CN104239698B - Solid propellant rocket vibrates the time series modification method of distorted signal - Google Patents

Solid propellant rocket vibrates the time series modification method of distorted signal Download PDF

Info

Publication number
CN104239698B
CN104239698B CN201410441024.0A CN201410441024A CN104239698B CN 104239698 B CN104239698 B CN 104239698B CN 201410441024 A CN201410441024 A CN 201410441024A CN 104239698 B CN104239698 B CN 104239698B
Authority
CN
China
Prior art keywords
vibration signal
vibration
time series
signal
modification method
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.)
Active
Application number
CN201410441024.0A
Other languages
Chinese (zh)
Other versions
CN104239698A (en
Inventor
卫莹
刘畅
夏虹
李蔚
贺晓芳
李瑛�
南林
陈涛
王忠颐
苏静
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
No401 Institute Fourth Research Academy Of China Aerospace Science And Technology Corp
Original Assignee
No401 Institute Fourth Research Academy Of China Aerospace Science And Technology Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by No401 Institute Fourth Research Academy Of China Aerospace Science And Technology Corp filed Critical No401 Institute Fourth Research Academy Of China Aerospace Science And Technology Corp
Priority to CN201410441024.0A priority Critical patent/CN104239698B/en
Publication of CN104239698A publication Critical patent/CN104239698A/en
Application granted granted Critical
Publication of CN104239698B publication Critical patent/CN104239698B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
  • Testing Of Engines (AREA)

Abstract

The present invention proposes a kind of time series modification method of solid propellant rocket vibration distorted signal, based on the autoregression model in time series, the coefficient of autoregression model is obtained by the method for Maximum-likelihood estimation, and then the vibration data point value at catastrophe point is obtained, and the multiple break value occurred on one section of longer period can be handled.The present invention solves the theoretical foundation weakness under the conditions of prior art, error is larger, use the problems such as dumb, with easy to operate, theoretical property is strong, the advantages that meeting actual data characteristic, available for Multiple Type engine vibration signal breakpoint reparation, effect is preferable, meets rocket engine ground test demand.

Description

Solid propellant rocket vibrates the time series modification method of distorted signal
Technical field
The invention belongs to Solid Rocket Engine Test observation and control technology field, and relating generally to one kind can send out solid-rocket Distortion analysis of vibration signal and modification method in motivation experiment.
Background technology
Influenceed because solid propellant rocket is vulnerable to the factors such as impact, temperature, malformation during experiment, experiment The vibration signal measured has a nonlinear and nonstationary feature, and in practical implementation, vibration signal is often regarded as steady random Signal.Vibration is one of reflection engine structure design parameter.During vibration signals collecting, easily by sensor, installation, The factors such as 50Hz Hz noises, amplifier and acquisition system interference influence, and cause vibration signal to distort, such as signal kick, zero The distortion situation such as line drift, trend.Therefore, in data analysis and process, invalid signals is rejected, distortion data is repaiied Just.
In rocket engine ground test, there is obvious Kick Abnormality situation sometimes in vibration signal, as shown in figure 1, existing Reparation be to take alternative, next section of value, i.e. x are substituted with the data value of the last periodi=xi+h, wherein, h is step-length, xi For i-th section of vibration signal value, xi+hFor the vibration signal value of the i-th+h sections.The method is only with certain section of normal data in the past to disconnected Signal value simple substitute at point, error are larger.Therefore, the breakpoint problem based on foregoing description, application time sequence are divided herein The method of analysis carries out repairing analysis.
Time series is variable order at timed intervals and the sequence of random variables that is formed, in recent years, time series point The application of analysis have penetrated into communications and transportation, intelligent control, neuron network simulation, biology, medical science, the hydrology, meteorology, economics, The natural sciences such as space science are among social field, just playing unrivaled significant role.Time series analysis early stage Research is divided into frequency domain (Frequency Domain) analysis method and time domain (Time Domian) analysis method.But frequency domain point It is typically complex to analyse procedure, is unfavorable for visual interpretation, there is larger limitation, so Time Domain Analysis is typically used, Time Domain Analysis is the sample autocorrelation function of analysis time sequence, and establishes parameter model, and then describes the dynamic of sequence Dependence relation.Britain statistician of visible nineteen twenty-seven Yule proposes autoregression (Autoregressive) to Time Domain Analysis earliest Model.Then, British mathematician, astronomer Walker introduce rolling average (Moving when analyzing India's air rule Average) model and auto regressive moving average (Autoregressive Moving Average) model.These models are established The basis of time series analysis Time Domain Analysis.In actual applications, the parameter model under linear normal state assumes is filled Decomposition is determined, and Nonlinear Time Series Analysis also attains full development, and Tong is proposed using the thought of piece-wise linearization tectonic model Threshold autoregressive model, has started the beginning of Nonlinear Time Series Analysis.
The content of the invention
Although existing breakpoint reparation can meet Solid Rocket Engine Test requirement, it is excessively simple to there is principle Singly, method is not reasonable, error larger grade shortcomings.The present invention proposes a kind of vibration number based on time series analysis According to breakpoint restorative procedure, handle vibration data using this method, can effectively repair kick point, have it is certain it is theoretical according to According to, and ensure error within the specific limits.
The technical scheme is that：
A kind of time series modification method of the solid propellant rocket vibration distorted signal, it is characterised in that：Including Following steps：
Step 1：Caused vibration signal when gathering Solid Rocket Engine Test, vibration signal meet autoregression mould Type；Exclude in vibration signal after Characteristics of Mutation caused by engine itself acts, if in the presence ofThen judge xtFor Kick point, wherein xt,xt-1The effective point value of the vibration signal at t and t-1 moment respectively in vibration signal；
Step 2：Kick point in vibration signal is repaired：
It is X to take the value repaired at processus aboralis hopt=α1Xt-1+α2Xt-2+…αpXt-p+εt, wherein Xt-1,Xt-2,…,Xt-pFor step Rapid 1 vibration signal collected is in t-1, t-2 ..., the point value at t-p moment, and p is exponent number, α1,α2,…αpFor autoregressive coefficient, Random disturbances { εtIt is white noise sequence, εtBetween it is independent mutually；α1,α2,…αpObtained using Maximum Likelihood Estimation.
Beneficial effect
The present invention, which has the special feature that, compared with prior art is：(1) there is the theory for preferably meeting vibration data characteristic Foundation；(2) it is easy to operate；(3) breakpoint on longer period can disposably be handled；(4) problem investigation is simple.
Brief description of the drawings
Fig. 1：The invention solves vibration signal kick point situation.
Fig. 2：The present invention chooses the data for including kick point at 7.3s -8.3s.
Fig. 3：7.3s -8.3s signal graph after present invention amendment.
Embodiment
Below by specific embodiment and with reference to accompanying drawing, the present invention is described in further detail：
Autoregression model in the catastrophe point restorative procedure usage time sequence analysis that the present embodiment is related to.
Step 1：Caused vibration signal when gathering Solid Rocket Engine Test.Vibration in rocket engine ground test Its statistical nature of signal does not change with time typically and changed, and value of the value at a time with preceding several moment It is relevant, meet autoregression model (Auto Regressive model).
Exclude in vibration signal after Characteristics of Mutation caused by engine itself acts, if in the presence ofThen judge xtFor kick point, wherein xt,xt-1The effective point value of the vibration signal at t and t-1 moment respectively in vibration signal.Fig. 1 show certain Caused vibration signal time history during rocket engine ground test, after judgement, there is kick point at three, be interference signal, Need to repair.
Step 2：Kick point in vibration signal is repaired：
For the point of vibration signal t, the value repaired at processus aboralis hop is taken to be
Xt=α1Xt-1+α2Xt-2+…αpXt-p+εt
E[εsεt]=0, s ＞ t
Wherein Xt-1,Xt-2,…,Xt-pIt is the vibration signal that step 1 collects in t-1, t-2 ..., the point value at t-p moment, p For exponent number, α1,α2,…αpFor autoregressive coefficient, random disturbances { εtIt is white noise sequence, εtBetween it is independent mutually.
P can be completed by the method for pattern-recognition, i.e., if the PARCOR coefficients (PACF) of vibration data are cut at p Tail, then we are it is determined that the sequence is AR (p) sequences, wherein factor alphaiAsked by the method for Maximum-likelihood estimation Solution, detailed process are as follows：
It is assumed that disturbance εn~N (C, δ2), and meet independent same distribution, so joint density
f(x1,x2…xn)=f1(x1)f2(x2)…fn(xn)
If (ε0, ε1... εn)TIt is an overall ε sample, is designated as θ=(C, δ2)T, then
That is L (θ) is (ε0, ε1... εn)TJoint density.
If (ε0, ε1... εn)TIt is unknown, (x-1,x-2…x-p),(x0,x1…xm) known to when, have
Remember θ=(C, δ2,α1,…αp)T, then its joint density function be
Then likelihood equation is
It is more than simultaneous that (p+1) individual equation, is designated as AX=b, wherein
X=(C, α1,α2…αp)T
Then X=A-1B, after solution, X=(C, α1,α2…αp)TAs its maximum likelihood estimation.
After Maximum Likelihood Estimation obtains autoregressive coefficient above, the point value X after repairing is drawnt, vibration signal As shown in figure 3, we can see that kick point is repaired from Fig. 3, effect is preferable.

Claims (1)

  1. A kind of 1. time series modification method of solid propellant rocket vibration distorted signal, it is characterised in that：Including following step Suddenly：
    Step 1：Caused vibration signal when gathering Solid Rocket Engine Test, vibration signal meet autoregression model；Row Except in vibration signal caused by engine itself acts after Characteristics of Mutation, if in the presence ofThen judge xtFor kick point, Wherein xt,xt-1The effective point value of the vibration signal at t and t-1 moment respectively in vibration signal；
    Step 2：Kick point in vibration signal is repaired：
    It is X to take the value repaired at processus aboralis hopt=α1Xt-1+α2Xt-2+αpXt-p+εt, wherein Xt-1,Xt-2,…,Xt-pAdopted for step 1 The vibration signal collected is in t-1, t-2 ..., the point value at t-p moment, and p is exponent number, α1,α2,…αpIt is random dry for autoregressive coefficient Disturb { εtIt is white noise sequence, εtBetween it is independent mutually；α1,α2,…αpObtained using following Maximum Likelihood Estimation：
    It is assumed that disturbance εm~N (C, δ2), and meet independent same distribution, so joint density
    f(x1,x2xm)=f1(x1)f2(x2)fm(xm)
    If (ε0,ε1,εm)TIt is an overall ε sample, is designated as θ=(C, δ2)T, then
    That is L (θ) is (ε0,ε1,εm)TJoint density；
    If (ε0,ε1,εm)TIt is unknown, (x-1,x-2…x-p),(x0,x1…xm) known to when, have
    Remember θ=(C, δ2,α1,αp)T, then its joint density function be
    Then likelihood equation is
    It is more than simultaneous that (p+1) individual equation, is designated as AX=b, wherein
    X=(C, α1,α2αp)T
    Then X=A-1B, after solution, X=(C, α1,α2αp)TAs its maximum likelihood estimation.
CN201410441024.0A 2014-09-01 2014-09-01 Solid propellant rocket vibrates the time series modification method of distorted signal Active CN104239698B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410441024.0A CN104239698B (en) 2014-09-01 2014-09-01 Solid propellant rocket vibrates the time series modification method of distorted signal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410441024.0A CN104239698B (en) 2014-09-01 2014-09-01 Solid propellant rocket vibrates the time series modification method of distorted signal

Publications (2)

Publication Number Publication Date
CN104239698A CN104239698A (en) 2014-12-24
CN104239698B true CN104239698B (en) 2018-03-16

Family

ID=52227748

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410441024.0A Active CN104239698B (en) 2014-09-01 2014-09-01 Solid propellant rocket vibrates the time series modification method of distorted signal

Country Status (1)

Country Link
CN (1) CN104239698B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109779791B (en) * 2019-03-24 2021-01-01 西安航天动力测控技术研究所 Intelligent diagnosis method for abnormal data in solid rocket engine

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1593831A2 (en) * 2004-04-15 2005-11-09 United Technologies Corporation Reduced gain thrust control valve
CN103116705A (en) * 2013-02-06 2013-05-22 中国航天科技集团公司第六研究院第十一研究所 Fault simulated analysis method for afterburning cycle rocket engine
CN103454089A (en) * 2013-09-12 2013-12-18 中国航天科技集团公司第四研究院四0一所 Device for measuring class of discontinuous parameters of solid rocket engine

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1593831A2 (en) * 2004-04-15 2005-11-09 United Technologies Corporation Reduced gain thrust control valve
CN103116705A (en) * 2013-02-06 2013-05-22 中国航天科技集团公司第六研究院第十一研究所 Fault simulated analysis method for afterburning cycle rocket engine
CN103454089A (en) * 2013-09-12 2013-12-18 中国航天科技集团公司第四研究院四0一所 Device for measuring class of discontinuous parameters of solid rocket engine

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
时间序列法在发动机振动故障诊断中的应用;郭刚 等;《振动、测试与诊断》;19920630;第12卷(第2期);第58-62页 *
液体火箭发动机故障检测和诊断中数据挖掘策略的分析;胡小平 等;《国防科技大学学报》;20051231;第27卷(第3期);第1-5页 *

Also Published As

Publication number Publication date
CN104239698A (en) 2014-12-24

Similar Documents

Publication Publication Date Title
CN110261109B (en) Rolling bearing fault diagnosis method based on bidirectional memory cyclic neural network
CN106934126B (en) Mechanical part health index construction method based on recurrent neural network fusion
CN112418013B (en) Complex working condition bearing fault diagnosis method based on meta-learning under small sample
CN106053330A (en) Soil compactness and moisture composite measuring method and device
CN103989462B (en) The extracting method of a kind of pulse wave fisrt feature point and second feature point
CN110824586B (en) Rainfall prediction method based on improved decision tree algorithm
CN110285909B (en) Method for calculating instantaneous cable force of cable-supported bridge based on synchronous compression transformation
CN107367552B (en) Based on the damage propatagtion method of real-time of FCM algorithm under the influence of time varying temperature
CN102680860A (en) Automatic fault-point locating method for traveling-wave based fault location of high-voltage electric power lines
CN110866448A (en) Flutter signal analysis method based on convolutional neural network and short-time Fourier transform
CN107886078A (en) A kind of Threshold Denoising method based on layered self-adapting threshold function table
CN104807534A (en) Equipment natural vibration mode self-learning recognition method based on online vibration data
CN107657110A (en) The fatigue damage evaluation method of large scale wind power machine blade
CN104881585A (en) Flutter boundary prediction method of two-degree-of-freedom wing
CN117332341B (en) Intelligent management method and system for airplane wheel guard data based on Internet of things
CN112418682B (en) Safety evaluation method for fusion of multi-source information
CN108549847B (en) Beam type structure crack damage identification method under non-reference data condition
CN111709350B (en) Low-frequency oscillation modal parameter identification method and system based on FCM clustering
CN104156775B (en) Meteorological disaster Forecasting Methodology based on arithmetic of linearity regression
CN107491579B (en) Joint probability calculation method and system for icing thickness and wind speed of power transmission line
CN104239698B (en) Solid propellant rocket vibrates the time series modification method of distorted signal
CN109815940B (en) Wavelet packet energy spectrum damage identification method
CN107392979B (en) The two dimensional visible state composition and quantitative analysis index method of time series
CN113128047A (en) Aero-engine thrust calculation method, storage medium and calculation device
CN116089865A (en) Method for classifying and identifying OPGW abnormal vibration based on improved grid search random forest algorithm

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
CB03 Change of inventor or designer information
CB03 Change of inventor or designer information

Inventor after: Wei Ying

Inventor after: Su Jing

Inventor after: Liu Chang

Inventor after: Xia Hong

Inventor after: Li Wei

Inventor after: He Xiaofang

Inventor after: Li Ying

Inventor after: Nan Lin

Inventor after: Chen Tao

Inventor after: Wang Zhongyi

Inventor before: Wei Ying

Inventor before: Liu Chang

Inventor before: Li Wei

Inventor before: Wang Zhongyi

Inventor before: He Xiaofang

Inventor before: Xia Hong

Inventor before: Su Jing

Inventor before: Nan Lin

Inventor before: Chen Tao

GR01 Patent grant
GR01 Patent grant