论文部分内容阅读
为了提高煤矿通风机故障诊断的准确性,提出了基于证据理论的混合诊断算法。即先用灰色建模方法实现故障特征量的累加处理,以增强数据的规律性。然后,采用2个并联的灰色BP网络进行故障局部诊断,获得彼此独立的证据。最后,再用证据理论融合算法对各证据进行融合,最终实现通风机的故障诊断。实例结果表明,该方法可有效提高诊断的可信度。
In order to improve the accuracy of coal mine ventilator fault diagnosis, a hybrid diagnosis algorithm based on evidence theory is proposed. That is to say, the gray modeling method is used to realize the accumulation of fault features to enhance the regularity of the data. Then, two parallel gray BP networks are used to diagnose the fault locally and obtain the independent evidence. Finally, the evidentiary theory fusion algorithm is used to fuse all the evidences to finally achieve the fan fault diagnosis. The experimental results show that this method can effectively improve the reliability of diagnosis.