Timely removal of traffic accidents has an important impact on normal road travel. This paper adopted the BP neural network model optimized by genetic algorithm to predict the duration of expressway traffic events. Data of 1256 traffic incidents occurred on some expressways in Shaanxi province in 2018 were collected, and factors that may affect the duration of traffic incidents were extracted. Three significant influencing factors were selected by principal component analysis. The optimization model was applied to predict the duration of traffic events. Results show while take the optimized connection weights and threshold of BP neural network model to predict the traffic incident duration, the forecast precision increase with the event long decrease. When the duration of a traffic event is within 120 minutes, the accuracy of model prediction is the highest, 77.3% of the prediction error of the duration of the traffic event can be controlled within 30% of the actual duration of the traffic event.
Prediction of Expressway Single Point Traffic Incident Time Based on Optimized GA-BP Neural Network
22.10.2021
849888 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Expressway traffic incident fusion sensing method based on Beidou space-time information
Europäisches Patentamt | 2023
|British Library Conference Proceedings | 2019
|