Reliability-Based Robust Design Optimization of Lithium-Ion Battery Cells for Maximizing the Energy Density by Increasing Reliability and Robustness

Lithium-ion batteries (LIBs) are increasingly employed in electric vehicles (EVs) owing to their advantages, such as low weight, and high energy and power densities. However, the uncertainty encountered in the manufacturing of LIB cells increases the failure rate and causes cell-to-cell variations, thereby degrading the battery capacity and lifetime. In this study, the reliability and robustness of LIB cells were improved using the design of experiments (DOE), and the reliability-based robust design optimization (RBRDO) approaches. First, design factors sensitive to the energy density and power density were selected as design variables through sensitivity analysis using the DOE. RBRDO was performed to maximize the energy density while reducing the failure rate and cell-to-cell variations. To verify the superiority of the reliability and robustness offered by RBRDO, the obtained results were compared with those from conventional deterministic design optimization (DDO), and reliability-based design optimization (RBDO). RBRDO increased the mean of the energy density by 33.5% compared to the initial value and reduced the failure rate by 98.9%, due to improved reliability, compared to DDO. Moreover, RBRDO reduced the standard deviation in the energy density (i.e., cell-to-cell variations) by 30.0% due to the improved robustness compared to RBDO.

ERGO: a new Robust Design Optimization Technique combining Multi-Objective Bayesian Optimization with Analytical Uncertainty Quantification

In this work, robust design optimization (RDO) is treated, motivated by the increasing desire to account for variability the design phase. The problem is formulated in a multi-objective setting with the objective of simultaneously minimizing the mean of the objective and its variance due to variability of design variables and/or parameters. This allows the designer to choose its robustness level without the need to repeat the optimization as typically encountered when formulated as a single objective. To account for the computational cost that is often encountered in RDO problems, the problem is fitted in a Bayesian optimization framework. The use of surrogate modeling techniques to efficiently solve problems under uncertainty has effectively found its way in the optimization community leading to surrogate-assisted optimization-under-uncertainty schemes. The surrogates are often considered cheap-to-sample black-boxes and are sampled to obtain the desired quantities of interest. However, since the analytical formulation of the surrogates is known, an analytical treatment of the problem is available. To obtain the quantities of interest without sampling an analytical uncertainty propagation through the surrogate is presented. The multi-objective Bayesian optimization framework and the analytical uncertainty quantification are linked together through the formulation of the robust expected improvement (REI), obtaining the novel efficient robust global optimization (ERGO) scheme. The method is tested on a series of test cases to examine its behavior for varying difficulties and validated on an aerodynamic test function which proves the effectiveness of the novel scheme.

Characterization of manufacturing uncertainties with applications to uncertainty quantification and robust design optimization

Methodologies to quantify the impact of manufacturing uncertainties in 3D CFD based design strategies have become increasingly available over the past years as well as optimization under uncertainties, aiming at reducing the systems sensitivity to manufacturing uncertainties. This type of non-deterministic simulation depends however strongly on a correct characterization of the manufacturing variability. Experimental data to characterize this variability is not always available or in many cases cannot be sampled in sufficiently high numbers. Principal Component Analysis (PCA) is applied to the sampled geometries and the influence of tolerances classes, sample size and number of retained deformation modes are discussed. It is shown that the geometrical reconstruction accuracy of the deformation modes and reconstruction accuracy of the CFD predictions are not linearly related, which has important implications on the total geometrical variance that needs to be retained. In a second application the characterization of manufacturing uncertainties to a marine propeller is discussed. It is shown that uncertainty quantification and robust design optimization of the marine propeller can successfully be performed on the basis of the derived uncertainties. This leads to a propeller shape that is less sensitive to the manufacturing variability and therefore to a more robust design.

Robust Design Optimization of an Industrial 1.5 Stage Axial Compressor Under Operational and Geometrical Uncertainties

A robust design optimization of a 1.5 stage axial compressor with secondary flows from Safran Aero Boosters is investigated. A total of 9 simultaneous operational and geometrical uncertainties are propagated for the nominal design point as well as for two off-design points, close to stall and choke conditions respectively. These uncertainties, including mass flow rates of the secondary flows, tip gap size of the rotor and highly correlated profiles on the inlet condition, are propagated by the Non-Intrusive Probabilistic Collocation method. In order to understand the effects of the uncertainties on the performances and to minimize the computational cost of the robust optimization, a preliminary uncertainty quantification (UQ) study of the original design is performed to identify and rule out less influential uncertainties. Contrary to what was expected, the imposed geometrical uncertainties on the tip gap are identified to have the relatively smallest influence on the performances by means of scaled sensitivity derivatives. The global objective of the robust design optimization is to minimize the standard deviations of the main compressor performances at all three operating points and to preserve the mean values of these performances. Because the objective functions are standard deviations, this study is only possible in a robust optimization setting, by propagating the simultaneous operational and geometrical uncertainties. A total of 9 stochastic objectives and 15 stochastic constraints are taken into account. The best optimal design preserves the mean performances of the compressor, while the standard deviations are minimized compared with the original design, ensuring a more robust operation. This effect is very pronounced in the off-design points.