基于蚁群算法优化反向传播神经网络的港口吞吐量预测

Throughput Prediction of Port Based on Back Propagation Neural Network Optimized by Ant Colony Algorithm

利用蚁群算法优化反向传播神经网络的初始权值、阈值,建立预测模型,对港口货物吞吐量进行预测。蚁群算法具有全局搜索能力,分布式计算和鲁棒性强等特点,有利于加快反向传播神经网络的收敛速度,避免易陷入局部极值的问题,提高建模精度。在港口吞吐量预测中的应用表明:蚁群算法优化BP神经网络模型、模糊神经网络预测模型、RBF预测模型及BP预测模型的平均绝对百分比误差分别为2.826%、3.734%、4.990%和6.566%;同时,蚁群算法优化BP神经网络模型收敛速度最快。

Port cargo throughput is an important index of port production and operation scale,and it is the basis for port construction and development.In order to maximize the role of port,it is necessary to make a reasonable and effective forecast for port cargo throughput.Ant colony algorithm is used to optimize the initial weight and threshold of BP neural network,and the prediction model is established to predict the port cargo throughput.Ant colony algorithm has the characteristics of global search,distributed computation and strong robustness,which is beneficial to accelerate the convergence speed of BP neural network,avoids the problem of easy to fall into local extremum,and improves the modeling accuracy.The application in port throughput prediction shows that the average absolute percentage errors of BP neural network model optimized by ant colony algorithm,fuzzy neural network prediction model,RBF prediction model and BP prediction model are 2.826%,3.734%,4.990%and 6.566%respectively;meanwhile,the convergence speed of BP neural network model optimized by ant colony algorithm is the fastest.