CN117972336A - Flight safety risk assessment method and device, electronic equipment and storage medium - Google Patents
Flight safety risk assessment method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117972336A CN117972336A CN202311680277.9A CN202311680277A CN117972336A CN 117972336 A CN117972336 A CN 117972336A CN 202311680277 A CN202311680277 A CN 202311680277A CN 117972336 A CN117972336 A CN 117972336A
- Authority
- CN
- China
- Prior art keywords
- flight
- data
- sample
- target
- safety risk
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000012502 risk assessment Methods 0.000 title claims abstract description 59
- 238000013058 risk prediction model Methods 0.000 claims abstract description 58
- 238000012549 training Methods 0.000 claims abstract description 53
- 230000010006 flight Effects 0.000 claims abstract description 22
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims description 32
- 238000012360 testing method Methods 0.000 claims description 14
- 238000011156 evaluation Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 238000013145 classification model Methods 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 230000008030 elimination Effects 0.000 claims description 4
- 238000003379 elimination reaction Methods 0.000 claims description 4
- 238000007726 management method Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000010801 machine learning Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000008520 organization Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Strategic Management (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of data processing, and provides a flight security risk assessment method, a device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring aviation flight data of a target flight; extracting features based on the feature templates and aviation flight data to obtain target aviation flight features of the target flights; the feature templates are generated based on sample features; inputting the target aviation flight characteristics into a safety risk prediction model to obtain a safety risk prediction result output by the safety risk prediction model; the safety risk prediction model is obtained by model training based on sample characteristics of sample flights and safety risk labels corresponding to the sample characteristics; the sample features are determined by feature selection of initial features based on an embedding method; the initial characteristics are obtained by extracting characteristics of historical aviation flight data of the sample flight; and carrying out security risk assessment based on the security risk prediction result. The application can improve the risk assessment efficiency of the fly safety.
Description
Technical Field
The application relates to the technical field of data processing, in particular to a flight security risk assessment method and device, electronic equipment and a storage medium.
Background
In order to implement the system and the effective civil aviation safety management, standardize the safety behavior of pilots, resist the safety risk of flying, reduce the unsafe events of flying, ensure the safety and normal operation of civil aviation, promote the comprehensive construction of aviation and need to evaluate the safety risk of flying. With the development of the big data age, the aviation industry has accumulated a lot of flight data, maintenance data, weather data, etc. However, the current flight safety risk assessment work is difficult to implement to assessment details, risk is predicted mainly by examining the quality condition of the work, flight personnel resist the risk mainly by experience, and the flight personnel have no unified standard and specification and lack certain systematicness and objectivity, so that the risk type prediction in the construction of a flight safety system is not comprehensive enough, the occurrence probability assessment is not accurate enough, and the efficiency in the current flight safety risk assessment is low.
Disclosure of Invention
The embodiment of the application provides a flight safety risk assessment method, a flight safety risk assessment device, electronic equipment and a storage medium, which are used for solving the problem of low efficiency when flight safety risk assessment is currently carried out.
In a first aspect, an embodiment of the present application provides a flight security risk assessment method, including:
Acquiring aviation flight data of a target flight;
Performing feature extraction on the basis of the feature template and the aviation flight data to obtain target aviation flight features of the target flights; the feature templates are generated based on sample features;
Inputting the target aviation flight characteristics into a safety risk prediction model to obtain a safety risk prediction result output by the safety risk prediction model; the safety risk prediction model is obtained by model training based on sample characteristics of sample flights and safety risk labels corresponding to the sample characteristics; the sample features are determined by feature selection of initial features based on an embedding method; the initial characteristics are obtained by extracting characteristics of historical aviation flight data of the sample flight;
And carrying out security risk assessment based on the security risk prediction result.
In one embodiment, the security risk prediction model is determined by:
acquiring historical aviation flight data of a sample flight and a security risk tag thereof;
Extracting features of the historical aviation flight data to obtain first features;
Training a preset classification model by using the security risk tag and the first feature through an embedding method to obtain an intermediate model and sample features;
Training the intermediate model based on the sample characteristics and the security risk label to obtain a security risk prediction model.
In one embodiment, the training the intermediate model based on the sample feature and the security risk tag to obtain a security risk prediction model includes:
dividing a data set formed by the sample features and the security risk tag into a training set and a testing set;
Model training is carried out on the intermediate model based on the training set;
If the prediction accuracy of the intermediate model reaches a preset accuracy threshold, performing model evaluation on the intermediate model based on the test set;
and if the accuracy rate of the intermediate model is greater than or equal to a preset accuracy rate threshold value, determining the intermediate model as a safety risk prediction model.
In one embodiment, the extracting the characteristics from the aviation flight data based on the characteristic template to obtain the target aviation flight characteristics of the target flight includes:
Repeating data elimination is carried out on the aviation flight data to obtain first data;
performing missing value processing on the first data to obtain second data;
performing outlier processing on the second data to obtain third data;
And obtaining a characteristic template, and carrying out characteristic extraction according to the characteristic template and the third data to obtain the target aviation flight characteristic of the target flight.
In one embodiment, the extracting the features according to the feature template and the third data to obtain the target aviation flight feature of the target flight includes:
Performing data conversion on the third data to obtain fourth data;
normalizing the fourth data to obtain fifth data;
And extracting features of the fifth data according to the feature templates to obtain target aviation flight features of the target flights.
In one embodiment, the security risk prediction result includes a prediction probability for each security risk type; the safety risk assessment based on the safety risk prediction result comprises the following steps:
respectively comparing the prediction probability of each safety risk type in the safety risk prediction result with a preset probability threshold;
And if any prediction probability is greater than or equal to the preset probability threshold, carrying out security risk early warning on the target flight.
In one embodiment, the aviation flight data includes at least one of flight records, aircraft sensor data, weather data, traffic management data.
In a second aspect, an embodiment of the present application provides a flight security risk assessment method apparatus, including:
The acquisition module is used for acquiring aviation flight data of the target flight;
The extraction module is used for carrying out feature extraction on the basis of the feature template and the aviation flight data to obtain target aviation flight features of the target flights; the feature templates are generated based on sample features;
The input module is used for inputting the target aviation flight characteristics into a safety risk prediction model to obtain a safety risk prediction result output by the safety risk prediction model; the safety risk prediction model is obtained by model training based on sample characteristics of sample flights and safety risk labels corresponding to the sample characteristics; the sample features are determined by feature selection of initial features based on an embedding method; the initial characteristics are obtained by extracting characteristics of historical aviation flight data of the sample flight;
and the evaluation module is used for carrying out security risk evaluation based on the security risk prediction result.
In a third aspect, an embodiment of the present application provides an electronic device, including a processor and a memory storing a computer program, where the processor implements the flight safety risk assessment method according to the first aspect when executing the program.
In a fourth aspect, an embodiment of the present application provides a storage medium, where the storage medium is a computer readable storage medium, including a computer program, where the computer program when executed by a processor implements the flight safety risk assessment method according to the first aspect.
According to the flight safety risk assessment method, the device, the electronic equipment and the storage medium, the initial characteristics are obtained by carrying out characteristic extraction on historical aviation flight data of a sample flight in advance, the sample characteristics are determined by carrying out characteristic selection on the initial characteristics based on an embedding method, the safety risk prediction model is obtained by carrying out model training on the sample characteristics of the sample flight and the safety risk labels corresponding to the sample characteristics, and meanwhile, the characteristic template is generated based on the sample characteristics, so that after the aviation flight data of a target flight are obtained, the aviation flight data can be rapidly and accurately subjected to characteristic extraction according to the characteristic template, the extracted characteristics are input into the safety risk prediction model, and the safety risk prediction result is rapidly and accurately obtained through the safety risk prediction model, and further, the safety risk assessment of the flight can be timely and accurately carried out according to the safety risk prediction result, and therefore the efficiency of the flying safety risk assessment can be improved.
Drawings
In order to more clearly illustrate the application or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the application, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method for risk assessment of flight safety according to an embodiment of the present application;
Fig. 2 is a schematic diagram of a determination flow of a security risk prediction model in a flight security risk assessment method according to an embodiment of the present application;
FIG. 3 is a schematic diagram of functional modules of an embodiment of a method and apparatus for risk assessment of flight safety according to the present application;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the embodiments of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The flight security risk assessment method, the flight security risk assessment device, the electronic equipment and the storage medium provided by the application are described in detail below with reference to the embodiments.
Fig. 1 is a schematic flow chart of a flight security risk assessment method according to an embodiment of the present application. Referring to fig. 1, an embodiment of the present application provides a flight security risk assessment method, which may include:
Step 100, acquiring aviation flight data of a target flight;
It should be noted that, the execution body of the flight security risk assessment method provided in the embodiment of the present application may be a server, a computer device, such as a mobile phone, a tablet computer, a notebook computer, a palm computer, a vehicle-mounted electronic device, a wearable device, an Ultra-mobile Personal Computer (UMPC), a netbook, a Personal digital assistant (Personal DIGITAL ASSISTANT, PDA), or the like. The flight safety risk assessment method device can be arranged or connected in the server or the computer equipment, and the flight safety risk assessment method can be completed by controlling the flight safety risk assessment method device.
The target flight in the application can be an airplane flight with a specified airplane model in a specified airline company. And more particularly may be flights flying for a specified period of time on a specified flight path. Or a general civil aviation flight.
Aviation flight data in the present application may include at least one of flight records, aircraft sensor data, weather data, traffic management data.
In one embodiment, the aviation flight data of the present application may include flight records, aircraft sensor data, weather data, traffic management data, and the like.
The flight records may include flight personnel attributes, flight numbers, departure times, landing times, flight distances, and the like, among others.
The aircraft sensor data may include aircraft equipment information, flight speed, altitude, barometric pressure, temperature, acceleration, and the like.
The meteorological data may include flight environment, temperature, humidity, wind speed, visibility, etc.
Traffic management data may include organization management, flight routes, flight density, and the like.
The flight personnel attributes may include theoretical basis and flight technology, mission experience and safety, lead trunk and technical backbone, training and following the century, physical attributes and psychological qualities, safety training and safety literacy, and the like. The above-mentioned items may also have more specific branch content, for example, the theoretical basis and flight technique may include aviation theory mastering conditions, capability conditions for solving flight practical problems by applying theory, technique maintenance and targeted training content implementation conditions, special condition handling method mastering conditions, and the like; the safety training and the safety literacy can comprise definite deviation, special conditions, life saving and other safety training content implementation conditions, flight habit and cultivation conditions, unit resource management implementation conditions and the like.
The aircraft equipment may include aircraft basic parameters, aircraft quality and safety assistance, aircraft failure and analysis, warranty capability and educational training, quality monitoring and technical notification implementation, aircraft instrumentation warranty, maintenance management organization leadership, and the like.
The flight environment may include aircraft logistics, technical logistics, emergency rescue capabilities, airport objective conditions, mission area environments, and the like.
Organization management may include leader function exertion, planning and implementation, training quality monitoring, flight safety cultural construction, training political works, and the like.
Step 200, extracting features based on the feature templates and aviation flight data to obtain target aviation flight features of a target flight;
According to the application, the historical aviation flight data of the sample flight can be subjected to feature extraction in advance to obtain initial features, the initial features are subjected to feature selection based on an embedding method to determine the sample features, model training is performed based on the sample features of the sample flight and the security risk labels corresponding to the sample features to obtain a security risk prediction model which can be used for carrying out security risk prediction according to the input features, and a feature template is generated based on the sample features.
The sample flight may be one or more flight types, which may be general civil aviation flight types or flight types of specified aircraft types in specified airlines, and may be specifically determined according to actual requirements, which is not specifically limited in the present application.
The security risk tag may be whether there is a security risk or not and a security risk type when there is a security risk. Types of security risks may include, but are not limited to, excessive angle of attack, stall, tail spin, etc.
In the application, proper characteristics can be selected from historical aviation flight data according to a predicted object, and a preset classification model is formed by combining a machine learning algorithm for classifying and regressing problems, wherein the machine learning algorithm can be a decision tree, a support vector machine, a random forest and the like.
Therefore, after the aviation flight data of the target flight are obtained, the feature template can be obtained, and the data formed based on the aviation flight data are subjected to feature extraction according to the feature template, so that the aviation flight feature of the target flight is obtained, and the aviation flight feature of the target flight is determined to be the target aviation flight feature.
Step 300, inputting the target aviation flight characteristics into a safety risk prediction model to obtain a safety risk prediction result output by the safety risk prediction model;
After the target aviation flight characteristics are obtained, the target aviation flight characteristics can be input into a safety risk prediction model, safety risk prediction is carried out according to the input characteristics through the safety risk prediction model, and a safety risk prediction result comprising the prediction probability of each safety risk type is obtained.
And 400, performing security risk assessment based on the security risk prediction result.
After the safety risk prediction result is obtained, the safety risk assessment can be carried out on the target flight according to the prediction probability of each safety risk type in the safety risk prediction result, and early warning can be carried out according to the assessment result, so that relevant personnel can conveniently carry out corresponding treatment on the target flight according to the early warning, and the probability of occurrence of flight accidents or accident symptoms is reduced to the greatest extent.
According to the flight safety risk assessment method provided by the embodiment of the application, the initial characteristics are obtained by carrying out characteristic extraction on the historical aviation flight data of the sample flight in advance, the characteristic selection is carried out on the initial characteristics based on the embedding method to determine the sample characteristics, the model training is carried out on the safety risk labels corresponding to the sample characteristics and the sample characteristics based on the sample characteristics of the sample flight to obtain the safety risk prediction model, and the characteristic template is generated based on the sample characteristics, so that after the aviation flight data of the target flight is obtained, the characteristic extraction can be carried out on the aviation flight data rapidly and accurately according to the characteristic template, the extracted characteristics are input into the safety risk prediction model, the safety risk prediction is carried out through the safety risk prediction model, and the safety risk assessment of the flight can be carried out timely and accurately according to the safety risk prediction result, so that the flight safety risk assessment efficiency can be improved.
In one embodiment, the security risk prediction model is determined by:
step 1, acquiring historical aviation flight data of a sample flight and a security risk tag thereof;
step 2, extracting features of historical aviation flight data to obtain first features;
step 3, training a preset classification model by using a security risk tag and a first feature through an embedding method to obtain an intermediate model and sample features;
And step 4, training the intermediate model based on the sample characteristics and the security risk label to obtain a security risk prediction model.
According to the application, the historical aviation flight data such as flight records, aircraft sensor data, meteorological data, traffic management data and the like of sample flights in a specified time period can be obtained, and the security risk labels of all flights in the time period can be obtained.
Further, an appropriate feature may be selected from the historical aviation flight data according to the predicted object as the first feature, for example, may be a flight altitude, a speed, a gesture, a heading, a wind speed, an air pressure, and the like.
Further, a predetermined classification model may be formed based on the selected first feature in combination with a machine learning algorithm for classifying and regressing the problem.
Furthermore, each first feature and the corresponding security risk label can be input into a preset classification model, feature selection is performed through an embedding method in the training process of the preset classification model, so that the embedding method is used for acquiring weight coefficients of the features and screening important features, features with important influences on model prediction targets can be identified and reserved, redundant features with small influences on prediction results are removed, and therefore the efficiency and accuracy of the model are improved.
Therefore, after training is completed, a trained model can be obtained as an intermediate model, and meanwhile, the characteristics obtained through screening can be obtained as sample characteristics.
Further, the sample features and the corresponding security risk labels may be formed into a data set and used to train the intermediate model, and after training is completed, a security risk prediction model for performing security risk prediction according to the input features of the same type as the sample features is obtained.
Further, training the intermediate model based on the sample features and the security risk tag to obtain a security risk prediction model, including:
step 41, dividing a data set formed by sample characteristics and security risk labels into a training set and a testing set;
step 42, model training is carried out on the intermediate model based on the training set;
Step 43, if the prediction accuracy of the intermediate model reaches a preset accuracy threshold, performing model evaluation on the intermediate model based on the test set;
And step 44, if the accuracy rate of the intermediate model is greater than or equal to the preset accuracy rate threshold, determining the intermediate model as a safety risk prediction model.
In the application, a data set formed by sample characteristics and corresponding security risk labels can be divided into a training set and a testing set. Specifically, the division may be performed by any means such as random division, hierarchical sampling, time series division, and cross-validation, and the present application is not limited thereto.
After the training set and the testing set are obtained, the intermediate model can be trained through the training set, and meanwhile, the accuracy rate of model prediction can be estimated in real time. Specifically, the current prediction accuracy of the model may be compared to a pre-set accuracy threshold.
When the prediction accuracy of the intermediate model is determined to be greater than or equal to the preset accuracy threshold, the intermediate model can be considered to achieve a better prediction effect, so that training of the model can be finished.
Further, the performance of the intermediate model may be evaluated by a test set, wherein the evaluated index may be any of accuracy, recall, precision. The present application will be described with reference to the accuracy.
The accuracy refers to the proportion of the classification model to the sample that is actually the positive example among the samples that are predicted to be the positive example.
In the model evaluation process, the accuracy of the intermediate model may be compared with a preset accuracy threshold.
If the accuracy rate of the intermediate model is determined to be greater than or equal to the preset accuracy rate threshold value, the model can be determined to have better prediction accuracy and stability, so that the intermediate model at the moment can be determined to be a safety risk prediction model.
According to the embodiment, the initial characteristics are obtained by carrying out characteristic extraction on historical aviation flight data of a sample flight in advance, the initial characteristics are subjected to characteristic selection based on an embedding method to determine the sample characteristics, model training is carried out based on the sample characteristics of the sample flight and the security risk labels corresponding to the sample characteristics to obtain a security risk prediction model, and meanwhile, a characteristic template is generated based on the sample characteristics, so that after the aviation flight data of a target flight are obtained, the aviation flight data can be rapidly and accurately subjected to characteristic extraction according to the characteristic template, the extracted characteristics are input into the security risk prediction model, security risk prediction is carried out through the security risk prediction model, and therefore a security risk prediction result is rapidly and accurately obtained, and further, security risk assessment of the flight can be timely and accurately carried out according to the security risk prediction result, and therefore the efficiency of flying security risk assessment can be improved.
In one embodiment, feature extraction is performed based on feature templates and aviation flight data to obtain target aviation flight features of a target flight, including:
Step 201, repeating data elimination is carried out on aviation flight data to obtain first data;
step 202, performing missing value processing on the first data to obtain second data;
step 203, performing outlier processing on the second data to obtain third data;
And 204, acquiring a feature template, and carrying out feature extraction according to the feature template and third data to obtain the target aviation flight feature of the target flight.
After the aviation flight data are obtained, the method can remove repeated data in the aviation flight data, specifically can detect whether identical records exist in the aviation flight data, and if so, deletes one group of data to obtain the first data so as to avoid repeated calculation or analysis.
Further, it may be possible to detect, for processing a missing value in the first data, in particular, whether a missing value is present in the first data, the missing value being characterized by no value in the record. The missing values, if any, may be processed by the following policies:
deleting records containing missing values: if the missing value is less, for example, below a number threshold, the record containing the missing value may be deleted directly.
Interpolation missing values: if the missing values are more, for example, greater than or equal to a number threshold, interpolation methods may be used to fill in the missing values, for example, using a mean, median, mode, regression model, or the like, to obtain the second data.
Further, an abnormal value in the second data may be processed, and in particular, whether or not an abnormal value exists in the second data may be detected, and the abnormal value may be a value significantly deviated from other records. The outliers can be handled by the following policies if any:
deleting the outlier: if the outlier is due to a data acquisition error or a data error input, the outlier may be deleted directly.
Replacement of outliers: if the outlier is due to a reasonable reason, the outlier may be replaced with a suitable replacement value, for example using an average, median or by interpolation, etc., thereby yielding third data.
Further, a predetermined feature template may be obtained.
Further, feature extraction can be performed on the data obtained through the third data processing according to the feature template, and the target aviation flight feature of the target flight is obtained.
Further, feature extraction is performed according to the feature template and the third data to obtain a target aviation flight feature of the target flight, including:
Step 2041, performing data conversion on the third data to obtain fourth data;
Step 2042, performing normalization processing on the fourth data to obtain fifth data;
and 2043, carrying out feature extraction on the fifth data according to the feature template to obtain the target aviation flight feature of the target flight.
After the third data is obtained, the data type conversion can be performed on the third data to ensure that the data type is correct.
Further, the fourth data may be subjected to data normalization processing, and the data normalization processing may eliminate dimensional differences between different features. In the application, the data can be converted into the same scale range by using a normalization method to obtain fifth data.
Further, corresponding features may be extracted from the fifth data according to the feature templates, and the extracted features may be determined as target aviation flight features of the target flights.
According to the embodiment, abnormal data processing can be performed on aviation flight data, the effectiveness of the data can be ensured, and further when feature extraction is performed from the aviation flight data according to the feature template, accurate aviation flight features can be obtained, and flight safety risk assessment can be accurately performed, so that flight safety risk assessment efficiency can be improved.
In one embodiment, performing a security risk assessment based on the security risk prediction result includes:
step 401, comparing the prediction probability of each security risk type in the security risk prediction result with a preset probability threshold;
Step 402, if any predicted probability is greater than or equal to a preset probability threshold, performing security risk early warning on the target flight.
The method and the device can extract the prediction probability corresponding to each security risk type in the security risk prediction result. Meanwhile, the probability threshold value can be set in advance for different security risk types, and the same probability threshold value can be set in advance for all security risk types.
Therefore, if the same probability threshold is set for all the security risk types in advance, the prediction probability of each security risk type in the security risk prediction result is respectively compared with the preset probability threshold, so that the magnitude relation between the prediction probability of each security risk type and the preset probability threshold is determined.
If the prediction probability of any safety risk type exists in each safety risk type, which is determined to be greater than or equal to the preset probability threshold, the safety risk early warning of the target flight can be carried out according to the safety risk type with the prediction probability greater than or equal to the preset probability threshold. For example, if the type of security risk with the prediction probability greater than or equal to the preset probability threshold is that the attack angle is too large, outputting early warning information of the too large attack angle. If the safety risk type with the prediction probability being greater than or equal to the preset probability threshold value is stall, stall early warning information is output.
According to the embodiment, after the safety risk prediction result is obtained rapidly and accurately, the safety risk assessment of the flight can be timely and accurately carried out according to the safety risk prediction result, so that the flight safety risk assessment efficiency can be improved.
Fig. 2 is a schematic diagram of a determination flow of a security risk prediction model in a flight security risk assessment method according to an embodiment of the present application. Referring to fig. 2, in one embodiment, historical aviation data may be collected, after the historical aviation data is preprocessed, feature selection may be performed on the preprocessed data, tag labels of the data may be further determined, and according to the historical aviation data, targets to be predicted, such as whether an aircraft fault occurs or not, flight path selection, and the like, may be determined, and these target tags may be mapped to the historical data to form a data set. Further, the data set may be partitioned into a training set and a testing set, typically with most of the data used for training and a small portion for testing. Further, the model evaluation can be performed by training the machine learning model through a training set, so as to judge whether the model reaches the target accuracy, if so, the model test is performed on the trained model through a test set, and the result comparison analysis is performed after the test result is output, so that the final safety risk prediction model is obtained. If the model does not reach the target accuracy, updating the training parameters, and continuously training the machine learning model through the training set until the model reaches the target accuracy.
Furthermore, the application also provides a flight safety risk assessment method and device.
Referring to fig. 3, fig. 3 is a schematic diagram of functional modules of an embodiment of a flight security risk assessment method apparatus according to the present application.
The flight safety risk assessment method device comprises the following steps:
An acquisition module 310, configured to acquire aviation flight data of a target flight;
the extracting module 320 is configured to perform feature extraction with the aviation flight data based on a feature template, so as to obtain a target aviation flight feature of the target flight; the feature templates are generated based on sample features;
the input module 330 is configured to input the target aviation flight characteristic into a security risk prediction model, and obtain a security risk prediction result output by the security risk prediction model; the safety risk prediction model is obtained by model training based on sample characteristics of sample flights and safety risk labels corresponding to the sample characteristics; the sample features are determined by feature selection of initial features based on an embedding method; the initial characteristics are obtained by extracting characteristics of historical aviation flight data of the sample flight;
and the evaluation module 340 is configured to perform security risk evaluation based on the security risk prediction result.
In one embodiment, the extraction module 320 is specifically configured to:
Repeating data elimination is carried out on the aviation flight data to obtain first data;
performing missing value processing on the first data to obtain second data;
performing outlier processing on the second data to obtain third data;
And obtaining a characteristic template, and carrying out characteristic extraction according to the characteristic template and the third data to obtain the target aviation flight characteristic of the target flight.
In one embodiment, the extraction module 320 includes an extraction unit to:
Performing data conversion on the third data to obtain fourth data;
normalizing the fourth data to obtain fifth data;
And extracting features of the fifth data according to the feature templates to obtain target aviation flight features of the target flights.
In one embodiment, the evaluation module 340 is specifically configured to:
respectively comparing the prediction probability of each safety risk type in the safety risk prediction result with a preset probability threshold;
And if any prediction probability is greater than or equal to the preset probability threshold, carrying out security risk early warning on the target flight.
According to the flight safety risk assessment method and device provided by the embodiment of the application, the initial characteristics are obtained by carrying out characteristic extraction on the historical aviation flight data of the sample flight in advance, the characteristic selection is carried out on the initial characteristics based on the embedding method to determine the sample characteristics, the model training is carried out on the basis of the sample characteristics of the sample flight and the safety risk labels corresponding to the sample characteristics to obtain the safety risk prediction model, and the characteristic template is generated based on the sample characteristics, so that after the aviation flight data of the target flight are obtained, the aviation flight data can be rapidly and accurately subjected to characteristic extraction according to the characteristic template, the extracted characteristics are input into the safety risk prediction model, and the safety risk prediction is carried out through the safety risk prediction model, so that the safety risk prediction result of the flight can be rapidly and accurately obtained, and the safety risk assessment of the flight can be timely and accurately carried out according to the safety risk prediction result, and the safety risk assessment efficiency of the flight can be improved.
Fig. 4 illustrates a physical schematic diagram of an electronic device, as shown in fig. 4, which may include: processor 410, communication interface (Communication Interface) 420, memory 430, and communication bus 440, wherein processor 410, communication interface 420, and memory 430 communicate with each other via communication bus 440. Processor 410 may call a computer program in memory 430 to perform the steps of a flight safety risk assessment method, including, for example:
Acquiring aviation flight data of a target flight;
Performing feature extraction on the basis of the feature template and the aviation flight data to obtain target aviation flight features of the target flights; the feature templates are generated based on sample features;
Inputting the target aviation flight characteristics into a safety risk prediction model to obtain a safety risk prediction result output by the safety risk prediction model; the safety risk prediction model is obtained by model training based on sample characteristics of sample flights and safety risk labels corresponding to the sample characteristics; the sample features are determined by feature selection of initial features based on an embedding method; the initial characteristics are obtained by extracting characteristics of historical aviation flight data of the sample flight;
And carrying out security risk assessment based on the security risk prediction result.
Further, the logic instructions in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, an embodiment of the present application further provides a storage medium, where the storage medium is a computer readable storage medium, where the computer readable storage medium stores a computer program, where the computer program is configured to cause a processor to execute the steps of the method provided in the foregoing embodiments, where the method includes:
Acquiring aviation flight data of a target flight;
Performing feature extraction on the basis of the feature template and the aviation flight data to obtain target aviation flight features of the target flights; the feature templates are generated based on sample features;
Inputting the target aviation flight characteristics into a safety risk prediction model to obtain a safety risk prediction result output by the safety risk prediction model; the safety risk prediction model is obtained by model training based on sample characteristics of sample flights and safety risk labels corresponding to the sample characteristics; the sample features are determined by feature selection of initial features based on an embedding method; the initial characteristics are obtained by extracting characteristics of historical aviation flight data of the sample flight;
And carrying out security risk assessment based on the security risk prediction result.
The computer readable storage medium may be any available medium or data storage device that can be accessed by a processor including, but not limited to, magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.), optical memory (e.g., CD, DVD, BD, HVD, etc.), and semiconductor memory (e.g., ROM, EPROM, EEPROM, nonvolatile memory (NAND FLASH), solid State Disk (SSD)), etc.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (10)
1. A method of flight safety risk assessment, comprising:
Acquiring aviation flight data of a target flight;
Performing feature extraction on the basis of the feature template and the aviation flight data to obtain target aviation flight features of the target flights; the feature templates are generated based on sample features;
Inputting the target aviation flight characteristics into a safety risk prediction model to obtain a safety risk prediction result output by the safety risk prediction model; the safety risk prediction model is obtained by model training based on sample characteristics of sample flights and safety risk labels corresponding to the sample characteristics; the sample features are determined by feature selection of initial features based on an embedding method; the initial characteristics are obtained by extracting characteristics of historical aviation flight data of the sample flight;
And carrying out security risk assessment based on the security risk prediction result.
2. The method of claim 1, wherein the security risk prediction model is determined by:
acquiring historical aviation flight data of a sample flight and a security risk tag thereof;
Extracting features of the historical aviation flight data to obtain first features;
Training a preset classification model by using the security risk tag and the first feature through an embedding method to obtain an intermediate model and sample features;
Training the intermediate model based on the sample characteristics and the security risk label to obtain a security risk prediction model.
3. The method according to claim 2, wherein training the intermediate model based on the sample features and the security risk tag to obtain a security risk prediction model comprises:
dividing a data set formed by the sample features and the security risk tag into a training set and a testing set;
Model training is carried out on the intermediate model based on the training set;
If the prediction accuracy of the intermediate model reaches a preset accuracy threshold, performing model evaluation on the intermediate model based on the test set;
and if the accuracy rate of the intermediate model is greater than or equal to a preset accuracy rate threshold value, determining the intermediate model as a safety risk prediction model.
4. The method for evaluating the flight security risk according to claim 1, wherein the feature extraction is performed with the aviation flight data based on a feature template to obtain the target aviation flight feature of the target flight, comprising:
Repeating data elimination is carried out on the aviation flight data to obtain first data;
performing missing value processing on the first data to obtain second data;
performing outlier processing on the second data to obtain third data;
And obtaining a characteristic template, and carrying out characteristic extraction according to the characteristic template and the third data to obtain the target aviation flight characteristic of the target flight.
5. The method for evaluating the risk of flying safety according to claim 4, wherein the extracting features according to the feature template and the third data to obtain the target aviation flight feature of the target flight comprises:
Performing data conversion on the third data to obtain fourth data;
normalizing the fourth data to obtain fifth data;
And extracting features of the fifth data according to the feature templates to obtain target aviation flight features of the target flights.
6. The method of claim 1, wherein the security risk prediction result includes a prediction probability for each security risk type; the safety risk assessment based on the safety risk prediction result comprises the following steps:
respectively comparing the prediction probability of each safety risk type in the safety risk prediction result with a preset probability threshold;
And if any prediction probability is greater than or equal to the preset probability threshold, carrying out security risk early warning on the target flight.
7. The method of claim 1-6, wherein the aviation flight data comprises at least one of flight records, aircraft sensor data, weather data, traffic management data.
8. A method and apparatus for risk assessment of flight safety, comprising:
The acquisition module is used for acquiring aviation flight data of the target flight;
The extraction module is used for carrying out feature extraction on the basis of the feature template and the aviation flight data to obtain target aviation flight features of the target flights; the feature templates are generated based on sample features;
The input module is used for inputting the target aviation flight characteristics into a safety risk prediction model to obtain a safety risk prediction result output by the safety risk prediction model; the safety risk prediction model is obtained by model training based on sample characteristics of sample flights and safety risk labels corresponding to the sample characteristics; the sample features are determined by feature selection of initial features based on an embedding method; the initial characteristics are obtained by extracting characteristics of historical aviation flight data of the sample flight;
and the evaluation module is used for carrying out security risk evaluation based on the security risk prediction result.
9. An electronic device comprising a processor and a memory storing a computer program, wherein the processor implements the flight safety risk assessment method of any one of claims 1 to 7 when executing the computer program.
10. A storage medium, which is a computer-readable storage medium, comprising a computer program, characterized in that the computer program, when executed by a processor, implements the flying safety risk assessment method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311680277.9A CN117972336A (en) | 2023-12-08 | 2023-12-08 | Flight safety risk assessment method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311680277.9A CN117972336A (en) | 2023-12-08 | 2023-12-08 | Flight safety risk assessment method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117972336A true CN117972336A (en) | 2024-05-03 |
Family
ID=90854503
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311680277.9A Pending CN117972336A (en) | 2023-12-08 | 2023-12-08 | Flight safety risk assessment method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117972336A (en) |
-
2023
- 2023-12-08 CN CN202311680277.9A patent/CN117972336A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113486938B (en) | Multi-branch time convolution network-based re-landing analysis method and device | |
| CN107818382B (en) | Flight arrival time prediction method | |
| Mangortey et al. | Application of machine learning techniques to parameter selection for flight risk identification | |
| CN108805337B (en) | Aviation operation control risk management and control method and system | |
| CN103745229A (en) | Method and system of fault diagnosis of rail transit based on SVM (Support Vector Machine) | |
| CN103080954A (en) | Method and system for analyzing flight data recorded during the flight of an airplane | |
| CN113344423B (en) | Pilot scene applicability diagnosis method and system based on machine learning | |
| Calle-Alonso et al. | A Bayesian-network-based approach to risk analysis in runway excursions | |
| Fernández et al. | Flight data monitoring (FDM) unknown hazards detection during approach phase using clustering techniques and AutoEncoders | |
| CN108957173A (en) | A kind of detection method for avionics system state | |
| Bleu-Laine et al. | Predicting adverse events and their precursors in aviation using multi-class multiple-instance learning | |
| CN117995027A (en) | Flight plan drop-preparation recommendation method, system, equipment and medium | |
| EP3462266B1 (en) | Aircraft maintenance message prediction | |
| Wang et al. | Airborne sensor data-based unsupervised recursive identification for UAV flight phases | |
| CN114331052A (en) | Determination method of aviation operation risk, computing device and storage medium | |
| WO2019110957A1 (en) | Estimating fatigue damage in a structure | |
| CN117972336A (en) | Flight safety risk assessment method and device, electronic equipment and storage medium | |
| CN115293225B (en) | Method and device for analyzing causes of pilot flat-floating ejector rod | |
| Mathur et al. | Prediction of aviation accidents using logistic regression model | |
| Saha et al. | Forecast analysis of visibility for airport operations with deep learning techniques | |
| Zhang et al. | Pilot maneuvering performance analysis and evaluation with deep learning | |
| CN112418730A (en) | Construction method of response index estimation model of transportation system | |
| CN112529416A (en) | Construction method of operation efficiency estimation model of transportation system | |
| İnan | Aircraft Damage Classification by using Machine Learning Methods | |
| Zhang et al. | A hybrid data-driven approach to analyze aviation incident reports |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |