Search Results

The safety, performance, and operational life of power dense Lithium-ion batteries used in Hybrid and Electric Vehicles are dependent on the operating temperature. Modeling and simulation are essential tools used to accelerate the design process of optimal thermal management systems. However, high-fidelity 3D computational fluid dynamics (CFD) simulation of such systems is often difficult and computationally expensive. In this paper, we demonstrate a multi-part coupled system model for simulating the heating/cooling system of the traction battery at a module level. We have reduced computational time by employing reduced-order modeling (ROM) framework on separate 3D CFD models of the battery module and the cooling plate. The order of the thermal ROM has also been varied to study the trade-off between accuracy, fidelity, and complexity. The ROMs are bidirectionally coupled to an empirical battery model built from in-house test data.

A lumped electro-thermal model for a battery module with 14 cells in series (14S1P), and with a cooling channel, is created by two-way coupling of an equivalent circuit model (ECM) and a linear time-invariant (LTI) method based thermal reduced order model (ROM). To create the ROM, a step response data in the form of temperature versus time curve is required. This data is obtained by running a transient full three-dimensional (3D) computational fluid dynamics (CFD) analysis for the full module. The thermal ROM accounts for the effect of the heat generated by the active cells, the joule heat generated in tabs and connectors, and the coolant inlet temperature. To create an ECM, data from hybrid pulse power characterization (HPPC) test is used. Such a lumped electro-thermal model for a battery module can run faster than a 3D CFD analysis and can be easily integrated in a system level model.

This paper presents a thermal electric two-way coupled li-ion battery pack model for an all-electric VW motorsport racer. It starts from the hybrid pulse power characterization (HPPC) test data at different state of charge (SoC) and temperature levels. Such information is used for cell level battery equivalent circuit model (ECM) parameter identification. Multiple cell ECMs are connected in series to create a module ECM. Battery thermal performance of the module is simulated first by computational fluid dynamics (CFD) for the module. Then, a thermal reduced order model (ROM) is created out of the CFD solution. The thermal ROM is then two-way coupled with the battery module ECM to form a complete battery module model. Multiple module models are connected to create a battery pack model. The complete pack model is then exported into Simulink for validation and simulation.

Due to growing interest in hybrid and electric vehicles, li-ion battery modeling is receiving a lot of attention from designers and researchers. This paper presents a complete model for a li-ion battery pack. It starts from the Newman electrochemistry model to create the battery performance curves. Such information is then used for cell level battery equivalent circuit model (ECM) parameter identification. 28 cell ECMs are connected to create the module ECM. Four module ECMs are connected through a busbar model to create the pack ECM. The busbar model is a reduced order model (ROM) extracted from electromagnetic finite element analysis (FEA) results, taking into account the parasitic effects. Battery thermal performance is simulated first by computational fluid dynamics (CFD). Then, a thermal linear and time-invariant (LTI) ROM is created out of CFD solution. The thermal LTI ROM is then two-way coupled with the battery pack ECM to form a complete battery pack model.

The thermal behavior of a fluid-cooled battery can be modeled using computational fluid dynamics (CFD). Depending on the size and complexity of the battery module and the available computing hardware, the simulation can take days or weeks to run. This work introduces a reduced-order model that combines proper orthogonal decomposition, capturing the variation of the temperature field in the spatial domain, and linear time-invariant system techniques exploiting the linear relationship between the resulting proper orthogonal decomposition coefficients and the uniform heat source considered here as the input to the system. After completing an initial CFD run to establish the reduction, the reduced-order model runs much faster than the CFD model. This work will focus on thermal modeling of a single prismatic battery cell with one adjacent cooling channel. The extension to the multiple input multiple output case such as a battery module will be discussed in another paper.

Due to growing interest in hybrid and electric vehicles, the battery, being one of the critical components, is receiving a lot of attention from designers and researchers. Two battery-modeling approaches, though seemingly different, share the same mathematical challenge of robust non-linear curve-fitting. The two methods are battery equivalent circuit model and battery system level thermal modeling using the linear time-invariant (LTI) method. Both modeling approaches involve curve-fitting testing data or data from advanced models to identify four parameters in a circuit model consisting of two pairs of RC elements. Such curve-fitting is mathematically a non-linear least-squares (LS) problem. Standard methods like the Levenberg-Marquardt (LM) method can be used for non-linear curve-fitting, but the LM method is known to be sensitive to initial conditions.

A commonly used physics based electrochemisty model for a lithium-ion battery cell was first proposed by professor Newman in 1993. The model consists of a tightly coupled set of partial differential equations. Due to the tight coupling between the equations and the 2d implementation due to the particle modeling, and thus called pseudo-2d in literature, numerically obtaining a solution turns out to be challenging even for a lot of commercial softwares. In this paper, the VHDL-AMS language is used to solve the set of equations. VHDL-AMS allows the user to focus on the physical modeling rather than numerically solving the governing equations. In using VHDL-AMS, the user only needs to specify the governing equations after spatial discretization. A simulation environment, which supports VHDL-AMS, can then be used to solve the governing equations and also provides both pre- and post- processing tools.

Battery thermal management for high power applications such as electrical/hybrid vehicles is crucial. Modeling is an indispensable tool to help engineers design better battery cooling systems. While Computational Fluid Dynamics (CFD) has been used quite successfully for battery thermal management, CFD models can be too large and too slow for repeated transient thermal analysis especially for a battery module or pack. An accurate but much smaller battery thermal model using a state space representation is proposed. The parameters in the state space model are extracted from CFD results. The state space model is then shown to provide identical results as those from CFD under transient power inputs. While a CFD model may take hours to run depending on the size of the problem, the corresponding state space model runs in seconds.

Computational Fluid Dynamics (CFD) is frequently used to predict complex flow phenomena and assist in engine design and optimization. The scavenge process within a 2-stroke engine is key to engine performance especially in high performance racing applications. In this paper, FLUENT CFD code is used to simulate the scavenging process within a 125cc single cylinder racing engine. A variety of different port designs are simulated and scavenge characteristics compared and contrasted. The predicted CFD results are compared with measured scavenge data obtained from the QUB single-cycle scavenge rig. These results show good agreement and provide valuable insight into the effect of port design features on the scavenging process.

Swirl/tumble are rotational flow inside the combustion chamber. Fluent Computational Fluid Dynamics (CFD) software has been successfully used to simulate engine swirl and tumble flow. Two mesh approaches are possible within Fluent software to calculate transient engine swirl and tumble. One approach uses hybrid mesh with remeshing, while the other approach uses hex/wedge mesh with layering. The hybrid method employs tetrahedral remeshing, and is easier to set up compared to hex/wedge method for which only layering is used. Being easier to use, the hybrid method raises some concerns about result accuracy due to higher numerical diffusion associated with tet elements compared to the corresponding hex/wedge elements used for layering approach. This paper examines the two mesh approaches in terms of result accuracy for two engines, one SI and one diesel. The results are compared with PIV data for the SI engine.

Modern pulse-turbocharged systems produce a turbine operating environment that is dominated by unsteady flow. Effective utilization of the unsteady exhaust gas energy content at the turbine inlet is critical to achieving optimum system efficiency. This work presents predictions for turbocharger unsteady performance from a model based on the Euler equations with source terms (EEST). This approach allows the time-accurate performance of the turbine to be determined, allowing comparisons of actual energy utilization and that estimated from steady flow performance maps.