基于贝叶斯网络的减速器异常振动故障诊断

Abnormal Vibration Fault Diagnosis of Reducer Based on Bayesian Network

为实现减速器异常振动的故障类型快速判断、降低巡检与维护成本,研发了一种减速器异常振动智能诊断模型。在历史异常振动数据不充足、不平衡的情况下,通过对历史故障履历资料的梳理,建立了减速器故障树,并映射为减速器异常振动贝叶斯网络结构。同时,对历史异常振动数据进行振动特征提取与标签化,选择期望最大化(EM)算法为参数学习方法,确定贝叶斯网络节点变量的概率分布。在减速器运行过程中,该模型处理实时振动数据后,融合故障知识转化的异常振动特征判别机制与分层吉布斯采样算法,对各节点变量进行故障概率推理,实现异常振动故障的及时定位。通过模型性能测试发现所提出的故障诊断模型相比于其他典型模型,在诊断结果准确性与区分正常与异常振动数据的精度方面均取得了较大的提升,并将模型集成至带式输送机智能运维系统中进行了工程验证。

In order to recognize the fault type of abnormal vibration of the reducer and reduce the cost of inspection and maintenance,an intelligent diagnosis model of abnormal vibration of the reducer is developed.In the case of insufficient and unbalanced historical abnormal vibration data,a fault tree of the reducer is established by combing the historical fault history data.It is then mapped to the Bayesian network structure of the abnormal vibration of the reducer.The historical abnormal vibration data are extracted and labeled.The expectation maximization(EM)algorithm is selected as the parameter learning method to determine the probability distribution of the node variables in the Bayesian network.After processing real-time vibration data,the model integrates the abnormal vibration feature discrimination mechanism of fault knowledge transformation and hierarchical Gibbs sampling algorithm to carry out fault probability inference for each node variable.Therefore,it can realize the timely location of abnormal vibration fault.Compared with other models,the proposed model has achieved great improvement in the accuracy of diagnosis results and distinguishing normal and abnormal data.The model is integrated into the intelligent operation and maintenance system of belt conveyor for engineering verification.