Abstract: An active control method and device for large deformation of a deep thin-bedded surrounding rock, the active control method includes: drill holes in the surrounding rock of a tunnel against a tunnel face, add the active control device containing a sleeve, a high-strength prestressed anchor bolt, a resin anchoring agent and a grouting device into drill holes, apply a preload to the anchor bolt when the resin anchoring agent has a certain strength, and carry out lag grouting in the surrounding rock through the grouting device after stress adjustment, and the high-strength prestressed anchor bolt and the grouting device are inserted in the sleeve with holes on the side. The active control device or method of the invention can effectively reduce the fracture depth and degree of a thin-bedded soft rock in deep engineering, and effectively inhibit the occurrence of large deformation disasters.

Abstract: A method for extracting a dam emergency event based on a dual attention mechanism is provided. The method includes: performing data preprocessing, building a dependency graph, building a dual attention network, and filling a document-level argument. The performing data preprocessing includes labeling a dam emergency corpus and encoding sentences. Building a dependency graph includes assisting a model to mine a syntactic relation based on a dependency. Building a dual attention network includes weighing and fusing an attention network based on a graph transformer network (GTN) and capturing key semantic information in the sentence. Filling a document-level argument includes filling a document-level argument by detecting a key sentence and ordering a semantic similarity. The method introduces a dependency and overcomes the long-range dependency problem based on the dual attention mechanism, thus achieving high identification accuracy and reducing a lot of labor costs.

Abstract: A method for extracting a dam emergency event based on a dual attention mechanism is provided. The method includes: performing data preprocessing, building a dependency graph, building a dual attention network, and filling a document-level argument. The performing data preprocessing includes labeling a dam emergency corpus and encoding sentences. Building a dependency graph includes assisting a model to mine a syntactic relation based on a dependency. Building a dual attention network includes weighing and fusing an attention network based on a graph transformer network (GTN) and capturing key semantic information in the sentence. Filling a document-level argument includes filling a document-level argument by detecting a key sentence and ordering a semantic similarity. The method introduces a dependency and overcomes the long-range dependency problem based on the dual attention mechanism, thus achieving high identification accuracy and reducing a lot of labor costs.