JP2564271B2 - Thermal power plant on-time shortest start method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a start-up system for a thermal power plant, and particularly to an operation start-up system for a specified start-up completion time, in compliance with operation constraint conditions in the plant start-up process. The present invention relates to a start-up method suitable for enabling start-up and completing start-up in a short time.

[Conventional technology]

The conventional method for starting a thermal power plant is, depending on the stop time before starting and the temperature condition of the equipment, the amount of fuel initially injected into the boiler, the time function of temperature rise and pressure rise of the main steam, the speed up and load of the turbine, and the load. In this method, the rising time function is determined as a start-up schedule and this start-up schedule is executed by a control system provided in each system of the plant. The most representative method is the Electrical World (19
February 66, pages 58 to 62, Electrical World, Vol.165,
No.6 paper Thermal Stress Influence Starting, Loadin
gof Boilers, Turbines).

This method is a method for uniquely determining the start-up schedule according to the initial state of a limited part of the plant.
That is, according to the initial values of the boiler steam pressure, the boiler outlet steam temperature, and the steam turbine casing temperature, the steam turbine acceleration rate, initial load amount, speed retention, and steam turbine warm-up time and load change rate due to load retention are determined. Is the way to do it. Although this method can keep the start-up completion time of the plant accurately, it absorbs the variation in the temperature rise characteristics of the boiler-generated steam as the margin of the start-up schedule, so the start-up schedule created is longer than necessary. It tends to be long. As another conventional method, USP3,44
6,224 and USP 4,228,359 are known. These are intended for rapid start of the steam turbine while monitoring the thermal stress generated in the steam turbine in real time online, but there is no mention of a specific method for keeping the specified start time and a method for starting the boiler. Absent.

Japanese Patent Laid-Open No. 59-157402 is known as a conventional method for reducing the startup time of a boiler. This method aims to rapidly raise the temperature of the steam generated by the boiler while monitoring the thermal stress generated in the boiler in real time online. However, this method also makes no mention of a specific method of keeping the specified start time and start of the turbine.

[Problems to be solved by the invention]

As described above, each of the conventional methods is a rapid start method focusing on only one of the boiler and the steam turbine. However, even if such individual methods are combined, there is no guarantee that the start completion time of the entire plant is accurate and the start time is the shortest. This is because the boiler and the steam turbine have very strong mutual interference, and the individual optimization does not always result in the overall optimization.

The present invention, while the above-mentioned conventional method was intended for one of the boiler or the steam turbine, in consideration of the start characteristics of both, the accuracy of the start completion time of the entire plant was improved and the start time was minimized. The purpose is to

[Means for solving problems]

The above-mentioned object focuses on the start-up characteristics of the entire plant system that has not been considered in the conventional method, and uses a model for predicting the start-up characteristics, and creates a start-up schedule that determines the shortest start-up schedule by a kind of hill climbing method. This is achieved by providing a function and adjusting a start start time such that the expected start completion time matches the designated time. Specifically, a start schedule creation function for creating a time-based start schedule related to the operation required to start the thermal power plant and the control target value setting before starting the plant, and actually according to the created start schedule. In a start-up system of the plant having a schedule execution function for starting a plant, the start-up schedule optimization function searches the start-up schedule that minimizes the start-up time under the condition that the start-up completion time is kept. Unit and a plant dynamic characteristic prediction unit, the plant dynamic characteristic prediction unit includes a plant dynamic characteristic model for predicting a start characteristic when the plant is assumed to be activated according to a start schedule, and the plant dynamic characteristic. The starting characteristics obtained by the model and measured from the actual plant before starting A stress calculation unit that calculates the stress of the plant based on the state value and a built-in stress calculation unit, and whether the stress value calculated by the plant dynamic characteristic prediction unit violates the operational constraint conditions in the schedule optimization unit. A constraint condition infringement determining unit that determines whether or not the start value is corrected to search for an optimum startup schedule in which the stress value does not violate the operation constraint conditions and has the shortest startup time, and the plant dynamic characteristic model is obtained. And a schedule setting function, wherein the schedule execution function is configured so that the error between the start completion time in the start schedule optimized by the schedule optimization unit and the start completion time specified in advance becomes less than or equal to an allowable value. There is a means to put the plant on standby.

[Action]

The plant dynamic characteristic model incorporated in the startup system of the thermal power plant is used for predicting the startup characteristics before the plant is actually started. The start-up characteristics under the operation constraint conditions initially set in this plant dynamics model are predicted, and the start-up schedule that minimizes the start-up time is obtained. When the difference between the expected start completion time and the designated time at this time is equal to or larger than the allowable value, the plant is put on standby, and when the difference is equal to or smaller than the customer value, the start is started. With such a procedure, the startup time can be minimized and the startup can be completed at the designated time.

Example of Invention

FIG. 1 shows the basic functional configuration of the thermal power plant shortest start method. The functions are roughly divided into a startup schedule creation function 1000 and a schedule execution function 2000. The former creates the optimal start-up schedule 101 that minimizes the plant start-up time, and the latter creates the actual plant 300.
In order to start 0 according to the optimum start schedule, the operation amount 201 for the plant is changed every moment.

The startup schedule creation function 1000 further includes a schedule optimization function 1100 and a plant dynamic characteristic prediction function 1200.
The former is further composed of an offline optimization function 1110 and an online optimization function 1120, and the latter is further a plant dynamics model.
1210, a boiler stress calculation function 1220 and a turbine stress calculation function 1230. The off-line optimization function 1110 assumes the startup schedule 111, and uses this as the plant dynamics model 121.
By setting to 0, the starting characteristic (213, 211, 212) is simulated, and the boiler stress calculation function 1220 and the turbine stress calculation function 1230 calculate the boiler stress 221 and the turbine stress 231 respectively. On the other hand online optimization function 1120
Sequentially optimizes the target operating state 121 of the turbine according to the turbine stress 231 when the plant dynamic model 1210 operates. The off-line optimization function 1110 evaluates the behavior of the process variables related to the operation restriction conditions calculated as a result and the start-up required time, newly assumes the start-up schedule 111, and sets this in the plant dynamic characteristic model 1210. By repeating such processing, the optimum start schedule 101 that can complete the start in the shortest time without the process state violating the operation limit is determined. The determined optimum start-up schedule 101 is set in the schedule effective function 2000 and becomes a target value for actually starting the plant. Here, plant dynamic characteristic model 1210, boiler stress calculation function
1220 and turbine stress calculation function 1230 have initial values
321, 322 and 323 are required, and these are process state values measured before starting.

Next, parameters that define the startup schedule in the embodiment will be defined. Since the start-up time of the plant basically depends on the temperature characteristic of the plant, one having a strong causal relationship with the temperature rise time of the plant should be selected as a parameter. Based on this basic concept, igniter ignition interval (T _IG), mill initiation interval (T _PLV), the four main steam NoboriAtsushiritsu (L _TMS) and reheat steam NoboriAtsushiritsu (T _RHH) Selected as a parameter.

The igniter ignition interval (T _IG ) is the interval at which when a boiler ignition command is generated, the igniters corresponding to each burner stage of the boiler are sequentially ignited at time intervals T _IG and the light oil burner is ignited.

The mill starting interval (T _PLV ) is a time interval when the pulverized coal mill is sequentially started after all the light oil burners are ignited. In this case, the mills up to the second unit supply 50% of the rated flow rate of pulverized coal according to the operation standard. When the 3rd mill is started, it operates so that the total flow rate of pulverized coal fed to the boiler is 40% of the rated value (the share of each mill is 67%). After that, the output of the generator that starts the turbine is
When it reaches 40%, it shifts to the normal load operation mode, the mill is started according to the load demand, and a maximum of 5 mills are operated. As each mill is started, the light oil burner is digested.

Main steam NoboriAtsushiritsu (L _TMS) is the normal load operating range (40 to 1
This is a parameter that regulates the rate of rise of the main steam temperature at (00% load), and has the function shown in the following formula for the main steam temperature target value T _MSSET calculated by the schedule execution function 2000.

Here, T _MS40 : Main steam temperature when reaching 40% load (℃) T _MSR : Main steam temperature rated value (℃) L: Load (%) That is, when the load reaches L _TMS (%) It means that the steam temperature reaches the rated value.

The reheat steam temperature rise rate (L _TRH ) is a parameter that regulates the rate of rise of the reheat steam temperature in the normal load operation region, similar to the main steam temperature, and is calculated by the schedule execution function 2000. It has a function shown in the following formula with respect to the temperature target value T _RHSET .

Here, T _RH40 : _Reheated steam temperature when reaching 40% load (℃) T _RHR : _Reheated steam temperature rated value (℃) L: Load (%) That is, load reaches L _TRH (%) This means that the reheat steam temperature reaches the rated value.

Next, the optimization algorithm in the offline optimization function 1110 will be described in detail.

FIG. 2 shows the basic processing procedure of the startup schedule optimization algorithm to which the complex method, which is one of the nonlinear optimization methods, is applied to the present invention. Where the schedule parameter is It is written as. Hereinafter, each processing function will be described.

(1) Initialization 100 Fig. 3 shows the symbols, values, units and meanings of constants and initial values used in the optimization algorithm.

(2) The initial simplex formation 200 processing procedure is shown in FIG. Initial trial point Is the set value Set and execute simulation. In this case, the simulation means to activate the plant dynamic characteristic prediction function 1200 and start Is to predict the starting characteristics when the plant is started according to. At this time, The driving restriction factor Y (N _MV ) for the implicit constraint condition Y
_{If L} (N _MV ) (see Fig. 5) is not violated, Form an initial simplex near.

Here, B _J is a pseudo-random number that satisfies −1.B _J 1. It is determined by the procedure shown in FIG.

In case of violating the implicit constraint condition, the trial point is corrected by the procedure shown in FIG.

(A) Random number generation 220 FIG. 6 shows a procedure for calculating a pseudo random number in which M is a variable.
This algorithm is a method that uses the number appearing fifth from the most significant digit of the square root.

(B) Trial points according to the initial simplex correction 240 extension magnification correction coefficient D (I) Is corrected as follows.

(C) Extension magnification correction coefficient determination 260 Sensitivity to operation limiting factors when operating parameters are changed differs from large, medium, small, and zero, as shown in FIG. Therefore, depending on which driving restriction factor violates the implicit constraint condition (see FIG. 5), the optimum value search efficiency is better when the trial point extension magnification is modified than when it is uniformly modified. It is expected to increase. Figure 9 shows the implicit constraint monitoring algorithm based on this concept (see Figure 10).
An algorithm for determining the extension magnification correction coefficient based on

(3) Characteristic evaluation ranking 300 As shown in FIG. 11, this function has K (in this example, K =
It is an algorithm for determining the following three characteristic points from the vertices of the simplex consisting of 8).

Operation parameter corresponding to the shortest activation vertex of K vertices And startup time (T _{X, Q} ). Operation parameter corresponding to the longest activation vertex of K vertices And start-up time (T _{X, S} ). Operation parameter corresponding to the vertex with the second longest startup time among the K vertices And start-up time (T _{X, S2} ) (4) Centroid calculation 400 As shown in Fig. 12, the worst vertex Geometric center of gravity of simplex consisting of (K-1) vertices excluding Ask for.

(5) New trial point determination 500 As shown in FIG. Is defined by coordinates that satisfy the following equation. This point is the worst vertex It is on a straight line that connects the It is in. Here, R is the extension magnification described in (7).

(6) Trial point retreat non-determination 600 When the extension magnification is corrected, the next trial point retreats toward the center of gravity, but instead of retreating indefinitely, as shown in Fig. 14, R-0.1R ₀ When this happens, the retreat is stopped, the entire simplex is degenerated, and a new search direction is found. This degeneration method will be described in (9).

(7) Extension rate modification 700 If the implicit constraint condition is violated, the extension rate R will be modified by the method shown in FIG.

(8) New trial point extension 800 New trial point The starting time T _{X, K + 1 at} is the shortest time T _{X, Q}
If it is shorter than the above, as shown in FIG. 16, further extend in the same direction to approach the optimum point. This re-extension point And

(9) Trial point retreat 900 New trial point If the startup time T _{X, K + 1 at} is longer than the second long time T _{X, S2 at that} time _, X _{K + 1} may have jumped over the optimum point. Therefore, as shown in FIG. 17, T _{X, K + 1}
In the case of T _{X, S} , the trial point moves backward toward the center of gravity (R
Is set to 0.5R), and when T _{X, K + 1} > T _{X, S} , it is largely retracted (set to R = −0.5). The trial point at this time And

(10) Degeneracy of simplex 1300 If no characteristic improvement point is found on the straight line connecting (T _{X, C}
<T _{X, S} Is the best point for the size of simplex By reducing in the direction of, a new possibility of approaching the optimum point is found. In this case, as shown in FIG. 18, the degeneracy rate is initially set to 1/2, but when each vertex violates the constraint condition, the degeneracy rate is set to 3/4. Nevertheless, the vertex that violates the constraint is returned to its original position. Here, the explicit constraints are the upper and lower limits of the optimization parameter itself, and Is. As shown in Fig. 19, the simulation is executed after confirming that all the parameters satisfy the explicit constraints.

(11) Worst point exclusion 1420,1440,1460 As shown in Fig. 20 Exclude To form a new simplex.

(12) Search count limit arrival judgment 1500 The search count is the number of simulations, and by limiting this, this algorithm will not fall into an infinite loop. FIG. 21 shows a processing procedure therefor, and the meanings of the signals are as follows.

N _T : Total number of simulations N _AD : Number of times simulation results were used as apex of simplex N _NG : Number of times simulation results were not used as apex of simplex N _SAD : The number of times the simulation result for degeneracy of the simplex was used as the apex of the simplex. N _SNG : Simulation for degeneration of simplex The number of times the result was not used as the peak of simplex. (13) Simulation 1600 The basic procedure of simulation is shown in FIG. At Simulation, the plant startup process was divided into three phases: boiler startup, speed increase, and load increase. The boiler start-up phase executes boost control from igniter ignition (this function is built into the plant dynamics model), and the start-up set pressure (main steam 94.9ata, reheat steam pressure 8.16ata)
The starting process until reaching is shown. Acceleration phase indicates the starting process until the metal match condition of the high-pressure turbine is satisfied by increasing the speed to the rated speed by the metal match control function including the speed increase control function. The load increase phase indicates the starting process until the load is included by the combination condition determination function and the rated load (the target load in actual operation) is reached by the load increase control function.

(14) Optimal point convergence judgment 1700 Optimal point, that is, the shortest start schedule satisfies the following equation. And

At this time It is written as.

This completes the detailed description of the offline optimization function 1110. Next, the online optimization function 1120 will be described in detail.

In order to realize quick start by online optimization,
We paid attention to the following points.

(1) Rapid pressure increase considering the rate of change of drum steam temperature An increase in the main steam pressure means an increase in the drum pressure, and an increase in the drum pressure is expressed as the above-mentioned saturation temperature determined by the pressure. Further, when the drum steam temperature changes, thermal stress is generated in the drum. In order to reduce the thermal stress of the drum to the allowable value or less, the change of the steam temperature and the maximum speed increase rate are sequentially determined to complete the speed increase in the shortest time.

(4) Early merger due to establishment of possible merger conditions Immediately after merger, the steam temperature generated by the boiler rises sharply. If this phenomenon is not taken into consideration and only the metal matching conditions of the high-pressure turbine are established, excessive thermal stress will be generated in the rotor despite the load being retained. As a result, the waiting time for joining is minimized, and the startup time is shortened.

(5) Rapid load increase considering the stress of high-pressure and medium-pressure turbines The stress (thermal stress + centrifugal stress) generated on the rotor surface and bore of high-pressure and medium-pressure turbines is kept below the allowable value, and the maximum load increase rate is By sequentially determining, the load increase is completed in the shortest time.

The processing method of the online optimization function 1120 created based on the basic idea described above will be described in detail.

(1) Step-up control It is necessary to keep the internal fluidization rate below the allowable value for the boiler drum when the plant is started. In the present invention, in consideration of the fact that the relationship between the pressure and the saturation temperature has non-linearity, the target pressure is determined so that the maximum allowable temperature change rate is always obtained. This minimizes the time required for boosting.

(2) Acceleration / ventilation by calculating optimal metal matching conditions The plant to be controlled was started at medium pressure (method to accelerate by a medium-pressure turbine). It is necessary to consider both the high-pressure turbine and the medium-pressure turbine as the metal match condition.
In the present invention, when the reheat steam temperature reaches the ventable temperature determined from the metal temperature of the intermediate pressure turbine, the intermediate pressure turbine is immediately ventilated and the speed is increased. When the speed-up is completed and the main steam temperature reaches the ventilation temperature that is determined by the metal temperature of the high-pressure turbine, the control is immediately advanced to the load increasing phase. As a result, the ventilation waiting time of the turbine is minimized.

(3) Rapid acceleration in consideration of the stress of the medium-pressure turbine The stress (thermal stress + centrifugal stress) generated on the rotor surface and bore of the medium-pressure turbine is kept below the allowable value, and the thermal stress changes as the body temperature changes. appear. At this time, in order to prevent the occurrence of excessive thermal stress, the rate of change of the internal fluid temperature must be kept below the allowable value. Since the internal fluid temperature is regarded as a saturation temperature that is uniquely determined by the pressure at that time, the allowable temperature change rate can be expressed by the allowable pressure change rate. As shown in FIG. 23, the relationship 1123 between the pressure P and the saturation temperature T _SAT is non-linear. Now, the set α (P) of saturation temperature change rates at pressure P is And the saturation temperature change rate allowable value is α _L , the allowable pressure change rate β (P) 1124 at pressure P is And the characteristic curve shown in FIG. 24 is obtained. This characteristic shows that the higher the pressure level, the larger the allowable pressure change rate. The control system block diagram 1125 shown in FIG. 25 makes use of this feature in boost control.

(2) Metal match control 1610 The basic processing procedure of metal match control is shown in FIG. Since this plant uses the medium pressure turbine starting method,
The reheat steam temperature T _RH is T _RMCHN (a value obtained by converting the metal match condition lower limit temperature of the intermediate pressure turbine into the reheat steam temperature,
(Hereinafter referred to as Negative Max for medium-pressure turbine), it means that the metal match condition has been established.
Acceleration by the medium-pressure turbine is possible. If it is low, wait for the temperature to rise in that state. However, the main steam temperature T _MS at the time when the metal match condition is established is T _MMCHP (a value obtained by converting the metal match condition upper limit temperature of the high-pressure turbine into the main steam temperature.
If it is higher than the above), it means that the temperature rise of the main steam has been too early and the load increase due to the high pressure turbine ventilation is impossible, so the speed increase by the intermediate pressure turbine is no longer meaningful.
That is, the metal match fails. In addition, T _MS >
If you _connect with T _MMCHP , you also fail in metal matching. If the main steam temperature T _MS after completion of the acceleration is lower than T _MMCHN (a value obtained by converting the metal match condition lower limit temperature of the high-pressure turbine to the main steam temperature, hereinafter referred to as the Negative Max value for the high-pressure turbine), Waiting for steam temperature rise. After that, when T _MS > T _{MMCHN and the metal} match condition is established, the process shifts to the concomitant possible condition determination function of the load increase phase. If you wait forever to establish the metal matching condition after completing the speed increase, limit the simulation time (T
_LIMIT ) and consider it as a startup failure. This saves the calculation time of the simulation.

 Next, a procedure for calculating the metal matching condition will be described.

i) Negative Max value (T _RMCHN ) 1611 for medium pressure turbine Fig. 27 shows Nega of reheat steam temperature for medium pressure turbine.
The calculation procedure of the _tive Max value (T _RMCHN ) will be shown. In this example,
The metal match lower limit temperature T _RSMIN of the steam temperature in the intermediate turbine ball during aeration is set to a value lower than the ball temperature T _IBO by 50 ° C. The figure shows the process is for calculating the reheat steam temperature T _RSMIN as steam temperature inside the ball is T _RSMIN. In this processing, the calculation routine (method of calculating the steam temperature in the bowl from the reheat steam temperature) included in the turbine stress calculation function 1230 is shared, and the method of conversely _obtaining T _RMCHN from T _RSMIN by convergence calculation is used.

ii) Positive Max value (T _MMCHP ) 1612 for high-pressure turbine Figure 28 shows Positi of main steam temperature for high-pressure turbine.
The calculation procedure of ve Max (T _MMCHP ) is shown. In this example, the metal match upper limit temperature T _MSMAX of the steam temperature after the first-stage high-pressure turbine during ventilation is set to a value higher by 50 ° C. than the rotor surface temperature (which is considered to be equal to the casing inner wall temperature). The process shown in the figure is intended for steam temperature after the first stage is to calculate the main steam temperature T _MMCHP such that T _MSmax. This process also uses the turbine stress calculation function 12 as in the previous section.
The calculation routine (method of calculating the steam temperature after the first stage from the main steam temperature) included in 30 is shared, and the method of conversely calculating T _MMCHP from T _MSMAX by the convergence calculation is _adopted .

iii) Negative Max value (T _MMCHN ) 1613 for high pressure turbine Fig. 29 shows Negati of main steam temperature for high pressure turbine.
The calculation procedure of the ve Max value (T _MMCHN ) is shown. This processing procedure is exactly the same as the previous section, and the metal match lower limit temperature T
_The main steam temperature T _MMCHN corresponding to _MSMIN is obtained.

(3) Speed-up control 1640 FIG. 30 shows a processing procedure of speed-up control. The characteristics of this processing are as follows.

i) The turbine can be started in a speed-up pattern in which the speed-up time is the shortest by predicting the stress generated in the medium-pressure turbine and sequentially obtaining the maximum speed-up rate at which the predicted value is equal to or less than the allowable value.

ii) To improve the accuracy of stress prediction, use the plant model as it is.

In this method, it is assumed that the speed is increased at the maximum speed increase rate DN (1) from the reference time T _IMEO to T _NVARY and then the speed is maintained, and the stress generated in the turbine is predicted until the time T _IMEO T _NUP . As a result, if the predicted value of the stress is less than or equal to the allowable value at any time, the time T _IMEO + T at the _speed increase rate DN (1).
Actually _speed up to _NVARY (as a simulation for startup). Conversely, if the predicted value is greater than or equal to the allowable value, 1
Set the speed up rate DN (2) under the rank to the model and predict the generated stress. Even in the case of DN (3), if the predicted stress value does not fall below the allowable value, set DN (4) to keep the speed. In this way, when the time T _IMED reaches T _NVARY , this time is set as the reference time T _IMEO again, and the same processing is performed. When the rated speed is reached by repeating the above process, the speed-up control is completed, and the process moves to the next process for determining the possible condition for merge.

(4) Judgment of possible conditions for merger No. 1620 shows the procedure for judging the condition for possible mergers.
The content of this processing is roughly divided into the following two, as indicated by the broken line in the figure.

i) State prediction after merger Predict the generated stress after inputting the initial load (L = 3%) using the plant model, and determine whether this is below the allowable value in the entire prediction section T _IL . The part to do. If it is less than or equal to the allowable value, load combination is performed.

ii) Waiting for concomitant conditions If the prediction result in the previous section is negative (if the predicted stress is above the allowable value), control cycle T _CS without load co-inclusion and without load operation (state prediction next) Wait until you do). During this period, if the main steam temperature T _MS exceeds the upper limit temperature T _MMCHP of the metal matching condition of the high-pressure turbine, the metal matching fails and the startup fails. If the metal match does not fail, the process after i) is predicted again by returning to the process of i) in the next control cycle. If the condition for enabling the parallel insertion is satisfied, the process proceeds to the next load increase control.

(5) Load increase control 1630 FIG. 32 shows the processing procedure of the load increase control. This control method is basically the same as the speed-up control method. In this method, the maximum load change rate D is from the reference time T _IMEO to T _LVARY.
It is assumed that the load increases at L (1) and then the load is maintained, and the stress generated in the turbine is predicted until time T _IMEO + T _LUP . As a result, if the predicted value of stress is less than or equal to the allowable value at any point in time, the load change rate DL (1) is taken as
_The load actually _increases up to _IMED + T _LVARY (as a startup simulation). On the contrary, when the predicted value exceeds the allowable value, the load change rate DL (2) one rank lower is set in the model and the generated stress is predicted. Even in the case of DL (3), if the stress prediction value still does not fall below the allowable value, set DL (4) and put it in the load holding state. In this way, when the time reaches the next control cycle T _IMEO + T _LVARY , this time is set as the reference time T _IMEO again, and the same processing is performed. When the target load is reached by repeating the above processing, the startup is completed.

As described above, the starting time required to reach the target load from the boiler ignition is T _X.

By the method explained above, the shortest start schedule Then, the activation time T _{X, Q} is obtained _, but the following processing is performed in order to realize the on-time activation. As shown in FIG.
At the current time T ₀ , the boiler is ignited and the shortest start schedule The expected start completion time T _PD when starting according to the following _formula is T _PD = T ₀ + T _{X, Q} (10). Therefore, the plant waits at a time interval of ΔT ₁ until the error between the designated start completion time T _SET and the above T _PD becomes equal to or less than the allowable value ε. When the error is less than the allowable value ε, the shortest start schedule Is set to the schedule execution function 2000 as. However, if the error between the specified time T _SET and the estimated time T _PD is greater than or equal to the predetermined value T _L , the optimality of the startup schedule cannot be guaranteed if this condition is kept waiting. Therefore, the system waits for the time step Δt ₂ and then restarts the optimal startup schedule again. Ask. However, ΔT ₂
<T _SET −T _PD

〔The invention's effect〕

According to the present invention, since the on-time start and the rapid start of the plant can be realized at the same time, there is an effect that the load adjustment accuracy in the power system is improved and the reliability of the power supply is improved. Furthermore, in each power plant, there is an effect that the start-up loss is reduced as the start-up time is shortened, and the burden on the operator is reduced.

[Brief description of drawings]

Fig. 1 shows the basic functional configuration of the thermal power plant shortest start method, Fig. 2 shows the basic processing procedure of the start schedule optimization algorithm to which the complex method is applied, Fig. 3 shows the meaning of symbols, and Fig. 4 shows Fig. 5 shows the operating conditions, Fig. 5 shows the operating conditions, Fig. 6 and Fig. 7 show the processing procedures, Fig. 8 shows the sensitivity to the driving limiting factors, and Fig. 9, Fig. 10, Fig. 11 show the monitoring algorithm. Fig. 12 is a flow chart for explaining the center of gravity, Fig. 13 is a flow for determining a new trial point, Fig. 14 is a flow for determining whether a trial point cannot be retracted, and Figs. FIG. 22 shows the processing flow, FIG. 22 shows the procedure flow of simulation, and FIG. 23 shows the relationship between pressure P and saturation sensitivity T _SAT .
Fig. 4 shows the relationship between stress P and β (P), Fig. 25 shows the control block diagram, Fig. 26 shows the basic procedure for metal match control, and Fig. 27 shows
The figure shows the procedure for calculating the reheated steam temperature, FIG. 28 shows the procedure for calculating the main steam temperature, and FIGS. 29 to 32 show the processing procedure flow for the main steam temperature and others. 1000: Start schedule creation function, 2000: Schedule, 3000: Actual plant, 1100: Schedule optimum function block.

Claims (2)

(57) [Claims] 1. A start-up schedule creating function for creating a time-dependent start-up schedule related to operation and control target value setting required for starting a thermal power plant before starting the plant, and actually executing the start-up schedule according to the created start-up schedule. In a start-up system of the plant having a schedule execution function for starting a plant, the start-up schedule optimization function searches the start-up schedule that minimizes the start-up time under the condition that the start-up completion time is kept. Unit and a plant dynamic characteristic prediction unit, the plant dynamic characteristic prediction unit includes a plant dynamic characteristic model for predicting a start characteristic when the plant is assumed to be activated according to a start schedule, and the plant dynamic characteristic. The starting characteristics obtained by the model and measured from the actual plant before starting A stress calculation unit that calculates the stress of the plant based on the state value and a built-in stress calculation unit, and whether the stress value calculated by the plant dynamic characteristic prediction unit violates the operational constraint conditions in the schedule optimization unit. A constraint condition infringement determining unit that determines whether or not the start value is corrected to search for an optimum startup schedule in which the stress value does not violate the operation constraint conditions and has the shortest startup time, and the plant dynamic characteristic model is obtained. And a schedule setting function, wherein the schedule execution function is configured so that the error between the start completion time in the start schedule optimized by the schedule optimization unit and the start completion time specified in advance becomes less than or equal to an allowable value. A thermal power plant on-time shortest start method, which is equipped with a means for putting the plant on standby. 2. The thermal power plant according to claim 1, wherein the stress calculation unit includes a stress calculation unit of a boiler and a stress calculation unit of a turbine which constitute the thermal power generation plant. Shortest time starting method.