Accurate short-term prediction of available parking space (APS) is the basic theory of parking guidance information system (PGIS). This study collected the data on parking availability at several on-street parking spaces on 12th Ave., Seattle, America to investigate the changing characteristics of APS, and predicted the APS based on an improved wavelet neural network (WNN). It presents an improved WNN algorithm with wavelet (WA) decomposition and particle swarm optimization (PSO). The original time series was decomposed and reconstructed by wavelet analysis, and the WNN algorithm finds the optimal threshold of initial weight through PSO. Compared with the methods of BPNN, WNN, PSO-WNN, the WA-PSO-WNN algorithm performs much better on predicting accuracy and stability.
Short–Term Prediction of Available Parking Space Based on Improved Wavelet Neural Network
20th COTA International Conference of Transportation Professionals ; 2020 ; Xi’an, China (Conference Cancelled)
CICTP 2020 ; 3383-3395
09.12.2020
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Short‐term forecasting of available parking space using wavelet neural network model
Wiley | 2015
|Short-term forecasting of available parking space using wavelet neural network model
IET | 2015
|