Engineering Design Optimization

Traditional surrogate-based design optimization (SBDO) is hitting a wall, especially with high-dimensional, complex designs. In this new paper, Dr. Namwoo Kang presents a next-gen framework using generative AI, integrating three key models: - Generative model (design synthesis) - Predictive model (performance estimation) - Optimization model (iterative or generative) Rather than optimizing directly in a high-dimensional design space (x), the workflow introduces a low-dimensional latent space (z) learned via generative models. ➡️ z → x → y z = latent variables x = CAD geometry y = performance (drag, stress, etc.) This means we’re no longer hand-coding design parameters or doing trial-and-error with simplified surrogate models. 🧠 Why this matters: - Parametric modeling is no longer a bottleneck - Complex shapes are learned directly from CAD - Dynamic and multimodal performance data (1D, 2D, 3D) can be used - Near real-time optimization is possible #AI #GenerativeDesign #CAE #DesignOptimization

They introduce a pioneering computational approach to enhance the morphological design of #robots. Their framework accepts a parameterized robot design and a motion plan, encompassing trajectories for end-effectors and, optionally, the body. The algorithm optimizes design parameters, including link lengths and actuator placements, while concurrently adjusting motion parameters like joint trajectories, actuator inputs, and contact forces. The key insight lies in establishing the intricate relationship between design and motion parameters through sensitivity analysis, assuming the robot's movements are modeled as spatiotemporal solutions to an optimal control problem. This connection between form and function enables automatic optimization of the robot design based on specifications expressed as a function of actuator forces or trajectories. The model undergoes evaluation by computationally optimizing four simulated robots utilizing various actuators. They further validate the framework by optimizing the design of two small quadruped robots and assessing their performances through hardware implementations. #authors: Sehoon Ha, Stelian Coros, Alex Alspach, James Bern, Joohyung Kim, Katsu Yamane #research #papers: https://lnkd.in/dde8udvV, https://lnkd.in/dgi8KNWE #robotics #projects #technology 

𝗔𝗲𝗿𝗼𝗱𝘆𝗻𝗮𝗺𝗶𝗰 𝗢𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗦𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗮𝗹 𝗜𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻 𝗶𝗻 𝗕𝘂𝗿𝗷 𝗞𝗵𝗮𝗹𝗶𝗳𝗮, 𝗥𝗲𝗱𝗲𝗳𝗶𝗻𝗶𝗻𝗴 𝗦𝗸𝘆𝘀𝗰𝗿𝗮𝗽𝗲𝗿 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝘁𝗵𝗿𝗼𝘂𝗴𝗵 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗱 𝗟𝗼𝗮𝗱 𝗗𝗶𝘀𝘁𝗿𝗶𝗯𝘂𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗪𝗶𝗻𝗱 𝗥𝗲𝘀𝗶𝘀𝘁𝗮𝗻𝗰𝗲 𝗦𝘁𝗿𝗮𝘁𝗲𝗴𝗶𝗲𝘀 !! 🏙️🗼🛠️ ✨𝗔𝗲𝗿𝗼𝗱𝘆𝗻𝗮𝗺𝗶𝗰 𝗗𝗲𝘀𝗶𝗴𝗻 𝗮𝗻𝗱 𝗪𝗶𝗻𝗱 𝗥𝗲𝘀𝗶𝘀𝘁𝗮𝗻𝗰𝗲.🍃 ✓ Y-shaped plan disrupts wind flow. ✓ Ensures stability under extreme loads. ✓ Curved edges create irregular geometry. ✓ Tapered profile prevents vortex formation. ✓ "Confusing the wind" enhances performance. ✓ Aerodynamic design reduces wind oscillations. ✓ Advanced structural engineering mitigates forces. ✨𝗙𝗹𝗼𝗼𝗿 𝗣𝗹𝗮𝘁𝗲 𝗖𝗼𝗻𝗳𝗶𝗴𝘂𝗿𝗮𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗟𝗼𝗮𝗱 𝗗𝗶𝘀𝘁𝗿𝗶𝗯𝘂𝘁𝗶𝗼𝗻.🏗️ ✓ 27 staggered floor plates used. ✓ Unique configurations per floor. ✓ Enhances lateral force resistance. ✓ Improves overall structural integrity. ✓ Optimizes structural load distribution. ✓ Withstands environmental stresses effectively. ✨𝗦𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗮𝗹 𝗦𝘆𝘀𝘁𝗲𝗺 𝗮𝗻𝗱 𝗠𝗮𝘁𝗲𝗿𝗶𝗮𝗹 𝗜𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻.♻️ ✓ Ensures durability at heights. ✓ Withstands high wind pressures. ✓ Lateral load resistance optimized. ✓ Efficient gravitational load distribution. ✓ Aluminum and stainless steel cladding. ✓ Buttressed mega-columns transfer loads. ✓ Endures extreme temperature fluctuations. ✓ High-performance reinforced concrete core. ✨𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗱 𝗠𝗼𝗱𝗲𝗹𝗶𝗻𝗴 𝗮𝗻𝗱 𝗘𝗻𝘃𝗶𝗿𝗼𝗻𝗺𝗲𝗻𝘁𝗮𝗹 𝗔𝗱𝗮𝗽𝘁𝗮𝘁𝗶𝗼𝗻.🌐 ✓ Wind tunnel testing refined form. ✓ Maintains iconic tower silhouette. ✓ Pinnacle of skyscraper engineering. ✓ Innovative design integration achieved. ✓ Endures harsh environmental conditions. ✓ Cutting-edge materials ensure durability. ✓ Advanced computational modeling used. ✓ Rigorous testing exemplifies engineering.

🎯 How can we use a low-fidelity optimization model to achieve similar performance to a high-fidelity model? Many decision-making algorithms can be viewed as tuning a low-fidelity model within a high-fidelity simulator to achieve improved performance. A great example comes from Cost Function Approximations (CFAs) by Warren Powell. CFAs embed tunable parameters, such as cost coefficients, into a simplified, deterministic model. These parameters are then refined by optimizing performance in a high-fidelity stochastic simulator, either via derivative-free or gradient-based methods. A similar philosophy appears in optimal control, where controllers are tuned using simulation optimization. ⚙️ Inspired by this paradigm, my student Asha Ramanujam recently developed the PAMSO algorithm. PAMSO—Parametric Autotuning for Multi-Timescale Optimization—tackles complex systems that operate across multiple timescales: High-level decision layer: makes strategic decisions (e.g., planning, design). Low-level decision layer: takes high-level inputs, makes detailed operating decisions (e.g., scheduling), applies detailed constraints and uncertainties, and computes the true objective. However, one-way top-down communication between layers often results in infeasibility or poor solutions due to mismatches between the high-level and the detailed low-level operating models. 💡 PAMSO augments the high-level model with tunable parameters that serve as a proxy for the complex physics and uncertainties embedded in the low-level model. Instead of attempting to jointly solve both levels, we fix the hierarchical structure: the high-level layer makes planning or design decisions, and then passes them down to the low-level scheduling or operational layer, which acts as a high-fidelity simulator. We treat this top-down hierarchy as a black box: The inputs are the tunable parameters embedded in the high-level model. The output is the overall objective value after the low-level simulator evaluates feasibility and performance. By optimizing these parameters using derivative-free methods, PAMSO is able to steer the entire system toward high-quality, feasible solutions. 🚀 Bonus: Transfer Learning! If these parameters are designed to be problem-size invariant, they can be tuned on smaller problem instances and transferred to solve larger-scale problems with minimal extra effort. ⚙️ Case studies demonstrate PAMSO’s scalability and effectiveness in generating good, feasible solutions: ✅ A MINLP model for integrated design and scheduling in a resource-task network with ~67,000 variables ✅ A massive MILP model for integrated planning and scheduling of electrified chemical plants and renewable energy with ~26 million variables Even solving the LP relaxation of these problems is beyond memory limits, and their structure is not easily decomposable for optimization techniques. https://lnkd.in/gDfcvDaZ

We are excited to share our latest research on a novel topology optimization method that integrates deformable meshes with the traditional SIMP (Solid Isotropic Material with Penalization) framework. 📌 Kyusoon Jung & Do-Nyun Kim, Density-based topology optimization using a deformable mesh, Computers and Structures, 316: 107879 (2025) https://lnkd.in/gt8raDBy 📌 Free access until August 20, 2025 https://lnkd.in/gNvw-iAA 🧠 Key Contributions 1. High-resolution results with coarse meshes: Nodal displacements are introduced as design variables alongside element densities, expanding the design space at high resolution without increasing mesh density. 2. No post-processing required: Unlike conventional SIMP methods that require iso-contour extraction and CAD reconstruction, our approach produces binary density distributions with smooth, fabrication-ready boundaries directly. 3. Checkerboard-free binarization: A novel patch-based density adjustment strategy enables clean structural patterns, eliminating intermediate densities and artifacts. 4. Enhanced computational efficiency: Up to 80% reduction in optimization time can be achieved while preserving or improving compliance compared to high-resolution SIMP. 5. Design flexibility: Exploration of multiple optimal configurations with comparable mechanical performance is possible by controlling a single parameter for the density threshold. We believe that the proposed method is a promising candidate for efficient structural design workflows, especially in applications where high-resolution and smooth geometry are critical.

🔬 Engineering design synthesis is moving from manual iteration to automated, data-driven approaches. MIT researchers map out how deep generative models are enabling this shift, with important technical implications for how we develop products in the reference paper below. The paper provides a systematic analysis across: 🎯 Design Problems: • Topology optimization • Materials & microstructure design • 2D/3D shape synthesis • Multi-component product design 💾 Data Representations: • Voxels & point clouds for 3D • Images for 2D designs • Parametric specs for manufacturing • Graphs for component relationships 🧮 Model Architectures: • GANs with various conditioning approaches • VAEs for latent space exploration • RL for sequential design decisions • Integration with physics-based simulation ⚖️ Loss Functions: • Performance metrics from simulation • Manufacturability constraints • Style transfer for design aesthetics • Multi-objective optimization 📊 Key Datasets: • UIUC airfoil database • ShapeNet/ModelNet for 3D shapes • BIKED bicycle design dataset • Material microstructure collections 📝 Reference: "Deep Generative Models in Engineering Design: A Review" by Regenwetter et al. https://lnkd.in/g_mMR-8y S2 Labs #EngineeringDesign #MachineLearning #TechnicalResearch

Next-Gen Lightweight Aircrafts ✈ The quest for ever-lighter, high aspect ratio wings (AR = b^2/S) presents a classic engineering dilemma: reduce weight for improved fuel efficiency and range, but risk compromising structural integrity. Traditional methods often struggle with:  a. Non-linear material behavior: Current optimization techniques frequently rely on simplified material models, potentially overlooking crucial real-world complexities like composite anisotropy or damage tolerance. b. Computational cost: Large-scale wing designs with intricate geometries can push traditional optimization algorithms to their limits, hindering design exploration and cycle times.    Here at BosonQ Psi (BQP), we're exploring the potential of advanced optimization methodologies to address these limitations:  We are exploring the potential of quantum algorithms. We're actively exploring the integration of quantum algorithms with our high-fidelity and ROM techniques. Our Novel Optimization approach offers the potential to revolutionize optimization on today's HPC This hybrid approach promises to:  1. Tackling highly complex design spaces  2. Uncover truly optimal designs by identifying material distributions and wing configurations that may be beyond the reach of traditional classical optimization methods.  3. Accelerate design cycles leading to faster innovation and time-to-market for next-generation aircraft. In the coming weeks, we will be showcasing the possibilities of our Quantum Inspired techniques to solve sticky problems that have plagued us for long. What do you think? Let me know in the comments #aerospace #optimization #quantumcomputing #simulation

Geometrically exact beam models are particularly useful for highly flexible structures like wind turbine blades and many aircraft applications. However, using them efficiently in design optimization applications poses some challenges. Taylor McDonnell just published a new model that reformulates the equations in a way that is more efficient and robust, smooths out singularities to allow for effective design iterations, and efficiently computes derivatives, via adjoints, for steady and unsteady problems. Journal paper: https://lnkd.in/gJJ2evZv Open-source code: https://lnkd.in/g65-hUfx Taylor was a rock star in the lab, developing all sorts of well documented code packages, and making advances in optimization and aeroelasticity. Excited to see him continue his career at AeroVironment #optimization, #adjoint, #aeroelasticity #design #windenergy #aeronautics