Diagnostic system of ship power plants using neural network models

In order to reduce the number of accidents in the domestic courts it is necessary to equip a vessel with monitoring systems of the state of the main and auxiliary machinery. One of the major shortcomings in the existing monitoring systems is the inability to determine the initial stage of the disruption of the system. The main function of most of the existing monitoring system is to remove the parameters from the sensors and to display the results of the crew and the shipowner. The use of neural network technology in the diagnosis of problems will help not only to fix the sensor readings and compare them with reference values, but also to analyze the parameters of the system obtained in the complex, predicting the possibility of occurrence of the failures of both separate elements and the whole system.