PSI - Issue 59

Victor Aulin et al. / Procedia Structural Integrity 59 (2024) 436–443

442

Victor Aulin et al. / Structural Integrity Procedia 00 (2019) 000 – 000

Fig. 4. Graph model of an artificial neural network of recognition of cylinder-piston group defects.

Fig. 5. Graph-model of an artificial neural network of recognition of defects in the valvetrain.

As a result of the testing of the built ANN models based on the data obtained in real production conditions, a combination of defects of the main resource-determining elements (CPG, KShM, timing belt) of the YMZ-238 was formed, which were entered into the database. These data made it possible to manage the condition of the CPG, KShM, timing belt, and therefore the engine resource, implementing the distribution of the optimal complex of technological maintenance and repair operations. 4. Conclusions The construction of structural-consequential models of relationships of controlled parameters for the main resource elements of the YMZ-238 engine was carried out on the basis of its design documentation and information on the functioning of its elements. This makes it possible to assess the technical condition at the stages of diagnosis and operational control during experimental research.