JPS623400B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for controlling the operation of a nuclear reactor, and particularly to a control rod for a forced circulation boiling water reactor, a recirculation flow rate,
The present invention relates to a method of controlling the operation of a nuclear reactor that is suitable for making a large change in output in the shortest time by controlling pressure and feed water temperature.

Conventionally, boiling water reactor (hereinafter referred to as BWR)
During operation, as the control rods move, local neutron flux distribution may be distorted, potentially exceeding thermal limits, so operators must constantly monitor the readings of the local neutron flux detector inside the reactor. Moreover, the operation method was determined by trial and error using the results of a calculation code that calculates the three-dimensional neutron distribution, and the start-up time was required from tens to hundreds of hours. Since various limiting conditions are set for the operation of control rods and recirculation flow rates, values such as three-dimensional neutron flux, three-dimensional maximum linear power density, and minimum critical heat flux are determined based on these limiting conditions. Finding a method to suppress the value to a desired value or less is a task that requires considerable time and effort. In particular, in the process of large-scale changes in reactor power, such as increasing the reactor power from extremely low power to rated power, operators must pay great attention to local monitoring within the reactor as described above. Overall measures in the process of significantly increasing the furnace power (for example from 10% in the shortest time
It was not possible to know how to increase the output to 100%, and it took an unnecessarily long time.

An object of the present invention is to eliminate the above-mentioned drawbacks of the prior art, and to be able to appropriately select a control means that can increase the reactor output to a predetermined output in a short time, thereby increasing the output up to the predetermined output. It is an object of the present invention to provide a method for controlling the operation of a nuclear reactor that can perform a rise in a short time.

A feature of the present invention is that the state quantities of the reactor are input into a single-point reactor model, and an operation command is generated that can change the current reactor output to the target reactor output in the shortest possible time.
A three-dimensional model predicts the local state of the reactor output when the reactor output is controlled in response to the operation command obtained using the one-point reactor model under the constraints of the one-point reactor model, and the predicted local state If the thermal constraint is satisfied, the output control means necessary to execute the operation command are selected and the output control means are sequentially operated.

Hereinafter, a preferred embodiment of the present invention will be described with reference to the drawings. Figure 1 shows the typical operating characteristics of a forced circulation BWR using a recirculation pump.
This is the core flow rate W _P map. A natural circulation line obtained when the control rod is withdrawn when the recirculation pump rotation speed is zero, a flow rate restriction line due to the upper limit of the recirculation pump rotation speed, a line for preventing pump cavitation, and a reactor core. The area bounded by the control rod withdrawal prevention line (), which represents a limit on the overall average heat flux, defines an operating tolerance range. Operational constraints include prohibition of control rod operation in high power ranges (for example, above 60% power), limits on the rate of increase in power due to recirculation flow rate, and limits on the linear power density of each fuel rod. In general, the control variables that can vary BWR reactor power include recirculation flow rate, reactor pressure, feedwater temperature, and control rods. FIG. 2 shows the basic configuration of this embodiment, which uses the control variables described above to achieve a large output change in the shortest time. The operation control device of this embodiment includes a data recording device 2 that always takes in and stores the latest plant state quantities (for example, neutron flux, thermal output, core flow rate, control rod position, and respective rate of change) from the reactor plant 1. In addition, the characteristics of the nuclear reactor determined from the plant state quantities mentioned above, especially Zenon (a nuclide produced by nuclear fission, which has a strong absorption cross section for neutrons and its response time are extremely long (more than 10 hours)) A single-point reactor model calculation unit 3 calculates the shortest control time using a single-point reactor approximation model that includes the dynamic characteristics of , which have a particularly large effect during startup, which lasts for tens to hundreds of hours, and takes into account the spatial extent of the reactor core. A three-dimensional distribution calculation device 4 calculates whether the above-mentioned shortest time control is achievable under various constraint conditions (control rod operation speed, heat flux limit, etc.), and a reaction given by the control rods. Reactivity estimator 5 that calculates the degree from actual plant data
and a comparison/judgment unit 6 which monitors the deviation between the actual reactivity and the target reactivity and sends a command to the arithmetic units 3 and 4 to correct the model. The single-point reactor model calculation device 3 calculates the heat output obtained from the data recording device 2;
Based on data such as core flow rate and control rod position, data for a single-point reactor model is created, and parameters are estimated that take into account the effects of changes in output level.
And based on this result, a single point model estimator 311 calculates operational constraints when viewed as a single point furnace, and (1)
It consists of a single point furnace model section 312 having a mathematical model expressed by equations (3) to (3).

dX(t)/dt=λ ₁ (t)−λ ₂ X(t) +γ ₂ Σ・φ(t)−σ ₂ φ(t)・X
(t) ...(1) d(t)/dt=-λ ₁ (t)+γ ₁ Σφ(t
)... (2) φ=(X(t), W _P (t), P(t), T _W (t), △ρ _rpd )...(3) However, φ is the neutron flux, X is the xenone concentration, is the iodine concentration, λ ₁ is the iodine decay constant, λ ₂ is the Zenon decay constant, γ ₁ is the direct generation rate of iodine, γ ₂ is the direct generation rate of Zenone, Σ is the fission cross section, σ
₂ is the Zenon micro absorption cross section, W _P is the recirculation flow rate, P is the core pressure, T _W is the feed water temperature, Δρ _rpd is the reactivity of the entire core average due to the control rods, and t is the elapsed time.

In particular, equation (3) shows that the output φ is the reactivity due to the control rod,
It is a functional equation that shows how it is affected by feed water temperature, reactor pressure, core flow rate, and Zenon, and is a static model that ignores transient phenomena of several minutes or less. The shortest time control law calculating device 32 uses the model obtained by the single-point reactor model section 312 to perform control that can change the output from the current output to the target output in the shortest time, using the one-point reactor model estimation section 311. It is calculated under the given constraint conditions and is obtained as a function of time as shown in FIG. As a method for calculating the shortest time control, for example, the method described in the reference document ("Application of Gradient Method to Process Optimization", Sayama, Oi) can be used. In addition,
In order to make it easier to understand the relationship between points S, A, B, D, E, and F in FIG. 1, which will be described below, the times corresponding to those points in FIG.
It is represented by dashed dotted lines indicated by D, E and F. Also,
The * mark above the symbol of each vertical axis item in FIG.
This indicates that the value was obtained from the target value.

The curve in Figure 3 is plotted on Figure 1.
This is a curve from point S to point F via points A, B, D, and E. The indicators shown beside the curves indicate that the control was performed by flow control FL, control rod control C, pressure control P, and feed water temperature control T, respectively. In particular, the reason why the solution is found in a loop like this is because we have imposed a restriction that prohibits control rod operation at high power. The reactivity is shown in Figure 3 as △ _ρrpd ), and from point A to point B, the output is increased at the flow rate W _P , and the neutron flux is held at a point as high as possible to accumulate Zenon. After that, the output is rapidly reduced to point D, and the dynamic characteristics of Zenon shown in Fig. 4 are used to further accumulate enough Zenon, and the thermal conditions such as the linear power density of the fuel rods are relaxed, and then the control is started. The rated power point F is reached by withdrawing the rod to point E and then increasing the core flow rate W _P . The feed water temperature T and pressure P are manipulated to more efficiently accumulate xenone at points B and D in a shorter time.

Up to this point, the nuclear reactor is only a single-point reactor approximation, and although we have provided guidelines on how to operate the reactor as a whole in order to complete startup in the shortest time, we have also given constraints on the single-point reactor model. Depending on how the conditions are given, there is no guarantee that the local thermal conditions are necessarily satisfied, and it is also unclear which of the many control rods to move and how.

Therefore, the three-dimensional distribution calculation device 4 calculates the distribution of neutrons, bubble distribution, heat flux distribution, average reactivity, etc. by solving the diffusion equation, which is given by the one-point reactor model calculation device 3. Determining whether the operation commands specified satisfy thermal constraints such as local linear power density, and determining which controls under thermal constraints are necessary to achieve the control rod reactivity specified as △ _ρrpd . It has a function to explore how far to move the stick. In other words, the operating condition restriction condition calculation unit 411
takes in necessary data from the data recording device 2 and determines operational constraints as a three-dimensional model. Based on the operation constraints and the operation command obtained by the single-point reactor model calculation device 3, the three-dimensional neutron distribution and control rod reactivity calculation device 412 calculates the distribution and average reactivity of neutrons, bubbles, and heat flux in three dimensions. Calculations are made using a model, and operating conditions are determined based on the calculated results. The obtained operating conditions are compared with operating constraints (thermal limitations), and it is determined whether the obtained results satisfy the thermal limitations. Calculation device 4
All information obtained in 12 (including judgment results)
is displayed on the CRT device 42. Calculation device 412
determines the specific control rod position for realizing the given Δρ _rpd through repeated interaction with the operator via the CRT device 42. Furthermore, after repeating the interaction with the operator several times, the calculation device 412 compares the obtained operating conditions with the thermal limits, and finally determines whether the calculated result satisfies the thermal limits. do. If it is determined that the condition is not satisfied, the calculation device 412 outputs a command to correct the one-point furnace model to the one-point furnace model calculation device 3. As a result of the above-described final determination, if the constraint conditions are satisfied, the calculation device 412 outputs the determined control rod position to the control rod pattern generation device 43. The control rod pattern generator 43 sequentially operates the control rods according to a given procedure. In this way, the single-point reactor model calculation device 3 always gives the shortest time control method from the current output to the target output, but the three-dimensional neutron distribution calculation device 4 receives data from the single-point reactor model calculation device 3 every period △T. In response to the command, a control means to realize Δρ _rpd at that point is searched for. After the control means is selected, the control rods, flow rate, pressure, and feed water temperature are operated based on the shortest time operation command obtained by the single point reactor model calculation device 3.

The reactor reactivity estimator 5 inputs measurement data of the reactor plant 1 and constantly calculates the reactivity Δρ _rpd due to the control rods added to the actual reactor. This reactivity △ρ _rpd is input to the comparison/judgment device 6 and is the optimum reactivity △ calculated by the single point reactor model calculation device 3.
ρ is compared with _rpd . The comparison/judgment unit 6 sends the three-dimensional neutron distribution calculation device 4 a three-dimensional signal when the actual reactivity △ρ _rpd does not reach the target value, the optimum reactivity △ρ _rpd , even after all control rod operation sequences are completed. Issue a command to modify the original model. As a result, the three-dimensional model is modified to a more accurate model, so that the accuracy of determining the control means of the control device of this embodiment thereafter is improved.

FIG. 5 is a chronological diagram showing the processing contents performed by the above-mentioned control device. The contents will be explained one by one.

First, data required for a one-point reactor model is imported from the data recording device 2 into the one-point model estimation unit 311 . Then, operating constraint conditions are determined.
Next, the single-point furnace model 312 establishes a single-point furnace model by entering operational constraints and determining coefficients of the single-point furnace model. The shortest time control law calculation device 32 determines the control that can change the output from the current output to the target output in the shortest time as described above, and also sends this control information and the obtained Δρ _rpd to the three-dimensional neutron distribution calculation device 4. Output. The three-dimensional neutron distribution calculation device 4 takes in data for a three-dimensional model from the data recording device 2, determines the most reliable operating constraint conditions, and establishes a three-dimensional model. After that, the three-dimensional neutron distribution calculation device 4 calculates the neutron distribution and reactivity in response to operating commands such as control rod and core flow rates given by the single-point reactor model, and further calculates operating conditions based on these calculation results. At the same time, the obtained operating conditions are compared with operating constraints (thermal constraints) to determine whether the operating conditions are acceptable. Further, through interactive processing via the CRT device 42, the three-dimensional neutron distribution calculation device 4 determines the control rod position for realizing Δρ _rpd . After such processing in the three-dimensional neutron distribution calculation device 4 is repeated, the three-dimensional neutron distribution calculation device 4 determines whether condition (a) shown in R ₁ in FIG. 5 is satisfied, that is, It is determined whether the obtained operating conditions satisfy operating constraints (thermal limitations). For example, one point reactor model calculation device 3
If the one-point reactor model built into the model or the operating limit conditions for the one-point reactor model are extremely inappropriate, the three-dimensional distribution calculation device 4 will be able to satisfy the thermal limitations and achieve the target △ρ _rpd . Even if you search the control rod array, it may not be found. In this case, the determination based on condition (a) indicates that the above-described operating conditions are not met, and the one-point furnace model and its constraint conditions are determined to be inappropriate, and the one-point furnace model is established.
When the operating conditions are determined to satisfy condition (a), the three-dimensional distribution calculation device 4 outputs an operation command to a control means such as a control rod drive device that operates the control rods. Even if all control rod operation sequences are completed, the actual reactivity △ρ _rpd obtained by the reactor reactivity estimator 5 does not reach the target value △ρ _rpd (R ₂ in Figure 5). When the condition (b) shown in
(=Δρ _rpd −Δρ _rpd ) is output to the three-dimensional distribution calculation device 4. The three-dimensional distribution calculation device 4 which receives the deviation signal δ improves the three-dimensional model and then calculates the
Search for new control rods to add and move. If Δρ _rpd reaches the target value Δρ _rpd during the control rod operation sequence, the control rod operation is stopped at that point according to condition (b). Next, other control rod manipulated variables such as recirculation flow rate, reactor pressure set point, and feed water temperature set point are sequentially manipulated to achieve the target core flow rate, reactor pressure, and feed water temperature. Therefore, at this stage, the shortest time control determined by the single-point furnace model has been applied at a certain control time. The output is monitored while the control is in place. In addition, from a certain control point up to the optimum point, the actual furnace output and the optimal solution (furnace output) found by the single-point furnace model calculation device 3 are calculated.
When the difference is smaller than the specified value (condition (c) indicated by R ₃ in Figure 5) is not satisfied, that is, when the above-mentioned difference exceeds the specified value, one point is set again. Calibrate the furnace model and find the shortest time control from the current output to the target output. In FIG. 5, each item and specific execution procedure are shown with time as the horizontal axis. However, in reality, the sampling period ΔT is approximately one hour, which is sufficient based on Zenon's transient characteristics shown in Figure 4. There is also plenty of time to get results. Also the time to operate the control rod,
Considering that the dynamic characteristics of Zenon are sufficiently gradual, the time required to change the flow rate, pressure, and feed water temperature without causing large transient fluctuations in the reactor and plant is not a problem, and the above-mentioned control variables The order in which they are operated does not matter much.

In this example, the output control of a nuclear reactor, which is a nonlinear system, can be determined specifically and quickly by appropriately using a single-point reactor model and a three-dimensional model, and by repeatedly using them as necessary. I can do it. Therefore, when changing the BWR's output significantly, which previously relied entirely on the operator's experience, it is possible to instruct the overall control strategy, that is, the appropriate control that can complete the output change in the shortest amount of time, in a short time. . Therefore, BWR output can be changed safely and in a short time, improving load followability to the power grid.

In particular, in this example, a control method suitable for increasing the reactor output to the target output is determined using a single-point reactor model, and the control method obtained using this single-point reactor model is input into a three-dimensional model. We are evaluating whether the control method is appropriate using a three-dimensional model. Therefore, an appropriate control method for increasing the reactor output to the target output can be obtained in a short time, and the memory capacity of the electronic computer can also be reduced. The control method obtained using the single-point reactor model is to increase the reactor output to the target output in the shortest possible time, and in this example, an appropriate control means for increasing the output can be found in a short time, so it is practical. The time required to increase the reactor power to the target power can be significantly reduced. The computer may also be small.

As the output of the single-point reactor model calculation unit 3, instead of the control rod reactivity change Δρ _rpd , it is also possible to use the average multiplication factor change Δk _rpd for the control rod over the entire core, and use equation (4). Therefore, it is clear that the same effect can be obtained with exactly the same configuration and method.

△ρ _rpd = △k _rpd −1/△k _rpd …(4) Also, if the three-dimensional model is highly accurate in all situations, the reactivity estimator 6 and its associated functions can be omitted. Of course.

According to the present invention, it is possible to instruct in a short time a control that can change the output of a nuclear reactor in the shortest possible time and safely. Therefore, changes in the reactor output can be completed in a short time. Furthermore, the memory capacity of the electronic computer can be reduced.

[Brief explanation of the drawing]

Figure 1 is a characteristic diagram showing the typical operating characteristics and operating tolerance range of a boiling water reactor, Figure 2 is a basic configuration diagram of the large output change device of the present invention, and Figure 3 is a single point reactor model. An explanatory diagram showing an example of the shortest time control law calculated by the calculation unit. Figure 4 is a diagram showing a transient change in Zenone concentration due to a stepwise decrease in output (or neutron flux). Figure 5 is a basic diagram. FIG. 3 is a characteristic diagram showing the execution status of each arithmetic function in FIG. 2, which is a configuration diagram. DESCRIPTION OF SYMBOLS 1...Nuclear plant, 3...Single point reactor model calculation device, 4...Three-dimensional distribution calculation device, 5...Reactivity estimation device, 6...Comparison judgment device, 31...Single point reactor model simulation element, 32... ... Minimum time control law calculation device, 41 ... Three-dimensional distribution calculation device, 42 ...
CRT dialogue device, 43...Control rod drive pattern generator, 311...Single point reactor model estimation unit, 312
...Single point reactor model section, 411...Operation limit condition calculation section, 412...Three-dimensional neutron distribution and control rod reactivity calculation device.

Claims (1)

[Claims] 1. Input the state quantities of the reactor into a single-point reactor model, and generate an operation command that can change the current reactor output to the target reactor output in the shortest time at the single-point reactor under the constraint conditions of the single-point reactor model. A three-dimensional model predicts the local state of the reactor output when the reactor output is controlled in accordance with the operation command obtained using the reactor model, and the predicted local state satisfies thermal constraints. If the predicted local state does not satisfy the thermal conditions, select the output control means necessary to execute the operation command and operate the output control means sequentially, and if the predicted local state does not satisfy the thermal condition, Alternatively, a nuclear reactor operation control method that repeats the above-mentioned process after modifying the three-dimensional model.