Modal properties optimisation of dynamically loaded car parts to enhance their fatigue life

To enhance the accuracy in predicting fatigue life of dynamically loaded parts that are usually sub-components in a larger structure. The fatigue life prediction will be based on multi-axial-loading and broad-band inputs as well as on additional histogram-based loadings.

As dynamically loaded products and parts are exposed to fatigue, and consequently, to fracture, our aim is to make a suitable numerical model that will enable us to predict the expected fatigue life of such products in advance, in the prototyping phase of the development. Additionally, realistic cases with broad-band and multi-axial loadings, complex models and arbitrary boundary conditions will be considered. However, although one-directional and mono-component sinusoidal loading and the calculation of the fatigue life is relatively well described in scientific literature, both theoretically and experimentally, there is no simple and straightforward approach on how to predict fatigue life in general, realistic cases as described above. For example, in the exploitation of a product, e.g. car engine holder, which is usually included into a complex dynamic system like an engine-engine holder-chassis, the loading is neither mono-component nor one-directional; e.g. in a car, the main source of excitation forces are an engine’s dynamic forces, which are broadband and multidirectional. In the exploitation of a car product, the excitation forces depend on the driving regime, e.g. the loading of an engine holder depends on the rotational speed of the crankshaft, driving on motor ways or off-road, etc. Furthermore, it is necessary to study the dynamics of a broader dynamic system/structure, and not just the dynamics of the parts/components under consideration, as the latter has different dynamic characteristics than the parts themselves. As a result of all the described complexities, the automotive industry (OEMs-Original Equipment Manufacturer- and suppliers) runs experimental testing, aimed to be as close to real conditions as possible. This testing is expensive, usually long-lasting and is possible only after the prototype has been manufactured. Therefore, the research will be focused on finding the necessary correlation between the experimental and numerical results as a way to quantify the results for the fatigue life of the model modeller to be developed. The numerical part will mainly involve the construction of a valid model for a dynamically loaded part (realistic, multi-axial and broad-band loadings given in the frequency domain, usually with additional histogram-based information, with realistic material properties and realistic boundary conditions) which will be a basis for the extraction and calculation of suitable, equivalent stresses of the part. These stresses are then used in a suitable fatigue calculation procedure to predict the fatigue life. The experimental part will in general serve to validate the numerical model and to validate the numerical results for the fatigue life. At the end, the final outcome of the project will be a product, a computer program, which will enable an engineer to calculate fatigue life predictions and simple, parametric what-if studies for realistic, dynamically loaded parts. Furthermore, the final product will also be suitable for integration to other commercial and non-commercial software packages, and for being extended/used in general, optimisation studies.
Project ID: 
4 480
Start date: 
Project Duration: 
Project costs: 
400 000.00€
Technological Area: 
Computer Software technology
Market Area: 
Motor Vehicles, Transportation Equipment and Parts

Raising the productivity and competitiveness of European businesses through technology. Boosting national economies on the international market, and strengthening the basis for sustainable prosperity and employment.