Application of scalar dissipation rate modelling to industrial burners in partially premixed regimes

The objective of this paper is to test various available turbulent burning velocity models on an experimental version of Siemens small scale combustor using the commercial CFD code. Failure of burning velocity model with different expressions for turbulent burning velocity is observed with an unphysical flame flashback into the swirler. Eddy Dissipation Model/Finite Rate Chemistry is found to over-predict mean temperature and species concentrations. Solving for reaction progress equation with its variance using scalar dissipation rate modelling produced reasonably good agreement with the available experimental data. Two different turbulence models Shear Stress Transport (SST) and Scale Adaptive Simulation (SAS) SST are tested and results from transient SST simulations are observed to be predicting well. SAS-SST is found to under-predict with temperature and species distribution.

Selecting a multi-agent system development tool for industrial applications: A case study of self-serving aircraft assets

Industrialists have few example processes they can benchmark against in order to choose a multi-agent development kit. In this paper we present a review of commercial and academic agent tools with the aim of selecting one for developing an intelligent, self-serving asset architecture. In doing so, we map and enhance relevant assessment criteria found in literature. After a preliminary review of 20 multiagent platforms, we examine in further detail those of JADE, JACK and Cougaar. Our findings indicate that Cougaar is well suited for our requirements, showing excellent support for criteria such as scalability, persistence, mobility and lightweightness. © 2010 IEEE.

Industrial ecology at factory level - A conceptual model

Growing environmental concerns caused by natural resource depletion and pollution need to be addressed. One approach to these problems is Sustainable Development, a key concept for our society to meet present as well as future needs worldwide. Manufacturing clearly has a major role to play in the move towards a more sustainable society. However it appears that basic principles of environmental sustainability are not systematically applied, with practice tending to focus on local improvements. The aim of the work presented in this paper is to adopt a more holistic view of the factory unit to enable opportunities for wider improvement. This research analyses environmental principles and industrial practice to develop a conceptual manufacturing ecosystem model as a foundation to improve environmental performance. The model developed focuses on material, energy and waste flows to better understand the interactions between manufacturing operations, supporting facilities and surrounding buildings. The research was conducted in three steps: (1) existing concepts and models for industrial sustainability were reviewed and environmental practices in manufacturing were collected and analysed; (2) gaps in knowledge and practice were identified; (3) the outcome is a manufacturing ecosystem model based on industrial ecology (IE). This conceptual model has novelty in detailing IE application at factory level and integrating all resource flows. The work is a base on which to build quantitative modelling tools to seek integrated solutions for lower resource input, higher resource productivity, fewer wastes and emissions, and lower operating cost within the boundary of a factory unit. © 2012 Elsevier Ltd. All rights reserved.

Gaussian Process training with input noise

In standard Gaussian Process regression input locations are assumed to be noise free. We present a simple yet effective GP model for training on input points corrupted by i.i.d. Gaussian noise. To make computations tractable we use a local linear expansion about each input point. This allows the input noise to be recast as output noise proportional to the squared gradient of the GP posterior mean. The input noise variances are inferred from the data as extra hyperparameters. They are trained alongside other hyperparameters by the usual method of maximisation of the marginal likelihood. Training uses an iterative scheme, which alternates between optimising the hyperparameters and calculating the posterior gradient. Analytic predictive moments can then be found for Gaussian distributed test points. We compare our model to others over a range of different regression problems and show that it improves over current methods.

Factor analysis based VTS discriminative adaptive training

Vector Taylor Series (VTS) model based compensation is a powerful approach for noise robust speech recognition. An important extension to this approach is VTS adaptive training (VAT), which allows canonical models to be estimated on diverse noise-degraded training data. These canonical model can be estimated using EM-based approaches, allowing simple extensions to discriminative VAT (DVAT). However to ensure a diagonal corrupted speech covariance matrix the Jacobian (loading matrix) relating the noise and clean speech is diagonalised. In this work an approach for yielding optimal diagonal loading matrices based on minimising the expected KL-divergence between the diagonal loading matrix and "correct" distributions is proposed. The performance of DVAT using the standard and optimal diagonalisation was evaluated on both in-car collected data and the Aurora4 task. © 2012 IEEE.