采用传感器节点的船舶动力装置状态监测研究

Research on condition monitoring of marine power plant using sensor nodes

为了解决传统船舶动力装置状态监测方法监测效率低、实时性差的缺陷问题,采用传感器节点对船舶动力装置状态监测方法进行研究。采用传感器节点对动力装置状态数据进行采集,通过过滤算法对其中无效数据进行滤除,以处理后的动力装置状态数据为依据,利用自回归AR算法构建动力装置状态预测模型,以预测模型为工具,通过梯度下降迭代算法对动力装置状态进行判定,实现了船舶动力装置状态的监测。通过测试得到,与传统的船舶动力装置状态监测方法相比较,提出的船舶动力装置状态监测方法极大地提升了监测效率与实时性,充分说明提出的船舶动力装置状态监测方法具备更好的监测效果。

In order to solve the problem of low efficiency and poor real-time performance of the traditional state monitoring method for ship power plant,the state monitoring method for ship power plant is studied by using sensor nodes.The sensor nodes are used to collect the state data of the power plant,and the invalid data are filtered by filtering algorithm.Based on the state data of the power plant after processing,the state prediction model of the power plant is constructed by using autoregressive AR algorithm.With the prediction model as a tool,the power plant is loaded by gradient descent iteration algorithm.The state of the ship power plant is determined and the state of the ship power plant is monitored.Through testing,compared with the traditional state monitoring method of ship power plant,the proposed state monitoring method of ship power plant greatly improves the monitoring efficiency and real-time,fully demonstrating that the proposed state monitoring method of ship power plant has better monitoring effect.