CN103335814B - Correction method for inclination angle measurement error data of experimental model in wind tunnel - Google Patents
Correction method for inclination angle measurement error data of experimental model in wind tunnel Download PDFInfo
- Publication number
- CN103335814B CN103335814B CN201310230816.9A CN201310230816A CN103335814B CN 103335814 B CN103335814 B CN 103335814B CN 201310230816 A CN201310230816 A CN 201310230816A CN 103335814 B CN103335814 B CN 103335814B
- Authority
- CN
- China
- Prior art keywords
- data
- model
- training
- correction
- curve
- 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.)
- Expired - Fee Related
Links
Landscapes
- Aerodynamic Tests, Hydrodynamic Tests, Wind Tunnels, And Water Tanks (AREA)
Abstract
The invention discloses an inclination angle measurement error data correction system and method of an experimental model in a wind tunnel when the experimental model is placed in the wind tunnel to be subjected to a blowing experiment. The inclination angle measurement data correction system comprises a data preprocessing module, a model training module and an error data correction module, and the correction method comprises three steps: data preprocessing, model training and error data correcting. The invention provides a multi-stage and multi-model fusion error correction method based on a neural network model as a basic model, the model fitting degree is higher, the calculation result is more accurate, the nonlinear error, caused by vibration of the model under the wind action, in data measured actually by an inclination angle sensor can be corrected well, and the correction error of the correction system can be controlled within 0.004-0.01 degree near the 0 degree.
Description
Technical field
The present invention relates to empirical model in a kind of wind-tunnel, empirical model measurement of dip angle error information modification method in wind-tunnel in blowing experiment process.
Background technology
At present, the measuring technique of supersonic speed and transonic wind tunnel model, in the production being employed for civilian goods more and more and test, to the requirement of measuring technology and precision also more and more higher.Because wind tunnel test process is a complicated aerodynamic experiment, test process link is many, in many ways the factor impact such as the technology grasp level of tested person instrument and test environment and tester, all brings impact and error to test result, measuring accuracy and accuracy.How to obtain test data accurately, reject the Accuracy that environment causes, this is most important to the measuring technique itself of supersonic speed and transonic wind tunnel.
Be fixed in wind tunnel model on support and have individual angle α with surface level, be referred to as inclination angle, the inner gravity sensor that can use a model is measured.Obliquity sensor is used for the actual angle of attack of measurement model after loading, is positioned at detection system foremost.Attack angle measurement value is one of basic data of other numerical operation of detection system, and its accuracy directly has influence on the accuracy of detection system test data.And in the wind tunnel test of reality, model can be subject to from external force impact all around, the moment loading that external force impact produces, on model, can make the gravity sensor at model inclination angle measure and produce error.Namely under the impact of various external force factor, there is error in the data measured by obliquity sensor.By analysis, this error mainly receives the oscillatory torque on different directions due to model, to such an extent as to the working environment of obliquity sensor is no longer stable, so the digital signal obtained by this sensor exists error, the angle of attack value of model can not be reflected really.Therefore, the raw digital signal obtained by obliquity sensor can not be directly used in other numerical evaluation, needs to carry out correction-compensation to this signal, to ensure that test data can truly reflect tested situation.
In order to reduce the margin of error, when bucking-out system modeling, usually adopting nonlinear mathematical model, and generally not adopting linear model.In nonlinear model system, conventional data error compensation method has successive approximation method, least square polynomial curve fitting method, adaptive filter method, error of fitting compensation formula method, cubic spline interpolation and the method etc. by neural network.
Due to model wind-engaging produce vibration produce complicacy and the certainty of moment, obliquity sensor data always exist because model vibrates the error produced, and the complicated function of the normally multiple variable of this error or model.If go to try to achieve this function by conventional Mathematical Fitting method, the key issue that so will solve is except carrying out rational mathematical description to possible variable, more to the variable existed in environment be excavated as much as possible, also need accurate mathematical model to describe the relation between these variablees in addition, to vibration error model of fit in wind-tunnel carry out people for explication be very difficult, and the extended capability of model also can not be guaranteed.In the recent period, data error correction technique field proposes to adopt nerual network technique to carry out the new technology of error compensation value matching.Because neural network does not need accurate mathematical model, non-linear mapping capability is good, is good at from the useful knowledge of inputoutput data learning, discloses data characteristics, process enchancement factor.The method of neural network is utilized to solve relation how between data of description environment and data error, by training fully neural network, utilize the ability of neural network self-organization Optimization Modeling, constantly carry out self-organization optimization with training data, thus form optimum and adaptable error compensation model, promote Accuracy Error data correction ability.
Summary of the invention
The object of the invention is to the performance improving existing Error Compensation Technology, there is provided a kind of and can effectively revise the nonlinearity erron caused due to vibration in obliquity sensor measured data, and the modification method of empirical model measurement of dip angle error information in the controlled wind-tunnel within 0 approximate angle 0.004 ~ 0.01 precision of inclination value round-off error.
The object of the invention is to be achieved through the following technical solutions: empirical model measurement of dip angle error information update the system in wind-tunnel, it comprises with lower module:
Data preprocessing module: pre-service is carried out to input data, extract vibrations error character vector, systematic features model training process and data correction process all comprise data preprocessing module, but pretreatment module completes different operating in VEC training process and data correction process;
Model training module: set up with angle, frequency and amplitude as input variable parameter according to the training data collected in advance and target data, error correction values is the joint error correction model exported, and this module comprises model eigenvectors and extracts submodule and model training submodule;
Data correction module: each section of input data obtained according to data preprocessing module in error correction process, extract the proper vector of each section of input data, and proper vector is inputted the joint error correction model that precondition obtains, obtain error correction values, modified value is added to original input, obtains revised data.
The modification method of empirical model measurement of dip angle error information in wind-tunnel, it comprises data prediction, model training and data correction three steps, model training and data correction all comprise data preprocessing module, but pretreatment module completes different operating in model training and data correction;
In described model training step, data prediction comprises following sub-step:
S101: loaded targets data and training data to update the system, and carry out down-sampling to target data curve and training data curve;
S102: carry out wavelet transformation and smoothing processing to the curve of target data curve and training data respectively, obtains their high and low frequency data;
S103: ask poor with the low-and high-frequency of all training data curves respectively to the low-and high-frequency data of target data curve, obtains the input (difference of high frequency) for training pattern and the conjunction of (difference of low frequency) output data set;
In described data correction step, data prediction comprises following sub-step:
S201: loaded targets data and each correction data to be repaired to system, and to aim curve and treat that fair curve carries out down-sampling;
S202: respectively to aim curve and treat that fair curve carries out wavelet transformation and smoothing processing, obtain their high and low frequency data;
S203: to the low-and high-frequency data of aim curve respectively with treat that the low-and high-frequency of fair curve asks poor, obtain input (difference of high frequency) data and the reference curve of VEC;
S204: the result according to end-point detection carries out segmentation to the input data obtained;
Described model training step comprises following sub-step:
S301: the input data obtained according to pre-service in model training step, extracts and revises feature;
S302: the output data obtained with pretreatment module in model training process are for output valve, the correction obtained in S301 is characterized as input value, composing training data acquisition, system is to revise every one dimension of proper vector value as an element variable, by the model training set that the training data subset division characterizing each element variable data variation in training data set is this element variable, train respectively with unit variable as input variable parameter, export N number of (N=element variable number) unit variation model that data are output valve, and be stored as intermediate result;
S303: to divide in each unit variance training data subset obtained point centered by each eigenwert by S302, within the specific limits to its interpolation, obtains the new characteristic value collection of the unit variable after inserting respectively;
S304: by characteristic value collection new for the unit variable that obtains in S303, the unit variation model bringing S302 acquisition respectively into obtains the output valve of each model, and by new unit characteristics of variables value set and each model output valve, together form the set of one dimension change training data, and store for follow-up model training;
S305: the one dimension change training data set utilizing S304 to obtain, based on Krging Algorithm for Surface Fitting, the M(M=that difference matching is three-dimensional coordinate with any two element variable values and corresponding model output valve
) individual space curved surface, in curved surface, each point is the vector be made up of 2 dimension element variable values and corresponding model output valve;
Each sampling K point on each curved surface in S306: the M generated in S305 each space curved surface, form the error information training set of KxM element, in set, each point is one or four dimensional vectors, with this error information training set training error correction model Q, and the system that is stored in is for error information makeover process;
Described data correction comprises following sub-step:
S401: extract the input data that in data correction step, data prediction obtains, and to each input segmentation, organize data into the long certain matrix of row;
S402: adopt same procedure to extract the eigenwert of every segment data with S301, and bring the eigenwert of each segment data into error correction values that the VEC trained obtains each segment data;
S403: the error correction values of each segment data is connected, forms complete error correction values, and carry out up-sampling and form final calibration corrections;
S404: final calibration corrections is added with the data of grandfather tape correction and obtains revised obliquity sensor data.
Element variable in described model training step is for comprising frequency h, amplitude f and angle a.
The invention has the advantages that:
1. build the multistage based on neural network model, multi-model merges error correcting method, constructed model has Fast Learning and can approach arbitrary internal model and internal model controller, thus well error correction is carried out to obliquity sensor data, make model-fitting degree higher, error compensation result is more accurate.
2., by the neural network correction model under the multiple vibration frequency of training and oscillation intensity, when inputting obliquity sensor data, based on multi-model and multistage data correction algorithm, error obliquity sensor data are revised; Can effectively revise the nonlinearity erron caused due to vibration in obliquity sensor measured data, and error of tilt modified value is controlled within 0 approximate angle 0.004 ~ 0.01 precision.
Accompanying drawing explanation
Fig. 1 is present system frame diagram;
Data prediction process flow diagram in Fig. 2 system model training process;
Data prediction process flow diagram in Fig. 3 system data makeover process;
Fig. 4 is model training method process flow diagram of the present invention;
Fig. 5 is the neural network structure schematic diagram of error information correction model;
Fig. 6 is data correction process flow diagram of the present invention.
Embodiment
Training dataset: in advance by hitting shake device, under different frequency, amplitude and inclination angle condition, empirical model being hit and encouraging vibrations, gather obtain under different frequency, various amplitude and different angle
individual obliquity sensor data.
Target data: the artificial free from error obliquity sensor data assert.
Further illustrate technical scheme of the present invention below in conjunction with accompanying drawing, but the content that the present invention protects is not limited to the following stated.
As shown in Figure 1, empirical model measurement of dip angle error information update the system in wind-tunnel, it comprises with lower module:
Data preprocessing module: pre-service is carried out to input data, extract vibrations error character vector, systematic features model training process and data correction process all comprise data preprocessing module, but pretreatment module completes different operating in VEC training process and data correction process;
Model training module: set up with angle, frequency and amplitude as input variable parameter according to the training data collected in advance and target data, error correction values is the joint error correction model exported, and this module comprises model eigenvectors and extracts submodule and model training submodule;
Data correction module: each section of input data obtained according to data preprocessing module in error correction process, extract the proper vector of each section of input data, and proper vector is inputted the joint error correction model that precondition obtains, obtain error correction values, modified value is added to original input, obtains revised data.
The modification method of empirical model measurement of dip angle error information in wind-tunnel, it comprises data prediction, model training and data correction three steps, model training and data correction all comprise data preprocessing module, but pretreatment module completes different operating in model training and data correction;
As shown in Figure 2, in described model training step, data prediction comprises following sub-step:
S101: loaded targets data and training data, to update the system, carry out target data curve and training data curve
times down-sampling, obtains the packed data of the target data after down-sampling and each training data;
S102: carry out continuous wavelet transform to each packed data respectively, then smoothing filtering process, obtain high and low frequency two parts frequency data of each packed data;
S103: the low-and high-frequency data of target data curve are asked poor with the low-and high-frequency of all training data curves respectively, obtains the input for training pattern
(difference of high frequency) and
(difference of low frequency) (
) output data set conjunction;
As shown in Figure 3, in data correction step, data prediction comprises following sub-step:
S201: loaded targets data and each correction data to be repaired to system, and to aim curve and treat that fair curve carries out
times down-sampling, obtains the packed data of the target data after down-sampling and correction data to be repaired;
S202: carry out continuous wavelet transform to the packed data of target data and correction data to be repaired respectively, then smoothing filtering process, obtain their high and low frequency two parts data;
S203: the low-and high-frequency curve of target data is asked poor with the low-and high-frequency of correction data to be repaired respectively, obtains the input (difference of high frequency) of VEC
data and reference curve.
S204: the result according to end-point detection will
be divided into
section, and every section is follow-uply compensated respectively;
As shown in Figure 4, correction model training step comprises following sub-step:
S301: the input data acquisition obtained according to pre-service in model training step
(
), extract and revise feature, concrete steps are as follows:
S3011: right
carry out Short Time Fourier Transform, obtain
the frequency data of each point;
S3012: feature extraction: build P 3 dimensional feature vectors
=
,
,
,
, P=
count.Wherein
for correspondence
of frequency data
point value,
for correspondence
of amplitude data
point value,
for of correspondence
the inclination value of individual training data;
S302: obtain with pre-service in model training step
(
) for exporting data, the correction feature obtained in S301
(
) be input value, composing training data acquisition, system is to revise proper vector value
every one dimension as an element variable, be the model training set of this element variable by the training data subset division characterizing each element variable data variation in training data set, namely
change training data subset
,
| belong to
change },
change training data subset
,
| belong to
change } and
change training data subset
,
| belong to
change }, train respectively with
for input variable parameter, output data are output valve
3 unit variation models
, and be stored as intermediate result;
S303: to divide in each unit variance training data subset obtained point centered by each eigenwert by S302, use following formula, within the specific limits to its interpolation, obtains the new characteristic value collection of the unit variable after inserting respectively:
=
=
=
S304: by characteristic value collection new for the unit variable that obtains in S303
bring the unit variation model that S302 obtains respectively into
obtain the output valve of each model
, and by new unit characteristics of variables value set and each model output valve, together formation one dimension change training data set
,
, and store for follow-up model training;
S305: the one dimension change training data set utilizing S304 to obtain, based on Krging Algorithm for Surface Fitting, under one dimension element variable situation is determined arbitrarily in matching respectively, other two element variable values and corresponding model output valve are three-dimensional coordinate, namely (
,
,
), (
,
,
) (
,
,
) 3 space curved surfaces, in curved surface, each point is the vector be made up of 2 dimension element variable values and corresponding model output valve;
S306: in 3 space curved surfaces generated in S305, each curved surface is respectively sampled
individual point (
), form
individual element error information training set, in set each point be one or four dimensional vectors
,
,
,
, with this error information training set, training as shown in Figure 5
,
,
be 3 yuan of inputs,
for the neural network VEC that unitary exports
, and the system that is stored in is for error information makeover process;
As shown in Figure 6, data correction comprises following sub-step:
S404: final calibration corrections is added with the data of grandfather tape correction and obtains revised obliquity sensor data.
S401: each segmentation of input data that data preprocessing module in data correction step is obtained carry out regularly for length be
matrix;
S402: adopt same procedure to extract the eigenwert of each point of every segment data with S301
(
) (
), and by eigenwert
bring the VEC trained into
obtain the error correction values of each point data;
S403: each point error correction values of each segment data is connected, forms complete error correction values, and carry out
times up-sampling forms final calibration corrections;
S404: final calibration corrections is added with the data of grandfather tape correction and obtains revised obliquity sensor data.
Element variable in described model training step is for comprising frequency h, amplitude f and angle a.
Claims (2)
1. the modification method of empirical model measurement of dip angle error information in wind-tunnel, it is characterized in that: it comprises data prediction, model training and data correction three steps, model training and data correction all comprise data preprocessing module, but pretreatment module completes different operating in model training and data correction;
In described model training step, data prediction comprises following sub-step:
S101: loaded targets data and training data to update the system, and carry out down-sampling to target data curve and training data curve;
S102: carry out wavelet transformation and smoothing processing to the curve of target data curve and training data respectively, obtains their high and low frequency data;
S103: the low-and high-frequency data of target data curve are asked poor with the low-and high-frequency of all training data curves respectively, obtains the input and output data acquisition for training pattern;
In described data correction step, data prediction comprises following sub-step:
S201: loaded targets data and each correction data to be repaired to system, and to aim curve and treat that fair curve carries out down-sampling;
S202: respectively to aim curve and treat that fair curve carries out wavelet transformation and smoothing processing, obtain their high and low frequency data;
S203: to the low-and high-frequency data of aim curve respectively with treat that the low-and high-frequency of fair curve asks poor, obtain input data and the reference curve of VEC;
S204: the result according to end-point detection carries out segmentation to the input data obtained;
Described model training step comprises following sub-step:
S301: the input data obtained according to pre-service in model training step, extracts and revises feature;
S302: the output data obtained with pretreatment module in model training process are for output valve, the correction obtained in S301 is characterized as input value, composing training data acquisition, system is to revise every one dimension of proper vector value as an element variable, by the model training set that the training data subset division characterizing each element variable data variation in training data set is this element variable, train respectively with unit variable as input variable parameter, export N number of unit variation model that data are output valve, and be stored as intermediate result, wherein, N=element variable number;
S303: divide in each unit variance training data subset obtained by S302 and put its interpolation centered by each eigenwert, obtains the new characteristic value collection of the unit variable after inserting respectively;
S304: unit variation model characteristic value collection new for the unit variable obtained in S303 being brought respectively into S302 acquisition obtains the output valve of each model, and by new unit characteristics of variables value set and each model output valve, together form the set of one dimension change training data, and store for follow-up model training;
S305: the one dimension change training data set utilizing S304 to obtain, based on Krging Algorithm for Surface Fitting, M the space curved surface that difference matching is three-dimensional coordinate with any two element variable values and corresponding model output valve, in curved surface, each point is the vector be made up of 2 dimension element variable values and corresponding model output valve, wherein, M=
;
S306: each sampling K point on each curved surface in M the space curved surface generated in S305, form the error information training set of KxM element, in set, each point is one or four dimensional vectors, with this error information training set training error correction model Q, and the system that is stored in is for error information makeover process;
Described data correction comprises following sub-step:
S401: extract the input data that in data correction step, data prediction obtains, and to each input segmentation, organize data into the long certain matrix of row;
S402: adopt same procedure to extract the eigenwert of every segment data with S301, and bring the eigenwert of each segment data into error correction values that the VEC trained obtains each segment data;
S403: the error correction values of each segment data is connected, forms complete error correction values, and carry out up-sampling and form final calibration corrections;
S404: final calibration corrections is added with the data of grandfather tape correction and obtains revised obliquity sensor data.
2. the modification method of empirical model measurement of dip angle error information in wind-tunnel according to claim 1, it is characterized in that, the element variable in described model training step is for comprising frequency h, amplitude f and angle a.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310230816.9A CN103335814B (en) | 2013-06-09 | 2013-06-09 | Correction method for inclination angle measurement error data of experimental model in wind tunnel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310230816.9A CN103335814B (en) | 2013-06-09 | 2013-06-09 | Correction method for inclination angle measurement error data of experimental model in wind tunnel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103335814A CN103335814A (en) | 2013-10-02 |
| CN103335814B true CN103335814B (en) | 2015-05-27 |
Family
ID=49244022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310230816.9A Expired - Fee Related CN103335814B (en) | 2013-06-09 | 2013-06-09 | Correction method for inclination angle measurement error data of experimental model in wind tunnel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103335814B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103940576B (en) * | 2014-04-22 | 2016-06-15 | 西南交通大学 | The appraisal procedure of a kind of wind1 tunnel trial balance based on acceleration signal approximate entropy |
| CN106650095B (en) * | 2016-12-21 | 2020-07-14 | 中国航天空气动力技术研究院 | Correction method of unmanned aerial vehicle control matrix based on wind tunnel test data and CFD calculation |
| CN108414185B (en) * | 2018-02-08 | 2020-01-31 | 上海机电工程研究所 | Zero error processing method for wind tunnel test data of symmetric aircraft |
| CN111309681B (en) * | 2020-01-19 | 2024-06-11 | 中国平安人寿保险股份有限公司 | Data modification method, device and computer readable storage medium |
| CN112100878A (en) * | 2020-08-18 | 2020-12-18 | 重庆地质矿产研究院 | Method for establishing time domain sequence stratigraphic section by using forward modeling of sedimentation |
| CN111999441A (en) * | 2020-08-28 | 2020-11-27 | 福建美营自动化科技有限公司 | Multi-channel extremely-low-concentration combustible and explosive gas rapid detector and gas discrimination method |
| CN112345199B (en) * | 2020-10-29 | 2022-07-22 | 中国空气动力研究与发展中心高速空气动力研究所 | Method for correcting impact of vibration of attack angle sensor of temporary-impulse high-speed wind tunnel |
| CN113029512B (en) * | 2021-04-01 | 2022-08-05 | 中国空气动力研究与发展中心高速空气动力研究所 | Air flow field model correction control method for temporary transonic wind tunnel |
| CN113155405B (en) * | 2021-04-27 | 2022-09-20 | 中国空气动力研究与发展中心设备设计与测试技术研究所 | Wind tunnel test attack angle mechanism pose parameter tracing method |
| CN113255577B (en) * | 2021-06-18 | 2021-12-14 | 中铁大桥科学研究院有限公司 | Active control intelligent data processing method for construction vibration parameters of cable-stayed bridge |
| CN113639956B (en) * | 2021-10-18 | 2022-01-18 | 中国空气动力研究与发展中心高速空气动力研究所 | Calibration device and calibration method for model inclination angle measurement device |
| CN114754973A (en) * | 2022-05-23 | 2022-07-15 | 中国航空工业集团公司哈尔滨空气动力研究所 | Wind tunnel force measurement test data intelligent diagnosis and analysis method based on machine learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5848073A (en) * | 1991-12-19 | 1998-12-08 | Lucent Technologies Inc. | Method and apparatus for predicting transmission system errors and failures |
| CN102122132A (en) * | 2010-01-11 | 2011-07-13 | 北京航空航天大学 | Intelligent control system for environmental simulation system based on a fuzzy neural network |
| CN102637742A (en) * | 2012-04-26 | 2012-08-15 | 北京大学 | Oxide semiconductor thin-film transistor and preparation method thereof |
| CN102880907A (en) * | 2012-08-24 | 2013-01-16 | 华锐风电科技(集团)股份有限公司 | Method and device for correcting wind speed |
-
2013
- 2013-06-09 CN CN201310230816.9A patent/CN103335814B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5848073A (en) * | 1991-12-19 | 1998-12-08 | Lucent Technologies Inc. | Method and apparatus for predicting transmission system errors and failures |
| CN102122132A (en) * | 2010-01-11 | 2011-07-13 | 北京航空航天大学 | Intelligent control system for environmental simulation system based on a fuzzy neural network |
| CN102637742A (en) * | 2012-04-26 | 2012-08-15 | 北京大学 | Oxide semiconductor thin-film transistor and preparation method thereof |
| CN102880907A (en) * | 2012-08-24 | 2013-01-16 | 华锐风电科技(集团)股份有限公司 | Method and device for correcting wind speed |
Non-Patent Citations (1)
| Title |
|---|
| 基于遗传神经网络模型实时误差修正任意角测量系统;叶兵;《中国博士学位论文全文数据库 工程科技Ⅰ辑》;20040915(第03期);第58-59页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103335814A (en) | 2013-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103335814B (en) | Correction method for inclination angle measurement error data of experimental model in wind tunnel | |
| CN104992225B (en) | A kind of compensation method of large deformation flexible body dynamic measuring data temperature and humidity | |
| CN104573248B (en) | The multiple dimensioned extreme learning machine training method of Fiber Optic Gyroscope Temperature Drift based on EMD | |
| CN108959794A (en) | A kind of structural frequency response modification methodology of dynamics model based on deep learning | |
| CN102759430B (en) | BP (Back Propagation) neural network based high-precision correction and test method for resonance cylinder pressure sensor | |
| CN109582988B (en) | Method for determining position of strain gauge for monitoring vibration stress of blade of aircraft engine | |
| CN104236893A (en) | Performance parameter test system and performance parameter test method of hydraulic damper | |
| CN104122031A (en) | Silicon pressure sensor temperature compensation method based on extreme learning machine | |
| CN115270239A (en) | Bridge reliability prediction method based on dynamic characteristics and intelligent algorithm response surface method | |
| CN110555231A (en) | Dynamic simulation model correction method | |
| CN105956318B (en) | Based on the wind power plant group of planes division methods for improving division H-K clustering method | |
| CN116029219A (en) | Aerodynamic heat prediction method, device, equipment and storage medium for aircraft | |
| CN113176022A (en) | Segmented neural network pressure sensor pressure detection method and system | |
| CN110705041A (en) | Linear structure working modal parameter identification method based on EASI | |
| US20150073730A1 (en) | Mechanical strain gauge simulation | |
| CN109738669A (en) | A kind of temperature drift compensation method of piezoelectric acceleration transducer | |
| Li et al. | An interpretable aerodynamic identification model for hypersonic wind tunnels | |
| CN109086501A (en) | A kind of flutter prediction technique | |
| CN104102127A (en) | Airborne aerodynamic parameter identification system | |
| CN106568463B (en) | Angular oscillation generating device and gyro frequency measurement method for gyro frequency characteristic | |
| CN113158315B (en) | Rock-soil body parameter three-dimensional non-stationary conditional random field modeling method based on static sounding data | |
| CN114791334A (en) | Calibration simplification method for pressure sensor | |
| CN107061032A (en) | The Forecasting Methodology and forecasting system of a kind of engine operating state | |
| CN111210877A (en) | Method and device for deducing physical property parameters | |
| CN105129073B (en) | A kind of aircraft gust alleviation adaptive feedforward control method based on GOFIR |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150527 Termination date: 20160609 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |