Font Size: a A A

Development Of Fault Diagnosis System For Analog Circuit Of Marine Power Plant Control Systerm

Posted on:2022-06-16Degree:MasterType:Thesis
Country:ChinaCandidate:Y LiFull Text:PDF
GTID:2532306839983159Subject:Instrumentation engineering
Abstract/Summary:PDF Full Text Request
As military strength continues to increase,the technology used in marine power plants is also constantly evolving.As a key component of the marine power plant control system,the analog board directly determines the precise operation of the marine power plant and whether it is used for a long time.Affected by factors such as high temperature and high humidity in the working environment of the marine power plant,circuit failures caused by component damage,drift,performance degradation and other reasons frequently occur,which seriously affects the reliability and safety of the entire power plant,which bothers the enhancement of the ship’s ability to perform missions.Therefore,it is of great significance to conduct fault diagnosis research on the analog board of the marine power plant control system.In response to the need for rapid and accurate identification of fault types during the maintenance and repair of the marine power plant control system analog board,this research takes the marine power plant analog board as the object,starts with the failure mechanism and degradation analysis of key components,carries out the research on the hard fault and soft fault diagnosis methods of the marine power plant control system,and develops the fault diagnosis system of the marine power plant control system simulation board.The main contents of research are as follows:1.Analyzed the composition and working principle of the analog board of the marine power plant control system,selected passive components such as resistors and capacitors and active components such as DC-DC power supplies to analyze the failure mechanism and characterization parameters,and established a typical hard fault mode database of the marine power plant control system analog board.On this basis,a hard fault diagnosis method is proposed,which realizes the accurate identification of the hard fault type of the analog board of the marine power plant control system.2.The soft fault diagnosis method of the analog board of the marine power plant control system is studied.Aiming at the problem of unbalanced fault data caused by a higher probability of single fault occurrence than double faults in the actual working process of the analog board,this paper proposed an analog circuit soft fault diagnosis method based on the Enhanced Boundary Equilibrium Generative Adversarial Network(EBEGAN).It introduced the Generative Adversarial Network into the field of analog circuit fault diagnosis,and at the same time add Conditional Variational Auto-encoder to the generator of the Generative Adversarial Network,so that the quality of the generated sample is guaranteed to be more diverse and the generated sample is closer to the real sample.It increased the classifier and classification loss,so that the Boundary Equilibrium Generative Adversarial Network can identify the type of faults.The simulation test results of the standard data set circuit and the actual control system analog board circuit show that the proposed method can effectively and stably diagnose the fault when the analog circuit is in an unbalanced data situation.3.Analyzed the hard fault and soft fault diagnosis requirements of the marine power plant control system analog board,analyzed the system development framework from the perspective of hardware and software,selected reasonable hardware products to build the fault diagnosis system;designed the fault diagnosis software according to the software engineering method,and realized the development of fault diagnosis software code;build a system test and verification platform,and carried out the function test and verification of the fault diagnosis system.The test and verification results indicate that the marine power plant control system analog board fault diagnosis system developed by this study can realize online real-time and fast diagnosis of hard faults,realize soft fault diagnosis under unbalanced data.That shorten the maintenance time and workload of ships,ensured the safety and stability of our country’s ship operations.
Keywords/Search Tags:Marine power plant, Analog circuit, Fault diagnosis, Generative Adversarial Networks, Conditional Variational Auto-encoder
PDF Full Text Request
Related items