CN109583593B - Low-altitude wind shear identification method based on automatic meteorological station - Google Patents
Low-altitude wind shear identification method based on automatic meteorological station Download PDFInfo
- Publication number
- CN109583593B CN109583593B CN201811285720.1A CN201811285720A CN109583593B CN 109583593 B CN109583593 B CN 109583593B CN 201811285720 A CN201811285720 A CN 201811285720A CN 109583593 B CN109583593 B CN 109583593B
- Authority
- CN
- China
- Prior art keywords
- wind
- data
- shear
- time
- wind shear
- 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 32
- 238000012544 monitoring process Methods 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000010801 machine learning Methods 0.000 claims abstract description 9
- 238000003908 quality control method Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims description 24
- 230000008859 change Effects 0.000 claims description 15
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 238000002790 cross-validation Methods 0.000 claims description 4
- 238000003066 decision tree Methods 0.000 claims description 4
- 238000007477 logistic regression Methods 0.000 claims description 4
- 238000012706 support-vector machine Methods 0.000 claims description 4
- 238000013479 data entry Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001360 synchronised 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P13/00—Indicating or recording presence, absence, or direction, of movement
- G01P13/02—Indicating direction only, e.g. by weather vane
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P5/00—Measuring speed of fluids, e.g. of air stream; Measuring speed of bodies relative to fluids, e.g. of ship, of aircraft
Landscapes
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Aerodynamic Tests, Hydrodynamic Tests, Wind Tunnels, And Water Tanks (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a low-altitude wind shear identification method based on an automatic weather station, which comprises the following steps: acquiring historical data of each wind element in a monitoring area; performing quality control and time synchronization on historical data to obtain a data list; selecting effective data of actually generated wind shear, and calculating the shear value of each wind element; scaling a standard value according to the distance between every two stations of the automatic meteorological station, and comparing the standard value with a wind shear judgment threshold value to obtain a label for judging whether wind shear occurs; using each wind element as input, using a tag of whether wind shear occurs as a result to form a data set, and using a SMOTE algorithm to balance the number of the two tag types; training by using a machine learning method, and outputting a recognition model; and reading the latest data of each automatic weather station, and inputting the latest data into the identification model to obtain an identification result. The method has the advantages of high identification precision and good local adaptability, can effectively identify the horizontal wind shear near the ground of the runway, and improves the guarantee efficiency of meteorological service.
Description
Technical Field
The invention relates to the field of data analysis of an automatic weather station, in particular to a low-altitude wind shear identification method based on the automatic weather station.
Background
Wind shear is a common atmospheric phenomenon, which refers to the rapid change in wind direction (or) velocity at the same or a short distance from different elevations. The wind shear may be classified into a horizontal shear of a horizontal wind, a vertical shear of a horizontal wind, and a shear of a vertical wind according to a change of a wind direction. In which wind shear, occurring mainly during the takeoff and landing phases of an aircraft, below a flight height of 600 meters, is called low-altitude wind shear. According to the statistics of the world meteorological organization and the international civil aviation organization, the low-altitude wind shear is the weather phenomenon which has the greatest threat to the flight safety of the airplane in the take-off and landing stage, and many major air accident accidents are caused by the low-altitude wind shear. While at aircraft landing, it is the heaviest of the observations to determine wind field conditions below altitude.
At present, the detection and identification of wind shear at home and abroad mainly depend on the distribution and the change condition of a space wind field in a detection area of related equipment. The automatic weather stations are generally deployed on two sides of a runway and used for visually acquiring a near-automatic horizontal wind field near the runway, so that ground wind horizontal shear can be calculated by using data of the automatic weather stations in a traditional wind shear identification method. Among them, the mature and effective system is the LLWAS system in the united states, but it has high requirements for the number and location of the deployment of the ground wind measuring devices, and has insufficient pertinence to local terrain, climate characteristics, and device characteristics. At present, ground wind measuring devices with the quantity and the installation positions required by the LLWAS are not arranged in the construction of many medium and small airports, the local adaptability of the LLWAS wind shear algorithm is not verified and reliable, the actual business mainly depends on the judgment of a forecaster, and the accuracy and the coverage degree still have a space for improvement.
Disclosure of Invention
According to the problems in the prior art, the invention provides the low-altitude wind shear identification method based on the automatic weather station, which has the advantages of high identification precision and good local adaptability, can effectively identify the horizontal wind shear on the ground close to the runway, and improves the guarantee efficiency of weather service.
The invention adopts the following technical scheme:
a low-altitude wind shear identification method based on an automatic weather station is characterized by comprising the following steps:
s1, acquiring historical data of each wind element acquired by all automatic weather stations in the monitoring area;
s2, performing quality control on the historical data to remove abnormal data to obtain effective data of each wind element, and performing time synchronization on the effective data according to set data time to obtain a data list corresponding to the effective data;
s3, selecting effective data in a time range set before and after the actual wind shear occurrence time in the data list, and calculating to obtain the shear value of each data time of each wind element corresponding between every two stations of all automatic meteorological stations in the monitoring area;
s4, scaling a standard value according to the distance between every two stations of the automatic meteorological stations in the monitoring area, comparing the scaled standard value with a wind shear judgment threshold value to obtain whether wind shear occurs at each data time, and giving a label whether wind shear occurs at each data time;
s5, forming a data set by taking each wind element of data at each moment as input and taking a label of whether wind shear occurs as a result, and expanding the data set by using an SMOTE algorithm to balance the number of the two types of labels of whether wind shear occurs to obtain a balanced data set;
s6, training the balanced data set by using a machine learning method, selecting the training result with the best recall ratio, and outputting a corresponding model as an identification model;
and S7, reading the latest data of the respective pneumatic stations in the monitoring area in real time, preprocessing the latest data by adopting the method of the step S2, and inputting the processed data into the recognition model to further obtain a recognition result.
Preferably, in step S1, the wind factors include an instantaneous wind speed, an instantaneous wind direction, a gust wind speed, a gust wind direction, a two-minute average wind speed, a two-minute average wind direction, a two-minute maximum wind speed, a two-minute maximum wind direction, a ten-minute average wind speed, a ten-minute average wind direction, a ten-minute maximum wind speed, and a ten-minute maximum wind direction.
Further preferably, in step S2, the quality control of the historical data includes data standardization check processing, climate limit check processing, extreme value range check processing, internal consistency check processing, and time consistency check processing; the data standardization check processing refers to deleting repeated data appearing in the historical data, and perfecting and unifying the file name, date and time corresponding to the historical data and the site information of the automatic weather station; the checking treatment of the climatological limit value refers to removing the historical data which are unlikely to appear according to the relevant standards of the meteorological industry and the local historical climatic characteristics of the monitored area; the extreme value range checking processing means that historical monthly average wind speed collected by each automatic meteorological station is selected as a basic unit, the standard deviation of the monthly average wind speed is calculated, the maximum wind speed plus twice the standard deviation of the monthly history is taken as a maximum value, the standard deviation minus twice the minimum wind speed of the monthly history is taken as a minimum value, and historical data which exceed the maximum value and are smaller than the minimum value are removed; the internal consistency checking treatment specifically refers to checking physical relations among the wind elements, namely abnormal data are removed according to judgment standards that the maximum wind speed is more than or equal to the maximum wind speed of two minutes for ten minutes, the maximum wind speed of two minutes is more than or equal to the instantaneous wind speed, and the wind speed is less than or equal to 0.2m/s when the wind direction is 0 degrees; the time consistency checking processing specifically refers to checking the change rate of corresponding historical data of a certain current time and five minutes before the time, eliminating the corresponding historical data of which the change rate is greater than an upper limit threshold, checking the change rate of corresponding historical data of a certain current time and one hour before the time, and eliminating the corresponding historical data of which the change rate is less than a lower limit threshold, wherein the upper limit threshold and the lower limit threshold are determined according to local historical climate characteristics of a monitored area.
More preferably, in step S2, the time synchronization of the valid data according to the set data time specifically means that a data list of the valid data is established with every half minute as the data time, that is, each entry in the valid data is collected into a data list corresponding to the half minute time closest to the original time.
Preferably, in step S3, effective data in twelve hours before and after the time when wind shear actually occurs in the data list are selected, an instantaneous vector wind is obtained according to an instantaneous wind speed and an instantaneous wind direction in the effective data, a vector gust is obtained according to a gust wind speed and a gust wind direction, a two-minute average vector wind is obtained according to a two-minute average wind speed and a two-minute average wind direction, a ten-minute average vector wind is obtained according to a ten-minute average wind speed and a ten-minute average wind direction, and vector differences of the instantaneous vector wind, the vector gust, the two-minute average vector wind, and the ten-minute average vector wind between all automatic weather stations in the monitoring area are calculated, so as to obtain a shear value of each data time between all automatic weather stations.
More preferably, in step S4, it is determined that wind shear occurs when the wind shear value between the two automated weather stations is greater than 15 hours, based on the international criterion for wind shear, that is, when the distance between the two automated weather stations is 4km, and the international criterion is set to float 10% to 20% below the threshold value for wind shear determination.
More preferably, in step S4, the distance between two stations of the automatic weather station in the monitoring area is scaled by a standard value, the scaled standard value is compared with a wind shear determination threshold, if the shear value of the instantaneous wind at a certain data time is greater than the wind shear determination threshold, or the shear value of the gust is greater than the wind shear determination threshold, the shear value of the average wind for two minutes is greater than the wind shear determination threshold, and the difference between the shear value of the instantaneous wind and the wind shear determination threshold is within 20%, the wind shear occurs at the data time, otherwise, the wind shear does not occur; and giving a corresponding label to whether wind shear occurs at each data moment, namely giving a label of 'yes' when wind shear occurs, and giving a label of 'no' when wind shear does not occur.
Still more preferably, in step S5, the balance between the numbers of the two types of labels whether wind shear occurs is specifically that the difference between the number of data labeled "yes" and the number of data labeled "no" is less than 5%.
More preferably, step S6 specifically includes inputting the balanced data set to train by using a machine learning method and using three models, i.e., a support vector machine, a logistic regression, and a decision tree, respectively, and performing cross validation by using 10-fold in the training process, selecting the training result with the best recall ratio, and outputting the corresponding model, which is used as the recognition model.
The invention has the advantages and beneficial effects that:
1) firstly, acquiring historical data of each wind element of an automatic meteorological station in a monitoring area, performing quality control on the historical data, and performing time synchronization on effective data according to set data time to obtain a data list corresponding to the effective data; selecting effective data in a time range set before and after the actual wind shear occurrence time in the data list, and calculating to obtain a shear value of each wind element at each data time; scaling a standard value according to the distance between every two stations of the automatic meteorological stations in the monitoring area, comparing the scaled standard value with a wind shear judgment threshold value, further obtaining whether wind shear occurs at each data time, and giving a label whether wind shear occurs at each data time; using each wind element of the data at each moment as input, using a tag of whether wind shear occurs as a result to form a data set, and expanding the data set by using an SMOTE algorithm to balance the number of the two tag types of whether wind shear occurs to obtain a balanced data set; then, training the balanced data set by using a machine learning method, selecting a training result with the best recall ratio, and outputting a corresponding model which is used as an identification model; and finally, reading the latest data of each pneumatic meteorological station in the monitoring area in real time, preprocessing the latest data by adopting the quality control method, and inputting the processed data into the identification model to further obtain an identification result. The method adopts the automatic weather station data to analyze and train, can effectively identify the low-altitude horizontal wind shear, innovatively adopts a machine learning mode to analyze historical data of each wind element compared with the traditional wind shear identification technology based on the automatic weather station data, and comprises consideration of influence factors which are difficult to directly correct or not clear in the existing research such as local terrain, weather characteristics, station position characteristics, data characteristics of each measuring device and the like in a monitored area, so that the accuracy of low-altitude wind shear identification is greatly improved, and powerful support is provided for weather guarantee service.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, a low-altitude wind shear identification method based on an automatic weather station includes the following steps:
s1, acquiring historical data of each wind element acquired by all automatic weather stations in the monitoring area;
specifically, the wind factors include an instantaneous wind speed, an instantaneous wind direction, a gust wind speed, a gust wind direction, a two minute average wind speed, a two minute average wind direction, a two minute maximum wind speed, a two minute maximum wind direction, a ten minute average wind speed, a ten minute average wind direction, a ten minute maximum wind speed, and a ten minute maximum wind direction.
S2, performing quality control on the historical data to remove abnormal data to obtain effective data of each wind element, and performing time synchronization on the effective data according to set data time to obtain a data list corresponding to the effective data;
specifically, the quality control of the historical data comprises data standardization check processing, climate limit check processing, extreme value range check processing, internal consistency check processing and time consistency check processing; the data standardization check processing refers to deleting repeated data appearing in the historical data, and perfecting and unifying the file name, date and time corresponding to the historical data and the site information of the automatic weather station; the checking treatment of the climatological limit value refers to removing the historical data which are unlikely to appear according to the relevant standards of the meteorological industry and the local historical climatic characteristics of the monitored area; the extreme value range checking processing means that historical monthly average wind speed collected by each automatic meteorological station is selected as a basic unit, the standard deviation of the monthly average wind speed is calculated, the maximum wind speed plus twice the standard deviation of the monthly history is taken as a maximum value, the standard deviation minus twice the minimum wind speed of the monthly history is taken as a minimum value, and historical data which exceed the maximum value and are smaller than the minimum value are removed; the internal consistency checking treatment specifically refers to checking physical relations among the wind elements, namely abnormal data are removed according to judgment standards that the maximum wind speed is more than or equal to the maximum wind speed of two minutes for ten minutes, the maximum wind speed of two minutes is more than or equal to the instantaneous wind speed, and the wind speed is less than or equal to 0.2m/s when the wind direction is 0 degrees; the time consistency checking processing specifically refers to checking the change rate of corresponding historical data of a certain current time and five minutes before the time, eliminating the corresponding historical data of which the change rate is greater than an upper limit threshold, checking the change rate of corresponding historical data of a certain current time and one hour before the time, and eliminating the corresponding historical data of which the change rate is less than a lower limit threshold, wherein the upper limit threshold and the lower limit threshold are determined according to local historical climate characteristics of a monitored area.
Specifically, the time synchronization of the valid data according to the set data time refers to establishing a data list of the valid data with every half minute as the data time, that is, each entry in the valid data is collected into a data list corresponding to the half minute time closest to the original time.
S3, selecting effective data in a time range set before and after the actual wind shear occurrence time in the data list, and calculating to obtain the shear value of each data time of each wind element corresponding between every two stations of all automatic meteorological stations in the monitoring area;
specifically, effective data in twelve hours before and after the actual wind shear occurrence moment in the data list are selected, instantaneous vector wind is obtained according to instantaneous wind speed and instantaneous wind direction in the effective data, vector gust is obtained according to gust wind speed and gust wind direction, two-minute average vector wind is obtained according to two-minute average wind speed and two-minute average wind direction, ten-minute average vector wind is obtained according to ten-minute average wind speed and ten-minute average wind direction, and vector differences of instantaneous vector wind, vector gust, two-minute average vector wind and ten-minute average vector wind between all automatic meteorological stations in the monitoring area are calculated, and then the shear value of each data moment between all automatic meteorological stations is obtained.
S4, scaling a standard value according to the distance between every two stations of the automatic meteorological stations in the monitoring area, comparing the scaled standard value with a wind shear judgment threshold value to obtain whether wind shear occurs at each data time, and giving a label whether wind shear occurs at each data time;
specifically, according to the international judgment standard of wind shear, namely when the distance between two automatic meteorological stations is 4km and the wind shear value between the two automatic meteorological stations is more than 15 hours, the wind shear is judged to occur, and the international judgment standard floats downwards by 10% to 20%, and is used as the wind shear judgment threshold value.
Scaling a standard value according to the distance between every two stations of the automatic meteorological station in the monitoring area, comparing the scaled standard value with a wind shear discrimination threshold value, if the shear value of instantaneous wind at a certain data moment is greater than the wind shear discrimination threshold value, or the shear value of gust is greater than the wind shear discrimination threshold value, the shear value of two-minute average wind is greater than the wind shear discrimination threshold value, and the difference value of the shear value of the instantaneous wind and the wind shear discrimination threshold value is within 20%, then wind shear is generated at the data moment, otherwise, wind shear is not generated; and giving a corresponding label to whether wind shear occurs at each data moment, namely giving a label of 'yes' when wind shear occurs, and giving a label of 'no' when wind shear does not occur.
S5, forming a data set by taking each wind element of data at each moment as input and taking a label of whether wind shear occurs as a result, and expanding the data set by using an SMOTE algorithm to balance the number of the two types of labels of whether wind shear occurs to obtain a balanced data set;
specifically, the balance between the number of the two types of the tags whether wind shear occurs is that the difference between the number of the data tagged with "yes" and the number of the data tagged with "no" is less than 5%.
S6, training the balanced data set by using a machine learning method, selecting the training result with the best recall ratio, and outputting a corresponding model as an identification model;
specifically, a machine learning method is utilized, three models of a support vector machine, a logistic regression model and a decision tree are respectively adopted, a balanced data set is input for training, cross validation is carried out by adopting 10 folds in the training process, a training result with the best recall ratio is selected, a corresponding model is output, and the model is used as an identification model.
And S7, reading the latest data of the respective pneumatic stations in the monitoring area in real time, preprocessing the latest data by adopting the method of the step S2, and inputting the processed data into the recognition model to further obtain a recognition result.
The process of the present invention is illustrated below with reference to examples.
8 pieces of automatic weather station equipment are arranged around a new bridge international airport runway in Hefei city, and an automatic weather station A and an automatic weather station B which are positioned at the north and south ends of the runway are taken as an example to explain the specific implementation mode of the method.
The distance between the automatic weather station A and the automatic weather station B is about 2.7km, historical data of the automatic weather station A and the automatic weather station B from station building to the present are collected, the historical data are preprocessed according to the method of the invention, the unreliable data are removed, the historical data are synchronized in time, and the available data are classified into the same time for utilization.
And paying attention to a time period which is 12 hours before and after the actual wind shear occurrence time of the report, and extracting concerned wind element information, wherein the concerned wind element information comprises instantaneous wind speed, wind direction, gust wind speed and wind direction, two-minute average wind speed and wind direction, ten-minute wind speed and wind direction. Scaling by distance with the international universal standard (shear greater than 15 knots (about 7.7m/s) over 4 km) and floating down 10%, i.e. the threshold W ═ 7.7 × (2.7/4) × 0.9 ≈ 4.6 m/s. And calculating the shear value of each wind element of the two stations, and using the threshold value W as a criterion for judging the shear of each wind element of the two stations of the automatic meteorological station A and the automatic meteorological station B.
In the judging process, considering the shear value of instantaneous wind, the shear value of gust wind, the shear value of two-minute average wind and the shear value of ten-minute average wind from large to small in sequence according to the weight by combining the experience of a forecaster, wherein the shear value of the instantaneous wind is larger than a wind shear judging threshold value, or the shear value of the gust wind is larger than the wind shear judging threshold value, the shear value of the two-minute average wind is larger than the wind shear judging threshold value, and the difference value between the shear value of the instantaneous wind and the wind shear judging threshold value is within 20%, marking as wind shear, otherwise, marking as no wind shear. The data set is formed by reporting wind shear tags of wind elements and markers of the automatic weather station A and the automatic weather station B which are 12 hours before and after the time when the actual wind shear is confirmed.
The data set is expanded using the SMOTE algorithm to extend the data entries marked as not windsheared to be consistent with or relatively close to the data entries marked as windsheared (the difference in the two numbers is controlled within 5%).
Inputting the expanded data set into three models of a support vector machine, a logistic regression and a decision tree for training, adopting 10-fold cross validation with recall ratio as priority, outputting an optimal result model, and taking the model as an identification model.
And acquiring real-time data of the automatic weather station A and the automatic weather station B, preprocessing the real-time data by the same method, and inputting the preprocessed real-time data into the trained recognition model to obtain a recognition result.
In conclusion, the invention provides the low-altitude wind shear identification method based on the automatic weather station, which has the advantages of high identification precision and good local adaptability, can effectively identify the horizontal wind shear on the ground near the runway, and improves the guarantee efficiency of the weather service.
Claims (9)
1. A low-altitude wind shear identification method based on an automatic weather station is characterized by comprising the following steps:
s1, acquiring historical data of each wind element acquired by all automatic weather stations in the monitoring area;
s2, performing quality control on the historical data to remove abnormal data to obtain effective data of each wind element, and performing time synchronization on the effective data according to set data time to obtain a data list corresponding to the effective data;
s3, selecting effective data in a time range set before and after the actual wind shear occurrence time in the data list, and calculating to obtain the shear value of each data time of each wind element corresponding between every two stations of all automatic meteorological stations in the monitoring area;
s4, scaling a standard value according to the distance between every two stations of the automatic meteorological stations in the monitoring area, comparing the scaled standard value with a wind shear judgment threshold value to obtain whether wind shear occurs at each data time, and giving a label whether wind shear occurs at each data time;
s5, forming a data set by taking each wind element of data at each moment as input and taking a label of whether wind shear occurs as a result, and expanding the data set by using an SMOTE algorithm to balance the number of the two types of labels of whether wind shear occurs to obtain a balanced data set;
s6, training the balanced data set by using a machine learning method, selecting the training result with the best recall ratio, and outputting a corresponding model as an identification model;
and S7, reading the latest data of the respective pneumatic stations in the monitoring area in real time, preprocessing the latest data by adopting the method of the step S2, and inputting the processed data into the recognition model to further obtain a recognition result.
2. The low altitude wind shear identification method based on the automatic weather station according to claim 1, characterized in that: in step S1, the wind factors include an instantaneous wind speed, an instantaneous wind direction, a gust wind speed, a gust wind direction, a two minute average wind speed, a two minute average wind direction, a two minute maximum wind speed, a two minute maximum wind direction, a ten minute average wind speed, a ten minute average wind direction, a ten minute maximum wind speed, and a ten minute maximum wind direction.
3. The low altitude wind shear identification method based on the automatic weather station according to claim 2, characterized in that; in step S2, performing quality control including data standardization check processing, climate limit check processing, extremum range check processing, internal consistency check processing, and time consistency check processing on the history data; the data standardization check processing refers to deleting repeated data appearing in the historical data, and perfecting and unifying the file name, date and time corresponding to the historical data and the site information of the automatic weather station; the checking treatment of the climatological limit value refers to removing the historical data which are unlikely to appear according to the relevant standards of the meteorological industry and the local historical climatic characteristics of the monitored area; the extreme value range checking processing means that historical monthly average wind speed collected by each automatic meteorological station is selected as a basic unit, the standard deviation of the monthly average wind speed is calculated, the maximum wind speed plus twice the standard deviation of the monthly history is taken as a maximum value, the standard deviation minus twice the minimum wind speed of the monthly history is taken as a minimum value, and historical data which exceed the maximum value and are smaller than the minimum value are removed; the internal consistency checking treatment specifically refers to checking physical relations among the wind elements, namely abnormal data are removed according to judgment standards that the maximum wind speed is more than or equal to the maximum wind speed of two minutes for ten minutes, the maximum wind speed of two minutes is more than or equal to the instantaneous wind speed, and the wind speed is less than or equal to 0.2m/s when the wind direction is 0 degrees; the time consistency checking processing specifically refers to checking the change rate of corresponding historical data of a certain current time and five minutes before the time, eliminating the corresponding historical data of which the change rate is greater than an upper limit threshold, checking the change rate of corresponding historical data of a certain current time and one hour before the time, and eliminating the corresponding historical data of which the change rate is less than a lower limit threshold, wherein the upper limit threshold and the lower limit threshold are determined according to local historical climate characteristics of a monitored area.
4. The method for identifying low altitude wind shear based on the automatic weather station as claimed in claim 3, wherein: in step S2, the time synchronization of the valid data according to the set data time specifically means that a data list of the valid data is established with every half minute as the data time, that is, each entry in the valid data is collected into a data list corresponding to the half minute time closest to the original time.
5. The method for identifying low altitude wind shear based on the automatic weather station as claimed in claim 4, wherein: in step S3, effective data in each twelve hours before and after the actual wind shear occurrence time in the data list are selected, instantaneous vector wind is obtained according to the instantaneous wind speed and the instantaneous wind direction in the effective data, vector gust is obtained according to the gust wind speed and the gust wind direction, two-minute average vector wind is obtained according to two-minute average wind speed and two-minute average wind direction, ten-minute average vector wind is obtained according to ten-minute average wind speed and ten-minute average wind direction, and the vector difference of the instantaneous vector wind, the vector gust, the vector difference of the two-minute average vector wind, and the vector difference of the ten-minute average vector wind between all automatic weather stations in the monitoring area are calculated, so as to obtain the shear value of each data time between all automatic weather stations.
6. The method for identifying low altitude wind shear based on the automatic weather station as claimed in claim 5, wherein: in step S4, it is determined that wind shear has occurred according to the international criterion of wind shear, that is, when the distance between two automatic weather stations is 4km and the wind shear value between two automatic weather stations is greater than 15 hours, and the international criterion is set to float 10% to 20% below, which is used as the wind shear determination threshold.
7. The method for identifying low altitude wind shear based on the automatic weather station as claimed in claim 6, wherein: step S4, scaling a standard value of the distance between two stations of the automatic weather station in the monitoring area according to a proportion, comparing the scaled standard value with a wind shear discrimination threshold, if the shear value of the instantaneous wind at a certain data time is greater than the wind shear discrimination threshold, or the shear value of gust is greater than the wind shear discrimination threshold, the shear value of the average wind for two minutes is greater than the wind shear discrimination threshold, and the difference between the shear value of the instantaneous wind and the wind shear discrimination threshold is within 20%, then wind shear occurs at the data time, otherwise, wind shear does not occur; and giving a corresponding label to whether wind shear occurs at each data moment, namely giving a label of 'yes' when wind shear occurs, and giving a label of 'no' when wind shear does not occur.
8. The method for identifying low altitude wind shear based on the automatic weather station as claimed in claim 7, wherein: in step S5, the balance between the two types of labels of whether wind shear occurs is specifically that the difference between the data labeled "yes" and the data labeled "no" is less than 5%.
9. The method for identifying low altitude wind shear based on the automatic weather station as claimed in claim 8, wherein: step S6 specifically includes inputting balanced data sets to train by using a machine learning method and using three models, i.e., a support vector machine, a logistic regression, and a decision tree, respectively, and performing cross validation by using 10-fold in the training process, selecting a training result with the best recall ratio, and outputting a corresponding model, which is used as an identification model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811285720.1A CN109583593B (en) | 2018-10-31 | 2018-10-31 | Low-altitude wind shear identification method based on automatic meteorological station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811285720.1A CN109583593B (en) | 2018-10-31 | 2018-10-31 | Low-altitude wind shear identification method based on automatic meteorological station |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109583593A CN109583593A (en) | 2019-04-05 |
| CN109583593B true CN109583593B (en) | 2020-11-13 |
Family
ID=65920911
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811285720.1A Active CN109583593B (en) | 2018-10-31 | 2018-10-31 | Low-altitude wind shear identification method based on automatic meteorological station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109583593B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113962462A (en) * | 2021-10-25 | 2022-01-21 | 中国科学院空天信息创新研究院 | Wind field stability prediction method and system based on convolutional neural network |
| CN115830380B (en) * | 2022-12-06 | 2023-08-01 | 中科三清科技有限公司 | Wind shear line identification method and device based on artificial intelligence, storage medium and terminal |
| CN115861811B (en) * | 2022-12-06 | 2023-06-20 | 中科三清科技有限公司 | Wind shear region identification method and device, storage medium and terminal |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105607063A (en) * | 2016-01-05 | 2016-05-25 | 北京无线电测量研究所 | Detection method and system for low-altitude wind shear at airport |
| CN106772387A (en) * | 2016-12-21 | 2017-05-31 | 中国航空工业集团公司雷华电子技术研究所 | A kind of wind shear recognition methods |
-
2018
- 2018-10-31 CN CN201811285720.1A patent/CN109583593B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105607063A (en) * | 2016-01-05 | 2016-05-25 | 北京无线电测量研究所 | Detection method and system for low-altitude wind shear at airport |
| CN106772387A (en) * | 2016-12-21 | 2017-05-31 | 中国航空工业集团公司雷华电子技术研究所 | A kind of wind shear recognition methods |
Non-Patent Citations (4)
| Title |
|---|
| Chaotic Oscillatory-Based Neural Network for Wind Shear and Turbulence Forecast With LiDAR Data;James N. K. Liu et al.;《IEEE》;20121130;第42卷(第6期);第1412-1423页 * |
| 双流机场风切变预警系统方案设计及LLWAS算法研究;郭忠立;《气象水文海洋仪器》;20180630(第2期);第28-31页 * |
| 基于仿真雷达图像的低空风切变类型识别研究;蒋立辉;《激光与红外》;20130331;第43卷(第3期);第334-338页 * |
| 复杂地形多测风塔综合地貌及风切变拟合修正的风资源评估方法研究;迟继峰 等;《华电技术》;20121130;第34卷(第11期);第75-81页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109583593A (en) | 2019-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109583593B (en) | Low-altitude wind shear identification method based on automatic meteorological station | |
| CN106771856B (en) | Electric power transmission line lightning stroke point determination method based on unmanned aerial vehicle technology | |
| CN109376605B (en) | Electric power inspection image bird-stab-prevention fault detection method | |
| CN106017551A (en) | Intelligent transmission line integrated monitoring analysis and early warning method | |
| CN110910440B (en) | Power transmission line length determination method and system based on power image data | |
| CN105279612A (en) | Poisson distribution-based power transmission line tripping risk assessment method | |
| CN111612315A (en) | Novel power grid disastrous gale early warning method | |
| CN109215399A (en) | A kind of termination environment intelligence stream interface driver generation method | |
| CN111695506A (en) | Wind-induced foreign matter short-circuit fault early warning method and system for power transmission line | |
| Rubnich et al. | An algorithm to identify robust convective weather avoidance polygons in en route airspace | |
| CN115373403A (en) | Construction machinery equipment inspection service system | |
| CN102903018A (en) | Air speed early warning information processing method of transmission line based on geographic information system (GIS) | |
| CN111126672A (en) | High-voltage overhead transmission line typhoon disaster prediction method based on classification decision tree | |
| CN113177678A (en) | Meteorological risk early warning method and device for different types of foreign body intrusion | |
| CN107944188A (en) | Typhoon eye of wind radius discrimination method near the ground based on weather station measured data | |
| CN106218910A (en) | Gather the system of high-voltage electric power circuit screen of trees defect, method and control terminal | |
| CN113866849A (en) | Intelligent inland river shipping meteorological service guarantee system | |
| CN109460923A (en) | A kind of ice covering on transmission lines probability forecasting method | |
| CN112683916A (en) | Method and device for identifying missing or mounting error of small hardware fittings of current collecting line tower | |
| CN116363537B (en) | Method and system for identifying hidden danger of hanging objects outside transformer substation | |
| CN105223472B (en) | A kind of ultra-high-tension power transmission line Fault Locating Method based on multi-source data set theory | |
| CN105203926B (en) | A kind of method for improving super high voltage direct current electricity transmission line fault location accuracy | |
| Wang et al. | Application of Intelligent Robot Inspection System in Power Transmission Project | |
| CN108764563B (en) | Squall line wind early warning method for power transmission line | |
| CN116383646A (en) | Airplane icing case identification method based on QAR data |
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 |