Traffic flow forecasting based on wavelet neural network optimized by GA

In the complex urban traffic system, real-time and accurate traffic flow forecasting is the key to achieve the traffic control and guidance system. Based on the basic theory of wavelet neural network (WNN), the paper introduces the genetic algorithm (GA) with global search capability and establishes the traffic flow forecasting model of wavelet neural network optimized by genetic algorithm (GA-WNN), which will search the global optimal solutions of parameters in wavelet neural network and avoid the shortcoming of falling into local minimum of gradient descent learning algorithm. To compare the forecasting effect and network performance, the paper also establishes the traffic flow forecasting models of BP neural network optimized by genetic algorithm (GA-BP) and WNN. Simulation results have demonstrated that the proposed method has better generalization and higher forecasting precision, and it is applicable to traffic flow forecasting.