CN-121998498-A - Multi-dimensional index comprehensive score management method and system for fire control training

CN121998498ACN 121998498 ACN121998498 ACN 121998498ACN-121998498-A

Abstract

The invention discloses a fire-fighting training multidimensional index comprehensive scoring management method and system, in particular relates to the technical field of fire-fighting training management, and aims to solve the problem of scoring attribution distortion caused by neglecting causal relation among cooperative actions in the existing scoring method; the method comprises the steps of constructing a knowledge graph representing causal conduction association between actions by acquiring individual action data of firefighters and team task results in real fire smoke thermal simulation training, identifying a key causal chain in the knowledge graph based on mutual exclusion constraint occupied by resources and consistency constraint of tactical targets, performing Monte Carlo disturbance simulation on an execution process of the key causal chain to generate a flow sensitivity index, evaluating contribution weights of all actions on the key causal chain to team results based on the index, and finally calculating multidimensional comprehensive scores of firefighters and teams according to contribution weight integration, so that more accurate evaluation of collaborative training effects is realized, and data support is provided for optimizing training schemes.

Inventors

SUN MING
SUN LIANG
LI CHANGHAI
XIA HUALONG

Assignees

长春市蓝程科技发展有限公司

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1. A fire training multidimensional index comprehensive score management method is characterized by comprising the following steps: S1, acquiring individual action behavior data and team task results of a firefighter in real fire smoke thermal simulation training; S2, judging whether causal conduction association exists between individual action behaviors of the firefighters based on individual action behavior data of the firefighters, if so, constructing a knowledge graph taking the individual action behaviors of the firefighters as nodes and taking the causal conduction association between the individual action behaviors of the firefighters as edges; s3, identifying a key causal chain from individual action behaviors of firefighters to team task results based on mutual exclusion constraint of resource occupation and consistency constraint of tactical targets in a knowledge graph; S4, carrying out Monte Carlo disturbance simulation on the execution process of the key causal chain to quantify the sensitivity degree of the key causal chain to the execution disturbance, and generating a flow sensitivity index; S5, evaluating contribution weights of individual action behaviors of firefighters on the key causal chain to team task results based on the flow sensitivity index; And S6, comprehensively calculating individual multidimensional comprehensive scores of firefighters and multidimensional comprehensive scores of teams according to the contribution weights.
2. The fire training multidimensional index comprehensive score management method as claimed in claim 1, wherein S1 includes: acquiring individual action behavior data of the firefighters through a training facility sensor and an audio/video recording device; recording team task results through a training management terminal; And performing time stamp alignment and associated storage on individual action behavior data of the firefighters and team task results.
3. The fire training multidimensional index comprehensive score management method as claimed in claim 1, wherein S2 includes: analyzing the dependency relationship between actions in individual action behavior data of firefighters; Judging whether a causal conduction association exists between the later action and the condition created by the former action according to the time sequence relationship; if causal conduction association exists, constructing a knowledge graph by taking individual action behaviors of firefighters as nodes; and in the knowledge graph, connecting corresponding individual action behavior nodes of the firefighters by taking the causal conduction association which is judged to exist as the edge.
4. The fire training multidimensional index comprehensive score management method as claimed in claim 1, wherein S3 includes: Traversing all causal conduction paths from individual action behavior nodes of firefighters to team task results in the knowledge graph; Screening causal conduction paths which meet the requirement that no resource occupation conflict exists between individual action behaviors of firefighters in time and space dimensions; Screening causal conduction paths in which tactical intentions of individual action behaviors of firefighters all point to the same team task result from causal conduction paths meeting the constraint of resource occupation mutual exclusion; Causal conduction paths meeting tactical objective compliance constraints are identified as key causal links.
5. The method for managing the comprehensive scores of the fire training multidimensional indexes according to claim 4, wherein screening the causal conduction paths in which tactical intentions of individual action behaviors of firefighters all point to the same team task result comprises analyzing a top-level tactical goal according to the team task result set by training, mapping each individual action behavior of firefighters in the causal conduction paths to a preset standard operation Cheng Xuku to obtain the corresponding standard tactical intentions, judging whether the standard tactical intentions corresponding to the individual action behaviors of all firefighters support the top-level tactical goal together, and reserving the causal conduction paths in which the standard tactical intentions support the top-level tactical goal together.
6. The fire training multidimensional index comprehensive score management method as claimed in claim 1, wherein S4 includes: defining a time disturbance parameter and a success rate disturbance parameter for individual action behaviors of firefighters contained in the key causal chain; based on the defined time disturbance parameter and success rate disturbance parameter, carrying out random sampling simulation for a plurality of times on the execution process of the key causal chain; Counting the ratio of the team task results which cannot be achieved in the random sampling simulation for a plurality of times; a process sensitivity index is generated based on the proportional calculations that the team task results fail to achieve.
7. The method for managing the comprehensive scores of the fire training multidimensional indexes according to claim 6 is characterized in that the method comprises the steps of carrying out random sampling simulation on the execution process of a key causal chain for a plurality of times based on defined time disturbance parameters and success rate disturbance parameters, including superposing a random time delay obeying specific distribution on the reference execution time of each individual action behavior of the key causal chain to simulate time disturbance, generating a random number for each individual action behavior of the firefighter according to the reference success rate of the individual action behavior of the firefighter to judge whether the action in the simulation is successful to simulate the success rate disturbance, and sequentially deducing the execution process of the whole key causal chain and the final team task result based on the randomly generated time delay and action success state.
8. The fire training multidimensional index comprehensive score management method as claimed in claim 1, wherein S5 includes: analyzing the flow sensitivity index to determine a disturbance sensitivity value of the key causal link; Setting weight distribution proportion of individual action behaviors of firefighters on the key causal chain according to the disturbance sensitivity value of the key causal chain; and calculating the contribution weight of each individual action behavior of the firefighters on the key causal chain to the team task result according to the weight distribution proportion.
9. The fire training multidimensional index comprehensive score management method as claimed in claim 1, wherein S6 includes: summarizing contribution weights of individual action behaviors of firefighters on a key causal chain for each firefighter to obtain individual contribution weight sum; Calculating individual multidimensional comprehensive scores of firefighters according to the sum of individual contribution weights; And integrating individual multidimensional comprehensive scores of all firefighters to obtain a team multidimensional comprehensive score.
10. A fire training multidimensional index comprehensive score management system for implementing a fire training multidimensional index comprehensive score management method as claimed in any one of claims 1 to 9, comprising: the data acquisition module is used for acquiring individual action behavior data of firefighters and team task results in real fire smoke thermal simulation training; The association judging module is used for judging whether causal conduction association exists between individual action behaviors of the firefighters based on individual action behavior data of the firefighters, if so, a knowledge graph taking the individual action behaviors of the firefighters as nodes and taking the causal conduction association between the individual action behaviors of the firefighters as edges is constructed; The causal chain identification module is used for identifying a key causal chain from individual action behaviors of firefighters to team task results based on mutual exclusion constraint of resource occupation and consistency constraint of tactical targets in the knowledge graph; The disturbance simulation module is used for carrying out Monte Carlo disturbance simulation on the execution process of the key causal chain to quantify the sensitivity degree of the key causal chain to the execution disturbance and generating a flow sensitivity index; the weight evaluation module is used for evaluating the contribution weight of individual action behaviors of firefighters on the key causal chain to the team task result based on the flow sensitivity index; and the scoring calculation module is used for comprehensively calculating individual multidimensional comprehensive scores and team multidimensional comprehensive scores of firefighters according to the contribution weights.

Description

Multi-dimensional index comprehensive score management method and system for fire control training Technical Field The invention relates to the technical field of fire training management, in particular to a method and a system for managing multi-dimensional index comprehensive scoring of fire training. Background The real fire smoke heat (CFBT) simulation training is one of core forms of actual combat training in the fire control field, the multi-person cooperative treatment is an important scene of the training, and the core aim is to improve the emergency rescue cooperative capacity of a fire control team by simulating the labor division cooperation in the real disaster environment. In a fire training management system, score management aiming at CFBT simulation training is a key link for supporting training effect evaluation, capability short-board recognition and training scheme optimization, and belongs to the important application category of personnel training and team workflow management. The existing related score management method is used for expanding independent quantitative accounting around dominant indexes such as individual operation specifications, single task completion conditions and the like, and individual and team scores are formed in a simple superposition or average mode so as to meet the basic requirements of data evaluation of training effects in fire training management. When the conventional fire control training scoring management method is applied to CFBT multi-person collaborative training scenes, causal conduction relations existing between individual action behaviors of firefighters in the collaborative process cannot be fully considered, scoring accounting is carried out only by means of individual independent index scores and team overall task results, so that actual influence weights of action behaviors of different firefighters on training effects cannot be accurately defined by the scoring results, further the problem of scoring attribution distortion is generated, the real level of team collaborative training cannot be objectively and comprehensively reflected, and accurate and effective data support cannot be provided for optimizing collaborative training schemes in a targeted mode and improving the overall emergency rescue capability of the team. Disclosure of Invention In order to overcome the defects in the prior art, the invention provides a multi-dimensional index comprehensive scoring management method and system for fire training, which are used for solving the problems in the prior art. In order to achieve the above purpose, the present invention provides the following technical solutions: a fire training multidimensional index comprehensive score management method comprises the following steps: S1, acquiring individual action behavior data and team task results of a firefighter in real fire smoke thermal simulation training; S2, judging whether causal conduction association exists between individual action behaviors of the firefighters based on individual action behavior data of the firefighters, if so, constructing a knowledge graph taking the individual action behaviors of the firefighters as nodes and taking the causal conduction association between the individual action behaviors of the firefighters as edges; s3, identifying a key causal chain from individual action behaviors of firefighters to team task results based on mutual exclusion constraint of resource occupation and consistency constraint of tactical targets in a knowledge graph; S4, carrying out Monte Carlo disturbance simulation on the execution process of the key causal chain to quantify the sensitivity degree of the key causal chain to the execution disturbance, and generating a flow sensitivity index; S5, evaluating contribution weights of individual action behaviors of firefighters on the key causal chain to team task results based on the flow sensitivity index; And S6, comprehensively calculating individual multidimensional comprehensive scores of firefighters and multidimensional comprehensive scores of teams according to the contribution weights. Further, S1 includes: acquiring individual action behavior data of the firefighters through a training facility sensor and an audio/video recording device; recording team task results through a training management terminal; And performing time stamp alignment and associated storage on individual action behavior data of the firefighters and team task results. Further, S2 includes: analyzing the dependency relationship between actions in individual action behavior data of firefighters; Judging whether a causal conduction association exists between the later action and the condition created by the former action according to the time sequence relationship; if causal conduction association exists, constructing a knowledge graph by taking individual action behaviors of firefighters as nodes; and in the knowledge graph, connecting corresponding indivi