CN1275024C - Time base varying monitoring method for large-scale construction damage status real time identification - Google Patents
Time base varying monitoring method for large-scale construction damage status real time identification Download PDFInfo
- Publication number
- CN1275024C CN1275024C CN 03132492 CN03132492A CN1275024C CN 1275024 C CN1275024 C CN 1275024C CN 03132492 CN03132492 CN 03132492 CN 03132492 A CN03132492 A CN 03132492A CN 1275024 C CN1275024 C CN 1275024C
- Authority
- CN
- China
- Prior art keywords
- measuring point
- data
- damage
- time
- real time
- 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
Images
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The present invention provides a time base varying monitoring technique for large-scale construction damage status real time identification, which comprises data collection, data analysis and postprocessing, wherein the data collection comprises monitoring point selection and the data collection; the data analysis comprises the selection of a reference monitoring point, the determination of a conductive value, the extraction of a free vibration signal of a monitoring point, the inspection of signal quality, the reading of vibration information, the determination of recognizing free degree, the determination of virtual expansion reference time, the calculation of average dynamic information, the calculation of the standard deviation of the whole information of the monitoring point and the extension time of a virtual monitoring point, the formation of a data matrix, false status removal, the calculation of a characteristic parameter, and the formation of a characteristic curve; the postprocessing comprises damage expression, error control, status prediction, damage statistic analysis, structural damage early warning and maintenance suggests thereof. The present invention provides a structural real time modal parameter for correcting damaged structural models, provides parameters for a structural life cycle and the control and the adjustment of a status, and ensures the normal operation of a structure in a service.
Description
(1) technical field
What the present invention relates to is a kind of at large-scale building hydraulic engineering structure objects, based on the large-scale data real-time processing method of model, and specifically a kind of method that the large scale structure damage status is monitored in real time.
(2) background technology
The importance and the market demand of large scale structure damage identification grow with each passing day.The damage of large scale structure identification at present mainly contains local and whole two kinds of technology.Because the whole monitoring technology can be found the hidden danger of structure early, easy to use, the monitoring expense is cheap, is the main means of future health monitoring, has very high theory and practical value.The whole monitoring technology is mainly carried out based on structural vibration parameter recognition technology.Damage monitoring based on vibration has two big class technology at present: frequency field monitoring technology and time domain detection technique.The frequency field technology is mainly used in small-sized (test) structure, and it generally requires the input of measurement structure, thereby can not carry out on-line monitoring to structure.The time domain technology can be carried out by real-time online, thereby is the major technique means of large-scale (reality) structure damage monitoring owing to do not need the input signal of structure.In the time domain technology, widespread usage be the not damaged system identification method at mechanical system of coming from machinery and aviation field conversion, they are: (1) Random Decrement method; (2) ITD method; (3) stochastic subspace method; (4) time series method; (5) based on the various clock signal analytical approachs of short-scale (as: wavelet analysis method).These method common features are: the structure number of degrees of freedom, is not very high, input signal steadily and be white noise, measuring point will ask many.Have only and satisfy these conditions and could obtain recognition effect preferably, obtain reflecting the dynamic parameter of system features.But for large scale structure,, thereby be difficult to satisfy the condition of traditional time domain method, directly use these technology can not on large scale structure, obtain structure dynamic feature coefficient accurately because complexity is far above the system of machinery or aviation field; On the other hand, conventional art is primarily aimed at the dynamic parameter identification of normal configuration, they do not consider that the feedback influence of damage back to recognition methods takes place structure, therefore the structural parameters of identification can not be followed the tracks of the bad change process with reflect structure well, be embodied in the on-the-spot result's of recognition technology poor repeatability, the damage positioning difficulty is big, and noise-sensitive degree height often causes damaging the failure of identification and location.For this reason, at growing to even greater heights that large scale structure monitoring requires, is badly in need of a kind of can real-time online recognition structure dynamic feature coefficient simultaneously again can tracking structure damage variation monitoring method.
(3) summary of the invention
The object of the present invention is to provide a kind ofly can provide the real-time modal parameter of structure accurately for structural damage model makeover process, the accurately response of predict under circumstances not known, for the operating mode real-time parameter that provides important is accurately controlled and adjusted to structure life cycle and state, guarantee large and complex structure interior health of phase under arms, be the time base variable monitoring method of large scale structure damage status Real time identification.
The object of the present invention is achieved like this: be made up of data acquisition, data analysis and post-processing step.Data acquisition comprises that the optimized choice of measuring point and multi-channel data gather in real time.The optimized choice of measuring point comprise at first carry out structural finite element analysis, form the structural modal information matrix, maximization structural information matrix obtains the measuring point Candidate Set, the tolerance interval that distributes of computation structure measuring point is determined the eigenwert second order sensitivity profile of distribution range, computation structure quality and the rigidity of measuring point, obtained the optimum measuring point of structure according to redundant measuring point at last then.Multi-channel data is gathered in real time and is comprised that at first the signal of degree of will speed up sensor is connected to adapter, is input to master processor again through signal condition card and data collecting card.Data analysis comprises: import sensitivity for analysis, selection is with reference to measuring point, determine conduction value, extract measuring point free vibration signal, carry out the signal quality inspection: if satisfy then transmit free vibration information, if do not satisfy then return and determine the conduction value process, again carry out the extraction of signal, read in the free vibration signal of extraction by time base variable data matrix structure, determine the identification number of degrees of freedom,, determine virtual extended reference time, calculate average dynamic information, calculate measuring point Global Information standard deviation, determine virtual measuring point time expand, form data matrix, reject false mode, the calculated characteristics parameter, form characteristic curve.Data Post comprises that expression, damage identification and analysis of Positioning Error, large scale structure damage identification statistical study, configuration state prediction, damage early warning and structural repair suggestion and large scale structure that the large scale structure damage is extracted design on-the-spot dynamic characteristic accumulation and checking.The present invention can also be included in the multi-channel data collection in real time, signal to acceleration transducer carries out the classification of monitoring scale earlier, for the measuring point number greater than 150 large-scale monitoring scale, adopt the integral collecting technical finesse, the host computer image data, the real-time video data of slave computer, monitoring sensor state, storage of collected data, the transmission image data is to master processor, upper and lower computer is driven under a mainboard by two CPU and two hard disk in a cabinet; Less than 100 medium monitoring scale, adopt the technical finesse of unit acquisition system for the measuring point number, data acquisition, demonstration, storage and transmission are all carried out on an industrial computer.
The invention has the advantages that: (1) is at line drawing large scale structure (as large bridge, dam body, head tower, marine structure, ships, launching tower, large-scale transportation system etc.) characteristic parameter, reflection and real-time follow-up structural health conditions change situation, determine that STRUCTURE DAMAGE LOCATION prevents that to greatest extent serious accident from taking place; (2) because empirical, the instability of large scale structure Theoretical Calculation, simultaneously the extreme environment load can not simulation, Theoretical Calculation has often implied extendible defective, in a single day the Service Environment of large scale structure has satisfied the condition that this defective takes place, to cause catastrophic results, so the correction technique of large and complex structure Theoretical Calculation is to guarantee structure healthy indispensable gordian technique of its work mission of finishing under the hypothesis environment.The present invention can provide the real-time modal parameter of structure accurately for the structural model makeover process, under the support of statistics and forecast model technology, the accurately response of structure under circumstances not known under the prediction statistical significance, for the real-time working condition parameter that provides very important is accurately controlled and adjusted to structure life cycle and state, guarantee the large and complex structure health in the phase under arms.
(4) description of drawings
Fig. 1 is ultimate principle figure of the present invention;
Fig. 2 is the ultimate principle figure of data acquisition;
Fig. 3 is the schematic diagram that measuring point is selected;
Fig. 4 is the schematic diagram of real time data acquisition;
Fig. 5,6, the 7th, the schematic diagram of data analysis;
Fig. 8 is the schematic diagram of aftertreatment.
(5) embodiment
For example the present invention is done in more detail below and describes:
The time base variable monitoring method of large scale structure damage status Real time identification mainly is made of three steps.
Data acquisition is made up of two step structures, and promptly the optimized choice of measuring point and multi-channel data are gathered in real time.Wherein measuring point selects to comprise that the finite element analysis of carrying out structure earlier, the modal information matrix of forming structure, maximization structural information matrix obtain the processes such as optimum measuring point that the tolerance interval that measuring point Candidate Set, computation structure measuring point distribute is determined the eigenwert second order sensitivity profile of distribution range, computation structure quality and the rigidity of measuring point, obtained structure with reference to redundant measuring point; Multi-channel data is gathered in real time and is comprised that the signal of acceleration transducer is connected to adapter, is input in the master processor through signal condition card and data collecting card again.
Hyperchannel is counted in real time and adopted is one of gordian technique of time base variable monitoring dynamic damage.After the structure measuring point was optimized, the measuring point scale still had tens to hundreds of for large and complex structure, and in order to finish the requirement of time base variable monitoring method, number adopts has also implemented following technical scheme:
(1), adopts the integral collecting technology for large-scale monitoring scale (the measuring point number is greater than 150).The host computer image data, the real-time video data of slave computer, the monitoring sensor state, the storage of collected data, the transmission image data is to master processor.Upper and lower computer is driven under a mainboard by two CPU and two hard disk in a cabinet, and DRAM provides frequency to be not less than 266MHz.The integrated collection auxiliary device of having simplified has greatly been strengthened the ability that several extraction systems are resisted rugged surroundings.
(2),, adopt unit acquisition system technology because the actual samples frequency of structure generally is not higher than 1000Hz for medium monitoring scale (the measuring point number is less than 100).Data acquisition, demonstration, storage and transmission are all carried out on an industrial computer.
(3) based on the synchronous collection method of data acquisition board (DAQ plate) high speed switching technique.Utilize the high-speed data frequency splitting technology that data-signal is connected to adapter, the signal of adapter is transferred to the DAQ plate, realize the synchronous acquisition of data based on frequency needs, utilize the adapter technology, can realize the frequency conversion setting of hardware, according to the variation of environment, adjust the sample frequency of number extraction system in real time.
(4) frequency conversion sampling method.Because the real time altering of environment, the single sample frequency of tradition is not only lost a large amount of useful informations but also is caused several extraction systems to waste in a large number.Time base variable is counted the variation of extraction system according to external environment condition for this reason, by the sample distribution table of setting, with the real-time automatically sample frequency of adjusting each passage of software, guarantees accurately obtaining of structural information.
(5), adopt and finished the data acquisition software of the real-time module of LabView in conjunction with the C language at large-scale collection scale.The real-time module of LabView is a real time operating system of independently supporting many CPU, utilizes this platform can efficiently finish the host computer task of data acquisition, adopts under the Data Transport Protocol at special number, and data are controllably transferred to slave computer reposefully.In addition, adopt the C language to work out sensor states supervisory programme and sampling variable frequency control program, and program is fired among the EPROM, be embedded into the integral collecting system, the host program real-time calling is provided, finishes the synchronous coordination of monitoring, frequency modulation and the collection of DAQ plate of data.
(6), adopt the LabWindows establishment and finished independently sampling system at medium collection scale.This system comprises sensor states equally and monitors and frequency conversion C language program curing module.
(7) above all hardware and software feature are applicable to two types of industrial computer and portable computers.
Data analysis is the large scale structure faulted condition core of monitoring in real time.Its key is the time base variable data matrix that forms the monitoring of structures feature, by the characteristic parameter of data matrix computation structure: frequency, damping and the vibration shape and the behavioral characteristics curve that obtains structure, correcting principle finite element FEM model.Its content can be divided into following a few step:
(1) imports acceleration signal into.According to the structure vibration shape and stress distribution, measurement structure vibration acceleration time series on location survey point most preferably then according to the demand of time base variable to the different measuring points acceleration signal, is imported the signal of collection into multithreading from slave computer.
(2) free vibration signal extraction.In structure military service process, selected measuring point be with reference to measuring point, determines the conduction value of Random Decrement, the free vibration signal of each measuring point of extraction structure according to reference measuring point time series variance.
(3) data matrix structure.The key in this step is the time series of constructing virtual measuring point.Because the theoretical measuring point of large scale structure is very huge, can not all measure all measuring points during actual measurement,, need compress and two kinds of methods of actual measuring point expansion theoretical measuring point in order to satisfy the completeness of modal information.When actual measuring point is expanded, general method is the vibration shape of measuring to be carried out interpolation according to mode orthogonality principle expand rank, if the expansion measuring point is more, can cause serious mode ill-condition equation, the real-time calculating effect of deterioration system, therefore the time base variable method adopts the time series of virtual measuring point technology to the actual measuring point constructing virtual of fraction, formation has the mode meaning and actual and non-existent virtual measuring point, thereby enlarge the measuring point scope, the interpolation of carrying out on the measuring point that enlarges among a small circle expands rank, guarantee the real-time of calculating and monitoring thereof, therefore the virtual measuring point seasonal effect in time series of time base variable building method is the core technology of data processing method in this patent, below the main process of concise and to the point narration time base variable.
We introduce the time base variable time variable: when producing virtual measuring point by actual measuring point, based on the expansion time that increases on the actual measuring point time series basis.The time base variable main task is construction data matrix [Z],
They form following equation:
By finding the solution formula (1), obtain the characteristic parameter of structure, and the behavioral characteristics curve.[Z],
Be to have comprised actual measuring point and virtual measuring point seasonal effect in time series information matrix.Its construction process is as follows:
(1) one on selected structure is calculated the whole standard deviation WSD of this measuring point time series with reference to measuring point.
N is a seasonal effect in time series length in the formula (2); H
iFor average dynamic information A DI, calculate according to following formula
P is actual measuring point number in the formula (3), x
i, x
jThe sampling time sequence data set that representative is chosen in selected reference measuring point, the principle of choosing is
In the formula (4), A
iBe measuring point seasonal effect in time series amplitude, M
n iBe i the time series set of selecting measuring point, std represents standard deviation.Promptly in the measuring point time series, come out greater than the time point signal of measuring point amplitude standard deviation is selected, form multidate information ADI, and then the expansion time that increases of the virtual measuring point of computation structure.
(2) model correction and damage location.Structure dynamic parameter according to identification, the inconsistent structural region of the actual identification parameter of correcting principle model analysis and structure, the rigidity of structure coefficient that obtains revising for the first time distributes and dynamic feature coefficient, and the parameter that will obtain for the first time is as the reference standard, repeating the structural modifications in (1)～(3) step in structure is on active service whole process calculates, comparative structure corrected parameter and canonical parameter are determined STRUCTURE DAMAGE LOCATION.
Aftertreatment is primarily aimed at 1: more to large scale structure measuring point number, the rigidity of structure of correction and mass matrix display unit are huge, are difficult for obviously expressing damage signal; 2: damage identification and analysis of Positioning Error; 3: large scale structure damage identification statistical study; 4: the configuration state prediction; 5: structure early warning and maintenance suggestion function, this five kinds of purposes and designing, the major technology characteristics are as follows:
(1) relation of direct visual representation damage information of correction matrix three-dimensional information expression and structure node is mainly adopted in the time base variable aftertreatment, application matrix shadow casting technique is with node and damage set of variations synthesizing one-dimensional state simultaneously, mark damage position with two-dimentional form and rectilinear figure, make the result simple and clear.
(2) the structural damage positioning error is represented with the false alarm rate of recognition confidence and macrolesion actual location, is provided with the check the value of false alarm rate in the program, and when false alarm rate rose, Automatic Program adjustment identification threshold and recognition efficiency guaranteed accuracy of identification.
(3) the configuration state prediction is according to the characteristic parameter changing condition of structure identification, uses the avatars based on model to come out to the regional stiffness variation trend of structure.
(4) time base variable large scale structure damage Real time identification has adopted the tri-layer maintenance program based on expert system, according to the damage status of identification, provides maintenance suggestion and decision-making automatically, if the damage of identification is higher than third level sign, will start warning.Wherein reporting to the police has 5 classification measures, and the warning of highest level need be keeped in repair immediately.The time base variable after-treatment system will change along with different monitoring targets and user's demand.
Except top five major functions, aftertreatment has the function of accumulation large scale structure characteristic, for design is verified afterwards.
Claims (2)
1, a kind of time base variable monitoring method of large scale structure damage status Real time identification, it is made up of data acquisition, data analysis and post-processing step, it is characterized in that:
1.1 data acquisition comprises that the optimized choice of measuring point and multi-channel data gather in real time;
1.1.1 the optimized choice of measuring point comprise at first carry out structural finite element analysis, form the structural modal information matrix, maximization structural information matrix obtains the measuring point Candidate Set, the tolerance interval that distributes of computation structure measuring point is determined the eigenwert second order sensitivity profile of distribution range, computation structure quality and the rigidity of measuring point, obtained the optimum measuring point of structure according to redundant measuring point at last then;
1.1.2 gathering in real time, multi-channel data comprises that at first the signal of degree of will speed up sensor is connected to adapter, is input to master processor again through signal condition card and data collecting card;
1.2 data analysis comprises: import sensitivity for analysis, selection is with reference to measuring point, determine conduction value, extract measuring point free vibration signal, carry out the signal quality inspection: if satisfy then transmit free vibration information, if do not satisfy then return and determine the conduction value process, again carry out the extraction of signal, read in the free vibration signal of extraction by time base variable data matrix structure, determine the identification number of degrees of freedom,, determine virtual extended reference time, calculate average dynamic information, calculate measuring point Global Information standard deviation, determine virtual measuring point time expand, form data matrix, reject false mode, the calculated characteristics parameter, form characteristic curve;
1.3 Data Post comprises that expression, damage identification and analysis of Positioning Error, large scale structure damage identification statistical study, configuration state prediction, damage early warning and structural repair suggestion and large scale structure that the large scale structure damage is extracted design on-the-spot dynamic characteristic accumulation and checking.
2, the time base variable monitoring method of large scale structure damage status Real time identification according to claim 1, it is characterized in that: in multi-channel data is gathered in real time, signal to acceleration transducer carries out the classification of monitoring scale earlier, for the measuring point number greater than 150 large-scale monitoring scale, adopt the integral collecting technical finesse, host computer image data, slave computer show image data, monitoring sensor state, storage of collected data in real time and transmit image data to master processor; Upper and lower computer is driven under a mainboard by two CPU and two hard disk in a cabinet; Less than 100 medium monitoring scale, adopt the technical finesse of unit acquisition system for the measuring point number, data acquisition, demonstration, storage and transmission are all carried out on an industrial computer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 03132492 CN1275024C (en) | 2003-07-09 | 2003-07-09 | Time base varying monitoring method for large-scale construction damage status real time identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 03132492 CN1275024C (en) | 2003-07-09 | 2003-07-09 | Time base varying monitoring method for large-scale construction damage status real time identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1566907A CN1566907A (en) | 2005-01-19 |
| CN1275024C true CN1275024C (en) | 2006-09-13 |
Family
ID=34469918
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 03132492 Expired - Fee Related CN1275024C (en) | 2003-07-09 | 2003-07-09 | Time base varying monitoring method for large-scale construction damage status real time identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1275024C (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100420944C (en) * | 2005-12-29 | 2008-09-24 | 西北工业大学 | Detection method for damage of random vibration structure based on correlation function amplitude vector |
| CN101197807B (en) * | 2006-12-13 | 2010-05-12 | 四川川大智胜软件股份有限公司 | Intelligent communication server |
| CN100468422C (en) * | 2006-12-31 | 2009-03-11 | 中国海洋大学 | Method for modifying structural model by cross modal of cross model |
| CN101122583B (en) * | 2007-09-06 | 2010-12-29 | 华中科技大学 | Sheared frame structure damage detection method |
| CN101221104B (en) * | 2007-10-16 | 2010-08-11 | 吴智深 | Structure health monitoring method based on distributed strain dynamic test |
| CN101488223B (en) * | 2008-01-16 | 2012-03-28 | 中国科学院自动化研究所 | Image curve characteristic matching method based on average value standard deviation descriptor |
| CN104112072B (en) * | 2014-07-15 | 2017-04-05 | 华侨大学 | Operational modal parameter recognition methods based on the principal component analysis of wavelet threshold denoising |
| CN104297004A (en) * | 2014-09-18 | 2015-01-21 | 天津大学 | Real-time bridge damage early-warning method based on AR-ARX model |
| CN104807661B (en) * | 2015-04-22 | 2018-07-31 | 中国十七冶集团有限公司 | A kind of high-rise and tall and slender structure Dynamic testing evaluation on bearing capacity method |
| CN104807607B (en) * | 2015-04-23 | 2017-10-24 | 成都畅达通检测技术股份有限公司 | The structures Defect inspection system and its detection method of spectral property are responded during based on excitation state |
| CN108801823B (en) * | 2018-06-25 | 2019-08-23 | 南京航空航天大学 | A kind of multiple dimensioned composite structure local fatigue appraisal procedure and system |
| CN109870404B (en) * | 2019-03-06 | 2021-07-13 | 石家庄铁道大学 | Rain shed structure damage identification method and device and terminal equipment |
| CN111254892B (en) * | 2020-01-20 | 2020-12-22 | 中国水利水电科学研究院 | Method for setting safety early warning threshold value of double-layer lining water delivery tunnel operation structure on soft foundation |
| CN111272764B (en) * | 2020-01-22 | 2023-04-28 | 哈尔滨工业大学 | Non-contact image identification mobile management and control system and method for large intelligent temporary platform |
| CN113255193B (en) * | 2021-06-17 | 2021-10-15 | 浙江中自庆安新能源技术有限公司 | Method and system for constructing virtual measuring points of rotating equipment based on finite element analysis |
| CN113465734B (en) * | 2021-09-02 | 2021-12-10 | 清华大学 | Real-time estimation method for structural vibration |
| CN115859708B (en) * | 2022-11-16 | 2024-03-19 | 大连理工大学 | Honeycomb sandwich structure damage model correction method based on distributed optical fiber measurement |
-
2003
- 2003-07-09 CN CN 03132492 patent/CN1275024C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1566907A (en) | 2005-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1275024C (en) | Time base varying monitoring method for large-scale construction damage status real time identification | |
| CN114893360B (en) | Method and system for identifying abnormal vibration of tower of wind turbine generator and monitoring running state | |
| CN114611778B (en) | Reservoir water level early warning method and system based on warehousing flow | |
| CN116380445B (en) | Equipment state diagnosis method and related device based on vibration waveform | |
| CN104915534A (en) | Deformation analysis and decision-making method of electric power tower based on sequence learning | |
| CN116245261A (en) | Wind field wind speed prediction method based on time sequence distribution adaptation | |
| CN113344275B (en) | Floating platform wave climbing online forecasting method based on LSTM model | |
| CN1361480A (en) | Vibration throw parameter monitoring and analyzing method for water power set | |
| CN116818246B (en) | Earthquake simulation vibrating table model prediction control method, system, medium and terminal | |
| Li et al. | Structural health monitoring data anomaly detection by transformer enhanced densely connected neural networks | |
| CN117891644A (en) | Data acquisition system and method based on digital twin technology | |
| CN110222384A (en) | A kind of Analysis of Bridge Cracks method, apparatus and storage medium | |
| CN117556670A (en) | Assembled structure damage identification method based on Bayesian theory | |
| CN117688721A (en) | Dynamic motion identification method and device for offshore floating platform | |
| CN116663126A (en) | Bridge temperature effect prediction method based on channel attention BiLSTM model | |
| CN116108749A (en) | Wind turbine flow field prediction method based on deep learning | |
| CN115630278A (en) | Network vibration damage detection method based on channel-space-time attention mechanism | |
| CN111428420B (en) | Method and device for predicting sea surface flow velocity, computer equipment and storage medium | |
| CN117058152B (en) | Intelligent detection method for construction quality of fabricated building based on point cloud data | |
| CN113343344B (en) | Structural response prediction method, device and equipment based on tower crane dynamic work | |
| CN117540453A (en) | Agent model optimization method, system and storage medium | |
| Seo et al. | Deep learning approach for predicting crack initiation position and size in a steam turbine blade using frequency response and model order reduction | |
| YOUNG et al. | COASTAL MODEL TEST BED EVALUATION OF CSHORE HYDRODYNAMIC PREDICTION PERFORMANCE | |
| CN118133376A (en) | Historic building digital twin processing method and device based on historic building information model | |
| CN117664569A (en) | Fault detection method and device and computer readable storage medium |
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 | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20060913 Termination date: 20090709 |