基于BP神经网络模型的福建海域赤潮预报方法研究
发布时间：2018-10-17 16:45
【摘要】：赤潮往往给渔业生产和人类的生命安全造成极大的危害,但由于赤潮的成因十分复杂,对其进行预报非常困难。本研究收集了福建海区2000年至2016年发生的219个赤潮案例有效数据,应用BP神经网络人工智能模型建立了其与气温、降水、风速、气压和日照5个气象因子的非线性关系,并将这些赤潮案例数据与相应的气象指标按闽东、闽中和闽南3个海区,分别输入模型进行学习、训练与预测。结果显示:1)闽东海区53个训练样本45个预测正确,正确率达84.91%,3个模拟预测样本全部正确;2)闽中海区69个训练样本58个预测正确,正确率达84.06%,4个模拟预测样本全部正确;3)闽南海区85个训练样本的运算预测结果63个正确,正确率74.12%,5个模拟预测样本全部正确,达到预期的结果。研究表明,以气象因子为自变量采用BP神经网络模型对赤潮的发生进行预测是可行的,该方法可为赤潮的预测提供新的途径。
[Abstract]:The red tide is often harmful to the production of fishery and the safety of human life, but it is very difficult to forecast the red tide because of its complexity. The data of 219 red tide cases occurred from 2000 to 2016 in Fujian sea area were collected, and the nonlinear relation with temperature, precipitation, wind speed, air pressure and sunshine was established by BP neural network artificial intelligence model. The data of these red tide cases and the corresponding meteorological indexes were studied, trained and predicted according to the input models respectively according to the three sea areas in the east, middle and south Fujian. The results showed that: 1) 53 training samples in Mindong area were correctly predicted, the correct rate was 81.91%, the three simulated predictive samples were all correct; 2) 69 training samples were correctly predicted, the correct rate was 84.6%, and the four simulated predictive samples were all correct. 3) The results of operation and prediction of 85 training samples in Minnan sea area are correct, the correct rate is 74. 12%, and the five analog predictive samples are all correct to achieve the expected results. The research shows that using the BP neural network model to forecast the occurrence of red tide is feasible by using meteorological factors as independent variables, and the method can provide a new way for the prediction of red tide.
【作者单位】： 福建省水产研究所;厦门市气象台;福建省海洋环境与渔业资源监测中心;厦门市海洋与渔业研究所;福建省海洋预报台;
【基金】：福建省海洋与渔业结构调整专项(2015) 福建省海洋与渔业厅科技外经外事处:基于BP神经网络的海区赤潮预警预报模型研究(闽海渔科2015005)~~
【分类号】：TP183;X55
[Abstract]:The red tide is often harmful to the production of fishery and the safety of human life, but it is very difficult to forecast the red tide because of its complexity. The data of 219 red tide cases occurred from 2000 to 2016 in Fujian sea area were collected, and the nonlinear relation with temperature, precipitation, wind speed, air pressure and sunshine was established by BP neural network artificial intelligence model. The data of these red tide cases and the corresponding meteorological indexes were studied, trained and predicted according to the input models respectively according to the three sea areas in the east, middle and south Fujian. The results showed that: 1) 53 training samples in Mindong area were correctly predicted, the correct rate was 81.91%, the three simulated predictive samples were all correct; 2) 69 training samples were correctly predicted, the correct rate was 84.6%, and the four simulated predictive samples were all correct. 3) The results of operation and prediction of 85 training samples in Minnan sea area are correct, the correct rate is 74. 12%, and the five analog predictive samples are all correct to achieve the expected results. The research shows that using the BP neural network model to forecast the occurrence of red tide is feasible by using meteorological factors as independent variables, and the method can provide a new way for the prediction of red tide.
【作者单位】： 福建省水产研究所;厦门市气象台;福建省海洋环境与渔业资源监测中心;厦门市海洋与渔业研究所;福建省海洋预报台;
【基金】：福建省海洋与渔业结构调整专项(2015) 福建省海洋与渔业厅科技外经外事处:基于BP神经网络的海区赤潮预警预报模型研究(闽海渔科2015005)~~
【分类号】：TP183;X55
【相似文献】
相关期刊论文 前6条
1 刘伟;;BP神经网络在海洋赤潮预测中的应用[J];海洋科学集刊;2010年00期
2 陈维;顾杰;李雯婷;秦欣;;基于BP神经网络的长江口北支河槽容积分析[J];海洋科学;2011年01期
3 朱智慧;;BP神经网络方法在海浪数值预报中的释用研究[J];大气科学研究与应用;2011年01期
4 李志龙;陈子q，
本文编号：2277307
本文链接：https://www.wllwen.com/kejilunwen/haiyang/2277307.html
