Transformation of Nonstationary Signals into Pseudostationary Signals for the Needs of Vehicle Diagnostics

Abstract

Research concerning the development of diagnostic methods of the state of the vehicle driving systems basic elements — have been carried on for many years. Modern diagnostic systems, regardless of being technically highly advanced or small degree only, concern about assessments of basic mechanical subsystems such as toothed gears or suspension elements. The information on state changes of mechanical systems can be obtained by an analysis of changes of nonlinear effects in such systems operations. Since the nonlinearity increase accompanies the system destruction, the separation of nonlinear parts of signals creates the possibility of the increase of an accuracy, reliability and application range of diagnostic methods. Technical state diagnostics for vehicles is much more difficult than for machines operating in stationary conditions. the main excitation source is an internal combustion engine, in which a rotational speed changes from an idle running (approximately 800 rpm) to a maximum rotational speed (e.g. above 5000 rpm). Thus, an application of classic analytical methods is not possible. the only solution is the transformation of the analysed signals into the “pseudostationary” ones, by taking into account changes in the engine rotational speed. One of the methods applied now a days is resampling (adaptation change of a signal time resolution in a digital form) performed on the basis of information concerning changes in an engine rotational speed. the signal from the sensor of the engine crankshaft location was used for the above application.