An emergency exercise evaluation method and system based on data fusion

Abstract

The application relates to the technical field of scheme evaluation, and provides an emergency drill evaluation method and system based on data fusion, which comprises the following steps: obtaining target emergency drill process information; performing node division on a target emergency drill scheme to obtain N target emergency drill nodes; obtaining N groups of node evaluation factors; performing weight analysis to obtain a factor weight analysis result; performing data fusion based on the N groups of node evaluation factors and the factor weight analysis result to obtain a characteristic drill evaluation model; obtaining real-time emergency drill information; inputting the real-time emergency drill information into the characteristic drill evaluation model to obtain a real-time emergency drill evaluation result. The method can solve the problem that an evaluation result is inaccurate due to unreasonable evaluation index setting in an emergency drill scheme evaluation process, can effectively and reasonably evaluate an emergency drill scheme, can compare and summarize shortcomings of the scheme, and can strengthen emergency management capability of an emergency event.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of scheme evaluation, in particular to an emergency drill evaluation method and system based on data fusion. BACKGROUND

[0002] Emergency drills take virtual emergencies as drill targets, simulate the disposal process of real rescue, and emergency drill evaluation can better summarize the content of emergency drills and enhance the emergency response capability of emergencies.

[0003] At present, there are many studies on emergency drill evaluation problems in China, however, most of the emergency drills do not combine the corresponding emergency plans, select appropriate methods for evaluation, and evaluators only make rough judgments on the corresponding content, without a set of perfect indicators for evaluation, resulting in inaccurate evaluation results and threatening the safety of people's life and property.

[0004] In summary, the prior art has the problem of inaccurate evaluation results due to unreasonable evaluation index setting in the emergency drill scheme evaluation process. SUMMARY

[0005] Therefore, it is necessary to provide an emergency drill evaluation method and system based on data fusion in view of the above technical problems.

[0006] An emergency drill evaluation method based on data fusion, the method comprises: performing feature recognition based on a target emergency drill scheme to obtain target emergency drill process information; performing node division on the target emergency drill scheme based on the target emergency drill process information to obtain N target emergency drill nodes, wherein N is a positive integer greater than 1; performing data mining based on the N target emergency drill nodes to obtain N groups of node evaluation factors; traversing the N groups of node evaluation factors to obtain factor weight analysis results; performing data fusion based on the N groups of node evaluation factors and the factor weight analysis results to obtain a feature drill evaluation model; performing real-time emergency drill based on the target emergency drill scheme to obtain real-time emergency drill information; inputting the real-time emergency drill information into the feature drill evaluation model to obtain real-time emergency drill evaluation results.

[0007] In one embodiment, it further comprises: traversing the N target emergency drill nodes to obtain a first target emergency drill node; performing feature extraction on the target emergency drill scheme based on the first target emergency drill node to obtain first node emergency drill information; performing drill factor extraction based on the first node emergency drill information to obtain a plurality of first drill factors; performing evaluation feature analysis based on the plurality of first drill factors to obtain a first group of node evaluation factors; adding the first group of node evaluation factors to the N groups of node evaluation factors.

[0008] In one embodiment, further comprising: performing principal component analysis based on the first node emergency drill information to obtain node standard emergency drill information; performing drill factor extraction based on the node standard emergency drill information to obtain a plurality of node emergency drill factors; and obtaining the plurality of first drill factors according to the plurality of node emergency drill factors.

[0009] In one embodiment, further comprising: obtaining a first node emergency drill feature data set according to the first node emergency drill information; performing decentralized processing on the first node emergency drill feature data set to obtain a second node emergency drill feature data set; obtaining a covariance matrix of a first node emergency drill feature according to the second node emergency drill feature data set; obtaining a first eigenvalue and a first eigenvector according to the covariance matrix of the first node emergency drill feature; and obtaining the node standard emergency drill information according to the first eigenvalue and the first eigenvector.

[0010] In one embodiment, further comprising: setting a drill factor as a retrieval feature and an evaluation index as a retrieval target; performing big data query based on the retrieval feature and the retrieval target to obtain a plurality of sample drill factors and a plurality of sample evaluation indexes; performing mapping relationship analysis based on the plurality of sample drill factors and the plurality of sample evaluation indexes to obtain a feature sample mapping relationship; obtaining an evaluation feature analysis model according to the plurality of sample drill factors and the plurality of sample evaluation indexes based on the feature sample mapping relationship; and inputting the plurality of first drill factors into the evaluation feature analysis model to obtain the first group of node evaluation factors.

[0011] In one embodiment, further comprising: setting weights of the N groups of node evaluation factors based on an emergency drill evaluation expert group to obtain an initial factor weight analysis result; performing value degree analysis based on the N groups of node evaluation factors to obtain a factor value degree analysis result; and performing weighted calculation on the initial factor weight analysis result based on the factor value degree analysis result to obtain the factor weight analysis result.

[0012] In one embodiment, further comprising: obtaining an emergency drill evaluation constraint condition; judging whether the real-time emergency drill evaluation result meets the emergency drill evaluation constraint condition; and generating an emergency drill early warning signal if the real-time emergency drill evaluation result meets the emergency drill evaluation constraint condition.

[0013] An emergency drill evaluation system based on data fusion, comprising:

[0014] A feature recognition module, the feature recognition module being configured to perform feature recognition based on a target emergency drill scheme to obtain target emergency drill process information;

[0015] a node division module, configured to divide the target emergency drill scheme based on the target emergency drill process information, to obtain N target emergency drill nodes, wherein N is a positive integer greater than 1;

[0016] a data mining module, configured to mine data based on the N target emergency drill nodes, to obtain N sets of node evaluation factors;

[0017] a weight analysis module, configured to analyze weights by traversing the N sets of node evaluation factors, to obtain a factor weight analysis result;

[0018] a characteristic drill evaluation model obtaining module, configured to perform data fusion based on the N sets of node evaluation factors and the factor weight analysis result, to obtain a characteristic drill evaluation model;

[0019] a real-time emergency drill module, configured to perform real-time emergency drill based on the target emergency drill scheme, to obtain real-time emergency drill information;

[0020] a real-time emergency drill evaluation result obtaining module, configured to input the real-time emergency drill information into the characteristic drill evaluation model, to obtain a real-time emergency drill evaluation result.

[0021] The above-mentioned emergency drill evaluation method and system based on data fusion can solve the problem of inaccurate evaluation results caused by unreasonable evaluation indexes in the evaluation process of an emergency drill scheme. By performing feature recognition based on a target emergency drill scheme, target emergency drill process information is obtained. The target emergency drill scheme is divided based on the target emergency drill process information, to obtain N target emergency drill nodes, wherein N is a positive integer greater than 1. Data mining is performed based on the N target emergency drill nodes, to obtain N sets of node evaluation factors. The N sets of node evaluation factors are traversed to analyze weights, to obtain a factor weight analysis result. Data fusion is performed based on the N sets of node evaluation factors and the factor weight analysis result, to obtain a characteristic drill evaluation model. Real-time emergency drill is performed based on the target emergency drill scheme, to obtain real-time emergency drill information. The real-time emergency drill information is input into the characteristic drill evaluation model, to obtain a real-time emergency drill evaluation result. The emergency drill scheme can be effectively and reasonably evaluated, and deficiencies in the scheme can be compared and summarized, to strengthen the emergency management capability for emergencies.

[0022] The above description is only a summary of the technical solutions of the present application. In order to make the technical means of the present application more clear, it can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 A flowchart of an emergency exercise evaluation method based on data fusion is provided for the present application;

[0024] Figure 2 A flowchart of obtaining N groups of node evaluation factors in an emergency exercise evaluation method based on data fusion is provided for the present application;

[0025] Figure 3 A flowchart of obtaining factor weight analysis results in an emergency exercise evaluation method based on data fusion is provided for the present application;

[0026] Figure 4 A structural diagram of an emergency exercise evaluation system based on data fusion is provided for the present application.

[0027] Reference signs: feature recognition module 1, node division module 2, data mining module 3, weight analysis module 4, feature exercise evaluation model obtaining module 5, real-time emergency exercise module 6, real-time emergency exercise evaluation result obtaining module 7. DETAILED DESCRIPTION

[0028] In order to make the purposes, technical solutions and advantages of the present application more clear, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0029] As shown in Figure 1 The present application provides an emergency exercise evaluation method based on data fusion, which comprises:

[0030] Step S100: feature recognition based on a target emergency exercise scheme to obtain target emergency exercise process information;

[0031] Step S200: node division based on the target emergency exercise process information to the target emergency exercise scheme to obtain N target emergency exercise nodes, wherein N is a positive integer greater than 1;

[0032] Specifically, a target emergency drill plan is obtained, which refers to a snowy weather traffic emergency drill plan. The target emergency drill plan is then subjected to process identification to obtain multiple process information segments, including five stages: early warning management, emergency preparedness, emergency response, summary and evaluation, and post-disaster handling. Based on the target emergency drill process information, the target emergency drill plan is divided into nodes, resulting in N target emergency drill nodes, where N is a positive integer greater than 1. Each target emergency drill node is a connection point among the multiple processes of the target emergency drill plan.

[0033] Step S300: Perform data mining based on the N target emergency drill nodes to obtain N sets of node evaluation factors;

[0034] like Figure 2 As shown, in one embodiment, step S300 of this application further includes:

[0035] Step S310: Traverse the N target emergency drill nodes to obtain the first target emergency drill node;

[0036] Step S320: Based on the first target emergency drill node, perform feature extraction on the target emergency drill plan to obtain the first node emergency drill information;

[0037] Specifically, a first target emergency drill node is extracted from the N target emergency drill nodes. This first target emergency drill node is any one of the N target emergency drill nodes. Then, each of the N target emergency drill nodes is extracted sequentially. Based on the first target emergency drill node, drill information is extracted from the target emergency drill plan to obtain the first node emergency drill information. The first node drill information is the specific content of the target emergency drill plan corresponding to the first target emergency drill node. For example, if the process information corresponding to the first target node is early warning management, then the first node emergency drill information includes information receiving methods, information reception, contact processing, information reporting, on-duty command, and early warning issuance.

[0038] Step S330: Extract exercise factors based on the emergency exercise information of the first node to obtain multiple first exercise factors;

[0039] In one embodiment, step S330 of this application further includes:

[0040] Step S331: Perform principal component analysis based on the emergency drill information of the first node to obtain the standard emergency drill information of the node;

[0041] In one embodiment, step S331 of this application further includes:

[0042] Step S3311: obtaining a first node emergency drill feature data set according to the first node emergency drill information;

[0043] Step S3312: performing decentralized processing on the first node emergency drill feature data set to obtain a second node emergency drill feature data set;

[0044] Step S3313: obtaining a covariance matrix of the first node emergency drill feature according to the second node emergency drill feature data set;

[0045] Step S3314: obtaining a first eigenvalue and a first eigenvector according to the covariance matrix of the first node emergency drill feature;

[0046] Step S3315: obtaining the node standard emergency drill information according to the first eigenvalue and the first eigenvector.

[0047] Specifically, the principal component analysis is performed on the first node emergency drill information. First, the feature data in the first node emergency drill information is subjected to data standardization processing, and a feature data set matrix is constructed to obtain the first node emergency drill feature data set. Then, each feature data in the first node emergency drill feature data set is subjected to decentralized processing. The decentralized processing refers to solving the average value of each feature in the first node emergency drill feature data set, and then subtracting the average value of each feature from itself for all samples, and then obtaining new feature values. The second node emergency drill feature data set is a data matrix composed of new feature values. The covariance formula is used to operate on the second node emergency drill feature data set to obtain the covariance matrix of the first node emergency drill feature of the second node emergency drill feature data set. Then, the eigenvalue and eigenvector of the covariance matrix of the first node emergency drill feature are obtained through matrix operation, and each eigenvalue corresponds to an eigenvector. In the obtained first eigenvector, the first K largest eigenvalues and the corresponding eigenvectors are selected, and the original features in the first node emergency drill feature data set are projected onto the selected eigenvectors to obtain the node standard emergency drill information after dimension reduction. The node standard emergency drill information includes information acceptance, contact processing, information reporting, on-site command, and warning release. Through the principal component analysis on the first node emergency drill information, the feature data in the first node emergency drill information can be reduced in dimension. On the premise of ensuring the information amount, the redundant data is eliminated, so that the sample amount of the feature data in the first node emergency drill information is reduced, thereby accelerating the operation speed of the training model for data.

[0048] Step S332: performing exercise factor extraction based on the node standard emergency exercise information to obtain a plurality of node emergency exercise factors;

[0049] Step S333: obtaining the plurality of first exercise factors according to the plurality of node emergency exercise factors.

[0050] Specifically, the node standard emergency exercise information is subjected to exercise factor extraction to obtain a plurality of node emergency exercise factors, the emergency exercise factors including exercise steps, emergency resource reserves, safety area settings and the like, data extraction is performed on the first node emergency exercise information according to the plurality of node emergency exercise factors to obtain a plurality of first exercise factors, the first exercise factors being specific implementation contents corresponding to the emergency exercise factors.

[0051] Step S340: performing evaluation feature analysis based on the plurality of first exercise factors to obtain a first group of node evaluation factors;

[0052] In one embodiment, the step S340 of the present application further includes:

[0053] Step S341: setting the exercise factors as retrieval features and setting the evaluation indexes as retrieval targets;

[0054] Step S342: performing big data query based on the retrieval features and the retrieval targets to obtain a plurality of sample exercise factors and a plurality of sample evaluation indexes;

[0055] Step S343: performing mapping relationship analysis based on the plurality of sample exercise factors and the plurality of sample evaluation indexes to obtain a feature sample mapping relationship;

[0056] Step S344: obtaining an evaluation feature analysis model according to the plurality of sample exercise factors and the plurality of sample evaluation indexes based on the feature sample mapping relationship;

[0057] Step S345: inputting the plurality of first exercise factors into the evaluation feature analysis model to obtain the first group of node evaluation factors.

[0058] Step S350: adding the first group of node evaluation factors to the N groups of node evaluation factors.

[0059] Specifically, the exercise factor is set as a retrieval feature, and an evaluation index is set as a retrieval target, the evaluation index being a completion standard of the exercise factor, such as exercise step timeliness, exercise step completion degree, emergency resource demand value, safety area size, and the like. Target search and query are performed based on the retrieval feature and the retrieval target based on big data technology to obtain a plurality of sample exercise factors and a plurality of sample evaluation indexes, the plurality of sample exercise factors and the plurality of sample evaluation indexes having a corresponding relationship, for example, when the sample exercise factor is an emergency resource demand value, the sample evaluation index can be set as 100, 100 tires, 2 tons of snow-melting agent, and the like. Based on a mapping relationship of the plurality of sample exercise factors and the plurality of sample evaluation indexes, an evaluation feature analysis model is constructed with the plurality of sample exercise factors and the plurality of sample evaluation indexes as content, the feature evaluation analysis model being used for analyzing the first exercise factor. Through data query based on big data technology, the evaluation feature analysis model is constructed, which can improve the accuracy of analysis and evaluation of the first exercise factor, thereby improving the accuracy of evaluation of the target emergency exercise scheme. Finally, the plurality of first exercise factors are input into the evaluation feature analysis model for evaluation and analysis to obtain a first group of node evaluation factors. Finally, the first group of node evaluation factors is added to the N groups of node evaluation factors until N groups of node evaluation factors are obtained.

[0060] Step S400: traversing the N groups of node evaluation factors to obtain factor weight analysis results;

[0061] As shown in FIG. 4, Figure 3 in one embodiment, the step S400 of the present application further includes:

[0062] Step S410: based on an emergency exercise evaluation expert group, the N groups of node evaluation factors are set with weights to obtain initial factor weight analysis results;

[0063] Step S420: based on the N groups of node evaluation factors, value degree analysis is performed to obtain factor value degree analysis results;

[0064] Step S430: based on the factor value degree analysis results, the initial factor weight analysis results are weighted to obtain the factor weight analysis results.

[0065] Specifically, an emergency exercise evaluation expert group is constructed, which is an emergency exercise evaluation expert database combined with artificial intelligence and databases, in which a large number of emergency exercise schemes, evaluation results and actual scheme effects are stored, and the evaluation expert database can be updated through continuous learning. According to the emergency exercise evaluation expert group, the N node evaluation factors are weighted, and different weights are set for the evaluation factors based on the importance of the evaluation factors in the scheme process, for example: 30%, 25%, 15%, 10%, 20%, etc., to obtain initial factor weight analysis results. The value degree analysis of the N group node evaluation factors is performed, which refers to setting different value coefficients according to the keyness of the evaluation factors in the scheme, for example: the configuration of ambulances is related to the safety of the people, so the coefficient is set to 100, and the importance of hot water is relatively low, which can be set to 20, etc. The value coefficient can be assigned by the person skilled in the art based on the actual situation, and the factor value degree analysis result is expressed by the value coefficient. Multiply the factor value degree analysis result by the corresponding initial factor weight analysis result to obtain the factor weight analysis result. Through obtaining the factor weight analysis result, support is provided for the next step of constructing a feature exercise evaluation model.

[0066] Step S500: data fusion based on the N group node evaluation factors and the factor weight analysis result to obtain a feature exercise evaluation model;

[0067] Step S600: real-time emergency exercise based on the target emergency exercise scheme to obtain real-time emergency exercise information;

[0068] Specifically, the N group node evaluation factors and the corresponding factor weight analysis result are matched to construct a feature exercise evaluation model, which is a multi-level fuzzy evaluation model. First, the target problem is refined into several levels, and then evaluated from low to high. According to the target emergency exercise scheme, the staff are arranged to perform real-time emergency exercise to obtain real-time emergency exercise information, which includes specific step information of the five processes of early warning management, emergency preparation, emergency response, summary evaluation and post-disposal.

[0069] Step S700: inputting the real-time emergency exercise information into the feature exercise evaluation model to obtain real-time emergency exercise evaluation results.

[0070] In one embodiment, the step S700 of the present application further comprises:

[0071] Step S710: obtaining an emergency exercise evaluation constraint condition;

[0072] Step S720: judging whether the real-time emergency drill evaluation result meets the emergency drill evaluation constraint condition;

[0073] Step S730: if the real-time emergency drill evaluation result meets the emergency drill evaluation constraint condition, generating an emergency drill warning signal.

[0074] Specifically, the real-time emergency drill information is finally input into the feature drill evaluation model for evaluation to obtain a real-time emergency drill evaluation result. A preset emergency drill evaluation constraint condition can be set based on actual conditions. For example: insufficient passenger evacuation vehicles, too long snow removal time, etc. According to the emergency drill evaluation constraint condition, the real-time emergency drill evaluation result is judged. When the real-time drill emergency result meets the emergency drill evaluation constraint condition, an emergency drill warning signal is generated. The emergency drill warning signal is used to remind the person in charge to make rectification to the target emergency drill scheme. Through the above method, the problem that the evaluation result is inaccurate due to unreasonable evaluation index setting in the emergency drill scheme evaluation process is solved. The emergency drill scheme can be effectively evaluated, the deficiencies of the scheme are compared and summarized, and the emergency management capability of the emergency is strengthened.

[0075] In one embodiment, as shown in Figure 4 An emergency drill evaluation system based on data fusion is provided, comprising: a feature recognition module 1, a node division module 2, a data mining module 3, a weight analysis module 4, a feature drill evaluation model obtaining module 5, a real-time emergency drill module 6, a real-time emergency drill evaluation result obtaining module 7, wherein:

[0076] The feature recognition module 1 is used for feature recognition based on the target emergency drill scheme to obtain target emergency drill process information;

[0077] The node division module 2 is used for node division of the target emergency drill scheme based on the target emergency drill process information to obtain N target emergency drill nodes, wherein N is a positive integer greater than 1;

[0078] The data mining module 3 is used for data mining based on the N target emergency drill nodes to obtain N groups of node evaluation factors;

[0079] The weight analysis module 4 is used for weight analysis of the N groups of node evaluation factors to obtain a factor weight analysis result;

[0080] a characteristic exercise evaluation model obtaining module 5, configured to obtain a characteristic exercise evaluation model based on data fusion of the N sets of node evaluation factors and the factor weight resolution results;

[0081] a real-time emergency exercise module 6, configured to perform real-time emergency exercise based on the target emergency exercise scheme, and obtain real-time emergency exercise information;

[0082] a real-time emergency exercise evaluation result obtaining module 7, configured to input the real-time emergency exercise information into the characteristic exercise evaluation model, and obtain a real-time emergency exercise evaluation result.

[0083] In one embodiment, the system further comprises:

[0084] a first target emergency exercise node obtaining module, configured to traverse the N target emergency exercise nodes, and obtain a first target emergency exercise node;

[0085] a characteristic extraction module, configured to perform characteristic extraction on the target emergency exercise scheme based on the first target emergency exercise node, and obtain first node emergency exercise information;

[0086] an exercise factor extraction module, configured to perform exercise factor extraction based on the first node emergency exercise information, and obtain a plurality of first exercise factors;

[0087] an evaluation characteristic analysis module, configured to perform evaluation characteristic analysis based on the plurality of first exercise factors, and obtain a first set of node evaluation factors;

[0088] a first set of node evaluation factor adding module, configured to add the first set of node evaluation factors to the N sets of node evaluation factors.

[0089] In one embodiment, the system further comprises:

[0090] a principal component analysis module, configured to perform principal component analysis based on the first node emergency exercise information, and obtain node standard emergency exercise information;

[0091] an exercise factor extraction module, configured to perform exercise factor extraction based on the node standard emergency exercise information, and obtain a plurality of node emergency exercise factors;

[0092] a plurality of first exercise factor obtaining module, configured to obtain the plurality of first exercise factors according to the plurality of node emergency exercise factors.

[0093] In one embodiment, the system further comprises:

[0094] a first node emergency drill feature dataset obtaining module, configured to obtain a first node emergency drill feature dataset according to the first node emergency drill information;

[0095] a second node emergency drill feature dataset obtaining module, configured to obtain a second node emergency drill feature dataset by decentralizing the first node emergency drill feature dataset;

[0096] a first node emergency drill feature covariance matrix obtaining module, configured to obtain a first node emergency drill feature covariance matrix according to the second node emergency drill feature dataset;

[0097] a first feature information obtaining module, configured to obtain a first eigenvalue and a first eigenvector according to the first node emergency drill feature covariance matrix;

[0098] a node standard emergency drill information obtaining module, configured to obtain the node standard emergency drill information according to the first eigenvalue and the first eigenvector.

[0099] In one embodiment, the system further comprises:

[0100] an information setting module, configured to set a drill factor as a retrieval feature and set an evaluation index as a retrieval target;

[0101] a big data query module, configured to perform big data query based on the retrieval feature and the retrieval target, to obtain a plurality of sample drill factors and a plurality of sample evaluation indexes;

[0102] a mapping relationship analysis module, configured to perform mapping relationship analysis based on the plurality of sample drill factors and the plurality of sample evaluation indexes, to obtain a feature sample mapping relationship;

[0103] an evaluation feature analysis model obtaining module, configured to obtain an evaluation feature analysis model based on the feature sample mapping relationship, according to the plurality of sample drill factors and the plurality of sample evaluation indexes;

[0104] The first group of node evaluation factor obtaining module is configured to input the plurality of first exercise factors into the evaluation feature analysis model to obtain the first group of node evaluation factors.

[0105] In one embodiment, the system further comprises:

[0106] The weight setting module is configured to set weights of the N groups of node evaluation factors based on a group of emergency exercise evaluation experts to obtain initial factor weight analysis results;

[0107] The value degree analysis module is configured to perform value degree analysis based on the N groups of node evaluation factors to obtain factor value degree analysis results;

[0108] The weighted calculation module is configured to perform weighted calculation on the initial factor weight analysis results based on the factor value degree analysis results to obtain the factor weight analysis results.

[0109] In one embodiment, the system further comprises:

[0110] The evaluation constraint condition obtaining module is configured to obtain emergency exercise evaluation constraint conditions;

[0111] The real-time emergency exercise evaluation result judgment module is configured to judge whether the real-time emergency exercise evaluation result meets the emergency exercise evaluation constraint conditions;

[0112] The emergency exercise early warning signal generation module is configured to generate an emergency exercise early warning signal if the real-time emergency exercise evaluation result meets the emergency exercise evaluation constraint conditions.

[0113] In summary, the emergency exercise evaluation method and system based on data fusion provided by the present application have the following technical effects:

[0114] 1. The problem of inaccurate evaluation results caused by unreasonable evaluation index setting in the emergency exercise scheme evaluation process is solved, and the emergency exercise scheme can be effectively and reasonably evaluated to compare and summarize the shortcomings of the scheme and strengthen the emergency management capability of unexpected events.

[0115] 2. By performing principal component analysis on the first node emergency exercise information, the feature data in the first node emergency exercise information can be reduced in dimension, redundant data can be removed under the premise of ensuring the amount of information, the sample size of the feature data in the first node emergency exercise information is reduced, and thus the operation speed of the training model for data is accelerated.

[0116] Any combination of the technical features in the above embodiments can be made, and for the sake of brevity, not all possible combinations are described above, however, as long as the combination of the technical features does not exist in contradiction, it shall be considered within the scope of the present disclosure.

[0117] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it shall not be understood as a limitation on the patent scope of the present application. It shall be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these shall be within the protection scope of the present application. Therefore, the patent protection scope of the present application shall be subject to the appended claims.

Claims

1. An emergency drill evaluation method based on data fusion, characterized in that, The method includes: Based on the target emergency drill plan, feature identification is performed to obtain target emergency drill process information; Based on the target emergency drill process information, the target emergency drill plan is divided into nodes to obtain N target emergency drill nodes, where N is a positive integer greater than 1; Data mining is performed on the N target emergency drill nodes to obtain N sets of node evaluation factors; The weights of the N sets of node evaluation factors are analyzed to obtain the factor weight analysis results. Data fusion is performed based on the N sets of node evaluation factors and the factor weight analysis results to obtain a feature training evaluation model; Conduct real-time emergency drills based on the aforementioned target emergency drill plan to obtain real-time emergency drill information; The real-time emergency drill information is input into the feature-based drill evaluation model to obtain the real-time emergency drill evaluation results; The weights of the N sets of node evaluation factors are analyzed to obtain the factor weight analysis results, including: Based on the emergency drill evaluation expert group, the weights of the N groups of node evaluation factors are set to obtain the initial factor weight analysis results. Based on the N sets of node evaluation factors, a value analysis is performed to obtain the factor value analysis results. The initial factor weight analysis results are weighted and calculated based on the factor value analysis results to obtain the factor weight analysis results. After obtaining the real-time emergency drill evaluation results, the method further includes: Obtain the constraints for emergency drill evaluation; Determine whether the real-time emergency drill evaluation results meet the emergency drill evaluation constraints. If the real-time emergency drill evaluation results meet the emergency drill evaluation constraints, an emergency drill early warning signal is generated.

2. The method as described in claim 1, characterized in that, The method further includes: Traverse the N target emergency drill nodes to obtain the first target emergency drill node; Based on the first target emergency drill node, feature extraction is performed on the target emergency drill plan to obtain the first node emergency drill information; Based on the emergency drill information of the first node, drill factors are extracted to obtain multiple first drill factors; Based on the multiple first training factors, an evaluation feature analysis is performed to obtain the first set of node evaluation factors; Add the first set of node evaluation factors to the N sets of node evaluation factors.

3. The method as described in claim 2, characterized in that, Based on the emergency drill information from the first node, drill factors are extracted to obtain multiple first drill factors, including: Principal component analysis is performed based on the emergency drill information of the first node to obtain the standard emergency drill information of the node. Based on the node standard emergency drill information, drill factors are extracted to obtain multiple node emergency drill factors. Based on the multiple node emergency drill factors, the multiple first drill factors are obtained.

4. The method as described in claim 3, characterized in that, Principal component analysis is performed based on the emergency drill information of the first node to obtain standard emergency drill information for the node, including: Based on the emergency drill information of the first node, obtain the emergency drill feature dataset of the first node; The emergency drill feature dataset of the first node is decentralized to obtain the emergency drill feature dataset of the second node. Based on the emergency drill feature dataset of the second node, obtain the covariance matrix of the emergency drill features of the first node; Based on the covariance matrix of the emergency drill characteristics of the first node, the first eigenvalue and the first eigenvector are obtained. Based on the first feature value and the first feature vector, the standard emergency drill information of the node is obtained.

5. The method as described in claim 2, characterized in that, Based on the evaluation feature analysis of the multiple first training factors, a first set of node evaluation factors is obtained, including: Set the exercise factors as search features and the evaluation indicators as search targets; Based on the search features and the search target, perform big data queries to obtain multiple sample training factors and multiple sample evaluation indicators; Based on the multiple sample training factors and the multiple sample evaluation indicators, a mapping relationship analysis is performed to obtain the feature sample mapping relationship; Based on the feature sample mapping relationship, and according to the multiple sample training factors and the multiple sample evaluation indicators, an evaluation feature analysis model is obtained; The multiple first training factors are input into the evaluation feature analysis model to obtain the first set of node evaluation factors.

6. An emergency drill evaluation system based on data fusion, characterized in that, The data fusion-based emergency drill evaluation system is used to implement the method described in any one of claims 1-5, including: The feature recognition module is used to perform feature recognition based on the target emergency drill plan to obtain target emergency drill process information. The node partitioning module is used to partition the target emergency drill plan into nodes based on the target emergency drill process information, and obtain N target emergency drill nodes, where N is a positive integer greater than 1. A data mining module is used to perform data mining based on the N target emergency drill nodes to obtain N sets of node evaluation factors. The weight parsing module is used to traverse the N groups of node evaluation factors to perform weight parsing and obtain factor weight parsing results. The feature training evaluation model acquisition module is used to perform data fusion based on the N sets of node evaluation factors and the factor weight parsing results to obtain the feature training evaluation model. A real-time emergency drill module is used to conduct real-time emergency drills based on the target emergency drill plan and obtain real-time emergency drill information. A real-time emergency drill evaluation result acquisition module is used to input the real-time emergency drill information into the feature-based drill evaluation model to obtain the real-time emergency drill evaluation result.

Images