CN112836905B - Flight event occurrence time prediction method, device, equipment and storage medium - Google Patents
Flight event occurrence time prediction method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN112836905B CN112836905B CN202110433578.6A CN202110433578A CN112836905B CN 112836905 B CN112836905 B CN 112836905B CN 202110433578 A CN202110433578 A CN 202110433578A CN 112836905 B CN112836905 B CN 112836905B
- Authority
- CN
- China
- Prior art keywords
- flight
- data
- time
- time sequence
- training
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000003860 storage Methods 0.000 title claims abstract description 14
- 238000012549 training Methods 0.000 claims abstract description 98
- 238000013528 artificial neural network Methods 0.000 claims abstract description 16
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 238000013135 deep learning Methods 0.000 claims abstract description 11
- 230000015654 memory Effects 0.000 claims description 21
- 230000008859 change Effects 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 11
- 238000013507 mapping Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 7
- 238000004140 cleaning Methods 0.000 claims description 4
- 239000013589 supplement Substances 0.000 claims description 4
- 230000010354 integration Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 230000010006 flight Effects 0.000 description 4
- 230000001360 synchronised effect Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G06Q50/40—
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a flight event occurrence time prediction method, a flight event occurrence time prediction device, computer equipment and a storage medium, relates to the technical field of aviation, and is used for improving the accuracy of flight event occurrence time prediction. The method mainly comprises the following steps: acquiring historical data of flight fixed information and time sequence message information, wherein the time sequence message information comprises the real time of a flight event; preprocessing all the historical data to obtain a training data set, wherein the training data set comprises a characteristic training set and a target training set, and the target training set is a data set of real time of flight events; carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model; and predicting the flight event occurrence time according to the flight event occurrence time prediction model.
Description
Technical Field
The present application relates to the field of aviation technologies, and in particular, to a method, an apparatus, a device, and a storage medium for predicting a flight event occurrence time.
Background
On one hand, the occurrence time of each event of the flight can be accurately predicted, so that the passengers can reasonably arrange time and the efficiency is improved; on the other hand, reliable reference information can be provided for air traffic control, airports and navigation departments, so that the reasonable decision can be made, and the method has great significance for aviation travel.
Currently, rule-based expert systems predict the time of flight events. The expert system comprises a knowledge base, a database and an inference engine. In a rule-based expert system, a rule is triggered when a condition of the rule is satisfied, and then an action is performed. The database contains a set of facts for matching conditions in the knowledge base, the inference engine performs inference, and the inference engine links rules in the knowledge base with facts in the database to give an inference result.
However, rule-based expert systems also have some drawbacks. First, rule searching is inefficient, all rules are searched for each inference, and when there are many rules, the speed is slow. Second, there is no learning ability, and the system cannot automatically update the knowledge base when rule changes result in mispredictions. Finally, it is inconvenient to maintain, the relationships between the rules are not transparent, there is a lack of hierarchical knowledge expression, and it is difficult to observe how a single rule works on the entire policy.
Disclosure of Invention
The embodiment of the application provides a flight event occurrence time prediction method, a flight event occurrence time prediction device, computer equipment and a storage medium, which are used for improving the accuracy of flight event occurrence time prediction.
The embodiment of the invention provides a flight event occurrence time prediction method, which comprises the following steps:
acquiring historical data of flight fixed information and time sequence message information, wherein the time sequence message information comprises the real time of a flight event;
preprocessing all the historical data to obtain a training data set, wherein the training data set comprises a characteristic training set and a target training set, and the target training set is a data set of real time of flight events;
carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model;
and predicting the flight event occurrence time according to the flight event occurrence time prediction model.
The embodiment of the invention provides a flight event occurrence time prediction device, which is applied to a server and comprises the following components:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring historical data of flight fixed information and time sequence message information, and the time sequence message information comprises the real time of a flight event;
the processing module is used for preprocessing all the historical data to obtain a training data set, wherein the training data set comprises a characteristic training set and a target training set, and the target training set is a data set of real time of flight events;
the training module is used for carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model;
and the prediction module is used for predicting the occurrence time of the flight event according to the flight event occurrence time prediction model.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the flight event occurrence time prediction method when executing the computer program.
A computer-readable storage medium, which stores a computer program, which when executed by a processor, implements the above-described flight event occurrence time prediction method.
The invention provides a flight event occurrence time prediction method, a flight event occurrence time prediction device, computer equipment and a storage medium, wherein historical data of flight fixed information and time sequence message information are firstly obtained, and the time sequence message information comprises the actual occurrence time of a flight event; then preprocessing all the historical data to obtain a training data set, wherein the training data set comprises a characteristic training set and a target training set, and the target training set is a data set of real time of flight events; carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model; and finally, predicting the time of the flight event according to the flight event occurrence time prediction model. Compared with the current rule-based expert system for predicting the occurrence time of the flight event, the method for predicting the occurrence time of the flight event comprehensively considers the flight fixed information and the time sequence message information to predict the occurrence time of the flight event and improves the accuracy of the prediction of the occurrence time of the flight event.
Drawings
Fig. 1 is a flowchart of a flight event occurrence time prediction method according to an embodiment of the present application;
FIG. 2 is a schematic structural diagram of an LSTM cell according to an embodiment of the present application;
fig. 3 is a timing message information diagram according to an embodiment of the present application;
FIG. 4 is a diagram of flight fixing information provided in accordance with an embodiment of the present application;
FIG. 5 is a graph of integrated profile data provided in accordance with an embodiment of the present application;
fig. 6 is a block diagram illustrating a structure of a flight event occurrence time prediction apparatus according to an embodiment of the present application;
fig. 7 is a schematic diagram of a computer device according to an embodiment of the present application.
Detailed Description
In order to better understand the technical solutions described above, the technical solutions of the embodiments of the present application are described in detail below with reference to the drawings and the specific embodiments, and it should be understood that the specific features of the embodiments and the embodiments of the present application are detailed descriptions of the technical solutions of the embodiments of the present application, and are not limitations of the technical solutions of the present application, and the technical features of the embodiments and the embodiments of the present application may be combined with each other without conflict.
Referring to fig. 1, a method for predicting flight event occurrence time according to a first embodiment of the present invention is shown, the method specifically includes steps S01-S04, and the details thereof are as follows:
and S01, acquiring historical data of flight fixed information and time sequence message information.
The historical data of the time sequence message information comprises the actual time of the flight event.
In the embodiment of the present invention, the flight fixed information is determined when the flight is initialized, and generally does not change with time, for example: planned departure date, departure airport, arrival airport, flight number, airline department, planned departure time, planned arrival time, and the like; the time sequence message information is generated by triggering various flight events in the whole life cycle of the flight, is time sequence data and changes along with time change, for example: event type, event information, update time, flight status, etc., and the embodiment of the present invention is not limited in particular.
The event type includes each event of the whole flight cycle of the flight, and if the door closing time is only predicted, only the event between the flight initialization and the door closing can be taken: initialization, change of predicted takeoff time, change of check-in state, change of equipment number, change of model, change of airplane position, luggage rotary table, check-in counter, change of boarding gate, change of flight state and closing of cabin door.
Wherein the flight status includes: planning, early warning, delaying, taking off, standby landing, arrival, cancellation, return voyage, lost connection and accident. Among the states for predicting the closed door are: planning, early warning, delaying and canceling.
And S02, preprocessing all the historical data to obtain a training data set.
Specifically, the preprocessing all the historical data to obtain a training data set includes:
and S021, cleaning, feature extracting and mapping historical data of the flight fixed information and the time sequence message information to form the training data set.
In the embodiment of the invention, all historical data of flights are cleaned, feature extracted and mapped to form a training data set. The training data set was divided into a training set and a test set, 80% for training and 20% for testing. The flight event detection method further comprises a characteristic training set, a target training set, a characteristic test set and a target test set, wherein the characteristic training set and the characteristic test set comprise a plurality of characteristic data, the target training set and the target test set are labels corresponding to the characteristic data, and the labels are the actual time of the flight event due to the fact that the flight event occurrence time is predicted in the embodiment.
S022, filtering the training data set according to preset conditions.
Specifically, domestic flight data for normal takeoff is selected from all training data sets, and foreign flights and cancelled flights are filtered out.
In the embodiment of the invention, in the time sequence message information of the flight, the event type comprises each event of the whole flight period of the flight, if only the closing door time is predicted, only the event between the flight initialization and the closing door is taken: initialization, change of predicted takeoff time, change of check-in state, change of equipment number, change of model, change of airplane position, luggage rotary table, check-in counter, change of boarding gate, change of flight state and closing of cabin door. And associating the flight fixed information and the time sequence message information of the same flight according to the same planned take-off date, take-off airport, arrival airport and flight number. And finally mapping all the data into numerical values, wherein when mapping is carried out on the time, the absolute time needs to be converted into relative time, the planned takeoff time is taken as a reference point, and other time characteristic mapping values are the difference values of the planned takeoff time and the time minutes.
And S03, carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model.
The embodiment selects LSTM (Long Short-Term Memory network) to process the training data set. The LSTM model structure is a layer of LSTM network followed by a fully connected layer (FC), the output of which is the final result. The LSTM neural network is formed by connecting a plurality of identical LSTM units in series, and the structural schematic diagram of each LSTM unit is shown in fig. 2. In this embodiment, the main initial parameters of the model may be set as: the method comprises the steps of inputting a feature dimension 10, hiding a layer dimension 128, outputting a dimension 1, feature sequence length 6, training cycle times 300 and learning rate 0.01. The loss function is the Mean Square Error (MSE) and the optimization function is Adam. And (4) bringing the data into a model for training for multiple times, and learning the neural network parameters until the value of the loss function is in a reasonable interval.
Specifically, the LSTM is formed by connecting a plurality of neural network units in series, each unit screens information by a forgetting gate, a memory gate and an updating gate, and the gate is a method for controlling information circulation and comprises a sigmoid neural network layer and a pointwise multiplication operation. The output value of the sigmoid layer is between (0,1) and is used for controlling the information traffic, wherein 0 represents that all information is blocked from passing through, and 1 represents that all information is allowed to pass through. tanh is used for mapping the state information to (-1,1), and pointwise multiplication operation is carried out on the output state information and the control information output by the sigmoid to realize the gating function.
Wherein C represents a cell state for storing long-term information,
is the state of the cell that is input,
is the cell state of the output and,
is a candidate cell state. h represents a hidden state for storing short-term information,
it is the input of the hidden state that,
is the output hidden state.
Is the information that is input by the current cell.
Forget gate representation
How much information is forgotten.
Memory gate representation
How much information in it is to be saved to
。
Updating a door representation
How much information in it is to be updated to
。
W is the matrix coefficient and b is the offset. W and b are parameters that the neural network needs to learn.
And S04, predicting the flight event occurrence time according to the flight event occurrence time prediction model.
Specifically, flight fixed information and time sequence message information of the flight to be predicted are obtained, then the flight fixed information and the time sequence message information are integrated to obtain time sequence characteristic data, and the time sequence characteristic data are input into a flight event occurrence time prediction model to obtain event occurrence time of the flight to be predicted. It should be noted that the event occurrence time in this embodiment may specifically be a door closing time, a boarding time, a flight departure time, and the like, and the embodiment of the present invention is not particularly limited.
In one embodiment of the invention, the predicting the flight event occurrence time according to the flight event occurrence time prediction model includes:
and S041, acquiring a data list corresponding to the target flight identifier from the key value database, wherein the data list comprises the characteristic data inserted according to the time sequence.
The key value database stores data lists respectively corresponding to different target flight identifications, each data list stores characteristic data inserted according to time sequence, namely the most recently inserted characteristic data is arranged at the top, and the elements of the lists are corresponding to the characteristic data, so that the characteristic data of the same planned takeoff date, a takeoff airport, an arrival airport and a flight number are sequentially inserted into the same list. The target flight identifier in this embodiment may be composed of a departure date, a departure airport, an arrival airport, and a flight number.
And S042, acquiring nearly N continuous characteristic data from the data list as time sequence characteristic data, wherein the characteristic data comprises flight fixed information and time sequence message information corresponding to the target flight identification.
And reading the data list, and acquiring near N continuous characteristic data from the data list as time sequence characteristic data. Namely, a data list is obtained from the key value database according to a key 'planned takeoff date _ takeoff airport _ arrival airport _ flight number' corresponding to the current characteristic data, and the data list comprises the latest N continuous characteristic data. N is a parameter of the model representing the length of the input time series feature, N being equal to 6 in this example.
Further, if N consecutive feature data do not exist in the data list, copying the existing feature data to supplement the time series feature data, so that the number of the time series feature data satisfies N.
If the number of the elements in the feature list is smaller than N, the sequential feature data needs to be supplemented to make the number of the elements N, and the strategy of supplementation in this example is to copy the first piece of inserted feature data as missing features. For example, a flight initialization message is received, the flight fixed information is associated to form integrated feature data, and the integrated feature data is stored in a data list, at this time, only one element of the flight in the database data list is needed, so that the feature needs to be copied for 5 times, and the length of the time sequence feature data reaches 6.
And S043, inputting the time sequence characteristic data into the flight event occurrence time prediction model to predict the occurrence time of the flight event.
Specifically, the list of time series feature data is converted into a two-dimensional tensor (N, M), where N is the length of the input time series feature, in this example 6, and M is the dimension of the feature, in this example 10. And acquiring the time t of the predicted value flight event, namely closing the door when the predicted flight is t minutes before the planned takeoff time, converting the t into the absolute time in the character string format, and sending the absolute time to a message queue for other programs to use.
The embodiment not only utilizes the current state characteristics of the flight, but also digs the relevance of the front state and the rear state from the time sequence change characteristics, and provides a method for more accurately predicting the time of closing the door of the flight. In addition, the flight time sequence message data are time sequence data generated by triggering of various flight events in the whole life cycle of the flight, and the method selects a proper neural network model to process the time sequence data, so that the method can be used for predicting a plurality of event times in the whole life cycle of the flight.
It should be noted that, before obtaining the data list corresponding to the target flight identifier from the key-value database, the method further includes:
acquiring flight time sequence message information one by one from a message queue; inquiring corresponding flight fixed information according to the take-off date, the take-off airport, the arrival airport and the flight number in the time sequence message information; performing correlation integration on the time sequence message information and flight fixed information of the same flight and filtering according to preset conditions to obtain characteristic data; and storing the characteristic data into a data list corresponding to the target flight identification in the key value database, wherein the target flight identification is the departure date, the departure airport, the arrival airport and the flight number.
For example, as shown in fig. 3, the time sequence message information of flights is acquired one by one from the message queue, and as shown in fig. 4, according to the departure date, the departure airport, the arrival airport, and the flight number in the time sequence message information, the corresponding flight fixed information is queried, and as shown in fig. 5, the time sequence message information and the flight fixed information of the same flight are associated and integrated, and are filtered according to the preset conditions, so as to obtain the characteristic data.
Wherein:
planned departure date of fd
adept take-off airport
an ade to an airport
fn flight number
et event type
ia event information
it insertion time
fc aeronautics Ltd
ptd planned takeoff time
pta planned arrival time
The invention provides a flight event occurrence time prediction method, which comprises the steps of firstly, obtaining historical data of flight fixed information and time sequence message information, wherein the time sequence message information comprises the actual occurrence time of a flight event; then preprocessing all the historical data to obtain a training data set, wherein the training data set comprises a characteristic training set and a target training set, and the target training set is a data set of real time of flight events; carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model; and finally, predicting the time of the flight event according to the flight event occurrence time prediction model. Compared with the current rule-based expert system for predicting the occurrence time of the flight event, the method for predicting the occurrence time of the flight event comprehensively considers the flight fixed information and the time sequence message information to predict the occurrence time of the flight event and improves the accuracy of the prediction of the occurrence time of the flight event.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
In one embodiment, a flight event occurrence time prediction device is provided, and the flight event occurrence time prediction device corresponds to the flight event occurrence time prediction method in the above embodiment one to one. As shown in fig. 3, the functional modules of the flight event occurrence time prediction apparatus are described in detail as follows:
the acquisition module 10 is configured to acquire historical data of flight fixed information and timing sequence message information, where the timing sequence message information includes real time of occurrence of a flight event;
the processing module 20 is configured to pre-process all the historical data to obtain a training data set, where the training data set includes a feature training set and a target training set, and the target training set is a data set of real time when a flight event occurs;
the training module 30 is configured to perform deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model;
and the prediction module 40 is used for predicting the flight event occurrence time according to the flight event occurrence time prediction model.
The processing module 20 is configured to perform cleaning, feature extraction, and mapping on historical data of the flight fixed information and the timing sequence message information to form the training data set; and filtering the training data set according to preset conditions.
Specifically, the prediction module 40 is configured to:
acquiring a data list corresponding to a target flight identifier from a key value database, wherein the data list comprises characteristic data inserted according to the time sequence;
acquiring near N continuous characteristic data from the data list as time sequence characteristic data, wherein the characteristic data comprises flight fixed information and time sequence message information corresponding to a target flight identifier;
inputting the time sequence characteristic data into the flight event occurrence time prediction model to predict the occurrence time of the flight event.
Further, the processing module 20 is further configured to:
acquiring flight time sequence message information one by one from a message queue;
inquiring corresponding flight fixed information according to the take-off date, the take-off airport, the arrival airport and the flight number in the time sequence message information;
performing correlation integration on the time sequence message information and flight fixed information of the same flight and filtering according to preset conditions to obtain characteristic data;
and storing the characteristic data into a data list corresponding to the target flight identification in the key value database, wherein the target flight identification is the departure date, the departure airport, the arrival airport and the flight number.
Further, the processing module 20 is further configured to:
and if the N continuous characteristic data do not exist in the data list, copying the existing characteristic data to supplement the time sequence characteristic data, so that the number of the time sequence characteristic data meets N.
Specifically, the flight event is a closed door, and the flight fixed information is a take-off date, a take-off airport, an arrival airport, a flight number, a navigation department, a planned take-off time and a planned arrival time; the time sequence message information includes: event type, event information, update time, flight status.
For specific limitations of the flight event occurrence time prediction device, reference may be made to the above limitations of the flight event occurrence time prediction method, which is not described herein again. The various modules in the above-described apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 4. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a flight event occurrence time prediction method.
In one embodiment, a computer device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
acquiring historical data of flight fixed information and time sequence message information, wherein the time sequence message information comprises the real time of a flight event;
preprocessing all the historical data to obtain a training data set, wherein the training data set comprises a characteristic training set and a target training set, and the target training set is a data set of real time of flight events;
carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model;
and predicting the flight event occurrence time according to the flight event occurrence time prediction model.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring historical data of flight fixed information and time sequence message information, wherein the time sequence message information comprises the real time of a flight event;
preprocessing all the historical data to obtain a training data set, wherein the training data set comprises a characteristic training set and a target training set, and the target training set is a data set of real time of flight events;
carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model;
and predicting the flight event occurrence time according to the flight event occurrence time prediction model.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention 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 solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (8)
1. A method for predicting flight event occurrence time, the method comprising:
acquiring historical data of flight fixed information and time sequence message information, wherein the time sequence message information comprises the real time of a flight event; the time sequence message information is generated by triggering various flight events in the whole life cycle of the flight, is time sequence data and changes along with time change;
preprocessing all the historical data to obtain a training data set, wherein the training data set comprises a characteristic training set and a target training set, and the target training set is a data set of real time of flight events;
carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model;
acquiring a data list corresponding to a target flight identifier from a key value database, wherein the data list comprises characteristic data inserted according to the time sequence;
acquiring the latest N continuous characteristic data inserted in sequence from the data list as time sequence characteristic data, wherein the characteristic data comprises flight fixed information and time sequence message information corresponding to the target flight identification;
if N continuous characteristic data do not exist in the data list, copying a first piece of inserted characteristic data to supplement the time sequence characteristic data, so that the number of the time sequence characteristic data meets N;
converting the list of time sequence feature data into a two-dimensional tensor (N, M), wherein N is the length of the input time sequence feature, and M is the dimension of the feature;
inputting the two-dimensional tensors (N, M) to the flight event occurrence time prediction model to predict the time of flight event occurrence.
2. The method for predicting flight event occurrence time according to claim 1, wherein the preprocessing all the historical data to obtain a training data set comprises:
cleaning, feature extracting and mapping historical data of the flight fixed information and the time sequence message information to form the training data set;
and filtering the training data set according to preset conditions.
3. The method of claim 1, wherein before obtaining the data list corresponding to the target flight identifier from the key-value database, the method further comprises:
acquiring flight time sequence message information one by one from a message queue;
inquiring corresponding flight fixed information according to the takeoff date, the takeoff airport, the arrival airport and the flight number in the time sequence message information;
performing correlation integration on the time sequence message information and flight fixed information of the same flight and filtering according to preset conditions to obtain characteristic data;
and storing the characteristic data into a data list corresponding to the target flight identification in the key value database, wherein the target flight identification is the departure date, the departure airport, the arrival airport and the flight number.
4. The method for predicting the occurrence time of a flight event according to any one of claims 1 to 3, wherein the flight event is a closed door, and the flight fixed information is a departure date, a departure airport, an arrival airport, a flight number, a driver, a planned departure time and a planned arrival time; the time sequence message information includes: event type, event information, update time, and flight status.
5. An apparatus for predicting flight event occurrence time, the apparatus comprising:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring historical data of flight fixed information and time sequence message information, and the time sequence message information comprises the real time of a flight event; the time sequence message information is generated by triggering various flight events in the whole life cycle of the flight, is time sequence data and changes along with time change;
the processing module is used for preprocessing all the historical data to obtain a training data set, wherein the training data set comprises a characteristic training set and a target training set, and the target training set is a data set of real time of flight events;
the training module is used for carrying out deep learning neural network training through the training data set to obtain a flight event occurrence time prediction model;
the prediction module is used for acquiring a data list corresponding to the target flight identifier from a key value database, wherein the data list comprises characteristic data inserted according to the time sequence;
acquiring the latest N continuous characteristic data inserted in sequence from the data list as time sequence characteristic data, wherein the characteristic data comprises flight fixed information and time sequence message information corresponding to the target flight identification;
if N continuous characteristic data do not exist in the data list, copying a first piece of inserted characteristic data to supplement the time sequence characteristic data, so that the number of the time sequence characteristic data meets N;
converting the list of time sequence feature data into a two-dimensional tensor (N, M), wherein N is the length of the input time sequence feature, and M is the dimension of the feature;
inputting the two-dimensional tensors (N, M) to the flight event occurrence time prediction model to predict the time of flight event occurrence.
6. The flight event occurrence time prediction device according to claim 5, wherein the processing module is configured to perform cleaning, feature extraction, and mapping on historical data of the flight fixed information and the timing message information to form the training data set; and filtering the training data set according to preset conditions.
7. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor when executing the computer program implements a flight event occurrence time prediction method according to any one of claims 1 to 4.
8. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements a flight event occurrence time prediction method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110433578.6A CN112836905B (en) | 2021-04-22 | 2021-04-22 | Flight event occurrence time prediction method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110433578.6A CN112836905B (en) | 2021-04-22 | 2021-04-22 | Flight event occurrence time prediction method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112836905A CN112836905A (en) | 2021-05-25 |
| CN112836905B true CN112836905B (en) | 2021-08-03 |
Family
ID=75929840
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110433578.6A Active CN112836905B (en) | 2021-04-22 | 2021-04-22 | Flight event occurrence time prediction method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112836905B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116306937B (en) * | 2023-03-22 | 2023-11-10 | 中航信移动科技有限公司 | Rule extraction method, medium and device based on time sequence offline data |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108898838A (en) * | 2018-08-03 | 2018-11-27 | 首都经济贸易大学 | A kind of aerodrome traffic congestion prediction technique and device based on LSTM model |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180012499A1 (en) * | 2016-07-07 | 2018-01-11 | Passur Aerospace, Inc. | System and Method for Predicting Aircraft Taxi Time |
| CN107818382B (en) * | 2017-10-27 | 2021-01-05 | 中航信移动科技有限公司 | Flight arrival time prediction method |
| CN109448445A (en) * | 2018-09-05 | 2019-03-08 | 南京航空航天大学 | Flight based on shot and long term Memory Neural Networks is delayed grading forewarning system method |
| CN110459056A (en) * | 2019-08-26 | 2019-11-15 | 南通大学 | A kind of public transport arrival time prediction technique based on LSTM neural network |
| CN110807558A (en) * | 2019-11-06 | 2020-02-18 | 深圳微品致远信息科技有限公司 | Method and device for predicting departure taxi time based on deep neural network |
| CN111401601B (en) * | 2019-12-23 | 2023-06-09 | 南京航空航天大学 | Delay propagation-oriented flight take-off and landing time prediction method |
| CN111178628B (en) * | 2019-12-30 | 2023-09-19 | 沈阳民航东北凯亚有限公司 | Luggage arrival time prediction method and device |
| CN111599219B (en) * | 2020-05-27 | 2021-12-24 | 中航信移动科技有限公司 | Multi-data-source flight takeoff time prediction method based on sequencing learning |
-
2021
- 2021-04-22 CN CN202110433578.6A patent/CN112836905B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108898838A (en) * | 2018-08-03 | 2018-11-27 | 首都经济贸易大学 | A kind of aerodrome traffic congestion prediction technique and device based on LSTM model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112836905A (en) | 2021-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11017677B2 (en) | Decision-making aid for revising a flight plan | |
| Carvalho et al. | On the relevance of data science for flight delay research: a systematic review | |
| Baumann et al. | Modeling of aircraft fuel consumption using machine learning algorithms | |
| Choi et al. | Artificial neural network models for airport capacity prediction | |
| GB2504081A (en) | Assessing performance of a system | |
| US20150186335A1 (en) | Assessing performance of a system | |
| CN112836905B (en) | Flight event occurrence time prediction method, device, equipment and storage medium | |
| CN110110804A (en) | Flight control system method for predicting residual useful life based on CNN and LSTM | |
| Ahmad et al. | Intelligent framework for automated failure prediction, detection, and classification of mission critical autonomous flights | |
| Ulm et al. | Functional federated learning in erlang (ffl-erl) | |
| US9153138B1 (en) | Agent-based airfield conflict resolution | |
| Dalmau et al. | A machine learning approach to predict the evolution of air traffic flow management delay | |
| Tamatsukuri et al. | Guaranteed satisficing and finite regret: Analysis of a cognitive satisficing value function | |
| Deshmukh et al. | Temporal logic learning-based anomaly detection in metroplex terminal airspace operations | |
| Mattioli et al. | Towards a holistic approach for AI trustworthiness assessment based upon aids for multi-criteria aggregation | |
| Laperrière-Robillard et al. | Supervised learning for maritime search operations: An artificial intelligence approach to search efficiency evaluation | |
| Lu et al. | Flight delay prediction using gradient boosting machine learning classifiers | |
| Zehai et al. | Remaining useful life prediction of integrated modular avionics using ensemble enhanced online sequential parallel extreme learning machine | |
| Mathur et al. | Prediction of aviation accidents using logistic regression model | |
| Hu et al. | Online updating with a probability-based prediction model using expectation maximization algorithm for reliability forecasting | |
| CN116596107A (en) | Method for predicting residual life of equipment in multiple failure modes based on small sample learning | |
| US20220406199A1 (en) | Method and device for supervising a traffic control system | |
| Wang | A dynamic fault tree approach for time-dependent logical modeling of autonomous flight systems | |
| Bandyopadhyay et al. | Prediction of air flight cancellation during COVID-19 using deep learning methods | |
| Fávero et al. | Analogy-based effort estimation: A systematic mapping of literature |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |