The biodiesel fuel is produced from vegetable oils and adipose tissue. Biodiesel with diesel fuel are used in
internal combustion engines in different ratios. Noise and vibration generated by combustion engines have
detrimental effects on users. This is more acute in diesel engines. A few researches of vibration biodiesel
blends exist in the world. Therefore, in this study, the vibrations of different blends biodiesel with diesel
fuel on a four-stroke diesel engine Perkins 1006-6 were evaluated. The experiments were performed in two
conditions, before and after the service engine. Artificial neural networks were used for modeling of the
vibrations. Because neural networks methods have more benefits than many of the usual definite statistical
methods. Results showed that the vibration values decrease considerably after the service engine. Fuel
blend, was significantly affected on vibration values. It was fixed that diesel engine with B40 and B20 fuel
blends have the lowest vibration and B15 and B30 fuel blends had the highest vibrations. The results demonstrated
that there is a good match between roots-mean- squares acceleration values and neural networks,
and the error rate approximately is close to zero in most patterns. Comparing the results of the neural network
simulation and experimental results showed that neural networks are a powerful tool for the modeling
of vibration in the engine.