In this paper an improved wavelet neural network (WNN) model is proposed to actualize dynamic forecast. An input data series, i.e. historical data, traffic condition and weather information about passenger flow surveyed from No.96 transit line in Dalian, China, is pre-processed via a fuzzy operator before being transferred to train and test the constructed network. A hybrid genetic algorithm and identical-dimension recurrence idea are constructed to optimize the structure and shape of WNN dynamically so as to enhance its forecast accuracy. The simulation result indicates the proposed WNN model can satisfy the precision request, accelerate the convergence speed, improve global generalization ability and possess practicality in transit dynamic scheduling. (a) For the covering entry of this conference, please see ITRD abstract no. E214938.
Samenvatting