Preplan flow chart automatic generation method based on semi-normalized description system

Abstract

The invention discloses an automatic plan flow chart generation method based on a semi-normalized description system. According to the plan flow chart automatic generation method, through ontology constraint, automatic flow construction and a reverse generation mechanism, an automatic conversion and standardized correction process from a natural language plan to a standardized flow chart is realized. The method comprises three steps: plan flow description specification based on ontology constraint, automatic generation of a plan flow chart based on semi-standardized description, and strict plan standardization correction based on a reverse generation mechanism. According to the method, accurate description of events, conditions and skips is ensured through ontology constraints, a semi-normalized description system is utilized to automatically construct a logical relationship between the events and the conditions and generate a flow chart meeting the plan requirements, and finally, a reverse generation mechanism is adopted to convert the flow chart back to natural language expression. And a correction suggestion which does not conform to the specification description is given, so that the standardization and the consistency of the plan are ensured. According to the method, through process automatic conversion and standardized correction, the compiling efficiency of the complex system maintenance plan is improved, the readability and the executability of the emergency plan are enhanced, and the method is suitable for modern complex emergency management scenes.

Description

Technical Field

[0001] The present invention relates to a method for automatically generating a pre - plan flow chart, and particularly to a method for automatically generating a pre - plan flow chart based on a semi - normalized description system. Background Art

[0002] As a disposal description for typical abnormal events of complex systems, the maintenance pre - plan provides low - threshold operation and maintenance guidance for system users and is an important guarantee for the long - term reliable operation of complex systems. Facing the complex assembly structure and signal association within the system, the maintenance pre - plan gradually shows industrial characteristics of long sequentiality and multiple branches. Traditional pre - plans in the form of text are difficult to effectively guide users to independently complete complex maintenance operations; emerging carriers represented by flow charts, with advantages such as operation unitization and intuitive process, have become an important form of expression for complex system maintenance pre - plans. However, although pre - plans in the form of flow charts have significant advantages for users, pre - plans in text form have unique advantages in terms of editing flexibility and storage simplicity. How to achieve automatic conversion from text form to flow chart form through a description specification friendly to editors is a key issue that takes into account the reading needs of users and the writing efficiency of manufacturers' maintenance pre - plans.

[0003] To meet the above requirements, this method proposes a method for automatically generating a pre - plan flow chart based on a semi - normalized description system, providing an automatic generation method from natural language to a pre - plan flow chart. This method constructs a description specification optimized for writers facing the main expression logic involved in the flow chart; further provides a mechanism for automatically correcting common non - compliant description forms, thereby improving the writing efficiency of complex system maintenance pre - plans and promoting the rapid application of emerging maintenance pre - plans. Summary of the Invention

[0004] The present invention provides a method for automatically generating a pre - plan flow chart based on a semi - normalized description system to solve the problems of non - intuitive description and unclear process existing in the existing pre - plan flow expression method in emergency management. The method effectively realizes the automatic generation of the pre - plan flow and strict normalized correction by introducing a semi - normalized description system and combining ontology constraints and reverse generation mechanisms, so as to adapt to the dynamic requirements in modern complex emergency management scenarios. Through the automatically generated pre - plan flow chart, the method ensures that the emergency plan is clearer, more efficient and compliant with standards when dealing with complex emergency events.

[0005] The method for automatically generating a pre - plan flowchart based on a semi - normalized description system includes three main steps: the pre - plan process description specification based on ontology constraints, the automatic generation of the pre - plan flowchart based on semi - normalized description, and the strict normalization and correction of the pre - plan based on the reverse generation mechanism. The step of the pre - plan process description specification based on ontology constraints ensures the accuracy and standardization of the pre - plan description by clarifying the relationships among events, conditions, and jumps, providing a basis for the subsequent flowchart construction; the step of automatically generating the pre - plan flowchart based on semi - normalized description constructs the edge relationships between nodes formed by events and conditions by identifying step identifiers, and generates a flowchart structure that meets the requirements of the pre - plan process according to these relationships; the step of strictly normalizing and correcting the pre - plan based on the reverse generation mechanism converts the automatically generated flowchart structure into natural language expressions, further correcting and removing the non - standard parts in the user's writing process to ensure the standardization and normalization of the pre - plan.

[0006] The features of the present invention are as follows:

[0007] (1) By means of ontology constraints, the internal relationships among elements such as events, conditions, and jumps are clarified, providing an accurate semantic basis for the automatic generation of the pre - plan flowchart and breaking through the limitations of traditional manual descriptions.

[0008] (2) Through the semi - normalized description system, the edge relationships among multiple events and conditions can be automatically constructed, thus efficiently generating a pre - plan flowchart that meets the actual requirements and avoiding the blind spots and inefficiencies of manual flowchart design.

[0009] (3) Through the reverse generation mechanism, strict normalization and correction of the automatically generated flowchart are achieved, ensuring the consistency of the style of the pre - plan written by users and improving the reading friendliness when users analyze multiple pre - plans for complex systems. Brief Description of the Drawings

[0010] Figure 1 It is the overall flowchart of a method for automatically generating a pre - plan flowchart based on a semi - normalized description system.

[0011] Figure 2 It is the flowchart of the algorithm for automatically generating a pre - plan flowchart based on semi - normalized description.

[0012] Figure 3 It is the flowchart of the algorithm for strictly normalizing and correcting the pre - plan based on the reverse generation mechanism. Detailed Description of the Preferred Embodiment

[0013] The following provides a detailed description of a method for automatically generating a pre - plan flowchart based on a semi - normalized description system provided by the present invention with reference to the drawings.

[0014] The present invention provides a method for automatically generating a pre - plan flow chart based on a semi - normalized description system. Generally speaking, this method realizes the transformation process from natural language to a pre - plan flow chart through three main stages: the pre - plan flow description specification based on ontology constraints, the automatic generation of the pre - plan flow chart based on semi - normalized description, and the strict normalization correction of the pre - plan based on the reverse generation mechanism. The method first defines events, conditions, and jumps through the specification based on ontology constraints to ensure the accurate description of the pre - plan process; then uses the semi - normalized description system to construct the edge relationship between events and conditions through step identification and automatically generate the flow chart; finally, adopts the reverse generation mechanism to transform the flow chart into a natural - language expression and automatically correct the writing that does not conform to the specification to ensure that the generated pre - plan meets the specification requirements.

[0015] 1. Specification of pre - plan process description based on ontology constraints

[0016] The steps of the pre - plan process description specification based on ontology constraints define and classify the three concepts of events, conditions, and jumps. Ontology constraints refer to the constraints and clear descriptions of events, conditions, and jumps, ensuring that in the expression of the pre - plan process, the writing and structure of events, conditions, and jumps can be standardized to ensure that the entire flow - chart generation process follows accurate and consistent description specifications, mainly including three steps: the definition of events, the definition of conditions, and the definition of jumps. The definitions and examples of the three types of steps are shown in Table 1.

[0017] 1.1. Definition of events

[0018] An event is an action that occurs or a change in the system state after a certain condition is triggered, such as performing a specific action after the trigger condition. The method defines and identifies events in the input text through regular expressions.

[0019] 1.2. Definition of conditions

[0020] A condition is a prerequisite for an event, which is embodied as a judgment or verification in the specific implementation. For example, in the statement "If condition A is met, then event B is executed", the part "If... is met" is the condition. The method uses keywords to determine whether the input text is a condition, thereby parsing different condition expressions and splitting the statement into a prerequisite condition and a subsequent action to ensure their correct logical matching with events.

[0021] 1.3. Definition of jumps

[0022] A jump represents the transfer logic in the flow chart after an event occurs. For example, if a certain condition is met, the process may jump to the next step or return to the previous step. The method manages the transfer relationship between events by setting up a state machine and a graph structure, and the extracted jump logic is embodied as an edge connecting the nodes before and after the jump.

[0023] Table 1 Definitions and examples of the three types of steps

[0024]

[0025] Through the definitions of these basic concepts, the described method ensures the strict distinction and standardized expression of events, conditions, and jumps. The described method uses an ontology knowledge extraction mechanism to extract this information from the plan document or database, ensuring that all concepts comply with the pre-set ontology constraints and avoiding non-standard expressions and incorrect relationship definitions.

[0026] 2. Automatic Generation of Plan Flowcharts Based on Semi-Normalized Descriptions

[0027] The automatic generation steps of the plan flowchart based on semi-normalized descriptions use step identifiers to automatically construct the edge relationships between events and conditions to form a flowchart, mainly including three steps: plan step identification and parsing, construction of relationships between events and conditions, and generation of plan flowcharts.

[0028] 2.1. Plan Step Identification and Parsing

[0029] The described plan step identification and parsing steps first traverse all steps in the plan document or database, replacing punctuation marks (such as commas, semicolons, etc.) with standard formats. Then, regular expressions are used to extract the node identifiers of each condition and event, decomposing each step in the input text into nodes and removing redundant parts in the text (such as redundant punctuation marks, etc.). For text containing "if", "then", "otherwise", or "goto", the prerequisite conditions, subsequent operations, or jumps are parsed, and finally, the condition and event are extracted through the separate condition event function to process subsequent branch conditions.

[0030] 2.2. Construction of Relationships between Events and Conditions

[0031] After the plan step identification and parsing are completed, the described construction steps of the relationships between events and conditions analyze the logical relationships between nodes using condition and jump information. The basic information of the edge includes the current node identifier, the target node identifier, and the description information of the edge. After parsing each condition or operation, the corresponding node is added to the node list; at the same time, based on the establishment or non-establishment of the condition, the corresponding edge is added to the edge list to represent the logical relationship between nodes. The method establishes edge connections between multiple events and conditions based on the above information and process, and each generated node represents a condition or event, and each edge represents the logical connection between each node.

[0032] 2.3. Generation of Plan Flowcharts

[0033] The nodes and edges generated in the above steps are aggregated into node and edge lists. The node list stores all node information, and the edge list stores all edge information. The generation steps of the pre-plan flowchart automatically generate a flowchart with a clear structure based on the information in the two lists. During generation, the check flow function is called to verify the correctness of the process. After completion, the post-processing flow function is called to make final corrections to the generated flowchart to ensure the integrity and correctness of the flowchart. If the number of nodes is less than two, it indicates that the flowchart is incomplete. At this time, a default start node and an event node are added, and the connection between them is added to the edge list, that is, a start node and an event node are automatically completed to ensure the integrity of the flowchart. When outputting the flowchart, the check flow function is called to check and correct the structure of the generated flowchart to ensure that all nodes and edges are logically coherent and no connections are lost. During the check, the dictionaries of nodes and edges are extracted from the flowchart structure. Each node is indexed according to its number, and the edges connect different nodes through the start node and the end node. Each node is recursively checked by the check flow function. For conditional nodes, specifically check whether there are two branches of the node to ensure that each branch has a correct connection. After correction, the normalized flowchart structure is returned and output.

[0034] 3. Strict normalization correction of the pre-plan based on the reverse generation mechanism

[0035] The strict normalization correction steps of the pre-plan based on the reverse generation mechanism generate natural language in reverse according to the generated flowchart, and automatically correct the user's non-standard descriptions according to the rules to achieve the normalization correction of the pre-plan, ensuring that the generated pre-plan description meets the normalization requirements, mainly including three steps: the conversion between the flowchart and natural language, the correction of non-standard writing, and the output of normalized natural language.

[0036] 3.1. Conversion between the flowchart and natural language

[0037] The conversion steps between the flowchart and natural language obtain the already created nodes and edges by parsing the generated flowchart, and convert the content of each node and its edge relationship into natural language descriptions according to the rules. For example, if the node is a conditional node, a natural language description such as "If it is satisfied..." is generated; if the node is an event node, a description such as "Execute..." is generated; the edges represent the transfer relationship between steps, and these relationships are converted into descriptions such as "Jump to..." during reverse generation. The above steps can automatically extract readable text information from the graphical structure to ensure that the language expression meets the specifications.

[0038] 3.2. Correction of non-standard writing

[0039] The above-mentioned steps for correcting non-standard writing detect non-standard writing parts in the pre-plan through reverse generation. Using the flowchart-to-text function, after the flowchart is reversely converted into text, the method can compare the converted text with the text before conversion and put forward targeted correction suggestions. The correction suggestion generation rules are shown in Table 2.

[0040] Table 2 Writing Correction Strategy

[0041]

[0042] 3.3. Output of Standardized Natural Language

[0043] The above-mentioned steps for outputting standardized natural language ensure that the final pre-plan process description will be standardized strictly in accordance with the rules formulated by experts through reverse generation and correction. After correction and standardization, the natural language descriptions of all nodes and the jump descriptions of edges, etc. will be combined into a complete text form to form a pre-plan description. This makes the pre-plan process not only comply with the ontology constraints, but also meet the logical rigor and executability in practical applications.

Claims

1. A method for automatically generating a pre - plan flow chart based on a semi - normalization description system, characterized in that: The described method for automatically generating the pre - plan flowchart realizes the automatic conversion and standardized correction from the natural - language pre - plan to the standardized flowchart through ontology constraints, automatic process construction, and reverse generation mechanism, so as to improve the compilation efficiency and readability of the emergency plan. The described method for automatically generating the pre - plan flowchart includes the steps of pre - plan process description specification based on ontology constraints, automatic generation of the pre - plan flowchart based on semi - standardized description, and strict standardization and correction of the pre - plan based on the reverse generation mechanism.

2. The method for automatically generating a pre-plan flow chart according to claim 1, wherein: The step of pre - plan process description specification based on ontology constraints ensures the accuracy and standardization of the pre - plan process by clarifying the relationships among three key elements: events, conditions, and jumps. An event is defined as an action that occurs after meeting specific conditions and is extracted using regular expressions. A condition is defined to describe the prerequisite for an event to occur, and keyword matching is used to ensure its clear logic. A jump is defined to represent the transfer logic of the process. The association between events is managed through state machines and graph structures to construct a coherent and standardized pre - plan process description, providing an accurate semantic basis for the automatic generation of flowcharts.

3. The method for automatically generating a pre-plan flow chart according to claim 1, wherein: The step of automatically generating the pre - plan flowchart based on semi - standardized description parses the pre - plan text, automatically identifies events, conditions, and their logical relationships, and constructs a standardized flowchart. This step uses regular matching and structured identifier parsing of the pre - plan text to extract the events, conditions, and jumps of each step. Based on the relevance of conditions, events, and jumps, the connections between nodes and edges are established. After forming a complete process structure, a flowchart generation algorithm is used to automatically draw a pre - plan flowchart that meets the specifications, and rule checking and automatic correction are used to ensure the logic, integrity, and readability of the process.

4. The method for automatically generating the pre - plan flowchart according to claim 1, wherein: The step of strict standardization and correction of the pre - plan based on the reverse generation mechanism converts the automatically generated flowchart back into natural language and corrects the non - compliant expressions to ensure the standardization and consistency of the pre - plan description. After parsing the flowchart structure, extracting the relationships between nodes and edges, and generating a text description that meets the specifications, rule matching is used to detect non - standard expressions and generate correction suggestions. Finally, a standardized correction suggestion for the pre - plan text is output to ensure its clear logic and standard expression, thereby improving the readability and executability of the emergency plan.

Images