A Two-Stage Hybrid Extreme Learning Model for Short-Term Traffic Flow Forecasting
Credible and accurate traffic flow forecasting is critical for deploying intelligent traffic management systems. Nevertheless, it remains challenging to develop a robust and efficient forecasting model due to the nonlinear characteristics and inherent stochastic traffic flow. Aiming at the nonlinear...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-06-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/12/2087 |
| _version_ | 1797484669866868736 |
|---|---|
| author | Zhihan Cui Boyu Huang Haowen Dou Yan Cheng Jitian Guan Teng Zhou |
| author_facet | Zhihan Cui Boyu Huang Haowen Dou Yan Cheng Jitian Guan Teng Zhou |
| author_sort | Zhihan Cui |
| collection | DOAJ |
| description | Credible and accurate traffic flow forecasting is critical for deploying intelligent traffic management systems. Nevertheless, it remains challenging to develop a robust and efficient forecasting model due to the nonlinear characteristics and inherent stochastic traffic flow. Aiming at the nonlinear relationship in the traffic flow for different scenarios, we proposed a two-stage hybrid extreme learning model for short-term traffic flow forecasting. In the first stage, the particle swarm optimization algorithm is employed for determining the initial population distribution of the gravitational search algorithm to improve the efficiency of the global optimal value search. In the second stage, the results of the previous stage, rather than the network structure parameters randomly generated by the extreme learning machine, are used to train the hybrid forecasting model in a data-driven fashion. We evaluated the trained model on four real-world benchmark datasets from highways A1, A2, A4, and A8 connecting the Amsterdam ring road. The RMSEs of the proposed model are 288.03, 204.09, 220.52, and 163.92, respectively, and the MAPEs of the proposed model are <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>11.53</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>10.16</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>11.67</mn><mo>%</mo></mrow></semantics></math></inline-formula>, and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>12.02</mn><mo>%</mo></mrow></semantics></math></inline-formula>, respectively. Experimental results demonstrate the superior performance of our proposed model. |
| first_indexed | 2024-03-09T23:07:43Z |
| format | Article |
| id | doaj.art-503585ac764242fd8420e65d2a14d602 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T23:07:43Z |
| publishDate | 2022-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-503585ac764242fd8420e65d2a14d6022023-11-23T17:49:31ZengMDPI AGMathematics2227-73902022-06-011012208710.3390/math10122087A Two-Stage Hybrid Extreme Learning Model for Short-Term Traffic Flow ForecastingZhihan Cui0Boyu Huang1Haowen Dou2Yan Cheng3Jitian Guan4Teng Zhou5Department of Computer Science, Shantou University, Shantou 515063, ChinaDepartment of Computer Science, Shantou University, Shantou 515063, ChinaDepartment of Computer Science, Shantou University, Shantou 515063, ChinaMedical College, Shantou University, Shantou 515063, ChinaMedical College, Shantou University, Shantou 515063, ChinaDepartment of Computer Science, Shantou University, Shantou 515063, ChinaCredible and accurate traffic flow forecasting is critical for deploying intelligent traffic management systems. Nevertheless, it remains challenging to develop a robust and efficient forecasting model due to the nonlinear characteristics and inherent stochastic traffic flow. Aiming at the nonlinear relationship in the traffic flow for different scenarios, we proposed a two-stage hybrid extreme learning model for short-term traffic flow forecasting. In the first stage, the particle swarm optimization algorithm is employed for determining the initial population distribution of the gravitational search algorithm to improve the efficiency of the global optimal value search. In the second stage, the results of the previous stage, rather than the network structure parameters randomly generated by the extreme learning machine, are used to train the hybrid forecasting model in a data-driven fashion. We evaluated the trained model on four real-world benchmark datasets from highways A1, A2, A4, and A8 connecting the Amsterdam ring road. The RMSEs of the proposed model are 288.03, 204.09, 220.52, and 163.92, respectively, and the MAPEs of the proposed model are <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>11.53</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>10.16</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>11.67</mn><mo>%</mo></mrow></semantics></math></inline-formula>, and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>12.02</mn><mo>%</mo></mrow></semantics></math></inline-formula>, respectively. Experimental results demonstrate the superior performance of our proposed model.https://www.mdpi.com/2227-7390/10/12/2087intelligent transportation systemtraffic flow modelingtime series analysismachine learningnoise-immune learning |
| spellingShingle | Zhihan Cui Boyu Huang Haowen Dou Yan Cheng Jitian Guan Teng Zhou A Two-Stage Hybrid Extreme Learning Model for Short-Term Traffic Flow Forecasting Mathematics intelligent transportation system traffic flow modeling time series analysis machine learning noise-immune learning |
| title | A Two-Stage Hybrid Extreme Learning Model for Short-Term Traffic Flow Forecasting |
| title_full | A Two-Stage Hybrid Extreme Learning Model for Short-Term Traffic Flow Forecasting |
| title_fullStr | A Two-Stage Hybrid Extreme Learning Model for Short-Term Traffic Flow Forecasting |
| title_full_unstemmed | A Two-Stage Hybrid Extreme Learning Model for Short-Term Traffic Flow Forecasting |
| title_short | A Two-Stage Hybrid Extreme Learning Model for Short-Term Traffic Flow Forecasting |
| title_sort | two stage hybrid extreme learning model for short term traffic flow forecasting |
| topic | intelligent transportation system traffic flow modeling time series analysis machine learning noise-immune learning |
| url | https://www.mdpi.com/2227-7390/10/12/2087 |
| work_keys_str_mv | AT zhihancui atwostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT boyuhuang atwostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT haowendou atwostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT yancheng atwostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT jitianguan atwostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT tengzhou atwostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT zhihancui twostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT boyuhuang twostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT haowendou twostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT yancheng twostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT jitianguan twostagehybridextremelearningmodelforshorttermtrafficflowforecasting AT tengzhou twostagehybridextremelearningmodelforshorttermtrafficflowforecasting |