改进遗传算法优化BP神经网络的旅游景区日客流量预测

Prediction for Tourist Attractions Daily Traffic Based on Modified Genetic Algorithm Optimized BP Neural Network

为提高BP神经网络预测模型的预测准确性,提出了一种基于改进遗传算法优化BP神经网络的预测方法.通过设计多层阶梯结构染色体,改进编码方式、适应度函数和遗传算子,引入自适应交叉、变异概率,实现对BP神经网络网络结构和初始网络权重的同步全局优化,提高BP神经网络预测模型的非线性学习和泛化能力.将该预测方法应用到黄山风景区日客流量实际预测中进行有效性验证,结果表明该方法对旅游景区日客流量具有更好的非线性拟合能力和预测准确性.

In order to improve forecasting model accuracy of BP neural network,an improved prediction method of optimized BP neural network based on modified Genetic Algorithm（ GA ） was proposed. We design new chromosomes with multi-storey Step-structure, improve the encoding mode, fitness function and genetic operator, and introduce the self-adaptive crossover and mutation probability, which optimizes the network structure and initial network weights of BP neural network synchronously. Moreover, the nonlinear learning and generalization abilities of BP neural network are enhanced. The availability of the modified prediction method was proved by predicting the daily passenger flow volume of Huangshan Scenic Spots. The computer simulations have shown that the nonlinear fitting and accuracy of the modified prediction methods are better than other prediction methods in forecast of daily passenger flow volume.