Development Of Fault Diagnosis Application System For Turbofan Engine Based On Data

Turbofan engines require acceptance test before delivery to customer. The test data need to beanalyzed to obtain the performances of the engine and its control system, then the conclusion to eachsystem of the engine whether normal or not can be made. However, test data are usually analyzed byoperators manually and several defects such as big labor intensity, high human and material costs andso on are emerged in this way. Therefore, it is very important significance and engineering practicalvalue to develop an application system based on data for turbofan engine fault diagnosis.Using the advanced database management technology, a specific engine condition monitoringand fault diagnosis system is designed in this paper. The turbofan engine sensor fault offline andonline diagnosis is explored based on support vector machine (SVM). And then the turbofan enginefault diagnosis system mainly including two functions: the test data abnormal monitoring and faultdiagnosis is established. This system uses B/S framework and Oracle10g database for the test datafault information processing, it realizes the fault diagnosis function of test data in different test modefollowing the functional requirements.Using the turbofan engine fault diagnosis system to test seven kinds of experiments includingself-checking, running-in, wet starting and so on, the results show that the system can accuratelydetect abnormal information and diagnose fault. The successful development of this system is notonly meeting the needs of the users, but also giving full play to the value of test data and providingreliable technical guarantee for the engine design, maintenance and experiment.