Abstract The present study applies computational intelligence techniques in the development of a hybrid model composed of Artificial Neural Networks (ANNs) and Genetic Algorithms (GAs) (MLP-GA) to estimate and fill in the gaps in the monthly variables of evaporation, maximum temperature and relative humidity to six regions in the state of Rio de Janeiro (RJ), Brazil. The results were evaluated using statistical techniques and compared with results obtained by the Multiple Linear Regression (RLM), Multilayer Perceptron (MLP) and Radial Basis Function (RBF) models and also compared with the data recorded by the weather stations. The correlation coefficient (r) between the evaporation estimates generated by MLP-GA with the recorded data showed a high relationship, remaining between 0.82 to 0.97. The average percentage error (MPE) ranged from 6.01% to 9.67%, indicating a accuracy between 90% to 94%. For the maximum temperature generated by MLP-GA the correlation with the recorded data remained between 0.97 to 0.99. It also presented the MPE between 0.95% to 1.57%, maintaining the accuracy of the estimated data between 98% to 99%. The correlation coefficient (r) between the relative humidity estimates generated with the MLP-GA remained between 0.89 a 0.97, the MPE between 1.15% to 1.89%, which guaranteed a rate higher than 98% of correctness in its estimates. Such results demonstrated gains in relation to the other applied models and allowed the accomplishment of the filling of most of the missing values.

Resumo O presente estudo aplica técnicas de inteligência computacional no desenvolvimento de um modelo híbrido composto por Redes Neurais Artificiais (RNAs) e Algoritmos Genéticos (AGs) (MLP-GA) para estimar e preencher lacunas nas variáveis mensais de evaporação, temperatura máxima e umidade relativa em seis regiões do estado do Rio de Janeiro (RJ), Brasil. Os resultados foram avaliados por meio de técnicas estatísticas e comparados com os resultados obtidos pelos modelos de Regressão Linear Múltipla (RLM), Perceptron de Multicamadas (MLP) e Redes de Função de Base Radial (RBF), além de serem comparados com os dados registrados pelas estações meteorológicas. O coeficiente de correlação (r) entre as estimativas de evaporação geradas pelo MLP-GA com os dados registrados mostrou uma relação elevada, permanecendo entre 0,82 e 0,97. O erro percentual médio (MPE) variou de 6,01% a 9,67%, indicando uma precisão entre 90% e 94%. Para a temperatura máxima gerada pelo MLP-GA, a correlação com os dados registrados permaneceu entre 0,97 e 0,99. Apresentou também o MPE entre 0,95% e 1,57%, mantendo a precisão dos dados estimados entre 98% e 99%. O coeficiente de correlação (r) entre as estimativas de umidade relativa geradas pelo MLP-GA permaneceu entre 0,89 e 0,97, com MPE entre 1,15% e 1,89%, garantindo uma taxa superior a 98% de acerto em suas estimativas. Tais resultados demonstraram ganhos em relação aos outros modelos aplicados e permitiram o preenchimento da maioria dos valores ausentes.

Resumo O alto consumo de água pela agricultura torna cada vez mais essencial o conhecimento da evapotranspiração de referência (ETo) para a realização do manejo da irrigação. Entretanto, definir um método adequado às diferentes localidades está associado à disponibilidade dos dados meteorológicos e a adaptação dos métodos às localidades aplicadas. Assim, o objetivo deste trabalho foi comparar e avaliar o desempenho dos métodos de ETo: Hargreaves-Samani, Jensen-Haise, Benavides & Lopez e Hamon com o método padrão Penman Monteith FAO-56, para estimar a ETo diária de seis regiões, três do estado do Rio de Janeiro e três do estado do Espírito Santo. As variáveis meteorológicas empregadas foram cedidas pelo Centro de Previsão de Tempo e Estudos Climáticos e do Instituto Nacional de Pesquisas Espaciais (CPTEC - INPE). O desempenho dos modelos foi avaliado por diferentes técnicas estatísticas onde o modelo que melhor se adaptou às localidades estudadas nos dois estados foi o de Jensen Haise, tendo obtido os índices de correlação (r) entre 0,73 a 0,94 e a confiabilidade (C) entre 0,60 a 0,90 com o modelo de Penman Monteith.

Abstract The high consumption of water by agriculture makes knowledge about reference evapotranspiration (ETo) for irrigation management more and more essential. However, define an appropriate method to the different locations is associated with the availability of meteorological data and the adaptation of the methods to the application sites. Thus, the objective of this study is to compare and evaluate the performance of the ETo methods: Hargreaves-Samani, Jensen-Haise, Benavides & Lopez and Hamon using the standard method Penman Monteith FAO-56, to estimate the daily ETo of six regions, three of the state of Rio de Janeiro and three of the state of Espírito Santo. The meteorological variables employed were provided by the Center for Weather Forecasting and Climate Studies and the National Institute of Space Research - (CPTEC - INPE). The performance of the models was evaluated by different statistical techniques where the model that best suited the studied locations in the two states was the Jensen Haise, obtaining correlation indexes (r) between 0.73 and 0.94 and reliability (C) between 0.60 and 0.90 with the Penman Monteith model.

Abstract The stability of colloidal suspensions of engineered nanoparticles in the environment is of great interest for the risk assessment and treatment of hazardous materials. The relevance of this study is due to the evidence that engineered nanoparticles have been produced and used in commercial products in large scale, especially ZnO and TiO2, which are transported through the environmental components interacting with soil and water, representing the high potential impact on the ecosystems. In this research, we newly investigated the influence of nanoparticles of zinc oxide (NPZnO) on the behavior of nanoparticles of titania (NPTiO2). The impact of parameters such as pH, zeta potential, ionic strength, electric conductivity, suspended nanoparticles size of the colloidal suspensions containing hazardous nanoparticles on its stability, were studied using experimental and theoretical approaches. Leaching experiments were carried out within a sandy soil column model. It was found that the NPTiO2 in suspensions enhance the stability of the nanoparticles colloidal suspensions, keeping higher nanoparticles concentrations, while NPZnO promotes rapid aggregation, improving the soil particle collector efficiency with a lower concentration in equilibrium in the colloidal suspension, which can be used for promoting environmental impact attenuation and mitigation.

Abstract Demand for iron ore worldwide has been steadily increasing, thus implying an increase in the volume of mining resulting in the extraction of ores with higher levels of silicon, phosphorus and sulfur impurities. An extensively used alternative is to perform a pre-treatment of the particles in order to reach the acceptable phosphorus limit in iron ore before its use in steel production. This study aims at the evaluation of the heating process and decrease of the phosphorus content of the particles using microwave energy in iron ore samples percolated by sulfuric acid followed by the leaching technique as a feasible route for the remotion of the phosphorus content in iron ore particles. Using the scanning electron microscopy (SEM) and statistical analysis techniques, it was possible to observe the characteristics of the iron ore particles after treatment with microwave energy. The results indicated that the developed process is feasible. The results of the kinetic simulation contribute to a better understanding of the phenomena of heating using microwave energy and also to the development of new cleaner technologies aiming to improve the efficiency of the acid baking - leaching technique for the reduction of the phosphorus content in iron ore particles.

Unlike the available mineral resources, the steelmaking processes have demanded raw material with lower phosphorus content to decrease the costs, energy use, and the residue generated within the steel plant. One alternative is to develop pretreatment of the iron ore concentrates producing raw materials with lower phosphorus. Depending on the mineral structure, a heat treatment combined with leaching could be an efficient way to achieve concentrates with low phosphorus (less than 0.01%), suitable for steelmaking processes. A fast and efficient way of applying energy to iron ore particles is the use of microwave to heat the particles. Thus, we propose a treatment using microwave heating while mixing with a dilute aqueous solution of sulfuric acid, followed by quenching during leaching with water, as a feasible route for the phosphorus removal from iron ore particles. We performed a design of experiment (DOE) to investigate the optimal conditions of heating and leaching, which maximize the rate of phosphorus removal. The structure of the iron ore particles after their treatment with microwave energy was observed by scanning electron microscopy (SEM). Thus, we analyze the optimal conditions for heating and leaching, the structure of the iron ore particles and which mechanism and rate equations are controlling the phosphorus removal. The results indicated that the average heating time is 10 min, the size of the crack of the order of 20µm and a leaching time of 8 min are adequate to achieve less than 0.01% of phosphorus. We demonstrated that under the most favorable combination of conditions for heating followed by leaching proposed in this study, the removal of the phosphorus content in the iron ore samples could reach 100%.

Abstract A numerical model based on transport equations for momentum, energy and chemical species for the gas and solid phases is proposed to simulate the inner phenomena in the direct reduction of the shaft furnace process for producing directly reduced iron (DRI). The model is verified using industrial data for productivity, raw materials and final composition of the DRI product. The model is used to evaluate operational practices using new raw materials and the composition of the reducing gas in the process. Three cases were considered, which correspond to available raw materials commercialized by different suppliers. The effects on the gas and solid inner temperatures, pressure and phase composition distributions are quantified. The simulation results indicated that good agreement for overall parameters of the process could be achieved and afterwards, detailed features of the inner conditions of the process are predicted.

Abstract This study evaluates the stability and transport behaviors of hazardous nanoparticles into soil landfills using experimental procedures to simulate the effects of natural silica nanoparticle suspensions during their percolation into the soil layers of municipal waste landfills. For this, we prepared stabilized suspensions of nanoparticle oxides containing silicon (NPSiO2), titanium (NPTiO2), copper (NPCuO) and zinc (NPZnO), which are recognized as hazardous to the environment, and we conducted leaching experiments within the soil column by simulating landfills layers and simulating the capture and attenuation of nanomaterials into municipal waste landfills. The results demonstrated that the presence of NPSiO2 in suspensions increases the stable concentrations of copper, zinc and titanium oxides and strongly decreases soil layer effectiveness. In contrast, NPZnO improves effectiveness due to its ability to promote agglomeration and setting conditions, allowing further complexation.

The properties of the super duplex stainless steels (SDSS) are strongly affected by the thermal history imposed by welding procedures. The controlled dual phase microstructure (ferrite and austenite) guarantee excellent mechanical properties such as mechanical strength and corrosion resistance, in addition to small thermal expansion coefficient and high thermal conductivity. In this paper, we newly proposed a model able to predict the thermal history of the welding pieces coupled with local mechanical properties developed during welding procedure that combine the effects of temperature and phase changes during welding. We applied inverse method to fit the thermophysical parameters based on measured data. The model was verified by comparing measured and predicted temperature profiles using thermocouples located within the heat affected zone. Thus, an inverse method was implemented to obtain the parameters fitting for the grain growth evolution compatible with the final microstructure and grain size measured using SEM images and stereological techniques. We demonstrated that very small amount of non-equilibrium deleterious phases and nanosized precipitated are expected during the welding procedures depending upon the local conditions of temperature, compositions and alloying dilution evolutions.