摘要
提出将量子粒子群优化(QPSO)算法融入尖点突变模型。由尖点突变理论势函数推导得到突变流形方程;针对道路交通应用,通过数学变换建立满足Maxwell约定的尖点突变模型;基于QPSO算法,对尖点突变模型参数求解过程进行优化。连续采样一组交通数据进行试验,表明基于QPSO算法的尖点突变模型对交通流速度的误差率为2.96%,明显低于基于散点拟合的尖点突变模型误差率11.56%。QPSO算法稳定好、收敛速度快,能够更好地实现交通流的高精度预测。
A strategy of integrating the quantum-behaved particle swarm optimization(QPSO)algorithm into a cusp catastrophe model is proposed.The catastrophe potential function is deduced from the potential function of cusp catastrophe theory.According to the road traffic application,a cusp catastrophe model satisfying the Maxwell's agreement is established through mathematical transformation.The parameters of the cusp catastrophe model are optimized based on the QPSO algorithm.A series of traffic data samples are continuously sampled to show that the error rate of traffic flow velocity based on the QPSO algorithm is 2.96%,which is significantly lower than the error rate of 11.46%for the cusp catastrophe model based on the scatter plot fitting.The QPSO algorithm has good stability and fast convergence speed,which can better achieve high precision prediction of traffic flow.
作者
胡志轩
张新晨
HU Zhi-xuan;ZHANG Xin-chen(College of Physical Science and Technology,Central China Normal University,Wuhan 430079,China)
出处
《工业技术创新》
2018年第2期90-93,共4页
Industrial Technology Innovation
关键词
尖点突变理论
量子粒子群优化(QPSO)算法
交通流
误差率
Cusp Catastrophe Theory
Quantum-behaved Particle Swarm Optimization(QPSO)Algorithm
Traffic Flow
Error Rate
