蚁群神经网络在船舶发电机故障诊断中的应用

Application of ant colony neural network in vessels fault diagnosis

发电机作为船舶关键部件之一,其运行状况关系船舶整机状态。随着技术的不断发展,船舶发电机结构更为复杂,同时其所发生的故障更为复杂和致命,因此研究高效可行的船舶发电机故障诊断技术已成为各研发机构的重要课题。本文通过分析算法模型建立了一种基于蚁群算法的人工神经网络,以发电机转子偏心问题为实例,在Matlab中应用该技术进行训练和故障诊断,并与BP算法进行比较。结果表明,蚁群算法训练速度快、精度高,是一种可靠的故障诊断方法。

As one of the key components in the vessels,the operation status of the generator is related to the state of the vessels. With the continuous development of technology,marine generator structure is more complex,and the fault is more complex and deadly. Therefore,the study of efficient and feasible vessel generator fault diagnosis technology has become an important subject in the research institutions. In this paper,by analyzing the algorithm model an artificial neural network based on ant colony algorithm was set up. Taking the generator rotor eccentricity as an example,training and fault diagnosis were carried out by MATLAB,and results were compared with that of BP algorithm. The results show that the ant colony algorithm is fast and accurate,and it is a reliable method for fault diagnosis.