H2 Norm Based Network Volatility Measures

in The proceedings of the American Control Conference (2014)

Motivated by applications in biology and economics, we propose new volatility measures based on the H2 system norm for linear networks stimulated by independent or correlated noise. We identify critical ... [more ▼]

Motivated by applications in biology and economics, we propose new volatility measures based on the H2 system norm for linear networks stimulated by independent or correlated noise. We identify critical links in a network, where relatively small improvements can lead to large reductions in network volatility measures. We also examine volatility measures of individual nodes and their dependence on the topological position in the network. Finally, we investigate the dependence of the volatility on different network interconnections, weights of the edges and other network properties. Hence in an intuitive and efficient way, we can identify critical links, nodes and interconnections in network which can shed light in the network design to make it more robust. [less ▲]

Heterogeneous agent models in economics: a study of heterogeneous productivity of sectors

in Proceedings of the 2008 American Control Conference (2008)

Macroeconomic modeling is undergoing a change from the ground up. Previously models based on fully rational representative agents were constructed to give macroeconomics soli microeconomic foundations ... [more ▼]

Macroeconomic modeling is undergoing a change from the ground up. Previously models based on fully rational representative agents were constructed to give macroeconomics soli microeconomic foundations. However the representative agent models have been shown to be inconsistent with empirical evidence and a new method of approach has emerged, one based on heterogeneity of agents. Recently heterogenous models have been used to simulate expected outcomes but due to their complexity little analytic work has been done. In this paper a basic model of the macro economy, with heterogeneous sectors differentiated by productivity, and driven by a jump Markov process, is investigated and steady state solutions for a sector’s output variance are discovered. We adjust the model to include a gain term, to represent a sector’s reaction to its error signal, excess demand, and then linearize the transition rates and apply the fluctuation dissipation theorem to solve the model. [less ▲]

Hierarchical a posteriori error estimation of Bank-Weiser type in the FEniCS Project

in Computers and Mathematics with Applications (2023), 131

In the seminal paper of Bank and Weiser [Math. Comp., 44 (1985), pp.283-301] a new a posteriori estimator was introduced. This estimator requires the solution of a local Neumann problem on every cell of ... [more ▼]

In the seminal paper of Bank and Weiser [Math. Comp., 44 (1985), pp.283-301] a new a posteriori estimator was introduced. This estimator requires the solution of a local Neumann problem on every cell of the finite element mesh. Despite the promise of Bank-Weiser type estimators, namely locality, computational efficiency, and asymptotic sharpness, they have seen little use in practical computational problems. The focus of this contribution is to describe a novel implementation of hierarchical estimators of the Bank-Weiser type in a modern high-level finite element software with automatic code generation capabilities. We show how to use the estimator to drive (goal-oriented) adaptive mesh refinement and to mixed approximations of the nearly-incompressible elasticity problems. We provide comparisons with various other used estimators. An open-source implementation based on the FEniCS Project finite element software is provided as supplementary material. [less ▲]

High-Level Modular Autopilot Solution for Fast Prototyping of Unmanned Aerial Systems

in IEEE Access (2020)

A redundant fast prototyping autopilot solution for unmanned aerial systems has been developed and successfully tested outdoors. While its low-level backbone is executed in a Raspberry Pi R 3 + NAVIO2 R ... [more ▼]

A redundant fast prototyping autopilot solution for unmanned aerial systems has been developed and successfully tested outdoors. While its low-level backbone is executed in a Raspberry Pi R 3 + NAVIO2 R with a backup autopilot, the computational power of an Intel R NUC mini-computer is employed to implement complex functionalities directly in Simulink R , thus including in-flight debugging, tuning and monitoring. Altogether, the presented tool provides a flexible and user-friendly high-level environment with enhanced computational capabilities, which drastically reduces the prototyping timespans of complex algorithms -between 50% and 75%, according to our long and proven experience in aerial robotics-, while preventing incidents thanks to its redundant design with a human-in-the-loop pilot on the reliable PX4. Three typical outdoor cases are carried out for validation in real-life scenarios, all mounted in a DJI © F550 platform. Full integration results and telemetry for more than 50 hours of outdoor flight tests are provided. [less ▲]

A holding control strategy for diverging bus lines

Scientific Conference (2018, July 24)

We introduce a holding criterion for network configurations with lines that operate jointly along a common corridor and then individually diverge. The proposed holding decision rule accounts for all ... [more ▼]

We introduce a holding criterion for network configurations with lines that operate jointly along a common corridor and then individually diverge. The proposed holding decision rule accounts for all different passengers groups in the overlapping segment and takes care of the transition to individual line operation. The holding rule is evaluated using simulation for different demand levels and segmentations and compared with other control schemes for a real-world network. Results show that gains in overall network performance as well as for specific passenger groups can be achieved under specific demand distributions. [less ▲]

How companies learn from design flaws: results from an empirical study of the german manufacturing industry

in Proceedings of the 15th International Conference on Engineering Design (2005)

Design flaws often become apparent at a time when the product is already in use and its development process, which in many cases includes extensive testing of parts, components and prototypes, is ... [more ▼]

Design flaws often become apparent at a time when the product is already in use and its development process, which in many cases includes extensive testing of parts, components and prototypes, is considered complete. Such flaws may reach from poor ergonomics to the total failure of the product. Often, especially when user safety is at risk, design flaws are so severe that companies are forced to announce a product callback. Petroski suggests that many (if not most) products, which we are familiar with today, have a long history of previously flawed designs [3]. This implies that designers did indeed learn from design flaws in both senses of the word “learn”: discovering the flaw and utilizing the knowledge gained about it to find a solution. As far as discovering a design flaw is concerned, it can be assumed that the feedback from those who interact with the physical products in practice – the individuals who maintain, repair, recycle but essentially use the products – plays an important role. In their previous work, the authors pointed out hat this feedback information could not only be vital for identifying potential product hazards but helps designers to review the effects of their design measures and therefore to improve their products from generation to generation [4]. In order to obtain a better understanding of how designers learn from design flaws, a mail survey was conducted that aimed at investigating company-, process- and product-related factors of this phenomenon and to answer (among others) the following research questions: • To what extent are design flaws of a company’s (or a competitor’s) product a driving force in the development of new products? • How do the designers of a company become aware of design flaws of their products? • How successful are companies in correcting design flaws? • How do successful and unsuccessful companies differ in terms of size, activity profile of their designers and characteristics of their products? • What are possible factors that influence the success in correcting a design fault? [less ▲]

HPC Multi-physics Biomass Furnace simulations as a Service

Scientific Conference (2020, November)

A hybrid MGA-MSGD ANN training approach for approximate solution of linear elliptic PDEs

E-print/Working paper (2020)

We introduce a hybrid "Modified Genetic Algorithm-Multilevel Stochastic Gradient Descent" (MGA-MSGD) training algorithm that considerably improves accuracy and efficiency of solving 3D mechanical problems ... [more ▼]

We introduce a hybrid "Modified Genetic Algorithm-Multilevel Stochastic Gradient Descent" (MGA-MSGD) training algorithm that considerably improves accuracy and efficiency of solving 3D mechanical problems described, in strong-form, by PDEs via ANNs (Artificial Neural Networks). This presented approach allows the selection of a number of locations of interest at which the state variables are expected to fulfil the governing equations associated with a physical problem. Unlike classical PDE approximation methods such as finite differences or the finite element method, there is no need to establish and reconstruct the physical field quantity throughout the computational domain in order to predict the mechanical response at specific locations of interest. The basic idea of MGA-MSGD is the manipulation of the learnable parameters’ components responsible for the error explosion so that we can train the network with relatively larger learning rates which avoids trapping in local minima. The proposed training approach is less sensitive to the learning rate value, training points density and distribution, and the random initial parameters. The distance function to minimise is where we introduce the PDEs including any physical laws and conditions (so-called, Physics Informed ANN). The Genetic algorithm is modified to be suitable for this type of ANN in which a Coarse-level Stochastic Gradient Descent (CSGD) is exploited to make the decision of the offspring qualification. Employing the presented approach, a considerable improvement in both accuracy and efficiency, compared with standard training algorithms such classical SGD and Adam optimiser, is observed. The local displacement accuracy is studied and ensured by introducing the results of Finite Element Method (FEM) at sufficiently fine mesh as the reference displacements. A slightly more complex problem is solved ensuring the feasibility of the methodology [less ▲]

A hyper-reduction method using adaptivity to cut the assembly costs of reduced order models

in Computer Methods in Applied Mechanics and Engineering (2021), 380

At every iteration or timestep of the online phase of some reduced-order modelling schemes, large linear systems must be assembled and then projected onto a reduced order basis of small dimension. The ... [more ▼]

At every iteration or timestep of the online phase of some reduced-order modelling schemes, large linear systems must be assembled and then projected onto a reduced order basis of small dimension. The projected small linear systems are cheap to solve, but assembly and projection are now the dominant computational cost. In this paper we introduce a new hyper-reduction strategy called reduced assembly (RA) that drastically cuts these costs. RA consists of a triangulation adaptation algorithm that uses a local error indicator to con- struct a reduced assembly triangulation specially suited to the reduced order basis. Crucially, this reduced assembly triangulation has fewer cells than the original one, resulting in lower assembly and projection costs. We demonstrate the efficacy of RA on a Galerkin-POD type reduced order model (RAPOD). We show performance increases of up to five times over the baseline Galerkin-POD method on a non-linear reaction-diffusion problem solved with a semi-implicit time-stepping scheme and up to seven times for a 3D hyperelasticity problem solved with a continuation Newton-Raphson algorithm. The examples are implemented in the DOLFIN finite element solver using PETSc and SLEPc for linear algebra. Full code and data files to produce the results in this paper are provided as supplementary material. [less ▲]

"I Personally Relate It to the Traffic Light": A User Study on Security & Privacy Indicators in a Secure Email System Committed to Privacy by Default

in Proceedings of the 36th Annual ACM Symposium on Applied Computing (2021)

Improving the usability and adoption of secure (i.e. end-to-end encrypted) email systems has been a notorious challenge for over two decades. One of the open questions concerns the amount and format of ... [more ▼]

Improving the usability and adoption of secure (i.e. end-to-end encrypted) email systems has been a notorious challenge for over two decades. One of the open questions concerns the amount and format of information that should be communicated to users to inform them of the security and privacy properties with respect to different messages or correspondents. Contributing to the ongoing discussion on the usability and effectiveness of security and privacy indicators, particularly in the context of systems targeting non-expert users, this paper sheds light on users' evaluation of traffic light-inspired indicators, as a metaphor to represent different privacy states and guarantees, provided by a new system for email end-to-end encryption called p≡p. Using a mixed-methods approach, based on input gathered from 150 participants in three online studies, we highlight the pros and cons of the traffic light semantic in p≡p's context and beyond, and discuss the potential implications on the perceived security and use of such systems. [less ▲]