Development of a test methodology for structurally integrated high-voltage storage devices

Predictive maintenance and online monitoring in vehicle construction require precise characterization of the components to be monitored. Together with BMW, Fraunhofer LBF has developed a new methodology for simplified testing and characterization of structure-integrated high-voltage batteries. This involves testing a part of the car body with an integrated high-voltage battery on a multiaxial vibration table. Bearings with adjustable stiffness replicate the stiffness of the rest of the car body, enabling different scenarios to be tested quickly.

Testing and characterization early in development

Ever-increasing time and cost pressures in development are also posing new challenges for component testing and the development of monitoring systems for vehicles. Increasing structural integration optimizes lightweight construction and space utilization in modern electric vehicles, but makes it more difficult to test the operational stability and characterize sub-structures such as high-voltage batteries. Test procedures that only test the entire vehicle in late development phases are cost-intensive and often lead to delays if weak points are only identified at a late stage. Isolated testing of individual sub-systems is often insufficient, especially in the case of structurally integrated high-voltage batteries, which also contribute to the structural integrity of the car body and are exposed to complex mechanical loads. However, early detection and correction of weaknesses can save considerable costs and shorten time-to-market.

MAST-based test method for high-voltage batteries

The methodology developed in collaboration with BMW for simplified testing and characterization of structurally integrated high-voltage batteries uses a multiaxial vibration table (MAST) to mechanically excite the test specimen, consisting of the high-voltage battery and the surrounding body parts. The use of a MAST significantly reduces the effort required compared to tests with the complete vehicle test bench.

The test specimen is connected to the MAST by means of eight bearings whose stiffness can be continuously adjusted between approx. 3 kN/mm and 20 kN/mm. The adjustable bearings allow different bearing conditions of the test specimen to be set with little effort. This makes it possible to simulate the stiffness of body parts that are not taken into account and to set different deformation states of the test specimen. This enables integration testing at an early stage of development. The LBF has developed the adjustable bearings and a structurally optimized overall design as well as a numerical optimization method. This methodology is used to determine the stiffness of the bearings to be set for the tests and to generate an optimized excitation signal for the MAST.

Successful proof of concept

As part of several projects, a numerical proof of concept for the test concept was first carried out in collaboration with BMW. This was followed by an experimental proof of concept on the Fraunhofer LBF's MAST and then on a BMW MAST. During these tests, various scenarios for connecting the test specimen were investigated and the test specimen behavior was evaluated on the basis of more than 150 measured strain and acceleration variables. In the current project phase, the methodology is being automated and transferred to BMW so that BMW will be able to use the methodology completely independently in the future.

The performance of the methodology has been demonstrated, in particular it has been shown that different stiffness settings can be used to achieve different deformation states of the test specimen. In addition, the strain distribution can be influenced by different stiffness configurations. This allows both stresses from body deformation and from the natural vibration behavior of the high-voltage storage assembly to be taken into account in the tests.

“By working with Fraunhofer LBF, we can develop our methods beyond the current state of the art. The results enable us to efficiently test high-voltage storage systems, and the methodology allows us to develop based on data.”

Sébastien Chéreau, BMW Group, Strategy & Methods Structural Durabiity