CN101900834A - Method for detecting ionized layer TEC exception - Google Patents

Abstract

The invention discloses a method for detecting ionized layer TEC exception. The method comprises the following steps of: A, performing smoothing filtration on an ionized layer TEC time sequence; selecting the TEC time sequence of any point according to the space-time characteristics of ionized layer TEC distribution, and performing smoothing filtration on the ionized layer TEC time sequence by adopting a moving average method to obtain a smoothed TEC time sequence; B, calculating a difference between a TEC smoothing value and an observation value so as to obtain a time sequence of a smoothed residual error; C, determining the arc length of a sliding window, calculating a mean value and a standard difference, adding 2 times standard difference to the mean value to obtain an upper limit value, subtracting 2 times standard difference from the mean value to obtain a lower limit value, performing exception detection on the time sequence of the residual error, comparing the data with the upper limit value and the lower limit value, and judging that the data between the upper limit value and the lower limit value is normal; and D, storing the detection result. The method is simple, has obvious effect, creatively performs the exception detection of the ionized layer TEC based on the smoothed residual error data, deducts the long periodic tendency of the TEC, and greatly improves the reliability of the detection result.

Description

A kind of method that detects ionized layer TEC exception

Technical field

The present invention relates to the monitoring technical field to space environment, more specifically relate to a kind of method that detects ionized layer TEC exception, it is a kind of monitoring method that is fit to space ionized layer TEC complicated and changeable.

Background technology

As the important component part of geospace environment, ionospheric activity and Human's production and life are closely bound up, and ionosphere is to modern radio engineering system and human space operation important influence; About solar activity, earth movements and terrestrial magnetic field change ionospheric existence and change in time and space.It is unusual that the outburst of solar flare can make ionosphere occur, and influences wireless communication, navigation and the business activity relevant with electromagnetism; The local dip of the sun can cause that also ionospheric normal configuration causes ionospheric storm, thereby makes the short wave communication signal decay significantly; In addition, seismic activity also can be brought out the generation and the propagation of acoustic-gravity wave in ionosphere.Therefore, the detection research to space ionized layer TEC exception activity becomes one of focus of current space environment study on monitoring.

Space environment and our many activities around the earth are closely related, therefore, in ionized layer TEC like this changeable today, press for and seek a kind of space environment of monitoring effectively and accurately and change and particularly discern the new approaches and methods whether ionized layer TEC ANOMALOUS VARIATIONS occurs.

Summary of the invention

The objective of the invention is at the ionized layer TEC around the present earth complicated and changeable, thereby remote effect are to Human's production and problem concerning life, a kind of method that detects ionized layer TEC exception has been proposed, this method is simple, be beneficial to programming, TEC residual error time series after choosing has smoothly greatly improved the accuracy of testing result as research object, has significant scientific research and use value.

In order to achieve the above object, the present invention adopts following technical measures:

The present invention includes smothing filtering to total electron content (TEC, Total Electronic Contents) data, to smooth value and observed reading do poor, select for use the sliding window method that residual error data is handled three parts to realize.A kind of method that detects ionized layer TEC exception, it comprises following four steps:

(1) adopt moving average method (" astronomical surveing processing method of data ") that the time series of ionized layer TEC grid points is carried out smothing filtering;

(2) observed reading of TEC and smooth value are done poor, obtained the time series of the TEC residual error after level and smooth;

(3) choose the Treatment Analysis method (those skilled in the art are very clear for this Treatment Analysis method) of sliding window: the sliding window segmental arc of both having chosen is 10 days, calculates TEC average and standard deviation in these 10 days;

(4) utilize the method for sliding window that the residual error time series after level and smooth is carried out abnormality detection: the standard deviation of choosing 2 times is as foundation, and the standard deviation of getting 2 times of average plus-minuss is as higher limit and lower limit; The part that then exceeds higher limit and lower limit promptly is considered as unusually.

The present invention compared with prior art has the following advantages and effect:

The first, this method is based on proposing on the grasp basis to the ionized layer TEC space-time characterisation, utilizes the TEC residual error data after level and smooth, from the source elimination the time cycle property that itself had, improved the reliability of testing result.

The second, behind the long periodicity item of deduction TEC, adopt the extracting method of sliding window, make that the background value of choosing is more sane, thereby further guaranteed the reliability of ionized layer TEC testing result.

The 3rd, this method principle is simple, and coding is simple, and is workable, is convenient to practical operation.

The present invention provides a kind of new monitoring method for ionized layer TEC, the inventive method is simple, and effect is obvious, carries out the abnormality detection of ionized layer TEC innovatively based on the residual error data after level and smooth, deduct the long period trend of TEC, significantly improved the reliability of testing result.This method innovation ground carries out the detection of abnormal movement based on the residual error data after level and smooth, elimination the cyclophysis of ionized layer TEC itself, reduced the interference of error source to testing result, increased substantially the reliability of testing result.The ionized layer TEC exception activity that utilizes this method to detect to obtain can be satisfied requirements such as space environment being continued detection accurately and reliably.

Description of drawings

Fig. 1 is a kind of method flow diagram that detects ionized layer TEC exception;

Fig. 2 is a kind of detected ionized layer TEC exception distribution plan.

Embodiment

Embodiment 1:

Below by the case description embodiments of the present invention, concrete workflow is described below as shown in Figure 1.

A kind of method that detects ionized layer TEC exception, it comprises the steps:

The first step, the ionized layer TEC time series is carried out smothing filtering 1; Because the space-time characteristic that ionized layer TEC distributes, we choose the time series X (T of any point TEC _j) (j=1～n) adopts the method for running mean that the ionized layer TEC time series is carried out smothing filtering, obtains the TEC time series X ' (T after level and smooth _j):

$X^{\′}\left(T_j\right)=\frac{1}{2n+1}\underset{k=1}{\overset{n}{\mathrm{\Σ}}}X\left(T_j\right),(j=\mathrm{1,2,3}........)$

The difference 2 of second step, calculating TEC smooth value and observed reading obtains the smoothly time series Δ X of back residual error _i

ΔX _i＝X(T _j)-X′(T _j)(j＝1，2，3......)

The 3rd step, determine the arc length of sliding window, computation of mean values and standard deviation are got average and are added, subtract 2 times of standard deviations and do upper limit value and lower limit value, and the time series of residual error is carried out abnormality detection 3, and (method that adopts sliding window is to residual sequence Δ X _iHandle); Specific implementation method is as follows:

If a chooses 30 days observation segmental arc, determine that form length is 10 days.

TEC residual error data Δ X in b, the 1st day to the 10th day the 10 days segmental arcs of calculating _iAverage and standard deviation;

$\stackrel{\‾}{X}=\frac{1}{N}\mathrm{\Σ\Δ}{X}_i,(i=\mathrm{1,2,3}......10)$

$\mathrm{\σ}=\sqrt{\frac{1}{N-1}\mathrm{\Σ}{({\mathrm{\ΔX}}_i-\stackrel{\‾}{X})}^2},(i=\mathrm{1,2,3}......10)$

Wherein: X is an average, and N is 10, Δ X _iBe the residual values of i days TEC, σ is a standard deviation.

C, determine unusual higher limit (Upper_bond) and lower limit (Lower_bond): higher limit and the lower limit of the standard deviation of 2 times of the plus-minuss of averaging during as abnormality detection:

$\left\{\begin{array}{c}\mathrm{Upper}\_\mathrm{bond}=\stackrel{\‾}{X}+2\mathrm{\σ}\\ \mathrm{Lower}\_\mathrm{bond}=\stackrel{\‾}{X}-2\mathrm{\σ}\end{array}\right.$

The 11st day data and higher limit and lower limit are compared, and what exceed higher limit is called the disturbance of ionized layer TEC positive anomaly, if being lower than lower limit then is called the disturbance of ionized layer TEC negative anomaly, is normal between higher limit and lower limit;

D, calculate the 2nd day to the 11st day average and standard deviation, then 2 times of standard deviations of average plus-minus are as detecting the standard whether the 12nd day ionosphere TEC abnormal disturbances occurs, copy pair the 12nd day ionized layer TEC of step 3 to carry out abnormality detection.

E, the rest may be inferred by analogy, and whether the TEC that just can detect in the 11st day to the 30th day abnormal disturbances occurred;

The 4th step, preservation testing result 4, testing result is asked for an interview Fig. 2.According to Fig. 1 Treatment Analysis method, the applicant detects the TEC in 23 1 months April 23 to Mays in 2009, the result shows April 29, No. 6 of May, No. 7, tangible abnormal disturbances has appearred in No. 9 and No. 21 ionized layer TECs, as shown in Figure 2, transverse axis is represented the time, and the longitudinal axis is the part that exceeds upper lower limit value, and unit is TEC _U, April 29 wherein, May 6, No. 7 is the TEC negative anomaly, May 9 and be for No. 21 the positive anomaly disturbance.

Claims (1)

1. a method that detects ionized layer TEC exception the steps include:

A, the ionized layer TEC time series is carried out smothing filtering (1); Consider the space-time characteristic that ionized layer TEC distributes, choose the time series X (T of any point TEC _j), adopt the method for running mean that the ionized layer TEC time series is carried out smothing filtering, obtain the TEC time series X ' (T after level and smooth _j):

$X^{\′}\left(T_j\right)=\frac{1}{2n+1}\underset{k=1}{\overset{n}{\mathrm{\Σ}}}X\left(T_j\right)$

B,, calculate the difference (2) of TEC smooth value and observed reading, obtain smoothly the time series Δ X of residual error afterwards _i

ΔX _i＝X(T _j)-X′(T _j)

C, determine the arc length of sliding window, computation of mean values and standard deviation are got average and are added, subtract 2 times of standard deviations and do upper limit value and lower limit value, and the time series of residual error is carried out abnormality detection (3), and step is as follows:

A, choose 30 days observation segmental arc, determine that form length is 10 days;

TEC residual error data Δ X in b, the 1st day to the 10th day the 10 days segmental arcs of calculating _iAverage and standard deviation;

$\stackrel{\‾}{X}=\frac{1}{N}\mathrm{\Σ\Δ}{X}_i$

$\mathrm{\σ}=\sqrt{\frac{1}{N-1}\mathrm{\Σ}{(\mathrm{\Δ}{X}_i-\stackrel{\‾}{X})}^2}$

Wherein:

Be average, N is 10, Δ X _iBe the residual values of i days TEC, σ is a standard deviation;

C, determine unusual higher limit and lower limit: higher limit and the lower limit of the standard deviation of 2 times of the plus-minuss of averaging during as abnormality detection:

$\left\{\begin{array}{c}\mathrm{Upper}\_\mathrm{bond}=\stackrel{\‾}{X}+2\mathrm{\σ}\\ \mathrm{Lower}\_\mathrm{bond}=\stackrel{\‾}{X}-2\mathrm{\σ}\end{array}\right.$

The 11st day data and higher limit and lower limit are compared, and what exceed higher limit is called the disturbance of ionized layer TEC positive anomaly, is lower than lower limit and just then is called the disturbance of ionized layer TEC negative anomaly, and being considered as between higher limit and lower limit is normal;

D, calculate the 2nd day to the 11st day average and standard deviation, 2 times of standard deviations of average plus-minus are as detecting the standard whether the 12nd day ionosphere TEC abnormal disturbances occurs, copy step (c) that the 12nd day ionized layer TEC is carried out abnormality detection;

E, the rest may be inferred by analogy, and whether the TEC that just detects in the 11st day to the 30th day abnormal disturbances occurred;

D, preservation testing result (4).

Images