CN109871412A - Lane flow analysis method based on K-Means cluster - Google Patents
Lane flow analysis method based on K-Means cluster Download PDFInfo
- Publication number
- CN109871412A CN109871412A CN201811600393.4A CN201811600393A CN109871412A CN 109871412 A CN109871412 A CN 109871412A CN 201811600393 A CN201811600393 A CN 201811600393A CN 109871412 A CN109871412 A CN 109871412A
- Authority
- CN
- China
- Prior art keywords
- cluster
- value
- data
- point
- class
- 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.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A kind of lane flow analysis method based on K-Means cluster includes the following steps: the vehicle flowrate data excessively for 1) obtaining certain road cross;2) data in certain week are obtained into each flow curve in certain week as smoothing processing with the method for sliding average;3) setting initial maximum clusters number of clusters, traverses each number of clusters value with the method for K-Means cluster;4) to cluster result processing, a number of clusters value is obtained with ancon rule;5) using this number of clusters value as cluster value, all Clustering Effects a certain to the end are clustered to obtain again with K-Means method.The present invention realizes that the cluster of single-revolution can distribute adaptive scheduling scheme to specific date even special time period, more to adapt to the timing scheme of the crossing wagon flow in road cross operation, the congestion of road is effectively relieved in the optimal adjustment effect for reaching wagon flow.
Description
Technical field
The present invention relates to traffic control engineerings, big data analysis application field, obtain especially with the method for sliding average
Matched curve, by being analyzed based on K-Means clustering algorithm daily flow curve, obtaining certain crossing can optimal adaptation timing
Scheme.
Background technique
With the development of urban economy, car population constantly rises per capita, and the congestion problems of road traffic are increasingly prominent,
As the continuous propulsion of urban infrastructure is also extremely urgent for the optimization of the timing scheme of belisha beacon.Due to lane vehicle
Rapidity, complexity and the uncertainty of changes in flow rate, wherein showing more outstanding was influenced by periods such as early evening peak
Greatly, vehicle flowrate variation is more obvious, frequent, complicated so as to cause city road net traffic state variation, adjustment problem, often in mixed
It closes in traffic behavior.And it is now based on classification, analysis and calculating of the K-Means clustering algorithm to single-revolution odd-numbered day vehicle flowrate, it will be right
Belisha beacon timing scheme does effective adjustment.Therefore in order to adapt at the crossing that specific date vehicle flowrate has significant change
Complicated traffic, the present invention proposes that the vehicle flowrate of crossing to certain section of Dates Study every day obtains daily flow line chart, then leads to
The method for crossing sliding average is fitted to a smooth curve, recycles K-Means algorithm to carry out curve sub-category, to adapt to not
Same timing scheme.Finally, using most suitable control strategy to reach the optimization of crossing wagon flow scheduling, mitigate the specific date
The road congestion conditions of special time period then realize the purpose of city road network coordinated control benefit.
Summary of the invention
In order to cook up more suitable for the specific timing scheme in certain date crossing, so as to better optimization of road joints traffic
Congestion problems accomplish the allocation plan for having personalized to each crossing.In view of the rapidity, random of city road network traffic flow amount
Property, complexity and period, every month, either large or small variation, previous period section timing can all occur for daily flow weekly
Scheme is not necessarily suitble to the flow of next period, it is therefore desirable to crossing flow analysis and feed back, in time to scheme into
The adjustment of Mobile state.The invention proposes the date of similar flow distribution is classified as same class using clustering algorithm, obtain more
Add accurate traffic classification result.
The technical solution adopted by the present invention to solve the technical problems is:
A kind of lane flow analysis method based on K-Means clustering algorithm, includes the following steps:
1) get intersection in the database crosses vehicle flowrate tables of data, counts a certain regulation step according to car data is crossed
The vehicle number set passed through in long time interval, is denoted as Pn, unit is vehicle number;N is 1,2,3 ... n;
The smoothed curve that method based on sliding average is fitted to obtain daily vehicle flowrate uses simple sliding average
Method is as follows:
Wherein At-nThe actual numerical value of Shi Dangqian, n are the number of sliding average, FtIt is after sliding average to next prediction
Value, the calculation method of simple sliding average are as follows:
2) set a certain maximum the number of iterations as n_max, the numerical value have to be larger than or be equal to after cluster it is getable most
Big number of clusters is then based on K-Means clustering algorithm, is clustered with given iteration cluster value, process is as follows:
2.1) given input value be the database comprising n data object and finally need to cluster cluster number K it
Afterwards, K-Means algorithm can iterate operation according to some distance function, until data are divided into K cluster, finally
When target function value convergence, final cluster result is obtained;Find the central point { c of K cluster1,c2,…,ck, thus
Make each data object point xiTo nearest cluster centre μiSquare distance and minimum;
2.2) initialize first to it: it has been known that there is the database of a n target sample, which is calculated as X=
{x1,x2,x3,…,xj,…,xn, wherein xj∈ R is randomly assigned the set C={ c of K initial cluster center point1,c2,…,
ck, each it is divided into a class ck, each class has a cluster centre point μi, it is denoted as cluster;K-Means algorithm is with Europe
Formula distance is surveyed as similarity, it is to seek corresponding a certain initial cluster center vector V optimal classification, so that evaluation index J reaches
The foundation of minimum judgement, using error sum of squares criterion function as clustering criteria function, so calculating every class sample point to cluster
Center μiSquare distance and;
2.3) second step, sample distribution: for the data point in data sample, according to them at a distance from cluster centre,
Using the criterion apart from minimum value as classification, they are distributed to and the highest cluster centre μ of its similarity respectivelyiRepresentative
Cluster;
2.4) it corrects cluster centre: calculating new cluster centre C of the mean value as class of all data points in each classification*,
Wherein niIt is sample set ciIn sample number;
2.5) calculate deviation size: wherein D be deviation, that is, data acquisition system class to cluster centre square distance with;
2.6) it checks whether to restrain, if deviation convergence stops algorithm while returning to the value of cluster centre, i.e.,
Reach convergent state, otherwise returns to step 2.3);
3) after the complete maximum cluster number of iteration, the cluster situation at each cluster value k is obtained, is obtained using ancon rule
And average discrete degree maximum to slope variation is K value corresponding to minimum point, and average discrete degree is exactly that data point arrives
The quadratic sum of the Euclidean distance of the cluster centre and the ratio of data point total number, the corresponding K value in position at inflection point, also
It is above-mentioned ancon, is just used as relatively optimal cluster numbers magnitude;
4) the opposite best clustering cluster numerical value judged using ancon rule is as new K value, then carries out second
Based on the algorithm of K-Means cluster, cluster result is finally exported;
5) optimal scheduling scheme is provided with reference to cluster result.
The invention has the benefit that obtaining more accurate traffic classification result.
Detailed description of the invention
Fig. 1 is the logical flow chart of K-Means clustering algorithm;
Fig. 2 is a certain practical road network schematic diagram of Taizhou of Zhejiang;
Fig. 3 is the curve graph based on simple sliding average algorithm;
Fig. 4 is the K value iteration line chart using the observation of ancon rule;
Fig. 5 is the cluster result schematic diagram based on K-Means clustering algorithm.
Specific embodiment
The invention will be further described below in conjunction with the accompanying drawings.
Referring to Fig.1~Fig. 5, a kind of lane flow analysis method based on K-Means cluster, comprising the following steps:
The first step, need to prepare m days certain crossing crosses vehicle flowrate tables of data, counts a certain regulation step according to car data is crossed
Long time interval Δ t, by vehicle number Pn, draw corresponding two-dimentional flow diagram.Recycle the method fitting of sliding average
M days m smoothed curves are obtained, wherein the method for sliding average has: simple moving average method and weighted moving average, it is believed that
Per unit time, that is, the weight of each step-length are all equal, so using the method for simple sliding average to be fitted
Curve, wherein At-1It is actual numerical value, n is the number of sliding average, FtIt is after sliding average to next predicted value, simply
The calculation method of sliding average is as follows:
Second step, set a certain maximum the number of iterations as n_max, which has to be larger than or equal to obtaining after cluster
Maximum number of clusters is then based on K-Means clustering algorithm, is clustered with given iteration cluster value;
1) K-Means algorithm is also referred to as K mean algorithm, is the database comprising n data object in given input value
After the number K for finally needing to cluster cluster, K-Means algorithm can iterate operation according to some distance function,
Until data are divided into K cluster, finally when target function value is restrained, final cluster result is obtained.In the algorithm
Thought wants the central point { c of K cluster to be found1,c2,…,ck, to make each data object point xiGather to nearest
Class center μiSquare distance and minimum;2) it is initialized, it is known to the database of a n target sample, the data set
Add up to X={ x1,x2,x3,…,xj,…,xn, wherein xj∈ R is randomly assigned the set C=of K initial cluster center point
{c1,c2,…,ck, each it is divided into a class ck, each class has a cluster centre point μi, it is denoted as cluster, K-
Means algorithm is surveyed using Euclidean distance as similarity, it be seek corresponding a certain initial cluster center vector V optimal classification so that
Evaluation index J reaches the foundation of minimum judgement, and algorithm uses error sum of squares criterion function as clustering criteria function, so meter
Every class sample point is calculated to cluster center μiSquare distance and;
3) sample distributes, minimum with distance according to them at a distance from cluster centre for the data point in data sample
It is worth the criterion as classification, respectively distributes to them and the highest cluster centre μ of its similarityiRepresentative cluster;
4) cluster centre is corrected, new cluster centre C of the mean value as class of all data points in each classification is calculated*。
Wherein niIt is sample set ciIn sample number;
5) calculate deviation size, wherein D be deviation, that is, data acquisition system class to cluster centre square distance with.
6) it checks whether to restrain, if deviation convergence stops algorithm while returning to the value of cluster centre, that is, reach
To convergent state, otherwise repeatedly second step sample is distributed;
Third step obtains the cluster situation at each cluster value k, utilizes ancon after the complete maximum cluster number of iteration
Rule obtains that slope variation is maximum and average discrete degree is K value corresponding to minimum point.Average discrete degree is exactly to count
Strong point to the cluster centre Euclidean distance quadratic sum and data point total number ratio, generally speaking, with the increasing of K value
Add, downward trend is presented with the longitudinal axis of average discrete distance and finally tends towards stability, then the corresponding K in the position at inflection point
Value, that is, above-mentioned ancon are just used as relatively optimal cluster numbers magnitude;
4th step, the opposite best clustering cluster numerical value judged using ancon rule is as new K value, then carries out
The secondary algorithm based on K-Means cluster, finally exports cluster result;
5th step provides corresponding optimal scheduling scheme with reference to cluster result
The present embodiment is using a certain crossing in the practical road network region in Taizhou plain city as examples of implementation, road network as shown in Figure 2, with
Continuous seven days of on May 30, of 24 days to 2017 May in 2017, in the vehicle data conduct that the crossing artificially marked as No. 77 passes through
Example is demonstrated, and the crossing marked as No. 77 is a certain crossing in Fig. 2, handles lane flow with K-means clustering algorithm
Result, comprising the following steps:
1) first step obtains in seven days of on May 30th, 24 days 1 May in 2017 in the mistake wagon flow at No. 77 crossings
Measure tables of data, 288 units will be divided within 24 hours one day as step-length using 5 minutes, that is, according to mistake car data table every
The primary vehicle quantity excessively of five minutes statistics draws line chart according to according to coordinate, is then intended using the method for simple sliding average
Resultant curve, wherein At-1It is actual numerical value, n is the number of sliding average, and the number of sliding average is set as n=by this example
6, the then F that sliding average obtainstIt is exactly after sliding average to next predicted value, the calculation method of this simple sliding average
It is as follows:
Then curve fitting obtained is as the flow curve in one day.It can be obtained with this seven days crosses in car data
Seven be fitted after curve, the curve being fitted is i.e. shown in Fig. 3;
2) then obtained curve is clustered using the method for K-means.It has been known that there is the numbers of a n target sample
According to library, which is denoted as X={ x1,x2,x3,…,xj,…,xn, wherein xj∈ R has been randomly assigned in K initial clustering
Set C={ the c of heart point1,c2,…,ck, each it is divided into a class ck, each class has a cluster centre point μi, remembered
For cluster.For K-Means algorithm using Euclidean distance as similarity measure, it is to seek corresponding a certain initial cluster center vector V most
Optimal sorting class, so that evaluation index J reaches the foundation of minimum judgement, algorithm is using error sum of squares criterion function as clustering criteria
Function, so calculating every class sample point to cluster center μiSquare distance and;
3) sample distributes, minimum with distance according to them at a distance from cluster centre for the data point in data sample
It is worth the criterion as classification, respectively distributes to them and the highest cluster centre μ of its similarityiRepresentative cluster;
4) size for calculating deviation, calculates new cluster centre of the mean value as class of all data points in each classification
C*, wherein niIt is sample set ciIn sample number;
5) calculate deviation size, wherein D be deviation, that is, data acquisition system class to cluster centre square distance with;
6) it checks whether to restrain, if deviation convergence stops algorithm while returning to the value of cluster centre, that is, reach
To convergent state, otherwise repeatedly 3) step sample is distributed;
7) after the complete maximum cluster number of iteration, the cluster situation at each cluster value K is obtained, as shown in figure 4, utilizing
Ancon rule obtains that slope variation is maximum and average discrete degree is K value corresponding to minimum point, the position at inflection point
Corresponding K value, that is, above-mentioned ancon are just used as relatively optimal cluster numbers magnitude, are with the optimum k value that this example obtains
3, as given clustering cluster value K=3, then second of algorithm based on K-Means cluster is carried out, finally exports cluster result;
8) curve after cluster is completed is finally obtained, 7 days vehicle flowrates excessively can be roughly divided into three classes, the first kind are as follows:
On May 24th, 2017,25 days, 26 days (Wednesday, four, five), the second class are May 27 (Saturday) in 2017, and third class is 2017
On May 28,29 days, 30 days (Sunday, one, two), Clustering Effect is as shown in Figure 5.
Described above is the excellent results that one embodiment that the present invention provides shows, it is clear that the present invention not only fits
Above-described embodiment is closed, it can under the premise of without departing from essence spirit of the present invention and without departing from content involved by substantive content of the present invention
Many variations are done to it and are implemented.
Claims (1)
1. a kind of lane flow analysis method based on K-Means cluster, which is characterized in that described method includes following steps:
1) get intersection in the database crosses vehicle flowrate tables of data, counts a certain regulation step-length according to car data is crossed
Time interval Δ t, unit are the second, by vehicle number set, be denoted as Pn, unit is vehicle number;N is 1,2,3 ... n.It is based on
The method of sliding average is fitted to obtain the smoothed curve of daily vehicle flowrate, wherein At-nThe actual numerical value of Shi Dangqian, n are that sliding is flat
Equal number, FtIt is after sliding average to next predicted value, the calculation method of simple sliding average is as follows:
2) a certain maximum the number of iterations is set as n_max, which has to be larger than or be equal to getable maximum cluster after cluster
Number, is then based on K-Means clustering algorithm, is clustered with given iteration cluster value;
2.1) after given input value is the database comprising n data object and finally needs to cluster the number K of cluster, K-
The operation that can be iterated according to some distance function of Means algorithm until data are divided into K cluster finally works as mesh
When offer of tender numerical value is restrained, final cluster result is obtained, finds the central point { c of K cluster1,c2,…,ck, to make each
A data object-point xiTo nearest cluster centre μiSquare distance and minimum;
2.2) it is initialized first.It has been known that there is the database of a n target sample, which is calculated as X={ x1,x2,
x3,…,xj,…,xn, wherein xj∈ R is randomly assigned the set C={ c of K initial cluster center point1,c2,…,ck, often
It is a to be divided into a class ck, each class has a cluster centre point μi, it is denoted as cluster;K-Means algorithm is with Euclidean distance
It is surveyed as similarity, it is to seek corresponding a certain initial cluster center vector V optimal classification, is sentenced so that evaluation index J reaches minimum
Disconnected foundation;Using error sum of squares criterion function as clustering criteria function, so calculating every class sample point to cluster center μi
Square distance and;
2.3) second step, sample distribution, for the data point in data sample, according to them at a distance from cluster centre, with away from
Criterion from minimum value as classification, respectively distributes to them and the highest cluster centre μ of its similarityiRepresentative cluster;
2.4) cluster centre is corrected, new cluster centre C of the mean value as class of all data points in each classification is calculated*, wherein ni
It is sample set ciIn sample number;
2.5) calculate deviation size, wherein D be deviation, that is, data acquisition system class to cluster centre square distance with;
2.6) it checks whether to restrain, if deviation convergence stops algorithm while returning to the value of cluster centre, that is, have reached
Convergent state, otherwise return step 2.3);
3) after the complete maximum cluster number of iteration, the cluster situation at each cluster value k is obtained, is obtained tiltedly using ancon rule
It is K value corresponding to minimum point that rate, which changes maximum and average discrete degree, and average discrete degree is exactly that data point is poly- to this
The quadratic sum of the Euclidean distance at class center and the ratio of data point total number, the corresponding K value in position at inflection point, that is, on
The ancon stated just is used as relatively optimal cluster numbers magnitude;
4) the opposite best clustering cluster numerical value judged using ancon rule is as new K value, then is based on for the second time
The algorithm of K-Means cluster, finally exports cluster result;
5) optimal scheduling scheme is provided with reference to cluster result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811600393.4A CN109871412A (en) | 2018-12-26 | 2018-12-26 | Lane flow analysis method based on K-Means cluster |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811600393.4A CN109871412A (en) | 2018-12-26 | 2018-12-26 | Lane flow analysis method based on K-Means cluster |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109871412A true CN109871412A (en) | 2019-06-11 |
Family
ID=66917098
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811600393.4A Pending CN109871412A (en) | 2018-12-26 | 2018-12-26 | Lane flow analysis method based on K-Means cluster |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109871412A (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110598755A (en) * | 2019-08-19 | 2019-12-20 | 北京建筑大学 | OD flow clustering method based on vector constraint |
CN110634287A (en) * | 2019-08-26 | 2019-12-31 | 上海电科智能***股份有限公司 | Urban traffic state refined discrimination method based on edge calculation |
CN110766043A (en) * | 2019-09-10 | 2020-02-07 | 国网浙江省电力有限公司杭州供电公司 | K-means clustering algorithm based on power grid information data |
CN110782667A (en) * | 2019-10-30 | 2020-02-11 | 北京百度网讯科技有限公司 | Signal lamp time-sharing timing method and device, electronic equipment and storage medium |
CN110796159A (en) * | 2019-09-12 | 2020-02-14 | 国网浙江省电力有限公司杭州供电公司 | Power data classification method and system based on k-means algorithm |
CN111315090A (en) * | 2020-02-24 | 2020-06-19 | 广西杰盛科技有限公司 | Wisdom city system based on street lamp |
CN111966951A (en) * | 2020-07-06 | 2020-11-20 | 东南数字经济发展研究院 | User group hierarchy dividing method based on social e-commerce transaction data |
CN112131606A (en) * | 2020-09-24 | 2020-12-25 | 合肥城市云数据中心股份有限公司 | Dynamic data difference privacy histogram publishing method based on K-means + + combined elbow method autonomous clustering technology |
CN112562336A (en) * | 2020-12-10 | 2021-03-26 | 重庆交通大学 | Real-time dynamic allocation method and system for intersection traffic signal duration |
CN113390925A (en) * | 2021-05-27 | 2021-09-14 | 中国农业大学 | Method for quantitatively detecting nitrogen in soil based on electrical impedance |
CN114779467A (en) * | 2022-04-27 | 2022-07-22 | 吉林大学 | Novel spectrometer membrane system combination selection method based on detector characteristics |
CN115310879A (en) * | 2022-10-11 | 2022-11-08 | 浙江浙石油综合能源销售有限公司 | Multi-fueling-station power consumption control method based on semi-supervised clustering algorithm |
CN116360956A (en) * | 2023-06-02 | 2023-06-30 | 济南大陆机电股份有限公司 | Data intelligent processing method and system for big data task scheduling |
CN116796214A (en) * | 2023-06-07 | 2023-09-22 | 南京北极光生物科技有限公司 | Data clustering method based on differential features |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1876757A1 (en) * | 2006-07-06 | 2008-01-09 | Alcatel Lucent | Effective bandwidth estimation of a bursty packet stream |
CN106408960A (en) * | 2016-10-29 | 2017-02-15 | 浙江大学 | Signal control time period dividing method based on ordered clustering |
CN106600965A (en) * | 2017-01-19 | 2017-04-26 | 上海理工大学 | Sharpness-based automatic identification method for morning and evening peak periods of traffic flows |
CN106875681A (en) * | 2017-03-22 | 2017-06-20 | 山东大学 | A kind of traffic congestion tendency Forecasting Methodology and system |
CN108492557A (en) * | 2018-03-23 | 2018-09-04 | 四川高路交通信息工程有限公司 | Highway jam level judgment method based on multi-model fusion |
-
2018
- 2018-12-26 CN CN201811600393.4A patent/CN109871412A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1876757A1 (en) * | 2006-07-06 | 2008-01-09 | Alcatel Lucent | Effective bandwidth estimation of a bursty packet stream |
CN106408960A (en) * | 2016-10-29 | 2017-02-15 | 浙江大学 | Signal control time period dividing method based on ordered clustering |
CN106600965A (en) * | 2017-01-19 | 2017-04-26 | 上海理工大学 | Sharpness-based automatic identification method for morning and evening peak periods of traffic flows |
CN106875681A (en) * | 2017-03-22 | 2017-06-20 | 山东大学 | A kind of traffic congestion tendency Forecasting Methodology and system |
CN108492557A (en) * | 2018-03-23 | 2018-09-04 | 四川高路交通信息工程有限公司 | Highway jam level judgment method based on multi-model fusion |
Non-Patent Citations (1)
Title |
---|
蒋帅: "K-均值聚类算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110598755B (en) * | 2019-08-19 | 2022-07-15 | 北京建筑大学 | OD flow clustering method based on vector constraint |
CN110598755A (en) * | 2019-08-19 | 2019-12-20 | 北京建筑大学 | OD flow clustering method based on vector constraint |
CN110634287B (en) * | 2019-08-26 | 2021-08-17 | 上海电科智能***股份有限公司 | Urban traffic state refined discrimination method based on edge calculation |
CN110634287A (en) * | 2019-08-26 | 2019-12-31 | 上海电科智能***股份有限公司 | Urban traffic state refined discrimination method based on edge calculation |
US11361658B1 (en) | 2019-08-26 | 2022-06-14 | Shanghai Seari Intelligent System Co., Ltd. | Edge computing-based method for fine determination of urban traffic state |
CN110766043A (en) * | 2019-09-10 | 2020-02-07 | 国网浙江省电力有限公司杭州供电公司 | K-means clustering algorithm based on power grid information data |
CN110796159A (en) * | 2019-09-12 | 2020-02-14 | 国网浙江省电力有限公司杭州供电公司 | Power data classification method and system based on k-means algorithm |
CN110782667A (en) * | 2019-10-30 | 2020-02-11 | 北京百度网讯科技有限公司 | Signal lamp time-sharing timing method and device, electronic equipment and storage medium |
US11527155B2 (en) | 2019-10-30 | 2022-12-13 | Apollo Intelligent Connectivity (Beijing) Technology Co., Ltd. | Time-segmented signal timing method and apparatus for traffic light, electronic device and storage medium |
CN111315090A (en) * | 2020-02-24 | 2020-06-19 | 广西杰盛科技有限公司 | Wisdom city system based on street lamp |
CN111966951A (en) * | 2020-07-06 | 2020-11-20 | 东南数字经济发展研究院 | User group hierarchy dividing method based on social e-commerce transaction data |
CN112131606A (en) * | 2020-09-24 | 2020-12-25 | 合肥城市云数据中心股份有限公司 | Dynamic data difference privacy histogram publishing method based on K-means + + combined elbow method autonomous clustering technology |
CN112562336A (en) * | 2020-12-10 | 2021-03-26 | 重庆交通大学 | Real-time dynamic allocation method and system for intersection traffic signal duration |
CN112562336B (en) * | 2020-12-10 | 2022-11-04 | 重庆交通大学 | Real-time dynamic allocation method and system for intersection traffic signal duration |
CN113390925A (en) * | 2021-05-27 | 2021-09-14 | 中国农业大学 | Method for quantitatively detecting nitrogen in soil based on electrical impedance |
CN114779467A (en) * | 2022-04-27 | 2022-07-22 | 吉林大学 | Novel spectrometer membrane system combination selection method based on detector characteristics |
CN115310879A (en) * | 2022-10-11 | 2022-11-08 | 浙江浙石油综合能源销售有限公司 | Multi-fueling-station power consumption control method based on semi-supervised clustering algorithm |
CN115310879B (en) * | 2022-10-11 | 2022-12-16 | 浙江浙石油综合能源销售有限公司 | Multi-fueling-station power consumption control method based on semi-supervised clustering algorithm |
CN116360956A (en) * | 2023-06-02 | 2023-06-30 | 济南大陆机电股份有限公司 | Data intelligent processing method and system for big data task scheduling |
CN116360956B (en) * | 2023-06-02 | 2023-08-08 | 济南大陆机电股份有限公司 | Data intelligent processing method and system for big data task scheduling |
CN116796214A (en) * | 2023-06-07 | 2023-09-22 | 南京北极光生物科技有限公司 | Data clustering method based on differential features |
CN116796214B (en) * | 2023-06-07 | 2024-01-30 | 南京北极光生物科技有限公司 | Data clustering method based on differential features |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109871412A (en) | Lane flow analysis method based on K-Means cluster | |
Wang et al. | The system of revenue sharing and fiscal transfers in China | |
CN100444210C (en) | Mixed controlling method of single dot signal controlling crossing | |
CN109800916B (en) | Modeling method for traffic flow of driving-in traffic in expressway service area | |
CN106504535B (en) | A kind of trip distribution modeling method of combination Gravity Models and Fratar models | |
CN111814079B (en) | Crowd sensing Internet of vehicles user screening method combining clustering and CMAB | |
CN107508287B (en) | Electricity grid substation load grouping method, device, storage medium and computer equipment | |
CN113744527B (en) | Intelligent targeting dredging method for highway confluence area | |
Bogue et al. | Migration and distance | |
CN105551250B (en) | A kind of urban road intersection operating status method of discrimination based on interval clustering | |
CN105374204B (en) | A kind of method that urban highway traffic detector is layouted | |
CN102200759A (en) | Nonlinear kernelled adaptive prediction method | |
Lin et al. | Tourism demand forecasting: Econometric model based on multivariate adaptive regression splines, artificial neural network and support vector regression | |
CN105427002A (en) | Logistic model-based passenger flow prediction method | |
CN115691138B (en) | Road network subarea division and subarea boundary flow control method | |
CN106202388B (en) | A kind of user gradation Automated Partition Method and system | |
CN110264007A (en) | Stock exchange control method and device | |
CN110516871A (en) | A kind of dynamic vehicle method for optimizing route based on fuzzy roll stablized loop strategy | |
Kulczycki et al. | Methodically unified procedures for a conditional approach to outlier detection, clustering, and classification | |
CN107316096A (en) | A kind of track traffic one-ticket pass passenger amount of entering the station Forecasting Methodology | |
CN102521271B (en) | Upscaling conversion method for discrete spatial data | |
CN109541729A (en) | A kind of prediction technique of the grassland in northern China area precipitation during growing season based on NARX | |
CN108898322A (en) | A kind of electric grid investment strategy benefit integrated evaluating method based on FCM algorithm | |
Kisiała et al. | Equalising or polarising: The centre–periphery model and the absorption of EU funds under regional operational programmes in Poland | |
CN108346287A (en) | Magnitude of traffic flow sequence pattern matching process based on analysis of Influential Factors |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190611 |
|
WD01 | Invention patent application deemed withdrawn after publication |