恶劣海况下船舶短期电力负荷预测方法  

Short-term Power Load Forecasting Method for Ships under Severe Sea Conditions

作　　者：谷义恩 杨智荣[2] 高海波[1] 唐元章 GU Yien;YANG Zhirong;GAO Haibo;TANG Yuanzhang(Wuhan University of Technology,Wuhan 430000,Hubei China;China Special Equipment Inspection and Research Institute,Beijing 101300,China)

机构地区：[1]武汉理工大学,武汉430000 [2]中国特种设备检测研究院,北京101300

出　　处：《海洋工程装备与技术》2023年第1期117-123,共7页Ocean Engineering Equipment and Technology

摘　　要：电力推进船舶在恶劣海况环境下运行时,整个电力系统的负荷会产生较大的波动,给船舶电网的能量管理带来很大的挑战。准确及时地预测电力负荷的变化趋势能有效保障电力系统的安全。本文建立BP神经网络、Elman神经网络、LSTM神经网络和改进的粒子群算法(IPSO)优化BP神经网络模型,对船舶短期电力负荷进行预测。结果表明,IPSO+BP神经网络模型的电力负荷预测精度最高,并且收敛速度快,性能最佳。When the electric propulsion ship operates under harsh sea conditions,the load of the entire power system will fluctuate greatly,which brings significant challenges to the energy management of the ship's power grid.Accurate and timely prediction of the trend of power load can effectively improve the safety of the power system.In this paper,BP neural network,Elman neural network,LSTM neural network and improved Particle Swarm Optimization(IPSO)that are used to optimize BP neural network model are compared in predicting ship short-term power load.The results show that the power load prediction accuracy of IPSO+BP neural network model is the highest and the effect is the best.

关 键 词：船舶电网 电力负荷预测 神经网络 IPSO+BP LSTM ELMAN 

分 类 号：U665.12[交通运输工程—船舶及航道工程] TP183[交通运输工程—船舶与海洋工程] TM715[自动化与计算机技术—控制理论与控制工程]