| Marine diesel engine is the power heart of ship.With the increasing of diesel engine strength,the possibility of failure increases correspondingly.Real-time condition monitoring and health assessment of the marine diesel engine is an important technical instrument which can increase the safety and the reliability of the ship.Based on vibration signals and thermal parameters of a marine diesel engine,the health monitoring and assessment system is developed and verified in the paper.The fault signs of the marine diesel engine can be found as early as possible by analyzing the processing signals measured and data mining.Therefore the failure parts and components of the marine diesel engine can be repaired in time.The research and development are of great significance for improving the reliability of marine diesel engine and of the engineering application value.The main research contents and conclusions are shown as follows:1、The development status of monitoring diagnosis and health status assessment technology for marine diesel engine are summarized.Finishing The overall design of the marine diesel engine health status monitoring and assessment data management system is carried out by a modular design method.The functions of the subsystems which are the embedded machine-side control,the on-line health monitoring and the off-line health analysis are designed.Also,the data interaction among subsystems is confirmed.2、An embedded control subsystem for marine diesel engine is developed.The components of each module of the subsystem and the internal and external communication modes are analyzed.The STM32 microcontroller is selected and designed as the hardware of the subsystem,and the functions of the data acquisition,the digital input and output are realized.According to the requirements of the control logics,the basic software flow is developed.It is verified that the embedded engine-side control subsystem of the marine diesel works reliably and stably by the tests of the diesel engine.It is also verified it’s the strong anti-electromagnetic interference ability.The embedded control subsystem meets the requirements of China Classification Society(CCS)Code for Classification of Steel Marine Vessels(2012).3、The hardware selection and software development of the on-line monitoring subsystem for the marine diesel engine health status are carried out.The vibration and temperature sensors and hardware acquisition platform are selected and designed.The vibration signal processing and the feature extraction methods in the time and frequency domains are studied.The vibration signal analysis and processing and feature extraction are carried out in the time and frequency domains respectively.The criterion of the health assessment and the number of the on-line signals are established.Based on the database,the real-time evaluation and data storage of the marine diesel engine health status are realized.4、The data mining technology of the vibration signals is studied by using Support Vector Machine and Fuzzy C Algorithm.The data mining of the thermodynamic parameters is studied by setting threshold generating rule table.Combining with the characteristic values of the diesel engine vibration signals and the parameter rule tables established,the off-line analysis subsystem software of the marine diesel engine health status is developed,and the fault information is deeply excavated.A relatively complete database is developed.All parameters,the data management and the storage of the marine diesel engine are realized.5、The experiments of the marine diesel engine health status monitoring and assessment data management system and the measurement system are carried out on the test bench of a type MAN 6L16/24 marine medium-speed diesel engine.The typical faults simulated of the marine diesel engine are carried out.The fault characteristic values of the vibration signals are extracted,and the feasibility of the analysis and the treatment of the on-line monitoring subsystem for the marine diesel engine is verified.The fault eigenvalues are imported into the off-line analysis subsystem,and the accuracies of the off-line analysis and diagnosis is verified by data mining.At the same time,the running data of the test bench is accumulated. |
