论文部分内容阅读
水电机组振动故障成因与故障征兆之间呈复杂的非线性关系,传统方法难以描述。当前研究常采用模式识别方法,如支持向量机、神经网络等实现振动故障诊断。该文在现有研究基础上,引进相关向量机(relevance vector machine,RVM)对诊断过程进行改进。相比传统方法,该文所提方法在学习过程中参数设置简单,在输出结果时给出了分类的可靠性,适合实际工程应用。同时,该方法在决策过程中,能够根据训练数据分布情况,自动选取决策结构,进一步提高诊断的速度与准确性。将该文所提诊断方法用于水电机组振动故障诊断实例,取得良好效果,验证了算法的有效性。
There is a complicated nonlinear relationship between the cause of fault and the symptom of fault in hydropower unit. The traditional method is difficult to describe. The current research often use pattern recognition methods, such as support vector machines, neural networks to achieve vibration fault diagnosis. Based on the existing research, this paper introduces the relevance vector machine (RVM) to improve the diagnosis process. Compared with the traditional method, the proposed method in the learning process parameter set is simple, the output reliability of the classification given, suitable for practical engineering applications. At the same time, this method can automatically select the decision structure according to the distribution of training data in the decision-making process, to further improve the speed and accuracy of diagnosis. The diagnostic method mentioned in this article is applied to the fault diagnosis of hydroelectric generating set. The good results are obtained and the effectiveness of the algorithm is verified.