Fast evolving neuro-fuzzy model and its application in online classification and time series prediction

A strategy for improving speed of the previously proposed evolving neuro-fuzzy model (ENFM) is presented in this paper to make it more appropriate for online applications. By considering a recursive extension of Gath?CGeva clustering, the ENFM takes advantage of elliptical clusters for defining validity region of its neurons which leads to better modeling with less number of neurons. But this necessitates the computing of reverse and determinant of the covariance matrices which are time consuming in online applications with large number of input variables. In this paper a strategy for recursive estimation of singular value decomposition components of covariance matrices is proposed which converts the burdensome computations to calculating reverse and determinant of a diagonal matrix while keeping the advantages of elliptical clusters. The proposed method is applied to online detection of epileptic seizures in addition to prediction of Mackey?CGlass time series and modeling a time varying heat exchanger. Simulation results show that required time for training and test of fast ENFM is far less than its basic model. Moreover its modeling ability is similar to the ENFM which is superior to other online modeling approaches.     

Virtual RF system platform dedicated to heterogeneous complex SoC design

In this paper, we propose to develop a hierarchical library associated with various simulators that can be used in a single platform, called TrustMe-ViP, which enables a unique simulation framework and full model interoperability. Such platform is dedicated to complex SoC design, such as trusted personal devices where cost and time-to-market are very important constraints. To validate this methodology, we present the estimation of BER and power consumption for a Bluetooth transceiver.     

Calcific mitral stenosis in the hemodialysis patient

The dialysis patient is prone to elevations in the calcium phosphorus product and hyperparathyroidism, which contributes to valvular and vascular calcification. We present the case of a young lady on chronic dialysis that developed mitral calcification complicated by severe mitral stenosis, caseous calcification and retinal embolization. She subsequently required mitral valve replacement.     

The role of mechanical forces in the planar-to-bulk transition in growing Escherichia coli microcolonies

Mechanical forces are obviously important in the assembly of three-dimensional multicellular structures, but their detailed role is often unclear. We have used growing microcolonies of the bacterium Escherichia coli to investigate the role of mechanical forces in the transition from two-dimensional growth (on the interface between a hard surface and a soft agarose pad) to three-dimensional growth (invasion of the agarose). We measure the position within the colony where the invasion transition happens, the cell density within the colony and the colony size at the transition as functions of the concentration of the agarose. We use a phenomenological theory, combined with individual-based computer simulations, to show how mechanical forces acting between the bacterial cells, and between the bacteria and the surrounding matrix, lead to the complex phenomena observed in our experiments—in particular the observation that agarose concentration non-trivially affects the colony size at transition. Matching these approaches leads to a prediction for how the friction between the bacteria and the agarose should vary with agarose concentration. Our experimental conditions mimic numerous clinical and environmental scenarios in which bacteria invade soft matrices, as well as shedding more general light on the transition between two- and three-dimensional growth in multicellular assemblies.     

A Multiphase Pore Scale Network Model of Gas Exchange in Apple Fruit

A multiphase pore scale network model was developed to describe mass transfer in apple fruit. The 3D microscale geometry of the tissue was reconstructed from synchrotron radiation tomography images. Individual cells and pores were identified using a watershed segmentation procedure on a Euclidean distance map of the tissue microstructure. Further morphological characteristics of each individual pore, including its volume, connections to the neighbors and the connected area between the pore and its neighbors, were determined. The tissue was represented by a network of nodes (simplified individual pores and cells) that were interconnected by tubes. The transport of the respiratory gases O₂ and CO₂ between two nodes was modelled using diffusion laws and irreversible thermodynamics, while respiration was taken into account in the individual cellular nodes. A numerical procedure was applied to simulate the gas transport within the discrete network and to compute the local diffusivities of the links in the network. The predicted overall gas diffusivities compared well to experimental data and results computed from a microscale continuum model, thereby validating the pore scale network model. This approach is a computationally attractive alternative to a continuum multiphase approach for modelling gas transport in fruit.     

Membrane treatment of side-stream cooling tower water for reduction of water usage

A pilot study was conducted to determine whether membrane treatment on a side stream of recirculating cooling-tower water could reduce overall water usage and discharge. The treated permeate was returned to the cooling tower while the concentrate was discharged to the sanitary sewer. Flow rates, pressures and water chemistry were monitored. The pilot demonstrated potential substantial water savings. Maximum make-up water and discharge reduction were 16% and 49%, respectively. As high as possible permeate recovery is needed to maximize water conservation. Silica scaling on the membranes limited water savings in this pilot. Development of membranes with a solute-rejection capacity less than the 92% average of the membranes used in the pilot would assist in optimizing water savings. Decreased water outlays compensated for the additional energy used by membrane treatment. Scaling control is critical for economic operation.     

3D Virtual Pome Fruit Tissue Generation Based on Cell Growth Modeling

A 3D virtual fruit tissue generator is presented that can distinctly define the microstructural components of a fruit tissue and that can be used to model important physical processes such as gas transport during controlled atmosphere storage. The model is based on the biomechanics of plant cells in tissues. The main merit of this algorithm is that it can account for typical differences in intercellular air space networks and in cell size and shape found between different fruit species and tissues. The cell is considered as a closed thin walled structure, maintained in tension by turgor pressure. The cell walls of adjacent cells are modeled as parallel, linear elastic elements which obey Hooke's law. A 3D Voronoi tessellation is used to generate the initial topology of the cells. Intercellular air spaces of schizogenous origin are generated by separating the Voronoi cells along the edges where three Voronoi cells are in contact; while intercellular air spaces of lysigenous origin are generated by deleting (killing) some of the Voronoi cells randomly. Cell expansion then results from turgor pressure acting on the yielding cell wall material. To find the sequence of positions of each vertex and thus the shape of the tissue with time, a system of differential equations for the positions and velocities of each vertex is established and solved using a Matlab ordinary differential equation solver. Statistical comparison with synchrotron tomography images of fruit tissue is excellent. The virtual tissues can be used to study tissue mechanics and exchange processes of important metabolites.     

Teachers' acceptance and use of an educational portal

In this study, teachers' acceptance and use of an educational portal is assessed based on data from two sources: usage data (number of logins, downloads, uploads, reactions and pages viewed) and an online acceptance questionnaire. The usage data is extracted on two occasions from the portal's database: at survey completion (T1) and twenty-two months later (T2). Framework for this study is C-TAM-TPB (Combined Technology Acceptance Model and Theory of Planned Behavior). 919 usable responses from teachers are obtained. Based on the observed use data at T1, four types of portal users are distinguished: ‘new’ (N = 37), ‘light’ (N = 641), ‘medium’ (N = 201), and ‘heavy’ (N = 40). Path analyses show that all predictor variables in C-TAM-TPB influence teachers' portal acceptance, but their significance level varies depending on the user type. The strongest predictors of behavioral intention to use the portal are attitude (‘new’) and perceived usefulness (‘light’, ‘medium’ and ‘heavy’), with variance explained ranging from .39 (‘medium’) to .71 (‘heavy’). The observed use data show that the portal is primarily used to search for and download material, rather than for sharing material or information. The use data at T2 show that teachers become more efficient in their search behavior and that the majority of the teachers use the portal more frequently. Guidelines are proposed to policymakers and school boards aiming to introduce a similar technology to teachers.     

Improved process control of an industrial sludge centrifuge-dryer installation through binary logistic regression modeling of the fouling issues

Biological wastewater treatment generates huge amounts of waste sludge which need to be dewatered and eventually dried to minimize transportation and incineration costs. A characteristic feature of sludge in this context is that it turns into a sticky substance during its drying process inducing fouling problems in the drying installation. At the wastewater treatment plant of Monsanto in Antwerp, Belgium, one enclosed centrifuge-dryer system is used to dry the sludge. In the past, this installation had to be shut down regularly due to dryer fouling problems. To avoid these operational problems, a binary logistic regression analysis is presented in this research based on a 5-year database, resulting in an empirical model for the evaluation of the dryer fouling risk as a function of the sludge feed characteristics. The model inputs are the sludge volume index (SVI) and the dosing of clay additive and tertiary (flotation) sludge, the latter containing polyaluminumchloride (PACl), to the sludge feed of this particular system.By exploiting the knowledge captured by this model, the derived control strategy is based on the value of the SVI. Whenever the SVI is high the original high clay dosing to the feed needs to be maintained. At moderate SVI values, implying an intrinsically better sludge dewaterability, the strategy dictates a reduction in the clay dosing to the sludge feed to have a reduced sludge solids dryness after dewatering, thereby avoiding that the sludge exhibits its most sticky phase when passing the most fouling sensitive part of the dryer. When the SVI is lower than 50 mL/g the control strategy states that conditioning of the sludge with PACl is required to mask the stickiness instead of postponing it, avoiding that the stickiness of the sludge already hampers the dewatering stage of the process.     

CIGI2011: A heuristic method for resource-constrained project scheduling with activity overlapping

The overlapping of activities is a common practice to accelerate the execution of engineering projects. This technique consists in executing in parallel two activities, normally executed in a sequential way, by allowing the downstream activity to start before the end of the upstream activity based on preliminary information. In this paper, we propose a constructive heuristic for the resource-constrained project scheduling problem with overlapping modes (RCPSP-OM). Given a set of activities to execute, the RCPSP-OM consists in determining the order of execution in time of a set of activities so as to minimize the total project duration, while respecting precedence relations, resource constraints and overlapping possibilities. The heuristic implies that rework tasks related to overlapping are added to downstream activities and that the consumption of the resources is constant throughout the execution of the project (including rework). The method also considers that the possible overlapping modes for every couple of activities and the duration of rework tasks associated with every mode are known in advance. Results show that, when the objective consists in minimizing the project duration, the consideration of the costs associated to activity overlapping allows to significantly reducing the cost of reworks. On the other hand, when the objective consists in maximizing the gains related to the project execution, the search for the best trade-off between acceleration and increase of project costs enables to avoid losses.