Neural network-based evaluation method for alignment state of air spring vibration isolation device

工况号	充气工况类型
Q1	根据压力均载原则充气至对中状态
Q2	循环充：每轮顺序为1→2→3→4→5→6→7→8，共8轮
Q3	先充首部：1→2→3→4充8次，后5→6→7→8充8次
Q4	先充右侧：1→3→5→7充8次，后8→6→4→2充8次
Q5	先充某一角：1→2→3→5充8次，后8→7→6→4充8次
Q6	先充前后端，后充中间：1→2→7→8充8次，3→4→5→6充8次

Table 2. Parameter settings for neural network model

View table

View in Article

Table 2. Parameter settings for neural network model

参数	参数设置
隐藏层节点数	第1层：128；第2层：64；第3层：32
学习率	0.002
迭代次数	3 000
激活函数	第1，2层：sigmoid()；第3层：relu( )
损失函数	torch.nn.MSELoss( )
优化器	torch.optim.Adam( )

Table 3. Prediction error of neural network model for small air spring vibration isolation device

View table

View in Article

Table 3. Prediction error of neural network model for small air spring vibration isolation device

数据集	误差值
数据集	x误差	z误差	α误差	γ误差	总体误差
测试集1（平滑处理）	0.075	0.361	0.018	0.147	0.199
测试集1（无平滑处理）	0.010	0.390	0.012	0.160	0.216
测试集2（平滑处理）	0.268	0.248	0.013	0.118	0.192

Table 4. Prediction accuracy of the two types of data in Test Set 3
View table
View in Article
Table 4. Prediction accuracy of the two types of data in Test Set 3
对中不对中
原始数据 1 373 1 304
预测结果 1 322 1 140
准确度/% 96.29 87.42

Table 5. Neural network prediction error for large air spring vibration isolation device
View table
View in Article
Table 5. Neural network prediction error for large air spring vibration isolation device
误差种类误差值
x 误差 0.011
z 误差 0.014
α 误差 0.003
γ 误差 0.001
总体误差 0.010

Tools

Get Citation

Copy Citation Text

Zhiwei LIU, Liang SHI, Song LIU. Neural network-based evaluation method for alignment state of air spring vibration isolation device[J]. Chinese Journal of Ship Research, 2024, 19(6): 117

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Theory and Method of Intelligent Design for Ship and Ocean Engineering

Received: May. 23, 2024

Accepted: --

Published Online: Mar. 14, 2025

The Author Email:

DOI:10.19693/j.issn.1673-3185.03933

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. The main typical inflation conditions

Table 1. The main typical inflation conditions

Table 2. Parameter settings for neural network model

Table 2. Parameter settings for neural network model

Table 3. Prediction error of neural network model for small air spring vibration isolation device

Table 3. Prediction error of neural network model for small air spring vibration isolation device

Table 4. Prediction accuracy of the two types of data in Test Set 3

Table 4. Prediction accuracy of the two types of data in Test Set 3

Table 5. Neural network prediction error for large air spring vibration isolation device

Table 5. Neural network prediction error for large air spring vibration isolation device