CN110287827A - A kind of bridge strain data outliers recognition methods based on data correlation - Google Patents
A kind of bridge strain data outliers recognition methods based on data correlation Download PDFInfo
- Publication number
- CN110287827A CN110287827A CN201910500098.XA CN201910500098A CN110287827A CN 110287827 A CN110287827 A CN 110287827A CN 201910500098 A CN201910500098 A CN 201910500098A CN 110287827 A CN110287827 A CN 110287827A
- Authority
- CN
- China
- Prior art keywords
- data
- strain
- difference
- acceleration
- potential
- 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.)
- Granted
Links
Classifications
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Algebra (AREA)
- Evolutionary Biology (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The bridge strain data outliers recognition methods based on data correlation that the invention discloses a kind of, comprising: obtain the sensor raw data for being installed on same position；Long period trend data is extracted to original strain data by wavelet decomposition, obtains the difference of original strain data Yu long period trend data；According to the mean value and standard deviation of strain difference data, the variation range of potential exceptional value in original strain data is obtained；Long period trend data is extracted to raw acceleration data by wavelet decomposition, obtains the difference of raw acceleration data Yu long period trend data；According to the mean value and standard deviation of acceleration difference data, the variation range of potential exceptional value in raw acceleration data is obtained；The appearance position for comparing potential exceptional value in the appearance position and raw acceleration data of potential exceptional value in original strain data judges to identify the real exceptional value of original strain data.The present invention precisely can efficiently handle strain data outlier problem in bridge health monitoring.
Description
Technical field
The invention belongs to bridge health monitoring data analysis and research fields, and in particular to a kind of bridge based on data correlation
Beam strain data outlier identification method.
Background technique
Domestic and abroad bridge health monitoring technique application gradually mature, large bridge structural healthy monitoring system function also by
Gradually improve, but due to the increasingly increase of its integral link and influence factor, break down it is potentially possible be also gradually increased, bridge
Contradiction between the chronicity of health monitoring task and the finite lifetime of monitoring device is more prominent, such as the failure of individual part
Chain reaction can often be caused, cause whole system that cannot run and even paralyse.Especially, prison caused by monitoring system faults itself
If measured data distortion cannot be found in time, the accuracy of security assessment result will affect, so that bridge structural health monitoring
Task cannot smoothly complete.It is more than 80% in existing highway bridge health monitoring systems according to the research of Houser and Aktan et al.
False alarm is all as caused by data distortion[1].Therefore, it is necessary to be cleaned to monitoring data, to realize sensing data
Value, improve the availability and efficiency of bridge health monitoring system.
Strain data exceptional value mainly due to power failure, instrument or other make signal generate momentary fluctuation, formed pulsation
It interferes and causes, be mainly shown as that signal fluctuation amplitude is excessive.But strain data itself fluctuates larger, will cause when vehicle passes through
Biggish signal fluctuation, such signal fluctuation and data outliers are difficult to differentiate from single signal source.Therefore, for above-mentioned number
The shortcomings that according to exceptional value, the method cleaned in real time there is an urgent need to develop a kind of pair of bridge monitoring dynamic strain data outliers.
Bibliography:
[1] He Dongdong, Zhang Li, Zhang Huili, Li Xianke " 3 σ criterion " are in the inspection of bridge health monitoring data singular value
Using the China and foreign countries [J] highway, 2013,33 (6): 107-110.
Summary of the invention
It is a kind of based on data correlation the technical problem to be solved by the present invention is to provide in view of the above shortcomings of the prior art
The bridge strain data outliers recognition methods of property, based on the bridge strain data outliers recognition methods energy of data correlation
Precisely efficiently handle strain data outlier problem in bridge health monitoring.
To realize the above-mentioned technical purpose, the technical scheme adopted by the invention is as follows:
A kind of bridge strain data outliers recognition methods based on data correlation, comprising the following steps:
Step 1, acquisition are installed on the sensor raw data of same position；
Step 2, by wavelet decomposition to original strain data extract long period trend data, obtain original strain data with
The difference of long period trend data；
The mean value and standard deviation of step 3, the strain difference data obtained according to step 2 obtain diving in original strain data
In the variation range of exceptional value；
Step 4 extracts long period trend data to raw acceleration data by wavelet decomposition, obtains original acceleration number
According to the difference with long period trend data；
The mean value and standard deviation of step 5, the acceleration difference data obtained according to step 4, obtain raw acceleration data
In potential exceptional value variation range；
It is potential different in the appearance position and raw acceleration data of potential exceptional value in step 6, the original strain data of comparison
The appearance position of constant value judges to identify the real exceptional value of original strain data.
Technical solution as a further improvement of that present invention, the specific steps of the step 1 are as follows:
Obtain the original strain data { x of the strain transducer acquisition at health monitoring systems same positioni, i=1,
2 ..., the raw acceleration data { y of n and acceleration transducer acquisitioni, i=1,2 ..., n, wherein n indicates data amount check.
Technical solution as a further improvement of that present invention, the specific steps of the step 2 are as follows:
Step 2.1, the long period trend data { s that original strain data is extracted using wavelet decompositioni, i=1,2 ..., n；
Step 2.2 calculates xiWith siDifference analyzed, be denoted as di=xi-si。
Technical solution as a further improvement of that present invention, the specific steps of the step 3 are as follows:
Step 3.1 calculates strain difference data { di, i=1,2 ..., the mean μ and standard deviation sigma of n；
Wherein:
Step 3.2, according to method of interval estimation, calculate the fluctuation range (μ-a σ, μ+a σ) of the normal value of strain difference,
Middle a is the threshold range coefficient of setting；
Step 3.3, according to strain difference normal value confidence interval, search strain difference data { di, i=1,
The data positional information of 2 ..., n other than confidence interval (μ-a σ, μ+a σ), note: { di, i=1,2 ..., the data of n < μ-a σ
Position sequence isWherein m1It is less than the data amount check of lower tantile for strain difference；{di, i=1,
2 ..., the Data Position sequence of n > μ+a σ isWherein m2It is greater than the number of upper tantile for strain difference
According to number；ThenFor the potential abnormal point in original strain data sequence
Location sets.
Technical solution as a further improvement of that present invention, the specific steps of the step 4 are as follows:
Step 4.1, the long period trend data { s that raw acceleration data is extracted using wavelet decompositioni', i=1,
2,…,n；
Step 4.2 calculates yiWith si' difference analyzed, be denoted as di'=yi-si'。
Technical solution as a further improvement of that present invention, the specific steps of the step 5 are as follows:
Step 5.1 calculates acceleration difference data { di', i=1,2 ..., the mean μ of n ' and standard deviation sigma '；
Wherein:
Step 5.2, according to method of interval estimation, calculate fluctuation range (μ '-the b σ ', μ '+b of the normal value of acceleration difference
σ '), wherein b is the threshold range coefficient of setting；
Step 5.3, the confidence interval according to the normal value of acceleration difference, search acceleration difference data { di', i=
The data positional information of 1,2 ..., n other than confidence interval (μ '-b σ ', μ '+b σ '), note: { di', i=1,2 ..., n < μ '-
The Data Position sequence of b σ ' isWherein n1It is less than the data amount check of lower tantile for acceleration difference；
{di', i=1,2 ..., the Data Position sequence of n > μ '+b σ ' isWherein m2It is big for acceleration difference
In the data amount check of upper tantile；ThenFor raw acceleration data sequence
Potential the abnormity point position set in column.
Technical solution as a further improvement of that present invention, the specific steps of the step 6 are as follows:
The element in the potential the abnormity point position set in original strain data sequence is traversed, if the element also adds original
In potential the abnormity point position set in speed data sequence, then the corresponding original strain data of the element is normal value, otherwise,
The corresponding original strain data of the element is exceptional value.
The invention has the benefit that the present invention is difficult in data mapping for dynamic strain data outliers and by vehicle
Signal fluctuation caused by load separates, and the comprehensive multi-data source of the present invention carries out information fusion, proposes based on data correlation
Strain data outlier identification method, has the advantages that
(1) exceptional value in dynamic strain data can be recognized accurately in the present invention, and erroneous judgement is few；
(2) present invention can carry out real time implementation to dynamic strain data, automation is cleaned, and have extensive engineer application valence
Value.
Detailed description of the invention
Fig. 1 is the one-dimensional time-histories figure of martyr river bridge strain transducer second channel data on April 9th, 2017.
Fig. 2 is the one-dimensional time-histories figure of martyr river bridge acceleration transducer second channel data on April 9th, 2017.
Fig. 3 is strain difference schematic diagram.
Fig. 4 is normal data and potential exceptional value figure in strain data.
Fig. 5 is acceleration difference schematic diagram.
Fig. 6 is normal data and potential exceptional value figure in acceleration information.
Fig. 7 is outlier identification figure in strain data.
Specific embodiment
A specific embodiment of the invention is further illustrated below according to Fig. 1 to Fig. 7:
A kind of bridge strain data outliers recognition methods based on data correlation, comprising the following steps:
1) sensor raw data for being installed on same position, is obtained.
Specific steps are as follows:
Obtain the original strain data { x of the strain transducer acquisition at health monitoring systems same positioni, i=1,
2 ..., the raw acceleration data { y of n and acceleration transducer acquisitioni, i=1,2 ..., n, wherein n indicates data amount check.
2) long period trend data, is extracted to original strain data by wavelet decomposition, obtains original strain data and length
The difference of cyclical trend data.
Specific steps are as follows:
A) the long period trend data { s of original strain data is extracted using wavelet decompositioni, i=1,2 ..., n；
B) x is calculatediWith siDifference analyzed, be denoted as di=xi-si。
3), according to the mean value and standard deviation of strain difference data, the variation of potential exceptional value in original strain data is obtained
Range.
Specific steps are as follows:
A) strain difference data { d is calculatedi, i=1,2 ..., the mean μ and standard deviation sigma of n；Wherein:
B) according to method of interval estimation, the fluctuation range (μ-a σ, μ+a σ) of the normal value of strain difference is calculated, wherein a is
The threshold range coefficient of setting；The fluctuation range for wherein straining the normal value of difference is also referred to as the normal fluctuation model for straining difference
It encloses；
C) according to the confidence interval of the normal value of strain difference, strain difference data { d is searchedi, i=1,2 ..., n is being set
Believe the data positional information other than section (μ-a σ, μ+a σ), note: { di, i=1,2 ..., the Data Position sequence of n < μ-a σ isWherein m1It is less than the data amount check of lower tantile for strain difference；{di, i=1,2 ..., n > μ+a σ
Data Position sequence beWherein m2It is greater than the data amount check of upper tantile for strain difference；ThenFor the potential the abnormity point position set in original strain data sequence.
The data that sensor acquires in the present embodiment are recorded according to time series, and there are the Data Position of formatting, the present embodiment
In position refer to Data Position when abnormal point occurs, i.e., when abnormal point occurs at the time of corresponding record.
4) long period trend data, is extracted to raw acceleration data by wavelet decomposition, obtains raw acceleration data
With the difference of long period trend data.
Specific steps are as follows:
A) the long period trend data { s of raw acceleration data is extracted using wavelet decompositioni', i=1,2 ..., n；
B) y is calculatediWith si' difference analyzed, be denoted as di'=yi-si'。
5), according to the mean value and standard deviation of acceleration difference data, potential exceptional value in raw acceleration data is obtained
Variation range.
Specific steps are as follows:
A) acceleration difference data { d is calculatedi', i=1,2 ..., the mean μ of n ' and standard deviation sigma '；Wherein:
B) according to method of interval estimation, the fluctuation range (μ '-b σ ', μ '+b σ ') of the normal value of acceleration difference is calculated,
Middle b is the threshold range coefficient of setting；Wherein the fluctuation range of the normal value of acceleration difference is also referred to as acceleration difference
Normal fluctuation range；
C) according to the confidence interval of the normal value of acceleration difference, acceleration difference data { d is searchedi', i=1,2 ...,
Data positional information of the n other than confidence interval (μ '-b σ ', μ '+b σ '), note: { di', i=1,2 ..., the number of n < μ '-b σ '
It is according to position sequenceWherein n1It is less than the data amount check of lower tantile for acceleration difference；{di', i=
1,2 ..., the Data Position sequence of n > μ '+b σ ' isWherein m2It is greater than upper quartile for acceleration difference
The data amount check of value；ThenIt is latent in raw acceleration data sequence
In the abnormity point position set.
It is potential different in the appearance position and raw acceleration data of potential exceptional value in step 6, the original strain data of comparison
The appearance position of constant value judges to identify the real exceptional value of original strain data.
Specific steps are as follows: judge whether potential exceptional value position is potential in raw acceleration data in original strain data
There is also identical the abnormity point positions for exceptional value position, it may be assumed that
The element in the potential the abnormity point position set P in original strain data sequence is traversed, is remembered: element p ∈ P, if p ∈
Q, then the corresponding original strain data of the element is normal value, and otherwise, the corresponding original strain data of the element is exceptional value.
Below by taking the bridge of martyr river as an example, illustrate specific implementation process of the invention:
Original strain data described in the present embodiment chooses strain transducer second channel in martyr river bridge on April 9th, 2017
Data, raw acceleration data chooses the acceleration transducer second channel data at same position, and sample frequency is
200Hz, totally 17280000 data, Fig. 1 illustrate the one-dimensional time-history curves of the original strain data, and it is original that Fig. 2 illustrates this
The one-dimensional time-history curves of acceleration information.Secondly, original strain data is divided into long period signal and high frequency letter using wavelet decomposition
Number, high frequency signal strains difference, and schematic diagram is as shown in Figure 3.Long period signal is mainly caused by temperature effect, with temperature
Spend positive correlation；High-frequency signal is mainly generated by vehicular load, but simultaneously also includes certain sensor problem itself
Or data outliers caused by external interference.
The mean value that strain difference is calculated is 0.0021, and standard deviation 1.4049, setting threshold range parameter is 8, is obtained
Normal fluctuation section to strain difference is (- 11.2372,11.2415), and then obtains the potential exceptional value of original strain data
It is 60.The distribution situation of the original potential exceptional value of strain data is as shown in Figure 4.
Then, wavelet decomposition is carried out to raw acceleration data, extracts its long period trend, obtains raw acceleration data
With the difference of long period trend data, as shown in Figure 5.Wherein, the mean value of acceleration difference is 0, standard deviation 6.8925, setting
Threshold range parameter is 8, and the normal fluctuation section for obtaining acceleration difference is (- 55.1397,55.1397), and then is obtained original
The potential exceptional value of acceleration information is 115.The distribution situation of the potential exceptional value of raw acceleration data is as shown in Figure 6.
Comparison strains the appearance position of potential exceptional value and the appearance position of the potential exceptional value of acceleration, judges to identify original
The real exceptional value of strain data.In the same position for finding original strain data and the potential jump point value of raw acceleration data
When, it is contemplated that there are the influences of certain response error and sensor for bridge structure itself, remember the potential jump point position of strain data
For pi, this section thinks in the potential jump point [p of acceleration informationi-2,piIt+2] is same position at position.If the potential jump of the two
The identical then corresponding strain data of point in point position is normal, does not deal with；Conversely, then the corresponding strain data of point is abnormal
Value.After comparison identification, 4 same positions, i.e. 4 normal datas are co-existed in, final original strain data exceptional value
Identification is as shown in Figure 7.
The above example shows that method proposed by the invention can effectively identify the exceptional value in strain data and for different
The erroneous judgement of constant value is few, can real time implementation, automation cleaning dynamic strain data in exceptional value, can be applied to Practical Project health
In monitoring data pretreatment.
Protection scope of the present invention includes but is not limited to embodiment of above, and protection scope of the present invention is with claims
Subject to, replacement, deformation, the improvement that those skilled in the art that any pair of this technology is made is readily apparent that each fall within of the invention
Protection scope.
Claims (7)
1. a kind of bridge strain data outliers recognition methods based on data correlation, which comprises the following steps:
Step 1, acquisition are installed on the sensor raw data of same position；
Step 2 extracts long period trend data to original strain data by wavelet decomposition, obtains original strain data and long week
The difference of phase trend data；
The mean value and standard deviation of step 3, the strain difference data obtained according to step 2, obtain potential different in original strain data
The variation range of constant value；
Step 4, by wavelet decomposition to raw acceleration data extract long period trend data, obtain raw acceleration data with
The difference of long period trend data；
The mean value and standard deviation of step 5, the acceleration difference data obtained according to step 4 obtain diving in raw acceleration data
In the variation range of exceptional value；
Potential exceptional value in the appearance position and raw acceleration data of potential exceptional value in step 6, the original strain data of comparison
Appearance position, judge to identify the real exceptional value of original strain data.
2. the bridge strain data outliers recognition methods based on data correlation, feature exist as described in claim 1
In the specific steps of the step 1 are as follows:
Obtain the original strain data { x of the strain transducer acquisition at health monitoring systems same positioni, i=1,2 ..., n and
Raw acceleration data { the y of acceleration transducer acquisitioni, i=1,2 ..., n, wherein n indicates data amount check.
3. the bridge strain data outliers recognition methods based on data correlation, feature exist as described in claim 1
In the specific steps of the step 2 are as follows:
Step 2.1, the long period trend data { s that original strain data is extracted using wavelet decompositioni, i=1,2 ..., n；
Step 2.2 calculates xiWith siDifference analyzed, be denoted as di=xi-si。
4. the bridge strain data outliers recognition methods based on data correlation, feature exist as described in claim 1
In the specific steps of the step 3 are as follows:
Step 3.1 calculates strain difference data { di, i=1,2 ..., the mean μ and standard deviation sigma of n；
Wherein:
Step 3.2, according to method of interval estimation, calculate the fluctuation range (μ-a σ, μ+a σ) of the normal value of strain difference, wherein a
For the threshold range coefficient of setting；
Step 3.3, according to strain difference normal value confidence interval, search strain difference data { di, i=1,2 ..., n exists
Data positional information other than confidence interval (μ-a σ, μ+a σ), note: { di, i=1,2 ..., the Data Position sequence of n < μ-a σ
ForWherein m1It is less than the data amount check of lower tantile for strain difference；{di, i=1,2 ..., n > μ+
The Data Position sequence of a σ isWherein m2It is greater than the data amount check of upper tantile for strain difference；ThenFor the potential the abnormity point position set in original strain data sequence.
5. the bridge strain data outliers recognition methods based on data correlation, feature exist as described in claim 1
In the specific steps of the step 4 are as follows:
Step 4.1, the long period trend data { s ' that raw acceleration data is extracted using wavelet decompositioni, i=1,2 ..., n；
Step 4.2 calculates yiWith s 'iDifference analyzed, be denoted as d 'i=yi-s′i。
6. the bridge strain data outliers recognition methods based on data correlation, feature exist as described in claim 1
In the specific steps of the step 5 are as follows:
Step 5.1 calculates acceleration difference data { d 'i, i=1,2 ..., the mean μ of n ' and standard deviation sigma '；
Wherein:
Step 5.2, according to method of interval estimation, calculate the fluctuation range (μ '-b σ ', μ '+b σ ') of the normal value of acceleration difference,
Wherein b is the threshold range coefficient of setting；
Step 5.3, the confidence interval according to the normal value of acceleration difference, search acceleration difference data { d 'i, i=1,
The data positional information of 2 ..., n other than confidence interval (μ '-b σ ', μ '+b σ '), note: { d 'i, i=1,2 ..., n < μ '-b
The Data Position sequence of σ ' isWherein n1It is less than the data amount check of lower tantile for acceleration difference；
{d′i, i=1,2 ..., the Data Position sequence of n > μ '+b σ ' isWherein m2It is big for acceleration difference
In the data amount check of upper tantile；ThenFor raw acceleration data sequence
Potential the abnormity point position set in column.
7. the bridge strain data outliers recognition methods based on data correlation, feature exist as described in claim 1
In the specific steps of the step 6 are as follows:
The element in the potential the abnormity point position set in original strain data sequence is traversed, if the element is also in original acceleration
In potential the abnormity point position set in data sequence, then the corresponding original strain data of the element is normal value, otherwise, this yuan
The corresponding original strain data of element is exceptional value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910500098.XA CN110287827B (en) | 2019-06-11 | 2019-06-11 | Bridge strain data outlier identification method based on data correlation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910500098.XA CN110287827B (en) | 2019-06-11 | 2019-06-11 | Bridge strain data outlier identification method based on data correlation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110287827A true CN110287827A (en) | 2019-09-27 |
| CN110287827B CN110287827B (en) | 2023-09-26 |
Family
ID=68003602
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910500098.XA Active CN110287827B (en) | 2019-06-11 | 2019-06-11 | Bridge strain data outlier identification method based on data correlation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110287827B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111189488A (en) * | 2019-12-13 | 2020-05-22 | 精英数智科技股份有限公司 | Sensor value abnormity identification method, device, equipment and storage medium |
| CN111707782A (en) * | 2020-03-30 | 2020-09-25 | 江苏方天电力技术有限公司 | Thermal power generating unit carbon dioxide emission concentration abnormity detection method based on oxygen amount |
| CN113866455A (en) * | 2021-09-30 | 2021-12-31 | 中铁桥隧技术有限公司 | Bridge acceleration monitoring data anomaly detection method, system and device based on deep learning |
| CN116308305A (en) * | 2023-05-25 | 2023-06-23 | 安徽省云鹏工程项目管理有限公司 | Bridge health monitoring data management system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102661716A (en) * | 2012-04-20 | 2012-09-12 | 武汉理工大学 | Method and system for detecting linearity and rigidity of bridges and tunnels on basis of fiber-optic gyroscope technology |
| US20140258533A1 (en) * | 2013-03-06 | 2014-09-11 | Vmware, Inc. | Method and System for Providing a Roaming Remote Desktop |
| CN105760934A (en) * | 2016-03-02 | 2016-07-13 | 浙江工业大学 | Bridge abnormity monitoring restoration method based on wavelet and BP neural network |
| CN107230150A (en) * | 2017-07-19 | 2017-10-03 | 西南大学 | A kind of method for detecting abnormality in stock market based on Non-negative Matrix Factorization |
| CN108170925A (en) * | 2017-12-21 | 2018-06-15 | 中国铁道科学研究院铁道建筑研究所 | A kind of bridge damnification method for early warning based on arma modeling |
-
2019
- 2019-06-11 CN CN201910500098.XA patent/CN110287827B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102661716A (en) * | 2012-04-20 | 2012-09-12 | 武汉理工大学 | Method and system for detecting linearity and rigidity of bridges and tunnels on basis of fiber-optic gyroscope technology |
| US20140258533A1 (en) * | 2013-03-06 | 2014-09-11 | Vmware, Inc. | Method and System for Providing a Roaming Remote Desktop |
| CN105760934A (en) * | 2016-03-02 | 2016-07-13 | 浙江工业大学 | Bridge abnormity monitoring restoration method based on wavelet and BP neural network |
| CN107230150A (en) * | 2017-07-19 | 2017-10-03 | 西南大学 | A kind of method for detecting abnormality in stock market based on Non-negative Matrix Factorization |
| CN108170925A (en) * | 2017-12-21 | 2018-06-15 | 中国铁道科学研究院铁道建筑研究所 | A kind of bridge damnification method for early warning based on arma modeling |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111189488A (en) * | 2019-12-13 | 2020-05-22 | 精英数智科技股份有限公司 | Sensor value abnormity identification method, device, equipment and storage medium |
| CN111189488B (en) * | 2019-12-13 | 2020-12-04 | 精英数智科技股份有限公司 | Sensor value abnormity identification method, device, equipment and storage medium |
| CN111707782A (en) * | 2020-03-30 | 2020-09-25 | 江苏方天电力技术有限公司 | Thermal power generating unit carbon dioxide emission concentration abnormity detection method based on oxygen amount |
| CN113866455A (en) * | 2021-09-30 | 2021-12-31 | 中铁桥隧技术有限公司 | Bridge acceleration monitoring data anomaly detection method, system and device based on deep learning |
| CN116308305A (en) * | 2023-05-25 | 2023-06-23 | 安徽省云鹏工程项目管理有限公司 | Bridge health monitoring data management system |
| CN116308305B (en) * | 2023-05-25 | 2023-08-08 | 安徽省云鹏工程项目管理有限公司 | Bridge health monitoring data management system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110287827B (en) | 2023-09-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110287827A (en) | A kind of bridge strain data outliers recognition methods based on data correlation | |
| KR101316486B1 (en) | Error detection method and system | |
| CN109974782A (en) | Equipment fault early-warning method and system based on big data sensitive features optimum option | |
| CN103499382B (en) | A kind ofly to merge and the Diagnosis Method of Transformer Faults of image recognition based on vibration data | |
| CN109375010A (en) | Winding in Power Transformer method for diagnosing faults based on vibration distribution characteristic model | |
| CN106568557B (en) | A kind of safe early warning method of high-speed railway bridge vehicle-bridge vibration performance | |
| CN109389325B (en) | Method for evaluating state of electronic transformer of transformer substation based on wavelet neural network | |
| Son et al. | Deep learning-based anomaly detection to classify inaccurate data and damaged condition of a cable-stayed bridge | |
| CN103267652B (en) | Intelligent online diagnosis method for early failures of equipment | |
| CN117538710B (en) | Intelligent early warning method and system for local dynamic discharge monitoring | |
| CN105626502A (en) | Plunger pump health assessment method based on wavelet packet and Laplacian Eigenmap | |
| CN104008294A (en) | System and method for detecting abnormality of bearing | |
| CN116520068B (en) | Diagnostic method, device, equipment and storage medium for electric power data | |
| CN117076955A (en) | Fault detection method and system for high-voltage frequency converter | |
| CN118091234B (en) | Current transformer for fault diagnosis processing | |
| Zhu et al. | Cost-effective bad synchrophasor data detection based on unsupervised time-series data analytic | |
| CN116012681A (en) | Method and system for diagnosing motor faults of pipeline robot based on sound vibration signal fusion | |
| CN117786349A (en) | Equipment fault early warning system and method based on data mining technology | |
| CN117630797A (en) | Ammeter health state detection method, system and storage medium based on working current | |
| CN102778632A (en) | Double normalization recognition method for directly forecasting and recognizing transformer winding fault type | |
| CN115310499B (en) | Industrial equipment fault diagnosis system and method based on data fusion | |
| CN102750445A (en) | Failure prediction method of multivariate gray model improved based on Complexification Simpson formula | |
| CN115221924A (en) | Time-series multi-mode-based industrial equipment anomaly detection intelligent recognition algorithm framework | |
| CN113255771B (en) | Fault diagnosis method and system based on multi-dimensional heterogeneous difference analysis | |
| CN114065651A (en) | Fault time prediction method for rotary equipment |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210423 Address after: 200120 building C, No. 888, Huanhu West 2nd Road, Lingang New Area, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai Applicant after: Shanghai Shenwu Intelligent Technology Co.,Ltd. Address before: 211200 Jiangsu, Nanjing, Lishui District, Yong Yang Street, Qinhuai Road, 288, Xing Zhuang science and Technology Industrial Park Applicant before: Nanjing Ruiyong Zhi Operations and Maintenance Engineering Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |