CN-121997534-A - Mine emergency rescue command training guiding method and system based on rule engine

CN121997534ACN 121997534 ACN121997534 ACN 121997534ACN-121997534-A

Abstract

The application discloses a rule engine-based mine emergency rescue command training guiding method and system, which belong to the technical field of emergency rescue training and comprise the steps of obtaining mine rescue history cases, calculating case similarity of the history cases based on disaster condition features and environmental features, generating a case database according to calculation results, structuring the history cases in the case database, carrying out rule regulation on mine rescue rule terms, constructing a rule base, generating dynamic treatment flow guiding mechanisms according to various rules in the rule base, generating prefabricated typical case flow templates of different mine disasters based on the dynamic treatment flow guiding mechanisms, selecting the prefabricated typical case flow templates by a user, automatically loading corresponding rules and physical parameters by the system, generating corresponding mine emergency rescue treatment flows, realizing intelligent guiding of the mine emergency rescue command training treatment flows, improving the authenticity and effectiveness of training exercises, and improving the standardized operation capability of rescue workers.

Inventors

ZHANG PENG
WANG ZHE
MA LONG
MA HONGTAO
LIU ZENGHUI
XU WENMING
YANG DAHAI
DENG RONG
QI HUI
YAN JUN

Assignees

中煤科工集团沈阳研究院有限公司

Dates

Publication Date: 20260508
Application Date: 20251211

Claims (9)

1. The mine emergency rescue command training guiding method based on the rule engine is characterized by comprising the following steps of: acquiring a mine rescue historical case, and calculating the similarity of the mine rescue historical case to generate a case database; acquiring mine rescue rule terms, and constructing a rule base based on the case database and the mine rescue rule terms, wherein the rule base comprises a rescue rule sub-base, a rule word base and a physical simulation rule sub-base; generating prefabricated typical case flow templates of different mine disasters based on a rule base; a dynamic disposal flow guiding mechanism is established, and a mine emergency rescue disposal flow is generated based on the dynamic disposal flow guiding mechanism.
2. The rule engine-based mine emergency rescue command training guidance method of claim 1, wherein the mine rescue history cases are obtained, the similarity of the mine rescue history cases is calculated, and a case database is generated, and the specific method is as follows: acquiring a mine rescue history case, and extracting disaster condition characteristic parameters and environment characteristic parameters; The disaster characteristic parameters comprise disaster type, occurrence place, influence range and initial parameters, wherein the initial parameters comprise harmful gas initial concentration, explosion initial impact force, initial water inflow, initial water level rising speed, initial fire source temperature or initial fire coverage area; The environment characteristic parameters refer to mine self and surrounding environment indexes affecting disaster development, rescue path selection, rescue resource waste and physical simulation results when mine disasters occur, and comprise mine external geographic environment parameters and mine internal facility environment parameters, wherein the mine external geographic environment parameters comprise surrounding rock lithology of a roadway, buried depth of the roadway, thickness of a mineral layer, geological structure, surface topography type and surface water distribution; performing similarity calculation on the mine rescue historical case based on disaster condition characteristic parameters and environment characteristic parameters, and performing labeling treatment on the mine rescue historical case information according to calculation results; And (3) correlating the labeled mine rescue historical case information by using a knowledge graph technology, constructing different types of historical rescue case knowledge graphs, and carrying out weighted fusion on the multiple types of historical rescue case knowledge graphs to generate a case database.
3. The mine emergency rescue command training guiding method based on the rule engine as claimed in claim 1, wherein the physical simulation rule sub-library construction method is as follows: Based on the environmental parameters of the facilities in the mine, constructing a three-dimensional tunnel model of the mine, combining physical theory related to mine disasters, and generating physical simulation rules, wherein the method comprises the following steps: Simulating the tunnel wind speed and flow direction by adopting a one-dimensional incompressible fluid equation to generate a wind flow rule; Dynamically calculating concentration distribution of different gases based on a Gaussian plume model to generate a gas diffusion rule; predicting a fire spreading path according to a heat conduction equation and fuel distribution, and generating a fire spreading rule; and adding the physical simulation rules into a physical simulation rule sub-library.
4. The mine emergency rescue command training guiding method based on the rule engine according to claim 1, wherein the construction method of the rescue rule sub-base is as follows: Classifying and extracting mine rescue historical cases from a case database, acquiring record data in each type of mine rescue historical cases, simulating an operation flow in a rescue process, and generating rescue rules according to the operation flow; The recorded data comprise an operation flow, disaster condition characteristic parameters and environment characteristic parameters in a historical rescue case; Constructing a three-dimensional dynamic simulation system according to the physical simulation rules, the mine three-dimensional roadway model, disaster condition characteristic parameters and environment characteristic parameters, importing the generated rescue rules into the three-dimensional dynamic simulation system, carrying out full-flow reproduction on the rescue rules by generating virtual accident scenes, analyzing the matching degree of the rescue rule execution results and the actual rescue results by adopting a confusion matrix, calculating the deviation rate, judging the rescue rules as failure rules when the deviation rate exceeds a preset threshold, and carrying out topological structure optimization on the failure rules by using a rule correction algorithm to obtain optimized rescue rules; The method comprises the following steps of: Receiving failure rule identification and deviation data, sending the failure rule identification to an expert knowledge base, inputting failure rule parameters and deviation data into a Bayesian network model, returning compliance constraint and generating a correction direction based on the failure rule identification by using the expert knowledge base; Combining the compliance constraint, the correction direction and the quantized correction suggestion to generate a correction optimization scheme which meets the compliance and has the highest prediction success rate; And carrying out topology result optimization on the failure rule by using a correction optimization scheme which meets compliance and has the highest prediction success rate to obtain an optimized rescue rule, and adding the optimized rescue rule and the rescue rule with the deviation rate not exceeding a preset threshold value into a rescue rule sub-library.
5. The mine emergency rescue command training guiding method based on the rule engine according to claim 1, wherein the rule sub-base construction method is as follows: Each rule term in the mine rescue rule terms is disassembled into executable logic with a clear logic relationship, the executable logic is converted into an atomization logic unit, the atomization logic unit is matched and conflict resolved by adopting a RETE algorithm, and rule rules are established; constructing a mine type tag system, calculating matching degrees of the atomization logic unit and different mine types by using a cosine similarity algorithm based on an atomization logic unit of mine rescue rule clauses, and generating a visual suitability thermodynamic diagram; matching and screening preset thresholds in the visual suitability thermodynamic diagram, adding warning marks for rule rules with matching degree lower than the preset thresholds, performing active shielding, and selecting a combination of rule rules with matching degree higher than the preset thresholds in the mine rescue cases of the same type; And adding the combination of the rule rules higher than the preset threshold value in the same type of mine rescue cases into a rule sub-library.
6. The rule engine-based mine emergency rescue command training guidance method of claim 1, wherein the treatment flow template comprises treatment steps, required resources, corresponding rules and physical parameters.
7. The rule engine-based mine emergency rescue command training guidance method of claim 1, wherein establishing a dynamic treatment flow guidance mechanism comprises: establishing a rescue resource database, wherein the rescue resource database comprises rescue team qualification, equipment performance parameters and material reserve; When a user receiving training selects a prefabricated typical case flow template, matching degree of resources required by the prefabricated typical case flow template selected by the user and actual adjustable resources is compared, if the matching degree is lower than a preset threshold value, a substitution template is generated based on rescue rules in a rule base, corresponding rules and physical parameters are loaded, and a corresponding mine emergency rescue treatment flow is generated; The dynamic treatment flow guiding mechanism also comprises a closed-loop optimization module, which is used for comparing the difference data of the historical optimal rescue cases in the mine case database with the mine emergency rescue treatment flow of the user after training is finished, establishing a difference analysis matrix, and generating targeted improvement suggestions by combining the rule base and the related knowledge in the case knowledge map, wherein the specific method comprises the following steps: The method comprises the steps of extracting key operation data of a historical optimal case from a case knowledge graph, wherein the key operation data comprise time sequence data, resource scheduling data and decision parameter data, the time sequence data comprise execution time of each treatment step and connection time between steps, the resource scheduling data comprise calling sequences of various resources and resource consumption, the decision parameter data comprise decision threshold values and operation logic of key nodes, and the key operation data are normalized and quantized to generate a historical optimal case matrix.
8. The rule engine-based mine emergency rescue command training guidance method of claim 7, wherein the mine emergency rescue treatment procedure is generated based on a dynamic treatment procedure guidance mechanism, and the specific method is as follows: Generating a historical optimal case matrix and a user operation matrix based on a dynamic treatment flow guiding mechanism, wherein the method for generating the user operation matrix is the same as the method for generating the historical optimal case matrix, and the user operation matrix is generated through the mine emergency rescue treatment flow of the user; mine emergency rescue treatment procedures for training guidance are generated based on the improvement advice.
9. The mine emergency rescue command training guidance system based on the rule engine is used for conducting mine emergency rescue command training guidance based on the method of claim 1 and is characterized by comprising a data storage module, a rule generation module, a template establishment module, a flow generation module and a closed loop optimization module; the data storage module is used for storing a mine rescue historical case, a historical rescue case knowledge graph, disaster condition characteristic parameters and environment characteristic parameters; The rule generation module is used for acquiring mine rescue rule terms, and constructing a rule base based on the case database and the mine rescue rule terms, wherein the rule base comprises a rescue rule sub-base, a rule word base and a physical simulation rule sub-base; The template building module is used for generating prefabricated typical case flow templates of different mine disasters; The flow generation module is used for loading corresponding rules and physical parameters according to the prefabricated typical case flow template selected by the user to generate a mine emergency rescue disposal flow; The closed-loop optimization module is used for comparing the difference data of the user operation and the historical optimal case after training is finished, establishing a difference analysis matrix, and combining the difference data with the rule base and the related knowledge in the case knowledge graph to generate a targeted improvement suggestion.

Description

Mine emergency rescue command training guiding method and system based on rule engine Technical Field The application belongs to the technical field of emergency rescue training, and particularly relates to a mine emergency rescue command training guiding method and system based on a rule engine. Background The mine emergency rescue is a high-risk operation, needs rescue workers to have strong comprehensive capacity in physical, psychological, knowledge, experience and the like, and is an important guarantee for safely and efficiently carrying out rescue and reducing accident loss. The rescue training is a basic way for improving the capability of rescue workers, and the main accident types of mine emergency rescue include natural environment disasters such as underground coal mine water penetration, fire, gas, coal dust explosion, roof caving and the like. However, the conventional mine emergency drilling system has the problem of disjointing the regulation and drilling flow, and lacks the automatic analysis and execution capability of regulation terms, so that the requirements of the rescue flow and the regulation are not met, and the rescue personnel are difficult to quickly master the standardized operation; meanwhile, the historical rescue cases are stored in static scripts, are not deeply bound with a physical engine and rules, cannot provide references for current training exercise through intelligent matching, have low multiplexing rate, are insufficient in physical simulation reality, are excessively simplified in physical simulation of a mine environment by a traditional training system, have larger deviation from actual catastrophe evolution rules, and influence the reliability of training exercise. Disclosure of Invention Aiming at the defects of the prior art, on the one hand, the invention provides a mine emergency rescue command training guiding method based on a rule engine, which comprises the following steps: acquiring a mine rescue historical case, and calculating the similarity of the mine rescue historical case to generate a case database; acquiring mine rescue rule terms, and constructing a rule base based on the case database and the mine rescue rule terms, wherein the rule base comprises a rescue rule sub-base, a rule word base and a physical simulation rule sub-base; generating prefabricated typical case flow templates of different mine disasters based on a rule base; a dynamic disposal flow guiding mechanism is established, and a mine emergency rescue disposal flow is generated based on the dynamic disposal flow guiding mechanism. Further, acquiring a mine rescue history case, calculating the similarity of the mine rescue history case, and generating a case database, wherein the concrete method comprises the following steps: acquiring a mine rescue history case, and extracting disaster condition characteristic parameters and environment characteristic parameters; The disaster characteristic parameters comprise disaster type, occurrence place, influence range and initial parameters, wherein the initial parameters comprise harmful gas initial concentration, explosion initial impact force, initial water inflow, initial water level rising speed, initial fire source temperature or initial fire coverage area; The environment characteristic parameters refer to mine self and surrounding environment indexes affecting disaster development, rescue path selection, rescue resource waste and physical simulation results when mine disasters occur, and comprise mine external geographic environment parameters and mine internal facility environment parameters, wherein the mine external geographic environment parameters comprise surrounding rock lithology of a roadway, buried depth of the roadway, thickness of a mineral layer, geological structure, surface topography type and surface water distribution; performing similarity calculation on the mine rescue historical case based on disaster condition characteristic parameters and environment characteristic parameters, and performing labeling treatment on the mine rescue historical case information according to calculation results; And (3) correlating the labeled mine rescue historical case information by using a knowledge graph technology, constructing different types of historical rescue case knowledge graphs, and carrying out weighted fusion on the multiple types of historical rescue case knowledge graphs to generate a case database. Further, the construction method of the physical simulation rule sub-library comprises the following steps: Based on the environmental parameters of the facilities in the mine, constructing a three-dimensional tunnel model of the mine, combining physical theory related to mine disasters, and generating physical simulation rules, wherein the method comprises the following steps: Simulating the tunnel wind speed and flow direction by adopting a one-dimensional incompressible fluid equation to generate a wind flow rule; Dyna