Short-term Traffic Flow Prediction based on GMO-DELM

Short-term traffic flow prediction is the premise of making full use of intelligent transportation system and plays an important role in urban traffic planning. In this paper, a deep extreme learning machine model based on grey wolf optimizer (GWO-DELM) is proposed. The grey wolf optimizer is used to optimize the input weights of hidden layers of deep extreme learning machine to improve the prediction accuracy of the model. In order to verify the effectiveness of the model, seven data sets of M25 High-speed Heathrow Airport were used to predict the traffic flow. Compared with the existing long short-term memory network model (LSTM), BP neural network model and unoptimized deep extreme learning machine model (DELM), the results show that: The traffic flow predicted by this optimization model fits the actual traffic flow best, and the prediction error is effectively reduced.