A laser radar real-time wind measurement data quality control system and method based on mobile platform

Abstract

The present invention relates to the field of laser radar technology, and discloses a laser radar real-time wind measurement data quality control system and method based on a mobile platform, wherein a laser radar real-time wind measurement data quality control method based on a mobile platform comprises the following steps: collecting vehicle-mounted wind measurement information; generating plane data based on the vehicle-mounted wind measurement information; inputting the plane data into a wind measurement model, and the wind measurement model outputs a wind field matrix representing the wind speed and wind direction at each position; the present invention collects the vehicle-mounted wind measurement information at a regular interval, and uses multiple vehicles to collaboratively obtain a high-precision global wind field map. Compared with the wind measurement data of a single vehicle, the wind field map obtained by the system includes the wind speed and wind direction information at each position and time.

Description

Technical Field

[0001] The present invention relates to the field of laser radar technology, and more specifically, to a laser radar real-time wind measurement data quality control system and method based on a mobile platform. Background Art

[0002] In recent years, vehicle-mounted LiDAR systems have been widely used in the field of autonomous driving. They can not only achieve high-precision environmental perception, but also obtain valuable wind measurement data by measuring the characteristics of airflow movement. However, due to changes in the surrounding environment and changes in the vehicle's motion state, the original wind measurement data often has certain errors and noise, and real-time correction is required to improve data quality and reliability. At the same time, when using multiple vehicles to collaborate on wind measurement data fusion, the wind measurement data of different vehicles may differ and be inconsistent. Summary of the invention

[0003] The purpose of the present invention is to provide a laser radar real-time wind measurement data quality control system and method based on a mobile platform in order to solve the above problems.

[0004] The present invention provides a method for analyzing an industry large model digital transformation system architecture, comprising the following steps:

[0005] Step 100, collecting vehicle-mounted wind measurement information, which is collected once at each time sampling point, including vehicle information, laser radar information and urban environment information;

[0006] Step 200, generating plane data based on vehicle-mounted wind measurement information, including a bin for storing data, a serial data stream indexed by a bin to a data object, and a relationship list for storing association relationships between data objects indexed by the bin, wherein the element value of the i-th row and j-th column of the relationship list represents a quantitative value of the association relationship between the data objects indexed by the i-th and j-th bins;

[0007] Data objects include vehicles, lidar, and urban environments;

[0008] A serial data stream No. 1 of a data object includes n data units sorted by time, and the t-th data unit represents the vehicle-mounted wind measurement information collected at the t-th moment of the data object;

[0009] Step 300, input the plane data into the wind measurement model, the wind measurement model includes a first backbone network, the first backbone network includes a first encoder, a second encoder and a first FC layer, wherein a serial data stream No. 1 is input into the first encoder, the first encoder outputs a first identification feature to the second encoder, the second encoder also inputs a relationship list, the second encoder outputs a second identification feature to the first FC layer, and the first FC layer outputs results representing wind speed and wind direction at each position.

[0010] Furthermore, the time sampling point refers to the moment at which data is collected at a predetermined time interval, and the default value of the predetermined time interval is 30 seconds.

[0011] Furthermore, the association relationship includes high and low, which correspond to two quantitative values ​​respectively; the data objects with high association relationship are defined as: the vehicle and its connected laser radar, the vehicle and its adjacent vehicles, and the vehicle and the urban environment in which it travels.

[0012] Furthermore, the entire urban area is divided into uniform grids, each grid representing a location.

[0013] Furthermore, the formula of the first encoder is as follows:

[0014] ;

[0015] ;

[0016] ;

[0017] in Represents the plane data The first bin corresponds to the first serial data stream The first identifying feature, Indicates The first bin corresponds to the first serial data stream The first identifying feature, and Respectively represent The first and second intermediate features of the facets, Indicates The first serial data stream of the bin data units, and are the first and second serial recognition weight matrices, is the first serial identification bias parameter, is the hyperbolic tangent function, represents the dot product, Represents the Sigmoid function;

[0018] The formula for the second encoder is as follows:

[0019] ;

[0020] in Indicates Layer The second facet feature of the facets, Representation and Facets A collection of facets with high associations. Indicates The first plane identification weight matrix of the layer, It is The surface element u of the layer The attention weight is calculated by the following formula:

[0021] ;

[0022] in is a learnable attention vector, represents the vector concatenation operation, Indicates The second plane identification weight matrix of the layer, represents transpose, express Activation function;

[0023] and Respectively represent Layer and The second facet feature of the facets, , F represents the total number of layers, hour , , hour is equal to the second identification feature of the face element v; and Respectively represent The object indexed by the uth and uth facets The first identifying feature, represents a total number of data items of the first serial data stream;

[0024] The formula for the first FC layer is as follows:

[0025] ;

[0026] in Represents the wind field matrix. An element of the wind field matrix represents the wind speed and direction at a location. Represents the plane data The second identification feature of the facet, Represents the collection of cells of all plane data, and FC means fully connected.

[0027] Furthermore, the loss function calculation method of the wind measurement model is:

[0028] ;

[0029] in, Indicates the wind loss value, is the number of wind measurement samples, is the number of locations where the urban area is divided. The wind measurement samples represent the actual values ​​of wind speed and direction at all locations within a time period. and Respectively represent Position No. The actual wind speed and direction of the wind measurement samples, and Respectively represent Position No. The wind speed and direction predicted by the wind measurement model for a wind measurement sample.

[0030] Further, the plane data is input into the heat island model, the heat island model includes a second backbone network, the second backbone network includes a third encoder, a fourth encoder, a feature fusion layer, a sixth encoder and a second FC layer, wherein the wind field matrix output by the first FC layer generates a second serial data stream, a data unit of the second serial data stream represents the wind field matrix information within a time period, and the order of the serial units is consistent with the time order of the wind field matrix information; the second serial data stream is input into the third encoder, the third encoder outputs the third identification feature to the feature fusion layer, the first serial data stream is input into the fourth encoder, the fourth encoder outputs the fourth identification feature to the feature fusion layer, the feature fusion layer outputs the fifth combination feature to the sixth encoder, the sixth encoder also inputs a relationship list, the sixth encoder outputs the sixth identification feature to the second FC layer, and the second FC layer outputs a result representing the probability of occurrence of the heat island effect at each location.

[0031] Furthermore, the formula of the third encoder is as follows:

[0032] ;

[0033] in Indicates A third identifying feature, Indicates A third identifying feature, Indicates the second serial data stream data units, and are the third and fourth serial recognition weight matrices, is the second serial identification bias parameter, yes Activation function;

[0034] The formula for the fourth encoder is as follows:

[0035] ;

[0036] in Represents the plane data The tth fourth identification feature of the serial data stream No. 1 corresponding to the facet, Indicates The t-1th fourth identification feature of the serial data stream No. 1 corresponding to the facet, Indicates The tth data unit of the serial data stream No. 1 of bins, and are the fifth and sixth serial recognition weight matrices, is the third serial identification bias parameter, yes Activation function;

[0037] The formula of the feature fusion layer is as follows:

[0038] ;

[0039] in, Represents the plane data The fifth combination feature corresponding to the face element is, Indicates A third identifying feature, represents the total number of data items of the second serial data stream, Represents the plane data The first bin corresponds to the first serial data stream The fourth identifying feature, represents the total number of data items of the first serial data stream, represents the feature fusion function;

[0040] The formula for the sixth encoder is as follows:

[0041] ;

[0042] Indicates Layer The sixth facet feature of the facets is Indicates The sixth panel feature of the u-th panel of the layer, and They represent the set of facets with high correlation with facet v and facet u respectively, represents the cardinality of a set, Indicates The sixth plane identification weight matrix of the layer, , Indicates the total number of layers, hour , The fifth combined feature of the data object representing the u-th facet index, hour is equal to the sixth identification feature of the face element v, is the sigmoid function;

[0043] The formula for the second FC layer is as follows:

[0044] ;

[0045] in represents the heat island matrix, and an element of the heat island matrix represents the probability of heat island effect at a location. Represents the plane data The sixth identification feature of the facets, Represents the set of facets of all plane data, and FC means fully connected.

[0046] Furthermore, the loss function calculation method of the heat island model is:

[0047] ;

[0048] in, represents the heat island loss value, is the number of heat island samples, is the number of locations where the urban area is divided. The heat island sample represents the actual heat island effect label of all locations within a time period. The value is 0 or 1. 1 indicates that the heat island effect occurs at the location, and 0 indicates that the heat island effect does not occur at the location. Indicates Position No. The actual heat island effect labels of heat island samples, The heat island model predicts Position No. The probability of occurrence of heat island effect for heat island samples, Represents the logarithmic function with the natural constant e as the base.

[0049] The present invention provides a mobile platform-based laser radar real-time wind measurement data quality control system, which is used to store computer-readable instructions. When the computer-readable instructions are read, the aforementioned mobile platform-based laser radar real-time wind measurement data quality control method can be executed.

[0050] The beneficial effect of the present invention is that the present invention collects vehicle-mounted wind measurement information at regular intervals and uses multiple vehicles to collaboratively obtain a high-precision global wind field map. Compared with single-vehicle wind measurement data, the wind field map obtained by the system includes wind speed and wind direction information at each location and time. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 The present invention is a flow chart of a method for quality control of real-time wind measurement data using a laser radar based on a mobile platform. DETAILED DESCRIPTION

[0052] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0053] At least one embodiment of the present invention discloses a method for quality control of real-time wind measurement data of a laser radar based on a mobile platform, such as Figure 1 As shown, the following steps are included:

[0054] Step 100, collecting vehicle-mounted wind measurement information, which is collected once at each time sampling point, including vehicle information, laser radar information and urban environment information;

[0055] Vehicle information includes vehicle motion status information (vehicle speed, direction angle, pitch angle and roll angle) and environmental information around the vehicle (temperature, humidity, air pressure);

[0056] LiDAR information includes raw measured wind speed and raw measured wind direction;

[0057] Urban environmental information includes urban topography, land use, and building distribution;

[0058] The time sampling point refers to the moment when data is collected at a predetermined time interval;

[0059] In one embodiment of the present invention, the default value of the predetermined time interval is 30 seconds.

[0060] Step 200, generating plane data based on vehicle-mounted wind measurement information, including a bin for storing data, a serial data stream indexed by a bin to a data object, and a relationship list for storing association relationships between data objects indexed by the bin, wherein the element value of the i-th row and j-th column of the relationship list represents a quantitative value of the association relationship between the data objects indexed by the i-th and j-th bins;

[0061] The association relationship includes high and low, corresponding to two quantitative values ​​respectively.

[0062] Data objects include vehicles, lidar, and urban environments;

[0063] The data objects with high correlation are defined as: the vehicle and its connected LiDAR, the vehicle and its neighboring vehicles, and the vehicle and the urban environment in which it is traveling;

[0064] A serial data stream No. 1 of a data object includes n data units sorted in time, and the t-th data unit represents the vehicle-mounted wind measurement information collected at the t-th moment of the data object.

[0065] Step 300, inputting plane data into a wind measurement model, the wind measurement model comprising a first backbone network, the first backbone network comprising a first encoder, a second encoder and a first FC (fully connected layer), wherein a serial data stream No. 1 is input into the first encoder, the first encoder outputs a first identification feature to the second encoder, the second encoder further inputs a relationship list, the second encoder outputs a second identification feature to the first FC layer, and the first FC layer outputs results indicating wind speed and wind direction at each position;

[0066] In one embodiment of the present invention, the entire urban area is divided into uniform grids, each grid representing a location;

[0067] In one embodiment of the present invention, the formula of the first encoder is as follows:

[0068]

[0069]

[0070]

[0071] in Represents the plane data The first bin corresponds to the first serial data stream The first identifying feature, Indicates The first bin corresponds to the first serial data stream The first identifying feature, and Respectively represent The first and second intermediate features of the facet, Indicates The first serial data stream of the bin data units, and are the first and second serial recognition weight matrices, is the first serial identification bias parameter, is the hyperbolic tangent function, represents the dot product, Represents the Sigmoid function.

[0072] In one embodiment of the present invention, the formula of the second encoder is as follows:

[0073]

[0074] in Indicates Layer The second facet feature of the facets, Representation and Facets A collection of facets with high associations. Indicates The first plane identification weight matrix of the layer, It is The surface element u of the layer The attention weight is calculated by the following formula:

[0075]

[0076] in is a learnable attention vector, represents the vector concatenation operation, Indicates The second plane identification weight matrix of the layer, represents transpose, express Activation function;

[0077] and Respectively represent Layer and The second facet feature of the facets, , F represents the total number of layers, hour , , hour is equal to the second identification feature of the face element v; and Respectively represent The object indexed by the uth and uth facets The first identifying feature, Represents the total number of data items of the first serial data stream.

[0078] In one embodiment of the present invention, the formula of the first FC layer is as follows:

[0079]

[0080] in Represents the wind field matrix. An element of the wind field matrix represents the wind speed and direction at a location. Represents the plane data The second identification feature of the facet, Represents the collection of cells of all plane data, and FC means fully connected.

[0081] In one embodiment of the present invention, the loss function calculation method of the wind measurement model is:

[0082]

[0083] in, Indicates the wind loss value, is the number of wind measurement samples, is the number of locations where the urban area is divided. The wind measurement samples represent the actual values ​​of wind speed and direction at all locations within a time period. and Respectively represent Position No. The actual wind speed and direction of the wind measurement samples, and Respectively represent Position No. The wind speed and direction predicted by the wind measurement model for a wind measurement sample.

[0084] An urban heat island effect prediction model is constructed based on the wind field matrix output by the wind measurement model to achieve accurate prediction of the urban heat island effect.

[0085] In one embodiment of the present invention, a mobile platform-based laser radar real-time wind measurement data quality control method further includes:

[0086] Step 400, inputting plane data into a heat island model, the heat island model includes a second backbone network, the second backbone network includes a third encoder, a fourth encoder, a feature fusion layer, a sixth encoder and a second FC layer, wherein the wind field matrix output by the first FC layer generates a second serial data stream, a data unit of the second serial data stream represents wind field matrix information within a time period, and the order of the serial units is consistent with the time order of the wind field matrix information; inputting the second serial data stream into the third encoder, the third encoder outputs the third identification feature to the feature fusion layer, inputting the first serial data stream into the fourth encoder, the fourth encoder outputs the fourth identification feature to the feature fusion layer, the feature fusion layer outputs the fifth combination feature to the sixth encoder, the sixth encoder also inputs a relationship list, the sixth encoder outputs the sixth identification feature to the second FC layer, and the second FC layer outputs a result representing the probability of occurrence of heat island effect at each position;

[0087] In one embodiment of the present invention, the formula of the third encoder is as follows:

[0088]

[0089] in Indicates A third identifying feature, Indicates A third identifying feature, Indicates the second serial data stream data units, and are the third and fourth serial recognition weight matrices, is the second serial identification bias parameter, yes Activation function.

[0090] In one embodiment of the present invention, the formula of the fourth encoder is as follows:

[0091]

[0092] in Represents the plane data The tth fourth identification feature of the serial data stream No. 1 corresponding to the facet, Indicates The t-1th fourth identification feature of the serial data stream No. 1 corresponding to the facet, Indicates The tth data unit of the serial data stream No. 1 of bins, and are the fifth and sixth serial recognition weight matrices, is the third serial identification bias parameter, yes Activation function.

[0093] In one embodiment of the present invention, the formula of the feature fusion layer is as follows:

[0094]

[0095] in, Represents the plane data The fifth combination feature corresponding to the face element is, Indicates A third identifying feature, represents the total number of data items of the second serial data stream, Represents the plane data The first bin corresponds to the first serial data stream The fourth identifying feature, represents the total number of data items of the first serial data stream, Represents the feature fusion function (splicing function).

[0096] In one embodiment of the present invention, the formula of the sixth encoder is as follows:

[0097]

[0098] Indicates Layer The sixth facet feature of the facets is Indicates The sixth panel feature of the u-th panel of the layer, and They represent the set of facets with high correlation with facet v and facet u respectively, represents the cardinality of a set, Indicates The sixth plane identification weight matrix of the layer, , Indicates the total number of layers, hour , The fifth combined feature of the data object representing the u-th facet index, hour is equal to the sixth identification feature of the face element v, is the sigmoid function;

[0099] In one embodiment of the present invention, the formula of the second FC layer is as follows:

[0100]

[0101] in represents the heat island matrix, and an element of the heat island matrix represents the probability of heat island effect at a location. Represents the plane data The sixth identification feature of the facets, Represents the set of facets of all plane data, and FC means fully connected.

[0102] In one embodiment of the present invention, the loss function calculation method of the heat island model is:

[0103]

[0104] in, represents the heat island loss value, is the number of heat island samples, is the number of locations where the urban area is divided. The heat island sample represents the actual heat island effect label of all locations within a time period. The value is 0 or 1. 1 indicates that the heat island effect occurs at the location, and 0 indicates that the heat island effect does not occur at the location. Indicates Position No. The actual heat island effect labels of heat island samples, The heat island model predicts Position No. The probability of occurrence of heat island effect for heat island samples, Represents the logarithmic function with the natural constant e as the base.

[0105] In at least one embodiment of the present invention, a mobile platform-based laser radar real-time wind measurement data quality control system is provided, which is used to store computer-readable instructions, and when the computer-readable instructions are read, the aforementioned mobile platform-based laser radar real-time wind measurement data quality control method can be executed.

[0106] The above describes an embodiment of the present invention, but this embodiment is not limited to the above-mentioned specific implementation mode. The above-mentioned specific implementation mode is merely illustrative and not restrictive. Under the guidance of this embodiment, ordinary technicians in this field can also make more forms of equivalent embodiments, all of which are within the protection of this embodiment.

Claims

1. A method for quality control of real-time wind measurement data using a laser radar based on a mobile platform, characterized in that: include: Step 100, collecting vehicle-mounted wind measurement information, which is collected once at each time sampling point, including vehicle information, laser radar information and urban environment information; vehicle information includes vehicle speed, direction angle, pitch angle and roll angle, and temperature, humidity and air pressure around the vehicle; laser radar information includes original measured wind speed and original measured wind direction; urban environment information includes urban topography, land use, and building distribution; Step 200, generating plane data based on vehicle-mounted wind measurement information, including a bin for storing data, a serial data stream indexed by a bin to a data object, and a relationship list for storing association relationships between data objects indexed by the bin, wherein the element value of the i-th row and j-th column of the relationship list represents a quantitative value of the association relationship between the data objects indexed by the i-th and j-th bins; Data objects include vehicles, lidar, and urban environments; A serial data stream No. 1 of a data object includes n data units sorted by time, and the t-th data unit represents the vehicle-mounted wind measurement information collected at the t-th moment of the data object; Step 300, input the plane data into the wind measurement model, the wind measurement model includes a first backbone network, the first backbone network includes a first encoder, a second encoder and a first FC layer, wherein a serial data stream No. 1 is input into the first encoder, the first encoder outputs a first identification feature to the second encoder, the second encoder also inputs a relationship list, the second encoder outputs a second identification feature to the first FC layer, and the first FC layer outputs results representing wind speed and wind direction at each position.

2. The method for quality control of real-time wind measurement data of a laser radar based on a mobile platform according to claim 1 is characterized in that: The time sampling point refers to the moment when data is collected at a predetermined time interval. The default value of the predetermined time interval is 30 seconds.

3. The method for quality control of real-time wind measurement data by a laser radar based on a mobile platform according to claim 1, characterized in that: The association relationship includes high and low, which correspond to two quantitative values ​​respectively; the data objects with high association relationship are defined as: the vehicle and its connected laser radar, the vehicle and its adjacent vehicles, and the vehicle and the urban environment in which it is traveling.

4. The method for quality control of real-time wind measurement data by a laser radar based on a mobile platform according to claim 1, characterized in that: The entire urban area is divided into uniform grids, each grid representing a location.

5. The method for quality control of real-time wind measurement data of a laser radar based on a mobile platform according to claim 1, characterized in that: The formula for the first encoder is as follows: ； ； ； in Represents the plane data The first bin corresponds to the first serial data stream The first identifying feature, Indicates The first bin corresponds to the first serial data stream The first identifying feature, and Respectively represent The first and second intermediate features of the facet, Indicates The first serial data stream of the bin data units, and are the first and second serial recognition weight matrices, is the first serial identification bias parameter, is the hyperbolic tangent function, represents the dot product, Represents the Sigmoid function; The formula for the second encoder is as follows: ； in Indicates Layer The second facet feature of the facets, Representation and Facets A collection of facets with high associations. Indicates The first plane identification weight matrix of the layer, It is The surface element u of the layer The attention weight is calculated by the following formula: ； in is a learnable attention vector, represents the vector concatenation operation, Indicates The second plane identification weight matrix of the layer, represents transpose, express Activation function; and Respectively represent Layer and The second facet feature of the facets, , F represents the total number of layers, hour , , hour is equal to the second identification feature of the face element v; and Respectively represent The object indexed by the uth and uth facets The first identifying feature, represents a total number of data items of the first serial data stream; The formula for the first FC layer is as follows: ； in Represents the wind field matrix. An element of the wind field matrix represents the wind speed and direction at a location. Represents the plane data The second identification feature of the facet, Represents the collection of cells of all plane data, and FC means fully connected.

6. The method for quality control of real-time wind measurement data of a laser radar based on a mobile platform according to claim 5, characterized in that: The loss function calculation method of the wind measurement model is: ； in, Indicates the wind loss value, is the number of wind measurement samples, is the number of locations where the urban area is divided. The wind measurement samples represent the actual values ​​of wind speed and direction at all locations within a time period. and Respectively represent Position No. The actual wind speed and direction of the wind measurement samples, and Respectively represent Position No. The wind speed and direction predicted by the wind measurement model for a wind measurement sample.

7. The method for quality control of real-time wind measurement data of a laser radar based on a mobile platform according to claim 1, characterized in that: Input the plane data into the heat island model, the heat island model includes a second backbone network, the second backbone network includes a third encoder, a fourth encoder, a feature fusion layer, a sixth encoder and a second FC layer, wherein the wind field matrix output by the first FC layer generates a second serial data stream, a data unit of the second serial data stream represents wind field matrix information within a time period, and the order of the serial units is consistent with the time order of the wind field matrix information; The second serial data stream is input into the third encoder, the third encoder outputs the third identification feature to the feature fusion layer, the first serial data stream is input into the fourth encoder, the fourth encoder outputs the fourth identification feature to the feature fusion layer, the feature fusion layer outputs the fifth combination feature to the sixth encoder, the sixth encoder also inputs the relationship list, the sixth encoder outputs the sixth identification feature to the second FC layer, and the second FC layer outputs the result representing the probability of occurrence of the heat island effect at each location.

8. The method for quality control of real-time wind measurement data of a laser radar based on a mobile platform according to claim 7, characterized in that: The formula for the third encoder is as follows: ； in Indicates A third identifying feature, Indicates A third identifying feature, Indicates the second serial data stream data units, and are the third and fourth serial recognition weight matrices, is the second serial identification bias parameter, yes Activation function; The formula for the fourth encoder is as follows: ； in Represents the plane data The tth fourth identification feature of the serial data stream No. 1 corresponding to the facet, Indicates The t-1th fourth identification feature of the serial data stream No. 1 corresponding to the facet, Indicates The tth data unit of the serial data stream No. 1 of bins, and are the fifth and sixth serial recognition weight matrices, is the third serial identification bias parameter, yes Activation function; The formula of the feature fusion layer is as follows: ； in, Represents the plane data The fifth combination feature corresponding to the face element is, Indicates A third identifying feature, represents the total number of data items of the second serial data stream, Represents the plane data The first bin corresponds to the first serial data stream The fourth identifying feature, represents the total number of data items of the first serial data stream, represents the feature fusion function; The formula for the sixth encoder is as follows: ； Indicates Layer The sixth facet feature of the facets is Indicates The sixth panel feature of the u-th panel of the layer, and They represent the set of facets with high correlation with facet v and facet u respectively, represents the cardinality of a set, Indicates The sixth plane identification weight matrix of the layer, , Indicates the total number of layers, hour , The fifth combined feature of the data object representing the u-th facet index, hour is equal to the sixth identification feature of the face element v, is the sigmoid function; The formula for the second FC layer is as follows: ； in represents the heat island matrix, and an element of the heat island matrix represents the probability of heat island effect at a location. Represents the plane data The sixth identification feature of the facets, Represents the set of facets of all plane data, and FC means fully connected.

9. The method for quality control of real-time wind measurement data of a laser radar based on a mobile platform according to claim 7, characterized in that: The loss function calculation method of the heat island model is: ； in, represents the heat island loss value, is the number of heat island samples, is the number of locations where the urban area is divided. The heat island sample represents the actual heat island effect label of all locations within a time period. The value is 0 or 1. 1 indicates that the heat island effect occurs at the location, and 0 indicates that the heat island effect does not occur at the location. Indicates Position No. The actual heat island effect labels of heat island samples, The heat island model predicts Position No. The probability of occurrence of heat island effect for heat island samples, Represents the logarithmic function with the natural constant e as the base.

10. A laser radar real-time wind measurement data quality control system based on a mobile platform, characterized in that: It is used to store computer-readable instructions, and when the computer-readable instructions are read, it can execute a mobile platform-based laser radar real-time wind measurement data quality control method as described in any one of claims 1-9.

Images