基于SVM的船舶上层建筑舱室噪声预报方法

摘 要：为对舰船舱室噪声进行精确预测，提出了基于SVM（支持向量机）的舱室噪声预测方法。采用RBF核函数和ERBF核函数，以某集装箱船上层建筑舱室噪声为训练样本，建立了两种集装箱船上层建筑舱室噪声的非线性回归模型；并应用两种模型对母型船及另一艘集装箱船上层建筑舱室噪声进行预测，并将预测结果进行了比较分析；在此基础上，应用效果较好的模型对一艘散装货船上层建筑舱室噪声进行预测。预测结果表明：应用SVM非线性回归模型对船舶上层建筑舱室噪声的预测是可行的，预测效果较为理想。

Prediction of noise in a ship' s superstructure cabins based on SVM method

In order to decrease cost, predicting noise before building is performed to improve a ship. Prediction of noise in a ship's superstructure cabins based on SVM method was presented here. Two nonlinear regression models of noise in superstructure cabins of a large container ship were established, based on a support vector machine(SVM), using RBF kernel function and ERBF kernel function, taking the actual measured noise of one container ship's superstructure cabins as the training sample. The two models were applied to predict noise in the parent ship and the another container ship's superstructure cabins. The results were compared and the better model was selected to predict the noise in a 174000DWT bulk carrier's superstructure cabins. The predicted results showed that the proposed method is feasible to predict noise in a ship's superstructure cabins and its effectiveness is good.