Instruments and Systems: Monitoring, Control, and Diagnostics Annotation << Back Method for determining the faulty cylinder of an internal combustion engine based on the time-frequency correlation analysis V.S. AVRAMCHU K, V.A. FAERMAN The article concerns the research and method development of digital signal processing applied to the problem of vibration defect diagnosis of cylinder-piston group of internal-combustion engine and particularly to the identification of fault cylinder. Known methods are not versatile, not well suitable for automation and have limited applicability. As a part of the research, the article provides the brief review of fundamental techniques of ICE vibration signal analysis, notably, frequency-domain, time-domain and time-and-frequency-domain approaches. As an alternative to the widely known methods of timeand- frequency analysis, authors suggest the method, based on novel mathematical apparatus of time-and-frequency correlation analysis. In order to estimate the feasibility of the method to the problem of fault cylinder identification, vibration signals were simulated in regular operation mode as well as in the presence of fuel feed subsystem malfunctions. The simulation and the analysis were performed with Mathcad 15, Matlab 2015b and with customized software, developed by the authors. Based on the analysis result, the set of fault symptoms was generated, which allows fault identification and detecting malfunctioning cylinder. The surfaces of time-and-frequency autocorrelation functions, which shapes indicate fault cylinder, were selected as fault symptoms. As a part of method validation, experimental studies of several internal-combustion engines were conducted, specifically, VAZ 2101. The results obtained in these studies confirm the conjecture on feasibility of time-and-frequency correlation analysis for problems of fault cylinder identification. Keywords: internal-combustion engine, vibration defect diagnosis, cylinder-piston group, signal processing, autocorrelation function, time-and-frequency analysis. Contacts: E-mail: avs@tpu.ru Pp. 37-43.