An airborne abnormal data determination method, device and electronic equipment

Abstract

The application provides an on-board abnormal data determination method and device and electronic equipment. The method comprises the following steps: selecting target on-board data corresponding to the number of independent variables of a data table; determining an on-board data average curvature sequence and an on-board data Gaussian curvature sequence corresponding to the target on-board data; determining an average curvature non-abnormal value interval and a Gaussian curvature non-abnormal value interval based on a relative deviation parameter, average curvature quartiles and average curvature quartile distance, Gaussian curvature quartiles and Gaussian curvature quartile distance; determining an average sequence abnormal data and a Gaussian sequence abnormal data based on an absolute deviation parameter, the average curvature non-abnormal value interval and the Gaussian curvature non-abnormal value interval; and combining the average sequence abnormal data and the Gaussian sequence abnormal data to obtain abnormal data of the target on-board data. The on-board abnormal data determination method, device and electronic equipment solve the problems of long time consumption, low reliability and low efficiency in checking on-board abnormal data by manual means.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a method and device for determining abnormal airborne data and an electronic device. BACKGROUND

[0002] The C919 project is a key civil aircraft project, and is a large jet civil aircraft independently developed according to international civil aviation regulations and with independent intellectual property rights. The C919 project has completed project demonstration, feasibility demonstration and pre-development stage work, and has entered the engineering development stage. The flight tube, performance and power of the aircraft are the key research fields for obtaining the TC (type certificate) of the C919 project. The flight management system (FMS) provides flight manual query and performance calculation support for the related crew of the three professions. In the FMS system, the airborne performance database (PDB) is used to realize data analysis and fault interpretation of the system, and the main functions include aircraft performance calculation, trajectory prediction, horizontal and vertical guidance. In order to make the airborne performance database data meet the data format and quality requirements specified by the flight management system (FMS) supplier AVIAGE / GE, the data quality needs to be judged. At present, the existing method for determining abnormal airborne data is to use the data validity function in the EXCEL or WPS table, and to judge the advantages and disadvantages of the data quality by manual observation.

[0003] However, when checking the abnormal airborne data in the above-mentioned manual manner, there are problems of complicated checking steps, long time consumption, low reliability and low efficiency. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a method and device for determining abnormal airborne data and an electronic device, so as to solve the problems of complicated checking steps, long time consumption, low reliability and low efficiency when checking the abnormal airborne data in a manual manner.

[0005] In a first aspect, an embodiment of the present application provides a method for determining abnormal airborne data, comprising:

[0006] selecting target airborne data corresponding to the number of independent variables of a target airborne data table;

[0007] determining an airborne data average curvature sequence and an airborne data Gaussian curvature sequence corresponding to the target airborne data;

[0008] determining an average curvature non-abnormal value interval and a Gaussian curvature non-abnormal value interval based on the selected relative deviation parameter, the average curvature quartile and the average curvature quartile distance corresponding to the airborne data average curvature sequence, and the Gaussian curvature quartile and the Gaussian curvature quartile distance corresponding to the airborne data Gaussian curvature sequence;

[0009] Determine the average sequence abnormal data corresponding to the average curvature sequence of the airborne data and the Gaussian sequence abnormal data corresponding to the Gaussian curvature sequence of the airborne data based on the selected absolute deviation parameter, the average curvature non-outlier interval and the Gaussian curvature non-outlier interval.

[0010] Combine the average sequence abnormal data and the Gaussian sequence abnormal data together as the abnormal data of the target airborne data.

[0011] Optionally, the determining of the average curvature sequence of the airborne data and the Gaussian curvature sequence of the airborne data corresponding to the target airborne data comprises: determining the average curvature and the Gaussian curvature of each data point in a two-dimensional surface in a three-dimensional space corresponding to the target airborne data; constructing an average curvature matrix of the airborne data from the average curvatures of all data points; constructing a Gaussian curvature matrix of the airborne data from the Gaussian curvatures of all data points; and performing dimension reduction processing on the average curvature matrix of the airborne data and the Gaussian curvature matrix of the airborne data to obtain the average curvature sequence of the airborne data and the Gaussian curvature sequence of the airborne data respectively.

[0012] Optionally, the determining of the average curvature non-outlier interval and the Gaussian curvature non-outlier interval based on the selected relative deviation parameter, the average curvature quartile and the average curvature quartile distance corresponding to the average curvature sequence of the airborne data, and the Gaussian curvature quartile and the Gaussian curvature quartile distance corresponding to the Gaussian curvature sequence of the airborne data comprises: determining the product of the relative deviation parameter and the average curvature quartile distance as the average curvature deviation, and the product of the relative deviation parameter and the Gaussian curvature quartile distance as the Gaussian curvature deviation; determining the difference between the first quartile of the average curvature and the average curvature deviation as the minimum average curvature non-outlier value, and the sum of the third quartile of the average curvature and the average curvature deviation as the maximum average curvature non-outlier value; determining the difference between the first quartile of the Gaussian curvature and the Gaussian curvature deviation as the minimum Gaussian curvature non-outlier value, and the sum of the third quartile of the Gaussian curvature and the Gaussian curvature deviation as the maximum Gaussian curvature non-outlier value; and constructing the average curvature non-outlier interval from the minimum average curvature non-outlier value and the maximum average curvature non-outlier value, and constructing the Gaussian curvature non-outlier interval from the minimum Gaussian curvature non-outlier value and the maximum Gaussian curvature non-outlier value.

[0013] Optionally, the determining of the average sequence abnormal data corresponding to the average curvature sequence of the airborne data and the Gaussian sequence abnormal data corresponding to the Gaussian curvature sequence of the airborne data based on the selected absolute deviation parameter, the average curvature non-outlier interval and the Gaussian curvature non-outlier interval comprises: regarding the target airborne data whose average curvature is outside the average curvature non-outlier interval and whose absolute value is greater than the absolute deviation parameter as the average sequence abnormal data; and regarding the target airborne data whose Gaussian curvature is outside the Gaussian curvature non-outlier interval and whose absolute value is greater than the square of the absolute deviation parameter as the Gaussian sequence abnormal data.

[0014] Optionally, after combining the average series abnormal data and the Gaussian series abnormal data together as the abnormal data of the target airborne data, the method further comprises: taking the target airborne data with the average curvature in the average curvature non-abnormal value interval or with the absolute value less than or equal to the absolute deviation parameter as average series non-abnormal data; taking the target airborne data with the Gaussian curvature in the Gaussian curvature non-abnormal value interval or with the absolute value less than or equal to the square of the absolute deviation parameter as Gaussian series non-abnormal data; and taking the intersection data of the average curvature non-abnormal data and the Gaussian curvature non-abnormal data as non-abnormal data.

[0015] Optionally, before selecting the target airborne data corresponding to the number of independent variables of the target airborne data table, the method further comprises: obtaining a to-be-judged airborne data table from a to-be-judged airborne data file; determining whether the to-be-judged airborne data table meets the independent variable number requirement based on the number of independent variables of the to-be-judged airborne data table; and if the to-be-judged airborne data table meets the independent variable number requirement, taking the to-be-judged airborne data table as the target airborne data table.

[0016] Optionally, the determination of whether the to-be-judged airborne data table meets the independent variable number requirement comprises: if the number of independent variables of the to-be-judged airborne data table is 1, determining that the to-be-judged airborne data table does not meet the independent variable number requirement; and if the number of independent variables of the to-be-judged airborne data table is 2 or 3, determining that the to-be-judged airborne data table meets the independent variable number requirement.

[0017] Optionally, the selection of the target airborne data corresponding to the number of independent variables of the target airborne data table comprises: if the number of independent variables of the target airborne data table is 2, selecting all values of the first independent variable, all values of the second independent variable, and all values of the dependent variable in the target airborne data table as the target data; if the number of independent variables of the target airborne data table is 3, selecting one of the three independent variables as a third independent variable, selecting one value from multiple values of the third independent variable as a target variable value, and taking the other two independent variables as the first independent variable and the second independent variable respectively; and selecting all values of the first independent variable, all values of the second independent variable, and all values of the dependent variable corresponding to the target variable value in the target airborne data table as the target data.

[0018] In a second aspect, an embodiment of the present application further provides a device for determining airborne abnormal data, and the device comprises:

[0019] a data obtaining module, configured to select target airborne data corresponding to the number of independent variables of a target airborne data table;

[0020] a series determining module, configured to determine an airborne data average curvature series and an airborne data Gaussian curvature series corresponding to the target airborne data;

[0021] The abnormal interval calculation module is configured to determine an average curvature non-outlier interval and a Gaussian curvature non-outlier interval based on the selected relative deviation parameter, the average curvature quartile and the average curvature quartile distance corresponding to the average curvature sequence of the airborne data, and the Gaussian curvature quartile and the Gaussian curvature quartile distance corresponding to the Gaussian curvature sequence of the airborne data, respectively.

[0022] The first abnormality determination module is configured to determine average sequence abnormal data corresponding to the average curvature sequence of the airborne data and Gaussian sequence abnormal data corresponding to the Gaussian curvature sequence of the airborne data based on the selected absolute deviation parameter, the average curvature non-outlier interval and the Gaussian curvature non-outlier interval, respectively.

[0023] The second abnormality determination module is configured to combine the average sequence abnormal data and the Gaussian sequence abnormal data together as abnormal data of the target airborne data.

[0024] In a third aspect, an electronic device is provided, which includes a processor, a memory and a bus, the memory stores machine readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, and the machine readable instructions are executed by the processor to perform the steps of the airborne abnormal data determination method as described above.

[0025] The embodiments of the present application have the following beneficial effects:

[0026] The airborne abnormal data determination method, device and electronic device provided by the embodiments of the present application can automatically select corresponding target airborne data according to the number of independent variables of the target airborne data table, and obtain the average curvature sequence and the Gaussian curvature sequence of the target airborne data by calculation. The abnormal data in the two sequences is determined by using the quartile, the quartile distance and the selected relative deviation parameter and absolute deviation parameter, so as to obtain the abnormal data of the target airborne data. The embodiments of the present application can automatically screen out abnormal data from different dimensional curvature sequences without manual inspection to determine the outliers in the airborne data. Compared with the prior art, the embodiments of the present application solve the problems of tedious inspection steps, long time consumption and low efficiency when manually checking the airborne abnormal data, and improve the reliability when checking a large amount of data.

[0027] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the following preferred embodiments are described in detail below, and the accompanying drawings are described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as a limitation to the scope. For those skilled in the art, other related drawings can also be obtained without creative labor.

[0029] Figure 1 A flow chart of the airborne abnormal data determination method provided by the embodiments of the present application is shown;

[0030] Figure 2 A structural schematic diagram of the airborne abnormal data determination device provided by the embodiments of the present application is shown;

[0031] Figure 3 A structural schematic diagram of the electronic device provided by the embodiments of the present application is shown. DETAILED DESCRIPTION

[0032] In order to make the purposes, technical solutions and advantages of the embodiments of the present application more clear, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, every other embodiment obtained by those skilled in the art without creative labor belongs to the scope of the present application.

[0033] It is worth noting that before the present application is proposed, the C919 project is a key civil aircraft project, which is a large jet civil aircraft with independent intellectual property rights developed in accordance with international civil aviation regulations. At present, the project has completed project demonstration, feasibility demonstration and pre-development stage work, and has entered the engineering development stage. The flight tube, performance and power of the aircraft are the key research fields for the TC (type certificate) of the C919 project. The flight management system (FMS) provides flight manual query and performance calculation support for the related crew of the three professionals. In the FMS system, the onboard performance database (PDB) is used to realize data analysis and fault interpretation of the system, and the main functions include aircraft performance calculation, trajectory prediction, horizontal and vertical guidance. In order to make the onboard performance database data meet the data format and quality requirements specified by the flight management system (FMS) supplier AVIAGE / GE, the data quality needs to be judged. At present, the existing onboard abnormal data determination scheme is to use the data validity function in the EXCEL or WPS table, and to judge the advantages and disadvantages of the data quality by manual observation. However, when checking the onboard abnormal data by the above-mentioned manual method, there are problems of tedious checking steps, long time consumption and low efficiency, and insufficient reliability.

[0034] Based on this, the embodiment of the present application provides an onboard abnormal data determination method to shorten the time consumed when checking the onboard abnormal data, and improve the checking efficiency and checking reliability.

[0035] Firstly, the technical terms involved in the present application are explained.

[0036] Normal curvature: the normal line passing through a point on a curved surface can make countless normal sections about the curved surface. The curvature of the intersection line of each normal section at the point is a normal curvature. There are countless normal curvatures at the point.

[0037] Principal curvature: among the normal curvatures, there is a maximum value (denoted as k1), and it can be proved that the curvature perpendicular to the maximum curvature surface is a minimum value (denoted as k2). The two curvatures are called principal curvatures, and the corresponding directions are called principal directions.

[0038] Gaussian curvature: the product of the two principal curvatures K≡k1k2, which reflects the intrinsic bending degree of a point.

[0039] Mean curvature: the arithmetic mean of the two principal curvatures H≡(k1+k2) / 2, which reflects the extrinsic bending degree of a point.

[0040] Please refer to Figure 1 , Figure 1A flowchart of an airborne abnormal data determination method provided by an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the airborne abnormal data determination method provided by the embodiment of the present application comprises the following steps. Figure 1

[0041] In step S101, target airborne data corresponding to the number of independent variables of a target airborne data table is selected.

[0042] In this step, the target airborne data table can refer to a data table in a to-be-judged airborne data file, and the target airborne data table is used to store airborne data.

[0043] The airborne data table comprises a data table name, a data version number, an update date of the data table, a data table description, a number of independent variables, data accuracy, independent variable identification, dependent variable identification, a value of the independent variable, and a value of the dependent variable. The independent variable identification can refer to an independent variable name, and the dependent variable identification can refer to a dependent variable name.

[0044] The number of independent variables can refer to the number of independent variables in the airborne data table. The airborne data table comprises at least one independent variable and one dependent variable.

[0045] For example, the independent variable can be flight speed, and the dependent variable can be flight drag coefficient.

[0046] The target airborne data can refer to flight-related data in the target airborne data table, and the target airborne data is used to generate a curvature image.

[0047] In the embodiment of the present application, the number of independent variables and the number of dependent variables in the target airborne data table are stored in the data table. The present application only performs abnormal data screening on data tables with two or three independent variables, and different numbers of independent variables correspond to different target airborne data selection methods. Therefore, the target airborne data selected for airborne data tables with different numbers of independent variables are also different. The selected target airborne data comprises independent variable data corresponding to at least two independent variables and dependent variable data corresponding to one dependent variable. A corresponding two-dimensional surface can be generated according to the independent variable data and the dependent variable data. Then, the average curvature and the Gaussian curvature of each data point in the two-dimensional surface are used to determine an airborne data average curvature sequence and an airborne data Gaussian curvature sequence.

[0048] In an optional embodiment, before the target airborne data corresponding to the number of independent variables of the target airborne data table is selected, the method further comprises the following steps: acquiring a to-be-judged airborne data table from a to-be-judged airborne data file; determining whether the to-be-judged airborne data table meets the independent variable number requirement based on the number of independent variables of the to-be-judged airborne data table; and if the to-be-judged airborne data table meets the independent variable number requirement, taking the to-be-judged airborne data table as the target airborne data table.

[0049] ​Here, the to-be-judged airborne data file can refer to a file including at least one to-be-judged airborne data table, and the to-be-judged airborne data file is used to store a plurality of to-be-judged airborne data tables.

[0050] For example, the to-be-judged airborne data file can be a CSV file, which is provided by a demander who needs to screen abnormal data.

[0051] When a plurality of to-be-judged airborne data tables are stored in one to-be-judged airborne data file, the to-be-judged airborne data tables are stacked in the file in a fixed format.

[0052] Specifically, the to-be-judged airborne data file is read into the server memory through the client GUI file selector. When the to-be-judged airborne data file includes a plurality of to-be-judged airborne data tables, different to-be-judged airborne data tables are distinguished according to table identifiers, and the to-be-judged airborne data tables are stored in the same format.

[0053] For example, the data table information of a single to-be-judged airborne data table is stored in the following order: data table name, data version number, data table update date, data table description, number of independent variables, data accuracy, independent variable identifier, dependent variable identifier, value of independent variable, and value of dependent variable.

[0054] When there are a plurality of independent variables, the independent variable identifiers are stored in sequence, one per row, and the values of the independent variables are stored in sequence, one per row.

[0055] When the to-be-judged airborne data file includes a plurality of to-be-judged airborne data tables, the number of independent variables of each to-be-judged airborne data table is extracted in sequence according to the storage order of the data tables, and it is determined whether the number of independent variables of the to-be-judged airborne data table is greater than or equal to a preset value. If the number of independent variables is greater than or equal to the preset value, the to-be-judged airborne data table is taken as a target airborne data table.

[0056] In an optional embodiment, determining whether the to-be-judged airborne data table meets the independent variable quantity requirement includes: if the number of independent variables of the to-be-judged airborne data table is 1, determining that the to-be-judged airborne data table does not meet the independent variable quantity requirement; and if the number of independent variables of the to-be-judged airborne data table is 2 or 3, determining that the to-be-judged airborne data table meets the independent variable quantity requirement.

[0057] Specifically, the server pre-processes each to-be-discriminated airborne data table through a preprocessing engine, i.e., performs screening and filtering processing. First, the screening processing is performed, which refers to screening processing of the data table. Only the data table meeting the condition is retained.

[0058] If the number of independent variables of the to-be-discriminated airborne data table is 1 and the number of dependent variables is 1, it indicates that the data image corresponding to the to-be-discriminated airborne data table is a line within the definition range of the independent variable. Since the present application is for discriminating abnormal data of a curved surface image, if the number of independent variables is 1, the to-be-discriminated airborne data table is filtered out.

[0059] If the number of independent variables of the to-be-discriminated airborne data table is 2 and the number of dependent variables is 1, it indicates that the data image corresponding to the to-be-discriminated airborne data table is a curved surface image of a three-dimensional space composed of various data points. Therefore, if the number of independent variables is 2, the to-be-discriminated airborne data table is retained and taken as a target airborne data table.

[0060] If the number of independent variables of the to-be-discriminated airborne data table is 3 and the number of dependent variables is 1, it indicates that the data image corresponding to the to-be-discriminated airborne data table is a four-dimensional space image composed of various data points. Since the prior art is difficult to directly display the thinking space image, the data points in the to-be-discriminated airborne data table can be regarded as a plurality of equal value surfaces of one of the independent variables, i.e., a plurality of curved surfaces of a three-dimensional space, which also meets the judgment requirement. Therefore, the to-be-discriminated airborne data table is retained and taken as a target airborne data table.

[0061] In an optional embodiment, the target airborne data corresponding to the number of independent variables of the target airborne data table is selected, including: if the number of independent variables of the target airborne data table is 2, selecting all values of the first independent variable, all values of the second independent variable, and all values of the dependent variable in the target airborne data table as target data; if the number of independent variables of the target airborne data table is 3, selecting one of the three independent variables as a third independent variable, selecting one value from the plurality of values of the third independent variable as a target variable value, and taking the other two independent variables as a first independent variable and a second independent variable respectively; and selecting all values of the first independent variable, all values of the second independent variable, and all values of the dependent variable corresponding to the target variable value in the target airborne data table as target data.

[0062] Specifically, after a to-be-discriminated airborne data table is determined as a target airborne data table, the table information of the target airborne data table is extracted. If the number of independent variables of the target airborne data table is 2, the data table name and the number of independent variables are extracted. If the number of independent variables of the target airborne data table is 3, the data table name, the number of independent variables, and all values of the third independent variable are extracted.

[0063] The extracted table information is sent to the client for display, and according to the table information of the plurality of target airborne data tables displayed by the client, the user can select which target airborne data table or which target airborne data tables to perform abnormal data judgment on. If the target airborne data table is of the self-variable quantity of 2, only the data table needs to be selected. If the target airborne data table is of the self-variable quantity of 3, not only the data table needs to be selected, but also a specific value of the third self-variable needs to be selected.

[0064] The server determines the selected target data according to the selection of the user. If the target airborne data table is of the self-variable quantity of 2, after the user selects the data table, the server takes all values of the first self-variable, all values of the second self-variable, and all values of the dependent variable in the data table as the target data, and constructs a three-dimensional image corresponding to the target data.

[0065] If the target airborne data table is of the self-variable quantity of 3, after the user selects the data table and the target variable value, the server takes all values of the first self-variable, all values of the second self-variable, and all values of the dependent variable corresponding to the target variable value in the data table as the target data, and constructs a three-dimensional image corresponding to the target data.

[0066] It should be noted that if the target airborne data table is of the self-variable quantity of 2, any self-variable can be used as the first self-variable or the second self-variable. If the target airborne data table is of the self-variable quantity of 3, a self-variable needs to be selected from the three self-variables as the third self-variable, and any one of the remaining two self-variables can be used as the first self-variable or the second self-variable.

[0067] In step S102, the airborne data average curvature sequence and the airborne data Gaussian curvature sequence corresponding to the target airborne data are determined.

[0068] In this step, the airborne data average curvature sequence can refer to a sequence corresponding to the average curvature of all data points in the two-dimensional surface corresponding to the target airborne data.

[0069] The airborne data Gaussian curvature sequence can refer to a sequence corresponding to the Gaussian curvature of all data points in the two-dimensional surface corresponding to the target airborne data.

[0070] In the embodiments of the present application, for the two-dimensional surface corresponding to the target airborne data, the average curvature and the Gaussian curvature corresponding to each data point on the surface are determined respectively, and based on the average curvature and the Gaussian curvature corresponding to all data points, the airborne data average curvature sequence and the airborne data Gaussian curvature sequence are determined.

[0071] In an optional embodiment, the determining of the average curvature series and the Gaussian curvature series corresponding to the target onboard data comprises: determining the average curvature and the Gaussian curvature of each data point in a two-dimensional surface in a three-dimensional space corresponding to the target onboard data; constructing an average curvature matrix of the onboard data by the average curvatures corresponding to all data points; constructing a Gaussian curvature matrix of the onboard data by the Gaussian curvatures corresponding to all data points; and performing dimension reduction on the average curvature matrix and the Gaussian curvature matrix to obtain the average curvature series and the Gaussian curvature series of the onboard data respectively.

[0072] Specifically, the target onboard data comprises data of two independent variables and data of one dependent variable, and a two-dimensional surface can be constructed by the data. The average curvature and the Gaussian curvature of each data point can be calculated by a curvature calculation engine on the server, and the average curvatures and the Gaussian curvatures corresponding to all data points form the average curvature matrix and the Gaussian curvature matrix of the onboard data respectively. The average curvature matrix and the Gaussian curvature matrix of the onboard data are both two-dimensional matrices.

[0073] Then, the average curvature matrix and the Gaussian curvature matrix of the onboard data are subjected to dimension reduction to obtain two one-dimensional series, i.e., the average curvature series and the Gaussian curvature series of the onboard data.

[0074] In step S103, the average curvature non-outlier interval and the Gaussian curvature non-outlier interval are determined respectively based on the selected relative deviation parameter, the average curvature quartile and the average curvature quartile distance corresponding to the average curvature series of the onboard data, and the Gaussian curvature quartile and the Gaussian curvature quartile distance corresponding to the Gaussian curvature series of the onboard data.

[0075] In this step, the relative deviation parameter can refer to a deviation degree coefficient relative to the first and last quartiles, and the relative deviation parameter is used to adjust the recognition range of the abnormal data.

[0076] The average curvature quartile can refer to three quartiles in the average curvature series of the onboard data.

[0077] The average curvature quartile comprises an average curvature first quartile, an average curvature second quartile and an average curvature third quartile.

[0078] The average curvature first quartile is the data at the 25% position in the sorted average curvature series, the average curvature second quartile is the data at the 50% position in the sorted average curvature series, and the average curvature third quartile is the data at the 75% position in the sorted average curvature series.

[0079] The average curvature quartile range can refer to a difference between the average curvature third quartile and the average curvature first quartile.

[0080] The Gaussian curvature quartiles can refer to three quartiles in a Gaussian curvature series of the onboard data.

[0081] The Gaussian curvature quartiles include a Gaussian curvature first quartile, a Gaussian curvature second quartile, and a Gaussian curvature third quartile.

[0082] The Gaussian curvature first quartile is a data at a 25% position in the sorted Gaussian curvature series, the Gaussian curvature second quartile is a data at a 50% position in the sorted Gaussian curvature series, and the Gaussian curvature third quartile is a data at a 75% position in the sorted Gaussian curvature series.

[0083] The Gaussian curvature quartile range can refer to a difference between the Gaussian curvature third quartile and the Gaussian curvature first quartile.

[0084] The average curvature non-outlier interval can refer to an interval range of non-outliers in an average curvature series of the onboard data, and the average curvature non-outlier interval is used to determine outliers in the average curvature series.

[0085] The Gaussian curvature non-outlier interval can refer to an interval range of non-outliers in a Gaussian curvature series of the onboard data, and the Gaussian curvature non-outlier interval is used to determine outliers in the Gaussian curvature series.

[0086] In the embodiments of the present application, after a user selects a target onboard data table on a client, the user can input a corresponding relative deviation parameter for the target onboard data table, and the relative deviation parameter is used to distinguish abnormal data and normal data. In an ideal state, the bending degree of a two-dimensional curved surface at each data point should be as small as possible, that is, the degree of deviation from a plane should be as small as possible. If the curvature of a data point is within a range corresponding to the relative deviation parameter, the onboard data corresponding to the data point is normal data. If the curvature of a data point is outside the range corresponding to the relative deviation parameter, the onboard data corresponding to the data point is abnormal data. Here, the curvature of the data point can be calculated to quantitatively measure the bending degree of the data surface at the data point. In addition, the greater the value of the set relative deviation parameter, the more tolerant the selection of abnormal data is, and the less abnormal data there is. Conversely, the smaller the value of the set relative deviation parameter, the more strict the selection of abnormal data is, and the more abnormal data there is.

[0087] In an optional embodiment, based on the selected relative deviation parameter, the average curvature quartiles and the average curvature quartile distance corresponding to the average curvature sequence of the on-board data, and the Gaussian curvature quartiles and the Gaussian curvature quartile distance corresponding to the Gaussian curvature sequence of the on-board data, the average curvature non-outlier interval and the Gaussian curvature non-outlier interval are determined respectively, including: determining the product of the relative deviation parameter and the average curvature quartile distance as the average curvature deviation, and determining the product of the relative deviation parameter and the Gaussian curvature quartile distance as the Gaussian curvature deviation; determining the difference between the first quartile of the average curvature and the average curvature deviation as the minimum average curvature non-outlier value, and determining the sum of the third quartile of the average curvature and the average curvature deviation as the maximum average curvature non-outlier value; determining the difference between the first quartile of the Gaussian curvature and the Gaussian curvature deviation as the minimum Gaussian curvature non-outlier value, and determining the sum of the third quartile of the Gaussian curvature and the Gaussian curvature deviation as the maximum Gaussian curvature non-outlier value; the average curvature non-outlier interval is composed of the minimum average curvature non-outlier value and the maximum average curvature non-outlier value, and the Gaussian curvature non-outlier interval is composed of the minimum Gaussian curvature non-outlier value and the maximum Gaussian curvature non-outlier value.

[0088] Specifically, the average curvature sequence of the on-board data and the Gaussian curvature sequence of the on-board data are sorted respectively, the data in the sequence is arranged in ascending order, and the quartiles corresponding to the two arranged sequences are determined respectively.

[0089] The average curvature quartiles include the first quartile of the average curvature, the second quartile of the average curvature, and the third quartile of the average curvature, which are denoted as Q 1_ , Q 1_ , and Q 1_ .

[0090] The Gaussian curvature quartiles include the first quartile of the Gaussian curvature, the second quartile of the Gaussian curvature, and the third quartile of the Gaussian curvature, which are denoted as Q 2_ , Q 2_ , and Q 2_3 .

[0091] The average curvature quartile distance is denoted as ΔQ1, and the Gaussian curvature quartile distance is denoted as ΔQ2, so ΔQ1=Q 1_ -Q 1_ , and ΔQ2=Q 2_ -Q 2_ .

[0092] The average curvature deviation = relative deviation parameter × ΔQ1, and the Gaussian curvature deviation = relative deviation parameter × ΔQ2.

[0093] The minimum average curvature non-outlier value is denoted as Q 1_ , and the maximum average curvature non-outlier value is denoted as Q 1_x , so Q 1_= Q 1_ - average curvature deviation, Q 1_x = Q 1_ + average curvature deviation, Q 2_min = Q 2_ - Gaussian curvature deviation, Q 2_x = Q 2_ + Gaussian curvature deviation.

[0094] Thus, the average curvature non-outlier interval and the Gaussian curvature non-outlier interval can be determined, the average curvature non-outlier interval being [Q 1_ , Q 1_x ], and the Gaussian curvature non-outlier interval being [Q 2_ , Q 2_x ].

[0095] In step S104, based on the selected absolute deviation parameter, the average curvature non-outlier interval, and the Gaussian curvature non-outlier interval, average sequence abnormal data corresponding to the average curvature sequence of the onboard data and Gaussian sequence abnormal data corresponding to the Gaussian curvature sequence of the onboard data are determined respectively.

[0096] In this step, the absolute deviation parameter can refer to the absolute deviation degree relative to the curvature surface, and the absolute deviation parameter is used to determine the identification range of the abnormal data together with the non-outlier interval.

[0097] In the embodiments of the present application, the curvature calculation engine of the server can not only calculate the average curvature and the Gaussian curvature of each data point, but also determine the average curvature non-outlier interval and the Gaussian curvature non-outlier interval according to the calculated average curvature and Gaussian curvature, and determine the average sequence abnormal data and the Gaussian sequence abnormal data by using the average curvature non-outlier interval and the Gaussian curvature non-outlier interval.

[0098] In an optional embodiment, based on the selected absolute deviation parameter, the average curvature non-outlier interval, and the Gaussian curvature non-outlier interval, the average sequence abnormal data corresponding to the average curvature sequence of the onboard data and the Gaussian sequence abnormal data corresponding to the Gaussian curvature sequence of the onboard data are determined respectively, including: regarding the target onboard data whose average curvature is outside the average curvature non-outlier interval and whose absolute value is greater than the absolute deviation parameter as the average sequence abnormal data; and regarding the target onboard data whose Gaussian curvature is outside the Gaussian curvature non-outlier interval and whose absolute value is greater than the square of the absolute deviation parameter as the Gaussian sequence abnormal data.

[0099] Specifically, if the average curvature of the data point on the two-dimensional surface corresponding to the target airborne data is outside the average curvature non-outlier interval, it indicates that the data point does not meet the normal value range requirement corresponding to the relative deviation parameter, if the absolute value of the average curvature of a data point is greater than the absolute deviation parameter, it indicates that the data point does not meet the normal value range requirement corresponding to the absolute deviation parameter, when the data point neither meets the normal value range requirement corresponding to the relative deviation parameter nor meets the normal value range requirement corresponding to the absolute deviation parameter, the target airborne data corresponding to the data point is determined as average series abnormal data.

[0100] If the Gaussian curvature of the data point on the two-dimensional surface corresponding to the target airborne data is outside the Gaussian curvature non-outlier interval, it indicates that the data point does not meet the normal value range requirement corresponding to the relative deviation parameter, if the absolute value of the Gaussian curvature of a data point is greater than the square of the absolute deviation parameter, it indicates that the data point does not meet the normal value range requirement corresponding to the absolute deviation parameter, when the data point neither meets the normal value range requirement corresponding to the relative deviation parameter nor meets the normal value range requirement corresponding to the absolute deviation parameter, the target airborne data corresponding to the data point is determined as Gaussian series abnormal data.

[0101] In step S105, the average series abnormal data and the Gaussian series abnormal data are combined together as the abnormal data of the target airborne data.

[0102] In this step, the abnormal data of the target airborne data is obtained by combining the average series abnormal data and the Gaussian series abnormal data, and a 3D curvature image that can interact with a mouse pointer is presented by a 3D visualization engine using a matplotlib algorithm library based on Python. The 3D curvature image can realize functions such as rotation, enlargement, reduction, saving, and real-time display of coordinates and images. In addition, multiple interactive windows can be opened at the same time, each interactive window corresponds to a 3D curvature image of a target airborne data table, and the abnormal data in multiple target airborne data can be identified at the same time.

[0103] In addition, in order to avoid repeated adjustment of the relative deviation parameter and the absolute deviation parameter, the adjusted deviation parameters can be applied to multiple target airborne data tables in batches, and the deviation parameters corresponding to each target airborne data table are displayed in the deviation parameter list of the client. The deviation parameter list corresponding to all target airborne data tables can also be saved in the server in the form of a CSV file. The CSV file can be directly found in the server, and the deviation parameters in the CSV file can be directly modified to be reimported into the client to update the display of the deviation parameter list.

[0104] Secondly, according to the identification result of the abnormal data, the abnormal data can be marked in the target airborne data table, for example, the abnormal data is displayed in a special color, so as to better distinguish the abnormal data.

[0105] In an optional embodiment, after the average sequence abnormal data and the Gaussian sequence abnormal data are combined together as the abnormal data of the target airborne data, the method further comprises: taking the target airborne data with the average curvature in the average curvature non-abnormal value interval or with the absolute value less than or equal to the absolute deviation parameter as the average sequence non-abnormal data; taking the target airborne data with the Gaussian curvature in the Gaussian curvature non-abnormal value interval or with the absolute value less than or equal to the square of the absolute deviation parameter as the Gaussian sequence non-abnormal data; and taking the intersection data of the average curvature non-abnormal data and the Gaussian curvature non-abnormal data as the non-abnormal data.

[0106] Here, the non-abnormal data refers to the data within the range of the relative deviation parameter and the absolute deviation parameter.

[0107] Specifically, when determining the non-abnormal data, first, the average sequence non-abnormal data and the Gaussian sequence non-abnormal data are determined according to the average curvature non-abnormal value interval, the Gaussian curvature non-abnormal value interval and the absolute deviation parameter. For any one of the average sequence non-abnormal data or the Gaussian sequence non-abnormal data, if the data belongs to both the average sequence non-abnormal data and the Gaussian sequence non-abnormal data, the data is determined as the non-abnormal data.

[0108] Compared with the prior art, the present application can automatically select the corresponding target airborne data according to the number of independent variables of the target airborne data table, and obtain the average curvature sequence and the Gaussian curvature sequence corresponding to the target airborne data by calculation. The abnormal data in the two sequences is determined by using the quartiles, the interquartile range, and the selected relative deviation parameter and absolute deviation parameter, so as to obtain the abnormal data of the target airborne data. The present application can automatically screen the abnormal data from the curvature sequences of different dimensions, without the need for manual inspection to determine the abnormal values in the airborne data, thereby solving the problems of tedious inspection steps, long time consumption and low efficiency when manually checking the airborne abnormal data, and improving the reliability when checking a large amount of data.

[0109] Based on the same inventive concept, the present application also provides an airborne abnormal data determination device corresponding to the airborne abnormal data determination method. Since the principle of solving problems in the device of the present application is similar to the above-mentioned airborne abnormal data determination method of the present application, the implementation of the device can be referred to the implementation of the method, and the repeated parts will not be described here.

[0110] Please refer to Figure 2 ,Figure 2 A structure diagram of an airborne abnormal data determination device provided by an embodiment of the present application is shown in FIG. 1. Figure 2 The airborne abnormal data determination device 200 includes:

[0111] A data acquisition module 201 is configured to select target airborne data corresponding to a number of independent variables of a target airborne data table.

[0112] A sequence determination module 202 is configured to determine an airborne data average curvature sequence and an airborne data Gaussian curvature sequence corresponding to the target airborne data.

[0113] An abnormal interval calculation module 203 is configured to determine an average curvature non-outlier interval and a Gaussian curvature non-outlier interval based on a selected relative deviation parameter, average curvature quartiles and average curvature quartile distance corresponding to the airborne data average curvature sequence, and Gaussian curvature quartiles and Gaussian curvature quartile distance corresponding to the airborne data Gaussian curvature sequence.

[0114] A first abnormality determination module 204 is configured to determine average sequence abnormal data corresponding to the airborne data average curvature sequence and Gaussian sequence abnormal data corresponding to the airborne data Gaussian curvature sequence based on a selected absolute deviation parameter, the average curvature non-outlier interval and the Gaussian curvature non-outlier interval.

[0115] A second abnormality determination module 205 is configured to combine the average sequence abnormal data and the Gaussian sequence abnormal data together as abnormal data of the target airborne data.

[0116] Referring to Figure 3 , Figure 3 A structure diagram of an electronic device provided by an embodiment of the present application is shown in FIG. 2. Figure 3 The electronic device 300 includes a processor 310, a memory 320 and a bus 330.

[0117] The memory 320 stores machine readable instructions executable by the processor 310, when the electronic device 300 is running, the processor 310 and the memory 320 communicate through the bus 330, and the machine readable instructions executed by the processor 310 can perform the steps of the airborne abnormal data determination method in the method embodiment shown in Figure 1 The specific implementation can refer to the method embodiment, and will not be described here.

[0118] An embodiment of the present application further provides a computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program is run by a processor to perform the airborne abnormal data determination method in the method embodiment shown in Figure 1The steps of the airborne abnormal data determination method in the method embodiment are specifically implemented as described above, and details are not described herein again.

[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the system, device and unit described above can refer to the corresponding processes in the foregoing method embodiments, and details are not described herein again.

[0120] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. The described device embodiments are only schematic, for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0121] The units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, that is, can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments.

[0122] In addition, each functional unit in the various embodiments of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.

[0123] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solutions of the present application or the part of the present application that essentially contributes to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, and various storage medium that can store program codes.

[0124] Finally, it should be noted that the above-described embodiments are merely specific embodiments of the present application, which are used to illustrate the technical solutions of the present application, but not to limit the same. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that any skilled person in the art can still modify or easily think of changes to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some of the technical features, within the technical scope disclosed by the present application. The modifications, changes or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for determining airborne anomaly data, characterized in that, include: Select target airborne data that corresponds to the number of independent variables in the target airborne data table; Determine the airborne data mean curvature sequence and the airborne data Gaussian curvature sequence corresponding to the target airborne data; Based on the selected relative deviation parameters, the mean curvature quartiles and the interquartile range of the mean curvature sequence of the airborne data, and the Gaussian curvature quartiles and the interquartile range of the Gaussian curvature sequence of the airborne data, the non-outlier intervals of the mean curvature and the Gaussian curvature are determined respectively. Based on the selected absolute deviation parameter, the non-outlier range of average curvature, and the non-outlier range of Gaussian curvature, the abnormal data of the average series corresponding to the average curvature series of the airborne data and the abnormal data of the Gaussian series corresponding to the Gaussian curvature series of the airborne data are determined respectively. The abnormal data of the mean sequence and the abnormal data of the Gaussian sequence are combined together as the abnormal data of the target airborne data. The process of determining the non-outlier intervals of the average curvature and the interquartile range of the Gaussian curvature based on the selected relative deviation parameters, the average curvature quartiles and the interquartile range of the Gaussian curvature corresponding to the average curvature sequence of the airborne data, and the Gaussian curvature quartiles and the Gaussian curvature range corresponding to the Gaussian curvature sequence of the airborne data, respectively, includes: The product of the relative deviation parameter and the interquartile range of the average curvature is determined as the average curvature deviation, and the product of the relative deviation parameter and the interquartile range of the Gaussian curvature is determined as the Gaussian curvature deviation. The difference between the first quartile of the mean curvature and the deviation of the mean curvature is determined as the minimum non-outlier value of the mean curvature, and the sum of the third quartile of the mean curvature and the deviation of the mean curvature is determined as the maximum non-outlier value of the mean curvature. The difference between the first quartile of Gaussian curvature and the deviation of Gaussian curvature is determined as the minimum non-outlier value of Gaussian curvature, and the sum of the third quartile of Gaussian curvature and the deviation of Gaussian curvature is determined as the maximum non-outlier value of Gaussian curvature. The interval of non-outlier average curvature is formed by the minimum non-outlier average curvature and the maximum non-outlier average curvature, and the interval of non-outlier Gaussian curvature is formed by the minimum non-outlier Gaussian curvature and the maximum non-outlier Gaussian curvature.

2. The method according to claim 1, characterized in that, Determining the airborne data mean curvature sequence and the airborne data Gaussian curvature sequence corresponding to the target airborne data includes: Determine the average curvature and Gaussian curvature of each data point in the two-dimensional surface in the three-dimensional space corresponding to the target airborne data; The average curvature matrix of airborne data is constructed from the average curvature corresponding to all data points. The Gaussian curvature matrix of airborne data is constructed from the Gaussian curvatures corresponding to all data points. The airborne data mean curvature matrix and airborne data Gaussian curvature matrix are subjected to dimensionality reduction processing to obtain the airborne data mean curvature sequence and airborne data Gaussian curvature sequence, respectively.

3. The method according to claim 1, characterized in that, The process of determining anomalous data in the mean curvature sequence and anomalous data in the Gaussian curvature sequence of the airborne data based on the selected absolute deviation parameter, the non-anomaly range of the mean curvature, and the non-anomaly range of the Gaussian curvature is described, respectively. Target airborne data whose average curvature is outside the range of non-outlier average curvature values ​​and whose absolute value is greater than the absolute deviation parameter are considered as outlier data in the average series. Airborne data whose Gaussian curvature is outside the non-outlier range of Gaussian curvature and whose absolute value is greater than the square of the absolute deviation parameter are considered as Gaussian sequence outlier data.

4. The method according to claim 1, characterized in that, After combining the average sequence outlier data and the Gaussian sequence outlier data as the outlier data of the target airborne data, the method further includes: Target airborne data whose average curvature is within the range of non-outlier average curvature values, or whose absolute value is less than or equal to the absolute deviation parameter, are considered as non-outlier average data. Target airborne data whose Gaussian curvature is within the non-outlier range of the Gaussian curvature, or whose absolute value is less than or equal to the square of the absolute deviation parameter, are considered as non-outlier data of the Gaussian sequence. The intersection of the average curvature non-anomaly data and the Gaussian curvature non-anomaly data is taken as non-anomaly data.

5. The method according to claim 1, characterized in that, Before selecting the target airborne data corresponding to the number of independent variables in the target airborne data table, the method further includes: Obtain the airborne data table to be judged from the airborne data file to be judged; Based on the number of independent variables in the airborne data table to be judged, determine whether the airborne data table to be judged meets the requirement for the number of independent variables; If the requirement for the number of independent variables is met, then the airborne data table to be judged is taken as the target airborne data table.

6. The method according to claim 5, characterized in that, Determining whether the airborne data table to be judged meets the requirement for the number of independent variables includes: If the number of independent variables in the airborne data table to be judged is 1, it is determined that the airborne data table to be judged does not meet the requirement for the number of independent variables; If the number of independent variables in the airborne data table to be judged is 2 or 3, it is determined that the airborne data table to be judged meets the requirement for the number of independent variables.

7. The method according to claim 5, characterized in that, The selection of target airborne data corresponding to the number of independent variables in the target airborne data table includes: If the number of independent variables in the target airborne data table is 2, select all values ​​of the first independent variable, all values ​​of the second independent variable, and all values ​​of the dependent variable in the target airborne data table as the target data; If the number of independent variables in the target airborne data table is 3, select one independent variable from the three independent variables as the third independent variable, and select one value from the multiple values ​​of the third independent variable as the target variable value. Use the other two independent variables as the first independent variable and the second independent variable, respectively. All values ​​of the first independent variable, all values ​​of the second independent variable, and all values ​​of the dependent variable corresponding to the target variable values ​​in the target airborne data table are selected as target data.

8. An airborne anomaly data determination device, characterized in that, include: The data acquisition module is used to select target airborne data corresponding to the number of independent variables in the target airborne data table; The sequence determination module is used to determine the airborne data mean curvature sequence and the airborne data Gaussian curvature sequence corresponding to the target airborne data; The abnormal interval calculation module is used to determine the non-abnormal value intervals of the average curvature and the non-abnormal value intervals of the Gaussian curvature based on the selected relative deviation parameters, the average curvature quartiles and the average curvature interquartile ranges corresponding to the average curvature sequence of the airborne data, and the Gaussian curvature quartiles and the Gaussian curvature interquartile ranges corresponding to the Gaussian curvature sequence of the airborne data, respectively. The first anomaly determination module is used to determine the average series anomaly data corresponding to the average curvature series of the airborne data and the Gaussian series anomaly data corresponding to the Gaussian curvature series of the airborne data, respectively, based on the selected absolute deviation parameter, the non-anomaly value range of the average curvature, and the non-anomaly value range of the Gaussian curvature. The second anomaly determination module is used to combine the average series anomaly data and the Gaussian series anomaly data together as the anomaly data of the target airborne data. The abnormal interval calculation module is specifically used for: The product of the relative deviation parameter and the interquartile range of the average curvature is determined as the average curvature deviation, and the product of the relative deviation parameter and the interquartile range of the Gaussian curvature is determined as the Gaussian curvature deviation. The difference between the first quartile of the mean curvature and the deviation of the mean curvature is determined as the minimum non-outlier value of the mean curvature, and the sum of the third quartile of the mean curvature and the deviation of the mean curvature is determined as the maximum non-outlier value of the mean curvature. The difference between the first quartile of Gaussian curvature and the deviation of Gaussian curvature is determined as the minimum non-outlier value of Gaussian curvature, and the sum of the third quartile of Gaussian curvature and the deviation of Gaussian curvature is determined as the maximum non-outlier value of Gaussian curvature. The interval of non-outlier average curvature is formed by the minimum non-outlier average curvature and the maximum non-outlier average curvature, and the interval of non-outlier Gaussian curvature is formed by the minimum non-outlier Gaussian curvature and the maximum non-outlier Gaussian curvature.

9. An electronic device, characterized in that, include: The device includes a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the airborne anomaly data determination method as described in any one of claims 1 to 7.

Images