VIBRATION MONITORING OF INTERNAL COMBUSTION ENGINE

Abstract

The work is devoted to the problem of diagnosing automobile internal combustion engines.
The problem of controlling the condition of internal combustion engines is now the most relevant
because of the increasing number of cars and tightening of environmental requirements. The paper
considers the consequences of a faulty internal combustion engine. The aim of the work is to
develop a method that can help to detect the fault most accurately and quickly. With the advent of
modern technologies the long known method of internal combustion engines condition estimation
by the sound can become the most advanced, as the human factor is excluded, for signal processing
the computer techniques of the sound spectrum analysis which is carried out by means of
artificial neural networks are applied. The application of artificial neural networks for analyzing
the sound spectrum has found application in speech recognition and for diagnosing respiratory
diseases. In the article the mechanisms which are capable to generate sound signals during internal
combustion engine work are considered, some of them are phased i.e. are connected to working
strokes, some of them are not phased. The proposed method of diagnostics allows selecting
"useful" sounds from the total number of engine noises and, after a comparative analysis, pointing
to a node whose sound differs from the reference, serviceable one. Scientific novelty lies in the fact
that the diagnostic process becomes automated, all the sounds recorded by sensors are processed
in a computer or a special scanner, the display shows information about the condition of certain
nodes, in contrast to traditional methods where the diagnosis is carried out visually or by ear. This
increases diagnostic accuracy and reduces overall labor intensity by eliminating partial or complete
engine disassembly