改进遗传算法优化灰色神经网络隧道变形预测

针对目前隧道变形预测方法的不足,该文提出了使用改进型遗传算法优化灰色神经网络的隧道变形预测模型。改进遗传算法策略:在种群繁衍过程中根据个体的适应度进行排序,再将排序后的种群均分为3个部分,按照比例对3个部分进行选择,最后从适应度较大的部分中随机选取个体在重新补充到种群中。改进型遗传算法可以避免陷入局部收敛成功找寻全局最优解,提高收敛速度。该文利用实际隧道监测数据进行实验,验证改进型遗传算法优化灰色神经网络的隧道变形预测模型。实验证明,改进型遗传算法优化灰色神经的隧道变形预测模型在进行隧道拱顶下沉量预测时有着更高的精度、更好的稳定性。

Tunnel deformation prediction based on grey neural network with improved genetic algorithm

In view of the shortcomings of current tunnel deformation prediction methods,an improved genetic algorithm is proposed to optimize the grey neural network model for tunnel deformation prediction.Improvement of genetic algorithm strategy:in the process of population reproduction,according to the fitness of individuals to sort,and then the sorted population are divided into three parts,according to the proportion of the three parts to select,and finally randomly select individuals from the larger part of the fitness to replenish to the population.The improved genetic algorithm can avoid falling into local convergence and find the global optimal solution successfully,and improve the convergence speed.The improved genetic algorithm is used to optimize the grey neural network,and the actual tunnel monitoring data are used for experiments.Experiments show that the improved genetic algorithm optimizes the grey neural network model for tunnel deformation prediction has higher accuracy and better stability in predicting tunnel vault settlement.