基于改进灰狼算法的铁路隧道射线跟踪模型校正

针对灰狼优化(grey wolf optimization, GWO)算法易陷入局部最优和收敛精度差的问题,提出了一种基于对立搜索和Levy飞行策略的改进灰狼优化算法——OLGWO算法.在算法初始化阶段,采用对立搜索策略以缩小可行解范围;在灰狼位置更新过程中,为避免算法陷入局部最优采用了Levy飞行策略. 4个标准测试函数的仿真实验表明,所提OLGWO算法在收敛速度及求解精度方面均优于GWO算法,可以较快且准确地搜索到目标函数的最优值.基于OLGWO算法对隧道射线跟踪传播模型进行校正的结果表明,校正后的模型在均方根误差和线性相关性方面具有较优的性能,能够实现铁路隧道环境中信号接收功率的精确预测.

Ray tracing model tuning based on grey wolf optimization in railway tunnels

To solve the problem that the gray wolf optimization (GWO) algorithm is easy to fall into the local optimum and the convergence precision is poor, this paper proposes an improved gray wolf optimization algorithm based on opposition search and Levy flight strategy(OLGWO). In the initial stage of the algorithm, the opposite search strategy is adopted to narrow the range of feasible solutions. And in the process of updating location of the gray wolves, Levy flight strategy is adopted to avoid the algorithm falling into the local optimum. By four standard test functions, simulation experiments show that the proposed OLGWO algorithm is superior to the GWO algorithm in terms of convergence speed and solution accuracy, and it can search for the optimal solution quickly and accurately. Next, based on the OLGWO optimization algorithm, the ray tracing propagation model of the tunnel is corrected. The results show that the corrected model has better performances in terms of RMS and linear correlation and can achieve precise prediction of signal receiving power in the railway tunnel environment.