基于改进神经网络的船舶主机温度预测

Ship main engine temperature prediction based on Improved Neural Network

由于传统神经网络存在收敛速度慢的缺陷,导致对船舶主机温度预测精度低,为了对船舶主机温度进行精确的预测,设计了基于改进神经网络的船舶主机温度预测模型。首先对当前船舶主机温度预测研究现状进行分析,找到引起预测效果差的因素,然后采集船舶主机温度变化的时间序列,并采用过程神经网络对船舶主机温度变化趋势进行估计,实现船舶主机温度预测,最后进行船舶主机温度预测验证性实验。结果表明,改进神经网络可以提高船舶主机温度预测精度,船舶主机温度预测误差远远小于传统神经网络,获得了比较满意的船舶主机温度预测结果。

Due to the slow convergence rate of the traditional neural network, the prediction accuracy of the temperature of the ship's host is low. In order to predict the temperature of the ship's host, the temperature prediction model based on the improved neural network is designed. First of all, the present situation of the current ship host temperature prediction research is analyzed, and the factors that cause the poor prediction result are found, then the time series of the temperature change of the ship host is collected, and the process neural network is used to estimate the temperature change trend of the ship host, and the temperature prediction of the ship host is realized. Finally, the ship main engine is carried out. The test is used to verify temperature prediction. The results show that the improved neural network can improve the prediction accuracy of the temperature of the ship's main engine. The temperature prediction error of the ship's main engine is far less than that of the traditional neural network, and a satisfactory temperature prediction result of the ship's main engine is obtained.