摘要
为了提高港口码头潮汐预报的精度,提出一种自适应变异的粒子群优化算法SAPSO,将SAPSO优化算法与BP神经网络结合,用以潮汐水位的实时预报。SAPSO-BP网络模型运用自适应变异的PSO算法优化BP神经网络的网络参数,克服了传统BP神经网络所具有的对初始权值阈值敏感、容易陷入局部极小值的缺点,最后选用Isabel港口的实测潮汐值数据进行潮汐水位的实时预报仿真试验,用以验证SAPSO-BP预测模型的实用性和可靠性。
In order to improve the accuracy of tidal level prediction in port and wharf, we propose a selfadapting particle swarm optimization(SAPSO) algorithm to optimize the back propagation(BP) neural network model. The model is referred to as SAPSO-BP model which employs PSO to adjust control parameters of BP network.This novel model overcomes the shortcoming of traditional BP neural network, which is sensitive to the initial weight threshold and is easy to trap in local minimum. The real-measured tidal level data of Isabel port is chosen as the test database to verify the practicability and reliability of the SAPSO-BP prediction model.
作者
张泽国
尹建川
柳成
张心光
ZHANG Ze-guo YIN Jian-chuan LIU Cheng ZHANG Xin-guang(Navigation College, Dalian Maritime University, Dalian 116026, China Automobile Engineering College, Shanghai University of Engineering Science, Shanghai 201620, China)
出处
《水运工程》
北大核心
2017年第1期34-40,共7页
Port & Waterway Engineering
基金
国家自然科学基金项目(51379002
51279106)
关键词
BP神经网络
自适应
粒子群优化
港口潮汐水位实时预测
调和分析
BP neural network
self-adapting
particle swarm optimization
real time tidal level forecast in port
harmonious analysis