CN109948540B - Time sequence QAR parameter feature extraction method based on curve interpolation and sampling - Google Patents

Abstract

The invention relates to a time sequence QAR parameter feature extraction method based on curve interpolation and sampling, and belongs to the field of machine learning. The method comprises the following steps of S1: calculating the difference between a critical QAR parameter curve of the flight segment and a group average value curve in the landing stage to obtain a deviation curve; s2: interpolation processing is carried out on deviation curves of key QAR parameters such as IVV, PITCH, HEIGHT and the like, and the deviation curves are converted into continuous curves; s3: and sampling continuous deviation curves of key parameters such as IVV, PITCH, HEIGHT and the like at equal intervals to obtain n-dimensional feature vectors for subsequent clustering and other operations. The invention provides a QAR curve feature extraction method based on interpolation smoothing for the first time, the previous feature extraction method usually extracts QAR parameters at a plurality of fixed moments as features or extracts features from a whole flight curve in a sampling manner, which is not beneficial to acquiring detail features at a curve level, and the interpolation smoothing method of the invention can well extract the detail features of local curves at a landing stage.

Description

Time sequence QAR parameter feature extraction method based on curve interpolation and sampling

Technical Field

The invention belongs to the field of machine learning, and relates to a time sequence QAR parameter feature extraction method based on curve interpolation and sampling.

Background

The heavy landing research method based on QAR parameter curve visualization and individual and group comparison analysis can well assist a pilot to analyze the generation reason of the heavy landing event by visually presenting the QAR parameters in the flight process of the airplane to the pilot and providing the differentiation comparison between individuals and groups, thereby assisting the pilot to make a decision. However, this method has a certain disadvantage that the curve result needs to be analyzed manually, and the analyst needs to have a certain flight-related professional knowledge, which increases the system cost to some extent. The present invention therefore addresses the problem of: how to let the machine automatically infer the cause of the heavy landing event by analyzing the QAR data on the premise that the aircraft is known to have the heavy landing event? The key to achieving this goal is how to extract the first two chapters of knowledge about the analysis of the cause of the heavy landing into a form that can be understood by the machine. The previous analysis can find that the main difference between the heavy landing and the non-heavy landing is that the "shapes" of some key QAR parameter curves are different, and if the traditional feature extraction method is adopted, for example, the QAR parameter value is extracted as a feature at some key time points, it is obviously impossible to capture the feature at such curve level.

Li et al have already conducted QAR-related research based on clustering, and they mainly use clustering and outlier detection methods to find flight curve patterns with abnormal parameters from massive flight QAR data, and although they also extract curve-level features, their feature extraction methods have certain drawbacks, such as: the characteristic vector is formed by sampling and taking partial data points from the whole flight curve, and the method is a coarse-grained characteristic extraction method, can only reflect the general trend of the flight curve, and cannot obtain the local detail characteristics of the curve. Therefore, the features extracted by this method cannot be used for fine-grained analysis, such as analysis of the cause of heavy landing.

Disclosure of Invention

In view of the above, the present invention provides a method for extracting a QAR parameter feature based on curve interpolation and sampling.

In order to achieve the purpose, the invention provides the following technical scheme:

a method for extracting QAR parameter characteristics of a time sequence based on curve interpolation and sampling comprises the following steps:

in the characteristic extraction stage, firstly, a key QAR parameter curve of the flight segment is obtained, the deviation between the parameter curve and the group average curve is calculated to obtain a deviation curve, then, interpolation smoothing operation is carried out on the curve, characteristic vectors are extracted from the smoothed curve, the characteristic vectors of different curves are merged, and finally, an n-dimensional characteristic vector is obtained.

Further, the feature extraction stage is as follows:

in order to obtain the feature vectors with the same length and ensure that the features in the QAR parameter curve are not lost as much as possible, a feature extraction method based on interpolation smoothing is provided;

because the original QAR parameter curve is composed of discrete points, after the deviation curve of a certain navigation section is obtained, the discrete deviation curve is converted into a continuous curve by an interpolation method, and then points are taken from any position on the curve to form a characteristic vector, and the characteristic vector is not influenced by the length of the curve.

Further, the interpolation method includes linear interpolation and spline interpolation;

(1) Linear interpolation

The basic assumption of linear interpolation is that any point in the curve between two adjacent points is on the line connecting the two points, and the two points are assumed to be (x) respectively ₁ ,y ₁ ) And (x) ₂ ,y ₂ ) And the interpolation point is (x, y), then the following is satisfied:

given an arbitrary x ₁ ＜x＜x ₂ And solving the formula to obtain a linear interpolation target value y corresponding to x as:

(2) Spline interpolation

In numerical analysis, spline interpolation interpolates by using a special piecewise polynomial called spline, thereby ensuring that the curve after interpolation is not only continuous but also smooth.

Further, after the interpolation processing, the deviation curve of the QAR parameter is changed from a discrete curve to a continuous curve; the following is the extraction of features: certain QAR parameter for certain segmentp, equally dividing the time from the 50 feet of the flight to the ground into n _p Equal parts, assuming a time duration of t, then starting at a point in time corresponding to a height of 50 feet, every other time

One data point in seconds is taken from the interpolation curve for a total of n _p Data points, constituting the following feature vector:

and finally, combining the feature vectors corresponding to different parameters to obtain a final feature vector.

The invention has the beneficial effects that: a curve feature extraction method based on interpolation smoothing is firstly proposed, the previous feature extraction method usually extracts QAR parameters at a plurality of fixed moments as features or extracts the features from a whole line segment curve in a sampling mode, the curve feature extraction method is not beneficial to obtaining detail features of a curve level, the interpolation smoothing method can well extract local curve detail features of a landing stage, and clustering accuracy can be greatly improved.

Drawings

In order to make the object, technical scheme and beneficial effect of the invention more clear, the invention provides the following drawings for explanation:

FIG. 1 is an overall frame of the present invention;

FIG. 2 is a IVV deviation curve for a heavy landing leg;

FIG. 3 is a graph of a linear interpolation effect;

FIG. 4 is a B-spline interpolation effect;

FIG. 5 shows the K-means clustering results of IVV deviation curve features.

Detailed Description

Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

Please refer to fig. 1, which is an overall framework of the present invention.

1.1 feature extraction

The main purpose of feature extraction is to extract key features from the QAR parameter curves, which contain knowledge about the cause of heavy landing, and need to be converted into a form of feature vector that the machine can understand. It is known through analysis that the major differences of different subtypes of heavy landing are the 'shapes' of the parameter curves such as radio altitude, IVV, PITCH, etc., so the core idea of feature extraction is to extract the 'shape' features of these curves. The curve of the time from 50 feet to the grounding is used as a decision, so that the emphasis is on how to extract features from the parameter curve of the interval from 50 feet to the grounding, and an intuitive idea is to directly use the QAR parameter time sequence data of the interval from 50 feet to the grounding as a feature vector, but the method has the following two problems:

the first problem is that the parameter values at different time points vary in different ranges, and if the QAR parameter values are used directly as the features, those features with large parameter values tend to "overwhelm" features with small parameter values, and although normalization processing can be performed on different features, such processing may result in loss of information. In addition, it is known through analysis that the characteristics of the heavy landing of different subtypes are mainly reflected in the difference between the QAR parameter curve (such as IVV curve) and the average curve, and if the QAR parameter value is directly used as the characteristic, it is difficult to reflect the relative difference. Therefore, to solve this problem, the average curve is subtracted from the QAR parameter curve of the leg to be analyzed, resulting in a deviation curve therebetween, as shown in fig. 2. The red curve in the left graph of fig. 2 is IVV curve of a certain heavy landing leg, the blue curve is an average curve, and the green curve in the right graph of fig. 2 is IVV curve minus the average curve to obtain a deviation curve. Fig. 3 shows the linear interpolation effect.

The second problem is that the lengths of the 50 feet to the ground of different flight segments are different, for example, the length of the 50 feet to the ground of the I-1 type heavy landing is only less than 5 seconds, while the length of the 50 feet to the ground of the I-2 type heavy landing may reach 10 seconds, so that the obtained eigenvectors are inconsistent in length and cannot be processed by the clustering algorithm. One solution is to sample the same number of points from the curves of different legs to form feature vectors, however, this approach may result in loss of features in the curves, and decrease the classification accuracy of the clustering algorithm. Therefore, in order to obtain the feature vectors with the same length and ensure that the features in the QAR parameter curve are not lost as much as possible, a feature extraction method based on interpolation smoothing is provided. Because the original QAR parameter curve is composed of discrete points, after the deviation curve of a certain navigation section is obtained, the discrete deviation curve is converted into a continuous curve by an interpolation method, and then points can be taken from any position on the curve to form a characteristic vector without being influenced by the length of the curve. Two classical interpolation algorithms are considered here: linear interpolation and spline interpolation.

(1) Linear interpolation

The basic assumption of linear interpolation is that any point in the middle of two adjacent points in the curve is on the connecting line of the two points, and the two points are assumed to be (x) respectively ₁ ,y ₁ ) And (x) ₂ ,y ₂ ) And the interpolation point is (x, y), then the following is satisfied:

thus, given an arbitrary x ₁ ＜x＜x ₂ By solving the above formula, the linear interpolation target value y corresponding to x is obtained as:

the effect of linear interpolation is shown in fig. 2, where the two purple vertical lines correspond to 50 feet height and ground time, respectively. The left side is an uninserted dispersion curve, the right side is a continuous curve after linear interpolation, and the effect of linear interpolation is equivalent to connecting adjacent points by straight lines, so the left and right graphs in fig. 2 have the same shape.

(2) Spline interpolation

In numerical analysis, spline interpolation is performed by using a special piecewise polynomial called spline, thereby ensuring that the curve after interpolation is not only continuous and smooth, and spline interpolation has the advantage that a small interpolation error can be realized by using a low-order polynomial pattern. B-spline interpolation is a commonly used spline interpolation method, and the effect is shown in fig. 4, and it can be seen that the curve after B-spline interpolation is obviously smoother compared with the linear interpolation in fig. 4.

After interpolation, the deviation curve of the QAR parameter is changed from a discrete curve to a continuous curve, and the next operation is to extract features. For a QAR parameter p for a leg, the time from 50 feet of altitude to ground for the leg (assuming a time duration of t) is divided equally into n _p Equal parts, then every other, starting at a point in time corresponding to a height of 50 feet

One data point in seconds is taken from the interpolation curve for a total of n _p Data points, constituting the following feature vector:

and finally, combining the feature vectors corresponding to different parameters to obtain a final feature vector.

2 analysis of results

Feature extraction is carried out on 29 first-class heavy landing legs in the data set, then feature vectors are input to a K-means clustering device, parameters K =3 of the clustering device are set, and clustering results are shown in fig. 5. Here, for the sake of intuition, only the features extracted from the IVV curve are used, and the abscissa in fig. 5 represents the i-th parameter of the feature vector, and the ordinate represents the corresponding feature value. The characteristic curve is IVV deviation curve, and the interpolation method is B spline interpolation. Through curve feature extraction, clustering accuracy can be greatly improved, and analysis errors are reduced.

As can be seen from FIG. 5, the K-means algorithm better groups the deviation curves into three categories, where the red curve corresponds to a type I-2 heavy landing, i.e., after entering 50 feet of altitude, the first half of the curve is significantly higher than zero because the pilot pulls too much to cause IVV to decrease rapidly, with the absolute value being less than the average (note that IVV is negative), and the second half of the curve is significantly higher than zero because the first half IVV decreases too rapidly to cause the aircraft to land in time, so the pilot pushes the second half of the curve to cause IVV to increase, and to be substantially greater than the average, so that the latter half of the curve will appear to be less than zero. The blue curve represents a type I-1 heavy landing, i.e., after entering 50 feet of altitude, the pilot has not pulled the stick in time to cause IVV to be consistently above average, and finally IVV lands without having fallen, so it can be seen that the blue curve is substantially below zero. The green curve corresponds to an I-3 type heavy landing, i.e. the first half IVV is better controlled, but the second half IVV does not have a tendency to continuously keep descending to cause heavy landing, so that the deviation curve is seen to be closest to zero.

And (3) aiming at different QAR parameters, extracting characteristics by adopting different interpolation methods, clustering 29 first class heavy landing samples, comparing with an artificial classification result, and sorting the accuracy of the classification result into a table 1. Table 1 shows the accuracy of the clustering result obtained by clustering using IVV, PITCH, and heigh deviation curves, respectively, where the number of features sampled by each curve is 50, that is, each sample is a feature vector with a length of 50.

TABLE 1 Re-landing Classification accuracy

Rate of accuracy	IVV feature	PITCH feature	HEIGHT characteristics
				Type I-1 accuracy	86.7％	93.3％	73.3％
Type I-2 accuracy	100％	25％	100％
				Type I-3 accuracy	100％	70％	90％
Overall rate of accuracy	93.1％	75.7％	82.8％

In 29 first-type heavy landing legs, 15, 4 and 10 manually classified I-1, I-2 and I-3 heavy landings are respectively arranged. From the classification result, the IVV curve feature has the best classification effect, only 2 errors are made, the total classification accuracy is as high as 93.1%, wherein the prediction accuracy of the I-2 and I-3 types is more 100%, the prediction accuracy of the I-1 type is 86.5%, 2 errors are made in 15 flight segments, and through observing the graphs of the two flight segments, we find that certain difficulty exists in classifying the two flight segments even if people classify the two flight segments, which further explains the effectiveness of the proposed clustering algorithm. The classification effect of the PITCH curve characteristics is worse, the overall accuracy is 75.9% (7 wrong divisions), wherein the prediction accuracy of the types I-1, I-2 and I-3 is 93.3%, 25% and 70% (the wrong divisions are 1/3/3), respectively, the prediction effect of the PITCH characteristics on the type I-1 heavy landing is best, and the prediction effect on the type I-2 heavy landing is worst. The overall classification effect of the HEIGHT characteristics is between IVV and PITCH, the overall accuracy is 82.8 (5 wrong scores), wherein the prediction accuracy rates of I-1, I-2 and I-3 types are 73.3%, 100% and 90% respectively (the wrong score numbers are 4/0/1 respectively), the HEIGHT characteristics have the best prediction effect on I-2 type heavy landing and the worst prediction effect on I-1 type.

In conclusion, it can be seen that the IVV curve feature has the best classification effect, and the PITCH curve feature has the good prediction effect on the I-1 type heavy landing, but it is difficult to identify the I-2 type heavy landing (the identification rate is only 25%), and the HEIGHT curve feature is better at predicting the I-2 type heavy landing. Overall, IVV and HEIGHT have better overall performance and perform more equally on different subtypes of heavy landing samples. The predicted performance characteristics of different parameter characteristics are different, and different parameter characteristics can be combined by some means in the future, so that the respective advantages of the different parameter characteristics are fully exerted.

Finally, it is noted that the above-mentioned preferred embodiments illustrate rather than limit the invention, and that, although the invention has been described in detail with reference to the above-mentioned preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined by the appended claims.

Claims (2)

1. A time sequence QAR parameter feature extraction method based on curve interpolation and sampling is characterized by comprising the following steps: the method comprises the following steps:

in the characteristic extraction stage, firstly, a key QAR parameter curve of the flight segment is obtained, the deviation between the parameter curve and the group average curve is calculated to obtain a deviation curve, then, interpolation smoothing operation is carried out on the curve, namely, a discrete deviation curve is converted into a continuous curve by an interpolation method;

after the interpolation processing, the deviation curve of the QAR parameter is changed from a discrete curve to a continuous curve; the following is the extraction of features: for a QAR parameter p for a leg, the time from 50 feet of altitude to ground for the leg is divided equally into n _p Equal parts, assuming a time duration of t, then starting at a point in time corresponding to a height of 50 feet, every other time

One data point in seconds is taken from the interpolation curve for a total of n _p Data points, constituting the following feature vector:

and finally, combining the feature vectors corresponding to different parameters to obtain a final feature vector.

2. The method of claim 1, wherein the method of extracting QAR parameters based on curve interpolation and sampling comprises: the interpolation method comprises linear interpolation and spline interpolation;

(1) Linear interpolation

The basic assumption of linear interpolation is that any point in the curve between two adjacent points is on the line connecting the two points, and the two points are assumed to be (x) respectively ₁ ,y ₁ ) And (x) ₂ ,y ₂ ) And the interpolation point is (x, y), then the following is satisfied:

given an arbitrary x ₁ ＜x＜x ₂ And solving the formula to obtain a linear interpolation target value y corresponding to x as:

(2) Spline interpolation

In numerical analysis, spline interpolation interpolates by using a special piecewise polynomial called spline, thereby ensuring that the curve after interpolation is not only continuous but also smooth.

Images