Studying the possibility of neural network application in the diagnostics of a small four-stroke petrol engine by wear particle content

U radu je prikazana primjena umjetne neuronske mreže (ANN) u dijagnostici motora. Primijenjena je unapređena neuronska mreža s povratnim rasprostiranjem greške s Levenberg-Marquardt algoritmom učenja na problemu procjene indeksa sadržaja čestica trošenja (WPC) s brojem sati rada motora kao ulaznim parametrom mreže. Generalizacijska svojstva razvijene umjetne neuronske mreže (ANN) su vrlo visoka, što je i potvrđeno vrlo dobrim međusobnim poklapanjem eksperimentalnih iznosa (nisu bili uključeni u proces učenja) i procijenjenih iznosa indeksa sadržaja čestica trošenja (WPC).This paper presents the application of artificial neural network (ANN) in engine diagnostics. One-layer feed-forward neural network which is trained using a back-propagation algorithm that updates the weights and biases values according to Levenberg-Marquardt algorithm has been established to predict the Wear Particle Content (WPC) using the number of working hours of the motor engine as input parameter. The generalization property of the developed ANN is very high, which is confirmed by a very good match between the predicted and the targeted values on a new data set that was not included in the training data set