Intelligent decision-making method and system for coal mine tunneling operation regulations based on semantic analysis

Abstract

The application discloses a coal mine tunneling operation regulation intelligent decision method and system based on semantic analysis, and relates to the field of data processing, wherein the method comprises: preprocessing, semantic analysis and entity relationship extraction of a coal mine tunneling operation regulation text, and constructing a structured tunneling operation knowledge representation; integrating coal mine operation safety regulations, tunneling equipment operation specifications and historical tunneling operation data for knowledge mining and integration, and building a tunneling operation prior knowledge base; based on the knowledge base, performing logical reasoning and compliance verification on the knowledge representation, and generating a tunneling operation decision scheme set; driving the scheme set to perform three-dimensional modeling and process simulation, constructing a coal mine tunneling operation intelligent agent, and performing digital rehearsal and regulation optimization decision. The application solves the technical problems of insufficient multi-source information fusion and lack of decision accuracy in existing coal mine tunneling operation regulation decision, and achieves the technical effects of realizing deep fusion of multi-source heterogeneous information in the field of coal mine tunneling, and improving decision accuracy.

Claims

1. A semantic parsing-based intelligent decision-making method for coal mine tunneling operation procedures, characterized in that, The method includes: Input the text of the coal mine tunneling operation procedure, perform preprocessing, semantic parsing and entity relation extraction on the text of the coal mine tunneling operation procedure, and construct a structured tunneling operation knowledge representation; By integrating coal mine operation safety regulations, tunneling equipment operation specifications and historical tunneling operation data, a prior knowledge base for tunneling operations is built. Based on the prior knowledge base of tunneling operations, logical reasoning and compliance verification are performed on the structured tunneling operation knowledge representation to generate a set of tunneling operation decision schemes; The system drives the set of tunneling operation decision schemes to perform three-dimensional modeling and process simulation, constructs an intelligent agent for coal mine tunneling operations, and performs digital pre-simulation and process optimization decisions based on the intelligent agent for coal mine tunneling operations. Constructing a structured knowledge representation for tunneling operations, including: A procedure text preprocessing program is constructed, which includes irrelevant character removal, text encoding standardization, stop word filtering, and terminology standardization. The coal mine tunneling operation procedure text is preprocessed according to the procedure text preprocessing program to obtain a usable coal mine tunneling operation procedure text; Obtain a coal mine terminology database, and perform entity recognition and semantic parsing on the available coal mine tunneling operation procedure texts based on the coal mine terminology database to determine the entity set and semantic set of the coal mine tunneling operation procedure texts; Based on the entity set and semantic set of the coal mine tunneling operation procedure text, entity relations are extracted, stored, and represented to construct a structured tunneling operation knowledge representation. Establish a prior knowledge base for tunneling operations, including: The coal mine operation safety regulations, tunneling equipment operation specifications, and historical tunneling operation data are standardized, and knowledge mining and integration are performed to obtain a knowledge set of coal mine operation safety regulations, a knowledge set of tunneling equipment operation specifications, and a knowledge set of historical tunneling operation data. Rule mining is performed on the knowledge set of coal mine operation safety regulations and the knowledge set of tunneling equipment operation specifications to obtain a coal mine tunneling operation rule base; Feature annotation is performed on the knowledge set of tunneling equipment operation specifications and the knowledge set of historical tunneling operation data to obtain a tunneling equipment operation parameter library and a coal mine tunneling operation case library; Based on the coal mine tunneling operation rule library, the tunneling equipment operation parameter library, and the coal mine tunneling operation case library, a prior knowledge base for tunneling operations is constructed.

2. The intelligent decision-making method for coal mine tunneling operation procedures based on semantic parsing as described in claim 1, characterized in that, Determine the entity set and semantic set of the coal mine tunneling operation procedure text, including: The available coal mine tunneling operation procedure text is segmented and part-of-speech tagging is performed using natural language processing algorithms to obtain a coal mine tunneling operation procedure word segmentation set; Based on the coal mine terminology database, entity recognition is performed on the word segmentation set of the coal mine tunneling operation procedure to obtain the entity set of the coal mine tunneling operation procedure; A pre-trained language model is used to perform semantic parsing on the word segmentation set of the coal mine tunneling operation procedure based on the entity set of the coal mine tunneling operation procedure, so as to obtain the semantic set of the operation procedure text; Define a semantic template for tunneling operations, and match and fill the semantic set of the operation procedure text based on the semantic template to determine the semantic set of the coal mine tunneling operation procedure text.

3. The intelligent decision-making method for coal mine tunneling operation procedures based on semantic parsing as described in claim 1, characterized in that, Based on the entity set and semantic set of the coal mine tunneling operation procedure text, entity relations are extracted, stored, and represented to construct a structured tunneling operation knowledge representation, including: Define the entity relationship type for tunneling operations based on the characteristics of coal mine tunneling operations; According to the entity relationship type of the tunneling operation, the entity relationship of the text semantic set of the coal mine tunneling operation procedure is extracted to obtain the entity relationship set of the tunneling operation procedure; The entity set of the coal mine tunneling operation procedure is fused according to the entity relationship set of the tunneling operation procedure to obtain the entity knowledge set of the tunneling operation procedure; Select a knowledge representation format, store and represent the entity knowledge set of the tunneling operation procedure based on the knowledge representation format, and construct a structured tunneling operation knowledge representation.

4. The intelligent decision-making method for coal mine tunneling operation procedures based on semantic parsing as described in claim 1, characterized in that, Generate a set of tunneling operation decision schemes, including: Based on the coal mine tunneling operation rule base and the tunneling equipment operation parameter base, forward and reverse logical reasoning is performed on the structured tunneling operation knowledge representation to obtain a set of candidate tunneling operation schemes; Based on the coal mine tunneling operation case library, the candidate tunneling operation scheme set is matched for similarity, and the set of available tunneling operation schemes is selected as the best. The available tunneling operation scheme set is subjected to multi-dimensional compliance verification and conflict constraint adjustment to generate a tunneling operation decision scheme set.

5. The intelligent decision-making method for coal mine tunneling operation procedures based on semantic parsing as described in claim 4, characterized in that, Obtain a set of candidate tunneling operation plans, including: The structured tunneling operation knowledge representation is subjected to forward condition and reverse target extraction to obtain the forward scenario conditions and reverse scenario targets of the tunneling operation. Based on the coal mine tunneling operation rule library and the tunneling equipment operation parameter library, forward and reverse logical reasoning is performed on the forward scenario conditions and reverse scenario targets of the tunneling operation to obtain a set of forward candidate operation schemes and a set of reverse candidate operation schemes; The positive candidate operation scheme set and the negative candidate operation scheme set are merged and deduplicated to obtain the candidate tunneling operation scheme set.

6. The intelligent decision-making method for coal mine tunneling operation procedures based on semantic parsing as described in claim 1, characterized in that, Constructing an intelligent agent for coal mine tunneling operations includes: Based on the geological data and work environment data in the prior knowledge base of tunneling operations, three-dimensional modeling is performed to generate a three-dimensional geological model and a tunneling work environment model. The three-dimensional geological model and the tunneling operation environment model are driven to perform process simulation fitting on the tunneling operation decision scheme set to construct a coal mine tunneling operation intelligent agent.

7. The intelligent decision-making method for coal mine tunneling operation procedures based on semantic parsing as described in claim 1, characterized in that, Based on the aforementioned intelligent agent for coal mine tunneling operations, digital pre-simulation and process optimization decisions are performed, including: Based on the aforementioned intelligent agent for coal mine tunneling operations, digital pre-simulation and effect evaluation are conducted to obtain simulation parameters for the effect of coal mine tunneling operations. The coal mine tunneling operation effect simulation parameters are used to iteratively optimize the coal mine tunneling operation procedure text, and the optimized coal mine tunneling operation procedure text is used to make tunneling operation decisions.

8. A semantic parsing-based intelligent decision-making system for coal mine tunneling operation procedures, characterized in that, The system is used to implement the intelligent decision-making method for coal mine tunneling operation procedures based on semantic parsing as described in any one of claims 1-7, and the system comprises: The structured processing module is used to input the text of the coal mine tunneling operation procedure, perform preprocessing, semantic parsing and entity relation extraction on the text of the coal mine tunneling operation procedure, and construct a structured tunneling operation knowledge representation. The module for building a prior knowledge base for tunneling operations is used to integrate coal mine operation safety regulations, tunneling equipment operation specifications, and historical tunneling operation data for knowledge mining and integration, and to build a prior knowledge base for tunneling operations. The tunneling operation decision scheme set generation module is used to perform logical reasoning and compliance verification on the structured tunneling operation knowledge representation based on the tunneling operation prior knowledge base, and generate a tunneling operation decision scheme set; The procedure optimization decision module is used to drive the set of tunneling operation decision schemes to perform three-dimensional modeling and process simulation, construct a coal mine tunneling operation intelligent agent, and perform digital pre-simulation and procedure optimization decision based on the coal mine tunneling operation intelligent agent.

Images