Dynamic Response Evaluation of a Chassis of a Generator Set Using FEA techniques

A Generator set is comprised of mainly an Engine, Alternator and Chassis. High Horse-Power Generator development is challenging, with lots of complexities in physical and virtual validations. Creating high fidelity analytical model is always beneficial and economical at the design stages as it avoids repetitive tests on various design concepts. This paper reports analytical methods of developing an FEA model of a Generator for locomotive application and its correlation with Test. Highlighted as well are some of the challenges faced in FE modeling of a large Generator model (60 liters engine capacity) with node count of around 4 million. In this technique, Modal Analysis is first performed to capture the dynamic behavior. More than 95 % correlation is achieved between the FEA and test natural frequencies (Bending modes). Harmonic Analysis with Modal Superposition is then applied to understand the dynamic response of a Chassis under the action of engine vibratory loads. In this paper commercial FEA software Ansys was used for evaluating chassis dynamic response using Modal and Harmonic analysis techniques. Engine Cylinder forces like Piston axial, Side forces and the Main bearing vertical and lateral forces are used as excitation forces in FEA model. Experimental Modal Analysis (Cold Engine) and Steady State Operating test (Hot/Running Condition) were used as physical validations using LMS Test.Lab. Despite limited availability of some input information, this method has shown encouraging correlation up to 90% when correlated with operating deflection shapes on a running Generator set data for lower order harmonics. Chassis design life is also validated for infinite vibratory load cycles using FE-Safe.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01703203
  • Record Type: Publication
  • Source Agency: SAE International
  • Report/Paper Numbers: 2019-26-0198
  • Files: TRIS, SAE
  • Created Date: Jan 15 2019 11:16AM