JP3335779B2 - Plant performance monitoring system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a plant performance monitoring system in a power plant, and more particularly to a data continuity system.
The present invention relates to a plant performance monitoring system that is highly reliable and reduces the load on operators.

[0002]

2. Description of the Related Art In the operation management of a plant, it is essential to monitor the performance of the plant to grasp the state of the plant. In a computer system for monitoring the performance of a plant, it is important to maintain data management of the plant. In a conventional plant performance monitoring system, a dual system is used to continuously record plant data, and the recording of plant data is continued.

In this system, a means for detecting the state of the other computer is provided between the computers to check the operation state between the computers, and when one system is stopped, the other system backs up its work. ing. FIG. 16 shows a configuration of a conventional plant performance monitoring system, and a case where one computer A stops will be described. In this case, a case will be described in which the computer has two systems consisting of A and B.

When an A-system abnormality is detected by the A-system abnormality detecting section 1, the A-system state sending section 2 is notified of the abnormality.
Upon receiving this notification, the A-state sending unit 2 turns off the contact signal (ON / OFF signal) indicating that an abnormality has occurred in the computer A, and sends a hard-wired signal to the A-system state receiving unit 7 on the computer B side. Send via. The A-system state receiving unit 7 receives the abnormality of the computer A, and notifies the A-system state determining unit 8 that there is an abnormality. The A-system state determination unit 8 determines the state of the computer A from the notification of the abnormality, and when it determines that the computer A is stopped, activates the A-system backup processing unit 10,
It backs up task A that was performed by the stopped computer. B-system backup processing is similar to the above processing.
The steps are performed in the order of 5, 6, and 9.

FIG. 17 shows a configuration diagram for releasing a conventional backup. Note that a system in which the business A and the business B operate on different CPUs for the purpose of load distribution in a computer having a redundant configuration is referred to as a load sharing system. Now, when computer B is restored in a state where computer B has been backed up (operations A and B are operating on computer A) due to a failure of computer B, application A activation requesting device 132 and application B A business recovery request is issued to both of the activation request devices 112.

[0006] In response to this, the service activation request devices 132, 112
Initializes each computer, activates the business A133 and business B113, and restores the original load share system. By the way, usually, data used for performing plant performance monitoring are plant data and operation history dependent data calculated from the plant data.

The current operation history dependent data is data calculated based on current plant data and past operation history dependent data, is a kind of integrated data, and is updated each time calculation is performed. Therefore, as a condition of the operation history dependent data to be stored, it is necessary that a certain number of data is always accurate and up-to-date at the same timing, that is, the data is sound.

If the past operation history dependent data does not satisfy the above conditions, the performance calculation cannot be continued and the plant monitoring cannot be continued. For this reason, in the current situation, the operator periodically saves the data on magnetic tape etc. in preparation for the destruction of the history-dependent data, and if the data is destroyed, manually restore the magnetic tape and resume plant performance monitoring Let me.

As an example, monitoring of core performance of a nuclear power plant will be described. The core performance is monitored by calculating the output of the nuclear fuel loaded in the core. To perform this calculation, the reactor pressure, reactor flow, and reactor The data from the neutron detector installed in the system is input, and the current core state is calculated based on the data and the past (normally, the result of the previous calculation).

Since the xenon data used as past data and the combustion data for each nuclear fuel are cumulative data (integrated data) from the start of the plant, some or all of the data may be lost due to a storage device failure or the like. If destroyed, the output of the nuclear fuel cannot be calculated, which may make it impossible to monitor the core performance.

Conventional core performance monitoring is performed with the configuration shown in FIG. Core (reactor) 201 and plant
Plant data obtained by sensors from 202
Are collected by the data receiving unit 204 and input to the core state calculating unit 206.

In the core state calculation unit 206, the data receiving unit 20
The current core state is calculated by inputting the past (previous) calculation result data 207 input from the plant data 205 input from 4 and the core state data storage unit 209,
The result 208 is sent to the core calculation result output unit 210 and output to a printer device or the like, while the data is sent to the core state data storage unit 209 for use as base data for the next calculation and stored.

Since the core data is extremely important data, the core state data is output from the magnetic tape output unit about once a week by an operator from the viewpoint of backup in case of destruction.
The calculation data 212 is output from the 213 to the magnetic tape 214 and stored. Also, in the present situation, if the computer goes down due to some cause, after the computer is restored, the calculation is restarted at the cycle timing again from the operation history dependent data calculated last.

For example, a case of a nuclear power plant will be described below. The state of the core is monitored by monitoring the power of the nuclear fuel loaded in the core, but the power of the nuclear fuel in the core cannot be directly measured.
At present, the output is calculated using a three-dimensional physical model.

In order to perform this calculation, the reactor pressure, reactor flow, and data from a neutron detector installed in the reactor, which are called plant data, are input, and the data is compared with the past (usually, The current core state is calculated based on the calculation result) data. The in-furnace fuel output is calculated from the current plant data together with the xenon data and the fuel burnup data, which are the previous core state data, and at the same time, the current xenon data and burnup data are calculated.

As described above, the reactor core monitoring system for calculating the power inside the reactor is an extremely important function in operating a nuclear power plant. Since it is necessary to continue the monitoring function, the monitoring function is automatically continued using the calculation data immediately before the computer is stopped.

FIG. 19 shows still another conventional example. In this example, when the calculation is performed in the core state calculation unit 306, the calculation process is performed according to the signal from the computer cycle management unit 313. At the time of calculation, the data 305 from the data receiving unit 304 and the data 30 from the core state data storage unit 309 were used.
It is natural to use 7. Also, core calculation result output section
Reference numeral 310, core state data magnetic tape input / output unit 311 and magnetic tape 312 are the same as those in FIG.

[0018]

In the above-mentioned conventional plant performance monitoring system, when the means for detecting the state of the partner computer fails, the two computers are connected to each other even though the computers are normal. In some cases, it is determined to be stopped, and the business is backed up, or the backup is not performed because the stop of the other party cannot be detected.
Furthermore, after the two computers have backed up, the recording device is simultaneously accessed to record the plant data, which causes a problem of inconsistency in the plant data.

In the past, when restoring the state of the load sharing system, it was not possible to normally terminate only one side of each business in the backup state without losing data consistency. For this reason, it is necessary to initialize the computer on the healthy side, and it is necessary to suspend the work that is not transferred to the recovery computer, and the function is stopped during the recovery work.

Further, it is conceivable that the constituent devices fail for some reason, and as a result, the computers constituting the plant performance monitoring system stop. If the computer system stops during the calculation of the periodic operation history dependent data, the calculation result data is only stored halfway. As a result, consistency between data cannot be obtained, meaning not only meaningless as data, but also data destruction may occur in some cases.

In this case, even if the computer is restored, since the sound operation history dependent data is not stored, the performance cannot be calculated, and the plant performance monitoring cannot be continued. Therefore, in order to restart the plant performance monitoring, the operator must restore the collected data from the magnetic tape and take a recovery procedure.

However, since it takes a lot of time for the operator to collect data every time the cycle is calculated, the data is actually collected only once a week. Therefore, the data to be restarted is often old, and there is a problem that the accuracy of the calculation result is deteriorated because the data is continued from the data to the present time.

At present, there is no method of restoring the cycle calculation of the operation history dependent data after the computer is stopped, and the length of the computer stop period and the fluctuation of the plant state during the stop period are not particularly checked. Further, in the calculation of the periodic operation history dependent data, it is not considered that the cycle period is long, and it is not assumed that there is a large change in the plant state during that period.

Therefore, when the computer suspension period is long, accurate calculation of the operation history-dependent data is not performed.
The error affects plant performance monitoring. At present, considering this effect, when the operator stops the computer, the period and the state of the plant are considered, and if necessary, the operation history dependent data is corrected.

For example, the core performance monitoring in a nuclear power plant will be described. The present reactor fuel output, xenon, and burnup calculation are based on the assumption that the plant state has changed linearly during the calculation cycle. It is not assumed that the plant data fluctuates beyond a certain level.

If the cycle calculation is normally continued, the calculation cycle is shorter than the temporal fluctuation of the plant data due to the plant operation. Therefore, there is no problem in performing the calculation assuming that the plant state changes linearly. . However, when the computer stops and the output changes non-linearly during that time, the behavior cannot be ignored and the calculation cannot be performed assuming that the output changes linearly.

In this case, xenon and burnup are not calculated correctly, and an error occurs between the data inside the core and the data obtained by calculation. As a result, the calculation is performed based on these calculated data. However, there is a problem that the power monitoring function of the existing reactor is affected.

The present invention has been made in view of the above circumstances, and has as its object to provide a plant performance monitoring system having data continuity and high reliability and capable of reducing the burden on operators. And

[0029]

A plant performance monitoring system according to claim 1 of the present invention uses a plurality of computers to calculate plant performance using data dependent on the operation history of the plant. A performance monitoring system, comprising: first detection means for detecting an abnormality of a counterpart computer; and second detection means for detecting an operation state of a task X running on the counterpart computer, wherein the first detection means Is configured so that the function of the service X and the data file used by the service X are automatically taken over by the local computer based on the AND condition between the operation of the second computer and the operation of stopping the operation of the other computer by the second detection means. did.

In the plant performance monitoring system according to the second aspect of the present invention, a plurality of computers mutually monitor the operating state of the other computer, and when an abnormality occurs in one of the computers, a backup process is performed by the healthy computer. A system performance monitoring system, comprising: a system state determination unit that reads an abnormal state of a computer of one system; and a shared memory that stores an abnormal state of the business X of each system. X is determined to be abnormal, and the following output processing is performed. Note (a) When the computer is abnormal and the operation X is abnormal, the backup processing of the operation X is performed on the other computer. (B) When the business X is normal due to a computer abnormality, an alarm process is performed to the effect that the abnormality detection unit has failed. (C) When the computer is normal and the operation X is abnormal, an alarm process indicating that an abnormality has occurred in the operation X is performed.

In the plant performance monitoring system according to the third aspect of the present invention, two computers share the tasks X and Y and process each of them, and when one computer fails, the other computer remains. In a computer system that processes both business X and business Y, when one of the computers recovers from a failure,
Means for issuing a suspension request to the business X, which was newly assigned and processed in response to a failure, between the business X and the business Y that were being processed by the other healthy computer, and In this case, the processing of the task X is transferred to the restored computer.

[0032] The plant performance monitoring system according to [claim 4] of the present invention according to [claim 3] further comprises means for arbitrarily designating a task to be transferred to the restored computer.

According to a fifth aspect of the present invention, there is provided a plant performance monitoring system for calculating plant performance using a plurality of computers by using data dependent on the operation history of the plant. Along with two or more sets of operation history dependent data storage areas used for calculation, an identifier indicating that the data in each area is sound and an identifier indicating the update time of the data are provided, respectively.
When starting the performance calculation, a data storage area that is sound and has the latest data update time is selected from the operation history dependent data, and data necessary for the performance calculation is input from this area.

A plant performance monitoring system according to a sixth aspect of the present invention is a plant performance monitoring system for calculating plant performance using a plurality of computers by using data dependent on the operation history of the plant. The plant state stored in the data storage area selected for the start of the calculation is compared with the current plant state input from the plant data input unit, and a difference between main parameters of the two plant states is determined. A computer restart calculation unit for determining whether or not the value is within the range is provided, and the performance calculation is automatically restarted when the difference between the main parameters is within a predetermined range.

[0035]

According to a first aspect of the present invention, a plant performance monitoring system comprises: a detection unit for detecting an abnormality of a computer; and a business operation status of the computer. The hardwire and a shared memory are provided between the computers. Grasping the state of the computer accurately, performing backup processing and continuing to record plant data.

In the plant performance monitoring system according to the second aspect of the present invention, when tasks are shared by a plurality of computers, each system status determination unit which reads the status of one of the computers reads an abnormality. Depending on the abnormality of the computer and the abnormality of the business, the backup is performed by another computer, the alarm of the failure of the abnormality detection unit, or the alarm of the business abnormality is separately output.

In the plant performance monitoring system according to the third aspect of the present invention, when the computer is restored, it is not necessary to initialize the healthy computer, and it is possible to prevent a functional stop due to the computer initialization.

In the plant performance monitoring system according to claim 4 of the present invention, by newly providing a business stop request device, it is possible to prevent a function stop of a business with a high data continuation request.

In the plant performance monitoring system according to claim 5 of the present invention, the data is managed for multiple generations by switching the storage destination of the data each time the calculation is performed. Even if there is an inconsistency, sound data remains one generation before, that is, the other, and after the computer is restored, the data is read from the previous generation and calculated, eliminating the need for the operator to restore the data. Becomes This eliminates the need for data recovery decisions and data recovery that were previously performed by operators, not only reducing the load on operators, but also monitoring plant performance by continuing immediately. Time loss is minimized, which contributes to safe operation of the plant.

The plant performance monitoring system according to claim 6 of the present invention can prevent execution of a calculation that causes an error even if there is a large change in plant data during a computer suspension period. Further, when the stop period has no influence, the cycle calculation is automatically continued, so that the plant performance can be monitored with high accuracy while reducing the load on the operator.

[0041]

An embodiment will be described below with reference to the drawings. An embodiment of a plant performance monitoring system according to [claim 1] of the present invention will be described below. In FIG. 1, the same parts as those in FIG. 11 is business A
A state detection unit, similarly, 12 is a business B state detection unit, and 13 is a shared memory having a business A state area and a business B state area. When the operation A becomes inoperable due to the abnormality of the computer A, the operation A state detection unit 11
The task A data is set as the status. Similarly, when the operation B cannot be operated due to the abnormality of the computer B, the operation B state detection unit 12 sets the data of the operation B abnormality in the operation B state of the shared memory 13.

The A-system status judging unit 8 is connected to the business A status area of the shared memory 13, and the B-system status judging unit 6 is also connected to the business B status area of the shared memory 13. Here, since the operation of each system abnormality detecting unit 1 or 3 is the same as that of the conventional example, the description is omitted, and the operation of the newly added portion will be described.

The A-system status judging unit 8 judges the status of the computer A from the notification from the A-system status receiving unit 7 and the operation A status of the shared memory 13, and when it judges that the computer A is stopped, the A-system backup processing unit 10 is started and the operation environment of the operation A is set, and the operation A is backed up. If one of the A-system status receiving unit 7 and the shared memory 13 notifies the
When set to the system status determination unit 8, an alarm is generated to the effect that the detection unit that has notified the stop has failed.

FIG. 2 is a flow chart showing the processing contents of the A-system state judging unit 8, and this part is described in [Claim 2]. First, in step 21, the A-system state determining unit 8 compares the state of the computer A received by the A-system state receiving unit 7 with the shared memory.
The status of business A from 13 is read (21, 22). Next, processing when the abnormality detection unit of the computer A becomes abnormal and when the operation A becomes abnormal will be described. First, when the abnormality detection unit of the computer A becomes abnormal, the state determination unit determines the received state of the computer A, and if this is the stop of the computer A (23), the state of the business A is determined (24). .

If the task A is also abnormal, the task A is backed up by the computer B (25). If the operation A is normal, an alarm for a failure of the abnormality detection unit is generated and the operator is notified (27). Next received computer A
If the status of the job is normal, the status of business A is determined (2
6) If the job A is abnormal, an alarm for the job A is generated (28) and the operator is notified.

Another embodiment of the plant performance monitoring system according to claim 1 of the present invention will be described with reference to FIG. In FIG. 3, the same portions as those in FIG. 3 is that a shared magnetic disk drive 14 is added. That is, in the above-described embodiment, the state of the computer is transmitted using the shared memory as the state transmission means to the other party. In this embodiment, instead of the shared memory, a medium which can be accessed simultaneously by each computer, for example, a magnetic disk It has been replaced with a device. It is clear that the present embodiment can achieve the same effect as the above embodiment.

An embodiment of a plant performance monitoring system according to claim 3 of the present invention will be described below with reference to FIGS. 4, 5, 6, and 7. 4 is a configuration diagram, FIG. 5 is a flowchart showing processing contents, FIG. 6 is a configuration diagram of a task state table, and FIG. 7 is a flowchart of task operation. A and B in FIG. 4 denote respective computers. The computer A includes a service activation request device 132 and a service recovery request device 121 for performing the service A133, and the computer B also has a configuration corresponding thereto.

The computer A has a service B 104 as a backup of the partner computer B. The task B has a task status table 105, a monitoring device 111, and a stop request flag.
It has 103. The computer B also has a configuration corresponding to this. In the normal state, computer A has job A,
In the computer B, the task B is operating.

In the situation according to the present invention, the computer B is down and the computer A is in the backup operation (the services A and B are operating).
Is the case. First, in FIG. 7, at the time of starting the task of the backed-up business B 104, it is checked whether the stop request flag 103 of the business B is set (see FIG. 7).
1) If the stop request flag 103 is set, the process is terminated without starting, and if not, the task is started.

The started task first sets the task identifier of its own task in the task status table 105 (72),
The data is updated (75), and at the end of the task, the own task identifier of the task status table 105 is reset (78). In this way, the task operation status of the task B is always registered. FIG. 6 shows the configuration of the task status table 105. The task status table 105 is monitored by the monitoring device 111.

The process when the computer B has recovered is shown in FIG. 5, and the business recovery request device 101 notifies the stop request flag 103 relating to the task start permission of the business B 104 of the recovery of the computer B (51). Upon receiving this recovery notification, the stop request flag 103 is set (53), and the activation of a new task of the task B 104 is prohibited (54). Business B104 to date
Although the operation status of the task is registered in the task status table, after the stop request flag 103 is set, the task B 104 only resets the task identifier of the task status table 105 (55).

After monitoring the task status table 105, the monitoring device 111 confirms the end of all tasks of the task B104, and then sends the task B10 to the task activation request device 112 of the computer B.
Notify the end of 4 (56). The service activation request device 112 starts the service B 113 (57). As a result, the load sharing system is restored without stopping the operation A (FIGS. 4101 to 113). The backup can be canceled by the same means at the time of backup by the computer B (FIG. 4121 to FIG. 4).
133).

An embodiment of a plant performance monitoring system according to claim 4 of the present invention will be described below with reference to FIGS. In FIG. 8, the same parts as those in FIG. In this embodiment, each computer is provided with a task A and a task B, and each of them is provided with a dedicated task status table 105, 108, 125, or 128.
Further, the service stop request devices 102 and 122 are provided, and similarly to the above, dedicated service stop request flags 103, 106, 123 and 126 are provided.

That is, in the load sharing state before the backup, a task state table dedicated to both the tasks A and B is provided, and the operation status of the task is always stored in the task state table by the same means as the setting and resetting of the task identifier. It is to be registered (see FIGS. 6 and 7).

In this state, computer B goes down and computer A
When the backup is in operation (operations A and B are running), the operation A 107 is stored in the operation A task state table 108 in the operation B 10.
No. 4 registers the operation status of all tasks in the business in the business B task status table 105.

FIG. 9 is a flow chart showing the processing contents when the computer B is restored. Now, when the computer B has recovered in this situation (91), the business recovery request device sends a computer B recovery notification to the business stop request device 102. In response to this, the business stop request device 102 determines the business to be stopped (92), the business A stop request flag 106 (93), and the business B stop request flag 10
3 Send a stop request to (93-1). Device for requesting stop operation by manual selection and operator input to determine stop operation
There is an automatic selection specified in 102. In addition, depending on the priority, the low-priority work is always performed in advance by the stop request
To be specified.

Thus, the service to be stopped is determined, and the service on one side is stopped in the same procedure as in [Claim 3]. The monitoring device 111 notifies the business start request device 112 of the stopped business. In response to this, the task activation request device 112 activates the task stopped on the computer A. As a result, the operation is restored to the load sharing system without stopping the operation on one side (FIGS. 8101 to 114). The backup can be canceled by the same means at the time of backup at the computer B (FIG. 8121 to FIG. 8).
134).

An embodiment of a plant performance monitoring system according to claim 5 of the present invention will be described below with reference to the drawings, taking a core performance monitoring device of a nuclear power plant as an example. FIG. 10 shows the configuration of the present embodiment. Plant data 203 obtained by sensors from the core (reactor) 201 and the plant 202 is collected by the data receiving unit 204 and input to the core state calculating unit 206.

In the core state calculation unit 206, the data receiving unit 20
4 to calculate the current core state by inputting the past (previous) calculation result data 207 input from the plant data 205 input from the core state data sampling determination unit 219, and output the result 208 to the core calculation result output unit. The data is sent to the printer 210 and output to a printer device or the like, while the data is sent to and stored in the core state data storage determination unit 220 so as to be used as base data for the next calculation.

The core state data sampling determination section 219 checks the data in the core state data storage section A 217 and the core state data storage section B 218 which are under two-generation management, and receives an input request 207 from the core state calculation section 206 and the core state. In response to an input request 212 from the state data magnetic tape input / output unit 213,
Data which is determined to be latest and sound (the determination method will be described later) is input 215 and output 207 and 212.

On the other hand, the core state data storage determination unit 220 responds to the output request 208 from the core state calculation unit 206 and the output request 211 from the core state magnetic tape input / output unit 213 in response to the unhealthy or old core. Output to state data storage unit A217 or core state data storage unit B218
216. Repeat the above.

The present invention will be described with reference to a block diagram of FIG. 11 and FIG. 10 which is an embodiment of the present invention, divided into a calculation execution timing and a recovery timing after the computer is stopped. (Calculation execution timing) If it is determined that the calculation is to be performed (151), the core state calculation unit 206 returns to the data reception unit.
The plant data is input from 204 (152). The data receiving unit 204 periodically receives plant data 203 from the core 201 and the plant 202.

After completion of the input, the core state calculation unit 206 requests the latest and sound data from the core state data collection determination unit 219. In the core state data collection judgment unit 219,
The data storage date and time, the data write start keyword, and the write completion keyword are input from the core state data storage unit A217 (154). Subsequently, the core state data storage unit B21
Input the same data from 8 (155).

First, it is determined whether the data write start keyword and the write completion keyword input from the core state data storage unit A217 match (156).
If the keyword matches, the stored data is determined to be sound (data inconsistency, no destruction). If the core state data storage unit A217 is determined to be sound, it is then determined by the same method whether the core state data storage unit B218 is sound (157).

If the core state data storage unit B218 is also judged to be sound, it is determined which is the latest from the data storage date and time recorded in the core state data storage unit A217 and the core state data storage unit B218 (158). If the core state data storage unit A217 is determined to be the latest in (158), the data in the core state data storage unit A217 is output to the core state calculation unit 206 (159).

When the core state data storage section B218 is determined to be the latest, the data in the core state data storage section B218 is output to the core state calculation section 206 (160). If it is determined in (156) that the core state data storage unit A217 is not sound, it is determined whether the core state data storage unit B218 is sound (161). If it is determined that the state is sound, the data in the core state data storage unit B218 is stored in the core state calculation unit 206.
Output to (162).

If it is determined in this determination that the core state data storage unit B218 is also not sound, the core state calculation unit 206 cannot perform the calculation, so that an error message is output and the operator is notified of abnormal and core state magnetic conditions. Tape input / output unit 213
Output a message requesting data input from
3).

The core state data storage unit A21 of (156)
If 7 is sound and core state data storage unit B218 is judged not to be sound, the data in core state data storage unit A217 is output to core state calculation unit 206 (164). The core state calculation unit 206 that has received the data from the core state data collection determination unit 219 calculates the state of the core (165).

When the calculation in the core state calculation unit 206 is completed, the data is output to the core calculation result output unit 210 and the result is output to the printer (166). To output the core state data (167). The core state data storage determination unit 220 inputs the data storage date and time, the data write start keyword, and the write completion keyword from the core state data storage unit A217 (168). Subsequently, the same data is input from the core state data storage unit B218 (169).

First, it is determined whether the data write start keyword and the write completion keyword input from the core state data storage unit A217 match (170). If not, the core state data storage unit A217 is sent to the core state data storage unit A217. Decide to record the data (171). That is, if the keywords do not match, the data is unreliable and cannot be used even if it is newer in time. Therefore, this unreliable data is deleted by overwriting.

When it is determined that the core state data storage unit A217 is sound, the core state data storage unit B21 is subsequently operated.
8 is judged to be sound (172), and if it is judged that B218 is not sound, it is decided to record data in the core state data storage unit B218 (173). In this way, by storing data on the unhealthy side, data on the healthy side can be left as past data.

When it is determined that both the core state data storage unit 217 and the core state data storage unit B218 are sound, the core state data storage unit A217 and the core state data storage unit B218
It is determined which is older from the date and time data recorded in (174). When it is determined that the record data in the core state data storage unit A217 is old, it is determined that the data is to be recorded in the core state data storage unit A217 (17).
Five ).

If it is determined that the core state data storage unit B218 is old, it is determined that data is to be recorded in the core state data storage unit B218 (176). In this way, when the core state storage unit that outputs data from the core state data storage determination unit 220 is determined, the data is output.

First, for example, a numerical value uniquely determined from time is recorded in the core state data storage unit as a write start keyword (177). Thereafter, data from the core state calculation unit is recorded (178). After all writing is completed,
The same data as the data recorded as the write start keyword is recorded as the write completion keyword (179
).

According to this mechanism, when the write start keyword and the write completion keyword do not match, it is possible to judge that a computer stop or the like has occurred during data recording, and that data may be inconsistent or destroyed. . In the case of (163), the core state data storage unit A
If a message indicating that neither 217 nor B218 is sound is output, the operator removes the magnetic tape 214 storing the core state data from the core state data magnetic tape storage unit 213 and the core state data from the core state data storage determination unit 220. It is necessary to restore the data to the storage unit A217 or the core state data storage unit B218.

(Recovery Timing after Stop of Calculation) At the recovery timing after stop, the core state data is taken out in the same manner as the normal calculation execution timing. That is, (156 to 16
By the core state data sampling judgment unit 219 in the same manner as 4). The core state data storage unit A217 and the core state data storage unit B218 are compared, and the calculation is restarted based on sound and latest data.

An embodiment of a plant performance monitoring system according to claim 6 of the present invention will be described with reference to the drawings, taking a core performance monitoring system of a nuclear power plant as an example. FIG. 13 shows a configuration diagram of the present invention. The feature of this embodiment is that a computer restart calculation unit 314 is provided. The configuration diagram of FIG. 13 will be described. At the cycle calculation timing, the calculation cycle management unit 313 sends a cycle calculation execution command to the core state calculation unit 306. The core state calculation unit 306 collects the plant data 303 of the core 301 and the plant 302 from the data receiving unit 304, and also extracts the previous core data 307 from the core state data storage unit 309.

The core state calculation unit 306 is based on the collected data.
Calculates the current core state. After the calculation, the core state calculation unit 306 transmits the current core data 308 to the core calculation result output unit 31.
Send to 0 and output the calculation result on paper. The core data 308 is stored in the core state data storage unit 309 this time. The core state data can be input / output to / from the magnetic tape 312 via the core state data magnetic tape input / output 311.

At the time of recovery after the computer is stopped, the computer restart calculation unit 314 stores the current data 316 for computer restart comparison from the data receiving unit 304 and the previous data 315 for computer restart comparison from the core state data storage unit 309. The sampling is performed, it is determined whether or not the cycle calculation can be executed, and a calculation inhibition flag is set.

FIG. 10 shows the flow at the time of restoration after the computer stops.
This will be described with reference to FIGS. When the computer recovers from the computer stopped state (351) (352), the computer restart calculation unit 31
4 (353), the plant data 31 at the time of the previous calculation
Collect 5 (354). In addition, plant data showing the current status of the plant to check for plant fluctuations
316 is collected from the data receiving unit 304 (355).

Based on the data obtained in (353) and (354), it is determined whether or not the behavior of the plant is within a range in which automatic start can be performed (356 to 359). For example: * The pattern of the control rods has not changed (356). * Fluctuation of core flow should be within ± 10% (357). * Reactor power change should be within ± 10% (358
). * Within 10 hours since the last calculation (359). Etc.

If even one of these conditions is not satisfied, that is, if it is determined that the behavior of the plant is too large, the computer restart calculation unit 314 sets a calculation inhibition flag in order to prevent the core state calculation unit from executing. (Eg CORM)
Is set (361), and a message is output to the operator so that the supplementary processing between computer stops is correctly performed (36).
3), end (364).

At the cycle calculation timing (368), the calculation cycle management section checks the above-mentioned calculation inhibition flag.
(369, 370).

The calculation inhibition flag is cleared after correction by the operator. When the calculation cycle management unit 313 detects that the calculation inhibition flag has been cleared at the cycle calculation timing, it activates the core state calculation unit 306 and calculates the current core state based on the data complemented by the operator. (369, 37
0).

[0085]

As described above, according to the present invention, the load on the operator can be reduced, and a plant performance monitoring system with high data continuity and high reliability can be provided.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a plant performance monitoring system according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing an operation of a computer state determining unit of the plant performance monitoring system according to the first embodiment of the present invention.

FIG. 3 is a configuration diagram of a plant performance monitoring system according to a second embodiment of the present invention.

FIG. 4 is a configuration diagram of a plant performance monitoring system according to an embodiment of [claim 2] of the present invention.

FIG. 5 is a block diagram showing a processing flow of a plant performance monitoring system according to an embodiment of [claim 2] of the present invention.

FIG. 6 is a configuration diagram of a task state table. .

FIG. 7 is a block diagram showing a flow of task operation.

FIG. 8 is a configuration diagram of a plant performance monitoring system according to a third embodiment of the present invention.

FIG. 9 is a block diagram showing a processing flow of the plant performance monitoring system according to the embodiment of [claim 3] of the present invention.

FIG. 10 is a configuration diagram of a plant performance monitoring system according to an embodiment of [claim 4] of the present invention.

FIG. 11 is a block diagram showing a processing flow of a plant performance monitoring system according to an embodiment of [claim 4] of the present invention.

FIG. 12 is a block diagram showing a processing flow of a plant performance monitoring system according to an embodiment of [claim 4] of the present invention.

FIG. 13 is a configuration diagram of a plant performance monitoring system according to an embodiment of [claim 5] of the present invention.

FIG. 14 is a block diagram showing a processing flow of the plant performance monitoring system according to the embodiment of [claim 5] of the present invention.

FIG. 15 is a block diagram showing a process flow of the plant performance monitoring system according to the embodiment of [claim 5] of the present invention.

FIG. 16 is a configuration diagram showing a related art of [claim 1] of the present invention.

FIG. 17 is a configuration diagram showing a related art of [claim 2] and [claim 3] of the present invention.

FIG. 18 is a configuration diagram showing a related art of [claim 4] of the present invention.

FIG. 19 is a configuration diagram showing a related art of [claim 5] of the present invention.

[Explanation of symbols]

 A, B computers 1 A system abnormality detection unit 2 A system state transmission unit 3 B system abnormality detection unit 4 B system state transmission unit 5 B system state reception unit 6 B system state determination unit 7 A system state reception unit 8 A system state Judgment unit 9 B-system backup processing unit 10 A-system backup processing unit 11 Business A status detection unit 12 Business B status detection unit 13 Shared memory 14 Shared magnetic disk devices 101, 121 Business recovery request devices 102, 122 Business stop request device 103, 106, 123, 126 business stop request flag 104, 113, 127, 134 business B 105, 108, 125, 128 business task status table 107, 114, 124, 133 business A 111 monitoring device 112, 132 business start request device 121 business Recovery request device 201, 301 Core 202, 302 Plant 204, 304 Data receiving unit 206, 306 Core state calculation unit 213, 311 Core state data magnetic tape input / output unit 217, 218, 309 Core state data storage unit 313 Computer synchronization management unit 314 total Machine restart calculator

────────────────────────────────────────────────── ─── Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI G06F 11/34 G06F 11/34 P // G05B 15/02 G05B 15/02 Z (72) Inventor Masahiro Fukudome Toshiba, Fuchu-shi, Tokyo No. 1, Toshiba Corporation Fuchu Plant (56) References JP-A-2-287796 (JP, A) JP-A-55-127652 (JP, A) JP-A-61-228535 (JP, A) (58 ) Fields surveyed (Int.Cl. ⁷ , DB name) G08B 31/00 G05B 7/02 G05B 9/03 G05B 23/02 301 G05B 23/02 302 G06F 11/34 G05B 15/02

Claims (6)

(57) [Claims] 1. A plant performance monitoring system that uses a plurality of computers to calculate plant performance using data dependent on the operation history of a plant, first detection means for detecting an abnormality of a partner computer, and a second detecting means for detecting the operating state of the business X running on the mating computer, the operation of the first detecting means, and the operation stops operation of the mating computer by said second detecting means Logical conjunction of
Matter, the plant performance monitoring system, characterized in that to take over the data file that functions and operations X is used in the business X automatically on their side computer. 2. In a plant performance monitoring system in which a plurality of computers mutually monitor the operating state of a partner computer and perform backup processing by a healthy computer when an abnormality occurs in one computer, an abnormality in a computer in one system each system state determining unit to read the status, the abnormal state of each system of the business X together and a shared memory for each storing therein, to determine an abnormality of the abnormality and business X of the calculation <br/> machine below A plant performance monitoring system, which carries out an output process. (A) When the computer is abnormal and the operation X is abnormal , the backup processing of the operation X is performed on the other computer. (B) When the computer X is normal due to a computer error, an alarm process is performed to the effect that the error detection unit is faulty. (C) When the computer is normal and the operation X is abnormal , an alarm process indicating that an abnormality has occurred in the operation X is performed. 3. A computer system in which two computers share a task X and a task Y and process each of them, and when one computer fails, the other computer processes the tasks X and Y together. When one of the computers recovers from the failure, the business X and the business Y that were processed by the other healthy computer
Out of which, a means for issuing a stop request to the business X that has been newly shared and processed in response to a failure, and transferring the processing of the business X requested to be stopped to the restored computer Plant performance monitoring system. 4. The plant performance monitoring system according to claim 3, further comprising means for arbitrarily designating a task to be transferred to said restored computer. 5. In a plant performance monitoring system that uses a plurality of computers to calculate plant performance using data dependent on the operation history of a plant, two or more sets of operation history dependent data storage areas used for performance calculation are provided. together comprising, an identifier that indicates the update time of the identifier and the data indicating that the data of each region is healthy each comprising, before
When starting the performance calculation, a data storage area that is sound and has the latest data update time is selected from the operation history dependent data, and data necessary for performance calculation is input from this area. Plant performance monitoring system. 6. A plant performance monitoring system that uses a plurality of computers to calculate plant performance using data dependent on the operation history of a plant, wherein the data is stored in a data storage area selected to start performance calculation. and plant state are plant input from the data input unit is compared with the current plant state, said one of the two plant state, the computer restarts computation difference key parameters to determine whether within a predetermined range The main parameters
A performance calculation system for automatically restarting the performance calculation when the difference between the parameters is within a predetermined range.