As a part of the COMPASS force field development, a number of small inorganic molecules were parametrized for condensed-phase applications. Using a simple valence model coupled with Coulomb energy and Lennard-Jones 9-6 functional terms, the parameters were optimized to yield accurate prediction of structural, vibrational, and thermophysical properties for these molecules. Extended validation on liquid nitrogen (N 2 ) and carbon dioxide (CO 2 ) in normal and supercritical conditions demonstrates that the present force field is capable of predicting various thermophysical properties in a very broad range of experimental conditions.

Building Information Modeling (BIM) is envisioned as an indispensable opportunity in the architecture, engineering, and construction (AEC) industries as a revolutionary technology and process. Smart construction relies on BIM for manipulating information flow, data flow, and management flow. Currently, BIM model has been explored mainly for information construction and utilization, but rare works pay efforts to information security, e.g., critical model audit and sensitive model exposure. Moreover, few BIM systems are proposed to chase after upcoming computing paradigms, such as mobile cloud computing, big data, blockchain, and Internet of Things. In this paper, we make the first attempt to propose a novel BIM system model called bcBIM to tackle information security in mobile cloud architectures. More specifically, bcBIM is proposed to facilitate BIM data audit for historical modifications by blockchain in mobile cloud with big data sharing. The proposed bcBIM model can guide the architecture design for further BIM information management system, especially for integrating BIM cloud as a service for further big data sharing. We propose a method of BIM data organization based on blockchains and discuss it based on private and public blockchain. It guarantees to trace, authenticate, and prevent tampering with BIM historical data. At the same time, it can generate a unified format to support future open sharing, data audit, and data provenance.

Direct prediction of material properties from microstructures through statistical models has shown to be a potential approach to accelerating computational material design with large design spaces. However, statistical modeling of highly nonlinear mappings defined on highdimensional microstructure spaces is known to be data-demanding. Thus, the added value of such predictive models diminishes in common cases where material samples (in forms of 2D or 3D microstructures) become costly to acquire either experimentally or computationally. To this end, we propose a generative machine learning model that creates an arbitrary amount of artificial material samples with negligible computation cost, when trained on only a limited amount of authentic samples. The key contribution of this work is the introduction of a morphology constraint to the training of the generative model, that enforces the resultant artificial material samples to have the same morphology distribution as the authentic ones. We show empirically that the proposed model creates artificial samples that better match with the authentic ones in material property distributions than those generated from a state-of-the-art Markov Random Field model, and thus is more effective at improving the prediction performance of a predictive structure-property model.

Crowdsourced evaluation is a promising method of evaluating engineering design attributes that require human input. The challenge is to correctly estimate scores using a massive and diverse crowd, particularly when only a small subset of evaluators has the expertise to give correct evaluations. Since averaging evaluations across all evaluators will result in an inaccurate crowd evaluation, this paper benchmarks a crowd consensus model that aims to identify experts such that their evaluations may be given more weight. Simulation results indicate this crowd consensus model outperforms averaging when it correctly identifies experts in the crowd, under the assumption that only experts have consistent evaluations. However, empirical results from a real human crowd indicate this assumption may not hold even on a simple engineering design evaluation task, as clusters of consistently wrong evaluators are shown to exist along with the cluster of experts. This suggests that both averaging evaluations and a crowd consensus model that relies only on evaluations may not be adequate for engineering design tasks, accordingly calling for further research into methods of finding experts within the crowd.

Several hybrid-electric vehicle architectures have been commercialized to serve different categories of vehicles and driving conditions. Such architectures can be optimally controlled by switching among driving modes, namely, the power distribution schemes in their planetary gear (PG) transmissions, in order to operate the vehicle in the most efficient regions of engine and motor maps. This paper proposes a systematic way to identify the optimal architecture for a given vehicle drive cycle, rather than parametrically optimizing one or more pre-selected architectures. An automatic generator of feasible driving modes for a given number of PGs is developed. For a powertrain consisting of one engine, two motors and two PGs, this generator results in 1116 modes. A heuristic search is then proposed to find a near-optimal pair of modes for a given driving cycle and vehicle specification. In a study this process identifies a dual-mode architecture with an 8% improvement in fuel economy compared to a commercially available architecture over a standard drive cycle.

In this paper we propose an indirect low-dimension design representation to enhance topology optimization capabilities. Established topology optimization methods, such as the Solid Isotropic Material with Penalization (SIMP) method, can solve large-scale topology optimization problems efficiently, but only for certain problem formulation types (e.g., those that are amenable to efficient sensitivity calculations). The aim of the study presented in this paper is to overcome some of these challenges by taking a complementary approach: achieving efficient solution via targeted design representation dimension reduction, enabling the tractable solution of a wider range of problems (e.g., those where sensitivities are expensive or unavailable). A new data-driven design representation is proposed that uses an augmented Variational Autoencoder (VAE) to encode 2D topologies into a lower-dimensional latent space, and to decode samples from this space back into 2D topologies. Optimization is then performed in the latent space as opposed to the image space. Established topology optimization methods are used here as a tool to generate a data set for training by changing problem conditions systematically. The data is generated using problem formulations that are solvable by SIMP, and are related to (but distinct from) the desired design problem. We further introduce augmentations to the VAE formulation to reduce unrealistic scattering of small material clusters during topology generation, while ensuring diversity of the generated topologies. We compare computational expense for solving a heat conduction design problem (with respect to the latent design variables) using different optimization algorithms. The new non-dominated points obtained via the VAE representation were found and compared with the known attainable set, indicating that use of this new design representation can simultaneously improve computational efficiency and solution quality.

Abstract-Network Function Virtualization (NFV) enables mobile operators to virtualize their network entities as Virtualized Network Functions (VNFs), offering fine-grained on-demand network capabilities. VNFs can be dynamically scale-in/out to meet the performance desire and other dynamic behaviors. However, designing the auto-scaling algorithm for desired characteristics with low operation cost and low latency, while considering the existing capacity of legacy network equipment, is not a trivial task. In this paper, we propose a VNF Dynamic Auto Scaling Algorithm (DASA) considering the tradeoff between performance and operation cost. We develop an analytical model to quantify the tradeoff and validate the analysis through extensive simulations. The results show that the DASA can significantly reduce operation cost given the latency upper-bound. Moreover, the models provide a quick way to evaluate the cost-performance tradeoff and system design without wide deployment, which can save cost and time.

We introduce a theory-driven mechanism for learning a neural network model that performs generative topology design in one shot given a problem setting, circumventing the conventional iterative process that computational design tasks usually entail. The proposed mechanism can lead to machines that quickly response to new design requirements based on its knowledge accumulated through past experiences of design generation. Achieving such a mechanism through supervised learning would require an impractically large amount of problem-solution pairs for training, due to the known limitation of deep neural networks in knowledge generalization. To this end, we introduce an interaction between a student (the neural network) and a teacher (the optimality conditions underlying topology optimization): The student learns from existing data and is tested on unseen problems. Deviation of the student's solutions from the optimality conditions is quantified, and used for choosing new data points to learn from. We call this learning mechanism "theory-driven", as it explicitly uses domain-specific theories to guide the learning, thus distinguishing itself from purely data-driven supervised learning. We show through a compliance minimization problem that the proposed learning mechanism leads to topology generation with near-optimal structural compliance, much improved from standard supervised learning under the same computational budget. example, the design of vehicle body-in-white is often done by experienced structure engineers, since topology optimization (TO) on full-scale crash simulation is not yet fast enough to respond to requests from higher-level design tasks, e.g., geometry design with style and aerodynamic considerations, and thus may slow down the entire design process 1 .Research exists in developing deep neural network models that learn to create structured solutions in a one-shot fashion, circumventing the need of iterations (e.g., in solving systems of equations [1], simulating dynamical systems [2], or searching for optimal solutions [3,4,5]). Learning of such models through data, however, is often criticized to have limited generalization capability, especially when highly nonlinear input-output relations or highdimensional output spaces exist [6,7,8]. In the context of TO, this means that the network may create structures with unreasonably poor physical properties when it responds to new problem settings. More concretely, consider a topology with a tiny crack in one of its trusses. This design would be far from optimal if the goal is to lower compliance, yet standard datadriven learning mechanisms do not prevent this from happening, i.e., they don't know that they don't know (physics).Our goal is to create a learning mechanism that knows what it does not know, and thus can self-improve in an effective way. Specifically, we are curious about how physicsbased knowledge, e.g., in the forms of dynamical models, theoretical bounds, and optimality conditions, can be directly injected into the learning of networks that ...