Application and comparison of RNN, RBFNN and MNLR approaches on prediction of flotation column performance

Evaluation of grade and recovery plays an important role in process control and plant profitability in mineral processing operations, especially flotation. The accurate measurement or estimation of these two parameters, based on the secondary variables, is a critical issue. Data-driven modeling techniques, which entail comprehensive data analysis and implementation of machine learning methods for system forecast, provide an attractive alternative. In this paper, two types of artificial neural networks(ANNs),namely radial basis function neural network(RBFNN) and layer recurrent neural network(RNN), and also a multivariate nonlinear regression(MNLR) model were employed to predict metallurgical performance of the flotation column. The training capacity and the accuracy of these three above mentioned types of models were compared. In order to acquire data for the simulation, a case study was conducted at Sarcheshmeh copper complex pilot plant. Based on the root mean squared error and correlation coefficient values, at training and testing stages, the RNN forecasted the metallurgical performance of the flotation column better than RBF and MNLR models. The RNN could predict Cu grade and recovery with correlation coefficients of 0.92 and 0.9, respectively in testing process.

Bolt-grouting combined support technology in deep soft rock roadway

Analyzing the mineral composition, mechanical properties and ground stress testing in surrounding rock,the study investigated the failure mechanism of deep soft rock roadway with high stress. The boltgrouting combined support system was proposed to prevent such failures. By means of FLAC3D numerical simulation and similar material simulation, the feasibility of the support design and the effectiveness of support parameters were discussed. According to the monitoring the surface and deep displacement in surrounding rock as well as bolt axial load, this paper analyzed the deformation of surrounding rock and the stress condition of the support structure. The monitor results were used to optimize the proposed support scheme. The results of field monitors demonstrate that the bolt-grouting combined support technology could improve the surround rock strength and bearing capacity of support structure, which controlled the great deformation failure and rheological property effectively in deep soft rock roadway with high stress. As a result, the long term stability and safety are guaranteed.

Investigation on leaching of malachite by citric acid

Leaching of an oxidized copper ore containing malachite, as a new approach, was investigated by an organic reagent, citric acid. Sulfuric acid is the most common reagent in the leaching of oxide copper ores, but it has several side effects such as severe adverse impact on the environment. In this investigation, the effects of particle size, acid concentration, leaching time, solid/liquid ratio, temperature, and stirring speed were optimized. According to the experimental results, malachite leaching by citric acid was technically feasible. Optimum leaching conditions were found as follows： the range of particle size, 105-150 μm; acid concentration, 0.2 M; leaching time, 30 min; solid/liquid ratio, 1：20 g/mL; temperature, 40℃; and stirring speed, 200 r/min. Under the optimum conditions, 91.61% of copper was extracted.

Partially decoupling and collar bonding of the encapsulated rebar rockbolts to improve their performance in seismic prone deep underground excavations

Rockbolt is widely employed all over the world as an effective ground reinforcement element in order to secure the underground workplaces.Ordinary encapsulated rebar or rebar rockbolt is most popular and commonly used as reinforcement in a ground support system because of its accessibility,cost effectiveness and easy practicability.Reinforcement elements in a seismic condition such as rock burst have to dissipate the energy release of the dynamic impact via their deformation and ultimate load capacity,knowing that the former is more important.In other words,achieving early stiff behaviour along with large deformation capacity in rockbolts are the goals for new development in rock reinforcement.Yielding rockbolts are expensive while some of them have large deformation capability with low ultimate load capacity.In this paper,modifications were made on encapsulation of rebar rockbolts to utilise it effectively as a yielding reinforcement in seismic conditions.Applying a sufficient decoupled length in the shank of rebar rockbolts which industry has regularly been using to control the bulking of the stress fractured ground,improves the deformation capacity of the bolt.Additionally,leaving a collar bonding underneath of the bearing pad and plate removes the weaknesses of the head anchorage of rockbolt.Therefore the dynamic performance of the bolt is improved by these easily applicable modifications.The behaviour and performance of encapsulated rockbolts have been discussed first,then the effects of modifications are illustrated.The proposed modification of the rebars is not only cost effective but also easy to apply in the field and improves the performance of reinforcements in seismic prone zones.

A method for extracting anomaly map of Au and As using combination of U-statistic and Euclidean distance methods in Susanvar district,Iran

Quantitative descriptions of geochemical patterns and providing geochemical anomaly map are important in applied geochemistry. Several statistical methodologies are presented in order to identify and separate geochemical anomalies. The U-statistic method is one of the most important structural methods and is a kind of weighted mean that surrounding points of samples are considered in U value determination. However, it is able to separate the different anomalies based on only one variable. The main aim of the presented study is development of this method in a multivariate mode. For this purpose, U-statistic method should be combined with a multivariate method which devotes a new value to each sample based on several variables. Therefore, at the first step, the optimum p is calculated in p-norm distance and then U-statistic method is applied on p-norm distance values of the samples because p-norm distance is calculated based on several variables. This method is a combination of efficient U-statistic method and p-norm distance and is used for the first time in this research. Results show that p-norm distance of p=2(Euclidean distance) in the case of a fact that Au and As can be considered optimized p-norm distance with the lowest error. The samples indicated by the combination of these methods as anomalous are more regular, less dispersed and more accurate than using just the U-statistic or other nonstructural methods such as Mahalanobis distance. Also it was observed that the combination results are closely associated with the defined Au ore indication within the studied area. Finally, univariate and bivariate geochemical anomaly maps are provided for Au and As, which have been respectively prepared using U-statistic and its combination with Euclidean distance method.

Usage of Iranian scoria for copper and cadmium removal from aqueous solutions

The competitive removal of copper and cadmium from aqueous solutions using scoria has been investigated. Scoria was characterized by various methods, such as XRD, XRF, FT-IR and SEM. The results show that scoria sample contained augite, enstatite, diopside, and olivine. These minerals were separated from each other and each mineral was then subjected to the adsorption experiments. It was found that the main absorbent constituent in scoria was augite. Finally, statistical experimental method was used to optimization of adsorption conditions(Initial concentration of copper and cadmium ions, the amount of scoria and temperature) for removal of Cu(II) and Cd(II) ions from solution in optimum conditions. The optimum conditions are obtained as follows: concentrations of Cu(II) and Cd(II) of 400×10-6 and 554×10-6, respectively; amount of scoria of 7 g; temperature of 38 °C. Under these conditions Cu(II) and Cd(II) ions are absorbed onto the scoria more than 79% and 16%, respectively.

Bioleaching of copper oxide ore by P seudomonas aeruginosa

Bioleaching is an environmentally friendly method for extraction of metal from ores. In this study, bioleaching of copper oxide ore by Pseudomonas aeruginosa was investigated. Pseudomonas aeruginosa is a heterotrophic bactermm that can produce various organic acids in an appropriate culture medium, and these acids can operate as leaching agents. The parameters, such as particle size, glucose percentage in the culture medium, bioleaching time, and solid/liquid ratio were optimized. Optimum bioleaching conditions were found as follows： particle size of 150-177 μm, glucose percentage of 6%, bioleaching time of 8 d, and solid/liquid ratio of 1：80. Under these conditions, 53% of copper was extracted.

采用人工神经网络基因复合算法预测氧化铜矿柱浸工艺铜浸出率（英文）

采用人工神经网络(ANN)以及人工神经网络和基因复合(GANN)算法来优化氧化铜矿柱浸工艺参数。采用三种高度的浸矿柱(2,4,6 m)和尺寸为<25.4 mm和<50.8 mm的两种矿物来进行浸出实验。在台架实验规模下,对浸矿柱高度、矿粒尺寸、硫酸流速、浸出时间等工艺参数对铜浸出率的影响进行研究,对浸出条件进行优化以得到最大的浸出率。研究结果表明,铜的浸出率随硫酸流速和浸出时间的增加而增加,随矿粒尺寸和浸矿柱高度的减小而增加。对人工神经网络(ANN)、人工神经网络和基因复合(GANN)算法的效率进行了比较,结果表明,人工神经网络和基因复合(GANN)算法比人工神经网络(ANN)算法更有效。采用新提出的算法模型来预测铜的浸出率误差更低。

Lead and zinc geochemical behavior based on geological characteristics in Parkam Porphyry Copper System, Kerman, Iran

Parkam(Sarah) porphyry system is located on the metallogenic belt of Kerman, Iran. Due to existence of some copper-rich resources in this region, finding out the exact statistical characteristics such as distribution of data population, mean, variance and data population behavior of elements like Cu, Mo, Pb and Zn is necessary for interpreting their geological behavior. For this reason, precise calculation of statistical characteristics of Pb and Zn grade datasets was performed and results were interpreted geologically. The natures of Pb and Zn distributions were initially identified and their distributions were normalized through statistical treatment. Subsequently, the variograms were calculated for each exploration borehole and show that both Pb and Zn geochemical variates are spatially correlated. According to the similarity of the behavior of Pb and Zn in these calculations, it is decided to measure their exact behavior applying K-means clustering method. K-means clustering results show that the Zn grade varies linearly relative to that of Pb values and their behavior is similar. Based on the geochemical behavior similarity of Pb and Zn, throughout the pervasive secondary hydrothermal activity, they are remobilized in the similar manner, from the deep to the shallow levels of the mineralization zones. However, statistical analysis suggests that hydrothermal activity associated with secondary waters in Parkam is effective in remobilizing and enriching both Pb and Zn since they have similar geochemical characteristics. However, the process does not result in generation of economic concentrations.

Comparative and Competitive Adsorption of Cu（II） and Cd（II） Using Scoria： Equilibrium, Kinetic and Thermodynamic Studies

Heavy metals, such as Cu（Ⅱ） and Cd（Ⅱ） are among the hazardous pollutants that lead to severe ecological problems and have a toxic effect on living organisms. This study was carried out on adsorption of Cu（Ⅱ） and Cd（Ⅱ） from single-and multi-component aqueous solutions with Iranian scoria. Two- and three-parameter isotherm models, such as Langmuir, Freundlich, Temkin, Dubinin-Radushkevich, Khan and Toth have been studied for single and bi- nary adsorption of Cu（Ⅱ） and Cd（Ⅱ） onto scoria. The best result was attained from Langmuir model for Cu（Ⅱ） and Cd（Ⅱ） ions in single and binary solution. Therefore, homogenous adsorption is dominated, which was emphasized by three-parameter isotherm models. Based on the value of the free energy of adsorption for Cu（Ⅱ） and Cd（Ⅱ）, the inte-raction between these ions and scoria is a physical adsorption. In order to investigate competitive adsorption behavior, modified and extended Langmuir and Freundlich models have been studied, which indicated that adsorption of Cu（Ⅱ） and Cd（Ⅱ） on scoria has synergistic behavior. Extended Freundlich model and modified Langmuir model described Cu（Ⅱ） and Cd（Ⅱ） adsorption onto scoria, respectively.