Rajive Ganguli

Determination of Rheological Behavior of Aluminum Oxide Nanofluid and Development of New Viscosity Correlations

Abstract Experimental investigations have been carried out to study the rheological behavior of aluminum oxide nanofluid. Nanoparticles with average particle size of 53 nm were dispersed in a base fluid of 60% (by mass) of ethylene glycol and water. Nanofluids of volumetric concentrations 1 to 10% were tested for determining the viscous properties. It was found that this nanofluid behaved as non-Newtonian at lower temperatures (-35°C to 0°C) and Newtonian at higher temperatures (0°C to 90°C). The data showed that the viscosity increases with an increase in concentration and decreases with increase in temperature. Two new correlations were developed expressing viscosity as a function of temperature and concentration.

Data Segmentation and Genetic Algorithms for Sparse Data Division in Nome Placer Gold Grade Estimation Using Neural Network and Geostatistics

Ore reserve estimation, based on sparse drill hole data, was conducted for a placer gold property in Nome, Alaska. A problem with sparse data is that random subdivision of the data into modelling and evaluation subsets (as is commonly done) becomes a problem, as random selection may result in biased/skewed subsets. Therefore, a technique that combined data segmentation with genetic algorithms (GA) was applied to divide the samples into three equivalent subsets: training, validation and testing. Data segmentation was done on the basis of the distribution of gold values. Neural network and a variety of kriging techniques were used to estimate gold grades. A multi-layer feed forward neural network along with “early/quick stop” training was used for neural network modelling. A comparative evaluation of kriging and neural network methods was then performed. The results revealed that neural network was generally superior to the kriging techniques for gold grade estimation in the Nome deposit.

A hybrid ensemble model of kriging and neural network for ore grade estimation

This paper presents a new hybrid methodology involving kriging and artificial neural network for ore grade estimation of two variables namely, Al2O3% and SiO2%, in a bauxite deposit. The dataset was divided into three statistically similar subsets: training, calibration and validation sets using a genetic algorithm. The proposed hybrid ensemble model was formed using a kriging model and several neural network models. The outputs of these component models were combined using two methods to produce a unified prediction. The results indicated that the hybrid model was not a better estimator than the kriging model for the variable Al2O3%. However, it provides slightly better performance in comparison to any of the other component models in the ensemble for the variable SiO2%.

Comparing the predictive performance of neural networks with ordinary kriging in a bauxite deposit

Abstract This paper investigates the performance of neural network and ordinary kriging techniques in estimating five variables: accumulated Al2O3, accumulated SiO2, percentage Al2O3, percentage SiO2, and thickness of a bauxite deposit in India. The assay values obtained from exploratory boreholes were compiled according to the geological composition of the deposit. Genetic algorithms were used to divide the dataset into model development and evaluation subsets, ensuring that the modelling subset and evaluation subset were similar and that performance evaluation was valid. The results indicate that neural networks and ordinary kriging performed equally well in this deposit, except for the variable accumulated Al2O3. However, the coefficients of determination (R2) of predictions were not very good.

Machine Learning Algorithms and Their Application to Ore Reserve Estimation of Sparse and Imprecise Data

Traditional geostatistical estimation techniques have been used predominantly by the mining industry for ore reserve estimation. Determination of mineral reserve has posed considerable challenge to mining engineers due to the geological complexities of ore body formation. Extensive research over the years has resulted in the development of several state-of-the-art methods for predictive spatial mapping, which could be used for ore reserve estimation; and recent advances in the use of machine learning algorithms (MLA) have provided a new approach for solving the prob-lem of ore reserve estimation. The focus of the present study was on the use of two MLA for estimating ore reserve: namely, neural networks (NN) and support vector machines (SVM). Application of MLA and the various issues involved with using them for reserve estimation have been elaborated with the help of a complex drill-hole dataset that exhibits the typical properties of sparseness and impreciseness that might be associated with a mining dataset. To investigate the accuracy and applicability of MLA for ore reserve estimation, the generalization ability of NN and SVM was compared with the geostatistical ordinary kriging (OK) method.

Expert System for Equipment Selection

An expert system was developed for equipment selection for various unit operations in open pit mining. For each unit operation, it considers a wide variety of equipment and ranks them based on their suitability. A major difference between this system and other published expert systems in the considered domain is that it incorporates the uncertainty that is an inherent characteristic of the factors affecting equipment selection. It also allows users to specify the site-specific relative importance of each of the governing factors. This makes the system relatively flexible and allows its adaptation to different mining conditions. The programming features of the expert system include uncertainty databases and object oriented programming. The expert system is validated with a case study (Malanjkhand Copper Mine, India).

General regression neural network residual estimation for ore grade prediction of limestone deposit

Abstract The aim of the present paper is to provide an improved estimator for the ore grade prediction of a limestone deposit in India. A generalised modelling framework with the help of general regression neural network and the ordinary kriging was formulated to capture the spatial variability of the deposit. In this platform, spatial variability of the deposit is assumed to be characterised by three major components: spatial trend component, regionalised component, and purely random component. The general regression neural network (GRNN) model was used to capture the spatial trend component, and the ordinary kriging technique was implemented to capture the regionalised component. The GRNN model was developed using the spatial coordinates (Northing, Easting and Elevation) as the input parameters and the grade attributes (CaO, Al2O3, Fe2O3 and SiO2) as the output parameters. The performance of the GRNN residual kriging model was tested using a testing data set, and the outputs of this model were compared with the outputs of the GRNN model, and the ordinary kriging. The comparative results show that the GRNN residual kriging model provided significant improvement over the ordinary kriging, however, it only shows a marginal improvement over the GRNN model.

Algorithms to control coal segregation under non-stationary conditions: Part II: Time series based methods

Abstract In this paper, procedures involving the use of time series models are developed for coal segregation control. In contrast to the methods of Part I of this paper, time series models directly accommodate the auto-correlated nature of the coal quality levels when estimating parameters to characterize the process. Moreover, they also provide forecasting capability that is useful in segregation control. Special attention is given in the paper to development of viable parameter updating strategies in order to deal with non-stationary applications. Performance of the time series methods is evaluated, and it compares favorably to the methods developed in Part I. Although more complicated to implement than the Part I methods, time series methods have the potential to be extended to applications where quality targets are to be maintained over small batches of coal (homogeneity control), whereas the other methods only apply to large batch quality targeting.

Exploration and Estimation of Gravel Resource Potential in Southeast Chukchi Sea Continental Shelf off Kivalina, Alaska

Frequent storm surges in the Alaskan arctic result in washovers and high erosion of barrier islands. The village council of Kivalina has resolved to relocate its present location on a barrier island in northwest arctic Alaska to an adjacent onshore site. The relocation plan envisages excavation of the upper 4 meters of the 25-km2 onshore permafrost ground and the construction of a foundation pad. The objective of this research is to estimate the gravel resource potential in the continental shelf off Kivalina. In this context, seismic surveys and sediment sampling were conducted. The seismic surveys were of limited use as they failed to resolve the upper 1–2 m of the seafloor. The lithostratigraphy indicated dominance of the 2.4–3.4 mm size fraction in the region north of Kivalina. The geostatistical analysis indicated an omnidirectional variogram fit to the data with ordinary kriging (OK) producing the best kriging estimate of the gravel resource potential. At least 20 × 106 m3 of gravel above 90% cut-off is present in the upper 0.5 m of the seafloor. The regional Pleistocene glaciation has affected the lateral variations in gravel abundance in the nearshore southeast Chukchi Sea.

Analysis of the twisting department at Superior Cable Corporation: A case study

Abstract The production of communication cables involves four major serial operations, namely tandem, twisting, stranding, and jacketing. Superior Cable Corporation manufactures such communications cables at the Elizabethtown, Kentucky, plant. The bottleneck operation in this facility is the twisting department. In this paper, we used simulation to analyze the operations in the twisting department and provide several recommendations to improve the production capacity of this department. Several of our recommendations are successfully implemented.