Autodesk (Canada)

Abstract As engineering systems grow more intricate and technological progress accelerates, traditional sources of design knowledge, such as historical data and expert intuition, struggle to keep pace with the complexity and the speed of knowledge generation. To address this challenge, additional sources of knowledge are necessary, particularly for designing unprecedented engineering systems lacking any design heritage. One promising approach involves analyzing optimized designs to extract valuable insights, enabling designers to break away from incremental improvements over existing designs. This article explores the extraction of design information from optimized designs in power flow systems using various classification machine learning methods, empowering designers to make informed decisions in future design endeavors. This design information can also serve as a foundation for synthesizing engineering system configurations that are more complex than those previously encountered. This approach offers several advantages over traditional methods, including its applicability in the absence of design heritage and its ability to provide normative guidance for system design. This article focuses on power flow systems that can be modeled as graphs with a tree structure, with the case study being multisplit fluid-based thermal management systems. The article presents four case studies demonstrating the effectiveness of using information from optimized designs to enhance the design of complex thermal management systems, in both human-directed and automated design processes. The results show that information extraction significantly improves the design process, with less than 1 percent error in approximating the true optimal configuration. This approach eliminates the need for solving complex control problems, leading to reduced computation costs.

Abstract This study introduces a graph-based framework developed for representing various aspects of optimal thermal management system design, with the aim of rapidly and efficiently identifying optimal design candidates. Initially, the graph-based framework is utilized to generate diverse thermal management system architectures. The dynamics of these system architectures are modeled under various loading conditions, and an open-loop optimal controller is employed to determine each system’s optimal performance. These modeled cases constitute the dataset, with the corresponding optimal performance values serving as the labels for the data. In the subsequent step, a Graph Neural Network (GNN) model is trained on 11,134 (30%) of the labeled data to predict the systems’ performance, effectively addressing a regression problem. Utilizing this trained model, we estimate the performance values for the remaining 26,195 (70%) of the data, which serves as the test set. The reason for larger number of test points was to ensure that the model is capable of predicting performance across all the diversity of inputs. In the third step, the predicted performance values are employed to rank the test data, facilitating prioritized evaluation of the design scenarios. Specifically, a small subset of the test data with the highest estimated ranks undergoes evaluation via an open-loop optimal control solver. This targeted approach concentrates on evaluating higher-ranked designs identified by the GNN, replacing the exhaustive search (enumeration-based) of all design cases. The results demonstrate a significant average reduction of over 92% in the number of system dynamic modeling and optimal control analyses required to identify optimal design scenarios.

Although the Futures Triangle, developed by Sohail Inayatullah in the 1990s, has become a staple mental model for the futures/foresight field, it has not been well explored in the academic literature in the context of other futures tools. Here, the authors use a variation of the Futures Triangle as a springboard for foresight work in groups of any size, with a particular focus on facilitating the mindset shift that precedes organizational transformation. The authors present directions and an 80-stakeholder use case of the Co-Creative Futures Triangle, an effective tool to help diverse members of a system to co-ideate unique, positive futures created from a pool of shared values and current realities.

We present a nonintrusive approach to robust structural topology optimization. Specifically, we consider optimization of mean- and variance-based robustness metrics of a linear functional output associated with the linear elasticity equation in the presence of probabilistic uncertainties in the loading and material properties. To provide an efficient approximation of higher-dimensional problems, we approximate the solution to the governing stochastic partial differential equations using the anchored ANOVA Petrov–Galerkin projection scheme. We then develop a nonintrusive quadrature-based formulation to evaluate the robustness metric and the associated shape derivative. The formulation is nonintrusive in the sense that it works with any level-set-based topology optimization code that can provide deterministic displacements, outputs, and shape derivatives for selected stochastic parameter values. We demonstrate the effectiveness of the proposed approach on various problems under loading and material uncertainties.

Reverse Engineering a CAD shape from other representations is an important geometric processing step for many downstream applications. In this work, we introduce a novel neural network architecture to solve this challenging task and approximate a smoothed signed distance function with an editable, constrained, prismatic CAD model. During training, our method reconstructs the input geometry in the voxel space by decomposing the shape into a series of 2D profile images and 1D envelope functions. These can then be recombined in a differentiable way allowing a geometric loss function to be defined. During inference, we obtain the CAD data by first searching a database of 2D constrained sketches to find curves which approximate the profile images, then extrude them and use Boolean operations to build the final CAD model. Our method approximates the target shape more closely than other methods and outputs highly editable constrained parametric sketches which are compatible with existing CAD software.

Parametric computer-aided design (CAD) is a standard paradigm used to design manufactured objects, where a 3D shape is represented as a program supported by the CAD software. Despite the pervasiveness of parametric CAD and a growing interest from the research community, currently there does not exist a dataset of realistic CAD models in a concise programmatic form. In this paper we present the Fusion 360 Gallery, consisting of a simple language with just the sketch and extrude modeling operations, and a dataset of 8,625 human design sequences expressed in this language. We also present an interactive environment called the Fusion 360 Gym, which exposes the sequential construction of a CAD program as a Markov decision process, making it amendable to machine learning approaches. As a use case for our dataset and environment, we define the CAD reconstruction task of recovering a CAD program from a target geometry. We report results of applying state-of-the-art methods of program synthesis with neurally guided search on this task.