CN113324646A - Big wind area electrified railway contact net positive feeder line galloping positioning algorithm - Google Patents
Big wind area electrified railway contact net positive feeder line galloping positioning algorithm Download PDFInfo
- Publication number
- CN113324646A CN113324646A CN202110578103.6A CN202110578103A CN113324646A CN 113324646 A CN113324646 A CN 113324646A CN 202110578103 A CN202110578103 A CN 202110578103A CN 113324646 A CN113324646 A CN 113324646A
- Authority
- CN
- China
- Prior art keywords
- signal
- imf
- galloping
- trend
- decomposition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
A big wind zone electrified railway contact net positive feeder galloping positioning algorithm comprises the following steps: s10, acquiring a monitoring point waving signal to obtain signal components corresponding to the x, y and z axes; s20, decomposing the signal component obtained in the step S10 to obtain an IMF sequence; s30, extracting trend items from the IMF sequence of the step S20; s40, subtracting the trend term obtained in the step S30 from the original signal of the signal component to obtain a corrected signal component; and S50, superposing the waveforms of the three-axis signal components to obtain the real galloping track of the monitoring point. The invention introduces an EEMD method to decompose the original signal component containing the trend item to obtain a series of IMF sequences, and provides a trend item extraction method with higher automation degree compared with the traditional method according to the characteristic that the waving signal is approximately symmetrical about a time axis, thereby realizing the correction of the waving signal waveform and obtaining the real waving track of the monitoring point. The invention is a galloping signal positioning algorithm with small calculated amount and high operation efficiency, and is suitable for processing the online galloping monitoring signal of the positive feeder of the contact network of the electrified railway.
Description
Technical Field
The invention relates to the technical field of overhead conductor monitoring, in particular to a positive feeder galloping positioning algorithm for a contact network of an electrified railway in a large wind area.
Background
The galloping of the positive feeder of the electrified railway contact net in the big wind area is self-excited vibration with low frequency and large amplitude under the action of wind excitation. In nature, this form of vibration is referred to as relaxation vibration; in terms of shape, it is called dancing because it flies up and down like dragon dance. The swinging of the positive feeder line leads to the aggravation of the abrasion of a connecting hardware fitting, the fatigue of a thread at a suspension point, strand breakage and line drop, insufficient safety distance discharge between wires, the tripping of a contact network and other faults, and the swinging of the positive feeder line becomes an important hidden danger threatening the safe operation of an electrified railway in a windy area.
In order to obtain amplitude and frequency information of the contact net positive feeder line galloping in real time, a wire galloping online monitoring system is built based on a wireless sensor network technology. And the monitoring terminal acquires acceleration information of the galloping monitoring point in the x, y and z directions in real time by using the triaxial acceleration sensor. After the acceleration signals of the monitoring points are preprocessed, integral operation can be carried out on the acceleration signals, and the speed and the displacement of the monitoring points are solved. Because the data processing methods of the three axial directions of x, y and z output by the triaxial acceleration sensor are similar, for avoiding redundancy, the x-axis monitoring data output by the monitoring terminal is taken as a processing object, the acceleration signals of the monitoring points are integrated for one time to obtain corresponding speed signals, and the speed signals are integrated again to obtain displacement signals of the monitoring points.
Theoretically, it is feasible to obtain the displacement by integrating the acceleration twice, but due to the environmental interference around the sensor, the collected result often generates a larger trend term, and in addition, the error accumulation in the integration process, the finally obtained displacement signal is even completely submerged. In order to accurately obtain the amplitude of the oscillation of the positive feeder line of the contact network, the trend item of the signal must be removed. Therefore, an efficient method for modifying the waveform of the integrated signal is needed.
At present, the common methods for extracting the signal trend item include a difference method, a low-pass filtering method and a polynomial fitting method. The difference method has small operation amount, the low-pass filtering method can only filter high-frequency noise, the polynomial fitting method has simple algorithm, and the three methods all need to presuppose the type of a trend item, such as linear, polynomial or exponential trend, and are not suitable for processing irregular or complex signals with no trend change. With the continuous and deep research, many scholars propose various new trend term extraction and removal methods in different fields, and the method mainly combines algorithms such as wavelet change and least square method and introduces the combined algorithms into the extraction of the signal trend term. However, in terms of analysis, the existing algorithm has high dependency on prior knowledge, high algorithm complexity and low automation degree for identifying the trend term.
Disclosure of Invention
In view of the above technical problems, embodiments of the present invention provide a positive feeder galloping positioning algorithm for a catenary of an electrified railway in a large wind area, which can identify a trend item, thereby implementing the correction of a galloping signal and obtaining a more accurate and true galloping signal waveform.
A big wind zone electrified railway contact net positive feeder galloping positioning algorithm comprises the following steps:
s10, acquiring a monitoring point waving signal to obtain signal components corresponding to the x, y and z axes;
s20, decomposing the signal component obtained in the step S10 to obtain an IMF sequence;
s30, extracting trend items from the IMF sequence of the step S20;
s40, subtracting the trend item extracted in the step S30 from the original signal component to obtain the waveform of the corrected signal component;
and S50, obtaining the real galloping track of the monitoring point according to the waveform of the triaxial signal component.
Step S20 includes:
s21, setting an amplitude coefficient k and a decomposition frequency M of white gaussian noise added to the integrated signal, and executing an initial frequency M of 1, and then M of M + 1;
s22, adding Gaussian white noise signal g (t) into original signal x (t) to obtain mixed signal x to be decomposedi(t)：
xi(t)＝x(t)+kg(t) (1)
S23, for xi(t) performing EMD decomposition to obtain n IMF sequences, wherein the ith IMF sequence obtained by the mth decomposition is as follows:
imfi,m(i＝1,2,...,n),n＝n+1 (2)
s24, if M < M, repeatedly executing S22 and S23;
s25, averaging IMF sequences obtained by M times of decomposition:
s26, the final decomposition result of EEMD is:
where δ is the remainder of the EEMD decomposition.
Step S30 includes:
s31, calculating the average value of the original signal x (t):
s32, calculating the mean sum of n signals after the jth IMF sequence reaches the remainder delta:
s33, when the ratio c of the two average values is judged to be within the allowable range of the error, determining the value of j:
s34, judging whether c meets the set condition, if true, finishing the judgment, outputting the j-th IMF sequence to the remainder delta and a trend term r (t) of the signal:
if the result is false, let j equal j-1, and go to step S32, continue judging until the condition is satisfied, and exit the loop.
The conditions set in step S34 are to determine whether c is between 0.95 and 1.05, where c represents the ratio of the average value of the original signal to the average value of the reconstructed trend term, and the range of c is an empirical value set according to a large number of research results. When c is within a set range of 0.95-1.05, it can be shown that the extracted trend term is correct.
Aiming at the defects of the prior art, the EEMD method is introduced to decompose the original signal component containing the trend item to obtain a series of IMF sequences, and the trend item extraction method with higher automation degree compared with the traditional method is provided according to the characteristic that the waving signal is approximately symmetrical about a time axis, so that the wave shape of the waving signal is corrected, and the real waving track of the monitoring point is obtained. The invention is a galloping signal positioning algorithm with small calculated amount and high operation efficiency, and is suitable for processing the online galloping monitoring signal of the positive feeder of the contact network of the electrified railway.
Drawings
The invention may be better understood from the following description of specific embodiments thereof taken in conjunction with the accompanying drawings, in which:
other features, objects and advantages of the invention will become apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings in which like or similar reference characters refer to the same or similar parts.
FIG. 1 is an x-axis direction raw velocity signal containing trend terms to be processed;
FIG. 2 is a sequence of IMFs after decomposition of the velocity signal EEMD;
FIG. 3 is a waveform diagram before and after a velocity signal elimination trend term;
FIG. 4 is a comparison graph of the effect of two methods on removing signal trend terms;
FIG. 5 is a graph of trend term difference values extracted by two methods;
FIG. 6 is a waveform diagram before and after the displacement signal removes the trend term;
FIG. 7 is a three-axis displacement output diagram of a wire galloping monitoring point;
FIG. 8 is a graph of the motion trajectory of the dance monitor point in a cross section perpendicular to the wire;
FIG. 9 is a scatter plot of the dancing monitor point signal in space;
FIG. 10 is a flow chart of the trend term extraction after EEMD decomposition in accordance with the present invention.
Detailed Description
Features and exemplary embodiments of various aspects of the present invention will be described in detail below. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced without these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples thereof. The present invention is in no way limited to any specific configuration and algorithm set forth below, but rather covers any modification, replacement or improvement of elements, components or algorithms without departing from the spirit of the invention.
Example embodiments will now be described with reference to the accompanying drawings, which may be embodied in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art.
In order to verify the effectiveness of the galloping positioning algorithm provided by the invention, simulation application analysis is carried out by taking an x-axis speed signal v (t) in fig. 1 as an example, a Gaussian white noise signal g (t) constructed by the invention is 1000 points, and the standard deviation is 0.2. The white noise amplitude coefficient k is 0.05, the decomposition frequency M is 100, and the white noise amplitude coefficient k and the decomposition frequency M are added to the velocity signal to obtain vi(t), EEMD decomposition is carried out to obtain IMF sequence as shown in FIG. 4.
According to the improved method for judging the trend term in the signal, the average value of the speed signal v (t) is obtained
Is 2.2063 m/s. In fig. 2, for the convenience of statistics, the remainder δ of the velocity signal in the x-axis direction of the waving monitoring point after EEMD decomposition is also regarded as an eigenmode function, i.e., IMF10 in fig. 2, and the average value of each IMF sequence obtained by calculation is shown in table 1.
TABLE 1 mean value of IMF sequences of each order after velocity signal decomposition
The mean sums of IMF8, IMF9, and IMF10 from Table 1
2.3014, then
The magnitude of the ratio c of (a) is 0.9587, which is within the allowable error range relative to the ideal ratio of 1, so the trend term of the velocity signal v (t) in fig. 1 is as follows:
according to the result of the component signal decomposition, after the trend term of the original signal is determined by using the positioning algorithm of the present invention, the trend term r (t) is subtracted from the original signal v (t) to obtain a signal without the trend term, i.e. a corrected signal, as shown in fig. 3.
As can be seen from FIG. 3, after the trend term of v (t) is removed, the signal fluctuation returns to the vicinity of the 0 axis, which illustrates that the trend term of the signal can be effectively removed by the method of the present invention. In order to check whether the method correctly extracts and eliminates the trend items, the invention selects a relatively mature moving average method to compare and verify the method.
The moving average method is a simple and effective method for eliminating the trend term. The method is characterized in that a smooth trend term curve is obtained by carrying out data smoothing processing on signals for multiple times, and the irregular trend of the signals can be eliminated by subtracting the trend term from the original signals. In order to effectively extract and eliminate the trend item in the speed signal, the invention carries out the sliding average processing with the sliding order of 30 and the smoothing number of 300 on the speed signal on the basis of a plurality of tests according to the characteristics of the waving signal. The trend term and the waveform after eliminating the trend term in the original velocity signal obtained by the moving average method are shown in fig. 4.
From fig. 4, the method and the moving average method of the present invention can both better realize the extraction and removal of the signal trend term, and intuitively, the trend term obtained by the two methods and the output curve after the trend term is eliminated have higher goodness of fit. The trend term signal extracted by the moving average method is subtracted by the method of the invention to obtain a relative error curve of the trend term signal and the moving average term signal, which is shown in figure 7.
The time domain characteristic statistics for the two trend term curves are shown in table 2. As can be seen from FIG. 5 and Table 2, the curves of the signal trend terms extracted by the method and the moving average method have small difference no matter from the viewpoint of quantitative analysis or qualitative analysis, and the error is between-0.5 and 0.5 from the viewpoint of data size. The mean value, the variance and the standard deviation of the trend items extracted by the two methods are similar in numerical value, and the relative error rates of the two methods are all within 5% on the basis of a moving average method, so that the method for identifying and extracting the signal trend items provided by the invention is verified to be correct and effective.
TABLE 2 comparison of trend term results extracted by the two methods
On the basis of proving that the extraction and elimination of the velocity signal trend term by the method of the invention are correct, in order to obtain the displacement signal of each galloping monitoring point, the velocity signal output after the trend term is eliminated in figure 3 is taken as the basis, the velocity signal is integrated again to obtain the required displacement oscillogram, a trend term is generated in the integration process, and the displacement waveform after the integration is necessarily deformed to a certain degree. Similarly, the method of the present invention is used to reconstruct and reject the trend term again for the displacement signal obtained by integration, so as to obtain a displacement signal without the trend term, as shown in fig. 6.
As can be seen from the graph 6, the amplitude of the galloping of the selected conductor galloping monitoring point in the x-axis direction is-1.5 m, and the amplitude and the frequency of the calculated displacement signal accord with the actual galloping condition of the positive feeder line, so that the method for extracting and removing the signal trend term is effective and reliable. Similarly, the method of the invention is used for carrying out integral and trend item elimination processing on the acceleration data of the y axis and the z axis output by the triaxial acceleration sensor of the waving monitoring point to obtain displacement signals of three axial directions, as shown in fig. 7.
It has been found that when the conductor is waved, the waved path can be replaced by a slightly windward inclined ellipse in a plane perpendicular to its cross-section. According to the principle of the least square method, the motion track of the node No. 2 is fitted in a section perpendicular to the axis of the positive feeder line, as shown in FIG. 8, and the distribution of the scattered points of the monitoring signal in a three-dimensional space is shown in FIG. 9.
As can be seen from fig. 8 and 9, after the acceleration signal output by the waving monitoring terminal is processed by the positioning algorithm of the present invention, the motion trajectory in the cross section perpendicular to the axis of the wire is approximately an oblique ellipse, and the major axis of the oblique ellipse is about 2m and the minor axis is about 1 m. The motion amplitudes of the overall trend of the fitted trajectory in the horizontal and vertical directions in fig. 8 have better coincidence than those in fig. 7. The distribution of a displacement signal scatter diagram obtained after the acceleration signals output by the monitoring points are positioned in the space accords with the actual situation, and the method verifies that the contact net positive feeder line galloping positioning algorithm is effective and feasible and can meet the requirements of the practical application of the engineering.
It will be appreciated by persons skilled in the art that the above embodiments are illustrative and not restrictive. Different features which are present in different embodiments may be combined to advantage.
Claims (4)
1. A big wind zone electrified railway contact net positive feeder galloping positioning algorithm comprises the following steps:
s10, acquiring a monitoring point waving signal to obtain signal components corresponding to the x, y and z axes;
s20, decomposing the signal component obtained in the step S10 to obtain an IMF (Intrinsic Mode Function) sequence;
s30, extracting trend items from the IMF sequence of the step S20;
s40, subtracting the trend term obtained in the step S30 from the original signal of the signal component to obtain the waveform of the corrected signal component;
and S50, obtaining the real galloping track of the monitoring point according to the waveform of the triaxial signal component.
2. The method of claim 1,
step S20 includes:
s21, setting an amplitude coefficient k and a decomposition frequency M of white gaussian noise added to the integrated signal, and executing an initial frequency M of 1, and then M of M + 1;
s22, adding Gaussian white noise signal g (t) into original signal x (t) to obtain mixed signal x to be decomposedi(t)：
xi(t)＝x(t)+kg(t) (1)
S23, for xi(t) performing EMD (Empirical Mode Decomposition) Decomposition to obtain n IMF sequences, wherein the ith IMF sequence obtained by the mth Decomposition is as follows:
imfi,m(i＝1,2,...,n),n＝n+1 (2)
s24, if M < M, repeatedly executing S22 and S23;
s25, averaging IMF sequences obtained by M times of decomposition:
s26, the final Decomposition result of EEMD (Ensemble Empirical Mode Decomposition) is:
where δ is the remainder of the EEMD decomposition.
3. The method of claim 1,
step S30 includes:
s31, calculating the mean value of the original signal x (t):
s32, calculating the mean sum of n signals after the jth IMF sequence reaches the remainder delta:
s33, when the ratio c of the two average values is judged to be within the allowable range of the error, determining the value of j:
s34, judging whether c meets the set condition, if true, finishing the judgment, outputting the j-th IMF sequence to the remainder delta and a trend term r (t) of the signal:
if the result is false, let j equal j-1, and go to step S32, continue judging until the condition is satisfied, and exit the loop.
4. The method of claim 3,
the conditions set in step S34 are to determine whether c is between 0.95 and 1.05, where c represents the ratio of the average value of the original signal to the average value of the extracted trend term, and the range of c is an empirical value set according to a large number of research results. When c is within a set range of 0.95-1.05, it can be shown that the extracted trend term is correct.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110578103.6A CN113324646A (en) | 2021-05-26 | 2021-05-26 | Big wind area electrified railway contact net positive feeder line galloping positioning algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110578103.6A CN113324646A (en) | 2021-05-26 | 2021-05-26 | Big wind area electrified railway contact net positive feeder line galloping positioning algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113324646A true CN113324646A (en) | 2021-08-31 |
Family
ID=77415118
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110578103.6A Pending CN113324646A (en) | 2021-05-26 | 2021-05-26 | Big wind area electrified railway contact net positive feeder line galloping positioning algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113324646A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102279084A (en) * | 2011-05-03 | 2011-12-14 | 西安工程大学 | Transmission line oscillation positioning system and method based on micro inertial measurement combination |
| CN103617356A (en) * | 2013-11-27 | 2014-03-05 | 国家电网公司 | Self-adaptive on-line monitoring data trend extraction method |
| CN106897740A (en) * | 2017-02-17 | 2017-06-27 | 重庆邮电大学 | EEMD DFA feature extracting methods under Human bodys' response system based on inertial sensor |
| CN112697263A (en) * | 2020-11-09 | 2021-04-23 | 山东柯瑞申智能科技有限公司 | EEMD multi-scale fluctuation analysis state monitoring method and device |
-
2021
- 2021-05-26 CN CN202110578103.6A patent/CN113324646A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102279084A (en) * | 2011-05-03 | 2011-12-14 | 西安工程大学 | Transmission line oscillation positioning system and method based on micro inertial measurement combination |
| CN103617356A (en) * | 2013-11-27 | 2014-03-05 | 国家电网公司 | Self-adaptive on-line monitoring data trend extraction method |
| CN106897740A (en) * | 2017-02-17 | 2017-06-27 | 重庆邮电大学 | EEMD DFA feature extracting methods under Human bodys' response system based on inertial sensor |
| CN112697263A (en) * | 2020-11-09 | 2021-04-23 | 山东柯瑞申智能科技有限公司 | EEMD multi-scale fluctuation analysis state monitoring method and device |
Non-Patent Citations (2)
| Title |
|---|
| 张军: "振动信号趋势项提取方法研究", 《电子学报》, vol. 45, no. 1, pages 22 - 28 * |
| 陈仁祥: "基于相关系数的EEMD转子振动信号降噪方法", 《振动、测试与诊断》, vol. 32, no. 4, pages 542 - 546 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111898447B (en) | Xin Jihe modal decomposition-based wind turbine generator fault feature extraction method | |
| Zhang et al. | Imbalanced fault diagnosis of rolling bearing using enhanced generative adversarial networks | |
| CN113610188B (en) | Bow net contact force non-section abnormality identification method and device | |
| CN101027565A (en) | Method and device for detecting electric arc phenomenon on at least one electric cable | |
| CN113177537B (en) | Fault diagnosis method and system for rotary mechanical equipment | |
| CN105534527B (en) | The recognition methods of intelligent wearable device special state and intelligent wearable device | |
| CN110909710A (en) | Self-adaptive main shaft performance degradation identification method based on S test piece | |
| CN113052000A (en) | Early weak fault signal characteristic diagnosis method for ship mechanical equipment | |
| CN113324646A (en) | Big wind area electrified railway contact net positive feeder line galloping positioning algorithm | |
| Hui et al. | A hybrid artificial neural network with Dempster-Shafer theory for automated bearing fault diagnosis | |
| CN110987433B (en) | Bearing fault early warning method based on high-frequency signal characteristic amplitude | |
| CN111461183A (en) | Flat wheel fault detection method for optimizing BP neural network by improving artificial fish swarm algorithm | |
| Song et al. | An improved structural health monitoring method utilizing sparse representation for acoustic emission signals in rails | |
| CN114066214A (en) | Power quality analysis method based on multi-fusion convolutional neural network | |
| CN110703013B (en) | Online identification method and device for low-frequency oscillation mode of power system and electronic equipment | |
| CN116304649A (en) | Motor fault signal feature extraction method, system, computer and storage medium | |
| CN113361427B (en) | Composite electric energy quality disturbance classification method based on WT and GA-PNN | |
| CN112104340B (en) | HMM model and Kalman filtering technology-based switching value input module BIT false alarm reduction method | |
| CN112083502B (en) | Magnetic different signal detection method based on parallel monostable stochastic resonance | |
| CN105447321B (en) | A kind of offline filtering method of reactivity meter electric current | |
| CN113688669A (en) | Hydroelectric generating set vibration signal energy characteristic identification method | |
| Chen et al. | Identification of typical partial discharge defects of distribution system equipment based on classification learner | |
| CN117828354A (en) | One-dimensional bearing fault signal data generation method integrating VAE and WGAN-GP | |
| CN117726436B (en) | Knowledge graph-based enterprise credit information link analysis method | |
| Chuangwen et al. | Fractal analysis of vibration signals for monitoring the condition of milling tool wear |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |