CN106599367A - Method for detecting abnormal state of spacecraft - Google Patents

Method for detecting abnormal state of spacecraft Download PDF

Info

Publication number
CN106599367A
CN106599367A CN201610997322.7A CN201610997322A CN106599367A CN 106599367 A CN106599367 A CN 106599367A CN 201610997322 A CN201610997322 A CN 201610997322A CN 106599367 A CN106599367 A CN 106599367A
Authority
CN
China
Prior art keywords
spacecraft
quick
association relation
model
abnormal state
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
Application number
CN201610997322.7A
Other languages
Chinese (zh)
Inventor
杨天社
赵静
王徐华
赵宜康
高宇
吴冠
王小乐
高波
刘兴淼
张海龙
邢楠
杨旭
张蔚
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.)
China Xian Satellite Control Center
Original Assignee
China Xian Satellite Control Center
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 China Xian Satellite Control Center filed Critical China Xian Satellite Control Center
Priority to CN201610997322.7A priority Critical patent/CN106599367A/en
Publication of CN106599367A publication Critical patent/CN106599367A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/30—Circuit design
    • G06F30/36—Circuit design at the analogue level
    • G06F30/367—Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/15—Vehicle, aircraft or watercraft design

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)

Abstract

The invention provides a method for detecting the abnormal state of a spacecraft, used for solving the technical problem that the early abnormity of the spacecraft is difficult to find in the prior art. The method for detecting the abnormal state of a spacecraft comprises the following steps: firstly, establishing a correlation model in the normal state by means of the satellite telemetry parameters; then, performing statistical analysis on the parameters of the correlation model in the normal state, obtaining the statistical characteristics and the change range of the parameters of the correlation model, and confirming the validity of the model through judging whether the values of the parameters of the correlation model are in the range of prediction; finally, inserting the data to be detected into the correct correlation model and realizing the abnormal state detection of the satellite by comparing the change rule of the detection data with the change rule of the normal state data. The abnormal state detection process provided by the invention is simple and clear, and can provide a powerful method support for the engineering application of abnormal state detection of the spacecraft.

Description

A kind of spacecraft abnormal state detection method
Technical field
The invention belongs to field of aerospace measurement and control, and in particular to a kind of side for the detection of spacecraft abnormal state Method.
Background technology
The abnormality detection technology of satellite be with satellite aerial mission extension, control accuracy improve and working life prolong Length etc. is required and grown up.The external spacefaring nation with the U.S. as representative is relatively early in the field starting, technically also leads It is first domestic a lot.IMS algorithms are a kind of mutation detection methods analyzed based on data association, are defined automatically using health data Incidence relation between systematic parameter, by same clan's relative analyses of measured data vector and knowledge base, reasoning, the row of prognoses system For whether normal.But IMS algorithms require to set up detailed knowledge base whether can just complete the behavior to prognoses system normal Detection.
The content of the invention
In order to overcome the deficiencies in the prior art, the present invention to provide a kind of method of Spacecraft anomaly state-detection, by closing Join foundation, the estimation of model parameter, the statistical characteristic analysis of association relation model and the normal condition model parameter of relational model Compare four steps with realistic model parameter, from spacecraft measured data, historical information Spacecraft anomaly state-detection is obtained As a result.
The technical solution adopted for the present invention to solve the technical problems is comprised the following steps：
Step one, by sun sensor output too quick output is denoted as, and sun sensor temperature is denoted as into too quick temperature, is set up Both normal condition association relation modelsWherein, y be too quick output, x1For too quick temperature, e is The error of association relation model；
Step 2, the too quick output of measurement setting period and too quick temperature, to sum of deviation squareAsk Lead and make it to be 0, obtain estimates of parameters b0、b1、b2；
Step 3：Select the numerical characteristic to be analyzed, including averageThe M such as median M, quantilep, variances sigma2, standard Difference σ and extreme difference R；
Step 4, measures the too quick output y ' and too quick temperature x of another period1', obtain association relation model y'=b0'+ b1'x1'+b2'(x1')2；When association relation model y'=b0'+b1'x1'+b2'(x1')2WithIn to divide The relative error of the numerical characteristic of analysis is when within 10%, it is believed that normal condition association relation model is correct, into step Five；Otherwise return to step one；
Step 5：Spacecraft measured data is substituted into into the association relation model set up of step one, by the curve chart for obtaining with The curve chart of normal condition association relation model is compared, when two curve chart Changing Patterns it is identical, and in measured data too The variance and average of quick output are less than 10% with the relative error of the variance of normal condition association relation model and average, then Judge spacecraft normal work；Otherwise it is assumed that spacecraft is abnormal state.
The invention has the beneficial effects as follows：Detect that spacecraft is likely to occur using the incidence relation between spacecraft parameter different Often situation, of the invention to set up association relation model between spacecraft two parameter according to the telemetry of spacecraft, calculates association and closes It is the numerical characteristic of model parameter, the correctness for setting up association relation model is verified by curve chart, finally, by what is finds Correlation properties between parameter are judging whether spacecraft occurs mutation.The abnormality detecting process of the present invention is short and sweet, can be The through engineering approaches application of Spacecraft anomaly state-detection provides powerful method and supports.
Description of the drawings
Fig. 1 is method of the present invention flow chart；
Fig. 2 is sun sensor temperature and the schematic diagram of sun sensor output in embodiment, wherein, it is (a) that the sun is quick Sensor temperature, (b) is sun sensor output；
Fig. 3 is sun sensor temperature and sun sensor output incidence relation figure；
Fig. 4 is real data and emulation data comparison schematic diagram.
Specific embodiment
With reference to the accompanying drawings and examples the present invention is further described, and the present invention includes but are not limited to following enforcements Example.
Technical scheme is specific as follows：
Step one：Either numeral or simulated solar sensor are can be seen that from sun sensor operation principle, substantially On be all circuit kit system.And the judgment criterion of its measured value, substantially change discontinuous.It is therefore possible to use statistics Analysis method, carries out the modeling of association in time relation.
Sun sensor to meeting relation exports (hereinafter referred to as too quick output) and sun sensor temperature is (hereinafter referred to as Too quick temperature) two telemetry parameters, set up the association relation model of normal condition：
Wherein, y be too quick output, x1For too quick temperature, e is the error of association relation model；
Step 2：The calculating of association relation model parameter, using least square method to estimate association relation model in Parameters.Its principle is to make sum of deviation square reach minimum, i.e.,
Reach minimum.To above formula (2) derivation and make it be 0, obtain final product estimates of parameters b0,b1,b2。
Step 3：The quick output of the present embodiment ether and the average of too quick temperatureAs analysis object, but as the present invention Extension, it is also possible to select different numerical characteristics to be analyzed, including：(1) locative characteristic parameter, locative number Word feature has averageThe M such as median M, quantilep；(2) statistical nature of dispersibility is represented, the parameter of data dispersibility is represented There is variances sigma2, standard deviation sigma, extreme difference R.
Step 4：The correctness of the association relation model that checking is set up, it is ensured that the foundation of this incidence relation is in satellite Present in the course of work, rather than the special case for sometime showing, need to be tested with the data of another time period Card, by this partial data formula (1) is substituted into, and is obtained
Y'=b0'+b1'x1'+b2'(x1')2…………………………………………(3)
For the association relation model that formula (1) and formula (3) are set up, the numeral for selecting the needs listed in step 3 to analyze is special Levy and be compared, as the digital spy of the association relation model of numerical characteristic and formula (1) foundation of the association relation model that formula (3) is set up The relative error levied is when within 10%, then it is assumed that it is correct that model is set up, and meets the incidence relation between parameter；Work as formula (3) relative error of the association relation model numerical characteristic that the numerical characteristic of the incidence relation set up is set up with formula (1) exceedes When 10%, then it is assumed that association relation model sets up the incidence relation not met between parameter, and it is incorrect that model is set up, and needs to return Return step one to model again.
Step 5：Test data is substituted into the association relation model set up, in formula (1), the incidence relation between two parameter Can be represented by curve chart, the Changing Pattern of the curve chart of model is different after actual curve chart is verified from step 4 When, according to the analysis to sun sensor operation principle, it is believed that in curve chart the variance and average of the too quick output of test data with The variance of normal condition association relation model and the relative error of average judge that spacecraft has asking for abnormal state more than 10%, then Topic；Otherwise it is assumed that spacecraft normal work.
By taking sun sensor association relation model as an example, this method is described further：
Need the telemetry first to obtaining to be chosen and divided before this method is carried out, select spacecraft normally to transport Sun sensor output and two parameters of sun sensor temperature during row, and three parts are splitted data into, telemetry 1 (is Totally 9000), telemetry 2 (10000) and test data (10000) data of three days are daily 3000,.
(Fig. 2) is as can be seen that too from the output of simulated solar sensor and simulated solar sensor temperature-time sequence The first two time period of positive area of illumination, i.e. data, two data have very strong negative correlation.Latter two observing and controlling time period, by In the visual field relation of simulated solar sensor, not in its visual field, output is approximately zero to the sun, and temperature is limited without this, So non-solar area of illumination, the relation is false.
Step 1-1, the sun sensor to meeting relation exports (hereinafter referred to as too quick output) and sun sensor temperature (hereinafter referred to as too quick temperature) two telemetry parameters, set up the correlation model of normal condition,
Wherein, S1For too quick output, T1For too quick temperature；Too quick output and the incidence relation figure of too quick temperature, such as Fig. 3, x Axle is too quick temperature, and y-axis is too quick output.
Step 1-2：By normal data 1 in units of day, average Sp1,Sp2,Sp3, Tp1,Tp2,Tp3, Sp1,Sp2,Sp3Point Wei not too quick output meansigma methodss of three days under normal condition；Tp1,Tp2,Tp3Too quick temperature three days is flat respectively under normal condition Average.
Step 1-3：By the calculated S of step 1-2p1,Sp2,Sp3,、Tp1,Tp2,Tp3, substitute into the incidence relation of step 1-1 3 formulas are obtained in model：
Calculating association relation model parameter is：B (0)=0.35613, b (1)=- 0.0204, b (2)=0.00029.
Step 1-4：Normal condition data 2 are brought in association relation model formula (4), corresponding too quick output valve is calculated Si, the curve of too quick output and too quick temperature is obtained, it is too quick defeated in telemetry as shown in the curve of square expression in Fig. 3 Go out shown in the curve that star represents in the relation such as Fig. 3 with too quick temperature, obtain its absolute error：0.0013, relative error is 6%, meet prediction and require, it can be seen that the dependency is set up, and obtains the curve chart of too quick temperature and Tai Min output relations.
Step 1-5：Test data is substituted in the association relation model of step 1-4, the too quick temperature of test data is obtained Curve chart such as Fig. 4 of too quick output, compares this curve chart and the curve chart in step 1-4, as can be seen from the figure step 1-5 Obtain curve chart to meet with step 1-4 curve characteristic, Changing Pattern is identical；The variance and average of two groups of data in calculated curve figure Obtain, normal condition data are 6% with the relative error of the variance of test data, the relative error of average is 5%, meets prediction Requirement, illustrate that spacecraft does not occur mutation.

Claims (1)

1. a kind of spacecraft abnormal state detection method, it is characterised in that comprise the steps：
Step one, by sun sensor output too quick output is denoted as, and sun sensor temperature is denoted as into too quick temperature, sets up both Normal condition association relation modelWherein, y be too quick output, x1For too quick temperature, e is association The error of relational model；
Step 2, the too quick output of measurement setting period and too quick temperature, to sum of deviation square Derivation simultaneously makes it be 0, obtains estimates of parameters b0、b1、b2；
Step 3：Select the numerical characteristic to be analyzed, including averageThe M such as median M, quantilep, variances sigma2, standard deviation sigma and Extreme difference R；
Step 4, measures the too quick output y ' and too quick temperature x of another period1', obtain association relation model y'=b0'+b1'x1' +b2'(x′1)2；When association relation model y'=b0'+b1'x1'+b2'(x′1)2WithIn to be analyzed The relative error of numerical characteristic is when within 10%, it is believed that normal condition association relation model is correct, into step 5；It is no Then return to step one；
Step 5：Spacecraft measured data is substituted into into the association relation model set up of step one, by the curve chart for obtaining with it is normal The curve chart of state relation relational model is compared, when two curve chart Changing Patterns it is identical and too quick defeated in measured data The variance for going out and average are less than 10% with the variance of normal condition association relation model and the relative error of average, then judge Spacecraft normal work；Otherwise it is assumed that spacecraft is abnormal state.
CN201610997322.7A 2016-11-14 2016-11-14 Method for detecting abnormal state of spacecraft Pending CN106599367A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610997322.7A CN106599367A (en) 2016-11-14 2016-11-14 Method for detecting abnormal state of spacecraft

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610997322.7A CN106599367A (en) 2016-11-14 2016-11-14 Method for detecting abnormal state of spacecraft

Publications (1)

Publication Number Publication Date
CN106599367A true CN106599367A (en) 2017-04-26

Family

ID=58590092

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610997322.7A Pending CN106599367A (en) 2016-11-14 2016-11-14 Method for detecting abnormal state of spacecraft

Country Status (1)

Country Link
CN (1) CN106599367A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107644148A (en) * 2017-09-19 2018-01-30 中国人民解放军国防科技大学 On-orbit satellite abnormal state monitoring method and system based on multi-parameter association
CN109506676A (en) * 2018-11-16 2019-03-22 中国西安卫星测控中心 Earth sensor method for diagnosing faults based on regression modeling
CN110348132A (en) * 2019-07-15 2019-10-18 长光卫星技术有限公司 Rail control effect fast evaluation method based on Bayes estimation
CN111966517A (en) * 2020-07-20 2020-11-20 北京控制工程研究所 On-orbit autonomous anomaly detection method for hierarchical spacecraft control system
CN112526560A (en) * 2020-12-03 2021-03-19 北京航空航天大学 Satellite key subsystem health state monitoring method based on relevance health baseline
CN112526559A (en) * 2020-12-03 2021-03-19 北京航空航天大学 System relevance state monitoring method under multi-working-condition

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101718864A (en) * 2009-11-13 2010-06-02 航天东方红卫星有限公司 System for real-time monitoring abnormal change of satellite telemetry parameters
CN104899456A (en) * 2015-06-15 2015-09-09 西北工业大学 Real-time correction method for on-orbit spacecraft model
US20160205117A1 (en) * 2015-01-13 2016-07-14 GM Global Technology Operations LLC Method and system for reflectometry based communication network monitoring, intrusion detection, and message authentication
CN106021845A (en) * 2016-04-18 2016-10-12 中国西安卫星测控中心 Spacecraft health condition assessment method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101718864A (en) * 2009-11-13 2010-06-02 航天东方红卫星有限公司 System for real-time monitoring abnormal change of satellite telemetry parameters
US20160205117A1 (en) * 2015-01-13 2016-07-14 GM Global Technology Operations LLC Method and system for reflectometry based communication network monitoring, intrusion detection, and message authentication
CN104899456A (en) * 2015-06-15 2015-09-09 西北工业大学 Real-time correction method for on-orbit spacecraft model
CN106021845A (en) * 2016-04-18 2016-10-12 中国西安卫星测控中心 Spacecraft health condition assessment method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
RYOHEI FUJIMAKI等: "An Anomaly Detection Method for Spacecraft Using Relevance Vector Learning", 《PAKDD》 *
吉林省农业学校: "《农业技术经济学》", 31 December 1988, 农业出版社 *
彭喜元: "航天器遥测数据异常检测综述", 《仪器仪表学报》 *
朱祥玲等: "一种基于多模型的航天试验效果实时评估方法", 《测控技术》 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107644148A (en) * 2017-09-19 2018-01-30 中国人民解放军国防科技大学 On-orbit satellite abnormal state monitoring method and system based on multi-parameter association
CN107644148B (en) * 2017-09-19 2020-09-22 中国人民解放军国防科技大学 On-orbit satellite abnormal state monitoring method and system based on multi-parameter association
CN109506676A (en) * 2018-11-16 2019-03-22 中国西安卫星测控中心 Earth sensor method for diagnosing faults based on regression modeling
CN110348132A (en) * 2019-07-15 2019-10-18 长光卫星技术有限公司 Rail control effect fast evaluation method based on Bayes estimation
CN110348132B (en) * 2019-07-15 2020-07-17 长光卫星技术有限公司 Rail control effect rapid evaluation method based on Bayes estimation
CN111966517A (en) * 2020-07-20 2020-11-20 北京控制工程研究所 On-orbit autonomous anomaly detection method for hierarchical spacecraft control system
CN111966517B (en) * 2020-07-20 2021-07-09 北京控制工程研究所 On-orbit autonomous anomaly detection method for hierarchical spacecraft control system
CN112526560A (en) * 2020-12-03 2021-03-19 北京航空航天大学 Satellite key subsystem health state monitoring method based on relevance health baseline
CN112526559A (en) * 2020-12-03 2021-03-19 北京航空航天大学 System relevance state monitoring method under multi-working-condition

Similar Documents

Publication Publication Date Title
CN106599367A (en) Method for detecting abnormal state of spacecraft
CN107357275B (en) Non-gaussian industrial process fault detection method and system
Liu et al. Robust modeling method for thermal error of CNC machine tools based on ridge regression algorithm
CN106951695B (en) Method and system for calculating residual service life of mechanical equipment under multiple working conditions
CN108960303B (en) Unmanned aerial vehicle flight data anomaly detection method based on LSTM
CN104483962B (en) The online fault detection method of batch process based on multidirectional orthogonal locality preserving projections
CN109165504B (en) Power system false data attack identification method based on anti-generation network
WO2016155241A1 (en) Method, system and computer device for capacity prediction based on kalman filter
CN112508105B (en) Fault detection and retrieval method for oil extraction machine
US11003738B2 (en) Dynamically non-gaussian anomaly identification method for structural monitoring data
CN109709934B (en) Fault diagnosis redundancy design method for flight control system
CN108829878B (en) Method and device for detecting abnormal points of industrial experimental data
CN111612048B (en) Unsupervised clustering anomaly detection method
CN104035431A (en) Obtaining method and system for kernel function parameters applied to nonlinear process monitoring
WO2021114320A1 (en) Wastewater treatment process fault monitoring method using oica-rnn fusion model
CN112882954A (en) Distributed database operation and maintenance dynamic threshold value warning method and device
CN111738348B (en) Power data anomaly detection method and device
CN110874685A (en) Intelligent electric energy meter running state distinguishing method and system based on neural network
CN111796233A (en) Method for evaluating secondary errors of multiple voltage transformers in double-bus connection mode
CN113298162A (en) Bridge health monitoring method and system based on K-means algorithm
CN110119579B (en) OICA-based complex industrial process fault monitoring method
CN110850358A (en) Electric energy meter comprehensive verification method and system based on stepwise regression algorithm
CN111091248A (en) Ship path planning method and system
US11676055B2 (en) System for detecting data drift in machine-learning process monitoring
CN106874531B (en) Method for automatically recovering abnormal measurement value data of atmospheric data system in case of failure

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
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20170426