Comparison of three artificial neural network approaches for estimating of slake durability index

Slake durability index (I _d2) is an important engineering parameter to assess the resistance of clay-bearing and weak rocks to erosion and degradation. Standard test sample preparation for slake durability test is difficult for some rock types and the test is time-consuming. The paper reports an attempt to define I _d2 using other parameters that are simpler to obtain. In this study, three different artificial neural network approaches, namely feed-forward back propagation (FFBP), radial basis function based neural network (RBNN), and generalized regression neural networks (GRNN) were used for estimating I _d2. The determination coefficient (R ²), root mean square error and mean absolute relative error statistics were used as evaluation criteria of the FFBP, RBNN, and GRNN models. The experimental results were compared with these models. The comparison results indicate that the GRNN models are superior to the FFBP and RBNN models in modeling of the slake durability index (I _d2).     

Creation of Histograms for Data in Various Mineral Resource and Engineering Problems: A Review of Existing Methods and a Proposed New Method to Define Bin Number

Histograms are widely used in geosciences for data analysis and visualization. In cases where a distribution is not fitted to data, histograms are often used to address various sampling- and interpolation-related aspects. However, the results of these applications are substantially affected by the histogram’s number of bins as determined by several binning methods. This paper proposes a new binning approach and compares it with various standard approaches to demonstrate the relative performance of the new approach. Cut-off grade optimization for polymetallic deposits, Monte-Carlo modeling, and derivation of conditional distribution, all of which use histograms, are used as case studies. The proposed technique is based on calculating the squared error for each bin in a histogram, and combining the error values to evaluate the total error for each histogram. The new technique then selects the bin number which minimizes the total error. The results showed that the new binning approach is well suited for binning small datasets and can be used in geoscience applications if needed.     

Heavy metal pollution and acid drainage from the abandoned Balya Pb-Zn sulfide Mine,NW Anatolia,Turkey

This study was conducted to determine the effects of the waste-rock dump (WRD) of the underground polymetallic Balya Mine on the Kocacay River and eventually on Lake Manyas in Turkey. Data presented in this paper include geochemical characteristics of various kinds of water (mine, surface and groundwater) and of suspended-particle samples in the vicinity of Balya. The more polluted mine waters have low pH and high conductivity, while high concentrations of Zn, Cd, Mn tend to be found in the dry and wet seasons. High concentrations of Pb, As, Cr, Cu and S appear only in the wet season. The sources of the heavy metal concentration within the Kocacay River are leached waste, surface run off, and overflow from the spillway of the WRD. To minimize the formation of acids and dissolved metal, and for the remediation of the harmful effects of extreme contamination conditions, it is recommended that lime or alkali materials and organic carbon be added to simulate the action of sulfate-reducing bacteria.     

Deformation of rubble-mound breakwaters under cyclic loads

Rubble-mound breakwaters usually consist of a core of small quarry-run rock protected by one or more intermediate layers or underlayers that separate the core from the cover layers, which are composed of large armor units. Failure of rubble-mound breakwaters may be due to effects such as removal or damage of the armor units, overtopping leading to scouring, toe erosion, loss of the core material, or foundation problems under waves. However, whether rubble mounds fail under seismic loads is unknown. High seismic activity can lead to large settlements and even to failure of the breakwaters. The design of coastal structures should take into account the most relevant factors in each case, including seismic loading. The objective of this study is to understanding the failure mechanisms of conventional breakwater structures under seismic loads on rigid foundations. Hence, an experimental study was carried out on conventional breakwater structures with and without toes, subjected to different dynamic loadings of variable frequencies and amplitudes, in a shaking tank. A shaking tank with a single degree of freedom was developed to study the simple responses of conventional rubble-mound breakwaters under cyclic loads. For each test, an automatic raining crane system was used to achieve the same relative density and porosity of the core material. The input motion induced horizontal accelerations of different magnitudes during the tests. The accelerations and the deformation phases of the model were measured by a data acquisition system and an image processing system. The experiments on the conventional rubble-mound type breakwater model were performed under rigid-bottom conditions. The model's scale was 1:50. Cyclic responses of breakwaters with toes and without toes were examined separately, and their behaviors were compared. The results were compared with a numerical study, and the material properties and failure modes were thus defined.     

Earthquake performance improvement of low rise RC buildings using high strength clay brick walls

Large number of vulnerable reinforced concrete (RC) buildings exists in earthquake prone areas. These low cost residential and/or commercial buildings, which are three to seven-stories high, usually do not receive essential engineering services during the construction phase. Finding cheap, easily applicable and occupant friendly retrofitting techniques are extremely important to reduce the seismic risk of these buildings. As an attempt to this, a particular type of high strength clay brick is studied to evaluate its potential for the structural retrofitting. A set of experiment was conducted to assess the important mechanical characteristics of the infill walls made from these bricks. Also the performance of two RC frames retrofitted with these walls, having different connection details between the wall and RC members was examined experimentally. The analytical nonlinear static analyses of these specimens have been performed using SeismoStruct to achieve some model parameters for representing the “infill wall model” in the program. Adaptive pushover and nonlinear time history analyses were conducted to investigate the performance of a six storey representative RC frame retrofitted with these walls. Evaluation of the results obtained in these analyses prove that this retrofitting technique introduces important strength and stiffness increments to the structure, regarding its seismic demands, which are similar to the results obtained from the experiments.     

Comprehensive study on automation of toxic gas measurement

The condition of the ambient air must be continuously checked to detect hazardous gas leakage and ensure the safety of employees in hazardous gas production and storage areas. Detection and elimination of hazardous gas leaks can be done with various complex measurement, safety, and disposal systems. These measurement systems consist of different types of sensor and require control and monitoring system. This leads to high investment and operating costs. In this study, semi-continuous measurement system automation with a single set of sensors working by taking air samples from different points over specific periods has been designed to reduce investment, operation, and maintenance repair costs. In the design, the measurement points are divided into zones and codes in the automation system. Thus, it is aimed to keep a constant record of what the air sample is coming from and what its content is. As a result, it has been determined that it is possible to establish a measurement and automation system similar to existing hazardous gas measurement systems with less investment and operating costs. The proposed automation system has been applied in the industrial storage area. In addition, a cost analysis has been performed and compared with conventional systems.