Determination of moisture diffusion coefficient in transformer paper using thermogravimetric analysis

This paper presents the methodology used and the results obtained in the determination of moisture diffusion coefficient of non-impregnated transformer insulating paper. In order to establish the diffusion coefficient, drying curves of paper samples were obtained by means of thermogravimetric experiments. The diffusion coefficient parameters were found by applying an optimization process based on genetic algorithms. The error function between measured and simulated curves was determined, and the parameters achieving the best correspondence between measured and estimated values were obtained. As a result, a new equation for the diffusion coefficient of non-impregnated insulating paper is proposed, depending on average moisture concentration, temperature and insulation thickness. The proposed coefficient was validated through experimental cases finding a good agreement between the experimental drying curves and those obtained by simulation using the diffusion coefficient. The proposed diffusion coefficient can be used for the determination of the time required to dry power transformers in factory.     

Ethylene production,respiration and gas exchange modelling in modified atmosphere packaging for banana fruits

A mathematical model was developed from experimental measurements to describe the evolution of the O₂, CO₂ and ethylene in a modified atmosphere packaging system for Cavendish bananas. The respiration and ethylene production in the fruits were experimentally obtained from a closed system method and then represented by Michaelis–Menten equations of enzyme kinetics. The gas transfer through the packaging was described by a Fick's diffusion equation, and the temperature dependence was represented based on the Arrhenius law. The model was validated by packaging the fruit in perforated bags of polypropylene and low density polyethylene at 12 °C for a period of 8 days. With the developed model it was possible to satisfactorily describe the experimental evolution of the gas content in the headspace of the packages, obtaining coefficients of determination (R²) of 0.93 for the O₂ levels, 0.90–0.91 for the CO₂ levels, and 0.89–0.93 for the ethylene levels.     

Hierarchical feature selection based on relative dependency for gear fault diagnosis

Feature selection is an important aspect under study in machine learning based diagnosis, that aims to remove irrelevant features for reaching good performance in the diagnostic systems. The behaviour of diagnostic models could be sensitive with regard to the amount of features, and significant features can represent the problem better than the entire set. Consequently, algorithms to identify these features are valuable contributions. This work deals with the feature selection problem through attribute clustering. The proposed algorithm is inspired by existing approaches, where the relative dependency between attributes is used to calculate dissimilarity values. The centroids of the created clusters are selected as representative attributes. The selection algorithm uses a random process for proposing centroid candidates, in this way, the inherent exploration in random search is included. A hierarchical procedure is proposed for implementing this algorithm. In each level of the hierarchy, the entire set of available attributes is split in disjoint sets and the selection process is applied on each subset. Once the significant attributes are proposed for each subset, a new set of available attributes is created and the selection process runs again in the next level. The hierarchical implementation aims to refine the search space in each level on a reduced set of selected attributes, while the computational time-consumption is improved also. The approach is tested with real data collected from a test bed, results show that the diagnosis precision by using a Random Forest based classifier is over 98 % with only 12 % of the attributes from the available set.     

uBench: exposing the impact of CUDA block geometry in terms of performance

The choice of thread-block size and shape is one of the most important user decisions when a parallel problem is written for any CUDA architecture. The reason is that thread-block geometry has a significant impact on the global performance of the program. Unfortunately, the programmer has not enough information about the subtle interactions between this choice of parameters and the underlying hardware. This paper presents uBench, a complete suite of micro-benchmarks, in order to explore the impact on performance of (1) the thread-block geometry choice criteria, and (2) the GPU hardware resources and configurations. Each micro-benchmark has been designed to be as simple as possible to focus on a single effect derived from the hardware and thread-block parameter choice. As an example of the capabilities of this benchmark suite, this paper shows an experimental evaluation and comparison of Fermi and Kepler architectures. Our study reveals that, in spite of the new hardware details introduced by Kepler, the principles underlying the block geometry selection criteria are similar for both architectures.     

A parallel evolutionary algorithm for technical market indicators optimization

This paper deals with the optimization of parameters of technical indicators for stock market investment. Price prediction is a problem of great complexity and, usually, some technical indicators are used to predict market trends. The main difficulty in using technical indicators lies in deciding a set of parameter values. We proposed the use of Multi-Objective Evolutionary Algorithms (MOEAs) to obtain the best parameter values belonging to a collection of indicators that will help in the buying and selling of shares. The experimental results indicate that our MOEA offers a solution to the problem by obtaining results that improve those obtained through technical indicators with standard parameters. In order to reduce execution time is necessary to parallelize the executions. Parallelization results show that distributing the workload of indicators in multiple processors to improve performance is recommended. This parallelization has been performed taking advantage of the idle time in a corporate technology infrastructure. We have configured a small parallel grid using the students Labs of a Computer Science University College.     

Matched disturbance suppression for nonlinear systems stabilizable by logic-based feedback

We focus on a class of nonlinear systems that can be stabilized at a desired compact set by means of logic-based continuous feedback and we show how to design regulators that are robust to matched exosystem-generated disturbances. The proposed design methodology, of interest also for purely continuous-time systems, combines recent advances in the field of hybrid control systems and output regulation for nonlinear systems. As an illustrative example, the problem of global stabilization of a 6-DOF rigid body affected by a force periodic disturbance of uncertain amplitude, phase and frequency is presented.