A nuclear power plant tracking and prediction system control middleware implementation system and method

Abstract

The present invention relates to the field of nuclear safety and nuclear emergency technology, and specifically to a nuclear power plant tracking and prediction system control middleware implementation system and method. In order to ensure the scientific nature of the nuclear emergency decision-making process, it is necessary to develop nuclear power plant tracking and prediction software to support scientific analysis of the occurrence and development process of accidents, and provide feasible accident mitigation measures to minimize the harm of accidents to the public. The nuclear power plant tracking and prediction system consists of two state tracking calculation modules, four identical prediction modules and a control middleware based on a real-time memory database. The implementation method includes selecting an operating mode; the tracking module performs tracking calculations; starting the prediction module; the prediction module performs prediction calculations; and user intervention. The present invention is a nuclear power plant tracking and prediction software that can be used to provide technical support and guidance for nuclear emergency decision-making and accident disposal.

Description

Technical Field

[0001] The present invention relates to the field of nuclear safety and nuclear emergency technology, and in particular to a system and method for realizing control middleware of a tracking and prediction system of a nuclear power plant. Background Art

[0002] Emergency preparedness and response for nuclear power plant accidents is the last barrier of its nuclear safety defense in depth. Correct nuclear accident mitigation measures can reduce accident risks, mitigate accident consequences, avoid or reduce the radiation risks that the public may be exposed to, and protect the health and safety of the surrounding public. Inappropriate nuclear accident emergency mitigation measures will also bring additional risks to the radiation safety of the public. Therefore, in order to ensure the scientific nature of the nuclear emergency decision-making process, it is necessary to develop nuclear power plant tracking and prediction software to support scientific analysis of the occurrence and development process of accidents, and provide feasible accident mitigation measures to minimize the harm of accidents to the public. Summary of the invention

[0003] 1. Purpose:

[0004] The purpose of the present invention is to provide a method for implementing a nuclear power plant tracking and prediction system control middleware based on a real-time database. The nuclear power plant tracking and prediction software implemented based on this method can be used to provide technical support and guidance for nuclear emergency decision-making and accident handling.

[0005] 2. Technical solution:

[0006] The nuclear power plant tracking and prediction system consists of two state tracking calculation modules (tracker), four identical prediction modules (predictor) and a management and control middleware based on a real-time memory database. The tracking calculation module and the prediction module are based on the internationally common severe accident analysis software MAAP5 as the calculation kernel.

[0007] The input data of this system can come from four external systems: real-time data of Unit 1, real-time data of Unit 2, full-range simulator of power plant, and severe accident simulator of power plant. 3D GUI and WEB GUI are user-exchange human-machine interfaces, enabling users to control the system and view display parameters through pictures and curves. The control middleware based on real-time memory database is the lowest-level service program of the entire system, which stipulates the workflow and coordination methods between various software functions, stipulates control commands and command formats, and manages system data in a unified manner.

[0008] The technical solution specifically includes:

[0009] A nuclear power plant tracking and prediction system management and control middleware implementation system includes a tracking and calculation module, a prediction module, a management and control middleware, a user interface and an external data source; bidirectional data exchange is performed between the management and control middleware and the tracking and calculation module and the prediction module; and the external data source inputs data to the management and control middleware.

[0010] The prediction modules are four identical modules, the tracking and calculation modules are two identical modules, and the kernel of each prediction module and tracking and calculation module is severe accident analysis software.

[0011] The external data sources include real-time data of the unit, a full-range simulator and a severe accident simulator.

[0012] The user interface includes a web user interface and a 3D user interface. The web user interface and the management and control middleware perform bidirectional data exchange, and the management and control middleware outputs data to the 3D user interface.

[0013] A method for implementing a nuclear power plant tracking and prediction system control middleware comprises the following steps: step 1: selecting an operation mode; step 2: a tracking module performs tracking calculations; step 3: starting a prediction module; step 4: the prediction module performs prediction calculations; and step 5: user intervention.

[0014] The step 1: selecting the operation mode specifically includes: (1) selecting the number of units; (2) determining the ID of the start tracking calculation module; and (3) determining the number of associated prediction modules and the prediction module ID.

[0015] The second step is that the tracking module performs tracking calculations, specifically including: (1) the control middleware interface issues a start command; (2) the control middleware starts the tracking calculation module, and the tracking calculation module starts the severe accident analysis software, cyclically obtains input from external data for calculation; (3) the tracking calculation module subscribes to the control commands from the real-time data area of ​​the control middleware, and regularly publishes the calculation results of the severe accident analysis software.

[0016] The step three: starting the prediction module, specifically includes: (1) after the tracking is successful, the user issues a command to start the prediction module in the web user interface and publishes it through the management and control middleware; (2) after the tracking calculation module receives the command to start the prediction module, it will generate a corresponding initialization file according to the settings.

[0017] The fourth step: the tracking calculation module performs predictive calculation, specifically including: (1) the control middleware starts the prediction module, and the prediction module obtains the initialization file generated by the tracking calculation module; (2) the severe accident analysis software is started and the calculation is performed in a loop; (3) the prediction module subscribes to the intervention data and control commands corresponding to the real-time data area of ​​the control middleware, and publishes the calculation results of each severe accident analysis software at a regular interval.

[0018] The step five: user intervention, specifically includes: (1) the user selects different prediction modules on the web user interface, inputs the intervention operation, and publishes it through the management and control middleware; (2) after the prediction module obtains the intervention data, it transmits the intervention data to the severe accident analysis software, thereby affecting the calculation effect.

[0019] 3. Effect:

[0020] The management and control middleware aggregates the software of each node into a system, so that the commands of the agreed format flow between the nodes and the data are shared between the nodes, so that each software can be executed according to the established workflow, thereby achieving the tracking and prediction effect. Through the WEB GUI, select the external data source as the input of the tracking calculation module, and through the MAAP calculation, track the accident status and identify the accident type. And give the five groups of initial running states required by the prediction analysis module: optimal, optimistic, sub-optimistic, conservative, and over-conservative. Predictor starts different MAAP instances with the initial state as input for parallel fast calculation. Perform different operations on each Predictor on the WEB GUI to affect the calculation logic of the MAAP instance. Through curve comparison and dynamic effect display, the effects of various intervention measures are analyzed, so that feasible accident mitigation measures can be provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a software structure diagram

[0022] Figure 2 Software collaborative workflow diagram

[0023] Figure 3 It is a parameter curve comparison chart DETAILED DESCRIPTION

[0024] The workflow and control relationship of each software can be found in Figure 2 .

[0025] When the user logs in to the WEB GUI, he first chooses to enter the drill mode or the accident mode.

[0026] Further determine whether it is a single unit or a dual unit. If it is a single unit, only one tracking calculation module will be started.

[0027] The ID of the started tracking and calculating module is further determined, and the number of prediction modules and the prediction module IDs associated with the tracking and calculating module are determined.

[0028] Once the above settings are confirmed, a start command is issued through the interface provided by the management and control middleware.

[0029] After receiving the command, the control middleware will start the tracking calculation module. The tracking calculation module will start the MAAP instance, loop to obtain input from external data, perform calculations, and track the source of data. The tracking calculation module will subscribe to the control commands from the real-time data area of ​​the control middleware and publish the calculation results of MAAP at a regular interval.

[0030] Furthermore, after successful tracking, the user issues a command to start the prediction module in the WEB GUI and publishes it through the management and control middleware.

[0031] After the tracking calculation module receives the command to start the prediction module, it will generate the corresponding initialization file according to the settings.

[0032] Furthermore, the control middleware starts the prediction module according to the settings. The prediction module will obtain the initialization file generated by the tracking calculation module according to the settings, start the MAAP instance, and perform calculations in a loop. The prediction module will subscribe to the intervention data and control commands corresponding to the real-time data area of ​​the control middleware, and publish the calculation results of each MAAP at a regular interval.

[0033] Furthermore, the user selects different prediction modules on the WEB GUI, enters the corresponding intervention operations, and publishes them through the management and control middleware. After the prediction module obtains them, it passes the intervention data to MAAP, thereby affecting the calculation effect.

[0034] The WEB GUI obtains the calculation structure of each MAAP regularly through the interface provided by the timing control middleware and displays it on the screen. Users can observe the operation curves of different parameters on the screen. Under different intervention measures, the MAAP calculation curves of different estimation input files can generate up to 15 curves for the same parameter. Through comprehensive information comparison, it can provide technical reference for the emergency control center experts to select appropriate severe accident management strategies.

[0035] In order to realize the above process and control, the management and control middleware memory database is designed as follows:

[0036]

[0037]

[0038]

[0039]

[0040] The delivery of command and control adopts publish / subscribe mechanism.

[0041] Issue the following command from the WEB GUI:

[0042] 1) Set the external system from which the data source of the specified tracking calculation module comes;

[0043] For example: publish cmd_track_calc_module1 1@1@5@Unit1

[0044] Meaning: Set the data source of tracking calculation module 1 to come from unit 1 and start tracking calculation module 1

[0045] 2) Start, run, pause, and stop each Predictor managed by the tracking calculation module;

[0046] For example: publish cmd_tracking_module1 1@1@5.04.001_3

[0047] Meaning: Start tracking the third MAAP of Predictor1 managed by calculation module 1

[0048] For example: publish cmd_tracking_module1 1@1@5.04.002_*

[0049] Meaning: Start tracking all MAAPs of Predictor2 managed by calculation module 1

[0050] For example: publish cmd_tracking_module2 1@4@5.04.003_*

[0051] Meaning: Stop all MAAPs of Predictor3 managed by tracke2

[0052] The tracking calculation module subscribes to the above commands. When receiving the commands, it controls the MAAP5 calculation instance used for tracking calculation and the MAAP5 calculation instance used for prediction to start, run, pause, stop according to the commands, and feedback the execution status.

[0053] The WEB GUI intervenes in the selected Predictor operation through the following commands:

[0054] For example: publish setdata_p1 "gui_INSGFW,0.56"

[0055] Predictor subscribes to the above commands. When receiving the commands, it intervenes in the calculation logic of the managed MAAP5 by setting variable data, thereby affecting the calculation effect.

[0056] The WEB GUI periodically calls the data acquisition command according to the selected MAAP instance of the Predictor. The WEB GUI subscribes to the real-time operation status of the Predictor and tracks the calculation module and displays it on the page.

[0057] WEB GUI gets data according to the selected Predictor and displays it on the WEB screen or curve, for example: hgetall Data_5.04.0001_1

[0058] The 3D GUI periodically calls the get data command according to the selected tracking calculation module, and the obtained data is displayed on the 3D screen, for example: hgetall Data_tracking calculation module 1.

[0059] After testing the selected cases, it was found that asynchronous computing can be 10 times faster than synchronous computing, which enables each MAAP5 process to calculate at the fastest speed allowed by the hardware, reach the set prediction time as early as possible, and display the prediction results faster. Therefore, the MAAP5 operation instances started by the Predictor of this system run asynchronously in parallel.

Claims

1. A nuclear power plant tracking and prediction system management and control middleware implementation system, characterized by: It includes tracking and calculation module, prediction module, management and control middleware, user interface and external data source; bidirectional data exchange is carried out between the management and control middleware and the tracking and calculation module and prediction module; the external data source inputs data to the management and control middleware; The prediction modules are four identical modules, the tracking and calculation modules are two identical modules, and the kernel of each prediction module and tracking and calculation module is severe accident analysis software; The external data sources include real-time data of the unit, a full-range simulator and a severe accident simulator; The user interface includes a web user interface and a 3D user interface. The web user interface and the management and control middleware perform bidirectional data exchange, and the management and control middleware outputs data to the 3D user interface.

2. A method for implementing a control middleware for a nuclear power plant tracking and prediction system, characterized in that: The method comprises the following steps: step 1: selecting an operation mode; step 2: the tracking module performs tracking calculation; step 3: starting the prediction module; step 4: the prediction module performs prediction calculation; step 5: user intervention; The step 1: selecting the operation mode specifically includes: (1) selecting the number of units; (2) determining the ID of the start tracking calculation module; (3) determining the number of associated prediction modules and the prediction module ID; The second step: the tracking module performs tracking calculation, specifically including: (1) the control middleware interface sends a start command; (2) the control middleware starts the tracking calculation module, and the tracking calculation module starts the severe accident analysis software, cyclically obtains input from external data for calculation; (3) the tracking calculation module subscribes to the control command from the real-time data area of ​​the control middleware, and regularly publishes the calculation results of the severe accident analysis software; The step three: starting the prediction module, specifically includes: (1) after the tracking is successful, the user issues a command to start the prediction module in the web user interface and publishes it through the management and control middleware; (2) after the tracking calculation module receives the command to start the prediction module, it will generate a corresponding initialization file according to the settings; The fourth step: the tracking calculation module performs prediction calculation, specifically including: (1) the control middleware starts the prediction module, and the prediction module obtains the initialization file generated by the tracking calculation module; (2) the severe accident analysis software is started and the calculation is performed cyclically; (3) the prediction module subscribes to the intervention data and control commands corresponding to the real-time data area of ​​the control middleware, and publishes the calculation results of each severe accident analysis software at a regular interval; The step five: user intervention, specifically includes: (1) the user selects different prediction modules on the web user interface, inputs the intervention operation, and publishes it through the management and control middleware; (2) after the prediction module obtains the intervention data, it passes the intervention data to the severe accident analysis software, thereby affecting the calculation effect.

Images