2026-10-19T01:06:31Zhttps://ebyu-rdlopen.rawdatalibrary.net/oai/request

oai:ebyu-rdlopen.rawdatalibrary.net:ebyu/104952026-10-18T20:33:47Zopenairedrivercom_ebyu_10000col_ebyu_10001

GMDH and NSGA-III based design optimization of bio-inspired cooling systems for prismatic LFP batteriesKalkan, OrhanBio-inspiredCFDCold plateGroup method of data handlingLi-ion batteryOptimizationBiomimeticsCooling systemsDropsGenetic algorithmsMultiobjective optimizationOptimal systemsScreeningThermal management (electronics)Cold platesDesign optimizationHigh-fidelityIon batteriesMulti-objectives optimizationNon-dominated sorting genetic algorithmsOptimal designOptimisationsLithium-ion batteriesPressure dropThis study presents a high-fidelity multi-objective optimization of bio-inspired cold plates (Ginkgo, Tilia, Tulip) for prismatic LFP battery thermal management. The coupled effects of mass flow rate (0.5–20 g s−1), inlet temperature (5–30°C), and glycol volume fraction (0–60%) on maximum battery temperature (Tmax), temperature non-uniformity (ΔT), and pressure drop (ΔP) are systematically examined under a 3C discharge rate. Numerical reliability is confirmed via GCI, with experimental validation showing a maximum error of 9.83%. To overcome the limitations of conventional regression, the Group Method of Data Handling (GMDH) is utilized to develop precise surrogate models, which are subsequently coupled with the Non-dominated Sorting Genetic Algorithm III (NSGA-III) for multi-objective optimization. The optimization identifies a robust, feasible Pareto front that strictly adheres to the safety constraints of 15°C ≤ Tmax≤35°C and ΔT≤5°C. Quantitative results reveal 10 candidate optimal designs, with Design 1 offering the highest efficiency at a minimum pressure drop of 1058.96 Pa, and Design 10 providing maximum thermal safety with a Tmax of 32.18°C. Furthermore, integrating freezing-point constraints enables climate-adaptive selection; designs with glycol fractions up to 58.43% ensure system integrity down to −49°C, albeit with a 58% increase in hydraulic penalty compared to the base optimal design. © 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.Elsevier B.V.2026Articlehttps://rdlopen.rawdatalibrary.net/r/ebyu/10495https://doi.org/10.1016/j.jpowsour.2026.2412560378-7753enJournal of Power Sourcesinfo:eu-repo/semantics/openAccessinfo:eu-repo/semantics/articlehttps://ebyu-rdlopen.rawdatalibrary.net/tr/items/ebyu/10495