神经网络北部湾海洋浪高预测

Beibu Gulf wave height prediction using neural network

为分析不同神经网络模型在海洋波高数据预测中的应用效果,基于北部湾单点浮标实测数据,首先对目标数据进行数字特征分析,根据分析结果选择离差标准化对输入数据进行预处理;接着通过实验分析对比了循环神经网络、长短时记忆网络、带卷积的循环神经网络和带卷积的长短时记忆网络4种网络模型方法在波高数据建模预测的效果。实验结果表明,4种网络建模方法对北部湾海域波高预测模型评估的平均均方根误差为0.20 m,平均相关系数为0.88,通过实验对比4种网络模型方法的预测分析结果表明,加入卷积后的网络模型预测结果准确度更好。

In order to analyze the application effect of different neural network models in the prediction of ocean wave height data,on the basis of the measured data of single point buoy in the Beibu Gulf,the digital feature of the target data are analyzed,and the deviation standardization is adopted according to the analysis results to preprocess the input data.And then,the prediction effect of the recurrent neural network(RNN),long short⁃term memory(LSTM)network,the RNN with convolution neural network(CNN)and the LSTM network with CNN in the wave height data modeling prediction was compared with experimental analysis.The experimental results show that the average root mean square error(RMSE)for the evaluation of prediction model of wave height in the Beibu Gulf by the four neural network modeling and prediction methods is 0.20 m and the average Pearson correlation coefficient(PCC)is 0.88.The predictive analysis results of the four neural network models were compared in the experiment.The results show that the accuracy of the prediction result of the network model adding convolution is higher.