JPH04326167A - Graphic information processor - Google Patents

Abstract

PURPOSE:To offer graphic information processing technique capable of obtaining very easily the evaluation information of plotting information from the plotting information. CONSTITUTION:A basic data part 4 consisting of a graphic.connection data storage part 1 to store the information of the shape, the meaning and the connection relation of a symbol and an attribute data storage part 3 to store attribute data 2 corresponding to the symbol, and a converting part 5 to convert data from the data part 4 into description related to factual data are provided. Besides, a knowledge base 8 consisting of a factual data storage part 6 to store the factual data converted by the converting part 5 and a predicative rule storage part 7 to stipulate a predicative rule by input from the outsides by declarative language is provided. A deductive inference part 10 to execute deductive inference by the knowledge base 8 is provided, and the above- mentioned inference part 10 executes the deductive inference from the predicative rule and the factual data of the above-mentioned knowledge base 8 on the basis of a question retrieval expression given from the outsides, and finds a retrieved answer corresponding to the contents of the above-mentioned question, and displays this retrieved answer on an operating part 11, and outputs the graphic to an output part 12.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention converts graphical information obtained from CAD, etc., obtained during the design, drawing, and input stages into a factual knowledge base together with attribute information, and enables checks in the design and input process by deductive reasoning, and・Driving and safety of input products,
The present invention relates to a graphical information processing device from which information such as equipment management can be obtained.

0002

2. Description of the Related Art In recent years, the capacity and computational processing power of EWS (engineering workstations) and the like have increased dramatically due to the increase in speed of memory technology and processors. Therefore, although the use of intelligent information processing such as expert systems is expected, it is not always possible for users to use them.

[0003] This is because it is extremely difficult to create a large-scale knowledge base, which is a prerequisite for practical intelligent information processing, especially a knowledge base containing a large amount of facts regarding actual objects. In other words, factual knowledge is sometimes expressed in various forms such as figures and shapes, and these are not necessarily suitable for intellectual information processing.

[0004] For example, when large-scale plant equipment is designed and drawn using CAD, in order to check the safety, operability, etc. of the equipment, conventional technology uses a CAD system to check the safety, operability, etc. of each element separately from the CAD drawings. It was necessary to separately input a large amount of information into a computer in a form suitable for knowledge processing and process it.

[0005] Furthermore, with respect to knowledge regarding rules and inferences, it is often found that it is necessary to newly create or add knowledge while creating a knowledge base and proceeding with intellectual processing. For example, when an accident occurs in plant equipment, it is necessary to search for hazardous locations in other existing equipment where a similar disaster may occur in order to prevent similar accidents.

[0006] However, with conventional technology such as equipment diagnosis, the type of accident is assumed first, and data and reasoning rules are determined accordingly, making it extremely difficult to deal with problems discovered later. Met.

[0007] Furthermore, according to the prior art, the design, drawing, and input using CAD, etc., and the knowledge base regarding the facts of the constituent elements appearing on the drawings are independent and separate, so that the design, drawing, and input are independent and separate from each other. Information designed and input using CAD cannot be used for checking, evaluating, searching, etc. regarding the stage, subsequent design, drawing/input objects, etc., and basically humans input data and give instructions to the computer. I had to.

[0008]

[Problem to be solved by the invention] For large-scale problems as described above, it is possible to input factual data for each target problem and program rules and knowledge.
It was difficult to develop software that required a huge amount of time and effort to input data, and it was also impractical in terms of time and cost.

[0009] In view of these points, the object of the present invention is to
The object of the present invention is to provide a graphical information processing technology that can extremely easily create a factual knowledge base from information inputted into a large-scale CAD design, etc., and can flexibly respond to new problems.

0010

[Means for Solving the Problems] The present invention provides a graphic/connection data storage section 1 that stores information on the shapes, meanings, and connection relationships of symbols, and an attribute data storage section 3 that stores attribute data 2 corresponding to the symbols. a basic data section 4 consisting of a basic data section 4, a conversion section 5 that converts the data from the data section 4 into a description regarding fact data, a fact data storage section 6 that stores the fact data converted by the conversion section 5, and an external Predicate rule storage unit 7 that defines predicate rules by inputting them in a declarative language from
and a deductive inference control unit 10 that performs deductive inference using the knowledge base unit 8. The inference unit 10 is configured to:
The gist is to perform deductive reasoning from the predicate rules and factual data in the knowledge base section 8, obtain a search answer corresponding to the question content, display this search answer on the operation section 11, and output it on the output section 12. .

[0011]

[Operation] A diagram of the principle of the present invention is shown in FIG. As shown in the figure, the configuration of this system includes a basic data section 4, a figure/connection data storage section 1 that stores shape/connection information of individual symbols, etc., and an attribute data storage section corresponding to each symbol, etc. It has 3.

Here, the graphic information of the graphic/connection data includes, for example, in the case of a piping instrumentation system, the shapes of equipment such as drums and towers, valves, flanges, piping parts, etc., and the straight lines that are their elements. , circular arcs, and other parts (primitives), as well as equipment such as distillation units, piping parts, and other aggregate information.

[0013] Furthermore, the connection data includes data showing the connection relationships of individual devices and piping, and the connection positions by corresponding parts (primitives).

In the case of a piping instrumentation system, the attribute data 2 is the design pressure, material, etc. of piping components, and in the case of an electric circuit, it is the rated current value of component elements such as electric elements. These are generated by designing, drawing, and inputting operations using CAD, etc.

Next, in the present invention, the information in the basic data section 4 is converted into fact data in a predicate format in the conversion section 5. The factual data generated here is, for example, factual information about the symbol such as the part element name, classification code, design pressure, etc. of the symbol on the CAD display.

[0016] Furthermore, the predicate rules given from the outside include a conceptual definition of the factual information (for example, a definition of a classification code, etc.),
Alternatively, a definition of deductive reasoning in the search (for example, an element whose voltage value when activated is higher than the rated voltage is dangerous) is prescribed.

The deductive inference control unit 10 performs deductive inference based on the given search query expression and fact data while referring to the predetermined word rule, and calculates the facts corresponding to the query expression, for example, equipment of equipment, piping, etc. Or search for dangerous points on the circuit.

[0018] The search answers obtained in this way are reflected in the display on the operation unit 11 and the drawings on the output unit 12. For example, by color-coding the relevant parts, the operator and user can easily find the search answers. I can understand it.

As described above, according to the present invention, a method is adopted in which factual data is extracted from figure/connection information of CAD, etc., and the deductive reasoning control unit 10 searches by deductive reasoning based on predicate rules created in advance. By doing so, firstly, CAD
It becomes very easy to generate a knowledge base of large-scale facts on which deductive reasoning can be based from a huge amount of data.

Second, since it is created in a predicate format, it is easy to generate and manage knowledge bases regarding inference rules and rules. Even if predicate rules are not determined in advance, by inputting predicate rules and search query expressions, new and diverse search contents can be executed without changing the original fact data.

[0021] Thirdly, by reflecting the search answers on the operation unit or the drawn drawings, operators and users can very easily grasp the contents of the search answers, such as dangerous areas of the system. I can do it.

[0022]

[Embodiment 1] The present invention will be specifically explained based on an embodiment in which the present invention is applied to the basic design of piping instrumentation in an actual plant.

FIG. 2 shows the device configuration of this embodiment. In the following description, description of the same parts as those in FIG. 1 described above will be omitted.

FIG. 3 shows a part of the equipment and piping system that are the objects of data stored in the graphic/connection data storage section 1 of the basic data section 4. In the figure, a column 13 and a drum 14 are connected to each other by a pipe 15. Design drawings as shown in the figure can be drawn using known graphic input means through the operating section 11 equipped with input means such as a mouse and digitizer.

The output section 12 is generally a display device such as a CRT, but it can also include an output means such as a plotter.

FIG. 4 shows a graphic information table, which may be in any format, but at least includes a specific identifier such as a node number for internally identifying a symbol, and a classification code for distinguishing the meaning of the symbol. It consists of an identifier and primitive information such as straight lines and arcs to indicate the shape. Further, if necessary, a part code indicating the meaning of each primitive, a group number identifying a set of symbols, etc. are assigned.

FIGS. 5 and 6 show a connection information table,
It shows the group information table in it, and clarifies the connection correspondence between symbols. Furthermore, if necessary, the positions where symbols are connected are indicated by parts (primitives).

FIG. 7 shows an example of an attribute data table stored in the attribute data storage section. In the figure, the design pressure of the drum represented by the identifier (number) "D-201" as attribute data is 10.3 kg/cm2G,
It shows that the design temperature is 200°C. Such attribute data can be registered in any known data format such as a table or field in a relational database.

The conversion unit 5 generates the following fact data based on the figure/connection data and attribute data 2 generated by the design, drawing, and input shown in FIG.
KEY(n102346,'D-201').
(Data 1)
node (n102346, c101)
．． (Data 2) node(n102345,c
201).
(Data 3) node (n1023
47, c300).
(Data 4) joint(
n102346, n102347).
(Data 5) jo
int(n102347, n102345).
(Data 6)
conn(n102346,b2.n102347
,b1). (Data 7)
conn(n102347,b1.n1
02345, b2). (Data 8) attr('D-201', d
p, ty0, un141, 10.3).
(Data 9) attr('D-20
1', op, ty0, un141, 12.7).
(Data 10) attr(
'D-201', dt, ty0, un11, 200. )
．． (Data 11) The meanings of each of the above data are as follows. Data 1: The machine number of symbol n102346 on the drawing is D-201. Data 2...The classification code of symbol n102346 is c1
01 (drum 14). Data 3...n102345 is c201 (tower 13). Data 4...n102347 is c300 (piping 15). Data 5...n102346 and n102347 are connected. Data 6...n102347 and n102345 are connected. Data 7...n102346 and n102347 are connected at site 2 and site 1, respectively. Data 8...n102347 and n102345 are connected at site 1 and site 2, respectively. Data 9...The design pressure of D-201 is 10.3 kg/c
It is m2G. Data 10...The design temperature of D-201 is 200°C. Data 11...The operating pressure of D-201 is 12.7 kg/c
It is m2G.

[0030] Here, nXXXXXX (X means a number) is a node number, which is a type of identification number unique to each symbol in the CAD drawing.

In the knowledge base section 8, the predicate rules in the predicate rule storage section 7 are registered in a declarative language format.

In other words, the expression "If A, then B" is expressed as "A:-B". Further, the expression "If A and B, then C" is expressed as "C:-A,B". This is specifically shown below. drum(X):-node(X.c101). (
Rule 1) equipment (X): -drum (X
). (Rule 2) un_safe_equipme
nt(X,Y):-equipment(X),key
(X, Y), attr(Y, dp, z.un141, p
d), attr(Y, op, z, un141, Po),
Pd<Po. (Rule 3) The meanings of each of the above rules are as follows.

Rule 1: The symbol (node) of classification code c101 represents a drum. Rule 2...Drums are instruments. Rule 3: Equipment whose operating pressure is higher than the design pressure is dangerous.

The predetermined word rule may be registered in advance in the predicate rule storage unit 7 as a general rule, or may be input as an additional rule when inputting a search question expression. Such a predicate rule given from the outside in a text format is translated into a format suitable for high-speed search by the compiling unit using a deductive database method, and is stored (registered) in the predicate rule storage unit 7.

Next, a specific example of a search question given to the deductive reasoning control section 10 will be shown below. In the questions below,
We are looking for the node number and machine number of unsafe equipment.

：-un_safe_equipment
(X, Y). (Question 1) Question 1 input in the predescriptive word format is compiled by the compiling unit 16 and translated into a high-speed search format using a deductive database search method, and then is provided to the deductive reasoning control unit 10. In response to the question, the deductive reasoning control unit 10 searches for a corresponding fact based on fact data and predicate rules, and generates the result as fact data.

[0037] Here, to briefly explain the arithmetic processing in the deductive inference control unit 10, first, deductive inference is performed by referring to each rule using the predicate rules in the predicate rule storage unit 7.

Here, Rule 3 will be referred to from Question 1 shown above. In other words, according to Rule 3, when searching for "dangerous equipment," it is sufficient to extract "equipment whose operating pressure is higher than the design pressure."

Next, according to rule 2, the drum/
Find a node. At this time, from rule 3, the classification code "
It turns out that the node defined by "c101" corresponds, and from data 2 it turns out that node "n102346" is a drum, and from data 1 it turns out that the machine number of this drum is "D-201". do.

The attribute data of the "D-201" drum is retrieved, and the design pressure and the operating pressure are compared from data 9 and data 11, and it is determined that the "D-201" drum corresponds to dangerous equipment.

Based on the above search, the deductive reasoning control unit 10
will output the following search results. un_safe_equipment(n102346
,■D-201'). (Output 1) The above output means that the node number of the unsafe device is "n102346" and its machine number is "D-201" (drum 14).

The deductive reasoning control unit 10 may display corresponding parts of the drawing displayed on the CRT in different colors based on the search output. In this case, display information may be output from the knowledge data section in response to a command from the deductive reasoning control section 10. In this case, a color code can be added to the node number as bundle information. for example"(
n102346, 7). , (n102345, 1). ”
Display data is output by pairing the node number and color code. Here, the unsafe node is “n102
346'', and the symbol of this node is displayed in red (color code: 7). Further, "n102345" is a safe node and is displayed in normal green (color code: 1).

In the above explanation, unsafe points on the piping 15 were indicated based on the retrieved factual data, but in addition to this, for example, a search in which all valves of the same type are displayed in different colors, Equipment/pipes connected without connection 1
Search for the same pressure range to find 5 groups, search for 15 ranges of equipment/pipes with low design pressure within the same pressure range, and search for the presence or absence of safety devices to search for safety valves connected to equipment, etc. is feasible.

Further, in the embodiment, the case where the present invention is applied to the piping 15 plant has been described, but the present invention is not limited to this, and can also be applied to the design of electric (electronic) circuits, the piping 15 of air conditioning equipment, etc.

[0045]

Embodiment 2 FIG. 8 shows a system configuration in Embodiment 2 of the present invention.

As shown in the figure, this embodiment is almost the same as the first embodiment (FIG. 2), except that the conversion section 5 is provided with a conversion control section 17.

The conversion control unit 5 includes a change information registration unit 18
It is equipped with When the contents of primary data such as flow sheet CAD or drawing database registered in the basic data section 4 are changed, this change information registration section 18 registers only the information that the data has been changed as a log. It has the function of keeping

The greatest feature of this embodiment is the data management method in the knowledge base section 8.

That is, in this embodiment, when a part of the primary data is changed due to a change in specifications, etc., the conversion unit 5 logs only the fact of the change in the change information registration unit 18 in the conversion control unit 17. The fact data in the knowledge base section 8 is not updated at this stage.

When the question search expression is given to the deductive reasoning control unit 10 and the search process is started, the conversion control unit 17 constantly refers to the knowledge base unit 8 and extracts the facts from the deductive reasoning control unit 10. Compare the access to the data storage unit 6 with the change information held by the change information registration unit,
If the primary data that is the basis of the accessed fact data has been changed, access from the deductive reasoning control unit 10 is temporarily stopped, and the conversion control unit 17 issues a conversion command to the conversion unit 5. The relevant primary data is output, converted, and registered in the fact data storage section 6 of the knowledge base section 8, and updated.

FIG. 9 is a flow diagram showing the above processing. In this way, in this embodiment, when a change occurs in the primary data in the basic data section 4, the conversion process is not performed immediately into fact data, but the conversion process is performed as necessary at the stage of the search process. I decided to do so.

That is, although a vast amount of data is generally accumulated in a knowledge base, the factual data that is accessed when performing actual search processing is only a small portion of it.

[0053] Therefore, if the conversion unit 5 executes conversion and reorganizes the factual data each time primary data is changed, which is relatively frequently done, the burden increases only on the conversion process, and the search processing speed decreases. It may cause problems.

Regarding this point, according to the present embodiment, the conversion process in the conversion unit 5 is executed at the stage when it becomes necessary for the search process, so that extremely efficient processing can be realized.

In particular, in a so-called distributed database type system in which the knowledge base section 8 is distributed in a plurality of locations, when the primary data in the first knowledge base section is changed, the information in the unrelated second knowledge base section is changed. Since the knowledge base section is not interrupted for conversion processing, the system can be operated efficiently.

In the above description of the second embodiment, the changed data is read out from the basic data section 4 each time a data changed location in the knowledge base section 8 is accessed, and converted into new fact data each time. As explained in , the conversion control unit 1
When the conversion control unit 17 is notified that the portion of the fact data accessed with respect to No. 7 corresponds to the changed portion, the changed range including the changed data may be batch-processed all at once.

Further, conversion and storage may be performed all at once by an external command or timer control.

[0058]

Effects of the Invention The present invention converts figure/connection information and attribute data into knowledge-based fact-related data during design, drawing, and input using CAD, etc., and searches by deductive reasoning based on predicate rules. It becomes easy to generate a knowledge base of large-scale facts on which to base inferences from a huge amount of data.

Furthermore, since the information is created in a predicate format, it is easy to generate and manage a knowledge base regarding rules.

Furthermore, by reflecting the search answers on drawings, users can easily understand the content, so they can review the design, safety, etc. at the design/planning stage, and identify dangerous areas of existing equipment. Various analyzes such as extraction can be performed extremely easily and flexibly.

[Brief explanation of drawings]

[Figure 1] Block diagram showing the principle of the present invention

[Figure 2] Block diagram showing the system configuration in Example 1

[Figure 3] Diagram showing plotting data in Example 1

[Figure 4]
Explanatory diagram showing a graphic information table in Example 1

FIG. 5 is an explanatory diagram showing a connection information table in the first embodiment.

FIG. 6 is an explanatory diagram showing a group information table in connection information in Embodiment 1.

[Fig. 7] Explanatory diagram showing an attribute data table in Example 1

[Fig. 8] Block diagram showing the system configuration in Example 2

[Fig. 9] Flow diagram showing the processing procedure in Example 2

[Explanation of symbols]

1...Graphic/connection data storage section 2. Attribute data 3. Attribute data storage section 4. Basic data section 5. Conversion section 6. Fact data storage section 7. Predicate rule storage section 8. Knowledge base department 10...Deductive reasoning control unit 11...Operation unit 12...Output section 13...Tower 14...Drum 15...Piping 16...Compile section 17...Conversion control unit 18...Change information registration department

Claims (1)

[Scope of Claims] [Claim 1] Graphic/connection information that represents the shapes, etc. of symbols and their elements as graphic information, and/or the meanings of symbols and their elements as codes, and their connection relationships as connection relationship information. a basic data section (4) consisting of a figure/connection data storage section (1) that stores symbols, and an attribute data storage section (3) that stores attribute data (2) that corresponds to symbols, sets of symbols, and element parts of symbols; and,
a converting unit (5) that converts data from the basic data unit (4) into factual data representing factual relationships;
); a knowledge base unit (8) comprising a fact data storage unit (6) that stores fact data converted by the method; and a predicate rule storage unit (7) that defines predicate rules based on external input;
It consists of a deductive reasoning control section (10) that performs deductive reasoning based on the information of the knowledge base section (8), and the deductive reasoning control section (10)
0) is based on a question search formula given from the outside or a question search formula started by operating the operation unit (11).
Deductive reasoning is performed based on the predicate rules and fact data of the knowledge base section (8) to obtain a search answer corresponding to the question content, display this search answer on the operation section (11), and output the output section (11).
2) A graphics processing information device, characterized in that the converting section (5) includes a change information registration section (
18), and when the data content of the basic data section (4) is changed, only the change information is registered in the change information registration section (18). In addition, the conversion control section (5) is configured to control the deductive reasoning control section (
Monitor access from 10) to the knowledge base part (8),
When a location corresponding to the changed data is accessed in the knowledge base unit (8), the search process is interrupted and the conversion unit (5) is instructed to convert the data and store it in the knowledge base unit (8). 2. The graphic information processing apparatus according to claim 1, wherein the graphic information processing apparatus provides an instruction. 3. The conversion of the portion corresponding to the changed data into fact data and its storage in the knowledge base portion (8) is performed by notifying the conversion control portion (5) that the accessed data portion is a changed portion. 3. The graphic information processing apparatus according to claim 2, wherein the change range including the change data is batch-processed once the change data has been changed. 4. The graphical information processing device includes a system design (process,
The figures and connection data are used for the design and drawing of piping and instrumentation diagrams, electrical diagrams, sequence diagrams, etc., and the symbols and their collections have unique node numbers, which can be linked to keys such as equipment numbers. By having a comparison table, it corresponds to the attribute data of the entity in the equipment, etc., and the attribute data (2)
is at least a symbol, a set of symbols, and a specified value representing a design specification and state of an element of a symbol, the predicate rule includes at least a concept definition and an inference rule, and the deductive inference control unit (10) , by inputting a question search formula, search for relevant data from fact data according to predicate rules, and by inputting this relevant fact data from the outside, it is possible to search for symbols, sets of symbols, and colors, line thicknesses, and lines of elemental parts of symbols. Depending on the display characteristics such as type, etc., the operation section (11)
4. The graphical information processing apparatus according to claim 1, wherein the graphical information processing apparatus is characterized in that the graphical information processing apparatus displays the graphical information on a display as an output unit and outputs the diagram on a plotter as an output unit.