Realistic Operating Conditions from Vehicle Modeling: An Application to Coolant Distributor Valves

Abstract

<div class="section abstract"><div class="htmlview paragraph">Vehicle electrification is bringing new challenges to the design of components for the automotive sector. New system requirements and functions are forcing either the development of new components or a complete redesign of the existing ones. In the absence of detailed pre-existing knowledge on operating conditions for these components, conservative requirements tend to result in overengineering. System modeling at vehicle level is a valuable approach in these circumstances, which can be used to efficiently estimate such conditions. With modeling, it is possible to define performance targets for components at an early development stage and to verify the impact of component design choices on vehicle performance. In this work we construct a full-vehicle model, which we use to frame the development of coolant distributor valves for electric powertrains.</div><div class="htmlview paragraph">In the first part of the work, we define the topology of the coolant circuit and the relevant interconnected systems (e.g. electrical power network, HVAC) based on vehicle teardown data. We identify representative operating conditions (e.g. driving cycle, ambient conditions). We combine the various systems into a vehicle global energy model.</div><div class="htmlview paragraph">In the second part of the work, we assess the influence of key design parameters for coolant distributor valves, such as the internal leakage, on global vehicle performance. The vehicle model includes a fluid-dynamic model of the valve calibrated on test measurements, and a simple control logic to define valve behavior as a function of the vehicle status (e.g. battery temperature, cabin requirements). We perform a parametric analysis for the internal leakage of the valve. With this analysis we can determine a leakage threshold up to which the energy efficiency of the vehicle and the quality of thermal management - expressed as time required to reach a temperature target - is not significantly affected. In turn, we show how realistic design constraints can be determined early in the development cycle of the system, avoiding overengineering and accelerating the development process.</div></div>