جستجوی مقالات مرتبط با کلیدواژه "طبقه بندی" در نشریات گروه "مهندسی معدن"

The purpose of grouting is to strengthen and improve the mechanical and hydraulic properties of the rock and soil. The fluid that is injected into the cavities and fissures of the environment is like a viscous liquid consisting of grains whose size is important in the grouting operation. Therefore, determining the groutability ratio in grouting operation is considered as an important parameter. Today, studies using data mining science show that the groutability of granular soils, in addition to grain size, is affected by various factors of the soil and the material of grout, which predicts groutability more accurately. Throughout history, many researchers have predicted groutability through experimental relationships. However, today, the capability of data mining methods in accurate predictions has shown that one approach in predicting groutability is to use a variety of data mining models and inferential systems.

Methodology and Approaches:

The purpose of this study is to evaluate several models of data mining methods, including ANN, SVM, KNN, RF and NB. For this purpose, a set of laboratory information related to groutability has been used in four literatures that include 87 data in order to develop efficient models for predicting groutability. Classification models are created in Orange software.

The output variable is a property of groutability, which as a binary variable has two states of zero meaning nongroutable and 1 meaning groutable. Input variables also include the ratio of water cement in the grout or viscosity (W/C), the relative density of the soil (Dr), grouting pressure (P), the percentage of the soil particles passing through a 0.6 mm sieve (FC), N1 = D15soil / D85 grout and N2 = D10 soil / D95 grout. The values of the evaluation criteria for the methods are almost close to each other. Based on the AUC index, the random forest is the best model and the k-nearest neighbor method has the lowest value of this index. However, in terms of other criteria, the artificial neural network is higher than other methods and the k-nearest neighbor method is very close to it. On the other hand, the random forest model has the lowest value of criteria. Ignoring the AUC criteria, ANN and KNN methods are the best methods.One of the capabilities of Orange software is to study the effect and importance of input variables on the prediction of the target variable, in other words, the sensitivity of the output variable to input variables. The results show that variable N2 is in the first level based on the three criteria of information gain, relative information gain and Gini index, and variable N1 is in the second level with a very small difference in the values of the criteria. In addition, in the last row, W/C has the lowest value of the criteria and shows a small role in the correct prediction of groutability.

Froth recognizing as the output of the flotation process can be efficient in better control of this process. Since the physical structure of the froth indicates the operational conditions of the flotation cell, the quality of the separation can be estimated according to its characteristics. Technological advances have led to monitoring and control operations using online systems. Image analysis method is one of the growing methods in this field. In addition to control the structural changes of the concentrate, using a device that can measure the quality of the froth online and report it regularly can keep the amount of chemicals consumed in the cell at optimal levels, and provide stability in the separation efficiency of the flotation unit. In this paper, changes in the froth structure of cleaner cell of Sungun copper processing plant were investigated. The area, circumference, elongation and diameter of the bubbles and their distribution, which are structural features of the froth, were calculated online to control flotation cell changes and used to classify the froth. Froth images were clustered in three classes of dry, wet and stiff froth using K-means algorithm and two characteristics of bubble fret area and diameter. The results showed that in dry froth, bubbles of froth surface are in a wider range of dimensions. This type of froth is in optimal condition in terms of load, mobility, stability and structure, also the grade of this type of froth is in the range of 25 to 29. Though, unlike dry froth, wet and hard froths are not in optimal condition in terms of bubble load, mobility, stability and structure due to the use of more or less excessive chemical additives. Also, the grades of wet and hard froths are in the range of 22 to 27 and 20 to 24, respectively.

Due to the role of recovery in calculating the economic value of ore blocks and the impact of the block's economic value on the design calculations of the final pit and production planning, determination of the amount of metal recovery from the ore material sent to the processing plant is very important. The aim of this study is to investigate the capability of estimating the recovery rate of ore in qualitative manner with three methods based on data classification from data mining techniques and quantitatively using multivariate regression and artificial neural networks. Hence, the Miduk copper mine was studied using 58 analyzed samples of the feed of the plant, including Cu, CuO and CuS grades, and the recovery rate of Cu in the final product of the plant. The process of predicting the total recovery of the reserve was made qualitatively by decision tree method, classification based on Bayes rule and k-nearest neighbor (kNN) classification algorithm. For quantitative estimation of recovery, multivariate regression and artificial neural network models were established between the mentioned grade parameters and recovery rates (For 47 samples of 58 samples) and with the 11 additional analyzed samples, the obtained models were validated. The coefficient of (R2), Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) in the regression model were 0.77, 0.027722 and 0.029722, respectively, and in the artificial neural network model, 0.82, 0.015753 and 0.024040, respectively. Therefore, the artificial neural networks model acts as a more accurate tool for predicting recovery versus the multivariable regression model. The results of sensitivity analysis of artificial neural network model showed that Cu grade is the most important factor and grade of CuO and CuS, respectively, as well as other factors influencing the changes in recovery rate.

Nowadays, very important issues are considered by geological and mining engineers in extraction process of minerals. The important goals are: productivity and high production rates, improve working conditions and lower operating costs. The most suitable option to achieve these goals is to replace the traditional extraction methods with full-mechanized methods. The mechanization of the extraction operations is time consuming process and it requires high investment costs. So, cause special sensitivity in all design and implementation phases. Therefore, it is important to design, investigate and identify the capability of mechanization of coal layers in the first stage. For this purpose, the engineering capabilities of the coal layers of different mines should be analyzed that is important and necessary to achieve the above goals. Therefore, criteria affecting the mechanization of coal layers were identified. Then, 17 Stopes were studied that located in the Eastern Alborz Basin and Tabas mine. The combined used method to evaluate mechanization capability is combination method of Fuzzy set theory and FDAHP. The results of the fuzzy classification system presented in current study show that the T layer of Tabas has a good quality, the K11 layer of the Takht mine is located in the middle class and other layers have not qualify for mechanization properly. In the following, the results of this study were validated with current mining conditions. The results of the study showed that the multifactorial fuzzy classification system can reliably use to classify the mechanization of coal layers based on geological conditions.

Since، in geomechanical evaluation of rock masses، many parameters affect the rock mass behavior simultaneously، rock engineering classifications are suitable approaches for studying and prediction of rock mass behaviors. Classic classifications in rock engineering have suffer from some practical limitations، particularly. These limitations are more obvious in when classification is made near the boundaries. y conditions. In this study in order to increase the abilities and expending the applications of rock mass drillability index (RDi)، fuzzy theory has been used. For this purpose، six parameters of the rock mass which are used in RDi classification، including uniaxial compressive strength (UCS)، joints dipping، Mohs hardness، joints aperture، joints spacing and grain size have been used. Then، some fuzzy functions have been defined on these parameters and finally the class of each rock mass has been identified. In order to compare the results of classic classification with results of fuzzy classification، a case study was done on rock masses of limestone mine of Shahrood cement factory. The results show that the fuzzy logic based classification produces clearer better results than classic system especially in rock masses with boundary condition. During this research، a software was prepared for doing theto calculate the RDi scoresions and classifying the rock mass. Theis program has been written by coded in Visual C++ and has a contains a graphical main window thatinterface with all input parameters which are related (rock mass characteristics) and also the out output parameters of the program (which are RDi score and class of rock mass). are illustrated in it.

In Aghdareh gold processing plant, grinding circuit product size has a significant effect on downstream gold leaching process. Therefore, the performance optimization of hydrocyclone package in order to decrease the size of overflow stream to less than 70 μm (present value) will considerably increase the gold recovery. In this research, optimization of Aghdareh’s hydrocyclone package performance was done by application of Plitt’s model and its calibration and simulation of classification process. Two plant sampling campaigns were carried out to collect samples from the streams around the hydrocyclone package. Screen analysis tests were performed to obtain particle size distributions (PSD). Then, measured PSD and flow rates of various streams were adjusted using NorBal data reconciliation software. Then, BMCS (BMCS-based Modular Comminution Simulator) was used to predict the PSD of overflow and underflow streams for the first sampling campaign and also calibrating Plitt’s model to operating Aghdareh hydrocyclone package. Then, BMCS simulation predictions based on the calibrated Plitt’s model were validated using the second sampling campaign. The simulations showed that by using vortex and spigot internal diameters equal to 75 and 70 mm, respectively and with 6 operating hydrocyclones, the P80 of the overflow product can be decreased to 68.5 μm. Also, simulations by using vortex and spigot internal diameters equal to 94 and 53, respectively with 6 operating hydrocyclones showed that the separation sharpness, m, can be increased from 1.18 to 2.81 which indicates a 2.38 times increase in classification accuracy.

Marl units generate the most sediment yield rate among the geological formations in the watershed basins. Therefore, they notably affect on useful lifetime of dams and water electric constructions, flooding processes, sedimentation in the water transport channels and the existence of aqueous creatures. Study and classification on the effective characteristics of erodibility of marls could notably help to remediation activities on them. In this research, the amount of sediment yield in each marl units were compared using a portable rainfall simulator at field. The amounts of runoff and produced sediments in each marl units at three repetitions were measured. The relation between physico- chemical characteristics of marl units as independent variables and the amounts of produced sediments as dependant variables was analyzed using multivariate linear regression method. Some characteristics such as SAR, EC, and K show 89, 3.5 and 1.5 percent of total variance of produced sediments at each marl units respectively. On the basis of ANOVA analysis of sediment data, the marl units have classified in four groups: A (include of Ols unit), B (include M3 unit), C (include M1 and M2) and D (include Olg and O-M units). The erodibility rate and sediment yield decrease from A to D group. Physico- chemical characteristics of four above mentioned groups have analyzed using variance analyze. The results show significant difference at level 0.5,. 01, and. 001 for some parameters such as SAR, EC, K+1, Na+1, Cl-1. Meanwhile, the mean amounts of sediment yield of various marl units obtained from rainfall simulator test show that evaporative marls of Neogene age (M1, M2 and M3) potentially produce sediment 27 to 100 times more than Qom formation marine marls and Kond formation lacustrine marls.