JPH04100104A - Execution order automatic determining system for data flow diagram type program - Google Patents

Abstract

PURPOSE:To set the execution order without manual operation by building a program, which automatically sets the execution order, in an editor which generates the application program of a controller. CONSTITUTION:A stack 105 is used when elements to be operated are successively retrieved upward from below. That is, element numbers of elements whose execution order is not settled are pushed to the stack for the purpose of holding them. Then, retrieval arrives at the uppermost element, and the element number of this element remains in the top of the stack 105. At this time, one route reaching the lowermost element from the uppermost element remains in the stack 105. The element number in the top of the stack 105 is popped and is transferred to an execution order output buffer 103 for output of the execution order. This processing is repeated to settle the execution order up to the lowermost element. Thus, the execution order is accurately determined in a short time by the computer.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a controller installed at a manufacturing site such as a steel plant, a chemical plant, a paper manufacturing plant, etc. This relates to a method for automatically determining the execution order of elements that perform control and calculations.

[Conventional technology]

Techniques using programs in data flow diagram format include Japanese Patent Application No. 1-193182 and Japanese Patent Application No. 1-219886 previously proposed by the present applicant.

Program in data flow diagram format (hereinafter referred to as drawing)
is shown in Figure 8.

Referring to FIG. 8, each square is a symbol of an element (hereinafter referred to as an "element") that performs control or calculation.

The numbers inside the element are numbers for identifying the element (hereinafter referred to as "
(referred to as "element number"). The number in parentheses above an element is the execution order of that element. Elements with lower numbers are executed first. Lines with arrows at the ends indicate paths along which data flows between elements. The starting end of the line is the output terminal of the element, and the ending end is the input terminal of the element. In the drawings described above, one element cannot operate correctly unless data at all of its input terminals are determined. In other words,
The elements must be executed in order starting from the upstream side of the data flow (the left side in FIG. 8).

In conventional data flow diagram format programs, elements are generated on a program creation screen (=drawing) such as a display device, the input and output terminals of these elements are connected with lines, and then the execution order of the elements (8th The numbers in parentheses in the figure) were set by a person.

[Problem to be solved by the invention] For this reason, it is fine if the number of elements in the drawing is small, but if the number of elements in the drawing increases, the execution order may overlap, or elements that should be executed first may be Mistakes such as having to execute it later can happen.

Therefore, in order to prevent the above-mentioned mistakes, the present invention has the following features:
By incorporating a program that automatically sets the execution order into an editor for creating a controller application program, a method is provided in which the execution order can be set without human intervention.

[Failure to solve the problem]

The present invention is directed to the point that in a data flow diagram format program, an element cannot operate correctly unless the data at the output terminal of an upstream element connected to its input terminal is determined. Starting from the most downstream element, it becomes a prerequisite for the correct operation of that element.'' - Search for the downstream element, and when the upstream element is found, the 1F flow, which is a prerequisite, is determined. The elements to be operated are sequentially searched from downstream to upstream, such as searching for elements on the side. Use the stack when exploring.

In other words, in order to suspend elements whose execution order is not determined,
Bush the element number onto the stack. Eventually, the device reaches the upstream end, and the device number of the upstream end device remains at the top of the stack.

At this time, one path is left in the stack from the element at the downstream end to the element at the upstream end. Next, pop the element number at the top of the stack and move it to the buffer that outputs the execution order. By sequentially repeating this process, the original execution order up to the downstream end element is determined. During this process, there is an element with multiple input terminals,
If there is a route that has not been determined, a search upstream from there is performed in the same manner as above. However, it is assumed that there are no elements forming a feedback loop.

[Effect]

With the above method, the execution order of any program, no matter how complex, can be automatically set in a short time, making it possible to improve the reliability and productivity of the program much more than when it is done manually.

〔Example〕

Hereinafter, specific examples of the present invention will be described.

FIG. 1 is an example of a functional block diagram for realizing automatic execution order setting. Referring to Figure 1, program 1
01 is a connection table (hereinafter referred to as rLsI) that represents the connection state between elements.
After inputting the contents of 102 (referred to as rLsOJ) and determining the execution order, it is output to an execution order output buffer (hereinafter referred to as rLsOJ) 1o3.

A write pointer (hereinafter referred to as rOPJ) 1o4 is a pointer indicating how far into the execution order output buffer 103 the element number has been written.

When determining the execution order, a stack (hereinafter referred to as "rSTK") 105 is used. Stack pointer (hereinafter referred to as rSPI
) 106 is a pointer indicating how far in the stack 105 the search information has been pushed.

FIG. 2 is an example of the connection table 102 in FIG. This table is a tabular representation of the connection relationships between the elements shown in FIG. Referring to FIG. 2, one row of the table shows the connection relationship of one element.

The first column 201 shows the element number. The second column 202 shows the number of input terminals of the element. From the third column onwards 203 is the element number (
(hereinafter referred to as "input element number"). Note that the input element numbers of the upstream end elements are 0, such as 204 and 205.

FIG. 3 is a flowchart showing the algorithm of the program 101 in FIG. The contents of each processing step in FIG. 3 will be explained below.

(Process 301) Initialize SP and OP.

(Process 302) The element number of the element at the downstream end is stored in STK.

FIG. 4 shows the result of executing step 302 in the case of FIG.

(Process 303) Determine whether STK is empty.

If it is empty, processing ends. If it is not empty, the process advances to step 304.

(Process 304) If the element number at the top of STK is Ol, that is, the input element number of the upstream end element, the process advances to process 307. Otherwise, the process advances to step 305.

(Process 305) The element number at the top of STK is LSO
If it is already included, the process advances to step 307. Otherwise, the process advances to step 306.

(Process! 306) Read the input element number of the element at the top of STK from the LSI and store it in STK. The process then proceeds to process 303. In FIG. 5, the process 80 in the case of FIG.
The results of executing step 6 are shown below.

(Process 307) If there is a [: mark second from the top of the STK, the process advances to process 308. Otherwise, process 31
Go to 1.

(Process 308) If the third element number from the top of the STK is already included in the LSO, the process advances to process 311. Otherwise, the process advances to process 309.

(Process 309) Store the third element number from the top of STK at the top of LSO. FIG. 6 shows the result of executing the process 309 in the case of FIG.

(Process 310) Advance the LSO write point by one.

(Process 311) The top of the STK and the second content from the top are excluded from the STK.

FIG. 4 shows that the ST
The information generated in K is shown. Referring to Figure 4, 4
01 and 403 are the element numbers of the elements at the downstream end. 402
is a mark (1) for separating two element numbers.

FIG. 5 shows the process 30 of FIG. S following the state of FIG. 4.
6 shows the information generated in the STK as a result of execution. The processing path immediately after the start is process 301 → process 302 → process 308 → process 304 → process 305 → process 306. Referring to FIG. 5, 501 is a symbol (=) indicating that the right side is the input element number list of the element on the left side. 502 and 504 are input element numbers.

503 is a mark (.) for separating input element numbers.

).

FIG. 6 shows a state in which the element number whose processing order is determined first is written into the LSO as a result of executing the process 309 in FIG. Referring to FIG. 6, 601 is the first element to be executed.

FIG. 7 represents a process in which the execution order is determined by tracing changes in the internal states of STK and LSO. Starting from 5TEPI, 5TEP
It ends at 24.

〔Effect of the invention〕

As explained above, according to the present invention, in conventional data flow diagram format programs, the execution order of functional elements that perform control and calculations is manually set, which takes time to create programs and reduces the occurrence of bugs. It becomes possible to quickly and accurately determine the order of execution using a computer.

[Brief explanation of the drawing]

Figure 1 is a block diagram of a program that determines the execution order.
Fig. 2 is a diagram showing an example of a table representing the connection state of elements, Fig. 3 is a flowchart showing the processing contents of a program that determines the execution order, and Fig. 4 is a diagram showing an example of a table representing the connection state of elements. Figure 5 shows an example of the information stored in the stack. Figure 5 shows an example of the state of the stack while searching for the execution order. Figure 6 shows the state of the execution order output buffer while searching for the execution order. Figure 7 is a diagram showing an example of a state. Figure 7 is a diagram showing an example of a process for searching the execution order expressed by tracing the status of the stack and execution order output buffer. Figure 8 is an example of a program in data flow diagram format. FIG. Patent applicant Anyo Electric Manufacturing Co., Ltd. Figure No. 4 Good size ('J ('JcOCf)
If) Cf) ■(f) (0(j) c
c+G. Ll") Lltsu LI"l hesitation U) BatsuUtsuL
i) lj tsuri r-tototori r-r-t'-r-r-r'
-r-r-t cO's nonono a: lωω's nononono's ■■'s ■■■■'s cn's ■■

Claims (1)

[Claims] In a program in data flow diagram format, after creating functional elements that perform control or calculations and completing the connections between these functional elements, the execution order of the functional elements is set to (1) the most downstream; Starting from a certain element, search for upstream elements that are a prerequisite for the correct operation of that element, (2) For elements whose execution order is not determined, the element number is placed in the stack and held, (3) When the element at the upstream end is reached, pop the element number at the top of the stack, move it to the buffer for outputting the execution order, and (4) repeat the processes from (1) to (3) above in sequence. During this process, if there is an element with multiple input terminals and there is an undefined path, search upstream from there using the same method as above and return to the original downstream element. An automatic execution order determination method for a program in a data flow diagram format, characterized by determining the execution order by determining the order.