The Research On Risk Assessment Technology Of Reciprocating Compressor Under Running Condition

The reciprocating compressor is a kind of mechanical equipment widely used in various professions, has the advantages of wide pressure range and high efficiency, while also has many shortcomings such as frequent failure, many damageable parts, etc. In the petroleum and petrochemical enterprises, the reciprocating compressor is key unit for diesel hydrotreating, hydrogenation residue and gasoline hydrogenation device, once there is a serious fault, there tend to be the hydrogen leakage, fire and explosion. Fault monitoring diagnosis and maintenance of reciprocating compressor has been the focus work of the relevant enterprises. Because the Time-Based Maintenance and Breakdown Maintenance of reciprocating compressor are the usually method for the relevant enterprises, may maintain them insufficiently or overly, often waste huge human, material and financial resources and economic loss of the enterprises, and has the poor correlation between maintenance decision and running state of reciprocating compressor.In order to solve these problems, this paper from these two-pronged approaches to carry out research work:one is to improve the traditional RCM method, using unit history check and maintenance data to quantify the influence factors of safety, environment, production loss and maintenance costs, to improve the accuracy of fault modes and effects analysis. On the other hand, the RCM method of reciprocating compressor is combined with the date of on-line monitoring system, to study the application method in reciprocating compressor fault prediction maintenance of dynamic RCM. The main contents include:(1) Carry out the dynamic risk assessment and decision making research of running state of the reciprocating compressor, set up the unit fault mode database and calculate the typical fault destructive power by semi-quantitative analysis method, and on this basis to complete and improvement fault mode and effect analysis, a semi quantitative analysis method of fault mode and effect is proposed; Forecast the future fault of the unit by using grey forecasting model and historical check and maintenance date, to provide theoretical support for the enterprise spare parts management, and then put forward a model of fault maintenance strategy based on dynamic RCM.(2) Carry out the research on reciprocating compressor fault monitoring and diagnosis methods. A signal processing method for reciprocating compressor based on signal similarity analysis is proposed for the importance of angle domain analysis method and the characteristics of the unit key phase sensor often damaged; The fault monitoring and warning method for the dynamic energy index of the piston rod is studied to improve the accuracy of fault warning, in light of shortcomings of the existing piston rod settlement monitoring have poor results in monitoring piston rod fracture; After the fault sensitive feature parameters is extracted, a kind of running state dimensionless index evaluation method based on the fault sensitive characteristics is proposed at last.(3) Integrated these above contents, carrying out the typical fault maintenance strategy research and using the static RCM and dynamic RCM assessment method in the engineering.