舰船上层建筑振动特性的数学模型设计

Mathematical model design of vibration characteristics of ship superstructure

针对当前数学模型无法描述舰船上层建筑振动特性的变化规律,为了提高舰船上层建筑振动特性预测精度,设计一种蚁群优化算法和神经网络相结合的舰船上层建筑振动特性预测数学模型。首先对当前各种舰船上层建筑振动特性预测数学模型的优缺点进行阐述,然后采用神经网络对舰船上层建筑振动特性变化规律进行拟合,并采用蚁群优化算法确定神经网络相关参数,最后进行舰船上层建筑振动特性预测数学模型的性能测试。结果表明,蚁群优化算法和神经网络相结合的舰船上层建筑振动特性预测精度高,不仅预测误差远低于当前其他舰船上层建筑振动特性预测数学模型,而且预测效率也得到了改善,为解决舰船上层建筑振动特性预测问题提供了一种新的研究方法。

In order to improve the prediction accuracy of the vibration characteristics of superstructure,a mathematical model for predicting the vibration characteristics of superstructure is designed,which combines ant colony optimization algorithm and neural network.Firstly,the advantages and disadvantages of various mathematical models for predicting the vibration characteristics of superstructure of naval ships are expounded.Then,the variation law of vibration characteristics of superstructure of naval ships is fitted by neural network,and relevant parameters of neural network are determined by ant colony optimization algorithm.Finally,the prediction mathematics of the vibration characteristics of superstructure of naval ships is carried out.The test results show that the prediction accuracy of vibration characteristics of warship superstructure based on ant colony optimization algorithm and neural network is high,prediction error of vibration characteristics of superstructure of warship is much lower than that of other current mathematical models for the prediction of the vibration characteristics of the superstructure of warship.Moreover,vibration characteristics prediction accuracy of the warship superstructure is much lower than that of other current mathematical models.The measurement efficiency has also been improved,which provides a new research method for the prediction of vibration characteristics of ship superstructure.