摘要
电力系统是船舶唯一的电力来源,一旦发生故障,电力供应就会立即中断,影响了船舶安全稳定运行。为此,针对目前提出的基于卷积神经网络、聚类分析、支持向量机的3种电力故障诊断系统鲁棒性不足的问题,在遗传算法的基础上,设计一种新的大型船舶电力系统短路故障诊断系统。该系统设计分为总体框架设计、硬件设计与软件设计3部分,为测试系统鲁棒性,进行压力测试。结果表明,与目前3种短路故障诊断系统相比,本系统面对不断增长的压力,系统失效次数更少,显现出来较强的鲁棒性,保证了船舶的安全稳定运行。
the power system is the only power source of the ship. Once there is a fault, the power supply will be interrupted immediately, which affects the safe and stable operation of the ship. For this reason, a new short-circuit fault diagnosis system for large-scale ship power system is designed on the basis of genetic algorithm, aiming at the problem of insufficient robustness of three kinds of power fault diagnosis systems based on convolution neural network, clustering analysis and support vector machine, which are put forward in current literature [1], literature [2] and literature [3]. The system design is divided into three parts, that is, the overall framework design, hardware design and software design. In order to test the robustness of the system, the pressure test is carried out. The results show that compared with the other three short-circuit fault diagnosis systems in literature [1], [2], [3], the system faces the increasing pressure, and the number of system failures is less,showing strong robustness, ensuring the safe and stable operation of the ship.
作者
王平
WANG Ping(Instrumentation Engineering Technology Research Center of Hebei Province,Chengde 067000,China)
出处
《舰船科学技术》
北大核心
2020年第2期88-90,共3页
Ship Science and Technology
基金
承德市科技支撑项目(201701A135)
关键词
电力系统
短路故障
诊断系统
power system
short circuit fault
diagnosis system
