ӭ
ղرվ
 
 
ǰλ<<վҳ<<ϸϢ

ʱηֲн콻վֲ㷨

ʱ䣺2016-12-21 1904


±ţ                              ͼţU491.12                  ױ־룺A

ʱηֲн콻վֲ㷨

Сз

ͬôѧ·뽻ͨ̽صʵңϺ 201804

ժҪнͨվķʱԤ⣬ϵӪƻĵ˿ͳзʽѡ񡢳ʱԤȣн·ģʽ¿зŻҪ塣зʽǶȷõۣ˳˿͹·γɵĽǶȷ,˽Ƕȷģ˿͵ʧVPnȷﵽֵµij˿ʧ,AFCȡĿΪ֧ţϽǶȷģͶԳ˿ʧļ,һֻʱεվֲԤⷽ,ŷվ̨򳵿ͨгʵؿ֮Ŀɣ˺򳵿-гؿӰ춯̬ģͣоģ㷨ijн·ʵ㣬ԤΪͨзŻṩۺͷ֧,ӪƻʱԤ⼰ģ͵IJṩο

ؼʣ·䣻Ԥ⣻̬ģͣģʽнͨ

Algorithms of Station Passenger Flow Forecast of Suburban Rail Transit Based on Distribution Time

Ding Xiao bing, Xu Xing fang

(The Key Laboratory of Road and Traffic Engineering, Ministry of Education, Shanghai, 201804)

Abstract: The stranded passenger flow forecasting of suburban rail station is related to the adjustment of operation plan, trip mode choice of passengers, and prediction of travel time, especially for operation scheme optimization based on fast-slow mode is of great importance. First of all, the theory of angle expenses was introduced, from the cost of the way of transit trip and subway lines, which determined the passenger choice. Used the passenger flow data obtained from AFC as data support, presented a forecasting method of passenger flow based on periods distribution. Secondly, analyzed the law of passenger flow and traffic exchange platform based on train passenger status, proposed the waiting passenger train ridership interaction model, and studied the algorithm. Finally, validate a suburban line, which showed that the accuracy of the prediction results improved. This can be as supplementary for railway traffic planning, guidance for passenger flow, and be of certain reference and practical.

Key words: railway transportationpassenger flow forecastdynamic interaction model of passenger flowexpress-local modeSuburban-urban line

1

йлȡóɹͬʱҲͬһĺ⣺شеӵԽأɱߣصλУǨȣн֮Ŀ֮нͨ·н·ͬ·Ÿǿʱշֲԣ߷峱ϫԣƽʱڿСվԡ߷岿ֳվ󣬸н·Ŀ֯гȫѹӰͨˮƽˣн·վĿʱԤ⣬ΪгƻļʱԼճָṩݡ

[1]ͨʱԷ˻ʱмȨģԤ߽ϴ㷨֮ۺ뽻ǽоҪơ[2]ͨ·ʵʱ̬ͨԤ·OD·ֵ·УδΪ̬޷ȡ·վ̨гϿʱֲ״,ʵʱϻоĿռƵ[3-5]ͳԤΧ˷ܽҪԤģعģʷ ƽģʱģʷƽģӦӳ֮ȷʱģȻʱ©ȱģʼĵҲΪһȡѧöԤԤԤȷȡõԤЧ

ͨ·վ̨ϵķӰվ̨ϺͼվǰʵؿͬվͨվԤҪϵͳ˿ؿǶȷó˿ʧʡ˿ȴʱ֮ϵ[6,7]ģӪʵʱԤҪҪһ̽о

ͨAFCվѰ˿̬ϵϻ·ĽǶȷģͣо˿͵ʧʼͨʽѡƫãӶվгģͣԤ⳵վʱοֲվ̨ԤΪͨзŻṩۺͷ֧,ӪƻʱԤ⼰ģ͵IJṩο

2 н·

н·ͨѾ߱һģĹ·ԽվгؿԤⲻ׼վ̨ӵ˿࣬ʧȣȫ⡣ˣнĻ·ѡΪн·ʱοԤҪ塣

2.1վɷֲ

Կվʱεǡ֣վ̨ɿɽΪӸֲָ

Ϊһ̣ͬķֲķǸʵ໥㹻СУ

ǷǿΪIJɹ

ͨ£н˿ͻ°гʱ̱վ򳵣ڿģʽ£˿вͬͣվµĶгοɹѡ񡣽˿ͺʱ䶨Ϊ˿͵վʱijһгʱֹʱ䡣

Ϊ˿ܼʱһг赱˿͵Ϊʱܵdz˸ôг

λ˿͵ѧΪôгȴʱѧ

                                               1

гʱΪУгΪ

ݷֲ壬ʾλ˿ʱǰ֣ٵλ˿ͣУ      2

ʾгԱϿʾгʣɷֱÿгΪгš

2.2˿·ѡ

˿͵·ѡϵͨij˿Ӱ쵽߷վij˿бҪԳ˿͵·ѡΪ趨 3

(1)ӵΪnʵʣΪгʱij˻һӳӵöԳ˿·ѡƫӰı仯

(2)ĻʱΪ߶Գ˿·ѡƫӰǶģ֮໥Ӱ졣

(3)ó·ʵֱƫ붨ΪǶȷӳ˿ѡƫ·ƫ̶ı仯ӳѡƫӦ淽ƫʵ[8-11]3ͬʱֵƫһ 0·ƫͼ 1ʾʽ3ʾ

ͼ1  Ƕȷʾ

Fig.1 Example of angle cost

                  3

ʾODԼm,յnĵk·ĽǶȷãλΪkmΪʵʳȣλΪkmΪƫյ㷽ƫǶȣȡֵΧΪΪܺ͡

ƵǶȷֵı仯ӦƫǶȵĵԣͼ1ʾΪijͨ·IJͼ·OO1-12-23-3DյD䳤ȷֱΪƫǶȷֱΪ

ֱã

ǶȷֵƫǶȵӶʱ仯ƫǶȵĵ˿ͶԽͨзʽѡĽǶȷӰʵ

ͨͷ˿ͳѡģͣ

У1~Ϊվ1վΪǷöµij˿ʧʣ   ֱʾվʵʺʵؿгԱλΪˣֱʾǶȷµijо롢ͨ·ʵʾ롣

3 ʱοԤģ͹㷨

3.1ģ

(1)н·ṹΪվ㼰

(2)гмƻ·ȫ쿪ʱ䡣ȫ췢Ϊ߷ƽʱ䡣Ϊг뿪վ̨ʱ̣ʾгվ̨ļʱ䡣

(3) վ̨˿ʵʱɣվ̨˿͵ʵʱɿͨվբȡָʱεĽվݣԲɼݴ벴ģͷ㡣

(4) г˿Ͷѡ񣬼˿ؽǶȷۣӵùƫǶȽСʱ˿Ϳѡзʽ˴Եʽͨӵͨʱ˿ɹתͨʣͨС¼

(5) վ̨ʱε˿ڸгϵķʾñеĵУֵ京Ϊվ̨ȷʱڵij˿гϵķ

3.2г-վ̨ʵʱԤģ͹

1 

վ˿͵Ԥ복վij˿гʽأ򹹽Գ˿͵Ϊĺ¼ijʱεij˿ͽվĿ

1 ʱεĽվϢ

Table 1 the amount of passengers in distribution time

ʱηֲ

ʣ

վ1

վ2

վ3

վ4

վj

6:00-9:00

9:00-17:00

17:00-19:00

19:00-

2г-վ̨˿ͷֲԤģ͹

ڸгijվʱʵʿؿվ̨򳵳˿ڲֵԸ гʵϿҲͬԸгվ̨ϳıΪѰҹɣ

гվ̨ϿͱΪΪʵϳ˿гͣվǰվ̨ȴ˿֮ȡһʱڣվ̨ij˿һгϵķΪ

                       4

ڼ2ʱڵվ̨һгؿͺʣij˿ɱʾΪ

ڹģʱվ̨ϣij˿½복վij˿ǻϾȷֲģվ̨δȡκκ֯ʩ˿ͬϳ

3.3ʵʱԤģ㷨

Step1ʼгͼʵ״̬һгһվ̨ʱ(j=12mԤ·ȫ߹mվ)

Step2òDzͨʽ·ͨߵƫǶμΪ

Step3Ƕȷ㳵վ˿͵ѡΪ

Step4гεִ·ϵһվ̨ʱгԱ

Step5ʽ(4)гվʱʵؿʣ

Step6г1Ϊٶ󣬰г1վ̨Ĵв£

˱ƣֱ·վ (12 j)Ƶг

6Ըгδÿһվ̨ʱΪʱڵ㣬гͣվڼļֻҪȡʱڵǰij˿͵ɣʱڵĵ޹ءÿǰһõϿԤķʱԡ

4 ʵ

4.1·ѡ

ԴϺͨ16߳˿ͳʾʵصʾ (20145·)Чջʾ3523ݣЧʾ7021ݡ˿ڸ߷ƽʱεijֱͬ߷(7:009:00 17:0019:00)ƽڣǰ߷ʱΣϴո߷ЧΪ5804飬ƽЧΪ4740顣òΪȷƫǶ˿ѡIJ궨VPn֧š

16·ͼ2ʾͨ·Ϊ16ߵͨзʽβù··ļнΪ

 

ͼ2 Ϻͨ16߽Ƕ

Fig.2 the angle direction of shanghai metro 16

ѡȡ16߷վբAFCΪֱȡ0.5h1hΪʱ2ʾ

2  16߳վ߷վ

Table 2 the amount of passengers during morning and evening in stations of metro 16

վ

/0.5h

/h

·վ

1 221

2 314

·վ

631

934

ɽ·վ

742

1 274

ֶվ

492

1 014

ɳվ

121

392

ͷվ

200

278

³վ

921

1 628

Ұ԰վ

325

591

վ

798

1 002

϶վ

617

1 108

Ժվ

705

957

ٸ۴

831

1 203

ˮ

1 102

2 140

 

ͨ OD·Ƕ

˼룬Ȼtֵ16ʽʧ VPnվԭʵ

վǰ 1/2ʱվ˿վһȨͨӰ[12]ʧģ 2.2ʽ3ʧ 16ɿƫվƫǶ 3ʾ

3  ·ƫǶȼ

Table 3 the angle and transportation instead of metro line 16

վ

ƫ

Wӵ

Wƽӵ

ܲǶȷõ·

·

1112 '

0.282

0.281

975 ·976 ·989 ·ררߣרƽ«ר 11 · 26 ·

·

2314 '

0.181

0.080

ϴ

ɽ·

536 '

0.312

0.122

170 ·790 ·983 ·987 · 35 ·

ֶ

3116 '

0.212

0.087

796 ·1080 ·

 

 

ɳ

1216 '

0.211

0.142

1066 ·1101 ·ݷ

ͷ

2328'

0.111

0.111

1067 ·

³

343'

0.201

0.187

628 ·רߣʯרߣר 1 ·

Ұ԰

2144 '

0.312

0.201

רߣר

 

 

5201'

0.221

0.200

1038 ·1073 ·

϶

4912 '

0.231

0.231

1073 ·

Ժ

1437 '

0.182

0.182

רߣ«רߣ 7 ·

ٸ

1321'

0.244

0.201

1077 ·ר

 

ˮ

1627 '

0.215

0.121

1043 ·ˮԴר

 2.2ʽ3 4߷ʧ VPƽʧVPƽ

4  ǶȷµСʱ˿ʧ복վ˿Աȣ/h

Table 4 the loss of passengers and normal amount under the cost of angle

վ

߷

߷ʧVPn

ƽ

ƽʧVPn

·վ

2 314

0.282

1 772

0.112

·վ

934

0.181

634

0.021

ɽ·վ

1 274

0.312

785

0.121

ֶվ

1 014

0.212

501

0.102

ɳվ

892

0.211

734

0.181

ͷվ

778

0.111

687

0.067

³վ

1 628

0.201

896

0.074

Ұ԰վ

891

0.312

901

0.171

վ

1 002

0.221

971

0.091

϶վ

1 108

0.231

839

0.085

Ժվ

957

0.182

724

0.121

ٸ۴

1 203

0.244

825

0.078

ˮ

2 140

0.215

1 801

0.121

 

ͼ3Ƕȷ³վ˿ʧ

Fig3 the loss of passengers under the cost of angle

 

5  Ƕȷµij˿ʧ복վ˿Աȱ/h

Table 5 the loss amount and normal amount of passengers under the cost of angle

վ

߷

G ʵ

ƽ

P ʵ

·վ

2 314

1 893

1 772

1 634

·վ

934

858

634

596

ɽ·վ

1 274

1 004

785

704

ֶվ

1 014

900

501

475

ɳվ

892

882

734

689

ͷվ

778

769

687

621

³վ

1 628

1 302

896

794

Ұ԰վ

891

615

901

846

վ

1 002

881

971

894

϶վ

1 108

963

839

795

Ժվ

957

783

724

694

ٸ۴

1 203

1 030

825

794

ˮ

2 140

1 894

1 801

1 756

 

ͼ4 Ƕȷµij˿ʧ복վ˿Ա

Fig.4 the loss amount and normal amount of passengers under the cost of angle

4.2ʱԤ

ÿĵ1гΪо󣬰гվ̨Ĵϱ45ݣв,ֱ·վƵгԤģ͵øʱοԤ6ʾ

6  ʱվԤ

Table 6 the forecast of passengers flow based on distribution time

   վ

߷

ƽ壨죩

߷

ƽ壨ҹ

·

12 031

8 610

11 201

8 721

·

9 574

7 141

9 845

6 321

ɽ·

9 414

6 571

9 216

5 014

ֶ

9 018

5 362

8 142

4 782

ɳ

8 124

5 217

9 345

3 587

ͷվ

10 250

6 057

9 714

5 214

³

9 217

4 325

8 521

4 241

Ұ԰

9 014

7 254

9 321

8 725

8 314

5 214

9 417

4 712

϶

11 401

8 472

11 247

8 914

Ժ

10 144

9 870

9 651

9 147

ٸ۴

9 524

8 901

9 041

8 521

ˮ

8 251

7 302

9 471

8 146

5

нͨ˿ͺ򳵷ʱοֲԤ⣬Ϊ˹ܲŶԿ֯ӪƻʵʱݣԳ߹滮óмƻԸվԤзʽ·ȡн·ֲԤ漰ڶ࣬۸໥Ӱ죬ĶЩӵӰϵ˷ͽģԽվ̨ͬʱε˿ڸгϵķ估վ̨˻վ̨-гԤģ͵ʵʱ㷨ͨʵ֤н·ģʽ³վԤ㷨Чԡԡ󽫶½վ˿ODֲԤⷽһо

οף

[1]Ŵ,𣮻ڿ˲ĹվʱԤ[J].ͨϵͳϢ,2011,11(4):154-159[ZHANG C H, SONG R. Kalman Filter-Based Short-Term Passenger Flow Forecasting on Bus Stop [J]. Journal of Transportation Systems Engineering and Information Technology, 2011, 11(4):154-159]

[2]𻪣йͨ·ֲʵʱԤⷽ[J]ͬôѧѧȻѧ,2011,39(6):857-861[XU R H. Real-time Forecast of Passenger Flow Distribution on Urban Rail Transit Line [J]. JOURNAL OF TONGJI UNIVERSITY (NATURAL SCIENCE), 2011, 39(6):857-861]

[3],.ͨģ㷨[J]. ͬôѧѧ:Ȼѧ,2004,32(9):1158-1159[WU X Y. Traffic Equilibrium Assignment Model Specially

for Urban Railway Network[J]. JOURNAL OF TONGJI UNIVERSITY (NATURAL SCIENCE), 2004, 32(9):1158-1159]

[4],︣ȣйͨ˿·ѡģͼ㷨 [J].ͨϵͳϢ,2009,9(2):85-90[LIU J F, SUN F L. Passenger Flow Route Assignment Model and Algorithm for Urban Rail Transit Network [J]. Journal of Transportation Systems Engineering and Information Technology, 2009, 9(2):85-90]

[5],˼̣״·гͼʵƽ̨۵о[J]ͨѧѧ,2002,26(05)12-18[MA J J, HU S J. Study on Foundational Theory of Laboratorial Platform for Train Working Diagram Based on Railway Network [J]. JOURNAL OF NORTHERN JIAOTONG UNIVERSITY, 2002, 26(05):12-18]

[5]Ԩ,ȣгͼѧģͼ㷨о[J]ѧ,2001,23(1)1-8

[PENG Q Y, WANG P. Study on general optimization model and its solution for railway network train diagram [J]. Journal of the China Railway Society, 2001, 23(1):1-8]

[6],,.ڶ·ijйֲͨģͼ㷨о[J]ѧ,2009,31(2):110-112[XU R H, LUO Q, GAO P. Passenger Flow Distribution Model and Algorithm for Urban Rail Transit Network Based on Multi-route Choice [J]. Journal of the China Railway Society, 2009, 31(2):110-112]

[7]ɽ,˼̵ȣ״·гͼо[J]ѧ,1998,20(5)15-21[ZHOU L S, HU S J. Network Hierarchy Parallel Algorithm of Automatic Train Scheduling [J]. Journal of the China Railway Society, 1998, 20(5):15-21]

[8]H. Chen and X. Yao, Multiobjective neural network ensembles based on regularized negative correlation learning, IEEE Trans. Knowl. Data Eng. 2010, 22(12):1738C1751.

[9]H. Dam, H. Abbass, C. Lokan. Neural-based learning classifier systems [J], IEEE Trans. Knowl. Data Eng. 2008, 20, (1):26C39

[10]S. Munder and D. Gavrila.An experimental study on pedestrian classification [J].IEEE Trans. Pattern Anal. Mach. Intel. 2006. 28(11):1863C1868

[11]J. R. Koza, Genetic Programming: On the Programming of Computers by Means of Natural Selection. Cambridge [M], MA: MIT Press, 1992.

[12] Y. Jin and B. Sendhoff. Pareto-based multiobjective machine learning: An overview and case studies [J]. IEEE Trans. Syst., Man, Cybern. C, Applicat. Rev, 2008. 38(3):397C415

 

߼飺

С1982-Уն̨ˣͬôѧͨ乤ѧԺʿоҪоΪͨӪ֯ŻE-maildxbsuda@163.comͨѶַϺɽ·88Ū2301ңʱࣺ201620 ϵ绰18017349689̶绰021-67791165

 

ʦз1963-Уڣʿʦλͬôѧ ͨ乤ѧԺо򣺹ͨ֯·з֯Żϵ绰021-69589372 E-mailxfx@tongji.edu.cn

 



ոڣ2015-09-05

ĿȻ׼61272029

ĿϺпѧίԱмƻĿ11170501400

һߣС1982-Уն̨ˣʿоҪо򣺹ͨг֯ŻE-maildxbsuda@163.com

ͨѶߣз1963-УճˣڣʿʦҪо򣺸·г֯E-mailxfx@tongji.edu.cn

 

ͳ