基于支持向量机的船舶电力负荷预测

SHIP POWER LOAD FORECASTING USING SUPPORT VECTOR MACHINE

船舶电力系统是一个独立的电力系统,需要根据准确的负荷预测来控制多台发电机组的运行。本文提出了一种基于支持向量机的船舶电力负荷短期预测方法。对某大型集装箱船舶在不同工况下的电力负荷数据,分别用基于径向基核函数的支持向量机方法、多层 BP 网络和 RBF 网络方法进行训练和预测计算,仿真结果表明支持向量机具有更高的预测精度,是船舶电力负荷预测的一种有效方法。

Ship power system is an isolated power system, and the several generators are controlled to run or stop according to accurate load forecasting respectively. A new short-term load forecasting method for ship power system based on support vector machine (SVM) is presented. Three methods of the load forecasting, the SVM based on radial basis function kernel, the multi-layer back-propagation neural network and the radial basis function neural network, are compared for actual load data sampled at different operation modes from a large-scale container ship. The simulation results show that the SVM method can achieve greater accuracy than other methods, and is effective for ship power load forecasting.