EM - Escola de Minas

URI permanente desta comunidadehttp://www.hml.repositorio.ufop.br/handle/123456789/6

Notícias

A Escola de Minas de Ouro Preto foi fundada pelo cientista Claude Henri Gorceix e inaugurada em 12 de outubro de 1876.

Navegar

Resultados da Pesquisa

Agora exibindo 1 - 10 de 12

Automatic system for visual detection of dirt buildup on conveyor belts using convolutional neural networks.
(2020) Santos, André Almeida; Rocha, Filipe Augusto Santos; Reis, Agnaldo José da Rocha; Guimarães, Frederico Gadelha
Conveyor belts are the most widespread means of transportation for large quantities of materials in the mining sector. Therefore, autonomous methods that can help human beings to perform the inspection of the belt conveyor system is a major concern for companies. In this context, we present in this work a novel and automatic visual detector that recognizes dirt buildup on the structures of conveyor belts, which is one of the tasks of the maintenance inspectors. This visual detector can be embedded as sensors in autonomous robots for the inspection activity. The proposed system involves training a convolutional neural network from RGB images. The use of the transfer learning technique, i.e., retraining consolidated networks for image classification with our collected images has shown very effective. Two different approaches for transfer learning have been analyzed. The best one presented an average accuracy of 0.8975 with an F-1 Score of 0.8773 for the dirt recognition. A field validation experiment served to evaluate the performance of the proposed system in a real time classification task.
A self-adaptive evolutionary fuzzy model for load forecasting problems on smart grid environment.
(2016) Coelho, Vitor Nazário; Coelho, Igor Machado; Coelho, Bruno Nazário; Reis, Agnaldo José da Rocha; Enayatifar, Rasul; Souza, Marcone Jamilson Freitas; Guimarães, Frederico Gadelha
The importance of load forecasting has been increasing lately and improving the use of energy resources remains a great challenge. The amount of data collected from Microgrid (MG) systems is growing while systems are becoming more sensitive, depending on small changes in the daily routine. The need for flexible and adaptive models has been increased for dealing with these problems. In this paper, a novel hybrid evolutionary fuzzy model with parameter optimization is proposed. Since finding optimal values for the fuzzy rules and weights is a highly combinatorial task, the parameter optimization of the model is tackled by a bio-inspired optimizer, so-called GES, which stems from a combination between two heuristic approaches, namely the Evolution Strategies and the GRASP procedure. Real data from electric utilities extracted from the literature are used to validate the proposed methodology. Computational results show that the proposed framework is suitable for short-term forecasting over microgrids and large-grids, being able to accurately predict data in short computational time. Compared to other hybrid model from the literature, our hybrid metaheuristic model obtained better forecasts for load forecasting in aMG scenario, reporting solutions with low variability of its forecasting errors.
Multi-objective energy storage power dispatching using plug-in vehicles in a smart-microgrid.
(2016) Coelho, Vitor Nazário; Coelho, Igor Machado; Coelho, Bruno Nazário; Cohen, Miri Weiss; Reis, Agnaldo José da Rocha; Silva, Sidelmo Magalhães; Souza, Marcone Jamilson Freitas; Fleming, Peter J.; Guimarães, Frederico Gadelha
This paper describes a multi-objective power dispatching problem that uses Plug-in Electric Vehicle (PEV) as storage units.We formulate the energy storage planning as a Mixed-Integer Linear Programming (MILP) problem, respecting PEV requirements, minimizing three different objectives and analyzing three different criteria. Two novel cost-to-variability indicators, based on Sharpe Ratio, are introduced for analyzing the volatility of the energy storage schedules. By adding these additional criteria, energy storage planning is optimized seeking to minimize the following: total Microgrid (MG) costs; PEVs batteries usage; maximum peak load; difference between extreme scenarios and two Sharpe Ratio indices. Different scenarios are considered, which are generated with the use of probabilistic forecasting, since prediction involves inherent uncertainty. Energy storage planning scenarios are scheduled according to information provided by lower and upper bounds extracted from probabilistic forecasts. A MicroGrid (MG) scenario composed of two renewable energy resources, a wind energy turbine and photovoltaic cells, a residential MG user and different PEVs is analyzed. Candidate non-dominated solutions are searched from the pool of feasible solutions obtained during different Branch and Bound optimizations. Pareto fronts are discussed and analyzed for different energy storage scenarios. Perhaps the most important conclusion from this study is that schedules that minimize the total system cost may increase maximum peak load and its volatility over different possible scenarios, therefore may be less robust.
A cognitive system for fault prognosis in power transformers.
(2015) Sica, Fernando Cortez; Guimarães, Frederico Gadelha; Duarte, Ricardo de Oliveira; Reis, Agnaldo José da Rocha
The power transformer is one of the most critical and expensive equipments in an electric power system.If it is out of service in an unexpected way, the damage for both society and electric utilities is verysignificant. Over the last decades, many computational tools have been developed to monitor the ‘health’of such an important equipment. The classification of incipient faults in power transformers via DissolvedGas Analysis (DGA) is, for instance, a very well known technique for this purpose. In this paper we presentan intelligent system based on cognitive systems for fault prognosis in power transformers. The proposedsystem combines both evolutionary and connectionist mechanisms into a hybrid model that can bean essential tool in the development of a predictive maintenance technology, to anticipate when anyequipment fault might occur and to prevent or reduce unplanned reactive maintenance. The proposedprocedure has been applied to real databases derived from chromatographic tests of power transformersfound in the literature. The obtained results are fully described showing the feasibility and validity ofthe new methodology. The proposed system can help Transformer Predictive Maintenance programmesoffering a low cost and highly flexible solution for fault prognosis.
NeuroDem - a neural network based short term demand forecaster.
(2001) Silva, Alexandre Pinto Alves da; Rodrigues, Ubiratan de Paula; Reis, Agnaldo José da Rocha; Moulin, Luciano Souza
The application of Neural Network (NN) based Short-Term Load Forecasting (STLF) has developed to sophisticated practical systems over the years. However, the question of how to maximize the generalization ability of such machines, together with the choice of architecture, activation functions, training set data and size, etc. makes up a huge number of possible combinations for the final NN design, whose optimal solution has not been figured yet. This paper describes a STLF system (NeuroDem) which has been used by Brazilian electric utilities for 3 years. It uses a non-parametric model based on a linear model coupled with a polynomial network, identified by pruninglgrowing mechanisms. NeuroDem has special features of data pre-processing and confidence intervals calculations, which are also described. Results of load forecasts are presented for one year with forecasting horizons from 15 min. to 168 hours ahead.
Enhancing neural network based load forecasting via preprocessing.
(2001) Silva, Alexandre Pinto Alves da; Reis, Agnaldo José da Rocha; El-Sharkawi, Mohamed A.; Marks II, Robert J.
The importance of Short-Term Load Forecasting (STLF) has increased, lately. With deregulation and competition, energy price forecasting has become a big business. Load bus forecasting is essential for feeding the analytical methods used for determining energy prices. The variability and nonstationarity of loads are getting worse due to the dynamics of energy tariffs. Besides, the number of nodal loads to be predicted does not allow frequent interventions from load forecasting specialists. More autonomous load predictors are needed in the new competitive scenario. Despite the success of neural network based STLF, techniques for preprocessing the load data have been overlooked. In this paper, different techniques for preprocessing a load series have been investigated. The main goal is to induce stationarity and to emphasize the relevant features of the series in order to produce more robust load forecasters. One year of load data from a Brazilian electric utility has been used to validate the proposed
Artificial neural network-based short-term demand forecaster.
(2003) Silva, Alexandre Pinto Alves da; Rodrigues, Ubiratan de Paula; Reis, Agnaldo José da Rocha; Moulin, Luciano Souza; Nascimento, Paulo Cesar do
The importance of Short-Term Load Forecasting (STLF) has been increasing lately. With deregulation and competition, energy price forecasting has become a big business. Bus load forecasting is essential to feed analytical methods utilized for determining energy prices. The variability and non-stationarity of loads are becoming worse due to the dynamics of energy tariffs. Besides, the number of nodal loads to be predicted does not allow frequent interventions from load forecasting experts. More autonomous load predictors are needed in the new competitive scenario. The application of neural network-based STLF has developed sophisticated practical systems over the years. However, the question of how to maximize the generalization ability of such machines, together with the choice of architecture, activation functions, training set data and size, etc. makes up a huge number of possible combinations for the final Neural Network (NN) design, whose optimal solution has not been figured yet. This paper describes a STLF system which uses a non-parametric model based on a linear model coupled with a polynomial network, identified by pruning/growing mechanisms. The load forecaster has special features of data preprocessing and confidence intervals calculations, which are also described. Results of load forecasts are presented for one year with forecasting horizons from 15 min. to 168 hours ahead
A hierarchical self-organizing map model in short-termload forecasting.
(2004) Carpinteiro, Otávio Augusto Salgado; Reis, Agnaldo José da Rocha
This paper proposes a novel neural model to the problem of short-term load forecasting. The neural model is made up of two self-organizing map nets|one on top of the other. It has been successfully applied to domains in which the context information given by former events plays a primary role. The model was trained and assessed on load data extracted from a Brazilian electric utility. It was required to predict once every hour the electric load during the next 24 hours. The paper presents the results, and evaluates them
A hierarchical hybrid neural model in short-termload forecasting.
(2004) Carpinteiro, Otávio Augusto Salgado; Reis, Agnaldo José da Rocha; Quintanilha Filho, Paulo Sergio
This paper proposes a novel neural model to the problem of short-term load forecasting. The neural model is made up o f two self-organizing map nets one on top of the other |,and a single-layer perceptron. It has application into domains in which the context information given by former events plays aprimary role. The model was trained and assessed onload data extracted from a Brazilian electric utility. It was required to predict once every hour the electric load during the next six hours. The paper presents the results, and evaluates them.
A SOM-based hierarchical model to short-term load forecasting
(2005) Carpinteiro, Otávio Augusto Salgado; Reis, Agnaldo José da Rocha

EM - Escola de Minas

Notícias

Navegar

Filtros

Configurações

Ordenar por

Resultados por página

Resultados da Pesquisa