Operating plan planning system for planning an operating plan for a power plant

Abstract

Planning system (1) for planning an operating schedule (100, 100D) of a power plant over a predetermined period of time, wherein the power plant is capable of generating energy by driving an energy generator using a turbine, and of executing any of an energy-selling operating mode which sells energy by operating the power plant in a state in which the turbine and the energy generator are connected, a zero-load operating mode which operates the power plant in a state in which the turbine and the energy generator are not connected, and a stop operating mode which stops the turbine, wherein the planning system (1) comprises: a determination unit (22) configured to determine whether a predicted energy selling price (Pe1, Pe2, Pe3) is higher than energy production costs (CP); and a planning unit (23) configured to calculate a loss index (J) for each of the energy selling operating mode, the zero-load operating mode, and the stop operating mode in the event that the predicted energy selling price (Pe1, Pe2, Pe3) is equal to or less than the energy production cost (CP) within the predetermined time period, using equation (1) based on an energy selling loss (J1) which is the difference between the expected energy selling price and the energy production cost (CP), a maintenance loss (J2), and an operating loss (J3) that does not contribute to an energy sale, and to schedule the operating plan (100, 100D) such that the power plant operates in an operating mode in which the loss index (J) is smallest, where the loss index (J) corresponds to the energy sales operating mode and includes the energy sales loss (J1) and the maintenance loss (J2), where the loss index (J) corresponds to the zero-load operating mode and the stop operating mode, respectively, includes the operating loss (J3) and the maintenance loss (J2), wherein the maintenance loss (J2) includes an operating time loss which consumes a durability based on an operating time of a component that configures the power plant, according to an accumulated operating time of the power plant, and a frequency of a start loss which consumes a frequency based on the frequency of starting the component according to an accumulated frequency of starting the power plant, where the maintenance loss (J2) of each of the energy sales operating mode and the no-load operating mode has the operating time loss as a major factor, where the maintenance loss (J2) of the stop mode has the frequency of a start loss as a major factor, and where the planning unit (23) is configured to: to determine the maintenance loss (J2) of each of the energy sales operating mode and the zero load operating mode using a first weighting coefficient (k1); to determine the maintenance loss (J2) of the stop mode using a second weighting coefficient (k2); to decrease the first weighting coefficient (k1) and increase the second weighting coefficient (k2), or to decrease the first weighting coefficient (k1) and increase the second weighting coefficient (k2) if the ratio of the accumulated frequency of a start to the frequency of a start increases; and to increase the first weighting coefficient (k1), decrease the second weighting coefficient (k2), or to increase the first weighting coefficient (k1) and decrease the second weighting coefficient (k2) if the ratio of accumulated operating time to durability relative to operating time increases, wherein Loss index = Energy sales loss + Maintenance loss + Operating loss that does not contribute to energy sales

Description

Technical field of the invention

[0001] The present invention relates to an operation plan planning system for a power plant. Background of the invention

[0002] In recent years, energy and electricity liberalization has been promoted. There are countries where approximately 30% of energy and electricity consumption is traded on a wholesale energy or electricity market. While the structure of a wholesale energy or electricity market varies from country to country, there is, for example, a market called the day-ahead market (DAM), where bidding takes place the day before for each unit of time. The selling price for energy or electricity on a wholesale energy or electricity market is determined based on the balance between supply and demand for energy or electricity and can change significantly over time.

[0003] Therefore, in order to obtain a return by participating in a bidding, it is important for an energy or power generation company to conduct optimal operations according to an energy or power sales price and fluctuations in energy or power demand.

[0004] Patent Document 1 discloses a technology that includes a step of setting a power sales price for each period, a step of detecting each operating condition that can be achieved by a power plant for each period, registering a yield for each transmission path that reaches the operating condition, and finally scheduling an operation plan of the power plant based on a transmission path that can maximize the yield. According to this technology, a yield for each period is obtained by reducing a yield generated by selling energy or electricity by a loss caused by power generation fluctuations (an event in which a contracted power generation amount and an actual power generation amount are different from each other).

[0005] Patent Document 2 discloses a power generation plan / power sales plan generation method that includes a process for acquiring a probability distribution according to uncertainty factors (fluctuations in energy demand, fluctuations in the unit price of fuel for power generation, and the like) for compensation generated according to power generation and power sales, and a process for displaying the probability distribution in a time order. According to this method, it is disclosed that a correlation between a return of a current-state operation plan and a risk can be acquired, and a search for a plan for resolving the risk of the current-state operation plan can be performed.

[0006] In Patent Document 3, a power plant operation support system has been disclosed, which includes an operating condition optimization calculation unit that calculates an optimal operating condition at which the power generation output value of a power plant is maximum based on a probability distribution of predicted power demand values ​​and predicted energy trading market price values, a risk evaluation unit that calculates and evaluates a risk value for damaging the optimization of the optimal operating condition, and a reservation power optimization unit.

[0007] The reservation energy optimization unit calculates optimal reserved energy of a power generation quantity capable of responding to a customer's load increase request in a short time, and calculates load conditions that are optimal in terms of cost for maintaining the reserved energy. One of these optimal load conditions is a rated overload, and Patent Document 3 describes that operating costs increase during operation under a rated overload, such as a reduction in device durability due to unnecessary loading or an increase in maintenance costs due to the application of a load that is not required for a facility.

[0008] Patent Document 3 discloses a power plant management system that plans an operation schedule such that a power generation yield is optimal with respect to durability consumption costs and the risk of unplanned plant shutdown (stop due to a malfunction).

[0009] Here, shelf life consumption costs are determined based on the amount of remaining shelf life consumed for shelf life costs (costs incurred until replacement after a certain component has been consumed from the time of a new product (or new acquisition time)), and are calculated as a product of a remaining shelf life consumption ratio (a ratio at which the remaining shelf life is consumed) and shelf life costs (shelf life consumption costs = remaining shelf life consumption ratio × shelf life costs).

[0010] Patent Document 4 discloses a control method for optimizing load shedding operation for a power plant with multiple generator units. The multiple generator units can each include either an on condition or an off condition during a selected operating period.The method may include: defining competing operating modes, wherein the competing operating modes include various possible combinations with respect to which of the generator units are in the on-state and which are in the off-state during the selected operating period; defining multiple cases for each of the competing operating modes, wherein the multiple cases include varying a value of an operating parameter; receiving performance targets including a cost function; receiving an environmental condition forecast; simulating the load shedding operation of the power plant with a power plant model given the value of the operating parameter and the environmental condition forecast; evaluating a simulation result according to the cost function to select a preferred case therefrom.

[0011] Patent Document 5 discloses that an optimal operation program calculation method and the apparatus therefor are provided with a market price forecasting part that forecasts the respective electricity prices in one or more electricity markets, a market determination part that determines a market whose forecast market price is the highest as the supply market, and an optimal operation program calculation part that calculates the cost of electricity generated by the respective generation plants and supplied to the determined market under the conditions of the intrinsic operation constraints of the respective generation plants and outputs an operation program that achieves a maximum profit based on the differences between the forecast market prices and costs.

[0012] Patent Document 6 discloses a method for controlling a gas turbine power plant, which comprises the steps of: setting the relationship between a predetermined reference time (e.g., ignition time) in a predetermined process (e.g.,the starting process) and a rotational speed of a turbine shaft according to a first function; prescribing the relationship between a rotational speed of the turbine shaft and a flow rate of fuel to be supplied to a combustor according to a second function; measuring an actual rotational speed of the turbine shaft at the elapsed time; rotating the turbine shaft by a starting device to drive the turbine shaft in such a manner that the measured actual rotational speed becomes the rotational speed corresponding to the elapsed time prescribed according to the first function; and obtaining a flow rate of fuel corresponding to the actual rotational speed according to the second function to supply the obtained fuel flow rate to the combustor.

[0013] Patent Document 7 discloses a system including a consulting system with a loss calculation engine configured to derive a total system loss for an industrial plant based on a first sensor positioned in a first industrial plant component and a first physical model of the first industrial plant component. The consulting system further includes a cost model configured to use a cost function to derive a cost based on the total system loss, and a control strategy system configured to derive a consulting report, a control correction factor, or a combination thereof based on the cost. A control system is configured to apply the control correction factor to control a process in the industrial plant.

[0014] Patent Document 8 discloses that a power procurement planning system is provided with a database that stores cost characteristic information showing a characteristic of the cost with respect to the power procurement amount.

[0015] Patent Document 9 discloses that an evaluation period setting part sets a target evaluation period. A power generation planning part establishes a power generation schedule, a bidding planning part establishes a bidding schedule, a power market transaction determination part determines a transaction amount and a transaction price, and a daytime power generation planning part establishes a daytime power generation schedule of a forecasted transaction day. A revenue calculation part performs revenue calculation based on the transaction amount, transaction price, and daytime power generation schedule of the forecasted transaction day, and a facility evaluation part evaluates the revenues of each generation set based on a revenue calculation result and power supply data.A scenario evaluation part lets an operation part simulate a power transaction at each data set based on a plurality of data sets determined in consideration of the difference in conditions and selected in a data management part, and evaluates the result of the simulation based on the difference in conditions. Patent specifications Patent Document 1: JP 2002- 281 665 A Patent specification 2: US 2004 / 0 181 491 A Patent specification 3: US 2005 / 0 015 283 A Patent specification 4: DE 10 2015 120 446 A1 Patent document 5: JP 2001- 86 645 A Patent specification 6: US 5,966,925 A Patent specification 7: US 2014 / 0 257 526 A1 Patent document 8: JP 2006- 271 148 A Patent document 9: JP 2007- 159 239 A

[0016] In addition, the following state of the art is available: MITRA, S.; SUN, L.; GROSSMANN, IE: “Optimal scheduling of industrial combined heat and power plants under time-sensitive electricity prices”; In: Energy, Issue 54, 2013, pages 194-211. Summary of the inventionTechnical problem

[0017] In the technology disclosed in Patent Document 1, while planning a power plant's operation schedule with a loss caused by focused power generation fluctuations, potential losses such as hot part consumption according to the power plant's operation and startup load are not considered. If a power plant's operation schedule is planned without considering such potential losses, a short-term profit can be obtained. However, the replacement interval of consumables and various maintenance intervals of the power plant are shortened, and in the case where a long-term profit / loss is considered, a high-profit operation schedule cannot be planned.

[0018] Furthermore, when considering a starting load or the amount of fuel consumed from the start of operation to the stabilization of operation, there are also cases where, during a period in which power sales are not performed, a no-load operation in which the operation of a turbine continues in a state where the turbine and a generator are not connected has a loss smaller than that in the case where the power plant is temporarily stopped when considering a long-term income / loss. However, in the technologies disclosed in Patent Documents 1 to 3, no-load operation is not considered.

[0019] The present invention has been made in consideration of the problems described above, and an object of the invention is to provide an operation plan planning system for a power plant and an operation plan planning method for a power plant, whereby an operation plan with high yield can be planned. Solution to the problem

[0020] According to the present invention, a planning system is provided according to independent claim 1. Advantageous modifications can be found in dependent claims 2 to 7. Advantageous effects of the invention

[0021] According to the present invention, for a profit period in which a predicted energy sales price is higher than power generation costs, an energy sales mode is selected, and for a loss period in which a predicted energy sales price is equal to or less than power generation costs, an operation mode with a lowest loss index is selected from the energy sales mode, a no-load mode, and a stop mode. Since the loss index considering maintenance loss and operation loss does not contribute to energy sales in addition to energy sales loss, a high-yield operation plan can be planned taking into account losses such as potential maintenance loss and operation loss that do not contribute to energy sales. Short description of the drawings Fig. 1 is a functional block diagram illustrating the configuration of an operation plan planning system for a power plant according to a first embodiment of the present invention. Fig. 2 is a schematic diagram illustrating a planning unit according to the first embodiment of the present invention, and it is a diagram in which a predicted energy sales price line L_Pe (predicted energy sales price Pe) and energy generation cost Cp are plotted on a graph with the energy unit price as the horizontal axis and time as the horizontal axis, and operation plans are coordinated with each other. Fig. 3 is a schematic diagram illustrating a predicted power sales price Pe, power generation cost Cp, power generation load Le, turbine rotation speed ωt, and a calculation equation of a loss index J to be adjusted in a power sales mode according to the first embodiment of the present invention. Fig. 4 is a diagram illustrating a predicted power sales price Pe, power generation cost Cp, a power generation load Le, a turbine rotation speed ωt, and a calculation equation of a loss index J to be adjusted in a no-load mode according to the first embodiment of the present invention. Fig. 5 is a diagram illustrating a predicted power sales price Pe, power generation cost Cp, power generation load Le, turbine rotation speed ωt, and a calculation equation of a loss index J to be adjusted in a stop mode according to the first embodiment of the present invention. Fig. 6 is a schematic flowchart illustrating an operation plan scheduling method of a power plant according to the first embodiment of the present invention. Fig. 7A and Fig. 7B are maps used for determining weighting coefficients according to a second embodiment of the present invention. Fig. 8 is a schematic diagram illustrating an additional function of a planning unit according to a third embodiment of the present invention, and is a diagram in which predicted energy sales prices Pe1, Pe2, and Pe3 and energy generation costs Cp are plotted on a graph with the energy unit price as the vertical axis and time as the horizontal axis, and operation plans to be adjusted are shown. Description of implementation examples

[0022] Embodiments of the present invention will be described below with reference to the drawings.

[0023] In each embodiment described below, an example is described in which an operation plan planning system for a power plant is applied to plan an operation plan of the power plant over 24 hours (predetermined period of time) when a bid of a day-before market is performed.

[0024] The embodiments described below are merely examples, and are not intended to preclude the application of various modifications or various technologies not clearly illustrated in the embodiments described below. Configurations of the embodiments shown below can be modified in various ways within a range that does not deviate from the concepts thereof, and they can be selected or combined as needed. [1. Embodiment][1-1. Configuration]

[0025] The overall configuration of an operation plan planning system for a power plant according to a first embodiment of the present invention will be described with reference to Fig. 1 to 5 described.

[0026] Fig. 1 is a functional block diagram illustrating the configuration of an operation plan planning system for a power plant according to the first embodiment of the present invention.

[0027] Fig. 2 is a diagram in which a predicted energy sales price line L_Pe (predicted energy sales price Pe) and energy generation cost Cp are plotted on a graph with the energy unit price as the horizontal axis and time as the horizontal axis, and operation plans are coordinated.

[0028] Fig. 3 is a diagram illustrating a predicted energy sales price Pe, energy generation cost Cp, energy generation load Le, turbine rotation speed ωt, and a calculation equation of a loss index J to be matched in an energy sales mode.

[0029] Fig. 4 is a diagram illustrating a predicted energy sales price Pe, energy generation cost Cp, energy generation load Le, turbine rotation speed ωt, and a calculation equation of a loss index J to be matched in a no-load operation mode.

[0030] Fig. 5 is a diagram illustrating a predicted energy sales price Pe, energy generation cost Cp, energy generation load Le, turbine rotation speed ωt, and a calculation equation of a loss index J to be matched in a stop mode.

[0031] The operation plan planning system for a power plant (hereinafter also referred to as a planning system) 1 according to this embodiment, as shown in Fig. 1 is configured to include: a planning device 2, an input unit (input device) 3 that inputs various types of information into the planning device 2, and a display unit (display device) 4 that outputs various types of information from the planning device 2.

[0032] In addition, the input unit 3 and the display unit 4 may be connected to the planning device 2 through a communication line such as the Internet or the like.

[0033] The planning device 2 is configured to include: a predicted energy sales price setting unit 21; a determination unit 22; a planning unit 23; and a storage unit 29. The storage unit 29 stores various programs and data required to control the predicted energy sales price setting unit 21, the determination unit 22, and the planning unit 23. In this embodiment, while the planning device 2 is configured by a single device (in other words, the predicted energy sales price setting unit 21, the determination unit 22, the planning unit 23, and the storage unit 29 are configured by a control function of the same device), the configuration is not limited to this.For example, the predicted energy sales price setting unit 21, the determination unit 22, the scheduling unit 23, and the storage unit 29 may be configured via a plurality of devices, and the plurality of devices may be connected by a communication line such as the Internet.

[0034] The predicted energy sales price setting unit 21, the determination unit 22 and the planning unit 23 are described below.

[0035] The predicted energy sales price setting unit 21 predicts a predicted energy sales price line L_Pe of a predicted target date (here, the next day of the bidding) based on a past result. The predicted energy sales price Pe changes with respect to time, and the predicted energy sales price line L_Pe represents a 24-hour transition of the predicted energy sales price Pe. Here, the predicted energy sales price line L_Pe is a step diagram formed by connecting predicted energy sales prices Pe estimated at a one-hour interval (see FIG. Fig. 2). The predicted energy selling price Pe is acquired by correcting an average of past results based on a predicted temperature and weather of a predicted target day.

[0036] Here, the predicted energy sales price setting unit 21 is not indispensable, but the predicted energy sales price information may be acquired from the outside and stored in the storage unit 29.

[0037] The determination unit 22 compares a predicted energy sales price Pe with the energy generation cost Cp per unit energy quantity for each hour and determines whether the predicted energy sales price Pe is higher than the energy generation cost Cp or not. The energy generation cost Cp is determined according to various specifications (the shape of the power plant, consumed fuel, the type of turbine drive source, and the like) of the power plant. Here, the energy generation cost Cp is appropriately input from the input unit 3 and stored in the storage unit 10. In the case where the energy generation cost Cp is predicted to be constant over a long period of time, the energy generation cost Cp may be stored in the storage unit 10 in advance as a fixed number.

[0038] The planning unit 23 performs selection control that selects an operation mode of a power plant for each unit time (here, for each hour), thereby planning an operation schedule over 24 hours (predetermined period).

[0039] This describes the operating mode of the power plant. The power plant generates energy by driving a power generator using a turbine. It represents a combined cycle power plant (GTCC) including a gas turbine, a flue gas boiler, a steam turbine, and a power generator. In the following description, a gas turbine and a steam turbine are also referred to as turbines.

[0040] There are three operating modes that can be selected by the power plant, including a power generation mode, a no-load mode, and a stop mode.

[0041] The power generation mode is a mode that operates in a state where the turbine is connected to the power generator (in other words, generating power or electricity). In the power generation mode, energy or electricity is generated, and accordingly, energy sales can be performed. Therefore, the power generation mode is hereinafter referred to as the power sales mode.

[0042] No-load mode is a mode of operation in which the turbine is not connected to the power generator (in other words, the turbine operates in a state where no load is connected). In no-load mode, no energy is sold because the power generator is not driven (no energy is generated).

[0043] Stop mode is an operating mode in which the fuel supply to the gas turbine is stopped, and the turbine and consequently the power generator are stopped. In stop mode, no energy is sold because the power generator is stopped (no power is generated).

[0044] In the following, planning unit 23 is described in particular with reference to Fig. 2 described.

[0045] The planning unit 23 selects the energy sales mode as an operation mode based on a result of the comparison performed by the determination unit 22 for a period of time (in other words, a revenue period te_p in which a revenue can be expected when energy is generated) in which the predicted energy sales price Pe is higher than the energy generation cost Cp. In the Fig. In the example shown in Figure 2, a yield period te_p is determined for each of a period from 7 a.m. to 5 p.m. and a period from 6 p.m. to 8 p.m., and the energy sales mode is selected.

[0046] On the other hand, the planning unit 23 selects one of the power sales mode, the no-load mode, and the stop mode as an operation mode based on a result of the comparison performed by the determination unit 22 for a period (in other words, a loss period te_r in which a loss occurs when power is generated) in which the predicted power sales price Pe is equal to the power generation cost Cp or less, by using a loss index (loss amount) J calculated using the following equation (1).

[0047] Specifically, the scheduling unit 23 calculates a loss index J using a power sales loss J1, a maintenance loss J2, and an operation loss (hereinafter simply referred to as operation loss) J3 that does not contribute to power sales, using the following equation (1) for each of three cases including a case where the power sales mode is performed, a case where the no-load mode is performed, and the stop mode is performed (for the case where the power plant is stopped). Then, the scheduling unit 23 selects an operation mode whose loss index J is the smallest. Loss index J=Energy sales loss J1+Maintenance loss J2+Operating loss J3

[0048] Here, the energy sales loss J1 is a loss (a so-called negative difference) that occurs when the predicted energy sales price Pe is lower than the energy generation cost Cp in the case where energy sales are performed. The maintenance loss J2 is a loss (a consumption loss of the durability of a part) when an increase in cost occurs as a result of replacing a required component according to a consumption of a component (part), in other words, a consumption of a durability, or the like. The operation loss represents fuel costs and the like consumed at the time when they do not contribute to energy sales (energy generation), such as at the time of starting up or at the time of performing no-load operation.

[0049] In the Fig. In the example shown in Figure 2, a loss period te_r is determined by the scheduling unit 23 for three periods of 0:00 a.m. to 7:00 a.m., 5:00 p.m. to 6:00 p.m., and 8:00 p.m. to midnight, and an operation mode is selected based on the loss index J for each loss period te_r. Accordingly, the stop operation mode is selected for 0:00 a.m. to 7:00 a.m., the energy sales operation mode is selected for 5:00 p.m. to 6:00 p.m., and the stop operation mode is selected for 8:00 p.m. to midnight. For the other periods, in other words, for the revenue period te_p, the energy sales operation mode is selected as an operation mode as described above.

[0050] In other words, in the Fig. In the example shown in Figure 2, a 24-hour operation plan 100 is planned in which the stop mode is set for 0:00 to 7:00, the energy sales mode is set for 7:00 to 20:00, and the stop mode is set for 20:00 to 24:00.

[0051] In addition, in the event that the Fig. 2, bidding is normally carried out from 7 a.m. to 8 p.m., during which energy is generated, and no bidding is carried out at other times.

[0052] In the following, the calculation of the loss index J in the energy sales mode, the no-load mode and the stop mode is further explained with reference to Fig. 3 to 5.

[0053] First, the energy sales mode is defined with reference to Fig. 3. In the case where the power sales mode is selected in the loss period te_r in which the predicted power sales price Pe is lower than the power generation cost Cp, the driving of the power generator continues according to the operation of the gas turbine (GT), and the turbine rotation speed ωt and the power generation load Le have values ​​of predetermined values ​​or greater. For this reason, no operation loss J3, which represents a cost, occurs at the time of startup or accompanying no-load operation. Accordingly, the loss index J in the power sales mode is a sum of the power sales loss J1 and the maintenance loss J2.

[0054] More specifically, the loss index J is calculated using the following equations (2) to (4) by using a loss ΔC (= power generation cost Cp - predicted energy sales price Pe) per unit power generation quantity, a power generation load Le for a loss period te_r, a maintenance interval INT_CI until CI is performed, cost C_CI related to CI, a maintenance interval INT_TI until TI is performed, cost C_TI related to TI, a maintenance interval INT_MI until MI is performed, cost C_MI related to MI, and a time Δt of a loss period te_r (= end time t2 - start time t1 of the loss period te_r). J=J1+J2 [Mathematical Formula 1] J1=∫t1t2(ΔC×Le)dt [Mathematical Formula 2] J2=C_CI×ΔtINT_CI+C_TI×ΔtINT_TI+C_MI×ΔtINT_MI

[0055] Here, CI is a combustion chamber inspection, and performs maintenance or replacement of a part (CI part) around the combustion chamber. TI is a turbine inspection, and performs maintenance or replacement of a part (TI part) around the turbine. MI is an entire power plant inspection, and is called a major inspection, and performs maintenance or replacement of a part (MI part) of the entire power plant.

[0056] Each of the inspection intervals (hereinafter referred to simply as intervals) INT_CI, INT_TI, and INT_MI is an interval until the next inspection is performed after a specific inspection is performed, and is based on an accumulated operating time. In other words, each inspection interval is a usage limit according to an accumulated operating time, and when it is the accumulated operating time, then repair or replacement must be performed according to the inspection. The intervals are short in the order of CI, TI, and MI (INT_CI < INT_TI < INT_MI), and the cost (repair cost or replacement cost of a part) is low in the order of CI, TI, and MI (C_CI < C_TI < C_MI).

[0057] Using the equation (3) described above, an energy sales loss occurring in a diagonal line area defined by a time Δt (in other words, the length of a loss period te_r) from a time t1 to a time t2 and ΔC, which is a differential between energy generation cost Cp and a predicted energy sales price Pe, is calculated.

[0058] Using equation (4) described above, a potential maintenance loss for this time Δt is calculated by respectively multiplying ratios of the time Δt of time t1 to time t2 (in other words, the length of the loss time period te_r) to the check intervals INT_CI, INT_TI, INT_MI by the costs C_CI, C_TI and C_MI required for the check CI, the check TI and the check MI.

[0059] In the following, the no-load mode is described with reference to Fig. 4. In the loss period te_r, when the no-load mode is selected, no power generation occurs (the power generation load Le is zero ("0")), and no power sales are performed. Therefore, no power sales loss JI is incurred.

[0060] Accordingly, the loss index J in the no-load mode is a sum of the maintenance loss J2 and the operating loss J3.

[0061] More specifically, the loss index J in the no-load operation mode is calculated using the following equations (5) to (7) using the check intervals INT_CI, INT_TI and INT_MI, the costs C_CI, C_TI and C_MI related to the check, the time Δt of the loss time period te_r (in other words, the length of the no-load operation time), and a fuel consumption amount F_wait and a fuel unit price Fee_fuel per unit time at the time of performing the no-load operation. J=J2+J3 [Mathematical Formula 3] J2=C_CI×ΔtINT_CI+C_TI×ΔtINT_TI+C_MI×ΔtINT_MI J3=F_wait×Δt×Fee_fuel

[0062] The fuel consumption quantity F_wait is proportional to the turbine rotation speed ωt, and an operating loss J3 calculated using the equation (7) described above is related to the size of the Fig. 4 shown diagonal line area.

[0063] The equation (6) described above corresponds to the equation (4) described above, and thus a description thereof is omitted.

[0064] The following describes the stop mode with reference to Fig. 5. In the loss period te_r, when the stop mode is selected, no power generation is performed (the power generation load Le is zero ("0")), and no power sales are performed. Therefore, no power sales loss J1 is caused.

[0065] Accordingly, the loss index J in the stop mode is a sum of the maintenance loss J2 and the operating loss J3.

[0066] Here, stopping the power plant requires the subsequent starting of the gas turbine, and in the stopping mode, any one of the maintenance loss J2 and the operation loss J3 is a starting-focused loss.

[0067] The following describes a maintenance loss J2 and an operation loss J3 in the stop mode.

[0068] First, the maintenance loss J2 is described. The maintenance loss J2 is correlated with the rate of wear and tear of the components that make up a power plant. The limits related to the wear and tear of the components are a limit based on the accumulated operating time and a limit based on the accumulated frequency of start-ups. While the maintenance loss J2 in the energy sales mode and no-load mode refers to the operating time limit, the maintenance loss J2 in the stop mode refers to the frequency limit of start-ups.

[0069] The operating loss J3 is described below. As can be seen from Fig. 5, after the gas turbine is started, it takes some time for the turbine rotation speed ωt to stabilize, and it also takes some time for the power generation load Le to stabilize. Accordingly, the fuel amount (hereinafter referred to as the start-up time fuel amount) required until the power generation load Le is stabilized from the start-up is the operation loss J3 in the stop mode.

[0070] Described more specifically, the loss index J in the stop mode is calculated using the following equations (8) to (10) using a marginal frequency of starting Strt_CI until CI is performed, cost C_CI related to CI, a marginal frequency of starting Strt_TI until TI is performed, cost C_TI related to TI, a marginal frequency of starting Strt_MI until MI is performed, cost C_MI related to MI, and a starting time fuel quantity F_strt and the fuel unit price Fee_fuel. J=J2+J3 [Mathematical Formula 4] J2=C_CI×1Strt_CI+C_TI×1Strt_TI+C_MI×1Strt_MI J3=F_strt×Fee_Fuel [1-2. Flowchart]

[0071] A method for planning an operation plan of a power plant according to this embodiment (hereinafter simply referred to as a planning method) will be described with reference to Fig. 6 described.

[0072] Fig. 6 is a schematic flowchart illustrating an operation plan scheduling method of a power plant according to this embodiment.

[0073] In step S1 (determination process), the process proceeds to step S6, and the power sales mode is selected based on a comparison of a predicted power sales price Pe with power generation cost Cp for a 24-hour profit period in which the predicted power sales price Pe is higher than the power generation cost Cp. On the other hand, for a loss period Te_r in which the predicted power sales price Pe is equal to or less than the power generation cost Cp, the process proceeds to step S2. In step S2, a loss index J of a case where each of the power sales mode, the no-load mode, and the stop mode is selected as a mode is calculated, and a mode whose loss index J is the smallest is selected as a mode of this loss period.In other words, in the case where the operation mode whose loss index J is the smallest is the energy sales mode as the operation mode of this loss period, the process proceeds to step S3, and the energy sales mode is selected. On the other hand, in the case where the operation mode whose loss index J is the smallest is the stop mode, the process proceeds to step S4, and the stop mode is selected. Furthermore, in the case where the operation mode whose loss index J is the smallest is the no-load mode, the process proceeds to step S5, and the no-load mode is selected. In other words, a scheduling process according to this embodiment is configured by steps S2 to S6. [1-3. Effect]

[0074] According to a scheduling system (scheduling method) for the first embodiment, for the profit period in which the predicted energy sales price Pe is higher than the power generation cost Cp, the energy sales mode is selected. On the other hand, for the loss period in which the predicted energy sales price Pe is equal to or less than the power generation cost Cp, a mode whose loss index is the smallest is selected from the energy sales mode, the no-load mode, and the stop mode.

[0075] Since the loss index takes into account maintenance loss and operating loss that do not contribute to energy sales, in addition to the energy sales loss, using a potential loss based on durability consumption, and using the operating loss, a loss based on operation at the time of restarting or at the time of performing no-load operation that does not contribute to energy sales is taken into account, an operation mode in the loss period is selected. Accordingly, a high-yield operation plan can be planned. [2. Embodiment][2-1. Configuration]

[0076] An operation plan planning system (planning method) for a power plant according to a second embodiment of the present invention will be described with reference to Fig. 7A and Fig. 7B.

[0077] Fig. 7A and Fig. 7B are maps used for determining a weighting coefficient according to a second embodiment of the present invention.

[0078] In the operation plan planning system (planning method) for a power plant according to the second embodiment of the present invention, functions are added to the planning unit (planning method) of the operation plan planning system (planning method) for a power plant according to the first embodiment. This point will be described below.

[0079] As described above in the first embodiment, the limits related to the consumption of the durability of a component are a limit according to the accumulated operating time and a limit according to the accumulated frequency of startup. From these limits, if any of them reach a limit, a check is performed, and repair or replacement must be performed.

[0080] For this reason, when the accumulated operating time reaches a limit (hereinafter also referred to as operating time durability), a corresponding component must be repaired or replaced, even if there is a gap in the accumulated starting frequency with respect to the limit (hereinafter also referred to as operating frequency durability). Similarly, when the accumulated starting frequency reaches a limit, a corresponding component must be repaired or replaced, even if there is a gap in the accumulated operating time with respect to the limit.

[0081] Accordingly, an operation is one in which the accumulated operating time and the accumulated starting frequency simultaneously reach the limits. Therefore, in this embodiment, a component is controlled to be used (consumed) such that both the accumulated operating time and the accumulated starting frequency are close to the limits.

[0082] In the following, the accumulated operating time or the accumulated frequency of starting close to the limit is shown as durability consuming.

[0083] Here, the durability consumption refers to the maintenance loss J2 among the elements of the loss index J. In each of the power sales mode and the no-load mode, the power plant does not support startup, and accordingly, in the maintenance loss J2, a loss (running time loss) according to the maintenance consumption related to the accumulated operating time is a major factor. In contrast, since the maintenance loss J2 of the stop mode supports restarting, the loss (the frequency of start-up loss) according to the durability consumption related to the accumulated frequency of startup is a major factor.

[0084] Thus, in this embodiment, for the energy sales mode and the no-load mode, the maintenance loss J2 is calculated using the following equation (11). In other words, the equation for calculating J2 according to the first embodiment is multiplied by a weighting coefficient (first weighting coefficient) k1. [Mathematical Formula 5] J2=(C_CI×ΔtINT_CI+C_TI×ΔtINT_TI+C_MI×ΔtINT_MI)×k1

[0085] Furthermore, for the stop mode, the maintenance loss J2 is calculated using the following equation (12). In other words, the equation for calculating the maintenance loss J2 according to the first embodiment is multiplied by a weighting coefficient (second weighting coefficient) k2. [Mathematical Formula 6] J2=(C_CI×1Strt_CI+C_TI×1Strt_TI+C_MI×1Strt_MI)×k2

[0086] The weighting coefficients k1 and k2 are calculated according to the Fig. 7A and Fig. 7B stored in the storage unit 29 in advance, using a ratio (the accumulated start time / the running time durability; hereinafter referred to as a running time durability consumption ratio) C1 of the accumulated start time to the running time durability and a ratio (the accumulated number of times of start / the number of times of start durability; hereinafter referred to as a number of times of start durability consumption ratio) C2 of the accumulated number of times of start to the number of times of start durability.

[0087] In other words, as in Fig. 7A, the weighting coefficient k1 is set to be smaller as the frequency of a starting durability consumption ratio C2 increases and as the operating life durability consumption ratio C1 decreases. In addition, as shown in Fig. 7B, the weighting coefficient k2 is set smaller as the running life durability consumption ratio C1 increases and as the frequency of a starting durability consumption ratio C2 decreases.

[0088] In addition, the weighting coefficients k1 and k2 can be calculated using functions of the durability consumption ratios C1 and C2 by switching between the maps or maps.

[0089] In addition, the weighting coefficients k1 and k2 can be determined using maps or functions according to the relationship between the running durability consumption ratio C1 and the frequency of a starting durability consumption ratio C2. [2-2. Effect]

[0090] According to the planning system (planning method) for the second embodiment, the following effects are present in addition to the effects of the first embodiment.

[0091] By setting the weighting coefficient k1 to be smaller as the frequency of a starting durability consumption ratio C2 increases and as the running-time durability consumption ratio C1 decreases, the maintenance loss J2 and, consequently, the loss index J of the energy sales mode and the no-load mode calculated using the weighting coefficient k1 are calculated to be smaller. Accordingly, compared to the case where no weighting coefficient is used, the energy sales mode or the no-load mode that consumes the running-time durability can be selected relatively easily.

[0092] In contrast, by setting the weighting coefficient k2 to be smaller as the running-time durability consumption ratio C1 increases and as the frequency of a starting durability consumption ratio C2 decreases, the durability loss J2 and, consequently, the loss index J of the stop mode calculated using the weighting coefficient k2 are calculated to be smaller. Accordingly, the stop mode that consumes the running-time durability can be selected relatively easily compared to the case where no weighting coefficient is used.

[0093] In this way, the energy sales mode or the no-load mode that consumes the operating time durability can be easily selected as the operating mode when the operating time durability consumption ratio C1 decreases, and the stop mode that consumes the frequency of a start durability can be easily selected as the operating mode when the frequency of a start durability consumption ratio C2 decreases. Accordingly, consumption is prevented from tending toward either the frequency of a start durability or the start-time durability, and thus, repair costs or replacement costs associated with component consumption can be avoided. [2-3. Others]

[0094] In the above-described embodiment, only the weighting coefficient k1 may be set smaller, or only the weighting coefficient k2 may be set larger as the frequency of a starting durability consumption ratio C2 increases, while the weighting coefficient k1 is set smaller and the weighting coefficient k2 is set larger as the frequency of a starting durability consumption ratio C2 increases.

[0095] Similarly, in the above-described embodiment, only the weighting coefficient k2 may be set smaller, or only the weighting coefficient k1 may be set larger as the running time durability consumption ratio C1 increases, while the weighting coefficient k2 is set smaller and the weighting coefficient k1 is set larger as the running time durability consumption ratio C1 increases. [3. Embodiment][3-1. Configuration]

[0096] An operation plan planning system (planning method) for a power plant according to a third embodiment of the present invention will be described with reference to Fig. 8 described.

[0097] Fig. 8 is a schematic diagram illustrating an additional function of a scheduling system (scheduling method) according to the third embodiment of the present invention, and is a diagram in which examples of predicted energy sales prices Pe1, Pe2, and Pe3 and energy generation cost Cp are plotted on a graph with the energy unit price as the vertical axis and time as the horizontal axis, and operation plans (an operation mode selected for each time point) are shown superimposed thereon.

[0098] In the operation plan planning system (planning method) for a power plant according to the third embodiment of the present invention, a function is added to a predicted power sales price setting unit for the first embodiment, and a function is added to a planning unit. These points will be described below.

[0099] The predicted energy selling price setting unit predicts three predicted energy selling price lines L_Pe1, L_Pe2, and L_Pe3 for a predicted target date (here, the next day of bidding) based on past results. These predicted energy selling price lines L_Pe1, L_Pe2, and L_Pe3 represent 24-hour transitions of predicted energy selling prices Pe1, Pe2, and Pe3, respectively.

[0100] The predicted energy sales price Pe2 is a price acquired by correcting an average of past results using weather, temperature, and the like. The predicted energy sales prices Pe1 and Pe3 are prices that predict fluctuations from the average of past results. The predicted energy sales price Pe1 is a highest predicted energy sales price acquired by adding a maximum fluctuation of the high price side to the average, and the predicted energy sales price Pe3 is a lowest predicted energy sales price acquired by subtracting a maximum fluctuation of the low price side from the average.Thus, the predicted energy sales price Pe1 is also called a highest predicted energy sales price Pe1, the predicted energy sales price Pe2 is also called a middle energy sales price Pe2, and the predicted energy sales price Pe3 is also called a lowest predicted energy sales price Pe3.

[0101] Similar to the first embodiment, the predicted energy sales price setting unit is not indispensable, but predicted energy sales price information may be acquired from the outside, and the acquired information may be stored in a storage unit.

[0102] The planning unit plans three operating plans using the predicted energy sales prices Pe1, Pe2 and Pe3.In other words, the planning unit plans a preliminary operation plan 100A based on a result of a comparison between the price cost Cp and the highest predicted energy sales price line L_Pe1 (highest predicted energy sales price Pe1) performed by the determination unit, the planning unit plans a preliminary operation plan 100B based on a result of a comparison between the price cost Cp and the average predicted energy sales price line L_Pe2 (average predicted energy sales price Pe2) performed by the determination unit, and the planning unit plans a preliminary operation plan 100C based on a result of a comparison between the price cost Cp and the lowest predicted energy sales price line L_Pe3 (lowest predicted energy sales price Pe3) performed by the determination unit 23.

[0103] Each of the preliminary operation plans 100A, 100B, and 100C is planned using a method corresponding to the method for generating the one-day operation plan based on a comparison between the price cost Cp and the predicted energy sales price line L_Pe (predicted energy sales price Pe) in the first embodiment.

[0104] Then, the planning unit determines a final operation plan (hereinafter also referred to as a main operation plan) 100D by forming such preliminary operation plans 100A, 100B, and 100C. In this formation, for each hour, an operation mode most frequently included in the operation modes configuring the preliminary operation plans 100A, 100B, and 100C is used as an operation mode at that time, and in the case where all the operation modes of the preliminary operation plans 100A, 100B, and 100C are different from each other, an operation mode of the preliminary operation plan 100B based on the average predicted energy sales price Pe2 is used as the operation mode at that time.

[0105] In the Fig.In the example shown in Figure 8, between midnight and 6 a.m., any one of the provisional operation plans 100A, 100B, and 100C is the stop mode, and accordingly, the stop mode is also selected in the main operation plan 100D. Between 6 a.m. and 7 a.m., the energy sales mode is used in the provisional operation plan 100A, and the stop mode is used in the provisional operation plans 100B and 100C, and thus, the most frequently used stop mode is selected in the main operation plan 100D. Between 7 a.m. and 10 a.m., the energy sales mode is used in the provisional operation plans 100A and 100B, and the stop mode is used in the provisional operation plan 100C, and thus, the most frequently used energy sales mode is selected in the main operation plan 100D.Similarly, at each hour from 10 a.m. to 5 p.m. and 6 p.m. to midnight, a most frequently used operating mode in the preliminary operating plans 100A, 100B and 100C is selected in the main operating plan 100D.

[0106] Between 5 p.m. and 6 p.m., the energy sales mode is selected in the preliminary operation plan 100A, the no-load mode is selected in the preliminary operation plan 100B, and the stop mode is selected in the preliminary operation plan 100C. Thus, the operation modes used in the preliminary operation plans 100A, 100B, and 100C are different from each other, and accordingly, the no-load mode used in the preliminary plan 100B is selected in the main operation plan 100D. [3-2. Effect]

[0107] According to the planning system (planning method) of the third embodiment, preliminary operation plans are planned based on three predicted energy sales prices, including the highest predicted energy sales price Pe1, the average predicted energy sales price Pe2, and the lowest predicted energy sales price Pe3, and a final operation plan is set by forming these preliminary operation plans. Accordingly, a responsive operation plan can be used even when an actual energy sales price deviates greatly from an average predicted value. [3-3. Others]

[0108] (1) In the above-described embodiments, an operation mode to be selected can be selected by a user using an input unit (selected operation mode acquisition unit) 3 in the case where all the operation modes of the preliminary operation plans 100A, 100B, and 100C are different from each other, while the operation mode of the preliminary operation plan 100B based on the average predicted energy sales price Pe2 is used as the operation mode at that time in the case where all the operation modes of the preliminary operation plans 100A, 100B, and 100C are different from each other. In such a case, an operation mode used by the preliminary operation plan 100B can be displayed on a display unit (recommended operation mode display unit) 4 as a recommended operation mode.

[0109] In such a case, the operation mode can be selected based on the user's knowledge in the case where all the operation modes of the preliminary operation plans 100A, 100B and 100C are different from each other.

[0110] (2) When the preliminary operation plans 100A, 100B and 100C are planned as in the second embodiment, the maintenance loss J2 can be calculated using the weighting coefficients k1 and k2. [4. Others]

[0111] (1) In each embodiment described above, while the power plant has been described as a gas turbine combined cycle power plant (GTCC), the power plant operation plan planning system (planning method) according to the present invention is not limited to the GTCC. For example, the planning system (planning method) can also be applied to a conventional thermal power plant configured by a steam boiler and a steam turbine.

[0112] (2) Even in the case where bidding is conducted for a period in which the energy sales mode is used based on an operation plan planned by the scheduling system (scheduling method), there are cases where a successful bidding is not carried out. In such cases, for a period in which a successful bidding is not carried out, the loss indexes J are calculated respectively for the case in which the no-load mode is selected and for the case in which the stop mode is selected, and an operation mode whose loss index J is smaller can be used by switching from the energy sales mode.

[0113] (3) When an operation mode is selected based on a comparison of loss indices J, in the case where a difference ΔJ between a smallest loss index and a second smallest loss index is within a reference value ΔJ0, then one of the operation mode whose loss index is the smallest and the operation mode whose loss index is the second smallest can be selected by a user. The above-described reference value ΔJ0 can be set, for example, based on a predicted accuracy of the predicted energy sales price Pe. List of reference symbols 1 Operational plan planning system for power plant (planning system) 2 Planning device 3 Input unit (selected mode acquisition unit) 4 Display unit (recommended operating mode display unit) 21 Predicted energy sales price setting unit 22 Determination unit 23 Planning Unit 29 Storage unit 100, 100D operating plan 100A, 100B, 100C preliminary operating plan C1 Frequency of a starting durability consumption ratio C2 Operating life-to-consumption ratio CP Energy generation costs C_CI Costs related to CI C_MI Costs related to MI C_TI Costs related to TI F_wait fuel consumption value Fee_fuel Fuel unit price F_strt Fuel size at the time of starting J Loss Index J1 Energy sales loss J2 Maintenance loss J3 Operating loss k1 weighting coefficient (first weighting coefficient) k2 weighting coefficient (second weighting coefficient) L_Pe, L_Pe1, L_Pe2, L_Pe3 predicted energy sales price line The energy generation load Pe1, Pe2, Pe3 predicted energy sales price Strt_CI Limit frequency of starting CI Strt_MI Limit frequency of starting MI Strt_TI Limit frequency of starting TI te_p yield period te_r loss time duration ΔC loss per unit energy generation size INT_CI Interval until CI is performed INT_MI Interval until MI is performed INT_TI Interval until TI is performed Le energy generation load in loss time duration te_r Δt time of loss time duration te_r ΔJ Difference between smallest loss index and second smallest loss index ΔJ0 Reference value of ΔJ ωt turbine rotation speed

Claims

[1] A planning system (1) for planning an operation schedule (100, 100D) of a power plant over a predetermined period of time, the power plant being capable of generating power by driving a power generator using a turbine, and executing any one of an energy sales mode that sells energy by operating the power plant in a state in which the turbine and the power generator are connected, a no-load mode that operates the power plant in a state in which the turbine and the power generator are not connected, and a stop mode that stops the turbine, the planning system (1) comprising: a determination unit (22) configured to determine whether a predicted energy sales price (Pe1, Pe2, Pe3) is higher than energy generation costs (CP) or not; and a planning unit (23) configured to calculate a loss index (J) of each of the energy sales mode, the no-load mode, and the stop mode in the case where the predicted energy sales price (Pe1, Pe2, Pe3) is equal to or less than the energy generation cost (CP) within the predetermined period of time, using equation (1) based on an energy sales loss (J1), which is the difference between the expected energy sales price and the energy generation cost (CP), a maintenance loss (J2), and an operation loss (J3) that does not contribute to energy sales, and to plan the operation plan (100, 100D) such that the power plant is operated in an operation mode in which the loss index (J) is the smallest, wherein the loss index (J) corresponding to the energy sales mode includes the energy sales loss (J1) and the maintenance loss (J2), wherein the loss index (J) corresponding to the no-load mode and the stop mode includes the operating loss (J3) and the maintenance loss (J2), respectively, wherein the maintenance loss (J2) comprises an operating time loss consuming a durability related to an operating time of a component configuring the power plant according to an accumulated operating time of the power plant and a frequency of a start-up loss consuming a frequency related to the frequency of starting the component according to an accumulated frequency of starting the power plant, wherein the maintenance loss (J2) of each of the energy sales mode and the no-load mode has the operating time loss as a main factor, wherein the maintenance loss (J2) of the stop mode has the frequency of a start loss as a main factor, and wherein the planning unit (23) is configured to: determine the maintenance loss (J2) of each of the energy sales mode and the no-load mode using a first weighting coefficient (k1); to determine the maintenance loss (J2) of the stop mode using a second weighting coefficient (k2); to decrease the first weighting coefficient (k1), to increase the second weighting coefficient (k2), or to decrease the first weighting coefficient (k1) and to increase the second weighting coefficient (k2) as a ratio of the accumulated frequency of starting to the frequency of starting increases; and to increase the first weighting coefficient (k1), to decrease the second weighting coefficient (k2), or to increase the first weighting coefficient (k1) and decrease the second weighting coefficient (k2) when a ratio of the accumulated operating time to the durability relative to the operating time increases, wherein Loss index = energy sales loss + maintenance loss + operating loss that does not contribute to energy sales [2] Planning system (1) according to claim 1, where three prices of a lowest predicted energy sales price, a middle predicted energy sales price and a highest predicted energy sales price are used as the predicted energy sales price (Pe1, Pe2, Pe3), wherein the determination unit (22) compares each of the lowest predicted energy sales price, the middle predicted energy sales price and the highest predicted energy sales price with the energy generation cost (CP) and determines whether the predicted energy sales price (Pe1, Pe2, Pe3) is higher than the energy generation cost (CP) or not, and wherein the planning unit (23) plans a preliminary operation plan (100A, 100B, 100C) for the predetermined period of time by performing the selection control for each of the lowest predicted energy sales price, the average predicted energy sales price, and the highest predicted energy sales price, and plans the operation plan (100, 100D) by forming a first preliminary operation plan based on the lowest predicted energy sales price, a second preliminary operation plan based on the average predicted energy sales price, and a third preliminary operation plan based on the highest predicted energy sales price. [3] The planning system (1) according to claim 2, wherein the planning unit (23) performs the forming by using the operation mode selected most frequently among the operation modes selected at the same time for each of the first preliminary operation plan, the second preliminary operation plan and the third preliminary operation plan. [4] The planning system (1) according to claim 3, wherein the planning unit (23) performs the forming by using the operation mode selected in the second preliminary operation plan in the case where the operation mode selected at the same time is different in each of the first preliminary operation plan, the second preliminary operation plan and the third preliminary operation plan. [5] The planning system (1) according to claim 3, further comprising a selection mode acquiring unit (3), wherein the planning unit (23) performs the forming by using the operation mode acquired by the selection mode acquiring unit (3) in the case where the operation mode selected at the same time is different in each of the first preliminary operation plan, the second preliminary operation plan and the third preliminary operation plan. [6] The planning system (1) according to claim 5, further comprising a recommended operation mode display unit (4), wherein the planning unit (23) causes the recommended operation mode display unit (4) to display the operation mode of the second preliminary operation plan as a recommended operation mode in the case where the operation mode selected at the same time is different in each of the first preliminary operation plan, the second preliminary operation plan and the third preliminary operation plan. [7] The scheduling system (1) according to any one of claims 1 to 6, wherein the operating loss (J3) includes fuel costs that do not contribute to power generation consumed at the time of starting the power plant and at the time of no-load operation.

Images