CN109961192B - Target event prediction method and device - Google Patents
Target event prediction method and device Download PDFInfo
- Publication number
- CN109961192B CN109961192B CN201910267176.6A CN201910267176A CN109961192B CN 109961192 B CN109961192 B CN 109961192B CN 201910267176 A CN201910267176 A CN 201910267176A CN 109961192 B CN109961192 B CN 109961192B
- Authority
- CN
- China
- Prior art keywords
- prediction
- time
- target event
- data
- reference data
- 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 85
- 238000012549 training Methods 0.000 claims description 16
- 238000013135 deep learning Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 10
- 238000005457 optimization Methods 0.000 claims description 10
- 230000002441 reversible effect Effects 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 208000025274 Lightning injury Diseases 0.000 description 29
- 238000010586 diagram Methods 0.000 description 12
- 230000000694 effects Effects 0.000 description 9
- 230000003449 preventive effect Effects 0.000 description 9
- 230000002411 adverse Effects 0.000 description 8
- 238000004891 communication Methods 0.000 description 6
- 230000010006 flight Effects 0.000 description 3
- 230000002829 reductive effect Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001681 protective effect Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 208000033999 Device damage Diseases 0.000 description 1
- 241000422846 Sequoiadendron giganteum Species 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 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"
-
- G06Q50/40—
Abstract
The embodiment of the invention provides a target event prediction method and a target event prediction device. The method comprises the following steps: acquiring reference data of a target event of a target area; the target area comprises at least two sub-areas, and the reference data is the number of times of the sub-areas that the target event occurs in a previous prediction period of a target time; the reference data comprises at least two groups of data at different preset moments; and inputting the reference data into a preset prediction model to obtain prediction data of the target event at the preset time in a prediction period after the target time and the prediction times of the target event in each sub-region. The target event prediction method and device provided by the embodiment of the invention can be used for predicting severe weather such as thunder and lightning in the field of aviation operation.
Description
Technical Field
The embodiment of the invention relates to the technical field of data processing, in particular to a target event prediction method and a target event prediction device.
Background
In recent years, with the rapid development of the civil aviation industry, the air traffic volume is increased rapidly, and the times of various flying and landing are continuously increased. While the amount of air transportation increases, the efficiency and safety of civil air transportation also need to be improved and preserved. Specifically, the dependence degree of civil aviation flight operation on weather is high, flight delay caused by adverse weather accounts for a large proportion of the whole flight delay, and the influence of thunder and lightning weather on the flight delay is particularly serious.
Thunder is a strong discharge phenomenon in nature, belongs to explosive natural hazards, and mainly shows a discharge phenomenon accompanied by thunder and lightning, and under different ground object conditions, the thunder and lightning show different performances. In terms of general ground and object conditions of lightning activities, isolated buildings, receiving antennas, power transmission lines, big trees and other equipment with point discharge characteristics in open lands are all easy to be struck by lightning; the harm is more obvious particularly to electronic equipment, informatization equipment and the like.
Taking the current civil aviation airport as an example, more electronic devices are gradually introduced in the construction of the modern airport, such as airplane runway navigation equipment, a program controlled switch, command equipment in a tower, satellite receiving equipment, command center equipment, high-frequency transceiver equipment, radar equipment, automatic retransmission equipment, microwave communication equipment and the like, and the electronic devices all keep high sensitivity to thunder and lightning. Although the performance of the current electronic equipment is continuously improved and adjusted, the performance of the current electronic equipment is controlled within a certain voltage range in operation, so that the performance of the current electronic equipment is difficult to deal with high voltage caused by lightning; once the electronic device is in an operating state, the electronic device is affected by lightning strike, the problem of device damage is likely to occur directly, and the related data recorded by the electronic device may disappear, so that the overall operation of the airport is affected, and the trip safety of the airplane is seriously affected.
Meanwhile, major safety accidents can be caused if lightning strikes occur in the airport operation process.
At present, the traffic management means adopted in the civil aviation field generally comprise a strategic phase and a tactical phase. The strategic stage is to reasonably adjust the airline structure, reasonably arrange and set a schedule and coordinate the flight time of the non-regular flights before the flight. The tactical operation is mainly that when severe weather occurs, scheduled sequential delay is generally adopted for non-taking off flights, and measures such as enlarging flight intervals, hovering in the air for waiting or waiting for landing to a designated airport are generally adopted for taking off flights; and when the flow in the control area is overlarge, the sectors are timely opened to limit the time for the airplanes in other control areas to enter the control area. To effectively complete the tactical phase, a prediction of inclement weather is required.
Therefore, in the prior art, prediction of severe weather such as lightning and the like has important significance for aviation operation.
Disclosure of Invention
The embodiment of the invention provides a target event prediction method and device, which are used for predicting severe weather such as thunder and lightning in the field of aviation operation.
In one aspect, an embodiment of the present invention provides a target event prediction method, where the method includes:
acquiring reference data of a target event of a target area; the target area comprises at least two sub-areas, and the reference data is the number of times of the sub-areas that the target event occurs in a previous prediction period of a target time; the reference data comprises at least two groups of data at different preset moments;
inputting the reference data into a preset prediction model to obtain prediction data of the target event at the preset time in a prediction period after the target time and the prediction times of the target event in each sub-region; the prediction model is obtained by deep learning of historical reference data of the target area.
In one aspect, an embodiment of the present invention provides a target event prediction apparatus, where the apparatus includes:
the acquisition module is used for acquiring reference data of a target event of a target area; the target area comprises at least two sub-areas, and the reference data is the number of times of the sub-areas that the target event occurs in a previous prediction period of a target time; the reference data comprises at least two groups of data at different preset moments;
the prediction module is used for inputting the reference data into a preset prediction model to obtain prediction data of the target event at the preset time in a prediction period after the target time and the prediction times of the target event in each sub-region; the prediction model is obtained by deep learning of historical reference data of the target area.
In another aspect, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, a bus, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps in the target event prediction method.
In still another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the steps in the target event prediction method.
According to the target event prediction method and device provided by the embodiment of the invention, the reference data of the target event of the target area is obtained, and the reference data is input into the preset prediction model, so that the prediction data of the prediction times of the target event in each sub-area at the preset time in the next prediction period of the target event is obtained, and corresponding preventive measures are taken in advance for the target event based on the prediction data, so that the adverse effect brought by the target event is reduced; especially, in the field of aviation operation, severe weather such as thunder and lightning is predicted, on one hand, dangerous operation is stopped, and safety accidents are avoided; on the other hand, when flight delay occurs, the scheme is deployed in advance to solve the problems of passenger overstock, airport detention and the like caused by flight delay, so that the safety and the order of passenger transportation work are ensured, and the quality of service guarantee is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a flowchart illustrating a target event prediction method according to an embodiment of the present invention;
FIG. 2 is one of the schematic diagrams of a first exemplary target area scene according to the embodiment of the present invention;
FIG. 3 is a second exemplary target area scene diagram according to the embodiment of the present invention;
FIG. 4 is a GRU model diagram of a second example of an embodiment of the present invention;
FIG. 5 is a second flowchart illustrating a target event prediction method according to an embodiment of the present invention;
FIG. 6 is one of the schematic views of a lightning strike event of a third example of embodiment of the invention;
FIG. 7 is a second schematic view of a lightning strike event of a third example of an embodiment of the invention;
FIG. 8 is a diagram illustrating a fourth exemplary airport flight operation, in accordance with an embodiment of the present invention;
FIG. 9 is one of the schematic diagrams of a fifth exemplary job node according to an embodiment of the present invention;
FIG. 10 is a second schematic diagram of a fifth exemplary operation node according to the embodiment of the present invention;
fig. 11 is a schematic structural diagram of a target event prediction apparatus according to an embodiment of the present invention;
fig. 12 is a schematic structural diagram of a server according to an embodiment of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments. In the following description, specific details such as specific configurations and components are provided only to help the full understanding of the embodiments of the present invention. Thus, it will be apparent to those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the invention. In addition, descriptions of well-known functions and constructions are omitted for clarity and conciseness.
It should be appreciated that reference throughout this specification to "an embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearances of the phrase "in an embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
In various embodiments of the present invention, it should be understood that the sequence numbers of the following processes do not mean the 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 the embodiments provided herein, it should be understood that "B corresponding to a" means that B is associated with a from which B can be determined. It should also be understood that determining B from a does not mean determining B from a alone, but may be determined from a and/or other information.
Fig. 1 is a flowchart illustrating a target event prediction method according to an embodiment of the present invention.
As shown in fig. 1, the target event prediction method provided in the embodiment of the present invention specifically includes the following steps:
Wherein, the target event can be a severe weather event, such as a lightning stroke, a gust, a rainstorm and the like; the target area may be an area that needs to be protected from a target event, such as an airport; to improve the prediction accuracy, the target region is divided into a plurality of sub-regions.
As a first example, referring to fig. 2, fig. 2 shows a target region, which is divided into 50 × 50 sub-regions, each rectangular region being a sub-region; as shown in the data in each sub-region in fig. 2, the reference data records the number of times of occurrence of the target event in a prediction period before the target time for each sub-region; taking the target event as a lightning stroke event as an example, the reference data is recorded as the number of times of lightning strokes of each sub-area in the time period, 1 lightning stroke event occurs in the first row and the first column, and 7 lightning stroke events occur in the first row and the second column.
The prediction period is a preset period, and the related data of the next period of the target time are predicted through the reference data of the previous prediction period of the target time; the reference data comprises at least two groups of data at different preset moments, and time continuity exists between each group of data.
Each prediction period may be preset with at least two preset moments, each preset period having a set of data. For example, the prediction period is one hour, and every ten minutes is used as a preset time, six groups of data are included in the reference data.
The prediction model is obtained by deep learning historical reference data of the target area, wherein the historical reference data is reference data of the target area in a previous prediction period; deep learning is carried out on historical reference data in a big data form, a model is continuously trained and optimized, and a prediction model meeting a preset accuracy requirement is finally obtained, wherein the preset accuracy requirement can comprise a specific numerical value of the accuracy requirement, such as 99%; and inputting the reference data into the prediction model to obtain the prediction data of the next prediction period after the target time. The prediction data correspondingly comprises data of each preset moment, and the data is prediction of occurrence times of the target event in each sub-region.
Still referring to fig. 2, inputting the reference data in fig. 2 into a prediction model to obtain prediction data; after one group of the prediction data is mapped into the target area, the schematic diagram of the target area is shown in fig. 3; in fig. 3, the predicted data at a predetermined time indicates that a lightning strike event is predicted to occur in each of columns 3 and 4 of the first row; three lightning events are predicted to occur in column 3 of the second row and one lightning event is predicted to occur in column 4 of the second row.
After the prediction data of the next prediction period of the target area is obtained, corresponding preventive measures can be taken according to the prediction data, and the deployment can be researched and judged in advance, so that adverse effects can be actively responded.
In the above embodiment of the present invention, by obtaining reference data of a target event in a target area and inputting the reference data into a preset prediction model, prediction data of the target event at the preset time in a prediction cycle after the target time and the prediction times of the target event occurring in each sub-area is obtained, so that corresponding preventive measures are taken in advance for the target event based on the prediction data, and adverse effects caused by the target event are reduced; especially, in the field of aviation operation, severe weather such as thunder and lightning is predicted, on one hand, dangerous operation is stopped, and safety accidents are avoided; on the other hand, when flight delay occurs, the scheme is deployed in advance to solve the problems of passenger overstock, airport detention and the like caused by flight delay, so that the safety and the order of passenger transportation work are ensured, and the quality of service guarantee is improved.
Optionally, in the above embodiment of the present invention, the method includes:
and training an original model through the historical reference data to obtain the prediction model.
Wherein, the historical reference data is the reference data of the target area in the previous prediction period; taking a target event as an example of a lightning strike event, historical thunderstorm process data of a specific airport guard area for years can be collected as example reference data. And deep learning is carried out on the historical reference data by adopting a big data form, the model is continuously trained and optimized, and finally the prediction model meeting the accuracy requirement is obtained.
Specifically, the step of training an original model by the historical reference data to obtain the prediction model includes:
firstly, inputting first historical reference data into the original model to obtain preliminary prediction data; the first historical reference data is the reference data of a historical prediction period;
secondly, performing reverse optimization on the original model through second historical reference data and the preliminary prediction data to obtain an optimized model; wherein the second historical reference data is reference data of a next prediction cycle of the historical prediction cycles;
and thirdly, iteratively inputting the next first historical reference data after the current reference data into the optimized model, and performing reverse optimization until the iteration times or the prediction precision meet a preset rule to obtain the prediction model.
After the historical reference data are obtained, the historical reference data are classified according to different belonged historical prediction periods. Firstly, acquiring historical reference data of a determined historical prediction period, namely first historical reference data, and inputting the first historical reference data into the original model to obtain preliminary prediction data, wherein the preliminary prediction data is prediction data of a next prediction period of the current historical prediction period; and in the second step, second historical reference data (namely actual data) of the next prediction period is obtained, reverse optimization is carried out on the original model through the difference between the second historical reference data and the preliminary prediction data, and the original model is adjusted to enable the prediction data output by the original model to be close to the actual data. And thirdly, iteratively inputting the next first historical reference data after the current reference data into the optimized model, and circularly executing the training and optimizing processes until the iteration times or the prediction precision meets a preset rule to obtain the prediction model.
Optionally, the raw model is a gated cyclic unit GRU model.
The Gated Recurrent Unit (GRU) is a commonly used Gated Recurrent Neural Network (GRNN), and GRNN can better capture a dependency relationship with a large time step distance in a time sequence.
As a second example, referring to fig. 4, fig. 4 shows an encoding (encoder) -decoding (decoder) network model of a deep learning based GRU, and the basic architecture is shown in fig. 4; wherein, data of 10 frames of Xi (x1, x2, x3, x4, x5, x6, x7, x8, x9 and x10) are input variables, each frame of data is a group of reference data, and a preset time interval is arranged between each frame of data;
and all 10 frames of data of Yi (y1, y2, y3, y4, y5, y6, y7, y8, y9 and y10) are output variables, and each frame of data is also the preset time interval.
Xi and Yi are sample data with a certain time interval, and are matrix vectors of 50 × 50.
Still taking airport lightning protection as an example, the time interval of each frame of reference data is TCLIP minutes, for example, x1 is reference data at the first TCLIP time of the current prediction period, and y1 is prediction data at the first TCLIP time of the next prediction period; x2 is reference data at the second TCLIP time of the current prediction cycle, and y2 is prediction data at the second TCLIP time of the next prediction cycle.
The number n of model cascade encoders and decoders is determined by the longest thunderstorm protection time (TMAX /) TCLIP of the airport, for example, if TMAX is 60 minutes and TCLIP is 6, then n is 10).
Thus, the lightning situation of the future TMAX hours after the current moment is predicted, and the prediction is realized by applying the encoder-decoder model.
Referring to fig. 2, the process of obtaining sample data is as follows:
(1) constructing a 50 x 50 grid structure with the center of an airport area as the center, wherein each small square (such as 1 square kilometer) in the grid is a sub-area, collecting thunderstorm process data of each specific airport in the process of historical reference data, starting 60 minutes before the beginning of the thunderstorm process, and taking one frame every TCLIP minutes until the end time of the thunderstorm process; a number of lightning strikes in the grid will mark the box as a number, for example: if n lightning strikes in a cell, the value of the cell is marked as n, and if no lightning strikes, the value of the cell is marked as 0, so that a 50 x 50 matrix is obtained, and the first frame of data is marked as x 1.
(2) According to the continuity of the time sequence, cutting is carried out once every TCLIP minute, and data of the TCLIP minute is processed in a mode of generating sample data (sample data, namely historical reference data), so that a 50 x 50 matrix is generated. The observation window was TMAX1 hours and the label window was TMAX2 hours.
(3) TMAX1 by 10 sample data, i.e., a matrix of TMAX 10 by 50, may be generated within an observation window TMAX1 hours; within this TMAX2 hours of the label window, 10 × TMAX2 sample data, i.e., TMAX2 × 10 matrices of 50 × 50, may be generated.
Secondly, the model training and optimizing process is as follows:
(1) inputting generated sample data into a constructed GRU encoder-decoder model for model training based on years of historical reference data of a specific airport; the sample data is divided into two parts, training sample data and test sample data.
The model is repeatedly trained by using the training samples until the mean square error is stable, the trained model is optimized by using the test samples to observe the training effect of the model, and finally the prediction model meeting the accuracy requirement is obtained.
Referring to fig. 5, an embodiment of the present invention further provides a further target event prediction method, where the method specifically includes the following steps:
Wherein the target event may be a severe weather event, such as a lightning strike, a gust, a rainstorm event; the target area may be an area that needs to be protected from a target event, such as an airport; to improve the prediction accuracy, the target region is divided into a plurality of sub-regions. As a first example, referring to fig. 2, fig. 2 shows a target area, which is divided into 50 × 50 sub-areas, each rectangular area is a sub-area, as shown by the data in each sub-area in fig. 2, and the reference data records the number of times of occurrence of a target event in a prediction period of each sub-area before a target time; taking the target event as a lightning stroke event as an example, the reference data is recorded as the number of times of lightning strokes of each sub-area in the time period, 1 lightning stroke event occurs in the first row and the first column, and 7 lightning stroke events occur in the first row and the second column.
The prediction period is a preset period, and the related data of the next period of the target time are predicted through the reference data of the previous prediction period of the target time; the reference data comprises at least two groups of data at different preset moments, and each group of data has time continuity; each prediction period may be preset with at least two preset moments, each preset period having a set of data. For example, the prediction period is one hour, and every ten minutes is used as a preset time, six groups of data are included in the reference data.
The prediction model is obtained by deep learning historical reference data of the target area, wherein the historical reference data is reference data of the target area in a previous prediction period; deep learning is carried out on historical reference data in a big data form, a model is continuously trained and optimized, and a prediction model meeting the accuracy requirement is finally obtained; and inputting the reference data into the prediction model to obtain the prediction data of the next prediction period after the target time. The prediction data correspondingly comprises data of each preset moment, and the data is prediction of occurrence times of the target event in each sub-region.
Still referring to fig. 2, inputting the reference data in fig. 2 into a prediction model to obtain prediction data; after one group of the prediction data is mapped into the target area, the schematic diagram of the target area is shown in fig. 3; in fig. 3, the predicted data at a predetermined time indicates that a lightning strike event is predicted to occur in each of columns 3 and 4 of the first row; three lightning events are predicted to occur in column 3 of the second row and one lightning event is predicted to occur in column 4 of the second row.
After the prediction data of the next prediction period of the target area is obtained, corresponding preventive measures can be taken according to the prediction data, and the deployment can be researched and judged in advance, so that adverse effects can be actively responded.
the starting time is a preset moment when the target event occurs for the first time in the next preset period; the end time is a preset moment when the target event occurs for the last time in the next preset period.
Wherein, after obtaining the prediction data, for each sub-region, determining the duration of its target event, so as to take corresponding preventive measures for the target event.
The duration is a time interval between a start time of the target event and an end time of the target event; specifically, taking a lightning strike event as an example, as a third example, referring to fig. 6, the starting time is a preset time when the target event occurs for the first time in the later preset period, that is, a first preset time when the lightning strike event occurs, that is, a first lightning strike enters a sub-area of the airport protection area.
It should be noted that, if the first lightning stroke event occurs between two preset times, the lightning stroke event is divided into the previous preset time.
The ending time is a preset moment when the target event occurs for the last time in the last preset period, referring to fig. 7, and the ending time is a moment when the last lightning stroke leaves a sub-region; it should be noted that, if the last lightning stroke event occurs between two preset times, the lightning stroke event is divided into the next preset time.
Aiming at a specific airport, the thunderstorm period of an airport protection area is that in a thunderstorm process, the time point of the first lightning stroke in the airport protection is taken as the beginning of the thunderstorm process, the last lightning stroke in the airport protection is taken as the end of the thunderstorm process, and the thunderstorm time in the airport protection area is the interval between the beginning time and the ending time.
Continuing with the second example above, from a series of total TMAX by 10 Yi outputs, the predicted airport shelter thunderstorm start and end times can be determined.
Designing an array lighting _ time _ arr, wherein the array has 10 elements in total TMAX; for Yi, if there is a lightning stroke distributed in at least one sub-area covered by the airport protection area, lighting _ time _ arr [ i ] is equal to 1, which means that the airport protection area is predicted to have the lightning stroke between (i-1) × 6 minutes and i × 6 minutes, otherwise, lighting _ time _ arr [ i ] is equal to 0, which means that the airport protection area is predicted to have no lightning stroke between (i-1) × 6 minutes and i × 6 minutes.
At a predetermined time, all the arrays are shown in table 1:
table 1:
| 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 |
assuming that the current time is T0, the start time of the thunderstorm process is predicted to be T0+12 minutes, and the end time is predicted to be T0+48 minutes, i.e. the preset time with the first value of 1 is taken as the start time of the thunderstorm, and the preset time with the last value of 1 is taken as the end time.
Several situations may occur for this array:
1) table 2, the predicted thunderstorm start time and end time are both between T0 and T0+ TMAX:
table 2:
| 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 |
the thunderstorm process is predicted to start at T0+6 minutes and end at T0+48 minutes.
2) Table 3, the lightning stroke is occurring at the current time, and the predicted thunderstorm ending time is between T0 and T0+ TMAX:
table 3:
| 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
the predicted thunderstorm process end time is T0+48 minutes.
3) Table 4, the predicted thunderstorm start time is between T0 and T0+ TMAX, and the end time is after T0+ TMAX:
table 4:
| 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
the last value of the array is 1, the start time of the predicted thunderstorm process is T0+12, and the assumed end time is T0+12+ TMAX 60.
4) Table 5, the current time the thunderstorm process is occurring, the predicted thunderstorm end time is after T0+ TMAX:
table 5:
| 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
the predicted thunderstorm process end time is T0+ TMAX 60 minutes.
5) Table 6, the array values have 0 between 1, the prediction results ignore the middle 0 time, and the prediction results are given as in tables 2 to 5;
table 6:
| 0 | 0 | 1 | 1 | 0 | 1 | 0 | 1 | 0 | 0 |
table 6 converts to table 7:
table 7:
| 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 |
according to the method, the predicted thunderstorm process time of the protective zone can be obtained, wherein the predicted thunderstorm process starting time is T0+12 minutes, and the predicted thunderstorm process ending time is T0+48 minutes.
Optionally, in the foregoing embodiment of the present invention, after step 503, the method further includes:
determining the occurrence density of the target event according to the prediction data; the occurrence density is the ratio of the total occurrence number of the target event in the duration to the duration;
and determining the occurrence intensity of the target event in the sub-area according to the occurrence density and a preset intensity level threshold value.
In order to predict the occurrence intensity of the target event, determining the occurrence density according to the prediction data; determining the occurrence intensity of the target event in the sub-area according to a preset intensity level threshold; the intensity level threshold is determined from historical reference data; specifically, taking a lightning stroke event as an example, in order to effectively estimate the strength of the process of generating the thunderstorm in the target area, the generation density of the thunderstorm is defined, the generation density is measured by the lightning stroke lightning falling frequency in the defined area, and the lightning stroke has detailed lightning stroke characteristics such as position information (longitude and latitude) and the like in available historical reference data. The occurrence density is defined as the total number of lightning strikes in the airport shelter (for example, in a circle with a radius of 10KM and with the center of the airport shelter as the center) divided by the duration of the thunderstorm process (minutes), for example, a total of 250 lightning strikes occur in the airport shelter in the course of 3 hours, and the thunderstorm density is 250/(3 × 60), i.e., 1.388 times per minute.
Taking two intensity levels as an example, a strong thunderstorm level and a weak thunderstorm level respectively; collecting a large number of historical reference processes (such as 2013-2018) of an airport protective area, wherein N thunderstorm processes are total, the thunderstorm density Di of each thunderstorm process is counted, and the average value P is calculated; and after obtaining the mean value P, taking the mean value P as a threshold, and if the thunderstorm density is greater than the threshold in a certain thunderstorm process, defining the thunderstorm process as a strong thunderstorm process, otherwise, defining the thunderstorm process as a weak thunderstorm process.
Optionally, in the foregoing embodiment of the present invention, after the step of determining the occurrence intensity of the target event in the sub-area, the method further includes:
acquiring events to be executed of the sub-regions;
and determining a target execution strategy corresponding to the occurrence strength according to a preset corresponding relation between the occurrence strength and an execution strategy of the event to be executed.
After the occurrence intensity of the target event in each sub-region is determined, the event to be executed in the sub-region is determined according to the duration, and the target execution strategy is determined according to the preset corresponding relation between the occurrence intensity and the execution strategy of the event to be executed.
The event to be executed may be an event to be executed in a subsequent prediction period or duration;
taking an airport as an example, taking corresponding preventive measures according to the prediction data to guarantee the operation. As a fourth example, referring to fig. 8, an airport flight job essentially includes what is shown in fig. 8; the wheel-mounting and wheel-dismounting device comprises a plurality of main lines of events to be executed, such as the butt joint of a gallery bridge/a passenger lift car, the opening of a passenger lift door, the opening of a cargo door, the adding of cabin oil, the inspection of a machine service and the like, wherein a plurality of sub-operations are respectively connected in series on each main line.
The specific execution strategy of the event to be executed under each occurrence intensity is specified in the preset corresponding relation; the execution policy may include stopping execution, suspending execution, and continuing execution.
Some contents in the preset correspondence are shown in table 8:
table 8:
optionally, in the foregoing embodiment of the present invention, the execution policy includes suspending execution and/or stopping execution;
the method further comprises the following steps:
and adjusting the predicted execution time of the event to be executed according to the target execution strategy.
After the target execution policy is determined, the predicted execution time of the event to be executed is adjusted according to the difference of the target execution policies, for example, if the target execution policy is to suspend execution and/or stop execution, the time required to be suspended is added to the predicted execution time.
Referring to fig. 8, still taking an airport lightning strike event as an example, from an upper wheel gear to a lower wheel gear, there are a plurality of main routes of the safeguard operation flow, which are parallel to each other, and within each route, there are 1 or more branch operation serial, wherein the maximum route (the sum of all serial branch operation times on one route) determines the estimated time of the lower wheel gear, and under the thunderstorm weather condition, a new estimation needs to be made for the flight safeguard node.
And obtaining the starting time and the ending time of the thunderstorm process of the airport according to the prediction data. As a fifth example, referring to fig. 9, assuming that the relationship between the event to be executed P1 and the originally estimated guaranteed job nodes is as shown in fig. 9, and P1 in the target execution strategy is stop execution, after the lightning event is ended, the event P1 is restarted, as shown in fig. 10; and recalculating the workflow time of the main line of the P1, and finally recalculating the new gear-withdrawing time.
In order to effectively estimate the time point of the airport cooperative decision, the embodiment of the invention provides a method for predicting the starting and ending time of the lightning stroke event passing through the airport and predicting the intensity of the thunderstorm process by using a prediction model, so that the nodes of the flight guarantee process are adjusted in time, and each decision party can more efficiently utilize the airport cooperative decision system to complete the operation task.
In the embodiment of the invention, by acquiring the reference data of the target event of the target area and inputting the reference data into the preset prediction model, the prediction data of the target event at the preset time in the next prediction period of the target time and the prediction times of the target event in each sub-area is obtained, and corresponding preventive measures are taken in advance for the target event based on the prediction data, so that the adverse effect brought by the target event is reduced; especially, in the field of aviation operation, severe weather such as thunder and lightning is predicted, on one hand, dangerous operation is stopped, and safety accidents are avoided; on the other hand, when flight delay occurs, the scheme is deployed in advance to solve the problems of passenger overstock, airport detention and the like caused by flight delay, so that the safety and the order of passenger transportation work are ensured, and the quality of service guarantee is improved.
The target event prediction method provided by the embodiment of the present invention is described above, and a target event prediction apparatus provided by the embodiment of the present invention will be described below with reference to the accompanying drawings.
Referring to fig. 11, an embodiment of the present invention provides a target event prediction apparatus, including:
an obtaining module 111, configured to obtain reference data of a target event in a target area; the target area comprises at least two sub-areas, and the reference data is the number of times of the sub-areas that the target event occurs in a previous prediction period of a target time; the reference data comprises at least two groups of data at different preset moments.
Wherein the target event may be a severe weather event, such as a lightning strike, a gust, a rainstorm event; the target area may be an area that needs to be protected from a target event, such as an airport; to improve the prediction accuracy, the target region is divided into a plurality of sub-regions. As a first example, referring to fig. 2, fig. 2 shows a target area, which is divided into 50 × 50 sub-areas, each rectangular area is a sub-area, as shown by the data in each sub-area in fig. 2, and the reference data records the number of times of occurrence of a target event in a prediction period of each sub-area before a target time; taking the target event as a lightning stroke event as an example, the reference data is recorded as the number of times of lightning strokes of each sub-area in the time period, 1 lightning stroke event occurs in the first row and the first column, and 7 lightning stroke events occur in the first row and the second column.
The prediction period is a preset period, and the related data of the next period of the target time are predicted through the reference data of the previous prediction period of the target time; the reference data comprises at least two groups of data at different preset moments, and each group of data has time continuity; each prediction period may be preset with at least two preset moments, each preset period having a set of data. For example, the prediction period is one hour, and every ten minutes is used as a preset time, six groups of data are included in the reference data.
The prediction module 112 is configured to input the reference data into a preset prediction model, so as to obtain prediction data of the target event at the preset time in a prediction cycle after the target time and the prediction frequency of the target event occurring in each sub-region; the prediction model is obtained by deep learning of historical reference data of the target area.
The prediction model is obtained by deep learning historical reference data of the target area, wherein the historical reference data is reference data of the target area in a previous prediction period; deep learning is carried out on historical reference data in a big data form, a model is continuously trained and optimized, and a prediction model meeting the accuracy requirement is finally obtained; and inputting the reference data into the prediction model to obtain the prediction data of the next prediction period after the target time. The prediction data correspondingly comprises data of each preset moment, and the data is prediction of occurrence times of the target event in each sub-region.
Still referring to fig. 2, inputting the reference data in fig. 2 into a prediction model to obtain prediction data; after one group of the prediction data is mapped into the target area, the schematic diagram of the target area is shown in fig. 3; in fig. 3, the predicted data at a predetermined time indicates that a lightning strike event is predicted to occur in each of columns 3 and 4 of the first row; three lightning events are predicted to occur in column 3 of the second row and one lightning event is predicted to occur in column 4 of the second row.
After the prediction data of the next prediction period of the target area is obtained, corresponding preventive measures can be taken according to the prediction data, and the deployment can be researched and judged in advance, so that adverse effects can be actively responded.
Optionally, in the above embodiment of the present invention, the apparatus includes:
and the model training module is used for training an original model through the historical reference data to obtain the prediction model.
Optionally, in the foregoing embodiment of the present invention, the model training module includes:
the primary prediction sub-module is used for inputting first historical reference data into the original model to obtain primary prediction data; the first historical reference data is the reference data of a historical prediction period;
the reverse optimization submodule is used for performing reverse optimization on the original model through second historical reference data and the preliminary prediction data to obtain an optimized model; wherein the second historical reference data is reference data of a next prediction cycle of the historical prediction cycles;
and the iteration submodule is used for iteratively inputting the next first historical reference data after the current reference data into the optimized model, and performing reverse optimization until the iteration times or the prediction precision meet a preset rule to obtain the prediction model.
Optionally, in the foregoing embodiment of the present invention, the original model is a gated cyclic unit GRU model.
Optionally, in the above embodiment of the present invention, the apparatus further includes:
a time determination module for determining, for each of the sub-regions,
determining the duration of the target event according to the prediction data; wherein the duration is a time interval between a start time of the target event and an end time of the target event;
the starting time is a preset moment when the target event occurs for the first time in the next preset period; the end time is a preset moment when the target event occurs for the last time in the next preset period.
Optionally, in the above embodiment of the present invention, the apparatus further includes:
the density determining module is used for determining the occurrence density of the target event according to the prediction data; the occurrence density is the ratio of the total occurrence number of the target event in the duration to the duration;
and the intensity determination module is used for determining the occurrence intensity of the target event in the sub-area according to the occurrence density and a preset intensity level threshold value.
Optionally, in the foregoing embodiment of the present invention, the intensity level threshold is determined according to the historical reference data.
Optionally, in the above embodiment of the present invention, the apparatus further includes:
the strategy determining module is used for acquiring the events to be executed of the sub-regions;
and determining a target execution strategy corresponding to the occurrence strength according to a preset corresponding relation between the occurrence strength and an execution strategy of the event to be executed.
Optionally, in the foregoing embodiment of the present invention, the execution policy includes suspending execution and/or stopping execution;
the device further comprises:
and the time adjusting module is used for adjusting the predicted execution time of the event to be executed according to the target execution strategy.
In the above embodiment of the present invention, the reference data of the target event in the target area is obtained by the obtaining module 111, the prediction module 112 inputs the reference data into a preset prediction model, so as to obtain the prediction data of the prediction times of the target event occurring in each sub-area at the preset time in a prediction cycle after the target time, and based on the prediction data, corresponding preventive measures are taken in advance for the target event, so as to reduce adverse effects caused by the target event; especially, in the field of aviation operation, severe weather such as thunder and lightning is predicted, on one hand, dangerous operation is stopped, and safety accidents are avoided; on the other hand, when flight delay occurs, the scheme is deployed in advance to solve the problems of passenger overstock, airport detention and the like caused by flight delay, so that the safety and the order of passenger transportation work are ensured, and the quality of service guarantee is improved.
In another aspect, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, a bus, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps in the target event prediction method.
For example, as follows, when the electronic device is a server, fig. 12 illustrates a physical structure diagram of the server.
As shown in fig. 12, the server may include: a processor (processor)1210, a communication Interface (Communications Interface)1220, a memory (memory)1230, and a communication bus 1240, wherein the processor 1210, the communication Interface 1220, and the memory 1230 communicate with each other via the communication bus 1240. Processor 1210 may call logic instructions in memory 1230 to perform the following method:
receiving program information and real-time heart rate information of a television program sent by an intelligent television, wherein the program information comprises: program playing time and program identification; and obtaining the playing process information of the television program according to the program information and the real-time heart rate information.
In addition, the logic instructions in the memory 1230 may be implemented in software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In still another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the steps in the target event prediction method.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: 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; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. A method of target event prediction, the method comprising:
acquiring reference data of a target event of a target area; the target area comprises at least two sub-areas, and the reference data is the number of times of the sub-areas that the target event occurs in a previous prediction period of a target time; the reference data comprises at least two groups of data at different preset moments;
inputting the reference data into a preset prediction model to obtain prediction data of the target event at the preset time in a prediction period after the target time and the prediction times of the target event in each sub-region; the prediction model is obtained by deep learning historical reference data of the target area;
training an original model through the historical reference data to obtain the prediction model; the original model is a gated cycle unit GRU model, and input variables and output variables of the GRU model have sample data of time intervals; the process of obtaining the sample data comprises the following steps:
constructing a grid structure taking the center of an airport area as the center, wherein each square in the grid is a sub-area, and collecting thunderstorm process data of each specific airport in the process of historical reference data;
cutting once every TCLIP minute, processing the data of the TCLIP minute according to a sample data generation mode, and generating a matrix vector, wherein TCLIP is the time interval of each frame of reference data;
generating the sample data within a viewing window and a label window;
wherein the step of training an original model through the historical reference data to obtain the prediction model comprises:
inputting first historical reference data into the original model to obtain preliminary prediction data; the first historical reference data is the reference data of a historical prediction period;
performing reverse optimization on the original model through second historical reference data and the preliminary prediction data to obtain an optimized model; wherein the second historical reference data is reference data of a next prediction cycle of the historical prediction cycles;
iteratively inputting the next first historical reference data after the current reference data into the optimized model, and performing reverse optimization until the iteration times or the prediction precision meet a preset rule to obtain the prediction model;
after the step of obtaining the preset time of the target event in a prediction cycle after the target time and the prediction data of the prediction times of the target event in each sub-region, determining the duration of the target event for each sub-region according to the prediction data; wherein the duration is a time interval between a start time of the target event and an end time of the target event;
the starting time is a preset moment when the target event occurs for the first time in the next preset period; the end time is a preset moment when the target event occurs for the last time in the next preset period;
wherein the duration of the target event is determined by setting an array comprising a plurality of elements;
each value in the array is a data record, and when a target event occurs in any one sub-area, the value in the array is recorded to be 1;
continuously recording data meeting a preset time period for multiple times, and determining the duration of the target event by recording the values in the array;
the starting time is a preset time when the target event occurs for the first time in the next preset period, and the ending time is a preset time when the target event occurs for the last time in the next preset period, and specifically includes:
and when the value of the first occurrence in the array is 1, the time recorded by the value is the starting time of the target event, and when the value of the last occurrence in the array is 1, the time recorded by the value is the ending time of the target event.
2. The method of claim 1, wherein after the step of determining the duration of the target event based on the prediction data, the method further comprises:
determining the occurrence density of the target event according to the prediction data; the occurrence density is the ratio of the total occurrence number of the target event in the duration to the duration;
and determining the occurrence intensity of the target event in the sub-area according to the occurrence density and a preset intensity level threshold value.
3. The method of claim 2, wherein the intensity level threshold is determined from the historical reference data.
4. The method of claim 2, wherein after the step of determining the occurrence intensity of the target event within the sub-region, the method further comprises:
acquiring events to be executed of the sub-regions;
and determining a target execution strategy corresponding to the occurrence strength according to a preset corresponding relation between the occurrence strength and an execution strategy of the event to be executed.
5. The method of claim 4, wherein the execution policy comprises suspending execution and/or stopping execution;
the method further comprises the following steps:
and adjusting the predicted execution time of the event to be executed according to the target execution strategy.
6. A target event prediction apparatus, the apparatus comprising:
the acquisition module is used for acquiring reference data of a target event of a target area; the target area comprises at least two sub-areas, and the reference data is the number of times of the sub-areas that the target event occurs in a previous prediction period of a target time; the reference data comprises at least two groups of data at different preset moments;
the prediction module is used for inputting the reference data into a preset prediction model to obtain prediction data of the target event at the preset time in a prediction period after the target time and the prediction times of the target event in each sub-region; the prediction model is obtained by deep learning historical reference data of the target area;
the model training module is used for training an original model through the historical reference data to obtain the prediction model; the original model is a gated cycle unit GRU model, and input variables and output variables of the GRU model have sample data of time intervals; the process of obtaining the sample data comprises the following steps:
constructing a grid structure taking the center of an airport area as the center, wherein each square in the grid is a sub-area, and collecting thunderstorm process data of each specific airport in the process of historical reference data;
cutting once every TCLIP minute, processing the data of the TCLIP minute according to a sample data generation mode, and generating a matrix vector, wherein TCLIP is the time interval of each frame of reference data;
generating the sample data within a viewing window and a label window;
wherein the step of training an original model through the historical reference data to obtain the prediction model comprises:
inputting first historical reference data into the original model to obtain preliminary prediction data; the first historical reference data is the reference data of a historical prediction period;
performing reverse optimization on the original model through second historical reference data and the preliminary prediction data to obtain an optimized model; wherein the second historical reference data is reference data of a next prediction cycle of the historical prediction cycles;
iteratively inputting the next first historical reference data after the current reference data into the optimized model, and performing reverse optimization until the iteration times or the prediction precision meet a preset rule to obtain the prediction model;
after the step of obtaining the preset time of the target event in a prediction cycle after the target time and the prediction data of the prediction times of the target event in each sub-region, determining the duration of the target event for each sub-region according to the prediction data; wherein the duration is a time interval between a start time of the target event and an end time of the target event;
the starting time is a preset moment when the target event occurs for the first time in the next preset period; the end time is a preset moment when the target event occurs for the last time in the next preset period;
wherein the duration of the target event is determined by setting an array comprising a plurality of elements;
each value in the array is a data record, and when a target event occurs in any one sub-area, the value in the array is recorded to be 1;
continuously recording data meeting a preset time period for multiple times, and determining the duration of the target event by recording the values in the array;
the starting time is a preset time when the target event occurs for the first time in the next preset period, and the ending time is a preset time when the target event occurs for the last time in the next preset period, and specifically includes:
and when the value of the first occurrence in the array is 1, the time recorded by the value is the starting time of the target event, and when the value of the last occurrence in the array is 1, the time recorded by the value is the ending time of the target event.
7. An electronic device comprising a memory, a processor, a bus, and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the target event prediction method according to any one of claims 1 to 5 when executing the program.
8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that: the program when executed by a processor implements the steps in the target event prediction method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910267176.6A CN109961192B (en) | 2019-04-03 | 2019-04-03 | Target event prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910267176.6A CN109961192B (en) | 2019-04-03 | 2019-04-03 | Target event prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109961192A CN109961192A (en) | 2019-07-02 |
| CN109961192B true CN109961192B (en) | 2021-11-26 |
Family
ID=67025595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910267176.6A Active CN109961192B (en) | 2019-04-03 | 2019-04-03 | Target event prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109961192B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110991625B (en) * | 2020-03-02 | 2020-06-16 | 南京邮电大学 | Surface anomaly remote sensing monitoring method and device based on recurrent neural network |
| CN113283589B (en) * | 2021-06-07 | 2022-07-19 | 支付宝(杭州)信息技术有限公司 | Updating method and device of event prediction system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101479751A (en) * | 2006-06-30 | 2009-07-08 | Nhn公司 | On-line e mail service system, and service method thereof |
| CN104316754A (en) * | 2014-11-06 | 2015-01-28 | 重庆大学 | Thunder and lightning counter remote monitoring early warning system and method |
| CN104915734A (en) * | 2015-06-25 | 2015-09-16 | 深圳市腾讯计算机系统有限公司 | Commodity popularity prediction method based on time sequence and system thereof |
| CN108427041A (en) * | 2018-03-14 | 2018-08-21 | 南京中科九章信息技术有限公司 | Lightning Warning method, system, electronic equipment and storage medium |
| CN108898258A (en) * | 2018-07-06 | 2018-11-27 | 广州供电局有限公司 | The analysis method and system of cascading failure in power system risk under Lightning Disaster weather |
-
2019
- 2019-04-03 CN CN201910267176.6A patent/CN109961192B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101479751A (en) * | 2006-06-30 | 2009-07-08 | Nhn公司 | On-line e mail service system, and service method thereof |
| CN104316754A (en) * | 2014-11-06 | 2015-01-28 | 重庆大学 | Thunder and lightning counter remote monitoring early warning system and method |
| CN104915734A (en) * | 2015-06-25 | 2015-09-16 | 深圳市腾讯计算机系统有限公司 | Commodity popularity prediction method based on time sequence and system thereof |
| CN108427041A (en) * | 2018-03-14 | 2018-08-21 | 南京中科九章信息技术有限公司 | Lightning Warning method, system, electronic equipment and storage medium |
| CN108898258A (en) * | 2018-07-06 | 2018-11-27 | 广州供电局有限公司 | The analysis method and system of cascading failure in power system risk under Lightning Disaster weather |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109961192A (en) | 2019-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109582034B (en) | Multitask route planning method and device and electronic equipment | |
| CN109961192B (en) | Target event prediction method and device | |
| CN108805337B (en) | Aviation operation control risk management and control method and system | |
| CN110570693A (en) | Flight operation time prediction method based on reliability | |
| Lim et al. | Re-planning of flight routes avoiding convective weather and the “three areas” | |
| CN112132366A (en) | Prediction system for flight clearance rate | |
| CN112132484A (en) | Flight recovery method and device | |
| CN113706931B (en) | Airspace flow control strategy recommendation method and device, electronic equipment and storage medium | |
| CN113408907B (en) | Threat impact analysis method, system and equipment for air traffic control system | |
| CN104656532A (en) | Water source pollution source monitoring method and water source pollution source monitoring system based on unmanned aerial vehicle | |
| Yang et al. | Stochastic scheduling of ground movement problem integrated with taxiway routing and gate/stand allocation | |
| Lopez | Vulnerability of airports on climate change: an assessment methodology | |
| JP4515980B2 (en) | Weather information provision system | |
| CN110825105B (en) | Satellite film pattern spot inspection method and device based on unmanned aerial vehicle | |
| CN116029534B (en) | Airport stand allocation method and device, electronic equipment and storage medium | |
| CN115662196B (en) | Method and device for processing information of forecast of ice accretion of airplane, electronic equipment and storage medium | |
| CN104464383B (en) | A kind of operation method for early warning and device | |
| CN114648895B (en) | Airport apron safety early warning method and system capable of reducing false alarm rate | |
| CN113570910B (en) | Air traffic flow management method and device and electronic equipment | |
| CN109215398A (en) | A kind of Path Planning for UAV and device | |
| Shao et al. | Air transportation delay propagation analysis with uncertainty in coloured–timed Petri nets | |
| CN112817333A (en) | Routing inspection planning method, device, equipment and computer readable medium | |
| CN111554126A (en) | Flight characteristic-based control instruction aided decision-making method and device | |
| CN110929881A (en) | Lightning stroke event statistical method | |
| Assaad et al. | Reducing fuel burn through air traffic flow optimisation incorporating wind models |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230116 Address after: Room 401, Floor 4, Xingfa Building, No. 45, Zhongguancun Street, Haidian District, Beijing, 100089 Patentee after: BEIJING JOZZON CAS SOFTWARE CO.,LTD. Address before: Room 501, building 2, Shuanglong science and Technology Industrial Park, No. 2, Shuanglong street, Qinhuai District, Nanjing, Jiangsu 210006 Patentee before: NANJING ZHONGKE JIUZHANG INFORMATION TECHNOLOGY Co.,Ltd. |