Fault Diagnosis Of Unmanned Roller Compactor Based On Hybrid Model And Functional Safety Mechanism

With the rapid development of intelligent and networked technology,it is increasingly important to develop unmanned technology of rolling mill in improving rolling efficiency and productivity.However,due to the high frequency vibration and a large amount of dust in the rolling process of the unmanned rolling machine,the whole electronic and electrical system of the rolling machine will be affected,which makes the key parts of the unmanned rolling machine prone to failure and out of control,and then leads to safety problems.Therefore,it is an urgent engineering demand to develop a fast and accurate fault diagnosis and treatment mechanism for unmanned rolling mill.Based on this purpose,this paper mainly does the following research:Firstly,a fault diagnosis method of unmanned rolling mill based on hybrid model is developed.In order to realize reliable fault diagnosis of the positioning system of unmanned rolling mill,this paper proposes a solution of total disturbance observer and support vector machine(SVM)classification,and establishes a multi-body kinematics model with heading angle and vehicle speed as inputs and longitude,latitude and heading angle as outputs.The total disturbance estimated online and the measured position or heading angle are input into the support vector machine for fault classification.Experimental results show that compared with the traditional solution based on support vector machine,the average accuracy of fault diagnosis is 95%,and the accuracy and calculation time are increased by 9% and 12% respectively.Secondly,according to the results of GPS fault diagnosis,a fault handling mechanism based on functional safety is designed.By analyzing the failure mode and influence of GPS sensor,the potential risks of unmanned rolling mill caused by GPS failure are obtained.At the same time,based on the functional safety mechanism and expert experience method,the ASIL safety level of dangerous events occurred in the working site of the rolling mill is classified,and the safety level is decomposed into GPS faults and failures corresponding to dangerous events,and then different redundant safety measures are designed based on the safety level of GPS faults and the scenes where dangerous events occur.The experimental results show that the fault can be handled within 1 second after GPS fault is detected,and the functional safety mechanism can ensure that the unmanned roller compactor can respond efficiently and accurately when faults with low safety level and high safety level occur,thus ensuring the safe operation of the roller compactor.To sum up,this paper puts forward that the GPS fault diagnosis and functional safety treatment mechanism of unmanned roller compactor can quickly and accurately diagnose the GPS sensor fault,and at the same time designs reasonable fault treatment measures to ensure the unmanned roller compactor to operate safely and efficiently.