Traffic Flow Forecasting Model of Typical Multi-Intersection for Urban Trunk Road Based on Dissimilation Particle Swarm Optimization

Based on the analysis of the standard Particle Swarm Optimization and the characteristic of typical multi-intersection for urban trunk road, a traffic flow forecasting model using dynamic recursion neural network is presented. The feature of this network is that the output of the hidden layer connects to the input of itself through the delay and storage of the context layer. The method of self-connection makes it sensitive to the historical data and improves the ability to process dynamic information. The initial inputs of the context unit and all the weights of the model are optimized by Dissimilation Particle Swarm Optimization. The algorithm mutates the local optimum based on the judging of local or global optimal value, and helps it jump out local convergence, so as to solve the defects which the found value are local optimal value and fall into the problem of prematurity of traditional Particle Swarm Optimization. An example demonstrates that this algorithm is simple, efficient, and fast.