KR101702912B1 - System for the Virtually Integrated Execution of the Engineering Analysis Codes using the Process Data Section - Google Patents

Abstract

The present invention relates to a virtually integrated driving system for engineering analysis codes using a process data section, and more specifically, to a technology in which when an operation of a total system is analyzed by integrating engineering programs for analysis verified in the analyzing performance through a permission or license, each of the programs directly accesses a data (or Heap) section in an individual space to change a value and dynamic-change time synchronization binary codes of each of the programs during an execution, so that a time synchronization is secured. According to the virtually integrated driving system for engineering analysis codes using a process data section, executing the time synchronization is integrally secured without damaging an original algorithm of the verified program, thereby outputting a result having high reliability.

Description

[0001] The present invention relates to a method and a system for analyzing a code,

The present invention relates to a virtual integrated analysis system for analytical codes using a process data section, and more particularly, to a method for analyzing a behavior of a comprehensive system by integrating analytical engineering programs that have been verified for analysis performance through licensing and the like , A technique of directly accessing the data (or Heap) section of each program in each independent memory space, changing the value, and dynamically changing the time synchronization binary code of each program during execution to ensure time synchronization, And more particularly, to a scientific analysis code virtual integrated drive system using a process data section that can achieve a high reliability result by ensuring time synchronization execution without compromising an algorithm.

When integrating existing codes developed for verification purposes of a specific field, such as safety analysis codes used in plant system design / safety analysis in the past, for use in analyzing the behavior of the integrated system as a whole, A method of merging program source codes of existing codes into one large code or a method of exchanging values of connection variables among a plurality of codes in the form of calling a function using a dynamic link library (DLL) method or a static library (LIB) method Or a more advanced method, extracts all the internal variables of each program and relocates it to one huge shared memory.

In the method of integrating the program source integration method or the DLL / LIB method, the type and kind of the linkage variables between programs are predefined in advance and the value is transferred as a function parameter. As the number of programs to be integrated increases And the linkage between the programs increases.

Therefore, the type of function to be transferred and the number of calls increase exponentially, and since the change of the source code is necessary as the connection variable is changed, the process of compiling by changing the source must be performed, There was a drawback that it lengthened.

In the case of a method of reorganizing the internal variables of each program using a shared memory, the storage space of the variables must be relocated to an external shared memory space rather than a method defined by the compiler. Therefore, There is a burden of having to redefine all the parts.

In the case where the programs for stability analysis of the safety related system, such as nuclear energy, are subjected to the license and the safety analysis is performed within the designated area, And the reliability of the result is seriously deteriorated.

For reference, codes used for design and safety verification of nuclear power plants should be submitted to the Nuclear Safety Commission over a period of 2 ~ 3 years for submission of a specific topic technical report of the code, To obtain a license.

The cost of this would be from billions to tens of billions of dollars.

Therefore, in order to analyze (or simulate) the behavior of the entire system by integrating and executing a plurality of system analysis programs, it is possible to analyze the behavior of the individual systems by changing the storage locations of the internal variables used by the respective programs (Or Heap) section in the data section, which is the storage space of each program's internal variables, rather than changing the source or reconfiguring the program sources into a huge program. (Memory-map) file information to read and write the values of corresponding internal variables without changing the program source, thereby providing an advantage of integrating and running a plurality of system analysis programs .

After each analysis program is loaded on a memory (RAM) for time-synchronization of a plurality of programs, an event synchronization object is generated and waits in a main execution loop (MEL) interval of each program. Code is inserted dynamically to change the binary code of the MEL section and then executed, thereby providing an advantage of time synchronization.

(Patent Document) Korean Patent Laid-Open No. 10-2007-0003578

SUMMARY OF THE INVENTION Accordingly, the present invention has been made keeping in mind the above problems occurring in the prior art,

A first object of the present invention is to provide a method and system for analyzing a plurality of analysis programs in a different process space by using memory map file information on a memory address space of internal variables used by analysis programs, Remote Permeation to the internal memory space to provide the boundary condition values of the boundary-condition variables needed by each analysis program so that the analysis programs can be executed in the same way as they execute alone have.

A second object of the present invention is to provide a method and system for generating an event synchronization object before and after a MEL section of analysis programs so that each analysis program can be driven at the same time in terms of time (Problem Time) To enable execution of all running analytical programs in a time-synchronized manner so as to perform a virtual integration for the Engineering Codes.

According to an aspect of the present invention, there is provided an analytical code virtual integrated drive system using a process data section according to an embodiment of the present invention,

A PVDB unit 3 that builds information on internal variables of analysis codes based on B-TREE,

A DB builder node unit 2 including a PIDB unit 4 constructed by building a structure including information of a process name, a handle value, and a dynamic memory start address value of an analysis code process on a simple loading scheme,

A data manager node unit (1) including a DB manager unit (5) for reading a current value of a designated variable and recording a specific value as a current value of the variable;

A process initializing node unit 8 for performing an initializing operation for integrated driving of analysis programs,

A boundary condition manager node unit 9 for filling the boundary condition values necessary for the analysis programs to operate,

And a real-time synchronization node unit (10) for performing time synchronization of programs for analysis and being able to be driven;

An engineering code execution node unit 11 for releasing the waiting state of analysis programs and simultaneously executing each MEL section of each analysis program to execute time synchronization when a time synchronization setting signal is generated in the execution controller node unit 7; ;

An execution UI unit 23 for starting, executing, and stopping the execution controller node unit 7;

And a user interface node unit 13 including a data UI unit 24 for acquiring and setting values of internal global variables used by engineering analysis programs through the execution controller node unit 7 .

The analytical code virtual integrated drive system using the process data section according to the present invention is a virtual integrated drive system that uses memory map file information for the memory address space of the internal variables used by the analysis programs without changing the source of the engineering analysis programs to be integrated By remotely penetrating the internal memory space of each analysis program in another process space and providing the boundary condition values of the boundary-condition variables required by each analysis program, analysis programs The effect of being executed in the same manner as when it is executed alone can be obtained.

In addition, an event synchronization object is generated before and after the MEL section of the analysis programs so that the analysis programs can be driven at the same time (Problem Time) of each analysis program, and the insertion code is dynamically inserted and changed And thus all of the running analysis programs can perform the virtual integration for the engineering processes so that the time synchronous execution can be performed.

In other words, when analyzing (or simulating) the behavior of the entire system by integrating a plurality of programs for system analysis, it is possible to change individual program sources to change the storage locations of the internal variables used by the respective programs, An external integrated control program is stored in the internal virtual space for analysis using the memory map file information containing the offset information in the data section which is the storage space of the internal variables of the respective programs, By reading and writing the variable values of the variables, it is possible to overcome many disadvantages such as the generation of a large amount of linked transfer functions used in the existing integration method, the inefficiency due to the increase of the number of linked function calls, and the change of numerous source codes for the linked variables , So that the integration period can be greatly shortened do.

In addition, it provides a system that integrates analytical programs with time synchronization without changing the source code of analysis programs. Therefore, it is possible to apply the licensed algorithms and scope of safety analysis codes as they are, For example, to make significant progress in nuclear safety analysis.

In the case of changing the safety analysis code that has been licensed as in the existing integration method, the long-term review period and the high cost according to the re-licensing have necessarily been accompanied. However, in the system according to the present invention, There is a big advantage that it does not.

Even if each analysis program is version-up, if the conventional integration method is used, it is required to undergo a lot of trial and error and a long integration process. However, according to the method of the present invention, , And time synchronization is ensured. Therefore, it is advantageous to relatively simplify the integration of each analysis program into the maintenance and version-up code.

In the conventional method of integrating sources, it is necessary to disclose one linking variable by specifying a linking variable in a program source dimension when a graphical linking program is developed. However, in the method according to the present invention, The present invention provides many convenience in developing a link program for other purposes such as a graphic display program because there is a function of inserting / deleting / inquiring / retrieving the global variables in the analysis program in a standardized manner.

FIG. 1 is an overall configuration diagram of a scientific analysis code virtual integrated drive system using a process data section according to an embodiment of the present invention, FIG. 2 is an example of a virtual memory space of an Engineering Code, FIG. 3 is a diagram showing the relationship between the memory map file and the analysis program virtual address space, FIG. 4 is a diagram showing the relationship between the PVDB unit and the PIDB unit, FIG. FIG. 7 is a flow chart of the MEL binary code change of the analysis program, FIG. 8 is a flowchart of the dynamic change analysis program MEL, and FIG. 10 is an RUN command flow chart of the boundary condition manager node unit of the execution controller node unit, The multiple parts of the flow chart node boundary conditions manager node portion FREEZE command, Figure 12 is a flow chart DB GET command manager portion, 13 is a flow chart DB SET command manager portion, 14 is a file process.ini implemented.

Hereinafter, embodiments of the analytical code virtual integrated drive system using the process data section according to the present invention will be described in detail.

1 is an overall configuration diagram of a scientific analysis code virtual integrated drive system using a process data section according to an embodiment of the present invention.

As shown in FIG. 1, the analytical code virtual integrated drive system of the present invention mainly includes a data controller node unit 1, an execution controller node unit 7, an engineering code execution node unit 11, (13).

The integration method described in the present invention is a method of exchanging variable values between analysis programs 12, 202, and 806 and a method of performing time synchronization between analysis programs 12, 202, and 806.

A method of exchanging variable values between the analysis programs 12, 202, and 806 will be described.

Typically, the variables used internally by the executable program are local variables, global variables, and dynamically allocated heap variables (Heap Variables).

The local variables consist mainly of stack variables that are used locally only in the function. They are used as variables that perform the role only as an intermediary medium for calculating the property to be analyzed, and are variables that are not worth connecting with other systems.

The parameters that are continuously survived in the execution program in the execution program are global variables (or Heap variables), and geometric information of the structures required for the execution of the analysis program, property values of constituent structures, It is stored in the global variable (or Heap variable) memory space.

For example, if an engineering analysis program is intended to simulate and analyze only specific systems, the property at the boundary is typically input by the input deck prior to execution of the analysis program.

Therefore, as the incident progression time progresses, the values of the property values at the boundary conditions are constant or simply calculated and input by predetermined rules.

However, when an integrated analysis of a comprehensive plant consisting of a number of systems such as a nuclear power plant is to be integrated, a number of systems are divided into a plurality of engineering analysis programs, and each program is synchronized in time, The properties must be dynamically filled by other programs.

At this time, there must be a way to transfer the values of the internal variables calculated by the analysis programs so that they can be used by other analysis programs.

In the present invention, the method of directly reading or writing the independent memory space of another program is used.

The global variable (or the Heap variable) is stored in the memory-map file 201,301 generated by the compiler when compiling the program source of the engineering analysis programs 12, 202, and 806, Information is provided in the form of an Offset address 204 under the data (or Heap) section 101, 102, 103, 203 in the program virtual space 202.

Therefore, when the analysis programs 12, 202, and 806 are loaded on the RAM, the start memory addresses 104 and 105 of these data (or Heap) sections 101, 102, And an offset address 204 of the internal variables provided by the memory map file are added to the virtual memory space 202 of the analysis programs 12, The value of the address (Logical Address) can be known.

pSrc = tagsvrDATA :: pProcessInfo-> pDataSectionAddr + tagsvrDATA :: src_var_info-> pOffset;

In the case of Table 1, the logical address calculation example code is shown.

As shown in FIGS. 12 to 13, by using the sample codes 1108, 1110, 1208, and 1210, the analysis programs 12, 202, The virtual memory space 202 of the virtual memory space 806 can be remotely accessed on the memory space of the other processes 22 so that a desired value can be written to or read from a specific address on an independent virtual memory space Do.

hProcess = OpenProcess (PROCESS_ALL_ACCESS, FALSE, dwPID);
WriteProcessMemory (hProcess, pRemoteBuf [0], (LPVOID) & param, dwSize, NULL);
(...)
ReadProcessMemory (hProcess, pRemoteBuf [3], (LPVOID) & param, dwSize, NULL);

Table 2 above shows an example code of a read / write function as an independent virtual memory space.

In the present invention, the memory map files 201, 202 defining the offset address value 204 of the global (or Heap) variables used by the analysis programs 12, 202, (TagsvrDATA structure 307) including various information of each variable (variable name, belonging process information, offset address value in the data section, etc.) based on the B-Tree method 311, 1113, 1213 (PIDB, 304) in the form of a structure (PROCESS_INFO structure, 309), and the analysis programs (12, 202, and 806, respectively, to be provided to each other.

3 is a diagram showing the relationship between the virtual address space 202 of the engineering analysis programs 12 and 202 and the memory map files 201 and 301. [

The address 204 indicates the offset address value 204 in the data (or Heap) section 101, 102, 103, 203 of the corresponding global variable in the relationship diagram shown in Fig.

Next, referring to FIG. 9, a description will be made of a method of performing time synchronization between the analysis programs 12, 202, and 806 described above.

In order to comprehensively analyze a system composed of complex systems such as a power plant, it is necessary to divide the system into systems to be analyzed by the engineering analysis programs 12, 202, and 806, to perform time synchronization of the engineering analysis programs, Should be.

For this purpose, in the present invention, binary execution codes 603 and 604 for synchronization are dynamically injected into the main execution loops (MEL, reference numerals 503, 602 and 702) of the respective engineering analysis programs 601 and 701 605) and causes synchronization signals 21, 805 to be generated in an external program (e.g., the execution controller node unit 7, 801) to ensure time synchronization between analysis programs 12, 202 have.

For example, as shown in reference numeral 501 in FIG. 6, all the engineering analysis programs include a process introduction unit 502, a process main execution loop unit 503, 602, 702, a process ending unit 504). ≪ / RTI >

In the process introduction unit 502, the geometric position information of the target system structure to be analyzed and the physical property values of the internal constituent materials are initialized in the form of an input deck, and the physical property values of the boundary systems are also input as initial values And initialization of internal global variables such as

The process main execution loop units 503, 602 and 702 denote the main execution loop of the process and abbreviated as "MEL ".

In the process main execution loop unit, the time (Advancement) 804 is advanced until a predetermined analysis end time, and the physical property values for all the physical spaces are numerically solved to perform repetitive calculation Corresponds to an execution loop.

The process ending unit 504 mainly performs printing of an execution result, such as recording the results of major variables in a file, for analyzing the execution result of the analysis program.

Of the three intervals, the MEL interval (the interval 705 to 706 in FIG. 8, the interval 807 to 808 in FIG. 9) of each of the analysis programs repeatedly executed according to the time-step 803 is the most important interval , It is impossible to synchronize the analysis programs 12, 202, and 806 with each other, if the MEL sections are not synchronized with each other in time, It becomes meaningless.

In other words, it should be ensured that analysis programs are analyzing at the same time, but it is meaningful to analyze the overall behavior of the analyzing plant.

In order to ensure time synchronization between the analysis programs 12, 202 and 806, in the present invention, a time synchronization event object (hereinafter referred to as a " time synchronization event object " (603), and inserting (605) a logic to wait (Set, 21,805) the event object (Wait, 604). Then, the MEL sections of the analysis programs (12, 202, 806) Thereby ensuring time synchronization (see FIG. 9).

For example, an external program independent of the analysis programs, such as the execution controller node unit 7, 801, may provide event synchronization signals 21, 805 so that the analysis programs 12, 202, (See FIGS. 8 and 9 and 3) and 707, so that the next MEL section can be executed at the same time, thereby achieving time synchronization.

9, an external control program for the event object (for example, the execution controller node unit 7, 801) sets an event synchronization object set signal 21, 805 for every? T period 803 to "PulseEvent ) Function, the analysis programs 12, 202, and 806 in the signal waiting state (refer to FIGS. 8 and 9 and the reference numeral 707) are provided for each MEL section (reference numerals 705 to 706, (809 to 808 intervals) are simultaneously executed and returned to the position of (3) (809) and are again in a standby state, thereby schematically illustrating a method of time synchronization of the analysis programs.

The following description will be made with reference to FIG. 1, which shows a virtual integrated analysis system for analysis codes using a process data section according to the present invention.

The analytical code virtual integrated drive system using the process data section comprises:

A PVDB unit 3 that builds information on internal variables of analysis codes based on B-TREE,

A DB builder node unit 2 including a PIDB unit 4 constructed by building a structure including information of a process name, a handle value, and a dynamic memory start address value of an analysis code process on a simple loading scheme,

A data manager node unit (1) including a DB manager unit (5) for reading a current value of a designated variable and recording a specific value as a current value of the variable;

A process initializing node unit 8 for performing an initializing operation for integrated driving of analysis programs,

A boundary condition manager node unit 9 for filling the boundary condition values necessary for the analysis programs to operate,

And a real-time synchronization node unit (10) for performing time synchronization of programs for analysis and being able to be driven;

An engineering code execution node unit 11 for releasing the waiting state of analysis programs and simultaneously executing each MEL section of each analysis program to execute time synchronization when a time synchronization setting signal is generated in the execution controller node unit 7; ;

An execution UI unit 23 for starting, executing, and stopping the execution controller node unit 7;

And a user interface node unit (13) including a data UI unit (24) for obtaining and setting values of internal global variables used by engineering analysis programs through the execution controller node unit (7) .

The data controller node unit 1 includes a DB builder node unit 2, 302, 1112, and 1212 and a DB manager unit 5.

At this time, the DB builder node unit (2, 302, 1112, 1212)

A PVDB unit 3, 303 that builds information 307 on internal variables of analysis codes based on B-TREE bases 311, 1113, 1213,

And a PIDB unit 4, 304 in which a structure 309 including the process name of the analysis code process, the handle value, and the dynamic memory start address value of the data section is constructed on the basis of a simple stacking method.

The PVDB is an abbreviation of Process Variable Data Base, and the PIDB is an abbreviation of Process Information Data Base.

The PVDB unit 3 or 303 stores the memory offset values of the analysis programs 12, 202 and 806 in the data sections 101, 102, 103 and 203 of the analysis program, (Variable name, offset value, belonging analysis program handle value, etc.) of individual global variables based on the memory map files 201 and 301 of the respective analysis execution target analysis programs representing the value information 204. The tagsvrDATA 307 ) Objects into a database based on the B-Tree (311, 1113, 1213).

The description of each structure constituting the B-Tree is shown in Tables 3 to 7, and the relationship among the structures is illustrated in FIG.

Field name meaning src_var_info Saving individual information about the internal global variables of the analysis program
(see Table 4 for a description of each field in the src_var_info object) pProcessInfo The global variable represented by the src_var_info object stores information about the original analysis program used (see Table 8 for a description of each field in the PROCESS_INFO structure) nTargetVarInfo If the global variable is used as an input value of another analysis program, the destination variable number to be provided as an input value target_var_info In the example of FIG. 14, a variable called src_var2 is a variable originally used in the analysis code of Engineering_Code_1, and a value of the variable src_var2 is a value of the variable of the analysis code of Engineering_Code_2 the value of nTargetVarInfo is set to 4, and the target_var_info field indicates the TARGET_VAR_INFO structure object for the variables tgt_var3, tgt_var4, tgt_var5, and tgt_var6, assuming that the tgt_var3 and tgt_var4 variables and Engineering_Code_3 analysis code should be provided as boundary condition values in tgt_var5 and tgt_var6. The value is specified as a value.

Table 3 above is a field description of tagsvrDATA structure (307).

Field name meaning name Global variable name Description Description string for global variable Units Engineering unit strings (eg, MPa, kg / s, etc.) Type Variable declaration type of global variable (eg char, bool, short, int, float, double, etc.) nDims Dimension declared in global variable Dims If the declared limit of each difference in the declared order (eg aa [2] [3], the first is 2, the second is 3). pOffset The offset address value of each global variable declared in the memory map file

Table 4 above shows the SRC_VAR_INFO structure 308 field description.

Field name meaning TargetVarName a variable name to receive a value from the src_var_info object 313 pData The tagvrDATA object address value with the TargetVarName variable. pAddr TargetVarName The memory address in the virtual address space in the analysis program to which the variable is assigned.
If TargetVarName is a variable name with order like tgt_var [2] [3], pData is the tagvrDATA object address that represents tgt_var variable, but tgt_var [2] [3] is not allocated to tgt_var address The address of tgt_var [2] [3] (not the address of tgtvar) should be indicated in pAddr.

Table 5 above is a description of the TARGET_VAR_INFO structure 310 field.

[NODE structure description]
- If the internal global variable object is database based on B-tree, it is used internally by B-tree implementation algorithm.
A structure that represents a bundle of ENTRY structures consisting of two pairs.

Table 6 above describes the NODE structure 305.

[ENTRY structure description]
- means an entry object constituting the NODE structure, and one unique ENTRY structure is allocated to each variable used in the analysis code. In particular, a data field value composed of fields represents an object corresponding to an internal variable 1: 1.

Table 7 above describes the ENTRY structure 306.

In addition, the PIDB unit (4, 304) constructs a structure (309) based on a simple loading method, including a process name, a handle value of the analysis code process, and information on a dynamic memory start address value of a data section.

That is, a database of a simple loading method 312 for acquiring an engineering analysis program list 1302 to be executed and storing the PROCESS_INFO object 309 corresponding to the analysis program in order in the database memory is constructed.

At this time, this database is called Process Information Data Base (PIDB) (4, 304) and stores the following information (PROCESS_INFO object 309 information) for each engineering analysis program.

Field name meaning ProcessName Analysis code string to be executed nHandle If the ProcessName code is actually executed, this process handle value pDataSectionAddr ProcessName The start address of the data section in the code's virtual memory space.

Table 8 is a description of the PROCESS_INFO structure 309 field.

In FIG. 14, it can be seen that the analysis code names to be executed are Engineering_Code_1, Engineering_Code_2 and Engineering_Code_N.

That is, these analysis code names are written in ProcessName 309, and when these three executable files are executed, their process handle values are written to nHandle 309, and these three data section start address values are written to pDataSectionAddr 309 ).

Also, the DB manager 5, 1106, and 1206 reads the current value of the designated variable and records the specified value as the current value of the corresponding variable (refer to 1105, 1107, 1205, 1207 Reference)

That is, the DB manager unit 5, 1106, and 1206 includes a function of reading the current value of the specified variable (see FIG. 12) and a function of recording the specified value as the current value of the variable (see FIG. 13) .

First, in order to read the current value of the variable, the tagsvrDATA object 307 for the variable is obtained from the PVDB unit 3,303 of the database for the variables, and the address space of the analysis program owning the variable (pProcessInfo ) 314 plus the data section memory address (104, 105) plus the offset value (204) in the data section.

tagsvrDATA * pData = PVDB :: search (variable name)
pDest = pData-> src_var_info.pProcessInfo + pData->src_var_info.pOffset;
ReadProcessMemory (hi.hProcess, pDest, & Value, nSize, ...);

Table 9 is an example code for reading variable values existing in different address spaces.

In particular, since a value must be obtained in the analysis program address space 202 having an address space that is separate from the execution controller node unit 7,801 that executes the acquisition instruction, it is possible to use a general memory copy function "memcpy () There is no need to use the "ReadProcessMemory ()" function 1110 provided by the code of FIG.

FIG. 12 is a diagram illustrating that variable values are acquired in the address space 202 of the analysis program when the "GET" button is pressed in order to obtain the current value of the designated variable. An example of the implementation code 1108 , 1110) are also provided.

Next, in order to record the value specified in the memory space of the variable, the tagvrDATA object 307 acquired in the PVDB unit 3 303 is used to write the value specified in the address space (pProcessInfo 314) .

tagsvrDATA * pData = PVDB :: search (variable name)
pDest = pData-> src_var_info.pProcessInfo + pData->src_var_info.pOffset;
WriteProcessMemory (hi.hProcess, pDest, & Value, nSize, ...);

Table 10 is an example code for recording variable values existing in different address spaces.

In this case, you can not use the "memcpy ()" function, which is a normal memory copy function, and you must use the "WriteProcessMemory ()" function that uses the process handle value of the analyzer that owns the variable.

13 illustrates a flow of recording a preset value in the address space in the data section of the address space 202 of the analysis process to which the variable belongs when the "SET" button is pressed in order to record a specific value in the designated variable Examples of implementations 1208 and 1210 are also provided.

Next, the execution controller node unit 7 will be described in detail.

The execution controller node unit (7, 801)

A process initialization node unit (8, 403) for performing an initialization operation for integrated operation of analysis programs,

A boundary condition manager node unit (9, 904, 1004) for filling boundary condition values necessary for the analysis programs to operate,

And a real-time synchronization node unit (10, 914) for performing time synchronization of analysis programs.

That is, a process initializer node for performing various initializations for the integrated operation of the analysis programs 12, 202, and 806 and a process initializer node for filling the boundary condition values necessary for the respective analysis programs to run A Boundary Condition Manager node, and a Real-Time Synchronizer node that enables each analysis program to be time-synchronized and driven.

9 to 9) of the PVDB unit 3, the PIDB unit 4 construction, the analysis programs 12, 202 and 806, and the boundary condition values (see FIG. 9 (See below).

That is, the execution controller node unit 7, 801 is a process that is independent of the engineering analysis programs 12, 202, 806. In the execution controller node unit, the value of each global variable owned by the engineering analysis programs is changed It is only necessary to use special memory copy functions (e.g., ReadProcessMemory 1110 / WriteProcessMemory (912, 1210) for Windows) for the separate address spaces 202, 1114 and 1214.

The process initialization node unit (8, 403) performs an initialization operation for integrated operation of analysis programs, and performs functions in the following order (refer to FIG. 5).

Step 1, execute each analysis program 1302 listed in the process.ini file (FIG. 14) in the CREATE_SUSPENDED mode (see code example 405, see 404 in FIG. 5).

The analysis programs listed in Fig. 14 are Engineering_Code_1, Engineering_Code_2, and Engineering_Code_N.

Step 2: Dynamically insert codes 603 and 604 for time synchronization into the MEL section (sections 705 to 706, sections 807 to 808) of the analysis programs 12, 202 and 806 suspended in the CREATE_SUSPENDED state 8) so that each engineering analysis program can be executed with the time synchronization function (see FIG. 9).

In other words, although there is no time synchronization function in the original MEL section (sections 705 to 706 and sections 807 to 808) of the respective engineering analysis programs 12, 202 and 806, the process initialization node section 8, 403 forcibly analyzes each analysis The time synchronization section is inserted in the MEL section of the programs so as to enable the time synchronization execution (see FIG. 9), and the time synchronization is performed in the real time synchronization node sections 10 and 914 of the execution controller node section 7 and 801 If the analysis instructions (21, 805) are given to the respective engineering analysis programs (see example code: 916), then all analytical programs 12, 202, 806 will execute their MEL sections simultaneously will be.

In Fig. 9, this principle is illustrated. Reference numeral 406 in FIG. 5 may be referred to.

Step 3: Based on the offset information 204 provided in the memory map files 201 and 301 of the internal global variables of the respective engineering analysis programs 12, 202 and 806, (PVB) (3, 303) the corresponding tagsvrDATA objects 307 on the B-Tree basis 311, 1113, and 1213.

At this time, the following field values are recorded in the tagsvrDATA object (see reference numeral 408).

        - data :: src_var_info (see Table 4 for a description of each of these fields)

        - data :: pProcessInfo address value

        - data :: nTargetVarInfo

 - data :: target_var_info (see Table 5 for a description of each of these fields)

Reference numeral 407 in FIG. 5 may be referred to.

Step 4, using the PROCESS_INFO object 309 to process.ini (1: 1 correspondence with one analysis program) by using the engineering analysis program list information 1302 to be executed provided in the process.ini file (see FIG. 14) (PIDB) (4, 304) with a simple loading method (312) in which they are stored in the database memory in the order listed in the file.

Reference numeral 409 in FIG. 5 may be referred to.

In step 5, the engine analysis programs 12, 202, and 806 that were CREATE_SUSPENDED for the initialization operation are resumed to execute step 3 in FIG. 9 to wait for the event synchronization signal. Code example: see reference numeral 412)

Reference numeral 411 in FIG. 5 may be referred to.

For example, the ResumeThread (pi.hThread) command is executed separately for the main threads of each engineering analysis program 12, 202, 806 in the execution controller node unit 7, 801, Allowing the engineering analysis programs 12, 202, 806 to resume execution.

That is, the engineering analysis programs do not resume themselves, but issue an instruction to resume from the outside at the execution controller node unit 7, 801.

If the event synchronization signal is set (21, 805) in the execution controller node unit 7, 801, the analysis programs are stored in the MEL section (705 to 706 section, 807 to 808 section) (3) In the state (807) of FIG. 9, the state is the state from the waiting state to the state (808), the state is moved to the state (803) again and is placed in the next standby state.

That is, by setting a single event synchronization object signal in the execution controller node unit 7, 801, all the analysis programs 12, 202, 806 waiting on the same event synchronization object can be executed simultaneously And advances the Problem Time 804 only by the same time interval to advance the time synchronization between the analysis programs 12, 202, and 806.

Here, the reason why the process is put into the standby state with CREATE_SUSPENDED is that the analytical programs dynamically insert (605) a code for time synchronization in the MEL parts 503, 602, and 702 before executing the respective MEL sections The PVDBs 3 and 303 based on the B-Tree 311, 1113, and 1213 are managed based on the memory map files 201 and 301 so that the analysis programs 12, 202 and 806 can achieve time synchronization, And prepares each field value of the tagsvrDATA structure 307 which stores each variable information in the PVDB 3 and 303.

If you do not set CREATE_SUSPENDED and start the process with a normal option, you can not insert dynamic time synchronization code for the MEL part, and because each analysis program is executed without time synchronization, the CREATE_SUSPENDED state To run the program.

In addition, the boundary condition manager node unit (9, 904, 1004) performs a function of filling boundary condition values necessary for the analysis programs to operate.

Each engineering analysis program 806 must know the property values of the desired boundary condition regions in order to perform its own execution.

Furthermore, when executed independently in the off-line, the boundary condition values are input only once with the input of Inputdeck at the beginning of execution. However, when a plurality of analysis programs provide boundary condition values to each other as in the present invention, The integrated analysis of the correct method will be possible only if the results calculated by the respective analysis programs 806 are provided as the boundary condition values of the other analysis programs for each step 803.

For this purpose, the boundary condition manager node unit 9, 904, and 1004 divides the values of the internal global variables calculated by the analysis programs 806 by each time-step basis 803 into desired target variables ( 310, 1303 and 1304 in FIG. 14), and executes the operation in the following order.

Step 1: Performs a Ready-to-Run status check to check whether all the analysis programs 806 are in the Wait state (reference numeral 906 in FIG. 10)

If any one of the analysis programs 806 fails to be in the Wait state because the execution of the MEL section (intervals 807 to 808 in FIG. 9) is not completed, the system waits until the periodic check time for the read-to-run. (Reference numerals 907 and 905 in Fig. 10)

That is, some analysis programs take a long execution time during execution of the MEL section (intervals 807 to 808 of FIG. 9) and are executed until the next ready-to-run check time 802 sent from the execution controller node unit 7, 801 If it is not completed, the execution controller node unit 7, 801 must not send a synchronization object setting (Set) signal 21, 805 to cause other analysis programs 806 already in the standby state to execute .

All of the other analysis programs 12, 202, and 806 have already completed the MEL section (sections 807 to 808 of FIG. 9), but if some of the analysis programs have not completed the MEL section (sections 807 to 808 of FIG. 9) , It is necessary to wait until it finishes and to execute it together so that time synchronization can be achieved.

Therefore, in order to synchronize the time, all the analysis programs check whether the execution of the one MEL interval (interval 807 to 808 of FIG. 9) is completed.

Step 2: If all the analysis programs 806 have completed the MEL section and are all in a wait state (see ③ and 707 in Figs. 8 and 9), then each analysis program 806 The required bounding area values should be updated.

This is because each analysis program 806 can execute with a newly calculated boundary condition value that is different from the previous time-step 803.

The method of updating the values of the boundary conditions is as follows.

All the internal variables (i.e., tagsvrDATA object 307) registered in the B-Tree 311, 1113, and 1213 of the PVDB 3,303 are navigated and the following steps are directly applied.

Copy all the src_var_info variable value (313) to the target address (315) that each of the internal variables should deliver.

That is, the src_var variable must be copied to all of the target variables whose values need to be copied, so that the boundary conditions are all filled (see 908, 909, 910, 911, and 912 in FIG. 10)

For reference, the following code is used to cruise all tagsvrDATA objects registered in the PVDB and write the values to the target address.

number = tagsvrDATA :: nTargetVarInfo;

       loop number> 0

  pSrc = tagsvrDATA :: pProcessInfo-> pDataSectionAddr + tagsvrDATA :: src_var_info-> pOffset;

       pDest = tagsvrDATA :: target_var_info [number-1] -> pAddr;

       WriteProcessMemory (pi.hProcess, pDest, pSrtc, nSize, ...);

       number = number - 1;

   end loop

In step 3, all the parameters registered in the B-Tree 311, 1113, and 1213 of the PVDB 3,303 are navigated to obtain target parameters 310 Time synchronization node unit 10, 914 (see reference numerals 913, 915 in FIG. 10).

In addition, the real-time synchronization node unit (10, 914) performs time synchronization of analysis programs to be able to be driven.

When all engineering programs are in a wait state 707 and all of their boundary condition values have been updated, all the engineering analysis programs simultaneously execute their respective MEL sections (sections 705-706, sections 807-808) The real-time synchronization node units 10 and 914 perform the functions of setting (Set) 21 and 805 the signal of the event synchronization object with respect to the standby state so that they can be started.

Since each of the analysis programs 806 has been updated with the new values of the boundary condition variable values by the step 1 and the boundary condition manager node unit 9, 904 and 1004, each of the analysis programs 806 has its own MEL interval 705 A set signal 21, 805, 915 of the event synchronization object is set so that the event synchronization object can be simultaneously executed in the period from 0 to 706 and from 807 to 808. (Reference Code: Reference 916) 9, the setting signals 21 and 805 are received by the analysis programs 806 and the wait state 707 is canceled so that the respective MEL sections of the analysis programs are simultaneously executed, 9 (913 and 915 in FIG. 10), which is the wait state 707, again.

Step 2, and the next synchronization set time 802. [0053] In the next periodic synchronization setting time 802, Ready-to-Run check and boundary condition value filling of the boundary condition manager node unit 9, 904 and 1004 and event synchronization signal setting 21, 805 of the real time synchronization node unit 10 ) Transmission operation is repeatedly performed, and the time automation execution is repeated (refer to steps 917 and 905 of FIG. 10)

In the case of FIG. 10, a flow chart of the boundary condition manager node unit 9 and the real-time synchronization node unit 10 is shown.

Meanwhile, when the time synchronization setting signal is generated in the execution controller node unit 7, the engineering code execution node unit 11 releases the wait states of the analysis programs and simultaneously executes each MEL section of each analysis program, As shown in FIG.

That is, if the time synchronization setting signal is generated (21, 805) by the execution controller node unit 7 with the nodes where the analysis programs 12, 806 are grouped, the engineering code execution node unit The wait states of the analysis programs 12 and 806 existing in the analysis program 11 are released and the respective MEL sections 705 to 706 and 807 to 808 of the analysis programs 12 and 806 It will run once at a time, making it time synchronized.

Meanwhile, the user interface node units 13, 401, 901, 1001, 1101,

An execution UI unit (23) for starting, executing, and stopping the execution controller node unit (7, 801)

And a data UI unit 24 for acquiring and setting values of internal global variables used by the engineering analysis programs through the execution controller node unit 7.

The execution UI unit 23 includes a START button 14 and a RUN button 15 and 902 for starting, And "FREEZE" buttons 16 and 1002, respectively.

Once the user presses the "START" button 14, 402, the process initializing node unit 8, 403 of the execution controller node unit 7 starts and the engineering code execution node unit 11, Activates the analysis codes 12, 202, 806 in CREATE_SUSPENDED mode and inserts (605) time synchronization coding into the MEL binary codes 503, 602, 702 of the respective engineering analysis codes (loaded in the RAM memory) , A B-Tree based database (PVDB) for the global variables of the engineering analysis codes, and a database (PIDB, 4) for the engineering analysis code process itself.

In Fig. 5, a control flow chart when the "START" button is pressed is provided together with a code example.

Secondly, when the "RUN" button 15, 902 is pressed, the boundary condition manager node unit 9, 904, 1004 of the execution controller node unit 7 starts and all the analysis programs 806 are in the Wait state To-run state check for checking the state of the analysis programs 806. When assuming a wait state, the update processing of the boundary region values of the analysis programs 806 is performed, and an event synchronization object set signal generation 21, 805) so that each of the analysis programs 806 operates the real-time synchronization node unit 10 to execute each MEL section (sections 705 to 706 and sections 807 to 808) one time at a time, The controller node unit (7, 801) enters the waiting state.

In Fig. 10, a control flow chart of the case where the "RUN" button is pressed (903) is provided together with a code example.

Thirdly, when the "FREEZE" button 16 1002 is pressed (1003), the periodic Ready-to of the boundary condition manager node unit (9, 904, 1004) And does not send the synchronization setting signal 21, 805 (regardless of the run check point 802, 905).

If all the analysis programs 806 are processed as described above, the analyzing program 806 is put into the waiting state 1005 as long as the waiting event synchronizing object does not receive the setting signal, It has exactly the same effect as making it.

If the user again presses the "RUN" button 15,902, it returns to the state of periodically transmitting the synchronization setting signal.

11, a control flow chart when the "FREEZE" button 16, 1002 is pressed is provided together with a code example (see reference numeral 1006).

In addition, the data UI unit 24 performs a function of acquiring and setting values of internal global variables used by the engineering analysis programs through the execution controller node unit 7.

In other words, a "GET" button 19, 1102 for acquiring the value of the internal global variable used by the engineering analysis programs through the execution controller node unit 7, 801 and a "SET" button (20, 1202).

First, when the user inputs a variable name for which the value is desired to be acquired in the "Varname" edit window 17, 1103 and presses the "GET" button 19, 1102, (1115, 1109, 1111), and records the value in the Value edit window (1104).

When the user presses the "GET" button 19 or 1102 in Fig. 12, a command flow chart is shown.

Second, when the user inputs a desired variable name into the "Varname" edit window 17, 1203 and then presses the "SET" button 20, 1202 after writing the specified value 18, 1204, The value of the variable is recorded (1211, 1209, 1215) by the method of the provided example code.

In FIG. 13, a command flow chart is shown when the user presses the "SET" button 20, 1202.

Through the present invention as described above, the memory map file information on the memory address space of the internal variables used by the analysis programs can be used without changing the source of the engineering analysis programs to be integrated, The analysis programs are remotely infiltrated into the internal memory space of the analysis programs and can provide the boundary condition values of the boundary-condition variables needed by the analysis programs, Effect.

In addition, an event synchronization object is generated before and after the MEL section of the analysis programs so that the analysis programs can be driven at the same time (Problem Time) of each analysis program, and the insertion code is dynamically inserted and changed And thus all of the running analysis programs can perform the virtual integration for the engineering processes so that the time synchronous execution can be performed.

It will be understood by those skilled in the art that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. It is to be understood, therefore, that the embodiments described above are to be considered in all respects as illustrative and not restrictive.

1: Data controller node part
7: Execution controller node part
11: Engineering code execution node section
13: User interface node section

Claims (4)

1. A system for analyzing a virtual analytical code using a process data section,
A PVDB unit 3 that builds information on internal variables of analysis codes based on B-TREE,
A DB builder node unit 2 including a PIDB unit 4 constructed by building a structure including information of a process name, a handle value, and a dynamic memory start address value of an analysis code process on a simple loading scheme,
A data manager node unit (1) including a DB manager unit (5) for reading a current value of a designated variable and recording a specific value as a current value of the variable;
A process initializing node unit 8 for performing an initializing operation for integrated driving of analysis programs,
A boundary condition manager node unit 9 for filling the boundary condition values necessary for the analysis programs to operate,
And a real-time synchronization node unit (10) for performing time synchronization of programs for analysis and being able to be driven;
When the time synchronization setting signal is generated in the execution controller node unit 7, the engineering code execution node (BS) 7 releases the waiting states of the analysis programs and simultaneously executes each main execution loop (MEL) (11);
An execution UI unit 23 for starting, executing, and stopping the execution controller node unit 7;
And a user interface node unit (13) including a data UI unit (24) for acquiring and setting values of internal global variables used by engineering analysis programs through the execution controller node unit (7) An analytical code virtual integrated drive system for scientific analysis using process data section.
The method according to claim 1,
The analytical code virtual integrated drive system,
When the analysis program is integrated, the memory map file information including the offset address information in the data section, which is the storage space of the internal global variables used by the analysis programs, And the internal variable values of the independent virtual memory space of the analytic code can be directly read and written so that analysis programs can be integrated without changing the source of the analysis program.
The method according to claim 1,
The analytical code virtual integrated drive system,
A binary code for generating and waiting for an event synchronization entity is inserted into a main execution loop section of the analysis program for time synchronization when the analysis programs are integrated, Wherein the plurality of analytical programs are time synchronized and can be executed integrally.
The method according to claim 1,
The PVDB unit (3)
The analysis programs 12,202 and 806 may determine that the internal global (or Heap) variables are in an integrated execution state, which indicates the memory offset value information 204 in the data sections 101, 102, 103, Based on the memory map files 201 and 301 of the target analysis programs, the tagsvrDATA 307 objects composed of the information (variable name, offset value, belonging analysis program handle value, etc.) of the individual global variables are stored in the B-Tree 311, 1113, and 1213. The virtual analytic code virtual integrated drive system using the process data section.

Images