PSI - Issue 57

Ewelina Czerlunczakiewicz et al. / Procedia Structural Integrity 57 (2024) 743–753 / Structural Integrity Procedia 00 (2019) 000 – 000

746

4

The effect of the main simulation parameters on the predicted simulated fatigue life were identified via pure linear simulations, and on an input loading that had a deterministic nature (sine sweep). This approach has shown to be useful to quantitatively rank the effect and the importance of the main simulation parameter, as described by Czerlunczakiewicz et al. (2023). In the present study, we aim at calculating the simulation fatigue life in a more representative scenario, by proposing a new model that can consider a random input load, the presence of nonlinear response among the components mounted on the carrier (the cooling radiator) and by using transient methods.

Methodology

2.

2.1. Simulation modeling of the ECM The system is mounted on the front end of the car, positioned on the chassis and supported by a top frame (the bolster). It consists of several heat exchangers responsible for the efficient thermal management of the powertrain, cabin, and battery cooling module in the case of hybrid or electric vehicles. The main component is the cooling radiator, which is attached to the top frame and the chassis using four decoupling isolators to filter vibrations generated by the cooling fan and transferred by the chassis. Additional exchangers, such as the condenser (as part of the AC loop), the air-cooled intercooler, and low temperature radiators, can be mounted on the radiator. These exchangers usually are mounted on the radiator using clipped brackets. The clipping connections need to be mathematically modeled in order to be implemented in the FEA. The FEA modeling of the bracket poses a significant challenge. The typical approach involves considering all contact points as rigid joints. In practice, subcomponents are mounted in a way that simulates bonding in the chosen direction to the other component. The advantage of this approach is that the displacement of each component is linear with the input vibration. The disadvantage is that it linearizes (simplifies) the connection. However, a more realistic approach involves describing the junction behavior with a non-linear definition. This includes defining the stiffness of the junction dependent on the acting force, as well as determining a clearance or gap between surfaces to allow components to touch and detach (micro shocks implementation) during the random load. The application of a non-linear friction definition is possible as well. The main challenges are how to correctly represent such complicated behavior and the necessity of using nonlinear dynamics simulation methods. 2.2. Applied techniques A dedicated case study was designed to evaluate the vibration fatigue life prediction using different calculation methods. The product undergoing the calculations is an ECM for a mild hybrid electric vehicle already in serial production. Engineers who are familiar with FEA dynamic calculations are aware that simulating complex industrial products, such as the engine cooling system, can result in very long calculation time. The objective of the study is to compare the results obtained from the different approaches to dynamic simulations when applied to durability design validation of an industrial cooling system undergoing vibration fatigue loadings. As a first step to reduce calculation time for FEA models, a super-element approach was used. This method significantly reduces the Finite Element model size by retaining only necessary data to represent the examined system (as shown in the Fig. 2). One can therefore consider a non-variable part of the model as stiffness, mass, and damping matrices representation. The disadvantage of this method is that structure inside the superelement is represented as fully linear.