基于神经网络的船舶交通流量预测研究

发布时间：2018-12-06 08:40

【摘要】：随着我国航运业的快速发展,海上交通变得越来越发达的同时,海上交通事故也逐渐增多。科学而准确的船舶交通流量预测能为海事机关和港航部门制定港口和航道规划提供数据支持和理论依据,是减少海上交通事故的关键因素之一。本文在总结现有船舶交通流量预测模型的基础上,对神经网络船舶交通流量预测模型进行研究,并提出基于遗传算法优化BP神经网络的船舶交通流量预测模型。首先,简述海上交通流的理论基础知识和船舶交通流量预测的基本概念,并给出船舶交通流量预测的评价指标。其次,为减少数据波动对预测精度的影响,运用五点三次平滑处理方法对采集的数据进行平滑处理和归一化处理。然后,建立基于BP神经网络的船舶交通流量预测模型,针对BP神经网络的固有缺陷,应用遗传算法对BP神经网络进行优化,建立基于遗传算法优化BP神经网络的船舶交通流量预测模型。最后,分别采用BP神经网络模型和遗传算法优化BP神经网络模型对深圳港的船舶交通流量进行预测。结果表明,在一定误差范围内,BP神经网络预测模型和遗传算法优化BP神经网络预测模型能较好的预测深圳港的船舶交通流量。对比分析上述两预测模型的预测结果,分析结果表明遗传算法能够有效避免BP神经网络的固有缺陷,应用遗传算法优化BP神经网络的船舶交通流量预测模型的预测精度更高,误差更小。
[Abstract]:With the rapid development of China's shipping industry, maritime traffic becomes more and more developed, and maritime traffic accidents are gradually increasing. Scientific and accurate prediction of ship traffic flow can provide data support and theoretical basis for maritime authorities and port and shipping departments to formulate port and channel planning. It is one of the key factors to reduce maritime traffic accidents. On the basis of summarizing the existing ship traffic flow forecasting model, this paper studies the ship traffic flow forecasting model based on neural network, and puts forward a ship traffic flow forecasting model based on genetic algorithm to optimize BP neural network. Firstly, the basic theoretical knowledge of marine traffic flow and the basic concept of ship traffic flow forecasting are briefly introduced, and the evaluation indexes of ship traffic flow prediction are given. Secondly, in order to reduce the influence of data fluctuation on prediction accuracy, 5.3 times smoothing processing method is used to smooth and normalize the collected data. Then, the ship traffic flow forecasting model based on BP neural network is established. Aiming at the inherent defects of BP neural network, the genetic algorithm is applied to optimize the BP neural network. A ship traffic flow forecasting model based on genetic algorithm (GA) optimization BP neural network is established. Finally, the BP neural network model and the genetic algorithm optimization BP neural network model are used to predict the ship traffic flow in Shenzhen Port. The results show that BP neural network prediction model and genetic algorithm optimization BP neural network prediction model can better predict the ship traffic flow in Shenzhen Port within a certain range of errors. The results show that genetic algorithm can effectively avoid the inherent defects of BP neural network, and the prediction accuracy of ship traffic flow forecasting model based on genetic algorithm optimization of BP neural network is higher. The error is smaller.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U692

【参考文献】