CN109992579A - A kind of data recovery method and system of highway infrastructures multi-resources Heterogeneous data - Google Patents
A kind of data recovery method and system of highway infrastructures multi-resources Heterogeneous data Download PDFInfo
- Publication number
- CN109992579A CN109992579A CN201910244035.2A CN201910244035A CN109992579A CN 109992579 A CN109992579 A CN 109992579A CN 201910244035 A CN201910244035 A CN 201910244035A CN 109992579 A CN109992579 A CN 109992579A
- Authority
- CN
- China
- Prior art keywords
- data
- sensor
- repaired
- source
- sequence
- 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
Abstract
The invention discloses the data recovery methods and system of a kind of highway infrastructures multi-resources Heterogeneous data, are related to data processing field, comprising steps of building multi-source data is associated with computation model；Obtain the deficiency of data of the sensor to be repaired and the partial data of other sensors, the corresponding data sequence for generating all the sensors；The data sequence of other sensors is input in the multi-source data association computation model, the interaction prediction sequence of values that sensor missing data to be repaired corresponds to time point is obtained；Interpolation calculation is carried out to the data sequence of sensor to be repaired using the method for spline interpolation, obtains the time prediction sequence of values that sensor missing data to be repaired corresponds to time point；According to the number of sensor consecutive miss data to be repaired and corresponding preset associated weights coefficient and time weighting coefficient, complex data to be repaired is obtained.The data recovery method of highway infrastructures multi-resources Heterogeneous data provided by the invention, it is more accurate that data are repaired.
Description
Technical field
The present invention relates to data processing fields, and in particular to a kind of data reparation of highway infrastructures multi-resources Heterogeneous data
Method and system.
Background technique
The pipe of Freeway Infrastructure, which is supported, at present relies on a variety of advanced sensor devices to obtain highway in real time
Frastructure state data are completed to construct section, bridge, tunnel, side slope etc. to realize the real-time monitoring to structure state
The Data Detection of object.The Freeway Infrastructure status data got mainly includes fixed detector data, video detection
Data, floating car data etc..
However, fixed detector and the sensor meeting in collection process used on highway and on service area monitoring point
There is a large amount of loss of data, for a part of reason from equipment for a long time by the influence of extreme climate environment, another part is former
Because more difficult derived from plant maintenance work.The missing of data produces certain influence to the management and operation of highway.
What existing data recovery technique generallyd use is the restorative procedure based on time series, and this method mainly passes through
Given data information in missing data itself surrounding time section, by based on the complement values such as interpolation, partial differential equation and total variation,
Approximating method predicts missing data.
It is above-mentioned based on the reparation of time series when continuous data miss rate is smaller, can be by based on the pre- of time series
Method of determining and calculating obtains preferable data prediction accuracy, but when data occur lacking more in continuous time period, single source
The self information of data is difficult accurately to restore the data of missing.
Summary of the invention
In view of the deficiencies in the prior art, the purpose of the present invention is to provide a kind of highway infrastructures multi-resources Heterogeneous
The data recovery method of data, it is more accurate that data are repaired.
To achieve the above objectives, the technical solution adopted by the present invention is that:
A kind of data recovery method of highway infrastructures multi-resources Heterogeneous data, which is characterized in that comprising steps of
S1: building multi-source data is associated with computation model, is got in advance using all the sensors complete multi-source heterogeneous
Data carry out learning training to multi-source data association computation model；
S2: using the sensor of missing data as sensor to be repaired, the deficiency of data of the sensor to be repaired is obtained
And the partial data of other sensors, in conjunction with the time point of sensor acquisition data, the corresponding data for generating all the sensors
Sequence；
S3: the data sequence of other sensors is input in the multi-source data association computation model, is obtained to be repaired
Sensor missing data corresponds to the interaction prediction sequence of values Y at time pointr；
S4: interpolation calculation is carried out to the data sequence of sensor to be repaired using the method for spline interpolation, is obtained to be repaired
Sensor missing data corresponds to the time prediction sequence of values Y at time pointp；
S5: according to the number i of sensor consecutive miss data to be repaired and corresponding preset associated weights factor alphaiWith
Time weighting factor betai, in conjunction with calculation formula YResult=αiYr+βiYp, obtain complex data Y to be repairedResult。
Based on the above technical solution, the preparatory multi-source heterogeneous data got using all the sensors are to this
Multi-source data association computation model carry out learning training detailed process include:
Using one of sensor as sensor of interest, the complete multi-source heterogeneous data that sensor of interest is got
As the target of multi-source data association computation model, the complete multi-source heterogeneous data that other sensors are got are as more
Source data is associated with the input of computation model, carries out learning training to multi-source data association computation model.
Based on the above technical solution, the preset associated weights factor alpha of the correspondenceiWith time weighting factor betai's
Calculating process are as follows:
Choose one section of partial data of all the sensors；
Several groups data are created, every group of data are identical as this section of partial data；
Continuous i data are randomly choosed in every group of data, and i data of selection are removed from every group of data, it is right
It should obtain the deficiency of data that several groups lack i data；
The predicted value of the i data lacked in every group of deficiency of data is calculated using the method for spline interpolation, and by i
The predicted value composition data forecasting sequence of data, obtains the time prediction sequence of values of all groups of deficiencies of data；
All groups of deficiencies of data are separately input in multi-source data association computation model, obtain all groups it is endless
The interaction prediction sequence of values of entire data；
Worked as according to partial data, time prediction sequence of values and interaction prediction sequence of values using least square method
Sensor lacks corresponding associated weights factor alpha when i dataiWith time weighting factor betai。
Based on the above technical solution, when sensor to be repaired, which is interrupted, lacks multiple data, by sensing to be repaired
The data of device are converted into multi-group data, only include the data of one section of consecutive miss in every group of data, and connect according in every group of data
The number of continuous missing data carries out data reparation respectively.
Based on the above technical solution, using the method for cubic spline interpolation in the step S4.
The present invention also provides a kind of data repair systems of highway infrastructures multi-resources Heterogeneous data, comprising:
Model construction module is used for: building multi-source data is associated with computation model, is got in advance using all the sensors
Complete multi-source heterogeneous data to the multi-source data association computation model carry out learning training；
Data processing module is used for: using the sensor of missing data as sensor to be repaired, obtaining the biography to be repaired
The deficiency of data of sensor and the partial data of other sensors, it is corresponding to generate in conjunction with the time point of sensor acquisition data
The data sequence of all the sensors；
Interaction prediction module, is used for: the data sequence of other sensors being input to the multi-source data association and is calculated
In model, the interaction prediction sequence of values Y that sensor missing data to be repaired corresponds to time point is obtainedr；
Time prediction module, is used for: being carried out using the method for spline interpolation to the data sequence of sensor to be repaired slotting
Value calculates, and obtains the time prediction sequence of values Y that sensor missing data to be repaired corresponds to time pointp；
Computing module is used for: according to the number i of sensor consecutive miss data to be repaired and corresponding preset association
Weight coefficient αiWith time weighting factor betai, in conjunction with calculation formula YResult=αiYr+βiYp, obtain complex data Y to be repairedResult。
Based on the above technical solution, the preparatory multi-source heterogeneous data got using all the sensors are to this
Multi-source data association computation model carry out learning training detailed process include:
Using one of sensor as sensor of interest, the complete multi-source heterogeneous data that sensor of interest is got
As the target of multi-source data association computation model, the complete multi-source heterogeneous data that other sensors are got are as more
Source data is associated with the input of computation model, carries out learning training to multi-source data association computation model.
Based on the above technical solution, the computing module further includes weight calculation unit,
The weight calculation unit is used for:
Choose one section of partial data of all the sensors；
Several groups data are created, every group of data are identical as this section of partial data；
Continuous i data are randomly choosed in every group of data, and i data of selection are removed from every group of data, it is right
It should obtain the deficiency of data that several groups lack i data；
The predicted value of the i data lacked in every group of deficiency of data is calculated using the method for spline interpolation, and by i
The predicted value composition data forecasting sequence of data, obtains the time prediction sequence of values of all groups of deficiencies of data；
All groups of deficiencies of data are separately input in multi-source data association computation model, obtain all groups it is endless
The interaction prediction sequence of values of entire data；
Worked as according to partial data, time prediction sequence of values and interaction prediction sequence of values using least square method
Sensor lacks corresponding associated weights factor alpha when i dataiWith time weighting factor betai。
Based on the above technical solution, when sensor to be repaired, which is interrupted, lacks multiple data, by sensing to be repaired
The data of device are converted into multi-group data, only include the data of one section of consecutive miss in every group of data, and connect according in every group of data
The number of continuous missing data carries out data reparation respectively.
Based on the above technical solution, the time prediction module using cubic spline interpolation method.
Compared with the prior art, the advantages of the present invention are as follows: the number of highway infrastructures multi-resources Heterogeneous data of the invention
According to restorative procedure, interaction prediction sequence of values and time prediction sequence of values are merged, and according to both preset power
Coefficient is weighed to carry out data reparation, more can accurately carry out the prediction of missing data, repairs its data more accurate.
Detailed description of the invention
Fig. 1 is a kind of process of the data recovery method of highway infrastructures multi-resources Heterogeneous data in the embodiment of the present invention
Figure.
Specific embodiment
Invention is further described in detail with reference to the accompanying drawings and embodiments.
Shown in Figure 1, the embodiment of the present invention provides a kind of data reparation side of highway infrastructures multi-resources Heterogeneous data
Method, comprising steps of
S1: building multi-source data is associated with computation model, is got in advance using all the sensors complete multi-source heterogeneous
Data carry out learning training to multi-source data association computation model；
S2: using the sensor of missing data as sensor to be repaired, the deficiency of data of the sensor to be repaired is obtained
And the partial data of other sensors, in conjunction with the time point of sensor acquisition data, the corresponding data for generating all the sensors
Sequence；
S3: the data sequence of other sensors is input in the multi-source data association computation model, is obtained to be repaired
Sensor missing data corresponds to the interaction prediction sequence of values Y at time pointr；
S4: interpolation calculation is carried out to the data sequence of sensor to be repaired using the method for spline interpolation, is obtained to be repaired
Sensor missing data corresponds to the time prediction sequence of values Y at time pointp；
S5: according to the number i of sensor consecutive miss data to be repaired and corresponding preset associated weights factor alphaiWith
Time weighting factor betai, in conjunction with calculation formula YResult=αiYr+βiYp, obtain complex data Y to be repairedResult。
Specifically, in embodiments of the present invention, the preparatory multi-source heterogeneous data pair got using all the sensors
The multi-source data association computation model carry out learning training detailed process include:
Using one of sensor as sensor of interest, the complete multi-source heterogeneous data that sensor of interest is got
As the target of multi-source data association computation model, the complete multi-source heterogeneous data that other sensors are got are as more
Source data is associated with the input of computation model, carries out learning training to multi-source data association computation model.
Specifically, in embodiments of the present invention, the preset associated weights factor alpha of the correspondenceiWith time weighting factor betai's
Calculating process are as follows:
Choose one section of partial data of all the sensors；
Several groups data are created, every group of data are identical as this section of partial data；
Continuous i data are randomly choosed in every group of data, and i data of selection are removed from every group of data, it is right
It should obtain the deficiency of data that several groups lack i data；
The predicted value of the i data lacked in every group of deficiency of data is calculated using the method for spline interpolation, and by i
The predicted value composition data forecasting sequence of data, obtains the time prediction sequence of values of all groups of deficiencies of data；
All groups of deficiencies of data are separately input in multi-source data association computation model, obtain all groups it is endless
The interaction prediction sequence of values of entire data；
Worked as according to partial data, time prediction sequence of values and interaction prediction sequence of values using least square method
Sensor lacks corresponding associated weights factor alpha when i dataiWith time weighting factor betai。
When the sensor used in highway infrastructures is tri- sensors of A, B, C, since sensor is obtaining number
According to when be all that fixed interval is quantized, all the sensors obtain data after, can be known according to the number of discrete point
Road sensor corresponds to the number i of missing data, associated weights factor alphaiWith time weighting factor betaiSample calculation it is as follows:
S501: take A, B, C sensor obtain one section of partial data, be denoted as Y, herein this section of partial data be interpreted as A,
B, the set for the data that C sensor obtains；
S502: creation K group data, every group of data are identical as this section of partial data, are denoted as Y respectively1, Y2, Y3... ...,
YK；
S503: randomly selecting continuous i data in every group of data, and the m group nth data of selection is denoted as Ymn,
Wherein positive integer of the m between 1~K, positive integer of the n between 1~i；
S504: i data of selection are removed from every group of data, and correspondence obtains K group and lacks the imperfect of i data
Data are denoted as X respectively1, X2, X3... ..., XK；
In embodiments of the present invention, the i position randomly selected in every group of continuous data can be overlapped, but not repeated,
And endpoint is not selected, and for example, this section of partial data is the sequence of values of 1~9 composition, when K=3, i=4, i of first group of removal
It is 2345 that continuous data, which can choose, and it is 4567 that i continuous data of second group of removal, which can choose, i of the removal of third group
It is 5678 that continuous data, which can choose,.
S505: calculating the predicted value of the i data lacked in every group of deficiency of data using interpolation method, with Ypmn table
Show the nth data predicted value of m group, wherein positive integer of the m between 1~K, positive integer of the n between 1~i, by every group
I number is it is predicted that value forms a number it is predicted that sequence, obtains K group data forecasting sequence, be denoted as Ypk；
S506: K group deficiency of data is separately input in multi-source data association computation model, the endless integer of K group is obtained
According to interaction prediction sequence of values, be denoted as Yrk；
S507: according to formula Yk=αiYrk+βiYpk, K simultaneous equations is obtained, using least squares estimate, is calculated
Weight coefficient α when the sensor lacks i data outiAnd βi。
In embodiments of the present invention, when sensor to be repaired, which is interrupted, lacks multiple data, by the number of sensor to be repaired
It only include the data of one section of consecutive miss in every group of data, and according to consecutive miss in every group of data according to multi-group data is converted into
The number of data carries out data reparation respectively.
For example, work as 2 data of data elder generation consecutive miss of A sensor, 3 data of consecutive miss behind one section of interval, then
Two groups of data are converted by the data of A sensor, only includes the data of one section of consecutive miss in every group of data, then distinguishes
Carry out data reparation.
Preferably, using the method for cubic spline interpolation in the step S4.
Since multi-resources Heterogeneous data often have semantic complementary characteristic, this shows there is one between the data in different data source
Determine correlation, therefore other data sources of missing data same period contain certain information content of missing data, i.e. missing data
The data of sensor and the data of other sensors have certain relevance.Meanwhile history number before and after sensor missing data
According to also having temporal associativity, missing data can centainly be predicted by data variation trend etc..
It is difficult to accurately be fitted curve when continuous time lacking more by the method that historical data is predicted, and multi-source
Data relation analysis result when multi-source data property difference is larger is difficult to restrain, so as to cause missing data repairing effect compared with
Difference.
In the data recovery method of the highway infrastructures multi-resources Heterogeneous data of the embodiment of the present invention, by above two feelings
Condition is merged, i.e., merges to interaction prediction sequence of values and time prediction sequence of values, and according to preset the two
Weight coefficient carries out data reparation, on the one hand, can promote the data in the sparse situation of missing data surrounding time data
On the other hand accuracy can carry out a periodical repair to result caused by multi-source data association analysis in conjunction with surrounding time information
Just, the temporal associativity of multi-source data association results is enhanced, and then more can accurately carry out the prediction of missing data, is made
It is more accurate that its data is repaired.
The embodiment of the invention also provides a kind of data repair systems of highway infrastructures multi-resources Heterogeneous data, comprising:
Model construction module is used for: building multi-source data is associated with computation model, is got in advance using all the sensors
Complete multi-source heterogeneous data to the multi-source data association computation model carry out learning training；
Data processing module is used for: using the sensor of missing data as sensor to be repaired, obtaining the biography to be repaired
The deficiency of data of sensor and the partial data of other sensors, it is corresponding to generate in conjunction with the time point of sensor acquisition data
The data sequence of all the sensors；
Interaction prediction module, is used for: the data sequence of other sensors being input to the multi-source data association and is calculated
In model, the interaction prediction sequence of values Y that sensor missing data to be repaired corresponds to time point is obtainedr；
Time prediction module, is used for: being carried out using the method for spline interpolation to the data sequence of sensor to be repaired slotting
Value calculates, and obtains the time prediction sequence of values Y that sensor missing data to be repaired corresponds to time pointp；
Computing module is used for: according to the number i of sensor consecutive miss data to be repaired and corresponding preset association
Weight coefficient αiWith time weighting factor betai, in conjunction with calculation formula YResult=αiYr+βiYp, obtain complex data Y to be repairedResult。
Specifically, the preparatory multi-source heterogeneous data got using all the sensors are associated with the multi-source data and calculate
Model carry out learning training detailed process include:
Using one of sensor as sensor of interest, the complete multi-source heterogeneous data that sensor of interest is got
As the target of multi-source data association computation model, the complete multi-source heterogeneous data that other sensors are got are as more
Source data is associated with the input of computation model, carries out learning training to multi-source data association computation model.
More specifically, the computing module further includes weight calculation unit,
The weight calculation unit is used for:
Choose one section of partial data of all the sensors；
Several groups data are created, every group of data are identical as this section of partial data；
Continuous i data are randomly choosed in every group of data, and i data of selection are removed from every group of data, it is right
It should obtain the deficiency of data that several groups lack i data；
The predicted value of the i data lacked in every group of deficiency of data is calculated using the method for spline interpolation, and by i
The predicted value composition data forecasting sequence of data, obtains the time prediction sequence of values of all groups of deficiencies of data；
All groups of deficiencies of data are separately input in multi-source data association computation model, obtain all groups it is endless
The interaction prediction sequence of values of entire data；
Worked as according to partial data, time prediction sequence of values and interaction prediction sequence of values using least square method
Sensor lacks corresponding associated weights factor alpha when i dataiWith time weighting factor betai。
In embodiments of the present invention, when sensor to be repaired, which is interrupted, lacks multiple data, by the number of sensor to be repaired
It only include the data of one section of consecutive miss in every group of data, and according to consecutive miss in every group of data according to multi-group data is converted into
The number of data carries out data reparation respectively.
Preferably, the time prediction module using cubic spline interpolation method.
The data repair system of highway infrastructures multi-resources Heterogeneous data of the invention, to interaction prediction sequence of values and when
Between prediction sequence of values merged, and data reparation is carried out according to both preset weight coefficient, can be more accurate
Ground carries out the prediction of missing data, repairs its data more accurate.
The present invention is not limited to the above-described embodiments, for those skilled in the art, is not departing from
Under the premise of the principle of the invention, several improvements and modifications can also be made, these improvements and modifications are also considered as protection of the invention
Within the scope of.The content being not described in detail in this specification belongs to the prior art well known to professional and technical personnel in the field.
Claims (10)
1. a kind of data recovery method of highway infrastructures multi-resources Heterogeneous data, which is characterized in that comprising steps of
S1: building multi-source data is associated with computation model, the complete multi-source heterogeneous data got in advance using all the sensors
Learning training is carried out to multi-source data association computation model；
S2: using the sensor of missing data as sensor to be repaired, obtain the sensor to be repaired deficiency of data and
The partial data of other sensors, in conjunction with the time point of sensor acquisition data, the corresponding data sequence for generating all the sensors；
S3: the data sequence of other sensors is input in the multi-source data association computation model, sensing to be repaired is obtained
Device missing data corresponds to the interaction prediction sequence of values Y at time pointr；
S4: interpolation calculation is carried out to the data sequence of sensor to be repaired using the method for spline interpolation, obtains sensing to be repaired
Device missing data corresponds to the time prediction sequence of values Y at time pointp；
S5: according to the number i of sensor consecutive miss data to be repaired and corresponding preset associated weights factor alphaiIt is weighed with the time
Weight factor betai, in conjunction with calculation formula YResult=αiYr+βiYp, obtain complex data Y to be repairedResult。
2. the data recovery method of highway infrastructures multi-resources Heterogeneous data as described in claim 1, which is characterized in that described
The multi-source heterogeneous data got in advance using all the sensors carry out learning training to multi-source data association computation model
Detailed process includes:
Using one of sensor as sensor of interest, the complete multi-source heterogeneous data that sensor of interest is got as
The target of the multi-source data association computation model, the complete multi-source heterogeneous data that other sensors are got are as multi-source number
According to the input of association computation model, learning training is carried out to multi-source data association computation model.
3. the data recovery method of highway infrastructures multi-resources Heterogeneous data as described in claim 1, which is characterized in that described
Corresponding preset associated weights factor alphaiWith time weighting factor betaiCalculating process are as follows:
Choose one section of partial data of all the sensors；
Several groups data are created, every group of data are identical as this section of partial data；
Continuous i data are randomly choosed in every group of data, i data of selection are removed from every group of data, to deserved
The deficiency of data of i data is lacked to several groups；
Calculate the predicted value of the i data lacked in every group of deficiency of data using the method for spline interpolation, and by i data
Predicted value composition data forecasting sequence, obtain the time prediction sequence of values of all groups of deficiencies of data；
All groups of deficiencies of data are separately input in the multi-source data association computation model, all groups of endless integers are obtained
According to interaction prediction sequence of values；
According to partial data, time prediction sequence of values and interaction prediction sequence of values, using least square method, obtain when sensing
Device lacks corresponding associated weights factor alpha when i dataiWith time weighting factor betai。
4. the data recovery method of highway infrastructures multi-resources Heterogeneous data as described in claim 1, which is characterized in that when to
When repairing sensor interruption and lacking multiple data, multi-group data is converted by the data of sensor to be repaired, in every group of data only
Data comprising one section of consecutive miss, and data reparation is carried out according to the number of consecutive miss data in every group of data respectively.
5. the data recovery method of highway infrastructures multi-resources Heterogeneous data as described in claim 1, which is characterized in that described
Using the method for cubic spline interpolation in step S4.
6. a kind of data repair system of highway infrastructures multi-resources Heterogeneous data characterized by comprising
Model construction module is used for: building multi-source data is associated with computation model, is got in advance using all the sensors complete
Whole multi-source heterogeneous data carry out learning training to multi-source data association computation model；
Data processing module is used for: using the sensor of missing data as sensor to be repaired, obtaining the sensor to be repaired
Deficiency of data and other sensors partial data, in conjunction with the time point of sensor acquisition data, it is corresponding generate it is all
The data sequence of sensor；
Interaction prediction module, is used for: the data sequence of other sensors being input to the multi-source data and is associated with computation model
In, obtain the interaction prediction sequence of values Y that sensor missing data to be repaired corresponds to time pointr；
Time prediction module, by: the data sequence of sensor to be repaired is carried out based on interpolation using the method for spline interpolation
It calculates, obtains the time prediction sequence of values Y that sensor missing data to be repaired corresponds to time pointp；
Computing module is used for: according to the number i of sensor consecutive miss data to be repaired and corresponding preset associated weights
Factor alphaiWith time weighting factor betai, in conjunction with calculation formula YResult=αiYr+βiYp, obtain complex data Y to be repairedResult。
7. the data repair system of highway infrastructures multi-resources Heterogeneous data as claimed in claim 6, which is characterized in that described
The multi-source heterogeneous data got in advance using all the sensors carry out learning training to multi-source data association computation model
Detailed process includes:
Using one of sensor as sensor of interest, the complete multi-source heterogeneous data that sensor of interest is got as
The target of the multi-source data association computation model, the complete multi-source heterogeneous data that other sensors are got are as multi-source number
According to the input of association computation model, learning training is carried out to multi-source data association computation model.
8. the data repair system of highway infrastructures multi-resources Heterogeneous data as claimed in claim 6, which is characterized in that described
Computing module further includes weight calculation unit,
The weight calculation unit is used for:
Choose one section of partial data of all the sensors；
Several groups data are created, every group of data are identical as this section of partial data；
Continuous i data are randomly choosed in every group of data, i data of selection are removed from every group of data, to deserved
The deficiency of data of i data is lacked to several groups；
Calculate the predicted value of the i data lacked in every group of deficiency of data using the method for spline interpolation, and by i data
Predicted value composition data forecasting sequence, obtain the time prediction sequence of values of all groups of deficiencies of data；
All groups of deficiencies of data are separately input in the multi-source data association computation model, all groups of endless integers are obtained
According to interaction prediction sequence of values；
According to partial data, time prediction sequence of values and interaction prediction sequence of values, using least square method, obtain when sensing
Device lacks corresponding associated weights factor alpha when i dataiWith time weighting factor betai。
9. the data repair system of highway infrastructures multi-resources Heterogeneous data as claimed in claim 6, which is characterized in that when to
When repairing sensor interruption and lacking multiple data, multi-group data is converted by the data of sensor to be repaired, in every group of data only
Data comprising one section of consecutive miss, and data reparation is carried out according to the number of consecutive miss data in every group of data respectively.
10. the data repair system of highway infrastructures multi-resources Heterogeneous data as claimed in claim 6, which is characterized in that institute
Time prediction module is stated using the method for cubic spline interpolation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910244035.2A CN109992579B (en) | 2019-03-28 | 2019-03-28 | Data restoration method and system for multisource heterogeneous data of highway infrastructure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910244035.2A CN109992579B (en) | 2019-03-28 | 2019-03-28 | Data restoration method and system for multisource heterogeneous data of highway infrastructure |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109992579A true CN109992579A (en) | 2019-07-09 |
| CN109992579B CN109992579B (en) | 2021-10-01 |
Family
ID=67131850
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910244035.2A Active CN109992579B (en) | 2019-03-28 | 2019-03-28 | Data restoration method and system for multisource heterogeneous data of highway infrastructure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109992579B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2593173A (en) * | 2020-03-16 | 2021-09-22 | British Telecomm | Modelling physical infrastructure |
| CN113642088A (en) * | 2021-10-13 | 2021-11-12 | 北京天辰信科技有限公司 | Method for feeding back construction progress information and displaying deviation of BIM (building information modeling) model in real time |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102110302A (en) * | 2009-12-25 | 2011-06-29 | 东软飞利浦医疗设备系统有限责任公司 | Method for compensating missing data in CT (Computed Tomography) scanning |
| CN102622880A (en) * | 2012-01-09 | 2012-08-01 | 北京捷易联科技有限公司 | Traffic information data recovery method and device |
| CN102819682A (en) * | 2012-08-02 | 2012-12-12 | 清华大学 | Data cleaning and recovering algorithm of floating car related to multi-threshold space |
| CN103971520A (en) * | 2014-04-17 | 2014-08-06 | 浙江大学 | Traffic flow data recovery method based on space-time correlation |
| CN105142164A (en) * | 2015-06-24 | 2015-12-09 | 北京邮电大学 | Data filling method and device of node to be estimated |
| CN106096302A (en) * | 2016-06-22 | 2016-11-09 | 江苏迪纳数字科技股份有限公司 | Based on time and the data recovery method of section dependency |
| US20170235848A1 (en) * | 2012-08-29 | 2017-08-17 | Dennis Van Dusen | System and method for fuzzy concept mapping, voting ontology crowd sourcing, and technology prediction |
| CN109063366A (en) * | 2018-08-22 | 2018-12-21 | 深圳市建筑科学研究院股份有限公司 | A kind of building performance online data preprocess method based on time and spatial weighting |
-
2019
- 2019-03-28 CN CN201910244035.2A patent/CN109992579B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102110302A (en) * | 2009-12-25 | 2011-06-29 | 东软飞利浦医疗设备系统有限责任公司 | Method for compensating missing data in CT (Computed Tomography) scanning |
| CN102622880A (en) * | 2012-01-09 | 2012-08-01 | 北京捷易联科技有限公司 | Traffic information data recovery method and device |
| CN102819682A (en) * | 2012-08-02 | 2012-12-12 | 清华大学 | Data cleaning and recovering algorithm of floating car related to multi-threshold space |
| US20170235848A1 (en) * | 2012-08-29 | 2017-08-17 | Dennis Van Dusen | System and method for fuzzy concept mapping, voting ontology crowd sourcing, and technology prediction |
| CN103971520A (en) * | 2014-04-17 | 2014-08-06 | 浙江大学 | Traffic flow data recovery method based on space-time correlation |
| CN105142164A (en) * | 2015-06-24 | 2015-12-09 | 北京邮电大学 | Data filling method and device of node to be estimated |
| CN106096302A (en) * | 2016-06-22 | 2016-11-09 | 江苏迪纳数字科技股份有限公司 | Based on time and the data recovery method of section dependency |
| CN109063366A (en) * | 2018-08-22 | 2018-12-21 | 深圳市建筑科学研究院股份有限公司 | A kind of building performance online data preprocess method based on time and spatial weighting |
Non-Patent Citations (3)
| Title |
|---|
| SHIFEN CHENG 等: "A Two-Step Method for Missing Spatio-Temporal data Reconstruction", 《INTERNATIONAL JOURNAL OF GEO-INFORMATION》 * |
| 孙玲 等: "考虑时空相关性的固定检测缺失数据重构", 《交通运输工程学报》 * |
| 遇今: "《航天器研制风险管理》", 31 December 2012, 航空工业出版社 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2593173A (en) * | 2020-03-16 | 2021-09-22 | British Telecomm | Modelling physical infrastructure |
| CN113642088A (en) * | 2021-10-13 | 2021-11-12 | 北京天辰信科技有限公司 | Method for feeding back construction progress information and displaying deviation of BIM (building information modeling) model in real time |
| CN113642088B (en) * | 2021-10-13 | 2022-01-07 | 北京天辰信科技有限公司 | Method for feeding back construction progress information and displaying deviation of BIM (building information modeling) model in real time |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109992579B (en) | 2021-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110108328B (en) | Method for acquiring water leakage amount of leakage area of water supply pipe network | |
| Gao et al. | Digital twin and its application in transportation infrastructure | |
| CN108106663B (en) | Online visual bridge monitoring platform for creating and managing | |
| CN103424654B (en) | A kind of appraisal procedure of voltage dip sensitivity of sensitive equipment | |
| CN102393877B (en) | Method for simulating random temperature field of steel box beam of bridge construction | |
| CN104063612A (en) | Tunnel engineering risk situation fuzzy evaluation method and system | |
| CN101793623B (en) | Progressive method for distinguishing slack supporting cable based on cable force monitoring during support settlement | |
| CN109992579A (en) | A kind of data recovery method and system of highway infrastructures multi-resources Heterogeneous data | |
| CN116822969B (en) | Water conservancy model cloud computing method and system based on model combination | |
| CN109150100A (en) | Fault detection method, device, equipment and the storage medium of photovoltaic plant | |
| CN115687983B (en) | Bridge health state monitoring method and system and electronic equipment | |
| CN111709184A (en) | Accumulated ice combined decision-making method based on pavement temperature prediction and simulation | |
| CN105824987A (en) | Wind field characteristic statistical distributing model building method based on genetic algorithm | |
| CN108415884A (en) | A kind of modal parameters real-time tracing method | |
| CN111241660B (en) | Civil engineering experiment detection system and method based on cloud computing technology | |
| CN116595327A (en) | Sluice deformation monitoring data preprocessing system and method | |
| CN110285330B (en) | Water network pipe burst detection method based on local outlier factor | |
| CN117236078A (en) | Highway traffic detector layout method based on double-layer constraint | |
| Cao et al. | Methodology to evaluate cost and accuracy of parking patrol surveys | |
| CN106708786A (en) | Method and system for calculating problem severity of iron tower based on sensor detection | |
| CN110852610A (en) | Road, bridge and tunnel health state and maintenance cost measuring and calculating method based on Markov model | |
| CN114611728B (en) | Sewage pipe network blockage monitoring method and system | |
| Wenner et al. | smartBRIDGE Hamburg: A digital twin to optimise infrastructure maintenance | |
| CN102359368B (en) | Working fluid level prediction method for submersible reciprocating-type oil pumping unit based on support vector machine | |
| CN111339699A (en) | Box-type structure weight deformation prediction method based on macroscopic stress and residual stress divergence |
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 |