JPH0413679B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a method and apparatus for predicting the state of a plant having a dynamic state such as a nuclear reactor plant or a chemical plant, and particularly to a method for predicting the state of a plant after an abnormality occurs. The present invention relates to a method and an apparatus for the same.

[Background of the invention]

For example, state prediction devices for boiling water reactors can be broadly classified into (1) those that have a static model that assumes a steady state, and (2) those that have a dynamic model that calculates transient states.

An example of the former is a core performance prediction device, which uses a phenomenon simulator (hereinafter abbreviated as simulator) having a static model. Then, data from the plant in a steady state is taken in, parameters included in the simulator are adjusted, and the state of the plant (power level, power distribution in the core, etc.) at any subsequent time is determined to indicate that the plant is in a steady state. It is assumed and obtained using a simulator.

On the other hand, the latter is used when the plant deviates from a steady state due to the occurrence of an abnormality, for example, and in boiling water reactors, there is a device that predicts the reactor water level, pressure, etc. after an abnormality occurs. This prediction device determines (initializes) the parameters included in the simulator using the actual measured values of various state quantities in the stationary equipment before the occurrence of an abnormality, and calculates transient changes in the plant state after the occurrence of an abnormality. .

The present invention is directed to the latter plant state prediction device.

FIG. 1 shows the latter prior art described above, and in the figure, time points t _p , ts and _te represent the time points at which an abnormality occurs, the start of prediction, and the end of prediction, respectively. This condition prediction device uses a simulator.
Since it is necessary to initialize with the actual measured value of the steady state before t _p , it is necessary to uselessly execute calculations from t _p to t _e by the simulator even though the actual measured value is available. Therefore, if many hours have passed from t _p to t _s , that is, if you start prediction after a long time has passed after the occurrence of an abnormality, the calculation time will be longer and the prediction will not be possible until a long time has elapsed since the start of prediction. The problem is that no results are obtained. Furthermore, since errors in the simulator's model accumulate over a long period of time, there is a problem in that the discrepancy between the obtained prediction results and the actual measurements becomes large. Generally, it is difficult to improve the accuracy of a dynamic simulator model compared to a static simulator model. In addition to it,
If the type of abnormality cannot be known in advance, the following problems arise.

For example, when a pipe rupture accident occurs, the area of the rupture, the quality of the rupture, etc. cannot be accurately known in advance. Therefore, the fracture area is determined by performing predictive calculations assuming the fracture area, comparing the plant data (data from t _p to t _s ) and the predicted results, and repeating the calculation until they match. This is a possible method. However, this method has the problem of requiring an enormous amount of calculation time because the long calculations described above must be repeated. Furthermore, due to the accumulation of errors in the simulator model, plant data and prediction results do not match, making it impossible to determine the fracture area and making prediction calculations impossible. Furthermore, the quality of the rupture is affected by changes in the reactor water level due to the rupture.
X _BRK (Quality of cooling water flowing out from the break)
In a situation where the value changes significantly ( _for example, from liquid _phase X _BRK = 0 to gas _phase Even if these values are determined by the method, if the liquid phase has changed to the gas phase, the values will differ significantly from the actual values at time _ts , and not only will they not be usable as subsequent data, but they will also be difficult to predict. There is a problem that the results differ significantly from the changes in the state of the plant after t _s .

[Purpose of the invention]

The purpose of the present invention is to solve the problems of the prior art described above, and to provide a method for predicting the state of a plant with a dynamic state that can accurately predict the state of the plant at any point in time after an abnormality occurs in a short calculation time; An object of the present invention is to provide a state prediction device that reliably implements a state prediction method.

[Summary of the invention]

In order to achieve this objective, the method for predicting the state of a plant with a dynamic state according to the present invention is based on the following method:
Changes in the various measured signals necessary to determine the state of the plant at any point in time after it deviates from the steady state are used to evaluate the various variables included in the physical model that describes the plant or a part thereof. The present invention is characterized by estimating the value and predicting changes in the state of the plant after the point in time using the physical model.

Further, the state prediction device for a plant having a dynamic state according to the present invention includes a plant data recording device that captures and stores various measurement signals necessary for determining the state of the plant when the plant deviates from a steady state; A prediction device comprising a part that estimates various variables included in a physical model describing a plant or a part thereof using data stored in the plant data recording device, and a part that predicts a state using the physical model. , a display device that displays information related to estimation of various variables included in the physical model, state prediction by the physical model, and data stored in the plant data recording device; and control of the plant data recording device and the display device. It is characterized by comprising an input device for inputting a request and data related to a plant operation system necessary for prediction by the prediction device.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to FIGS. 2 to 8.

FIG. 2 is a diagram _{showing the} basic state prediction method of the present invention in a nuclear reactor plant.
t _f and t _e respectively indicate the times of occurrence of an abnormality, activation of the prediction device, completion of estimation of unmeasurable parameters, and completion of prediction. Further, Δt _i is a predetermined time range. The state prediction method of the present invention starts automatically or at the request of an operator when the occurrence of an abnormality is detected by a scram signal, etc.
Start importing actual measured data on plant conditions. Then, the latest data is always stored for only Δt _i . When a prediction start command is input, the prediction device starts (t _s ) and uses the measured data stored from (t _s −Δt _i ) to t _s to calculate the unmeasurable parameters included in the simulator model. Estimate,
Changes in plant conditions such as reactor water level and pressure after t _f are predicted using the actual measurement data or predicted values by the simulator at time t _f when parameter estimation is completed as initial values. Then, when the prediction ends at time t _p , the results are displayed.

FIG. 3a schematically shows an actual nuclear reactor, and FIG. 3b schematically shows a reactor model of a simulator included in the state prediction device of the present invention. Figure 3a
, 1 is the core, 2 is the lower plenum, 3 is the recirculation system, 4 is the jet pump, 5 is the steam separator, 6 is the
1 is a water supply system, 7 is a main steam system, 8 is a safety relief valve, 9 is an isolation cooling system, 10 is an emergency core cooling system, 11 is a rupture in the piping connected to the pressure vessel 16, 1
2 is a pressure gauge, 13 is a water level gauge, 14 is a downcomer, and 15 is a shroud wall.

The simulator of this embodiment models this actual nuclear reactor as shown in FIG. 3b. That is, it is assumed that the inside of the pressure vessel is divided into two regions filled with saturated water 17 and saturated steam 18, and the pressure is uniform within the pressure vessel. Also, consider a heating element 19 that simulates heat generation in the reactor core. Regarding the flow of cooling water into and out of the pressure vessel, the actual cooling system of a nuclear reactor, breaks, etc. are summarized in the three flow paths shown in the figure. That is, they are grouped into three: a saturated steam outlet channel 20, a saturated water outlet channel 21, and a cooling water inlet channel 22.

Considering the mass balance and energy balance for the above model system, the mass balance of saturated water is as shown in the following equation.

dm _f / dt＝W _l,io −W _f,put −m〓 _ev ……(1) Here, t: Time m _f : Saturated water mass W _l,io : Flow rate W of cooling water flowing into the pressure vessel _f,put : Flow rate of saturated water flowing out from the pressure vessel m〓 _ev : Amount of evaporation of saturated water per unit time The mass balance of saturated steam is as follows.

dm _g /dt=m〓 _ev −Wg, out...(2) where, m _g : Saturated steam mass Wg, out: Saturated steam flow rate flowing out from the pressure vessel Also, the energy of cooling water in the pressure vessel Considering balance, using equations (1) and (2), and considering that the volume inside the pressure vessel is constant, the following equation can be obtained.

∂P／∂t=−v _f (W _l,io −W _f,put )−W _g,put v _g +v _g −v _f
/h _g −h _f {Q〓＋W _l,io (h _l,io −h _f )}／m _f ∂v _f /∂P+
m _g ∂v _h ／∂P−v _g −v _f ／h _g −h _f (m _f ∂h _f ／∂P＋m _g ∂h _g ／
∂P−αV……(3) Where, P: Reactor pressure v _f : Saturated water specific volume v _g : Saturated steam specific volume h _f : Saturated water enthalpy h _g : Saturated steam enthalpy Q〓: Cooled from heating element Amount of heat transferred to water per unit time h _l,io : Enthalpy of inflowing cooling water α : Work equivalent of heat V : Volume of pressure vessel When solved in combination with the equation (steam table), the change in pressure can be found from equation (3), and the change in the mass of saturated water, that is, the change in water level, can be found from equation ( ₁ ). Once determined, it can be found from the steam table, and from this, the volume and water level of saturated water can be found). In order to solve the above equation, it is first necessary to find the values of pressure P, saturated water mass m _f , and saturated steam mass m _g at the time of starting the prediction, and their time derivatives. . These can be determined from actual measurements as follows. In other words, the pressure can be determined using a pressure measurement value using a pressure gauge. Further, the saturated water mass can be determined by using the water level measurement value using a water level gauge, the results regarding the shape and pressure of the pressure vessel, and the steam table. Saturated steam can be determined based on the results for the saturated water mass, the shape of the pressure vessel, and the steam table.

Another matter necessary to solve the above equation will be explained next.

In equations (1), (2), and (3), W _l,io , W _f,put , W _g,put ,
h _l,io and m〓 _ev are expressed as follows.

W _l,io = W _FE +W _RCIC +W _ECCS ……(4) W _l,put = (1−X _BRK )・A _BRK・G _BRK (P, X _BRK ) ……(5) W _g,put =X _{BRK ・ A BRK ・ G BRK ( P , _ _ _ _ _} +W _ECCS ) ……(7) m〓 _ev =1/h _g −h _f {Q〓＋W _l,io (h _l,io −h _f ) −∂P/∂t(m _f・∂h _f ／∂ P＋m _g ∂h _g ／∂P−αV)}
...(8) Here, G: Mass flux W: Flow rate A: Area h: Enthalpy MSL: Main steam system Here, equation (8) is derived from equations (1) and (2) and the energy balance.

In equations (4) to (8), since the flow rate and enthalpy of each system are measured in the plant, these values can be obtained from actual measurements at each point before the prediction start point. Also, regarding the point in time when there is no actual measured value after the start of prediction,
Regarding flow rate, if you input the system operation (operation for operation) status and use the interlock and control system model that determines automatic start/stop of the system,
You can ask for it. On the other hand, assuming that changes in cooling system enthalpy can be ignored (changes are generally gradual), the enthalpy of cooling water from each cooling system is also determined. Furthermore, regarding the mass flow rate flowing out from the rupture port, pressure and rupture _quality
Since we already have a model that asks for it from X _BRK
By determining , G _BRK can be obtained. Also,
The area of rupture A _BRK also needs to be determined in order to obtain W _f,put and W _g,put . Regarding Q〓 in equation (7), there is already a model that gives it as a function of time after the scram occurs.

That is, in order to solve equations (1) to (3) in combination with equations (4) to (8), it is necessary to determine two unmeasurable parameters, A _BRK and X _BRK .

FIG. 4 is a schematic diagram showing the configuration of the state prediction device in this embodiment. In the figure, 23 is a nuclear reactor plant, 24 is a data recording device, 25 is a prediction device, 26 is an operator console, and 27 is a display device. The prediction device 25 includes a simulator 28 having the above model, and the simulator 28
is connected to the parameter estimation unit 29 and the phenomenon prediction unit 30 that estimate the above-mentioned unmeasurable parameters A _BRK and X _BRK . The plant data recording device 24 is
It is activated automatically when an abnormality occurs or manually by a activation signal from the operator console 26, and the reactor pressure and water level from the reactor plant, the cooling water flow rate of the emergency core cooling system and other systems, and the enthalpy ( temperature) and the actual data for a predetermined time (Δt _i ).
Always remembers the latest data. When a prediction start command is input from the operator console 26, the prediction device 25 is activated and the parameter estimator 2
9 takes in actual measurement data of the plant stored in the plant data recording device 24, and uses the data to estimate unmeasurable parameters included in the simulator. When this estimation is completed and input of manual operation schedule data for the cooling system from the operator console 26 is completed, the phenomenon prediction unit 30
starts, and after setting the initial values of the simulator 28, it starts predicting the phenomenon. When the prediction is completed, the display device 27 displays the predicted results of the reactor water level and pressure, as well as the estimated A _BRK , X _{BRK ,} etc.

FIG. 5 is a schematic diagram chronologically showing the processing by the state prediction device. As shown, the activation signal is t _s
First, from the recorded (t _s −Δt _i ), t _s
Using actual plant data of the time Δt _i ,
Start estimating the unmeasurable parameters A _BRK and X _BRK . At the same time, system operation data input by the operator is also captured. parameters
_{When the} estimation _of A _{BRK ,}
Set the initial value at , and set the reactor pressure (P) after t _f ,
Predict the water level (l). And when the prediction is finished,
The results are displayed and the process ends (t _e ). Here, if the input of the system operation data by the operator is later than the parameter determination, the above-described plant data recording, prediction, etc. are performed with the time of input of the system operation data as t _e .

FIG. 6 is a schematic diagram chronologically showing the processing by the state prediction device when changes in reactor water level and pressure are predicted assuming three types of system operations. As shown in the figure, when the device is activated at time t _s , estimation of the parameters A _BRK and X _BRK is started as described above, and these parameters are determined at time t ¹ _f . Since the input of system operation data (1) has already been completed at time t ¹ _f , the plant data at t ¹ _f is immediately imported, and then the phenomenon prediction and result display after t ¹ _f are performed. Implemented.

At the display end time t ¹ _e , when a request to end the prediction is received from the operator, the prediction device shown in FIG. 4 stops. However, if this request is not received, the system waits for the next system operation data to be input. Then, when the input of system operation data (2) is completed (t ² _f ), the plant data at that time t ² _f is taken in, and the pressure and water level after t ² _f are predicted. Input of system operation data is possible even while predictive calculations are being performed. This is shown by the system operation data (3) input, and in this case, as soon as the prediction and result display after t ² _f is completed, the next system operation is started.
The process for prediction assuming (3) is started in the same manner as above.

FIG. 7 shows the flow of processing for determining unmeasurable parameters A _BRK and X _BRK in the parameter estimation section. When the parameter estimator starts up (t _s ), it first assumes A _BRK and X _BRK . Measurements such as reactor pressure and water level are then used to determine the time (t _s
−Δt _i ), the values of the reactor pressure P, the saturated water mass m _f , and the saturated steam mass m _g and their time differential values are determined. On the other hand, the flow rate and enthalpy of the emergency core cooling system and other cooling systems are determined from plant data in the time range from (t _s −Δt _i ) to t _s . Then, by solving equations (1) to (8) in combination with the steam table, changes in reactor pressure and water level from (t _s −Δt _i ) to t _s are calculated. Once the calculation result is obtained, it is compared with the measured reactor pressure and water level values from ( _ts - Δt _i ) to _ts . As a result, if the difference between the calculated value and the measured value is larger than the predetermined value ε, A _BRK and X _BRK are assumed again and the above process is repeated. This process is repeated until the deviation becomes equal to or less than ε, and the values of A _BRK and X _BRK when the process is completed are adopted as the final A _BRK and X _BRK . Here, this device uses nonlinear programming to modify the assumed values of A _BRK and X _BRK .

FIG. 8 is a diagram showing the predicted results of the reactor pressure and water level when the present device is activated in the event of total loss of water supply and stuck safety relief valve in the reactor. Here, an offline simulator, which is generally used for nuclear reactor safety analysis, is used to simulate signals from the plant. In the figure, _tp indicates the time point at which total water supply loss occurred, and _ts indicates the time point at which the device was activated. Further, in this example, Δt _i was set to 1 minute. The device starts up at t _s , and the time from t _s to the time t _e at which the unmeasurable parameters are determined is approximately 1 second (the calculation time on a general-purpose computer);
This is a sufficiently practical time.

Next, a second embodiment of the present invention will be described with reference to FIGS. 9 to 12.

FIG. 9a schematically shows an actual nuclear reactor, and FIG. 9b schematically shows a reactor model of a simulator included in the state prediction device of the present invention. In the simulator model of this embodiment, as shown in FIG. 9b, the pressure vessel has a region filled with saturated steam 31;
2 of the region filled with saturated water 33 and voids 32
It is assumed that the pressure is uniform within the pressure vessel. In addition, a heating element 1 that simulates heat generation in the reactor core
Think of 9. Regarding the flow of cooling water into and out of the pressure vessel, the actual cooling system of a nuclear reactor, breaks, etc. are summarized in the three flow paths shown in the figure. That is, the saturated steam outlet flow path 2
0, a saturated water outlet flow path 21, and a saturated water inlet flow path 23.

The above model system differs from the model system of the first embodiment only in that voids in saturated water are taken into consideration. Therefore, if the mass of the total saturated water in the pressure vessel is m _f , the mass of the total saturated steam is m _g , and the mass of the saturated steam that exists as voids among the saturated steam is m _g , then Equations (1) to (8) used in the first example
holds true as is. Regarding m〓, if we assume that the voids are uniformly distributed in the height direction and that their average rising speed is V _slip , the following equation can be obtained from the mass balance.

dm〓／dt＝m〓 _ev −2V _slip /l×m〓 ……(9) Here, l: Water level In equation (9), V _slip can be calculated using the experimental formula or the value measured by experiment. You can decide. Also,
The water level l can be determined if the saturated water mass m _f and the void mass m〓 are known. After all, in order to combine and solve equations (1) to (9), it is sufficient to find the initial value of m〓 in addition to A _BRK and X _BRK .

FIG. 10 is a schematic diagram showing the configuration of the state prediction device in this embodiment. In the figure, 23 is a nuclear reactor plant, 24 is a data recording device, 34 is a prediction device, 26 is an operator console, and 27 is a display device. In this device, the data recording device 24, the operator console 26, and the display device 27 have exactly the same functions as those in the device of the first embodiment described above. Prediction device 34
are unique to this embodiment, and include a simulator 35 with a model that takes the above-mentioned voids into consideration, fracture area A _BRK , fracture quality X _BRK , and void mass m〓 at the time of starting the simulator. It is composed of a parameter estimation section 36 that estimates unmeasurable parameters and a phenomenon prediction section 37.

FIG. 11 shows the processing in the apparatus according to this embodiment in chronological order. When the start signal is input at time t _s , this device first captures the actual plant measurement data recorded in the data recording device for the period Δt _i from time (t _s −Δt _i ) to time t _s , and then records it. to start estimating the unmeasurable parameters A _BRK , X _BRK and m〓(t _s −Δt _i ). In parallel, system operation data entered by operators from the operator console is also captured. Parameters A _BRK , X _BRK and
When the estimation of m〓(t _s −Δt _i ) ends at t _f , the time t _f
Take in the plant data at and estimate the initial value at t _f of the simulator. Only the value of the void mass m〓(t _f ) at t _f cannot be determined by plant data;
A simulator is used to calculate from (t _s −Δt _i ) to t _f , and the value of m〓(t _f ) obtained as a result is adopted as the initial value. After setting these initial values, the reactor pressure P and water level l after t _f
Start predicting. Then, display the obtained results.

Figure 12 shows the determination of unmeasurable parameters A _BRK , X _BRK , m〓(t _s −Δt _i ) in the parameter estimator and
This figure shows the flow of processing for determining m〓(t _f ). In the parameter estimation part, first
_{Assuming A BRK , _ _} Find the values and differential values of P, saturated water mass m _f , saturated steam mass m _g , and the time differential value of void mass m〓. On the other hand, the flow rate and enthalpy of the cooling system are determined from the plant data in the time range from ( _ts - Δt _i ) to _ts . Then, equations (1) to (9) can be solved in combination with the steam table, and changes in reactor water level and pressure from (t _s −Δt _i ) to t _s can be calculated. Next, this calculation result is compared with the actually measured values of reactor pressure and water level. Then, if the deviation between the calculated value and the measured value is larger than a predetermined value ε, A _BRK , X _BRK , and m〓(t _s −Δt _i )
Reassume and repeat the above process. This _process is repeated until the difference between the calculated _value and the measured value becomes less than ε, and the values _of A _BRK , Adopted as. Next, using these determined values, a simulator calculates the time point t _f from (t _s −Δt _i ) when the value is determined.
Perform calculations up to and calculate the void mass m〓 at t _f
Determine (t _f ). Note that in this device as well, as in the device according to the first embodiment, nonlinear programming is used to modify the assumed values of _ABRK and others.

Further, a third embodiment of the present invention will be explained with reference to FIGS. 13 and 14. FIG. 13a schematically shows an actual nuclear reactor, and FIG. 13b schematically shows a nuclear reactor model of a simulator included in the state prediction device of the present invention. In the simulator model of this embodiment, the pressure vessel is divided into a total of four regions, two regions outside the shroud wall 15 and two regions inside the shroud wall 15, as shown in FIG. 13b. That is, the inside of the shroud wall is divided into a lower plenum region 39 and a core region 40, and the outside of the shroud wall is divided into a downcomer region 38 and a steam dome region 41. Reference numerals 42 to 46 indicate channels flowing into and out of the respective regions. We are also considering a heating element in the core region that simulates the heat generated by the fuel. Consider energy balance and mass balance in each of the above areas. Further, assuming that the pressure is uniform within the pressure vessel, differential equations for the pressure within the pressure vessel, the mass of cooling water in each region, and enthalpy can be obtained.

Regarding the above model, if it is assumed that the cooling water in the pressure vessel is saturated at the time (t _s −Δt _i ) and that the void 47 exists only in the core, then
_There are three unmeasurable parameters: fracture area A _BRK , fracture quality It becomes possible to predict reactor water level and pressure. Furthermore, in the case of the apparatus according to the present embodiment, since the core void ratio, the core water level, etc. are also calculated, the present apparatus can be used to estimate these state quantities that cannot be measured.

Figure 14 shows the cooling water amount _MSP and cooling water temperature of the suppression pool which is a part of the containment vessel.
An example applied to the prediction of T _SP is shown in Fig. 14a.
1 is a schematic diagram of a reactor containment vessel, and FIG. In the figure, 4
8 is the reactor pressure vessel, 49 is the reactor containment vessel, 5
0 is a dry well, 51 is a suppression pool, 52 is various cooling systems using the suppression pool as a water source, 53 is a heat exchanger for the cooling system for cooling the suppression pool, 54 is a vent pipe, 55 is a safety relief valve, 56 indicates a rupture in the piping connected to the reactor pressure vessel. In the apparatus according to this embodiment, the suppression pool is modeled as shown in FIG. 13b. In other words, the cooling water that enters the suppression pool 57 (the cooling water from the relief safety valve 55 that enters the suppression pool 51 in FIG. (corresponding to the suppression pool) is shown by a flow path 60, and the cooling water discharged from the suppression pool 57 by various cooling systems is shown by a flow path 58.
The amount of heat removed by the pool cooling system 59 is defined as Q. Considering the amount of cooling water and the amount of energy flowing in and out of these channels, considering the energy balance and mass balance of the cooling water in the suppression pool 57, and making appropriate assumptions, the mass and temperature of the cooling water can be calculated. A first-order ordinary differential equation for time is obtained.

In the above equations, the initial values for mass and temperature are determined from plant data. In addition, the flow path 5
The flow rate and heat removal amount Q in No. 8 are obtained from the plant data before the prediction start time, and from the plant data after the prediction start time.
It can be determined by giving the operating status of the system. Therefore, by making the flow rate and energy from flow path 60 two unmeasurable parameters, the mass and temperature of the suppression pool can be predicted in exactly the same manner as in the previously described embodiment. It is possible.

In addition, the suppression of the device according to this embodiment
By combining the pool model and the reactor model in the embodiment shown above to create a model that comprehensively describes the system consisting of the containment vessel and pressure vessel shown in Figure 14a, the reactor pressure, water level, It is possible to create a device that can predict changes in the suppression pool water temperature, its cooling water mass, etc. all at once.

Further, by combining the device according to the present invention with a driving guide device, etc., it can be effectively used to determine an appropriate driving method, a margin time until operation, and the like.

Incidentally, although the above-mentioned embodiments are all applied to a nuclear reactor plant, the same functions and effects as described above can be achieved even when applied to a chemical plant.

〔Effect of the invention〕

As explained above, according to the state prediction method of the present invention, when an abnormality occurs in a plant with a dynamic state, predictions of subsequent phenomena can be made in a short time by calculation at any point after the abnormality occurs. , and can be performed with high precision. This makes it possible to know in advance the margin time until cooling system operation and the effects of operation, especially in the case of nuclear reactor plants, which is useful for determining operating methods to prevent the spread of abnormalities and safely shut down the reactor. .

Moreover, according to the state prediction device of the present invention, the above-described state prediction method can be reliably implemented.

[Brief explanation of drawings]

FIG. 1 is a diagram explaining a conventional state prediction device;
2 to 8 show a first embodiment of the present invention, FIG. 2 is a diagram explaining the basic method of the present invention, FIG. 3 a shows an actual nuclear reactor, and FIG. 4 is a schematic diagram showing the reactor model of the device of the present invention, and FIG. 4 is a schematic diagram showing the configuration of the device of the present invention.
5 and 6 are schematic diagrams showing the processing in the device of the present invention in chronological order, FIG. 7 is a diagram showing the flow of processing in the parameter estimation section, and FIG. 8 is the prediction result by the device of the present invention. Diagrams showing, Figures 9 to 1
2 shows a second embodiment of the present invention, FIG. 9a is a schematic diagram showing an actual nuclear reactor, FIG. FIG. 11 is a schematic diagram showing the configuration of the device of the present invention; FIG.
Figure 2 shows the flow of processing in the parameter estimating section, and Figure 13 shows a third embodiment of the present invention, where a shows an actual nuclear reactor and b shows a diagram of the device of the present invention. Schematic diagram showing each nuclear reactor model, No. 14
The figures show a fourth embodiment of the present invention, in which Figure a is a schematic diagram showing an actual containment vessel, and Figure b is a schematic diagram showing a containment vessel model possessed by the apparatus of the present invention. 1...Reactor core, 2...Lower plenum, 3...Recirculation system, 4...Jet pump, 5...Steam water separator, 6...Water supply system, 7...Main steam system, 8...Relief safety valve, 9... Isolation cooling system, 10... Emergency core cooling system, 11... Fracture, 12... Pressure gauge, 13
... Water level gauge, 14 ... Downcomer, 15 ... Shroud wall, 16 ... Pressure vessel, 17 ... Saturated water, 18 ... Saturated steam, 19 ... Heating element, 20 ...
... Saturated steam outlet channel, 21... Saturated water outlet channel,
22... Cooling water inlet channel, 23... Nuclear reactor plant, 24... Plant data recording device, 25...
Prediction device, 26... Operator console, 2
7...Display device, 28...Simulator, 29...
...Parameter estimation unit, 30...Phenomenon prediction unit, 31
... Saturated steam, 32 ... Void, 33 ... Saturated water, 34 ... Prediction device, 35 ... Simulator,
36...Parameter estimation section, 37...Phenomenon prediction section, 38...Downcomer region, 39...Lower plenum region, 40...Core region, 41...Steam dome region, 42-46...Cooling water flow path, 47...
Void, 48...Reactor pressure vessel, 49...Reactor containment vessel, 50...Dry well, 51...Suppression pool, 52...Various cooling systems, 5
3... Heat exchanger for cooling the suppression pool, 54... Vent pipe, 55... Safety relief valve, 5
6...Rupture, 57...Suppression pool,
58... Channel, 59... Heat removal, 60... Channel.

Claims (1)

[Claims] 1. A state in which when a plant in a dynamic state deviates from a steady state and shifts to a transient state, changes in the state of the plant are predicted from the values of various measurement signals necessary for determining the plant state. The prediction method includes determining changes in the various measurement signals within a predetermined time range immediately before the start of plant state prediction after deviation, and calculating the values of various variables included in a physical model that describes the plant or a part thereof. A method for predicting the state of a plant having a dynamic state, comprising estimating from the change and using the estimated values of the various variables to predict a change in the plant state after the start of the plant state prediction. 2. Claims characterized in that a change in the state of the plant when the system operation is performed is predicted from the operation data when the hypothetical system operation is performed and the estimated values of the various variables. A method for predicting the state of a plant having a dynamic state as described in item 1. 3 Physics that represents the state inside the reactor pressure vessel using changes within the predetermined time range of measurement signals related to the pressure inside the reactor pressure vessel, the cooling water level, and the operating status of various systems installed in the nuclear power plant. The state of a plant having a dynamic state according to claim 1, characterized in that the values of various variables included in the model are estimated and changes in the pressure in the reactor pressure vessel and the water level of the cooling water are predicted. Prediction method. 4 Using changes within the predetermined time range of measurement signals related to the state of cooling water in the reactor containment vessel and the operating status of various systems installed in the nuclear power plant, to create a physical model representing the state inside the reactor containment vessel. 2. A method for predicting the state of a plant having a dynamic state according to claim 1, characterized in that the values of various variables included are estimated to predict changes in the state of cooling water in a reactor containment vessel. 5 Using changes within the predetermined time range of measurement signals related to the pressure in the reactor pressure vessel, the cooling water level, the cooling water state in the reactor containment vessel, and the operating status of various systems installed in the nuclear power plant, Estimating the values of various variables included in a physical model representing the state inside the reactor pressure vessel and the state inside the reactor containment vessel, and comprehensively predicting the state of cooling water in the reactor pressure vessel and the containment vessel. A method for predicting the state of a plant having a dynamic state as claimed in claim 1. 6. A state prediction method for predicting changes in the state of the plant from the values of various measurement signals necessary for determining the plant state when a plant in a dynamic state deviates from a steady state and shifts to a transient state, comprising: a plant data recording device that measures and stores changes in the various measurement signals within a predetermined time range immediately before the start of plant state prediction after deviation; and measured values of the various measurement signals stored in the plant data recording device. an estimation device that estimates the values of various variables included in a physical model that describes the plant or a part thereof from changes; and a prediction device that uses the estimated values of the estimation device to predict changes in the state of the plant after the start of the plant state prediction. A state prediction device for a plant having a dynamic state, comprising a prediction device and a display device that displays the prediction result and the estimated value. 7. An input device for inputting hypothetical system operation data, the prediction device predicting a change in the plant state when the system device is performed from the estimated value and the hypothetical system operation data, and the display device 7. An apparatus for predicting the state of a plant having a dynamic state according to claim 6, characterized in that the prediction result is displayed on the apparatus.