CN116165274B - Urban orbit damage identification method based on Bayesian global sparse probability principal component analysis - Google Patents
Urban orbit damage identification method based on Bayesian global sparse probability principal component analysis Download PDFInfo
- Publication number
- CN116165274B CN116165274B CN202310129979.1A CN202310129979A CN116165274B CN 116165274 B CN116165274 B CN 116165274B CN 202310129979 A CN202310129979 A CN 202310129979A CN 116165274 B CN116165274 B CN 116165274B
- Authority
- CN
- China
- Prior art keywords
- data
- monitoring data
- global
- damage
- vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000006378 damage Effects 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000000513 principal component analysis Methods 0.000 title claims abstract description 25
- 238000012544 monitoring process Methods 0.000 claims description 69
- 239000013598 vector Substances 0.000 claims description 52
- 239000011159 matrix material Substances 0.000 claims description 34
- 238000013519 translation Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000005259 measurement Methods 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 230000036541 health Effects 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 7
- 239000013307 optical fiber Substances 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000008602 contraction Effects 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 abstract description 8
- 238000011156 evaluation Methods 0.000 abstract description 3
- 238000012423 maintenance Methods 0.000 abstract description 3
- 238000002372 labelling Methods 0.000 abstract description 2
- 208000027418 Wounds and injury Diseases 0.000 description 9
- 208000014674 injury Diseases 0.000 description 9
- 238000001514 detection method Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 3
- 230000003902 lesion Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 102100037651 AP-2 complex subunit sigma Human genes 0.000 description 1
- 101000806914 Homo sapiens AP-2 complex subunit sigma Proteins 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012847 principal component analysis method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/04—Analysing solids
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/44—Processing the detected response signal, e.g. electronic circuits specially adapted therefor
- G01N29/4472—Mathematical theories or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2291/00—Indexing codes associated with group G01N29/00
- G01N2291/02—Indexing codes associated with the analysed material
- G01N2291/023—Solids
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Mathematical Optimization (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Immunology (AREA)
- General Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Theoretical Computer Science (AREA)
- Pathology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Acoustics & Sound (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
- Machines For Laying And Maintaining Railways (AREA)
Abstract
The application provides a method for identifying urban rail damage based on Bayesian global sparse probability principal component analysis. According to the method, firstly, the vehicle-passing vibration data are preprocessed and aligned, so that the global features of the vehicle-passing vibration data are more obvious, the principal component analysis and identification are facilitated, and then the global sparse mode is solved by adopting Bayesian global sparse probability principal component analysis, namely, which data are related to the global features are selected. Because the damage characteristic is non-global, the global sparse mode is a damage sensitive index, data related to the global characteristic can be judged to be in a lossless state, and other data correspond to possible track damage, so that the aim of unsupervised track damage identification is fulfilled. The method does not need high-cost damage diagnosis data labeling, has the advantages of global property, real-time property and the like, can timely find and effectively detect hidden or unobvious damage, provides a new way for exploring and realizing full-time global orbit intelligent diagnosis and evaluation, and has good reference value for urban orbit operation and maintenance.
Description
Technical Field
The application belongs to the technical field of urban rail operation and maintenance and structural health monitoring, and particularly relates to a method for identifying urban rail damage based on Bayesian global sparse probability principal component analysis.
Background
With the continuous improvement of the urban level in recent years, urban rail transit is built and put into use in a large scale, and by the end of 2022, 51 cities in the whole country are provided with urban rail transit, and the operation mileage reaches 8012.85 km. At present, the rail state mainly adopts modes of manual appearance detection, static rail inspection vehicle detection and dynamic rail inspection vehicle detection, and an intelligent and real-time monitoring means is lacked. The method has the limitation that the damage cannot be found in time, the damage is difficult to effectively detect for hidden or unobvious damage, the whole line and the whole coverage in space are difficult to realize, and the like.
Structural health monitoring has become an important way to ensure the service safety of urban rail transit infrastructure in recent years by installing sensors on the structure, sensing, transmitting and processing monitoring data in real time. The vibration-based monitoring has the advantages of global property, real-time property and the like, and is an important means for monitoring track damage. The urban rail universe route is long, the damage variety is various, the concealment is strong, the damage characteristics have diversity and complexity, the coupling influence mechanism of the rail system is complex, the cost for marking the data of the on-line monitoring supervision type damage diagnosis model is high, and a new unsupervised intelligent diagnosis method is urgently needed to be researched.
Disclosure of Invention
The application aims to solve the problems in the prior art and provides a method for identifying urban rail damage based on Bayesian global sparse probability principal component analysis. The method is suitable for urban rail damage identification.
The application is realized by the following technical scheme, and provides a method for identifying urban rail damage based on Bayesian global sparse probability principal component analysis, which comprises the following steps:
firstly, arranging an optical fiber sensor on an urban rail, and carrying out health monitoring on measuring points along the way to obtain measurement data of average dynamic strain of each measuring point when the vehicle passes each time;
step two, preprocessing alignment vehicle-passing vibration data, taking monitoring data of one train as a standard signal, and performing translation and expansion and contraction on a time domainFinishing the starting time and train speed of other subway vibration time sequences, performing interpolation correction on the data with the dislocation after adjustment, selecting the result with the maximum correlation with the standard signal as an alignment result, and forming a track monitoring data matrix
Step three, based on the track monitoring data matrix X and the dimension M of the potential space z thereof, executing an EM algorithm, and obtaining uncertain parameters { v, alpha, sigma) by initial value iteration 2 ，μ 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D The most probable value of sigma;
step four, converting the continuous vector v obtained in the step three into a binary vector o epsilon {0,1} D The global sparse mode of the characteristic is characterized, and the selected related variable in the data matrix X corresponds to a non-zero term in the sparse mode vector o;
and fifthly, judging whether each piece of monitoring data corresponds to the damage state or the lossless state by utilizing the binary element of the vector o obtained in the step four.
Further, the second step specifically comprises:
step 2.1, selecting monitoring data of a train as standard signals;
step 2.2, selecting a proper interpolation method for other subway vibration time sequences, and serializing discrete monitoring data;
2.3, carrying out translation and scaling on the continuous monitoring data in a reasonable range in a time domain, wherein the range selection is as small as possible while ensuring that the monitoring data can be aligned with a standard signal;
step 2.4, discretizing the continuous monitoring data subjected to translation and scaling according to the time data of the standard signal, and calculating the correlation between the continuous monitoring data and the standard signal;
step 2.5, selecting a translation scaling result with the maximum correlation with the standard signal as an alignment result;
step 2.6, common part forming track monitoring on the time domain of the alignment resultData matrix
Further, the process of obtaining the most probable value of the uncertain parameter in the third step is based on the variational EM algorithm, and the hidden variable { mu ] of the control variational distribution q (Z, P) is iteratively updated in turn according to the track monitoring data matrix X and the dimension M of the data potential space Z 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D Sigma and uncertainty parameter θ= { v, α, σ 2 Zero, realize the free energy of variation A process of minimization in which H [ q (Z, P)]The differential entropy of the variation distribution q (Z, P) is represented.
Further, the process specifically comprises the following steps:
step 3.1, initializing uncertainty parameters { v, α, σ } 2 ，μ 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D ，∑}；
Step 3.2, calculating E, updating the variation distribution q (Z, P) =q (Z) q (P), wherein The calculation process is for all i e { 1..N } and k e { 1..D } calculated according to the following formula:
wherein m= (M 1 ，...，m D ) T ，
And 3.3, calculating M steps through the following formula, and updating the uncertain parameter theta:
for all k e {1,..d },
step 3.4, repeating steps 3.2 to 3.3 until free energyThe variation of (c) is smaller than a set threshold value, resulting in the most probable value of each uncertain parameter.
Further, the step 3.1 specifically includes:
step 3.1.1, initializing each element in the vector v to 1;
step 3.1.2, taking the parameter alpha as 0.1, 1 and 10 respectively;
step 3.1.3 parameter { μ } 1 ，...，μ N ，m 1 ，...，m D Initializing to singular vectors of a data matrix X;
step 3.1.4, parameter { S } 1 ，...，S D Sigma is initialized to Sigma=I M ，S 1 ＝...＝S D ＝α -2 I M ；
And 3.1.5, respectively performing a plurality of preliminary iterations for different initialization schemes formed by three values of the parameter alpha, and selecting a scheme with the lowest free energy of the iteration result as the initialization setting of each parameter.
Further, the fourth step specifically comprises:
step 4.1, taking binary vector o for all k e {1,., D } (k) Corresponding position elements of the first k largest elements in the corresponding continuous vector v are 1, and the other elements are 0;
step 4.2, evaluate each binary vector o (k) Is defined by the marginal likelihood function p (x|o, alpha, sigma) 2 )＝∫p(x|P，o，σ 2 ) P (P|alpha) dP, where
Step 4.3, selecting the binary vector o with the maximum marginal likelihood function (k) As binary conversion result o of the successive vector v.
Further, the binary elements in the fifth step include 0 and 1, where the element value of the damaged state is 0 and the element value of the lossless state is 1.
The application provides a city orbit damage recognition system based on Bayesian global sparse probability principal component analysis, which comprises:
and a data acquisition module: arranging optical fiber sensors on the urban rail, and carrying out health monitoring on the along-road measuring points to obtain measurement data of average dynamic strain of each measuring point when each measuring point passes through the vehicle;
and a pretreatment module: preprocessing alignment vehicle vibration data, taking monitoring data of one train as a standardThe signals, the initial time and train speed of other subway vibration time sequences are adjusted through translation and scaling in the time domain, interpolation correction is carried out on the data which are misplaced after adjustment, and the result with the maximum correlation with the standard signals is selected as an alignment result to form a track monitoring data matrix
The execution module: based on the track monitoring data matrix X and the dimension M of the potential space z, an EM algorithm is executed, and uncertain parameters { v, alpha, sigma 2, mu ] are obtained by initial value iteration 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D The most probable value of sigma;
and a conversion module: converting the continuous vector v obtained in the execution module into a binary vector o e {0,1} D The global sparse mode of the characteristic is characterized, and the selected related variable in the data matrix X corresponds to a non-zero term in the sparse mode vector o;
and a judging module: and judging whether each piece of monitoring data corresponds to the damage state or the lossless state by using the binary element of the vector o obtained in the conversion module.
The application provides electronic equipment, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the urban rail damage identification method based on Bayesian global sparse probability principal component analysis when executing the computer program.
The application provides a computer readable storage medium for storing computer instructions which when executed by a processor realize the steps of the urban rail damage identification method based on Bayesian global sparse probability principal component analysis.
The application has the beneficial effects that:
1. compared with the traditional track state detection method, the structure health monitoring method based on vibration has the advantages of global property, real-time property and the like, and can timely discover and effectively detect hidden or unobvious damage;
2. compared with a supervised learning method, the method does not require marking the data, and avoids the problem of high marking cost of the damage diagnosis data;
3. the application creatively realizes urban orbit damage identification based on the sparse mode vector obtained by the Bayesian global sparse probability principal component analysis method, and provides a new way for exploring and realizing full-time global orbit intelligent diagnosis and evaluation;
4. the application can process the problem of unsupervised feature selection of the high-dimensional structural health monitoring data, and is suitable for automatic identification of the abnormal track state based on a large amount of vibration data.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present application, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for identifying urban rail damage based on Bayesian global sparse probability principal component analysis;
FIG. 2 is a flowchart of a process for obtaining sparse mode vector o based on iterative optimization parameters of a variational EM algorithm according to the present application;
fig. 3 is a representative graph of measured data of average dynamic strain in urban rail transit area A, B according to an embodiment of the application: wherein (a) is pre-injury measurement data and (b) is post-injury measurement data;
fig. 4 is a diagram of representative alignment results of urban rail transit area A, B measurement data in an embodiment of the application: wherein (a) is a pre-injury measurement data alignment result and (b) is a post-injury measurement data alignment result;
fig. 5 is a confusion matrix diagram of the damage recognition result for the urban rail transit area A, B according to an embodiment of the present application: wherein (a) is a confusion matrix of the recognition result of the region A, and (B) is a confusion matrix of the recognition result of the region B.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
In the problem of damage identification, the characteristic change caused by the damage is not a global characteristic of the data matrix in consideration of the fact that track damage is unlikely to occur and is various. Therefore, the application adopts Bayesian global sparse probability principal component analysis to unsupervised robust mining of common characteristics of rail vibration monitoring data caused by each train, and eliminates abnormal data so as to realize unsupervised rail damage identification. Firstly preprocessing and aligning the vibration data of the passing vehicle, so that the global characteristics of the vibration data are more obvious, the analysis and the identification of principal components are facilitated, and then, the Bayesian global sparse probability principal component analysis is adopted to solve the global sparse mode, namely, which data are related to the global characteristics are selected. Because the global sparse mode is a damage sensitive index, the data related to the global features can be judged to be in a lossless state, and other data correspond to possible track damage, so that the aim of unsupervised track damage identification is fulfilled.
Referring to fig. 1-5, the application provides a method for identifying urban rail damage based on bayesian global sparse probability principal component analysis, which comprises the following steps:
firstly, arranging an optical fiber sensor on an urban rail, and carrying out health monitoring on measuring points along the way to obtain measurement data of average dynamic strain of each measuring point when the vehicle passes each time;
preprocessing alignment train passing vibration data, specifically, taking monitoring data of one train as a standard signal, adjusting the starting time and train speed of other subway vibration time sequences through translation and expansion and contraction in a time domain, performing interpolation correction on the data with dislocation after adjustment, and selecting the data with the largest correlation with the standard signalAs a result of the alignment, a track monitoring data matrix is formed
Step three, based on the track monitoring data matrix X and the dimension M of the potential space z thereof, executing an EM algorithm, and obtaining uncertain parameters { v, alpha, sigma) by initial value iteration 2 ，μ 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D The most probable value of sigma;
step four, converting the continuous vector v obtained in the step three into a binary vector o epsilon {0,1} D The global sparse mode characterizing features, in particular, the selected related variables (columns) in the data matrix X correspond to non-zero terms in the sparse mode vector o;
and fifthly, judging whether each piece of monitoring data corresponds to a damage state (the element value is 0) or a lossless state (the element value is 1) by utilizing the binary elements (0 and 1) of the vector o obtained in the step four.
The second step is specifically as follows:
step 2.1, selecting monitoring data of a train as standard signals;
step 2.2, selecting a proper interpolation method, such as cubic spline interpolation, for other subway vibration time sequences, and specifically adopting piecewise linear interpolation to continuously acquire discrete monitoring data;
step 2.3, carrying out translation and scaling on the continuous monitoring data in a reasonable range in a time domain, wherein the range is selected to be as small as possible while ensuring that the monitoring data can be aligned with a standard signal;
step 2.4, discretizing the continuous monitoring data subjected to translation and scaling according to the time data of the standard signal, and calculating the correlation between the continuous monitoring data and the standard signal;
step 2.5, traversing the translation and scaling ranges at small intervals, and selecting a translation scaling result with the largest correlation with the standard signal as an alignment result; the selection method can adopt small-interval traversal selection or optimization algorithms such as genetic algorithm and the like;
step 2.6Taking common parts in alignment result time domain to form track monitoring data matrix
The third step is specifically as follows:
the process of obtaining the most probable value of the uncertain parameter in the third step is based on a variational EM algorithm, and the hidden variable { mu ] of the control variational distribution q (Z, P) is iteratively updated in turn according to the track monitoring data matrix X and the dimension M of the data potential space Z 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D Sigma and uncertainty parameter θ= { v, α, σ 2 Zero, realize the free energy of variation A process of minimization in which H [ q (Z, P)]Differential entropy representing the variation distribution q (Z, P); the process specifically comprises the following steps:
step 3.1, initializing uncertainty parameters { v, α, σ } 2 ，μ 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D ，∑}；
Step 3.2, calculating E, updating the variation distribution q (Z, P) =q (Z) q (P), wherein The calculation process is, for all i e {1,..N } and k e {1,..D }, calculated according to the following formula:
wherein m= (M 1 ，...，m D ) T ，
And 3.3, calculating M steps through the following formula, and updating the uncertain parameter theta:
for all k e {1,..d },
step 3.4, repeating steps 3.2 to 3.3 until free energyThe variation of (a) is smaller than a set threshold (e.g. 10 -6 ) The most probable value of each uncertain parameter is obtained.
The step 3.1 specifically comprises the following steps:
step 3.1.1, initializing each element in the vector v to 1;
step 3.1.2, taking the parameter alpha as 0.1, 1 and 10 respectively;
step 3.1.3,Parameter { mu } 1 ，...，μ N ，m 1 ，...，m D Initializing to singular vectors of a data matrix X;
step 3.1.4, parameter { S } 1 ，...，S D Sigma is initialized to Sigma=I M ，S 1 ＝...＝S D ＝α -2 I M ；
And 3.1.5, respectively performing several (such as 5 times, not more than 5 times is recommended) preliminary iterations aiming at different initialization schemes formed by three values of the parameter alpha, and selecting a scheme with the lowest free energy of the iteration result as initialization setting of each parameter.
The fourth step is specifically as follows:
step 4.1, taking binary vector o for all k e {1,., D } (k) Corresponding position elements of the first k largest elements in the corresponding continuous vector v are 1, and the other elements are 0;
step 4.2, evaluate each binary vector o (k) Is defined by the marginal likelihood function p (x|o, alpha, sigma) 2 )＝∫p(x|P，o，σ 2 ) P (P|alpha) dP, where
Step 4.3, selecting the binary vector o with the maximum marginal likelihood function (k) As binary conversion result o of the successive vector v.
The application also provides a system for identifying urban rail damage based on Bayesian global sparse probability principal component analysis, which comprises:
and a data acquisition module: arranging optical fiber sensors on the urban rail, and carrying out health monitoring on the along-road measuring points to obtain measurement data of average dynamic strain of each measuring point when each measuring point passes through the vehicle;
and a pretreatment module: preprocessing alignment train passing vibration data, taking monitoring data of one train as a standard signal, and adjusting the starting of other subway vibration time sequences through translation and scaling in the time domainThe initial time and the train speed, interpolation correction is carried out on the data which are misplaced after adjustment, and the result with the maximum relativity with the standard signal is selected as the alignment result to form a track monitoring data matrix
The execution module: based on the track monitoring data matrix X and the dimension M of the potential space z, an EM algorithm is executed, and uncertain parameters { v, alpha, sigma } are obtained by initial value iteration 2 ，μ 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D The most probable value of sigma;
and a conversion module: converting the continuous vector v obtained in the execution module into a binary vector o e {0,1} D The global sparse mode of the characteristic is characterized, and the selected related variable in the data matrix X corresponds to a non-zero term in the sparse mode vector o;
and a judging module: and judging whether each piece of monitoring data corresponds to the damage state or the lossless state by using the binary element of the vector o obtained in the conversion module.
According to the urban rail damage identification method and system based on Bayesian global sparse probability principal component analysis, urban rail damage identification is realized by utilizing Bayesian global sparse probability principal component analysis based on non-global property of damage characteristics, high-cost damage diagnosis data labeling is avoided, and a new approach is provided for exploring and realizing full-time global rail intelligent diagnosis and evaluation. As a track state detection method, the method has the advantages of global property, real-time property and the like, and can timely find and effectively detect hidden or unobvious damage, thereby better serving the field of urban track operation and maintenance.
Examples
Referring to fig. 3-5, the urban railway unsupervised damage identification method based on Bayesian global sparse probability principal component analysis is utilized to carry out urban railway unsupervised damage identification aiming at crack damage of a ballast bed plate on a certain section of 60kg/m urban railway in China. This example shows the lesion recognition results for each of 115 pre-lesion and 15 post-lesion measurement data for two 5 meter long regions (denoted as region a and region B) adjacent to the lesion.
The method for identifying the urban rail damage based on Bayesian global sparse probability principal component analysis is used for identifying the unsupervised damage of the urban rail:
the first step is specifically as follows: an optical fiber sensor is arranged on a certain urban rail, an optical cable is arranged on the surface of a track bed plate, each measuring area along the way is 5 meters long, the average dynamic strain in a train passing area is acquired according to the phase difference change of the sensor when a train passes through, the sampling frequency is 1000Hz, and representative measuring data of an area A, B are shown in fig. 3;
the second step is specifically as follows: taking the monitoring data of the first train as a standard signal, setting a translation range to be within 2 seconds of the midpoint of other monitoring data and the midpoint of the standard signal, and setting a scaling range to be a section [0.8,1.05 ]]Selecting an interpolation result with the largest correlation with the standard signal as an alignment result to form a track monitoring data matrixWhere n=6321, d=130, representative alignment results for the regions A, B are shown in fig. 4;
the third step is specifically as follows: inputting a track monitoring data matrix X, setting the dimension M=10 of the potential space z, setting the iteration number of the initialization scheme in the step 3.1.5 to be 5, and setting the free energy change threshold in the step 3.4 to be 10 -6 Executing EM algorithm to obtain uncertain parameters { v, alpha, sigma } by initial value iteration 2 ，μ 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D The most probable value of sigma;
the fourth step is specifically as follows: based on the continuous vector v obtained in the step three, a binary sparse mode vector o of a detection area A is obtained through conversion A ＝[0 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 0 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 1 1 1 1 0 1 0 0 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 1 1 1 1 1 1 0 1 1 0 1 1 1 1 1 1 0 1 1 1 1 1 1 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]And a binary sparse mode vector o for region B B ＝[0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 0 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 0 1 0 0 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 0 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]The first 115 and the last 15 elements of each sparse mode vector respectively correspond to the measured data before and after the 15 injuries of the 115 areas;
the fifth step is specifically as follows: using the vector o obtained in step four A O B The binary elements (0 and 1) in the data are used for judging whether each piece of monitoring data corresponds to a damage state (the element value is 0) or a lossless state (the element value is 1), and the confusion matrix of the judgment result is shown in fig. 5.
In the above embodiment, although a small amount of pre-injury data is identified as injury, the accuracy of identifying the post-injury data is 100%, so that rail injury can be effectively identified and missed judgment can be avoided, and the method is suitable for practical engineering application.
The application provides electronic equipment, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the urban rail damage identification method based on Bayesian global sparse probability principal component analysis when executing the computer program.
The application provides a computer readable storage medium for storing computer instructions which when executed by a processor realize the steps of the urban rail damage identification method based on Bayesian global sparse probability principal component analysis.
The memory in embodiments of the present application may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The nonvolatile memory may be a Read Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an electrically Erasable EPROM (EEPROM), or a flash memory. The volatile memory may be random access memory (random access memory, RAM) which acts as an external cache. By way of example, and not limitation, many forms of RAM are available, such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous DRAM (SLDRAM), and direct memory bus RAM (DR RAM). It should be noted that the memory of the methods described herein is intended to comprise, without being limited to, these and any other suitable types of memory.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the processes or functions described in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a high-density digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a Solid State Disk (SSD)), or the like.
In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The steps of a method disclosed in connection with the embodiments of the present application may be embodied directly in a hardware processor for execution, or in a combination of hardware and software modules in the processor for execution. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method. To avoid repetition, a detailed description is not provided herein.
It should be noted that the processor in the embodiments of the present application may be an integrated circuit chip with signal processing capability. In implementation, the steps of the above method embodiments may be implemented by integrated logic circuits of hardware in a processor or instructions in software form. The processor may be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, or discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method.
The urban rail damage recognition method based on Bayesian global sparse probability principal component analysis provided by the application is described in detail, and specific examples are applied to explain the principle and the implementation mode of the application, and the description of the above examples is only used for helping to understand the method and the core idea of the application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.
Claims (6)
1. The urban orbit damage identification method based on Bayesian global sparse probability principal component analysis is characterized by comprising the following steps of: the method comprises the following steps:
firstly, arranging an optical fiber sensor on an urban rail, and carrying out health monitoring on measuring points along the way to obtain measurement data of average dynamic strain of each measuring point when the vehicle passes each time;
preprocessing aligned train passing vibration data, taking one train of monitoring data as a standard signal, adjusting the starting time and train speed of other subway vibration time sequences through translation and scaling in a time domain, performing interpolation correction on the data with dislocation after adjustment, and selecting a result with the greatest correlation with the standard signal as an alignment result to form a track monitoring data matrix
Step three, based on the track monitoring data matrix X and the dimension M of the potential space Z thereof, executing an EM algorithm, and obtaining uncertain parameters { v, alpha, sigma) by initial value iteration 2 ，μ 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D The most probable value of sigma;
step four, converting the continuous vector v obtained in the step three into a binary vector o epsilon {0,1} D The global sparse mode of the characteristic is characterized, and the selected related variable in the data matrix X corresponds to a non-zero term in the sparse mode vector o;
step five, judging whether each piece of monitoring data corresponds to a damage state or a nondestructive state by utilizing the binary element of the vector o obtained in the step four;
the process of obtaining the most probable value of the uncertain parameter in the third step is based on a variational EM algorithm, and the hidden variable { mu ] of the control variational distribution q (Z, P) is iteratively updated in turn according to the track monitoring data matrix X and the dimension M of the data potential space Z 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D Sigma and uncertainty parameter θ= { v, α, σ 2 Zero, realize the free energy of variationA process of minimization in which H [ q (Z, P)]Differential entropy representing the variation distribution q (Z, P);
the process specifically comprises the following steps:
step 3.1, initializing uncertainty parameters { v, α, σ } 2 ，μ 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D ，∑}；
Step 3.2, calculating E, updating the variation distribution q (Z, P) =q (Z) q (P), wherein The calculation process is for all i e { 1..N } and k e { 1..D } calculated according to the following formula:
wherein m= (M 1 ，...，m D ) T ，
And 3.3, calculating M steps through the following formula, and updating the uncertain parameter theta:
for all k e {1,..d },
step 3.4, repeating steps 3.2 to 3.3 until free energyThe variation of (2) is smaller than a set threshold value, and the most possible value of each uncertain parameter is obtained;
the step 3.1 specifically comprises the following steps:
step 3.1.1, initializing each element in the vector v to 1;
step 3.1.2, taking the parameter alpha as 0.1, 1 and 10 respectively;
step 3.1.3 parameter { μ } 1 ，...，μ N ，m 1 ，...，m D Initializing to singular vectors of data matrix X；
Step 3.1.4, parameter { S } 1 ，...，S D Sigma is initialized to Sigma=I M ，S 1 ＝…＝S D ＝α -2 I M ；
Step 3.1.5, respectively performing a plurality of preliminary iterations for different initialization schemes formed by three values of the parameter alpha, and selecting a scheme with the lowest free energy of the iteration result as the initialization setting of each parameter;
the fourth step is specifically as follows:
step 4.1, taking binary vector o for all k e {1,., D } (k) Corresponding position elements of the first k largest elements in the corresponding continuous vector v are 1, and the other elements are 0;
step 4.2, evaluate each binary vector o (k) Is defined by the marginal likelihood function p (x|o, alpha, sigma) 2 )＝∫p(x|P，o，σ 2 ) P (P|alpha) dP, where
Step 4.3, selecting the binary vector o with the maximum marginal likelihood function (k) As binary conversion result o of the successive vector v.
2. The method according to claim 1, characterized in that: the second step is specifically as follows:
step 2.1, selecting monitoring data of a train as standard signals;
step 2.2, selecting a proper interpolation method for other subway vibration time sequences, and serializing discrete monitoring data;
2.3, carrying out translation and scaling on the continuous monitoring data in a reasonable range in a time domain, wherein the range selection is as small as possible while ensuring that the monitoring data can be aligned with a standard signal;
step 2.4, discretizing the continuous monitoring data subjected to translation and scaling according to the time data of the standard signal, and calculating the correlation between the continuous monitoring data and the standard signal;
step 2.5, selecting a translation scaling result with the maximum correlation with the standard signal as an alignment result;
step 2.6, the common part on the time domain of the alignment result is taken to form a track monitoring data matrix
3. The method according to claim 1, characterized in that: the binary elements in the fifth step comprise 0 and 1, wherein the element value of the damaged state is 0, and the element value of the lossless state is 1.
4. The recognition system of the urban rail damage recognition method based on the Bayesian global sparse probability principal component analysis as claimed in claim 1, wherein the recognition system is characterized by: the system comprises:
and a data acquisition module: arranging optical fiber sensors on the urban rail, and carrying out health monitoring on the along-road measuring points to obtain measurement data of average dynamic strain of each measuring point when each measuring point passes through the vehicle;
and a pretreatment module: preprocessing aligned train passing vibration data, taking one train of monitoring data as a standard signal, adjusting the starting time and train speed of other subway vibration time sequences through translation and expansion and contraction in a time domain, performing interpolation correction on the data which are misplaced after adjustment, selecting a result with the largest correlation with the standard signal as an alignment result, and forming a track monitoring data matrix
The execution module: based on the track monitoring data matrix X and the dimension M of the potential space Z, an EM algorithm is executed, and uncertain parameters { v, alpha, sigma ] are obtained by initial value iteration 2 ，μ 1 ，...，μ N ，m 1 ，...，m D ，S 1 ，...，S D Most probable value of }, sigma；
And a conversion module: converting the continuous vector v obtained in the execution module into a binary vector o e {0,1} D The global sparse mode of the characteristic is characterized, and the selected related variable in the data matrix X corresponds to a non-zero term in the sparse mode vector o;
and a judging module: and judging whether each piece of monitoring data corresponds to the damage state or the lossless state by using the binary element of the vector o obtained in the conversion module.
5. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1-3 when the computer program is executed.
6. A computer readable storage medium storing computer instructions which, when executed by a processor, implement the steps of the method of any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310129979.1A CN116165274B (en) | 2023-02-17 | 2023-02-17 | Urban orbit damage identification method based on Bayesian global sparse probability principal component analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310129979.1A CN116165274B (en) | 2023-02-17 | 2023-02-17 | Urban orbit damage identification method based on Bayesian global sparse probability principal component analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116165274A CN116165274A (en) | 2023-05-26 |
| CN116165274B true CN116165274B (en) | 2023-11-14 |
Family
ID=86419599
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310129979.1A Active CN116165274B (en) | 2023-02-17 | 2023-02-17 | Urban orbit damage identification method based on Bayesian global sparse probability principal component analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116165274B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117076992B (en) * | 2023-10-16 | 2024-01-19 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | Structural member damage detection method and system based on signal processing |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104634872A (en) * | 2015-01-10 | 2015-05-20 | 哈尔滨工业大学（威海） | Online high-speed railway steel rail damage monitoring method |
| WO2018122806A1 (en) * | 2016-12-30 | 2018-07-05 | 同济大学 | Travel time distribtuion-based multimodal traffic anomaly detection method |
| CN109086604A (en) * | 2018-07-05 | 2018-12-25 | 成都信息工程大学 | Android malicious act software identification method and system based on sparse Bayesian model |
| CN112528564A (en) * | 2020-12-08 | 2021-03-19 | 中国矿业大学 | Bridge structure damage identification method based on sparse Bayesian learning |
| CN112949131A (en) * | 2021-03-05 | 2021-06-11 | 哈尔滨工业大学 | Probability damage positioning vector method for continuous bridge cluster damage diagnosis |
| KR20210087875A (en) * | 2020-01-03 | 2021-07-13 | 삼성전자주식회사 | Method and apparatus for processing sparse points |
| CN113297790A (en) * | 2021-05-19 | 2021-08-24 | 哈尔滨工业大学 | High-speed rail response prediction method based on sparse Bayesian width learning |
| CA3135732A1 (en) * | 2020-10-23 | 2022-04-23 | 10353744 Canada Ltd. | Method and device for processing credit data |
| WO2022141213A1 (en) * | 2020-12-30 | 2022-07-07 | 中南大学 | Gene prediction method and system for fault of autonomous rail rapid transit vehicle in smart city |
| CN114896712A (en) * | 2021-12-07 | 2022-08-12 | 中铁五局集团建筑工程有限责任公司 | Structural damage identification method based on reference Bayesian principle and sparse regularization |
| CN115470838A (en) * | 2022-08-15 | 2022-12-13 | 浙江大学 | Robust dynamic variational Bayesian dictionary learning-based process monitoring method |
| CN115618209A (en) * | 2022-09-16 | 2023-01-17 | 哈尔滨工业大学 | Railway track state evaluation method based on sparse extreme learning machine and hypothesis testing |
| CN115700494A (en) * | 2022-09-16 | 2023-02-07 | 哈尔滨工业大学 | Rail transit monitoring data cleaning method and system based on Bayesian inference |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140272914A1 (en) * | 2013-03-15 | 2014-09-18 | William Marsh Rice University | Sparse Factor Analysis for Learning Analytics and Content Analytics |
| US20210065052A1 (en) * | 2019-08-26 | 2021-03-04 | Nvidia Corporation | Bayesian optimization of sparsity ratios in model compression |
-
2023
- 2023-02-17 CN CN202310129979.1A patent/CN116165274B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104634872A (en) * | 2015-01-10 | 2015-05-20 | 哈尔滨工业大学（威海） | Online high-speed railway steel rail damage monitoring method |
| WO2018122806A1 (en) * | 2016-12-30 | 2018-07-05 | 同济大学 | Travel time distribtuion-based multimodal traffic anomaly detection method |
| CN109086604A (en) * | 2018-07-05 | 2018-12-25 | 成都信息工程大学 | Android malicious act software identification method and system based on sparse Bayesian model |
| KR20210087875A (en) * | 2020-01-03 | 2021-07-13 | 삼성전자주식회사 | Method and apparatus for processing sparse points |
| CA3135732A1 (en) * | 2020-10-23 | 2022-04-23 | 10353744 Canada Ltd. | Method and device for processing credit data |
| CN112528564A (en) * | 2020-12-08 | 2021-03-19 | 中国矿业大学 | Bridge structure damage identification method based on sparse Bayesian learning |
| WO2022141213A1 (en) * | 2020-12-30 | 2022-07-07 | 中南大学 | Gene prediction method and system for fault of autonomous rail rapid transit vehicle in smart city |
| CN112949131A (en) * | 2021-03-05 | 2021-06-11 | 哈尔滨工业大学 | Probability damage positioning vector method for continuous bridge cluster damage diagnosis |
| CN113297790A (en) * | 2021-05-19 | 2021-08-24 | 哈尔滨工业大学 | High-speed rail response prediction method based on sparse Bayesian width learning |
| CN114896712A (en) * | 2021-12-07 | 2022-08-12 | 中铁五局集团建筑工程有限责任公司 | Structural damage identification method based on reference Bayesian principle and sparse regularization |
| CN115470838A (en) * | 2022-08-15 | 2022-12-13 | 浙江大学 | Robust dynamic variational Bayesian dictionary learning-based process monitoring method |
| CN115618209A (en) * | 2022-09-16 | 2023-01-17 | 哈尔滨工业大学 | Railway track state evaluation method based on sparse extreme learning machine and hypothesis testing |
| CN115700494A (en) * | 2022-09-16 | 2023-02-07 | 哈尔滨工业大学 | Rail transit monitoring data cleaning method and system based on Bayesian inference |
Non-Patent Citations (2)
| Title |
|---|
| A Hybrid Optimization Algorithm with Bayesian Inference for Probabilistic Model Updating;Hao Sun等;《Computer-Aided Civil and Infrastructure Engineering》;全文 * |
| 基于概率主成分分析的结构健康监测数据修复方法研究;马帜等;《振动与冲击》;第40卷(第21期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116165274A (en) | 2023-05-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111094956B (en) | Processing thermal imaging images with neural networks to identify subsurface erosion on insulation (CUI) | |
| CN110909046B (en) | Time-series abnormality detection method and device, electronic equipment and storage medium | |
| CN106950276B (en) | Pipeline defect depth inversion method based on convolutional neural network | |
| CN116165274B (en) | Urban orbit damage identification method based on Bayesian global sparse probability principal component analysis | |
| CN113009447B (en) | Road underground cavity detection and early warning method based on deep learning and ground penetrating radar | |
| CN109752744B (en) | Multi-satellite combined orbit determination method based on model error compensation | |
| CN116050571A (en) | Water quality prediction method based on transducer sparse attention mechanism | |
| CN113343402B (en) | Pipeline corrosion grade evaluation method based on multilayer convolution sparse coding | |
| CN113670616B (en) | Bearing performance degradation state detection method and system | |
| Jiang et al. | Anomaly detection of industrial multi-sensor signals based on enhanced spatiotemporal features | |
| Jiang et al. | THMS-Net: A two-stage heterogeneous signals mutual supervision network for MFL weak defect detection | |
| CN115343703A (en) | Pipeline identification method of 3D-CNN ground penetrating radar three-dimensional image based on self-training | |
| CN115700494B (en) | Rail transit monitoring data cleaning method and system based on Bayesian reasoning | |
| CN115618209B (en) | Railway track state evaluation method based on sparse extreme learning machine and hypothesis test | |
| CN113468804B (en) | Underground pipeline identification method based on matrix bundles and deep neural network | |
| WO2021219259A1 (en) | System, method and computer program for determining estimated sensor data | |
| CN112033406A (en) | Navigation method, device and storage medium based on lightweight network | |
| CN118224552A (en) | Outdoor pipeline heat preservation effect monitoring system in job site | |
| CN117590476A (en) | Transient electromagnetic scanning detection system and method | |
| CN112348076B (en) | Multi-sensor data fusion method suitable for road information acquisition | |
| Bian et al. | Disturbances prediction of bit error rate for high-speed railway Balise transmission through persistent state mapping | |
| CN116304789A (en) | VP inclinometer fault diagnosis method and device | |
| CN115980697A (en) | Method for inverting boundary layer height by using laser radar under different weather conditions | |
| CN108482428A (en) | A kind of train locating method based on curvature array correlation | |
| CN115963788A (en) | Multi-sampling-rate industrial process key quality index online prediction method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |