基于免疫粒子群神经网络的短时交通流量预测

免疫理论中的基于浓度选择机制能避免粒子群算法在群体收敛性和个体多样性平衡问题上的不足，使改进后的粒子群算法优化BP神经网络参数的配置，提高短时交通流量预测的准确性。仿真实验表明：免疫粒子群优化后的BP神经网络可有效提高短时交通流量的预测精度，减小预测误差。

Short Term Traffic Flow Forecasting Based on Immune Particle Swarm Optimization in Neural Network

T he problem w hich the balance of population convergence and individual diversity in the particle swarm optimization algorithm can be solved by the concentration selection mechanism of immune theory . The improved particle swarm optimization algorithm can optimize BP neural network configuration parameters and improve the accuracy of short‐term traffic flow forecasting . The simulation experiment shows that the BP neural network model based on immune particle swarm optimization algorithm can effectively improve the prediction accuracy of short‐term traffic flow and reduce the prediction error .