Research on Fault Diagnosis Method of Scraper Conveyor in Start Stop Stage Based on DDNN

Aiming at the existing problem of excessive dependence on manual experience,low accuracy rate,and poor real-time performance in the traditional scraper conveyor start stop stage fault diagnosis,a DDNN based scraper conveyor start stop stage fault diagnosis method is studied.Five types of common start stop faults in the scraper conveyor start stop stage are analyzed,and the DDNN model theory is analyzed.The obtained monitoring data of the scraper conveyor start stop stage are deployed at the edge and cloud according to the difficulty level using digital graph transformation and convolution feature bags;Implement 5 types of start stop fault diagnosis through cloud edge collaborative reasoning mechanism.The experiment results show that after adopting the cloud edge collaborative reasoning mechanism,the fault diagnosis accuracy can be improved to 99.52%,the communication cost can be reduced by 14.88%,and the goal of fault diagnosis during the start stop stage of the scraper conveyor is achieved without relying on manual experience,with high diagnostic accuracy and good real-time performance.