Development of a self­diagnostics subsystem of the information­measuring system using anfis controllers

• Main Authors: Ihor Zhukovyts’kyy, Ihor Kliushnyk
• Format: Article
• Language: English
• Published: PC Technology Center 2018-02-01
• Series: Eastern-European Journal of Enterprise Technologies
• Subjects: hydraulic transmission of diesel locomotive; information-measuring system; parameter measurement sensors; neural-fuzzy controller
• Online Access: http://journals.uran.ua/eejet/article/view/123591

A hybrid self-diagnostic system was designed to evaluate correctness of functioning of sensors of the information-measuring system of testing hydraulic transmissions of diesel locomotives of UHP 750 type. The system features the possibility of checking certain four parameters in steady-state operation conditions using known mathematical dependencies. For the other 14 parameters (for which mathematical dependencies were not studied and which have a high complexity of calculations), 14 neural-fuzzy ANFIS networks were developed. Self-diagnostic algorithms using ANFIS controllers were elaborated. The algorithms provide prediction of individual system parameters with the help of ANFIS controllers and a further comparison of the predicted parameters with the measured parameters. The ANFIS controller structure with the proposed Sugeno rule set was constructed and its efficiency was shown. Network training and test of the diagnostic subsystem were performed using the data sets obtained in a series of tests of hydraulic transmissions conducted at Promteplovoz diesel locomotive repair plant. The test results have shown that application of the proposed procedure ensures obtaining of correct result of the self-diagnostic subsystem operation.