Litcius/Paper detail

Design and optimization of lithium-ion battery protector with auxetic honeycomb for in-plane impact using machine learning method

Michael Alfred Stephenson Biharta, Sigit Puji Santosa, Djarot Widagdo

2023Frontiers in Energy Research18 citationsDOIOpen Access PDF

Abstract

The lithium-ion battery is becoming a very important energy source for vehicles designated as electric vehicles. This relatively new energy source is much more efficient and cleaner than conventional fossil fuel. However, lithium-ion batteries have a high risk of fire during a crash, where the large deformation on the battery during the crash may cause thermal runaway. This research explores that idea by studying the design and optimization of sandwich-based auxetic honeycomb structures to protect the pouch battery cells for the battery pack system of electric vehicles undergoing axial impact load using machine learning methods. The optimization was done using Artificial Neural Network (ANN), and Non-Dominated Sorting Genetic Algorithm Type II (NSGA-II) combined with Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). Artificial Neural Network predicted the sandwich structure’s specific energy absorption (SEA) and the maximum battery stress during deformation. NSGA-II combined with TOPSIS optimized the design using both of the predictors. Both creations of the training data and validation were done using the non-linear finite element method. The optimized design has a geometric shape of Double-U, a length of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="m1"><mml:mrow><mml:mn>6</mml:mn><mml:mtext> </mml:mtext><mml:mi>m</mml:mi><mml:mi>m</mml:mi></mml:mrow></mml:math> , a width of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="m2"><mml:mrow><mml:mn>4.2</mml:mn><mml:mtext> </mml:mtext><mml:mi>m</mml:mi><mml:mi>m</mml:mi></mml:mrow></mml:math> , cross section’s thickness of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="m3"><mml:mrow><mml:mn>0.6</mml:mn><mml:mtext> </mml:mtext><mml:mi>m</mml:mi><mml:mi>m</mml:mi></mml:mrow></mml:math> , and consists of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="m4"><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:math> layer. The optimum design has a specific energy absorption of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="m5"><mml:mrow><mml:mn>47,997.84</mml:mn><mml:mtext> </mml:mtext><mml:mi>J</mml:mi></mml:mrow></mml:math> and can maintain the battery’s von Mises stress to a maximum of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="m6"><mml:mrow><mml:mn>43.16</mml:mn><mml:mtext> </mml:mtext><mml:mi>M</mml:mi><mml:mi>P</mml:mi><mml:mi>a</mml:mi></mml:mrow></mml:math> , well below the designated battery’s von Mises stress limit of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="m7"><mml:mrow><mml:mn>67.97</mml:mn><mml:mtext> </mml:mtext><mml:mi>M</mml:mi><mml:mi>P</mml:mi><mml:mi>a</mml:mi></mml:mrow></mml:math> .

Topics & Concepts

Battery (electricity)Computer scienceMaterials scienceArtificial neural networkAlgorithmBattery packArtificial intelligenceMachine learningMechanical engineeringPhysicsThermodynamicsEngineeringPower (physics)Cellular and Composite StructuresAdvanced Battery Technologies ResearchAutomotive and Human Injury Biomechanics