João Azevedo

3-D numerical model for predicting NOx emissions from an industrial pulverized coal combustor

The paper describes a three-dimensional computer simulation developed to predict the behaviour of industrial pulverized coal furnaces. The model was applied to a 300 MWe front-wall fired boiler under different operating conditions. The main characteristics of the current model are the use of the Lagrangian framework to describe the particulate phase, the discrete transfer model to handle the radiation transmission, and the standard k−e model for turbulence treatment. The turbulent dispersion of particles is described by a new stochastic self-stable strategy. The NOx formation mechanism is calculated in a post-processor routine. Measurements of NOx emissions were made in the flue gases; the results agree well with those obtained by the simulations.

Seismic Tests and Numerical Modeling of a Rolling-ball Isolation System

For the seismic isolation of light structures, the use of laminated rubber bearings is neither economical nor, for most cases, technically suited. For the isolation of this type of structure a new system, consisting of steel balls rolling on rubber tracks, has been developed at TARRC (Tun Abdul Razak Research Centre). This article presents the results of experimental tests carried out for the characterization of the behavior of this new device. A numerical model is also proposed that can be used to assess the seismic response of structures with this isolation system. Comparison of the predictions of the numerical model with the experimental data shows that the model is adequate to perform the correct assessment of the seismic response of isolated structures. The results of the experimental campaign of shaking-table tests, as well as the numerical simulations, show that there is an effective reduction of the acceleration levels induced in the isolated structures.

CO-COMBUSTION OF PULVERIZED COAL, PINE SHELLS, AND TEXTILE WASTES IN A PROPANE-FIRED FURNACE: MEASUREMENTS AND PREDICTIONS

This paper describes an experimental and numerical investigation on the co-combustion of propane with pulverized coal, pine shells, and textile wastes. Measurements have been performed in a large-scale laboratory furnace fired by an industrial-type swirl burner. Data are reported for in-flame major gas-phase species concentration, including NOx, in-flame gas temperature, and overall char burnout for three flames: a propane/coal flame, a propane/pine shells flame, and a propane/textile wastes flame. For comparison purposes, data are also reported for a pure propane flame. The experimental results show that CO and unburned hydrocarbon emissions from propane cofiring with pine shells and textile wastes can be important due to the relatively large sizes of the solid particles and that NOx emissions data reproduce the impact on them of the fuel nitrogen content via fuel-NO formation for the three solid fuels studied. Also, the cofiring of propane with pine shells and textile wastes yields particle burnout values much higher than that of the propane/coal flame despite the similarities of the three flames revealed by the in-flame data. This is because of the higher volatile matter content of the pine shells and textile wastes, in spite of their much larger particle sizes, compared with coal. The experimental conditions were then simulated numerically. For the numerical predictions, gas-phase calculations were based on the Eulerian approach whereas the particulate phase predictions were based on a stochastic Lagrangian approach. The turbulence model used herein was the standard two-equation eddy viscosity model. The gaseous fuel and volatiles oxidation was assumed to be controlled by both turbulent mixing and chemical kinetics. Radiation was accounted for. The biomass/waste devolatilization was simulated by a single-reaction, first-order decomposition model. Thermal-NO, prompt-NO, and fuel-NO mechanisms were all considered as potential contributors to the NO formation. From the comparisons of the predictions with the experimental data, a good agreement was generally found, except in the near-burner region close to the furnace symmetry axis. The discrepancies are due to the limitations of the k–ϵ turbulence model to simulate high swirling flows. Moreover, the simplified fuel-NO model used herein requires further improvements to yield more accurate results.

A model for predicting the failure of structural steel elements

Abstract A numerical model for predicting the failure of structural steel elements subjected to cyclic loading is presented. The cyclic response is considered to be inelastic to seismic loading. The model for damage accumulation is expressed as a function of the inelastic strain and the dissipated hysteretic energy. It is shown that damage accumulation does not depend on the maximum strain but does depend on the inelastic deformation. The numerical model is used for the evaluation of the behavior coefficient of steel structures.

Hysteretic behaviour of steel members: Analytical models and experimental tests

Abstract The seismic performance of steel structures is highly dependent on the hysteretic behaviour of their members. The assessment of this behaviour cen be done by means of experimental tests and the use of analytical models that take into account the main phenomena involved in a non-linear response. An overview of experimental methodologies for cyclic tests, including a more detailed investigation of the ECCS recommendations, is presented. Analytical models for the simulation of hysteretic behaviour and comparisons with some experimental results are shown. Finally, it is demonstrated how a cumulative damage model for failure assessment allows for the evaluation of q factors for structural design.

An API proposal for integrating sensor data into web apps and WebRTC

Today we use smartphones and personal computers everywhere to communicate in ways made possible by the ubiquity of the Internet: we exchange emails and instant messages, make voice and video calls and participate in social media. WebRTC furthers the possibilities of audio and video communications and true peer-to-peer communications as it brings these capabilities also to an universal platform: the browser. But communication is more than just audio or video sharing, communication is about providing features capable of filling the distance void. Today we have sensors that can provide data to enrich communication by enabling the usage of information about the context of the remote peer, e.g.: providing temperature information, speed or acceleration. Mobile devices, being equipped with a range of sensors, such as accelerometers, gyroscopes and magnetometers are prime terminals to enrich communications with context data, but all types of devices can make use of external sensing devices or even reach out to sensor networks. As the browser becomes a fundamental platform, sensor data is also relevant to non WebRTC web applications. In this paper, we present an API proposal to enable web applications to access sensor data and an extension to WebRTC to enable peer-to-peer exchange of sensor data, bringing nearby sensor streams to web applications and multimedia communication over the web. Telemedicine, Meteorology or Seismology are prime examples of the applicability of this technology, that also enables new types of context-aware and context-based applications and communications.

Assessment of q factors for seismic design of low-rise steel buildings

A method to evaluate the q factors for seismic design is presented and applied to low-rise steel buildings. Three structural systems are analysed: cantilever structures, diagonally braced cantilever structures and simple frames. Real and artificial accelerograms, with duration of 15 and 25 seconds, are employed to simulate different seismic loading conditions. Failure is characterised by means of a numerical index based on damage accumulation. This index can be obtained from static tests and used to assess failure in dynamic studies. The results of parametric studies on the influence of different natural frequencies, slenderness and axial load level on the q factors are presented.

METHODOLOGY FOR THE PROBABILISTIC ASSESSMENT OF q FACTORS: A DAMAGE INDEX APPROACH

The objective of the present work is to present a methodology for the identification of relevant limit states, namely ultimate limit states leading to structural collapse, and for the assessment of design q factors (or force reduction factors) for reinforced concrete structures under seismic loading. It follows a probabilistic approach based on damage indices. The utilised nonlinear models, as well as the damage indices, which are those proposed by Miner and by Park and Ang, are.described. The methodology of analysis is presented emphasising its probabilistic characteristics. Some parametric studies are carried out, including the analysis of one regular plane frame reinforced concrete structure, designed for three different ductility classes (those proposed by Eurocode 8) and assuming different q factors in design. Results show how the chosen damage indices can be used as parameters to characterise the structural response and how the proposed methodology can be used to assess the design q factors. It is also shown that, for moderate seismic input, the three ductility classes are essentially equivalent in terms of maximum damage indices, but that for higher seismic levels the differences are evident, justifying the use of different q factors.

NOX EMISSIONS FROM A FRONT WALL FIRED PULVERISED COAL BOILER

The present work is concerned with both axissymmetric and three dimensional pulverised coal combustors. The axissymmetric case is used to test and develop a numerical model which is confronted to experimental results including N related species distributions. NO emissions were measured for several operating conditions of a front wall fired pulverised coal boiler 720 MWt. These experimental results are compared to model predictions showing satisfactory overall agreement from the engineering point-of-view.