Research Of Remote Fault Diagnosis For Marine Vessels Based On Immune Clonal Algorithm

With the rapid development of modern science, technology and manufacturing process, the marine vessels system is more and more systematic, large-scale and complex, the equipment of marine vessels continuous working under high load, large power and harsh environment, fault problems will inevitably arise various, light is delay normal production cycle, heavy then endanger the safety of crews’ life and property on vessels, which caused serious economic losses and casualties. The fault monitoring and diagnosis system can detect and prevent the fault timely, and guide the process of shooting fault problems and hidden dangers, to make sure vessels equipment operate in safety, high efficiency and stable methods, reduce the rate of vessels equipment accident, prolong the service life of ship equipment. Based on the above analysis, this paper made research as follows.First of all, the present status of the vessels fault diagnosis at domestic and abroad is discussed, the shortcomings and limitations of the traditional fault diagnosis method to vessels is analyzed, and the drawbacks of previous vessels fault diagnosis means exist, combined with the characteristics of vessels, proposed the scheme of remote fault diagnosis for marine vessels, and discusses the various technical details and the overall feasibility of the scheme.Secondly, expounds how to improve the accuracy of fault diagnosis and the diagnostic efficiency for marine vessels. In order to achieve the purpose of the research, introduce the Immune Clonal Algorithm in Artificial Immune System. In order to make the algorithm more suitable for processing vessels equipment operating parameters, the clone and mutation operator in the algorithm are improved, enhanced the ability of escape and global optimization for the algorithm. And after the data pre-processing stage, before the formal process to the data use immune clone algorithm stage, the data is processed by clustering which introduce competition mechanism, make each excellent antibody in the cluster (each little search zone) can obtain the chance of cloning amplification for affinity maturation. Increase the diversity of antibody group in distribution, the algorithm has achieved a balance between depth and breadth in searching. Effectively avoid the occurrence of the phenomenon that the algorithm trapped in local optimal solution and result in the emergence of premature convergence. At the same time, this can effectively solve the problem of slow convergence of the Immune Clonal Algorithm, and effectively improve the accuracy of fault diagnosis.Finally, the Immune Clonal Algorithm which had improved is applied to one example of fault diagnosis for marine vessels. SQL Server 2005 is applied to establish database, which used to storage historical data and fault rule knowledge. MATLAB GUI is used as the tool to develop the system, the database toolbox in MATLAB is used to manipulate the data stored in the database. Using the above mentioned techniques, a simplified remote fault diagnosis system for marine vessels is developed. And through the example analysis, the practicability, accuracy and effectiveness of the remote fault diagnosis system for the marine vessels is verified.