from the start / stop change target, that is, the calculation target, when the calculation for updating the start / stop plan is repeated. Thus, in S104 to S106 according to the present embodiment, the target generator determination unit 25 functions as a correction target determination unit.

  In the process S107, the real number relaxation unit 26 determines the start / stop change target generator determined in the processing unit S106, the start / stop plan stored in the optimal solution storage area M106, and the start / stop plan to be corrected stored in the strategy storage area M103. Based on the strategy for determining the range of the above, the time and generator for relaxing the start / stop variable are determined and stored in the correction target storage area M107. Here, “relaxation” refers to changing the constraint so that the start / stop variable, which is originally limited to “0” or “1”, can be a real number between 0 and 1. Further, the real number relaxation unit 26 sets a constraint condition and an objective function using the time and generator stored in the correction target storage area M107, and stores them in the objective function / constraint condition storage area M110. The secondary coefficient a, the linear coefficient b, the constant term c, and the startup cost of the fuel cost function of the thermal power generator are values corrected using the fuel cost function coefficient and the startup cost correction strategy as illustrated in FIG. Is used.

Add the above constraints, formulate and calculate the result. The objective function is a cost calculation formula for creating a power plant operation plan in which the power generation cost TC consisting of the fuel cost and the start-up cost is minimized. The power generation cost TC includes the second-order coefficient ai, the linear coefficient bi and the constant term ci of the fuel cost function of the thermal power generator i, the starting cost SUCi, the power generator number i of the thermal power generator, the output Pit at time t, and the start / stop variable uit. Is represented by the following formula (1).

  In equation (1), the fuel cost is expressed as a quadratic function of the generator output Pit, such as “aitPit2 + biPit + circuit”, but may be expressed linearly. Since the generator output that has stopped is zero, the operating state uit for starting and stopping (t is time) only needs to be applied to the constant term of the fuel cost. The start-up cost is expressed using the start / stop state uin at the minimum demand time N and the start / stop state uip at the maximum demand time P. The formula (1) is a formula for one day, but by adding the formula for one day for the required number of days, it can be used for a weekly plan, for example.

式（２）において、第１項は火力発電機による供給力、第２項は揚水発電と揚水ポンプによる供給力（負荷）、第３項は水力、第４項は融通電力や発電事業者から購入するパターンによる供給力である。左辺は時刻ｔにおける需要である。 A typical constraint condition expression is shown below. Equation (2) is a supply-demand balance constraint that demand and supply power match.

In Equation (2), the first term is the supply power from the thermal power generator, the second term is the supply power (load) from the pumped-storage power generation and the pump, the third term is the hydropower, the fourth term is from the accommodative power and the power generation company It is the supply power by the pattern to purchase. The left side is the demand at time t.

発電機は、運転停止しているときはｕｉｔがゼロとなるために、発電機出力はゼロとなり、運転しているときは出力下限Ｐｉ ｍｉｎと出力上限Ｐｉ ｍａｘの間となる。また、起動停止状態ｕｉｔが０から１のときはその値に応じて出力上下限が変化する。 Equation (3) is a constraint equation for the upper and lower output of the thermal power generator.

Since the generator is zero when the operation is stopped, the generator output is zero, and when the generator is operating, it is between the output lower limit Pi min and the output upper limit Pi max. When the start / stop state uit is from 0 to 1, the output upper and lower limits change according to the value.

運転予備力とは、需要予測誤差あるいは供給力の単一事故が発生したときでも、停電が発生しないように予想需要より多めに最大供給力を確保しておくものであり、式（４）においてＲｔで示されている。火力発電機による出力増加可能分を予備力としている。この出力増加可能分は起動停止状態により変化する。停止しているときは出力上限はゼロで出力もゼロとなるために、出力増加可能分もゼロとなる。 Equation (4) shows the operating reserve constraint conditions.

Operation reserve is to secure a maximum supply capacity that is greater than the expected demand so that a power failure does not occur even if a demand forecast error or a single accident occurs in the supply capacity. Indicated by Rt. The reserve for the increase in output by the thermal power generator is used. This increase in output varies depending on the start / stop state. When it is stopped, the output upper limit is zero and the output is zero, so the amount of increase in output is also zero.

式（５）のＱｔで示されるように、火力発電機の出力からその発電機の最小出力を差し引いた値がその発電機の下げ余力である。 Equation (5) shows the constraint condition for the lowering margin.

As indicated by Qt in the equation (5), a value obtained by subtracting the minimum output of the generator from the output of the thermal power generator is the remaining power of the generator.

  Even in the present situation, when the actual demand becomes smaller than the demand forecast value, it is necessary to balance the supply and demand by suppressing the output of thermal power, which is a variable supply capacity, and suppressing the output of a pump or hydraulic power. Here, only the thermal power is targeted, and the lowering margin is ensured. However, the pumping pump and the hydraulic power suppression may be considered. The government's policy is to introduce a large amount of wind power and solar power in the future. These supply capacities may deviate greatly from the values predicted in advance, and it is important to ensure sufficient reserve capacity when the output becomes much larger than the predicted supply capacities. The same situation occurs when the ratio of nuclear power with zero CO2 emission accompanying power generation increases.

Expression (6) shows an example of the restriction on the time change in the start / stop state, and is an example when the change in demand with time is used.

  Thermal power generators have restrictions on the minimum continuous operation time and the minimum continuous stop time, which are expressed by the second expression of Expression (6). This means that once it is operated (started up), it must be continuously operated for a minimum of several hours, or once stopped, it must be stopped continuously for a minimum of several hours. This constraint condition determines that the start / stop variable does not change for a continuous number of hours (mdt) with a small demand centered on the minimum demand time N. Similarly, a continuous number of hours (mut) with a large demand around the maximum demand time P defines that the start / stop variable does not change.

  Thermal power generators cannot be operated repeatedly starting and stopping every hour considering thermal stress such as boilers. Alternatively, when such an operation is performed, a start-up cost is required, so that the power generation cost is very high. If the start / stop plan of the thermal power generator is planned independently, for example, every hour, the temporal continuity of the plan cannot be considered. For this reason, a change constraint on the start / stop state is set.

  For example, in the demand data of FIG. 5, the demand decreases from 1 o'clock to 5 o'clock, increases from 5 o'clock to 17 o'clock, and then decreases toward 24 o'clock. Such periodicity is in power demand. Here, in order to provide the time continuity of the start / stop plan, a method of making the time change of the output of the thermal power generator the same as the periodicity described above can be considered.

  However, since the demand from 8 o'clock to 10 o'clock increases rapidly, several thermal power generators are started during this time period, so the number of operating units increases, and the output of the generator that was operating continuously increases. May decrease. At this time, although the demand is increasing, the generator output decreases, and the generator output cannot be used in order to consider the continuity of time. Therefore, paying attention to the start / stop variables of the generator, as mentioned earlier, when the demand from 8 o'clock to 10 o'clock increases rapidly, several thermal power generators are started during this time period and operated continuously. The startup / shutdown variable of the generator was still in operation.

  Similarly, when demand decreases. That is, it is possible to consider temporal continuity by changing the start / stop variable of the generator as the demand changes. Here, in order to avoid the start and stop due to a temporary decrease in demand due to the lunch break at 13:00 and a temporary lighting peak that occurs around 18:00 in spring and autumn, the minimum demand time N to the maximum demand time of the daily demand To P, the constraint that the start / stop state does not decrease (increase or the same) is expressed as the first expression of Equation (6), from 1 o'clock to the minimum time N and from the maximum demand time P to 24:00. Thus, the restriction that the start / stop variable does not increase is expressed by the third expression of Expression (6).

  The starting and stopping states that can finally be taken are “1” in the operating state and “0” in the stopping state, but here they are treated as continuous values and are real values between 0 and 1, so the starting and stopping states If the value does not decrease, not only 0 to 1 but also 0.3 to 0.5, etc. are included.

  In the demand increase time zone UPT, the start / stop state does not decrease, the start / stop state does not change during the minimum continuous time centered on the minimum time N, the start / stop state does not change during the minimum continuous operation time centered on the maximum time P, and demand There is a restriction that the start / stop state does not increase in the decrease time zone DNT. It should be noted that by applying the constraint of equation (6), the start-up cost can be calculated without error using the two items of equation (1).

Equation (7) shows the fuel consumption constraint equation. The LNG fuel is transferred to the LNG tank when the LNG ship enters the port. At this time, if there is no free space in the LNG tank, it cannot be transferred from the LNG ship. Further, if the LNG tank becomes empty, the LNG thermal power generator cannot generate power. For this reason, it is necessary to consume LNG fuel systematically, and the upper and lower limits of fuel consumption are required.

  “FLmin” is the lower limit of the output during the planning period of all generators using the target fuel, and “FLmax” is the upper limit of the output during the planning period of all generators using the target fuel. “FL” is a calculated value of the output during the planning target period of all the thermal power generators using the target fuel, that is, the total of Pit in the equation (1). In addition, the said target period is 24 hours as shown in FIG. 5, FIG. 6, for example. Therefore, the value of “FL” is a value obtained by integrating “Pi” over the target period.

  Here, FIG. 19 and FIG. 20 are diagrams showing a case where a plurality of generators in one area are considered, that is, a case where a power flow restriction on transmission lines between different areas is not considered. As shown in FIG. 19, when tidal constraints are not taken into consideration and LNG (Liquid Natural Gas) is considered as a fuel consumption constraint, a generator that is not subject to correction is shown in FIG. It can be considered that the generators LNG201 and LNG202 are replaced with one integrated thermal generator LNG200.

  In this case, the total value of the output upper limit and the output lower limit of each thermal power generator represented by the expression (3) is used as the output upper power limit and the power lower limit of the thermal power generator replaced with the thermal power generator LNG200. In principle, the variable “i” in equation (1) is the number of all generators, and when calculating equation (1), it is necessary to calculate the value of “aitPit2 + biPit + ciuit” for all generators. On the other hand, by consolidating a plurality of generators as described above, the number of variables “i” can be reduced, and the load of calculation processing can be reduced.

  FIGS. 21 and 22 are diagrams illustrating a case where a plurality of generators in a plurality of areas are considered, that is, a case where a power flow restriction on transmission lines between different areas is considered. As shown in FIG. 21, when tidal current restrictions are considered and LNG is considered as fuel consumption restrictions, generators NG201, LNG202, LNG301, and LNG302 that are not subject to correction are shown in FIG. It can also be considered by replacing the generators LNG200 and LNG300 that are aggregated for each area. In this case as well, as in the case of FIGS. 19 and 20, the number of variables can be reduced to reduce the load of calculation processing.

ＰＦｉ ｍｉｎは送電線ｉの潮流下限値、ＰＦｉ ｍａｘは送電線ｉの潮流上限値であり、ＰＦｉは送電線ｉの潮流値である。 Equation (8) shows the tidal current constraint equation. The upper and lower limits of power that can be sent to the transmission lines connecting the areas are determined. Since power flow has a direction, lower limit constraints are also required.

PFi min is a tidal lower limit value of the transmission line i, PFi max is a tidal upper limit value of the transmission line i, and PFi is a tidal value of the transmission line i.

ＷＬｉ ｍｉｎは揚水池あるいはダムｉの水位の下限、ＷＬｉ ｍａｘは揚水池あるいはダムｉの水位の上限、ＷＬｉｔは揚水池あるいはダムｉの水位である。 Equation (9) shows the upper and lower limit constraints of the reservoir or hydraulic dam level. Although the time change of the water level is calculated from the inflow and outflow, this formula is not described here.

WLi min is the lower limit of the water level of the pond or dam i, WLi max is the upper limit of the water level of the pond or dam i, and WLit is the water level of the pond or dam i.

ｕｉ ｍｉｎは、火力発電機ｉの起動停止変数の下限、ｕｉ ｍａｘは火力発電機ｉの起動停止変数の上限、ｕｉｔは火力発電機ｉの起動停止変数である。起動が確定している発電機のｕｉ ｍｉｎおよびｕｉ ｍａｘは１、停止が確定している発電機のｕｉ ｍｉｎおよびｕｉ ｍａｘは０である。未確定の発電機のｕｉ ｍｉｎは０でｕｉ ｍａｘは１である。 Expression (10) is the upper and lower limit constraints of the start / stop variable of the thermal power generator.

ui min is a lower limit of the start / stop variable of the thermal power generator i, ui max is an upper limit of the start / stop variable of the thermal power generator i, and uit is a start / stop variable of the thermal power generator i. The ui min and ui max of the generator whose start is determined is 1, and the ui min and ui max of the generator whose stop is determined is 0. The undefined generator ui min is 0 and ui max is 1.

  In addition, by setting the uit whose start is determined by the start / stop variable to be 1 and the uit whose stop is determined to be 0, the equations (1), (3), (4), (5), By reducing the variables of the equations (6) and (10), the optimization processing unit 27 can speed up the optimization calculation processing S108 and processing S110 by mathematical programming.

  The process S108 is an optimization calculation process based on mathematical programming, and is processed by the optimization processing unit 27. In S108, the optimization processing unit 27 reads out the objective function and the constraint conditions such as the above expressions (1) to (10) from the objective function / constraint condition storage area M110 and reads them from the solver storage area M111. Input into an optimization program of mathematical programming and execute optimization calculation. The optimization calculation is to obtain the values of “Pi” and “ui” that satisfy each of the above-described constraints and minimize the calculated cost TC.

  As described above, in the calculation of S108, the start / stop variable ui of the generator to be corrected easily converges to “0”, and the start / stop variable ui of the non-target generator easily converges to “1”. Each coefficient has been changed. Thereby, in the calculation of S108, the start / stop variable ui is selected so that the start / stop variable ui of the generator to be corrected converges to “0” as much as possible and the output of the other generator approaches the output upper limit. It is. The optimization processing unit 27 stores the power plant operation plan obtained as a result of the optimization calculation in the current solution storage area M104, and outputs the calculation result to the display device 7 by the calculation result output unit 32. FIG. 11 shows an example of the screen output displayed on the output device 7 for the start / stop variable of the relaxed power plant operation plan.

  In FIG. 10, the generator 7 is relaxed to a real variable from 9:00 to 22:00, but in the result of FIG. 11, it is 0 stop from 9:00 and 17:00 to 22:00, and from 10:00 to 16:00. The values up to 0 are larger than zero. That is, in the initial plan, the generator 7 has not been determined to run / stop from 9:00 to 22:00, but the start / stop variable ui is “0” from 9:00 and 17:00 to 22:00 according to the calculation of S108. It turned out that it was possible to stop the apparatus.

  Process S 109 is a thermal power generator activation / deactivation number determination process, which is executed by the activation / deactivation determination unit 28. The start / stop execution unit 28 reads the start / stop variable from the current solution storage area M104, replaces the start / stop variable larger than 0 with 1, and stores the result in the current solution storage area M104. As a result, the start / stop variable in FIG. 11 is “1” indicating operation from 10:00 to 16:00 as shown in FIG. That is, in the present embodiment, the real number relaxation unit 26, the optimization processing unit 27, and the activation stop determination unit 28 function as an activation state calculation unit.

  When the process of S109 is completed, the objective function and the constraint condition formula are created based on the start / stop variables of the generator characteristic storage area M101, the constraint condition storage area M102, and the current solution storage area M104, and the objective function / constraint condition is created. Store in the storage area M110.

  The process S110 is the same process as the process S108, and is executed by the optimization processing unit 27. However, in S110, since the formal power generation cost is calculated based on the result of S108, the original value is used for the coefficient and start-up cost of the fuel cost function instead of the values corrected in FIG. Further, as described above, the current solution storage area M104 stores the start / stop variable updated by the processing of S108 and S109. Therefore, the updated start / stop variable is used in the processing of S110. Above, the power generation cost is calculated. That is, in S110, the optimization processing unit 27 functions as a power generation cost calculation unit.

  Furthermore, in S110, since the purpose is to calculate the official power generation cost, the method of reducing the calculation processing load by aggregating a plurality of generators that are not subject to correction described in FIGS. Cannot be used. The calculated power generation cost is stored in the current solution storage area M104.

  The best solution check unit 30 executes the process S111. The best solution check unit 30 reads the power generation cost of the current solution storage area M104, reads the power generation cost of the best plan storage area M105, compares the power generation cost of the current solution with the power generation cost of the best plan, and determines the power generation cost of the current solution. If it is cheaper, the process proceeds to step S112, and if not, the process proceeds to step S113. Here, the generator check end flag of the generator that is the start / stop change target stored in the generator group storage area M108 is set to zero.

  The best solution check unit 30 executes the process S112. The best solution check unit 30 stores the power plant operation plan in the current solution storage area M104 in the best solution storage area M105. The generator number corrected when the best plan is reached is also stored.

  The best solution check unit 30 executes the process S113. The best solution check unit 30 reads the generator check end flag of the group generator currently being checked in the generator group storage area M108, and if the generator check end flag is 1, only the current target generator is a candidate. It is determined that all are checked, and the process proceeds to step S114. Otherwise, the generator check end flag of the current target generator is set to 0.

  The process S114 is executed by the optimal solution check unit 31. The optimal solution check unit 31 reads the power generation cost of the best solution storage area M104, reads the power generation cost of the optimal plan storage area M105, compares the power generation cost of the best solution with the power generation cost of the optimal plan, and determines the power generation cost of the best solution. If it is cheaper, the process proceeds to process S115, and if not, the process proceeds to process S116. When the process proceeds to step S115, the update flag of the generator group storage area M108 of the generator number when the best plan of the best solution storage area M104 is set is set to 0.

  The optimum solution check unit 31 executes the process S115. The optimum solution check unit 31 stores the power plant operation plan in the best solution storage area M104 in the best solution storage area M105. That is, in S111, S112, S114, and S115, the best solution check unit 30 and the optimum solution check unit 31 function as an activation plan update unit. The optimal solution check unit 31 executes the process S116. The optimum solution check unit 31 reads the generator group storage area M108, determines whether there is a group check end flag other than the current group, and proceeds to processing S117 if there is none, otherwise Set the group check end flag of the group to 0.

  The process S117 is a process for outputting the optimum plan, and is executed by the output result output unit 32. The output calculation result output unit 32 reads the optimum operation plan from the optimum solution storage area M106 and outputs the start / stop plan, the power generation cost, the output of each generator, or the like to the display device 7. Moreover, as a calculation result so far, FIG. 13 is a screen which displays the power generation cost when the start-up / stop plan of the generator corrected in each group is corrected. FIG. 14 shows the transition of the power generation cost every time the optimum solution is updated (the number of repetitions). FIG. 15 shows the constraints and the values that are the upper limit or lower limit of the constraint conditions for the optimum operation plan or the operation plan created in step S110, for confirming which constraint is severe. Is.

  According to the embodiment of the present invention, at most one generator start / stop variable is relaxed to a real number at one time, and the fuel cost function and start cost of the generator at the relaxed time are set to be higher than the original values. Set the fuel cost function and start-up cost of the generator for which start / stop is confirmed to be lower than the original values, create an optimal start-stop plan by quadratic programming method or linear programming method, If the optimum value of the stop variable is larger than 0, the start-up plan is created by determining the start-up 1 and otherwise the stop-off 0. It is possible to create a power generation commercial plan that satisfies all the constraints and minimizes the power generation cost by repeating the relaxation and determination of the start / stop variables similar to the above for the obtained optimal start / stop plan. it can.

  In the embodiment, as described with reference to FIG. 7, a case where a correction value such as a coefficient is manually input in advance has been described as an example. In addition, a mode in which the coefficient of the target generator is automatically multiplied by an integer in S107, or a mode in which the coefficient of the non-target generator is automatically set to zero is also possible.

  Further, in the above embodiment, when the optimal solution is obtained in the generator with the generator number “7” as a result of calculation for the group 2 as the correction target, the start / stop variable of the generator “7” is set to a predetermined value. A start / stop plan with “0” in time is used as the optimal solution. Here, instead of targeting group 1 immediately after that, the generator check end flag of the generator of “7” is set to “0”, and the same process is repeated again with group 2 as the correction target. Thus, it is preferable to confirm whether the generators “6” and “8” can be further stopped. Such an aspect can be realized by the update processing of the generator check end flag and the group check end flag in S115.

1 Power plant operation plan creation device,
2 CPU,
3 main memory,
4 input devices,
5 external storage device,
6 input devices,
7 Display device,
8 printing device,
9 database,
10 reading device,
11 Power system,
12 Power system,
20 calculation processing unit,
21 Control part,
22 Planning data reading section,
23 Plan creation strategy setting department,
24 Initial plan reading part,
25 Target generator determination unit,
26 Real number relaxation part,
27 Optimization processing unit,
28 Start / stop determination unit,
30 Best solution check section,
31 Optimal solution check section,
32 Calculation result output section

Claims (16)

A generator operation plan determination device that determines a startup state of a plurality of generators based on a time-series predicted value of power demand,
A demand forecast acquisition unit for obtaining a demand forecast that is information of time series forecast values of power demand;
An initial plan for acquiring a generator start plan that is information in which an initial value of a start state value, which is information indicating a state of either start or stop of each of the plurality of generators by a numerical value of 1 or 0, is determined. An acquisition unit;
A correction target determination unit that determines a target generator that is a generator to be corrected to correct the startup state value based on the generator startup plan;
A formula for calculating a power generation cost for calculating a power generation cost for an output for each of the plurality of generators, and the demand forecast is satisfied and calculated in the time series based on a power generation cost calculation formula including the activation state value An activation state calculation unit that obtains activation state values of the plurality of generators in time series so that the cost to be minimized is included,
When the activation state calculation unit obtains the activation state value in time series, the activation state value of the target generator is likely to converge to 0, and activation of an untargeted generator that is a generator other than the target generator The power generation cost calculation formula is calculated by changing the coefficient of the power generation cost calculation formula so that the state value easily converges to 1 and setting the start state value constraint as a real number from 0 to 1, and starting the target generator The generator operation plan determination device, wherein the state value is 1 when calculated as a value other than 0. The initial value of the activation state value defined in the generator activation plan includes a real number greater than 0 and less than 1 that is a value indicating that activation / deactivation of the generator is not confirmed,
2. The generator operation plan determination device according to claim 1, wherein the correction target determination unit determines a generator having a startup state value other than 0 and 1 in the initial plan as a target generator. .   The initial plan acquisition unit calculates the power generation cost calculation formula so that the start state value constraint is a real number from 0 to 1, and the demand prediction is satisfied and the calculated cost is minimized in the time series. The generator operation plan determination apparatus according to claim 2, wherein the startup state value obtained in this way is acquired as the initial plan.   The activation state calculation unit changes a coefficient of a power generation cost calculation formula of the target generator so that a power generation cost of the target generator is calculated to be larger than an original power generation cost calculation formula. Item 4. The generator operation plan determination device according to any one of Items 1 to 3.   The activation state calculating unit is configured to change a coefficient of a power generation cost calculation formula of the non-target generator so that a power generation cost of the non-target power generator is calculated smaller than an original power generation cost calculation formula. The generator operation plan determination apparatus according to any one of claims 1 to 4.   The said starting state calculation part changes the coefficient of the power generation cost calculation formula of the said non-target generator so that the power generation cost of the said non-target generator is calculated as zero. Generator operation plan decision device. Applying the activation state value calculated by the activation state calculation unit and the activation state value in the generator activation plan to the power generation cost calculation formula so that the demand prediction is satisfied and the calculated cost is minimized. A power generation cost calculation unit for calculating the power generation cost of the plurality of generators;
When the power generation cost calculated by applying the calculated start state value is lower than the power generation cost calculated by applying the start state value in the generator start plan, the calculated start state value causes the The generator operation plan determination device according to any one of claims 1 to 6, further comprising a startup plan update unit that updates the generator startup plan.   The generator operation plan determination device according to claim 1, wherein the power generation cost calculation formula is based on at least one of a linear programming method, a quadratic programming method, and an equal lambda method.   9. The power generation according to claim 1, wherein the correction target determination unit divides the plurality of generators into a plurality of groups having different priorities, and determines the target generator according to the priorities. Machine operation plan decision device. The initial value of the activation state value defined in the generator activation plan includes a real number greater than 0 and less than 1 that is a value indicating that activation / deactivation of the generator is not confirmed,
The correction target determination unit divides generators whose start is confirmed and generators whose start / stop is not determined into different groups, and sets the priority of the group of generators whose start / stop is not determined 10. The generator operation plan determination apparatus according to claim 9, wherein the generator operation plan determination apparatus is set to be higher than the priority order of the determined generator group.   The correction target determination unit includes a generator that is determined to stop in at least a part of the time series of the generator startup plan and a generator that is determined to start in all time. Divide into different groups and set the priority of the generator group that has been confirmed to stop for a certain period higher than the generator group that has been confirmed to start all the time The generator operation plan determination device according to claim 9 or 10.   The correction target determination unit determines, as the target generator, a generator that is determined to stop in at least a part of a period in the time series of the generator startup plan. 11. The generator operation plan determination device according to any one of 11 above.   The activation state calculation unit obtains activation state values of the plurality of generators so that the demand prediction is satisfied and the calculated cost is minimized, and further, an upper limit value and a lower limit value of an output in each generator are satisfied. And the total of the upper and lower limits of the output of each of the two or more generators that are part of the plurality of non-target generators, and the two or more generators are virtually used as one generator to generate the power generation cost. The generator operation plan determination device according to any one of claims 1 to 12, wherein a calculation formula is calculated.   The activation state calculation unit virtually sets the two or more generators when there is a fuel consumption restriction with respect to a total of fuel consumed by the two or more generators that is a part of the plurality of non-target generators. The generator operation plan determination apparatus according to claim 13, wherein the power generation cost calculation formula is calculated so that the fuel consumption constraint is satisfied as one generator. A generator operation plan determination method for determining a startup state of a plurality of generators based on a time-series predicted value of power demand,
Get demand forecast, which is information of time series forecast value of power demand,
Obtaining a generator start plan which is information in which an initial value of a start state value which is information indicating either a start or stop state of each of the plurality of generators by a numerical value of 1 or 0,
Determine the target generator that is the generator to be corrected to correct the startup state value based on the generator startup plan,
A formula for calculating a power generation cost for calculating a power generation cost for an output for each of the plurality of generators, and the demand forecast is satisfied and calculated in the time series based on a power generation cost calculation formula including the activation state value The starting state values of the plurality of generators are determined in time series so that the cost to be generated is minimized,
When obtaining the activation state value in time series, the activation state value of the target generator easily converges to 0, and the activation state value of the non-target generator that is a generator other than the object generator converges to 1. The power generation cost calculation formula is calculated by changing the coefficient of the power generation cost calculation formula to make it easier to set the activation state value constraint as a real number from 0 to 1,
The generator operation plan determination method according to claim 1, wherein when the activation state value of the target generator is calculated as a value other than 0, the target generator is set to 1. A generator operation plan determination program for determining the startup states of a plurality of generators based on time-series predicted values of power demand,
Obtaining a demand forecast that is information of time series forecast values of power demand;
Obtaining a generator activation plan that is information in which an initial value of an activation state value, which is information indicating one of the numerical values of 1 and 0, indicating either the activation or deactivation state of each of the plurality of generators; ,
Determining a target generator that is a generator to be corrected to correct the startup state value based on the generator startup plan; and
A formula for calculating a power generation cost for calculating a power generation cost for an output for each of the plurality of generators, and the demand forecast is satisfied and calculated in the time series based on a power generation cost calculation formula including the activation state value Determining the startup state values of the plurality of generators in time series so that the cost to be generated is minimized,
When obtaining the activation state value in time series, the activation state value of the target generator easily converges to 0, and the activation state value of the non-target generator that is a generator other than the object generator converges to 1. Changing the coefficient of the power generation cost calculation formula so as to facilitate the calculation of the power generation cost calculation formula with the constraint of the activation state value as a real number from 0 to 1,
A generator operation plan determination program that causes the generator operation plan determination device to execute a step of 1 when the activation state value of the target generator is calculated as a value other than 0.

Images