Refrigerated Container Ship Based On Neural Network Fault Diagnosis

Reefer containers transportation developed rapidly with the development of the economy, but the number of the case about the cargo loss has been increasing. So it is necessary to develop a fault diagnose system based on information technology. When this system is used on board a ship, it can decrease the work of the crew.The dissertation begins with fault diagnoses and then introduce the reefer containers, such as their specification management and so on . Third, because reefer containers have the character of being transported in the world, it is not easy to get the full data of specified containers. Based on the fault reports and repair reports on board, we get the frequencies and the repair expenses of the major faults of the refrigerating unit. And we find out that the compressor of the refrigerating unit is the most serious fault. Fourth, the author introduce the basic frame of the diagnose system of fault based on the neural network. Fifth, the author develops the fault example through the experiment. At last, according to the test data above, the author develop the fault diagnose system of reefer containers based on neural network with Visual Basic language.The fault can be divided into a hard fault and soft fault according to its severity. Hard trouble is generally greater and sudden, so it is easy to be measured The soft trouble usually takes place gradually.it is difficult to be measured in the initial stage . In a sense, the danger of the soft trouble is greater than the hard trouble. So we should pay more attention to the soft fault.Because the time of the study is limited ,the parameter example and hard fault example collected are not perfect. If the LAN is set up on ocean-going ship and this software is Installed in the local computer This software can predict the fault. When the hard fault happens, the engineer on board can carry on the treatment according to guidance on the software.