Reconstrução e reconhecimento de imagens binárias utilizando o algoritmo Máquina de Boltzmann

The objective of this work is to use algorithms known as Boltzmann Machine to rebuild and classify patterns as images. This algorithm has a similar structure to that of an Artificial Neural Network but network nodes have stochastic and probabilistic decisions. This work presents the theoretical framework of the main Artificial Neural Networks, General Boltzmann Machine algorithm and a variation of this algorithm known as Restricted Boltzmann Machine. Computer simulations are performed comparing algorithms Artificial Neural Network Backpropagation with these algorithms Boltzmann General Machine and Machine Restricted Boltzmann. Through computer simulations are analyzed executions times of the different described algorithms and bit hit percentage of trained patterns that are later reconstructed. Finally, they used binary images with and without noise in training Restricted Boltzmann Machine algorithm, these images are reconstructed and classified according to the bit hit percentage in the reconstruction of the images. The Boltzmann machine algorithms were able to classify patterns trained and showed excellent results in the reconstruction of the standards code faster runtime and thus can be used in applications such as image recognition.

Avaliação da soldagem GMAW mecanizada e automatizada para união de dutos utilizando fontes convencional e com controle de curto- circuito

Welding is one of the most employed process for joining steel pipes. Although, manual welding is still the most used one, mechanized version and even automatized one have increased its demand. Thus, this work deals with girth welding of API 5L X65 pipes with 8” of nominal diameter and 8.0 mm thickness, beveled with V-30º narrow gap. Torch is moved by a bug carrier (mechanized welding) and further the parameters are controlled as a function of angular position (automatized welding). Welding parameters are presented for filling the joint with two-passes (root and filling/capping passes). Parameters for the root pass were extracted from previous author´s work with weldments carried out in plates, but validated in this work for pipe welding. GMAW processes were assessed with short-circuit metal transfer in both conventional and derivative modes using different technologies (RMD, STT and CMT). After the parameter determination, mechanical testing was performed for welding qualification (uniaxial tension, face and root bending, nick break, Charpy V-notch impact, microhardness and macrograph). The initially obtained results for RMD and CMT were acceptable for all testing and, in a second moment, also for the STT. However, weld beads carried out by using the conventional process failed and revealed the existence of lack of fusion, which required further parametrization. Thus, a Parameter-Variation System for Girth Welding (SVP) was designed and built to allow varying the welding parameters as a function of angular position by using an inclinometer. The parameters were set for each of the three angular positions (flat, vertical downhill and overhead). By using such equipment and approach, the conventional process with parameter variation allowed reducing the welding time for joint accomplishment of the order of 38% for the root pass and 30% for the filling/capping pass.