Research On Short-term Traffic Flow Prediction Method Based On IWOA-WNN

In recent years,with the rapid development of social economy,people's material living standards have been continuously improved,and the demand for automobiles has also increased.As a result,road traffic has increased,and traffic problems such as road congestion have become increasingly serious.In order to alleviate such problems,with science With the development of technology,Intelligent Transportation Systerm.ITS has emerged.Among them,short-term traffic flow prediction plays an important role in intelligent transportation systems.The accuracy of short-term traffic flow prediction determines the traffic of ITS in traffic.Control and traffic guidance functions are good or bad.Therefore,research on short-term traffic flow prediction is still a hot spot in the field of intelligent transportation.The research on short-term traffic flow prediction in this paper mainly completed the following work:(1)Inductive analysis of the status quo of short-term traffic flow forecasting at home and abroad,and classification and comparison of common forecasting methods based on short-term traffic flow data characteristics.The study found that long-term and short-term memory is a characteristic of short-term traffic flow nonlinearity.(LSTM)neural network has become a hotspot of short-term traffic flow prediction research with its superior feature learning ability,wavelet neural network(WNN)with its good self-learning,self-adaptation and ability to approximate arbitrary nonlinear mapping.In this paper,LSTM and WNN are used to predict the short-term traffic flow.The simulation results show that the wavelet neural network has higher prediction accuracy and better fit to short-term traffic flow data.Therefore,wavelet neural network is selected as short-term traffic flow.The basic model of predictive research.(2)Based on the establishment of the IWOA-WNN prediction model.In the wavelet neural network,the gradient descent method is sensitive to the initial value of weight and wavelet factor,and it is easy to fall into the local minimum value,thus affecting the prediction performance of wavelet neural network.An improved wavelet neural network(IWOA-WNN)is proposed to improve the whale optimization algorithm.The model is used for short-term traffic flow prediction to improve the prediction accuracy of the model.(3)An improved whale optimization algorithm(IWOA)is proposed.Aiming at the shortcomings of low convergence precision and slow convergence speed of traditional whale optimization algorithm,the nonlinear convergence factor a and Tent chaotic map are introduced to improve the basic whale optimization algorithm.In order to verify the optimization performance of the improved algorithm,the basic whale optimization algorithm is used.The simulation results of five benchmark tests are carried out.The results show that the improved whale optimization algorithm proposed in this paper has faster optimization speed and higher precision.(4)Short-term traffic flow prediction based on IWOA-WNN model.The public transportation flow data set was obtained from the California Transportation Performance Measurement System(PeMS),and the data set was subjected to preprocessing such as missing repair,wavelet soft threshold denoising,phase space reconstruction and normalization,and short-term after pre-processing.The traffic flow data is used to verify the prediction performance of the proposed IWOA-WNN combined model.The IWOA-WNN prediction model is compared with the WNN prediction model and the GA-WNN,GWO-WNN and CS-WNN combined prediction models.For the calculation and analysis of the mean absolute error(MAE),mean absolute percentage error(MAPE)and root mean square error(RMSE)of the prediction results,it can be seen that the prediction model proposed in this paper has higher prediction accuracy.