Research And Application Of Typical Faults Mechanisms And Monitoring And Diagnosis Method For Priston Engine

Piston engine, which is represented by diesel engine and gas engine, is widely used in oil fields, gas fields, underground gas storage and other industrial fields. As the core power equipment of these processes, the safe and stable operation of the piston engine can ensure the smooth operation of oil and gas exploitation and natural gas transportation. Piston engine structure is complex, with the features of more moving parts, more excitation sources and more wearing parts. The unit contains a large number of unbalanced inertia and outbreak forces, so that the failure rate is high. Once the piston engine has serious faults, such as the breakage of connecting rod, serious wear and tear of thrust bearing, broken teeth of gear box and piston cylinder failure, the engine will not recover for a short time. And this will have a bad influence on the production, causing huge economic losses to the enterprise and society, even producing a fire accident, endangering personal safety and affecting the reputation of the enterprise. At present, regular maintenance and breakdown maintenance are the main method in domestic piston engine maintenance.Judgment of the fault mainly depends on the ear, hand touch, thermal parameters monitoring and off-line measurement of body vibration, which a higher level of monitoring is using a single point of the vibration sensor for the whole vibration level monitoring. In this mode, the regular maintenance costs are high, and there are problems such as excessive or inadequate maintenance. Thermal parameter monitoring can not effectively reflect the mechanical fault of the unit, for example thermal parameters are not sensitive to cylinder scuffing, cylinder collision and connecting rod hit wear fault.Off-line measurement of body vibration can not reflect the real-time operation of the units; the monitoring effect is lagging behind. On-line single-point vibration monitoring is only measuring a single amount; it can not be timely warning. And when it worked, the equipment had been damaged. Therefore, it is of great significance to study how to effectively monitor the running state of piston engine in real time and develop an intelligent diagnosis system.This paper, based on the typical fault mechanism and characteristics study of piston engine, is aiming at ensuring the safe and stable operation of piston engine and avoiding the occurrence of major safety accidents of the engine, and researching on the on-line monitoring, fault diagnosis methods and monitoring and diagnosis system of piston engine. Then complete the experiment and practical application. The main contents of this paper are as follows:(1) The typical fault mechanism of piston engine was studied. Based on modeling and simulation, the typical fault mechanism and characteristic analysis of piston engine were carried out. Through researching on the engine thrust bearings abrasion fault, abnormal valve clearance fault, connecting rod bearing wear fault and engine combustion failure, the paper has laid the foundation for the monitoring and diagnosis method and the typical fault monitoring scheme.(2) Research on typical fault monitoring and diagnosis method of piston engine was conducted. According to the signal characteristics of piston engine,a new multi signals fusion diagnosis method of fault diagnosis based on wavelet packet and KPCA, a fault diagnosis method based on instantaneous speed and impact signal, and a diagnosis technique of abnormal valve clearance fault based on corner domain sensitive feature were presented. The effectiveness of these methods has been verified by practical case.(3) The research on the online monitoring and diagnosis system of piston engine fault was conducted. This paper has put forward the layout of the online monitoring system and the overall framework of the software, and studied the automatic fault diagnosis method for the typical fault of the engine.A 6-cylinder and 12-cylinder engine simulation bench was established, and typical fault simulation experiments of the engine were carried out. The research results have been applied to the actual field unit, which verified the practicability and reliability of the piston engine online monitoring and diagnosis system.