Research On Intelligent Operation And Maintenance Management Method Of Marine Diesel Generator

With the development of modern science and technology,ships are constantly moving in the direction of intelligence,digitization and networking.As the core equipment of the ship,the marine diesel generator directly affects the safe operation of the entire ship.Due to its complex structure and many moving parts,the maintenance and management of the marine diesel generator becomes extremely complicated.It is the focus of ship intelligence research to develop the intelligent operation and maintenance management strategy of marine diesel generators through the guidance of modern scientific intelligent diagnostic technology.This study focuses on the intelligent operation and maintenance management strategy of marine diesel generators.By analyzing the reliability of diesel generators,various factors related to the health factors of marine diesel generators are established,and principal component analysis is built based on these factors.PCA)and BP neural network model are used to evaluate the health status of marine diesel generators.A probabilistic neural network(PNN)model is also established to identify diesel generator faults,and diesel power generation is based on ship diesel generator health status assessment and fault identification.Machine intelligent operation and maintenance management strategy,the specific work is as follows:(1)Establish a reliability index system for marine diesel generators,comprehensively consider the three factors of environment,personnel and equipment to classify the health status of marine diesel generators into five risk levels,and determine the health indicators of11 diesel generators..And through the quantitative processing,a neural network training matrix for the health risk level of marine diesel generators is constructed.The neural network training matrix of PCA is used for dimensionality reduction.The BP neural network is trained by the training matrix after dimensionality reduction and the health status of the marine diesel generator is evaluated.The evaluation results show that the PCA-BP neural network has higher evaluation accuracy and evaluation efficiency,which can meet the requirements of ship diesel generator health assessment.(2)Using the diesel generator vibration signal as the sample data of fault diagnosis,select the time domain frequency domain characteristic value of the vibration signal and use the stack self-encoder to select and reduce the selected vibration signal characteristic value.The probabilistic neural network algorithm is then used to identify the faults of the collected diesel generator fault data.According to the fault identification result of the marine diesel generator,the corresponding intelligent operation and maintenance management method is formulated.(3)Combined with the health status assessment and fault identification of diesel generators,the intelligent operation and maintenance database and operation management strategy are designed.Under the normal monitoring mode,the corresponding intelligent operation and maintenance management strategies are formulated according to the different health status of the marine diesel generators,and the diesel oil is evaluated through the neural network.The generator health status makes appropriate adjustments to the maintenance schedule.When the fault occurs,the fault diagnosis method is used to make a preliminary diagnosis and alarm.In the fault repair,the appropriate maintenance guidance scheme is given based on the currently evaluated diesel generator health status.