CN109992579B - Data restoration method and system for multisource heterogeneous data of highway infrastructure - Google Patents
Data restoration method and system for multisource heterogeneous data of highway infrastructure Download PDFInfo
- Publication number
- CN109992579B CN109992579B CN201910244035.2A CN201910244035A CN109992579B CN 109992579 B CN109992579 B CN 109992579B CN 201910244035 A CN201910244035 A CN 201910244035A CN 109992579 B CN109992579 B CN 109992579B
- Authority
- CN
- China
- Prior art keywords
- data
- sensor
- repaired
- source
- sensors
- 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
Images
Landscapes
- Testing Or Calibration Of Command Recording Devices (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses a data restoration method and a data restoration system for multisource heterogeneous data of highway infrastructure, which relate to the field of data processing and comprise the following steps: constructing a multi-source data association calculation model; acquiring incomplete data of the sensor to be repaired and complete data of other sensors, and correspondingly generating data sequences of all the sensors; inputting data sequences of other sensors into the multi-source data correlation calculation model to obtain correlation prediction numerical value sequences of the missing data of the sensor to be repaired at corresponding time points; carrying out interpolation calculation on a data sequence of the sensor to be repaired by using a spline interpolation method to obtain a time prediction numerical sequence of a time point corresponding to missing data of the sensor to be repaired; and obtaining the data to be repaired according to the number of the continuous missing data of the sensor to be repaired and the corresponding preset associated weight coefficient and time weight coefficient. The data restoration method for the multisource heterogeneous data of the highway infrastructure, provided by the invention, has the advantage that the data restoration is more accurate.
Description
Technical Field
The invention relates to the field of data processing, in particular to a data restoration method and system for multi-source heterogeneous data of highway infrastructure.
Background
At present, management and maintenance of highway infrastructure relies on various advanced sensor devices to acquire state data of the highway infrastructure in real time so as to realize real-time monitoring of the state of a structure and complete data detection of structures such as road sections, bridges, tunnels, slopes and the like. The acquired highway infrastructure state data mainly comprises fixed detector data, video detection data, floating car data and the like.
However, the fixed detectors and sensors used on the highway and at the monitoring points in the service area lose a lot of data during the acquisition process, partly because the equipment is affected by the severe weather environment for a long time, and partly because the maintenance work of the equipment is difficult. The lack of data has a certain influence on the management and operation of the highway.
The existing data restoration technology generally adopts a restoration method based on time series, and the method mainly predicts the missing data through known data information in the time periods before and after the missing data per se and through complementary values and a fitting method based on interpolation, partial differential equations, total variation and the like.
When the continuous data loss rate is low, the time-series-based repair can obtain good data prediction accuracy through a time-series-based prediction algorithm, but when the data are lost more in a continuous time period, the self information of the data from a single source is difficult to accurately recover the lost data.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a data repairing method for multisource heterogeneous data of highway infrastructure, which is more accurate in data repairing.
In order to achieve the above purposes, the technical scheme adopted by the invention is as follows:
a data restoration method for multi-source heterogeneous data of a road infrastructure is characterized by comprising the following steps:
s1: constructing a multi-source data association calculation model, and performing learning training on the multi-source data association calculation model by using complete multi-source heterogeneous data obtained by all sensors in advance;
s2: taking a sensor with missing data as a sensor to be repaired, acquiring incomplete data of the sensor to be repaired and complete data of other sensors, and correspondingly generating data sequences of all the sensors by combining time points of data acquisition of the sensors;
s3: inputting data sequences of other sensors into the multi-source data correlation calculation model to obtain a correlation prediction numerical value sequence Y of the missing data of the sensor to be repaired at a corresponding time pointr；
S4: the spline interpolation method is used for carrying out interpolation calculation on the data sequence of the sensor to be repaired,obtaining a time prediction numerical value sequence Y of the corresponding time point of the missing data of the sensor to be repairedp；
S5: according to the number i of continuous missing data of the sensor to be repaired and the corresponding preset associated weight coefficient alphaiAnd a time weight coefficient betaiIn combination with the calculation formula YResult＝αiYr+βiYpObtaining data Y to be repairedResult。
On the basis of the technical scheme, the specific process of using the multi-source heterogeneous data acquired by all the sensors in advance to carry out learning training on the multi-source data association calculation model comprises the following steps:
and taking one sensor as a target sensor, taking the complete multi-source heterogeneous data acquired by the target sensor as a target of the multi-source data association calculation model, taking the complete multi-source heterogeneous data acquired by other sensors as the input of the multi-source data association calculation model, and performing learning training on the multi-source data association calculation model.
On the basis of the technical scheme, the corresponding preset associated weight coefficient alphaiAnd a time weight coefficient betaiThe calculation process of (2) is as follows:
selecting a section of complete data of all sensors;
creating a plurality of groups of data, wherein each group of data is the same as the complete data;
randomly selecting continuous i data in each group of data, removing the selected i data from each group of data, and correspondingly obtaining a plurality of groups of incomplete data missing i data;
calculating the predicted values of i missing data in each group of incomplete data by using a spline interpolation method, and forming the predicted values of the i data into a data prediction sequence to obtain time prediction numerical value sequences of all groups of incomplete data;
respectively inputting all groups of incomplete data into the multi-source data correlation calculation model to obtain correlation prediction numerical value sequences of all groups of incomplete data;
predicting numerical sequence from complete data, timeUsing least square method to obtain the corresponding associated weight coefficient alpha when the sensor lacks i dataiAnd a time weight coefficient betai。
On the basis of the technical scheme, when the sensor to be repaired is intermittently missing a plurality of data, the data of the sensor to be repaired is converted into a plurality of groups of data, each group of data only comprises a section of continuously missing data, and data repair is respectively carried out according to the number of the continuously missing data in each group of data.
On the basis of the above technical solution, the method of cubic spline interpolation is adopted in step S4.
The invention also provides a data recovery system of multisource heterogeneous data of highway infrastructure, which comprises the following steps:
a model building module to: constructing a multi-source data association calculation model, and performing learning training on the multi-source data association calculation model by using complete multi-source heterogeneous data obtained by all sensors in advance;
a data processing module to: taking a sensor with missing data as a sensor to be repaired, acquiring incomplete data of the sensor to be repaired and complete data of other sensors, and correspondingly generating data sequences of all the sensors by combining time points of data acquisition of the sensors;
a relevance prediction module to: inputting data sequences of other sensors into the multi-source data correlation calculation model to obtain a correlation prediction numerical value sequence Y of the missing data of the sensor to be repaired at a corresponding time pointr；
A temporal prediction module to: interpolation calculation is carried out on the data sequence of the sensor to be repaired by using a spline interpolation method to obtain a time prediction numerical value sequence Y of the time point corresponding to the missing data of the sensor to be repairedp；
A computing module to: according to the number i of continuous missing data of the sensor to be repaired and the corresponding preset associated weight coefficient alphaiAnd a time weight coefficient betaiIn combination with the calculation formula YResult＝αiYr+βiYpObtaining data Y to be repairedResult。
On the basis of the technical scheme, the specific process of using the multi-source heterogeneous data acquired by all the sensors in advance to carry out learning training on the multi-source data association calculation model comprises the following steps:
and taking one sensor as a target sensor, taking the complete multi-source heterogeneous data acquired by the target sensor as a target of the multi-source data association calculation model, taking the complete multi-source heterogeneous data acquired by other sensors as the input of the multi-source data association calculation model, and performing learning training on the multi-source data association calculation model.
On the basis of the technical scheme, the calculation module further comprises a weight calculation unit,
the weight calculation unit is configured to:
selecting a section of complete data of all sensors;
creating a plurality of groups of data, wherein each group of data is the same as the complete data;
randomly selecting continuous i data in each group of data, removing the selected i data from each group of data, and correspondingly obtaining a plurality of groups of incomplete data missing i data;
calculating the predicted values of i missing data in each group of incomplete data by using a spline interpolation method, and forming the predicted values of the i data into a data prediction sequence to obtain time prediction numerical value sequences of all groups of incomplete data;
respectively inputting all groups of incomplete data into the multi-source data correlation calculation model to obtain correlation prediction numerical value sequences of all groups of incomplete data;
obtaining a corresponding correlation weight coefficient alpha when the sensor lacks i data by using a least square method according to the complete data, the time prediction numerical value sequence and the correlation prediction numerical value sequenceiAnd a time weight coefficient betai。
On the basis of the technical scheme, when the sensor to be repaired is intermittently missing a plurality of data, the data of the sensor to be repaired is converted into a plurality of groups of data, each group of data only comprises a section of continuously missing data, and data repair is respectively carried out according to the number of the continuously missing data in each group of data.
On the basis of the technical scheme, the time prediction module adopts a cubic spline interpolation method.
Compared with the prior art, the invention has the advantages that: according to the data restoration method for the multisource heterogeneous data of the highway infrastructure, the correlation prediction numerical value sequence and the time prediction numerical value sequence are fused, and data restoration is performed according to the preset weight coefficients of the correlation prediction numerical value sequence and the time prediction numerical value sequence, so that missing data can be more accurately predicted, and the data restoration is more accurate.
Drawings
Fig. 1 is a flowchart of a data recovery method for multi-source heterogeneous data of a road infrastructure according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Referring to fig. 1, an embodiment of the present invention provides a data recovery method for multi-source heterogeneous data of a road infrastructure, including:
s1: constructing a multi-source data association calculation model, and performing learning training on the multi-source data association calculation model by using complete multi-source heterogeneous data obtained by all sensors in advance;
s2: taking a sensor with missing data as a sensor to be repaired, acquiring incomplete data of the sensor to be repaired and complete data of other sensors, and correspondingly generating data sequences of all the sensors by combining time points of data acquisition of the sensors;
s3: inputting data sequences of other sensors into the multi-source data correlation calculation model to obtain a correlation prediction numerical value sequence Y of the missing data of the sensor to be repaired at a corresponding time pointr；
S4: interpolation calculation is carried out on the data sequence of the sensor to be repaired by using a spline interpolation method to obtain a time prediction numerical value sequence Y of the time point corresponding to the missing data of the sensor to be repairedp；
S5: according to the number i of continuous missing data of the sensor to be repaired and the corresponding preset associated weight coefficient alphaiAnd a time weight coefficient betaiIn combination with the calculation formula YResult＝αiYr+βiYpObtaining data Y to be repairedResult。
Specifically, in the embodiment of the present invention, the specific process of performing learning training on the multi-source data association calculation model by using the multi-source heterogeneous data acquired by all the sensors in advance includes:
and taking one sensor as a target sensor, taking the complete multi-source heterogeneous data acquired by the target sensor as a target of the multi-source data association calculation model, taking the complete multi-source heterogeneous data acquired by other sensors as the input of the multi-source data association calculation model, and performing learning training on the multi-source data association calculation model.
Specifically, in the embodiment of the present invention, the corresponding preset associated weight coefficient αiAnd a time weight coefficient betaiThe calculation process of (2) is as follows:
selecting a section of complete data of all sensors;
creating a plurality of groups of data, wherein each group of data is the same as the complete data;
randomly selecting continuous i data in each group of data, removing the selected i data from each group of data, and correspondingly obtaining a plurality of groups of incomplete data missing i data;
calculating the predicted values of i missing data in each group of incomplete data by using a spline interpolation method, and forming the predicted values of the i data into a data prediction sequence to obtain time prediction numerical value sequences of all groups of incomplete data;
respectively inputting all groups of incomplete data into the multi-source data correlation calculation model to obtain correlation prediction numerical value sequences of all groups of incomplete data;
obtaining the current sensor missing by using a least square method according to the complete data, the time prediction numerical value sequence and the correlation prediction numerical value sequenceAssociated weight coefficient alpha corresponding to i dataiAnd a time weight coefficient betai。
When the sensors used in the highway infrastructure are A, B, C three sensors, since the sensors are all discrete values at fixed intervals when acquiring data, after all the sensors acquire data, the number i of the sensors corresponding to the missing data and the associated weight coefficient alpha can be known according to the number of discrete pointsiAnd a time weight coefficient betaiAn example of the calculation of (c) is as follows:
s501: taking a segment of complete data acquired by A, B, C sensors, denoted as Y, where the segment of complete data is understood to be a set of data acquired by A, B, C sensors;
s502: creating K groups of data, wherein each group of data is the same as the complete data and is respectively marked as Y1，Y2，Y3，……，YK；
S503: randomly selecting continuous i data in each group of data, and recording the nth data of the selected mth group as YmnWherein m is a positive integer between 1 and K, and n is a positive integer between 1 and i;
s504: removing the selected i data from each group of data, correspondingly obtaining K groups of incomplete data missing i data, and respectively recording the K groups of incomplete data as X1，X2，X3，……，XK；
In the embodiment of the present invention, the randomly selected i positions in each group of consecutive data may overlap, but do not repeat, and the end point is not selected, for example, the complete data segment is a sequence of values 1 to 9, when K is 3 and i is 4, the first group of i consecutive data removed may be selected as 2345, the second group of i consecutive data removed may be selected as 4567, and the third group of i consecutive data removed may be selected as 5678.
S505: calculating the predicted value of i missing data in each group of incomplete data by using an interpolation method, expressing the nth data predicted value of the mth group by using Ypmn, wherein m is a positive integer between 1 and K, n is a positive integer between 1 and i, forming the i data predicted values of each group into a data prediction sequence, and obtaining a K group data prediction sequenceIs marked as Ypk；
S506: respectively inputting the K groups of incomplete data into a multi-source data correlation calculation model to obtain a correlation prediction numerical sequence of the K groups of incomplete data, and recording the correlation prediction numerical sequence as Yrk；
S507: according to formula Yk＝αiYrk+βiYpkK simultaneous equations are obtained, and the weight coefficient alpha when the sensor lacks i data is calculated by using a least square estimation methodiAnd betai。
In the embodiment of the invention, when the sensor to be repaired is intermittently missing a plurality of data, the data of the sensor to be repaired is converted into a plurality of groups of data, each group of data only comprises a section of continuously missing data, and data repair is respectively carried out according to the number of the continuously missing data in each group of data.
For example, when the data of the sensor a continuously lacks 2 data, and continuously lacks 3 data after a period of interval, the data of the sensor a is converted into two groups of data, each group of data only includes a section of continuously missing data, and then data restoration is performed respectively.
Preferably, the step S4 uses a cubic spline interpolation method.
Since the multi-source heterogeneous data often has semantic complementary characteristics, which indicate that data of different data sources have certain correlation, other data sources of the missing data in the same time period contain certain information amount of the missing data, that is, the data of the missing data sensor and the data of other sensors have certain correlation. Meanwhile, historical data before and after the missing data of the sensor also has time relevance, and the missing data can be predicted to a certain extent through data change trend and the like.
The method for predicting historical data is difficult to accurately fit a curve when a large amount of missing data exists in continuous time, and the result of multi-source data correlation analysis is difficult to converge when the difference of multi-source data characteristics is large, so that the repairing effect of the missing data is poor.
In the data restoration method for the multisource heterogeneous data of the highway infrastructure, the two situations are fused, namely, the association prediction numerical value sequence and the time prediction numerical value sequence are fused, and data restoration is performed according to the preset weight coefficients of the association prediction numerical value sequence and the time prediction numerical value sequence, so that on one hand, the data accuracy under the condition that the time data before and after the missing data is sparse can be improved, on the other hand, the result generated by the multisource data association analysis can be corrected to a certain extent by combining the time information before and after the missing data, the time association of the multisource data association result is enhanced, further, the missing data can be more accurately predicted, and the data restoration is more accurate.
The embodiment of the invention also provides a data recovery system of multisource heterogeneous data of highway infrastructure, which comprises the following steps:
a model building module to: constructing a multi-source data association calculation model, and performing learning training on the multi-source data association calculation model by using complete multi-source heterogeneous data obtained by all sensors in advance;
a data processing module to: taking a sensor with missing data as a sensor to be repaired, acquiring incomplete data of the sensor to be repaired and complete data of other sensors, and correspondingly generating data sequences of all the sensors by combining time points of data acquisition of the sensors;
a relevance prediction module to: inputting data sequences of other sensors into the multi-source data correlation calculation model to obtain a correlation prediction numerical value sequence Y of the missing data of the sensor to be repaired at a corresponding time pointr；
A temporal prediction module to: interpolation calculation is carried out on the data sequence of the sensor to be repaired by using a spline interpolation method to obtain a time prediction numerical value sequence Y of the time point corresponding to the missing data of the sensor to be repairedp；
A computing module to: according to the number i of continuous missing data of the sensor to be repaired and the corresponding preset associated weight coefficient alphaiAnd a time weight coefficient betaiIn combination with the calculation formula YResult＝αiYr+βiYpObtaining data Y to be repairedResult。
Specifically, the specific process of performing learning training on the multi-source data association calculation model by using the multi-source heterogeneous data acquired by all the sensors in advance includes:
and taking one sensor as a target sensor, taking the complete multi-source heterogeneous data acquired by the target sensor as a target of the multi-source data association calculation model, taking the complete multi-source heterogeneous data acquired by other sensors as the input of the multi-source data association calculation model, and performing learning training on the multi-source data association calculation model.
More specifically, the calculation module further includes a weight calculation unit,
the weight calculation unit is configured to:
selecting a section of complete data of all sensors;
creating a plurality of groups of data, wherein each group of data is the same as the complete data;
randomly selecting continuous i data in each group of data, removing the selected i data from each group of data, and correspondingly obtaining a plurality of groups of incomplete data missing i data;
calculating the predicted values of i missing data in each group of incomplete data by using a spline interpolation method, and forming the predicted values of the i data into a data prediction sequence to obtain time prediction numerical value sequences of all groups of incomplete data;
respectively inputting all groups of incomplete data into the multi-source data correlation calculation model to obtain correlation prediction numerical value sequences of all groups of incomplete data;
obtaining a corresponding correlation weight coefficient alpha when the sensor lacks i data by using a least square method according to the complete data, the time prediction numerical value sequence and the correlation prediction numerical value sequenceiAnd a time weight coefficient betai。
In the embodiment of the invention, when the sensor to be repaired is intermittently missing a plurality of data, the data of the sensor to be repaired is converted into a plurality of groups of data, each group of data only comprises a section of continuously missing data, and data repair is respectively carried out according to the number of the continuously missing data in each group of data.
Preferably, the temporal prediction module adopts a cubic spline interpolation method.
The data restoration system for the multisource heterogeneous data of the highway infrastructure fuses the correlation prediction numerical value sequence and the time prediction numerical value sequence and restores the data according to the preset weight coefficients of the correlation prediction numerical value sequence and the time prediction numerical value sequence, so that missing data can be more accurately predicted, and the data restoration is more accurate.
The present invention is not limited to the above-described embodiments, and it will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the principle of the present invention, and such modifications and improvements are also considered to be within the scope of the present invention. Those not described in detail in this specification are within the skill of the art.
Claims (4)
1. A data restoration method for multi-source heterogeneous data of a road infrastructure is characterized by comprising the following steps:
s1: constructing a multi-source data association calculation model, and performing learning training on the multi-source data association calculation model by using complete multi-source heterogeneous data obtained by all sensors in advance;
s2: taking a sensor with missing data as a sensor to be repaired, acquiring incomplete data of the sensor to be repaired and complete data of other sensors, and correspondingly generating data sequences of all the sensors by combining time points of data acquisition of the sensors;
s3: inputting data sequences of other sensors into the multi-source data correlation calculation model to obtain a correlation prediction numerical value sequence Y of the missing data of the sensor to be repaired at a corresponding time pointr；
S4: interpolation calculation is carried out on the data sequence of the sensor to be repaired by using a spline interpolation method to obtain a time prediction numerical value sequence Y of the time point corresponding to the missing data of the sensor to be repairedp；
S5: according to the number i of continuous missing data of the sensor to be repaired and the corresponding preset associated weight coefficient alphaiAnd a time weight coefficient betaiIn combination with the calculation formula YResult＝αiYr+βiYpObtaining data Y to be repairedResult；
The corresponding preset associated weight coefficient alphaiAnd a time weight coefficient betaiThe calculation process of (2) is as follows:
selecting a section of complete data of all sensors;
creating a plurality of groups of data, wherein each group of data is the same as the complete data;
randomly selecting continuous i data in each group of data, removing the selected i data from each group of data, and correspondingly obtaining a plurality of groups of incomplete data missing i data;
calculating the predicted values of i missing data in each group of incomplete data by using a spline interpolation method, and forming the predicted values of the i data into a data prediction sequence to obtain time prediction numerical value sequences of all groups of incomplete data;
respectively inputting all groups of incomplete data into the multi-source data correlation calculation model to obtain correlation prediction numerical value sequences of all groups of incomplete data;
obtaining a corresponding correlation weight coefficient alpha when the sensor lacks i data by using a least square method according to the complete data, the time prediction numerical value sequence and the correlation prediction numerical value sequenceiAnd a time weight coefficient betai；
The specific process of using the multi-source heterogeneous data acquired by all the sensors in advance to carry out learning training on the multi-source data association calculation model comprises the following steps:
taking one sensor as a target sensor, taking complete multi-source heterogeneous data acquired by the target sensor as a target of the multi-source data association calculation model, taking complete multi-source heterogeneous data acquired by other sensors as input of the multi-source data association calculation model, and performing learning training on the multi-source data association calculation model;
when the sensor to be repaired is discontinuously lacked of a plurality of data, converting the data of the sensor to be repaired into a plurality of groups of data, wherein each group of data only comprises a section of continuously lacked data, and respectively repairing the data according to the number of the continuously lacked data in each group of data.
2. The method for restoring data of multisource heterogeneous data of road infrastructure according to claim 1, wherein the step S4 is a cubic spline interpolation method.
3. A data recovery system for multi-source heterogeneous data of a highway infrastructure, comprising:
a model building module to: constructing a multi-source data association calculation model, and performing learning training on the multi-source data association calculation model by using complete multi-source heterogeneous data obtained by all sensors in advance;
a data processing module to: taking a sensor with missing data as a sensor to be repaired, acquiring incomplete data of the sensor to be repaired and complete data of other sensors, and correspondingly generating data sequences of all the sensors by combining time points of data acquisition of the sensors;
a relevance prediction module to: inputting data sequences of other sensors into the multi-source data correlation calculation model to obtain a correlation prediction numerical value sequence Y of the missing data of the sensor to be repaired at a corresponding time pointr；
A temporal prediction module to: interpolation calculation is carried out on the data sequence of the sensor to be repaired by using a spline interpolation method to obtain a time prediction numerical value sequence Y of the time point corresponding to the missing data of the sensor to be repairedp；
A computing module to: according to the number i of continuous missing data of the sensor to be repaired and the corresponding preset associated weight coefficient alphaiAnd a time weight coefficient betaiIn combination with the calculation formula YResult＝αiYr+βiYpObtaining data Y to be repairedResult；
The calculation module further comprises a weight calculation unit,
the weight calculation unit is configured to:
selecting a section of complete data of all sensors;
creating a plurality of groups of data, wherein each group of data is the same as the complete data;
randomly selecting continuous i data in each group of data, removing the selected i data from each group of data, and correspondingly obtaining a plurality of groups of incomplete data missing i data;
calculating the predicted values of i missing data in each group of incomplete data by using a spline interpolation method, and forming the predicted values of the i data into a data prediction sequence to obtain time prediction numerical value sequences of all groups of incomplete data;
respectively inputting all groups of incomplete data into the multi-source data correlation calculation model to obtain correlation prediction numerical value sequences of all groups of incomplete data;
obtaining a corresponding correlation weight coefficient alpha when the sensor lacks i data by using a least square method according to the complete data, the time prediction numerical value sequence and the correlation prediction numerical value sequenceiAnd a time weight coefficient betai；
The specific process of using the multi-source heterogeneous data acquired by all the sensors in advance to carry out learning training on the multi-source data association calculation model comprises the following steps:
taking one sensor as a target sensor, taking complete multi-source heterogeneous data acquired by the target sensor as a target of the multi-source data association calculation model, taking complete multi-source heterogeneous data acquired by other sensors as input of the multi-source data association calculation model, and performing learning training on the multi-source data association calculation model;
when the sensor to be repaired is discontinuously lacked of a plurality of data, converting the data of the sensor to be repaired into a plurality of groups of data, wherein each group of data only comprises a section of continuously lacked data, and respectively repairing the data according to the number of the continuously lacked data in each group of data.
4. The system for data recovery of multi-source heterogeneous data of a road infrastructure of claim 3, wherein the time prediction module adopts a cubic spline interpolation method.
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 CN109992579A (en) | 2019-07-09 |
| CN109992579B true 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) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP4122165A1 (en) | 2020-03-16 | 2023-01-25 | British Telecommunications public limited company | Software defined network specification |
| GB2593173A (en) * | 2020-03-16 | 2021-09-22 | British Telecomm | Modelling physical infrastructure |
| 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 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| 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 |
Family Cites Families (4)
| 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 |
| CN102819682B (en) * | 2012-08-02 | 2015-01-14 | 清华大学 | Data cleaning and recovering algorithm of floating car related to multi-threshold space |
| US9461876B2 (en) * | 2012-08-29 | 2016-10-04 | Loci | System and method for fuzzy concept mapping, voting ontology crowd sourcing, and technology prediction |
| 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 |
-
2019
- 2019-03-28 CN CN201910244035.2A patent/CN109992579B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109992579A (en) | 2019-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109992579B (en) | Data restoration method and system for multisource heterogeneous data of highway infrastructure | |
| CN109726355B (en) | Ship track repairing method based on vector interpolation | |
| CN104599292A (en) | Noise-resistant moving target detection algorithm based on low rank matrix | |
| CN110781266A (en) | Urban perception data processing method based on time-space causal relationship | |
| CN106651731B (en) | Communication tower to-be-solved problem set generation method and system based on historical data | |
| CN115951014A (en) | CNN-LSTM-BP multi-mode air pollutant prediction method combining meteorological features | |
| CN115600051B (en) | Intelligent track maneuvering detection method and device based on short arc space-based optical observation | |
| CN104156579A (en) | Dynamic traffic abnormal data detection and recovery method | |
| ATE478399T1 (en) | METHOD FOR COUNTING OBJECTS IN A MONITORED ENVIRONMENT AND APPARATUS THEREOF | |
| CN115267945A (en) | Thunder and lightning early warning method and system based on graph neural network | |
| Kordmahalleh et al. | Hurricane trajectory prediction via a sparse recurrent neural network | |
| CN105512464A (en) | Method of combining satellite and sites for observation and retrieval of space-time continuous PM2.5 concentration | |
| CN105627978A (en) | Rainy season land slide deformation monitoring method and rainy season land slide deformation monitoring system | |
| CN114861774A (en) | False data identification method and system in power grid | |
| CN109145773A (en) | A kind of failure prediction method of multi-source tendency information fusion | |
| CN112948462A (en) | Ultra-short-term wind speed prediction method based on improved singular spectrum analysis and Bp neural network | |
| CN117371571A (en) | Regional air quality prediction model based on multi-scale dynamic synchronous diagram mechanism | |
| CN113687192B (en) | Method for collecting and positioning discharge signal of power transmission line | |
| CN114611728B (en) | Sewage pipe network blockage monitoring method and system | |
| CN108121878A (en) | The impulsive neural networks model of autocoding source signal | |
| CN101984560A (en) | Centralized multi-source combined Viterbi data interconnection tracker | |
| CN113810226B (en) | Wide area network single-point flow prediction method combining discrete feature coding and cluster correction | |
| CN112419751B (en) | Signalized intersection lane queuing length estimation method based on single-section electric alarm data | |
| CN109946096B (en) | High-speed train air pipe fault diagnosis method based on model space | |
| CN106918677A (en) | For the apparatus and method that pollutant is traced to the source |
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 |