MVM 2022 - Volume 48 - Number 2
  1. A NEURAL NETWORK-BASED CONTROL ALGORITHM FOR A HYDRAULIC HYBRID POWERTRAIN SYSTEM
    Authors: Sonja Kostić, Bogdan Živković, Aleksandra Ivanović, Dragan Džunić, Suzana Petrović Savić,Vladimir Kočović
    DOI: 10.24874/mvm.2022.48.02.05
    Abstract

    Many automotive parts are exposed to complex dynamic stresses during their exploitation. For the design of such elements as shafts, axles, etc., it is necessary to perform fatigue testing in order to obtain permanent dynamic strength of the part. For the purposes of such tests, a device has been developed that provides the possibility of dynamic testing of parts loaded only by torsion, only on bending or parts that are loaded with combined bending and torsional stress. The device provides the possibility of testing cylindrical workpieces. The obtained experimental data show that the maximum error of measuring the dynamic strength of materials at complex stresses is 5%.


    Keywords: body structures, fatigue strength, devices for testing material, dynamic durability, combined stress