以提升船舶航行安全性为目的，研究基于传感器采集信息的船舶发动机状态智能检测方法。该方法利用转速传感器、压力传感器、温度传感器等组成传感器阵列，采集船舶发动机状态信息，基于核函数的主元分析方法提取发动机状态信息特征，将发动机状态信息特征输入到最小二乘支持向量机内，通过建立最小二乘支持向量机优化目标函数、设置约束条件和建立检测输出函数，完成船舶发动机状态智能检测过程。实验结果表明：该方法采集的船舶发动机震动信息最大数值与最小数值与其实际数值完全重合，采集船舶发动机状态信息能力较强；提取发动机震动信号的子带能量特征和谱能量特征分布较为一致，提取发动机声信号特征较为精准；检测发动机不同类型故障错误概率数值较低，其智能检测效果显著。
In order to improve the navigation safety of ships, the intelligent detection method of ship engine state based on the information collected by sensors is studied. This method uses the speed sensor, pressure sensor and temperature sensor to form a sensor array. After collecting the state information of ship engine, the main component analysis method based on kernel function is used to extract the state information characteristics of ship engine; The engine state information features are input into the least squares support vector machine, and the intelligent detection process of the ship engine state is completed by establishing the optimization objective function of the least squares support vector machine, setting the constraint conditions and establishing the detection output function. The experimental results show that the maximum value and minimum value of the vibration information collected by this method coincide with the actual value, and the ability of collecting the state information of the ship engine is strong; The sub-band energy feature and spectral energy feature distribution of the engine vibration signal are more consistent, and the engine acoustic signal feature is more accurate; The error probability of detecting different types of faults of marine engines is low, and its intelligent detection effect is remarkable.
2022,44(17): 118-121 收稿日期：2022-05-04
DOI：10.3404/j.issn.1672-7649.2022.17.023
分类号：TN913
作者简介：沈大伟(1979-)，男，博士，讲师，研究方向为测试计量技术及仪器、动态测试与智能仪器、电气工程与智能控制
参考文献：
[1] 邹强, 田颖, 李红松, 等. 基于支持向量机的燃料电池发动机氢气泄漏检测方法[J]. 北京交通大学学报, 2020, 44(1): 84–90
[2] 胡廷智, 张忠清, 肖木峥, 等. 基于激光自准直的发动机轴孔同轴度在线检测方法[J]. 推进技术, 2019, 40(9): 2099–2104
[3] 黄功, 赵永平, 谢云龙. 基于局部密度的加权一类支持向量机算法及其在涡轴发动机故障检测中的应用[J]. 计算机应用, 2020, 40(3): 917–924
[4] 强子健, 鲁峰, 常晓东, 等. 基于二阶鲁棒滑模观测器的民用涡扇发动机气路故障诊断[J]. 推进技术, 2020, 41(6): 1411–1419
[5] 柳长源, 车路平, 毕晓君. 基于TWSVM算法的发动机故障识别方法[J]. 内燃机学报, 2019, 37(1): 84–89
[6] 陈鲲, 茆志伟, 张进杰, 等. 基于和声搜索优化栈式自编码器的柴油发动机故障诊断[J]. 机械工程学报, 2020, 56(11): 132–140
