Search

There has been in recent years a trend towards adopting deep neural networks for addressing earth science problems. Of the various deep neural networks applied to different problems in earth sciences, this study aimed to demonstrate how to apply the group method of data handling (GMDH) neural networks for mineral prospectivity mapping (MPM). GMDH neural networks are sophisticated, multilayered, robust tools for addressing complex regression problems. However, labeled data for MPM, which constitute multivariate attributes of known deposit and non-deposit sites, are often (if not always) insufficient to train neural networks adequately. Such issue triggers networks' poor generalization; that is, the networks developed fit the labeled data perfectly (i.e., low bias) but cannot predict unseen data accurately (i.e., high variance). Given this, GMDH neural networks were, in this study, coupled with a window-based data augmentation technique, an approach to generate additional geologically constrained labeled samples for MPM, and applied to a district hosting few but giant porphyry copper deposits. It was recognized that coupling the data augmentation technique with GMDH neural networks yielded robust predictive models that can handle the bias–variance tradeoff, making this combined methodology a viable option for MPM in terrains that host few but large mineral deposits.

An Nb-Ti low carbon HSLA steel with the application of Coiled Tubing steel in petroleum industry, was subjected to hot rolling followed by the subsequent cooling procedures including cooling and coiling. The effect of cooling rate and coiling temperature on final microstructure and texture was investigated via optical microscopy, SEM, and EBSD analyses. It was found that rolling at 800 °C results in the appearance of elongated ferrites due to the activation of the deformation-induced ferrite transformation mechanism. Cooling from the finish rolling with different rates of 2.4 to 15 °C/s leads to the formation of polygonal ferrite and retained austenite. Higher cooling rate increases the retained austenite, decreases the ferrite grain size and improves the grain size homogeneity. Coiling at 600 °C results in coarser ferrite with the texture of γ fiber which is due to the recrystallization of ferrite. On the other hand, coiling at 500 °C intensifies the {332} $$\left\langle {113} \right\rangle$$ and {113} $$\left\langle {110} \right\rangle$$ components indicating deformed austenite transformed to ferrite.

A dislocation assisted self-consistent model based on Tandon and Weng approach and Bergstrom dynamic recovery model for particulate-reinforced composites has been extended to consider the matrix evolution during high-temperature deformation on flow stress. The impact of main influential processing parameters such as temperature, strain, and strain rate in addition to reinforcement characteristics, including particle size, and volume fraction, were successfully taken into account in the constitutive model. Moreover, the effect of particle fracture, diffusion relaxation around particles, dynamic recrystallization, and dynamic recovery of the matrix during deformation were precipitated as the softening factors in the presented model. It was found that the occurrence of particle stimulating nucleation mechanism can destroy the load transfer mechanism, which results in flow curve softening for a limited range of deformation conditions. It was shown that deformation mechanisms in single-phase alloy and metal matrix composite are the same, which are viscous glide and dislocation climb.

Geochemical sampling media, be they rock chip, soil or stream sediment samples, routinely collected in mineral exploration, are key to detecting mineralisation-related geochemical dispersion patter...

The effects of hot rolling and homogenisation treatment on the microstructure and mechanical properties of low-alloy transformation-induced plasticity (TRIP) steel were studied. With increasing the...

Stream sediments remain the preferred geochemical sampling media for regional mineral exploration programs in most parts of the world where there is a well-defined drainage system. In Au exploration the analysis of either Bulk Leach Extractable Gold (BLEG) or acid-extractable Au in the −80# ( mineral deposit types and geological settings. Using a geochemical database derived mainly from company exploration programs, a study is made on the efficiency of singularity mapping technique in detecting the known mineral deposits or prospects. The composite media of samples are representative of sampling locations but also are extremely diluted. Therefore, recognition and mapping of significant geochemical anomalies of composite media still remains a challenge. In this study, geochemical assay results were subjected to multifractal analyses to detect gold geochemical anomalies, which are related to gold mineralization . In this regard, a singularity mapping technique was applied to enhance the single element geochemical signatures of BLEG sand −80# stream sediment data. Since the singularity mapping technique does not define threshold values for delineation of anomalies, number-size (N-S) and concentration-area (C-A) fractal models were applied to delimit the geochemical anomalies as target areas for further exploration. A hundred percent of the known gold deposits and prospects discovered in the study area are delineated by BLEG Au stream sediment anomalies by N-S and C-A multifractal modeling. Moreover, 89% of known gold prospects of the study area are delineated within the recognized anomalous zones of Au in −80# stream sediment data. The comparison of the delineated geochemical anomalies of BLEG and −80# stream sediment with those of rock chip samples demonstrated the efficiency of the applied methodology.

Aluminum rod (AA1100) is severely plastic-deformed by one pass of the ECAP-Conform method. Optical and electron backscatter diffraction (EBSD) methods are used for microstructural characterizations. EBSD examinations of in process and processed samples reveal that grain sizes are not changed appreciably and they are mostly elongated by the plastic deformation. The EBSD data shows an increase in the number of subgrains and inhomogeneous dislocation distributions. Estimation of the geometrically necessary dislocation (GND) densities is done by using EBSD data. Estimated average GND density of the annealed material is as low as 5 × 107 m−2. For severely deformed Al specimen, the estimated GNDs and SSDs densities are 2.9 × 1014 m−2 and 4.6 × 1014 m−2 respectively with an averaged total dislocation density of 7.5 × 1014 m−2. Mechanical properties of samples at different stages of deformation are examined through hardness and tensile tests. Microhardness results show the inhomogeneous distribution of hardness at these sections due to the complex and varied loading conditions. Despite hardness inhomogeneity, macro tensile test results show that ECAP-Conform process is very effective in increasing the strength of the material at least by the factor of three. This increase in strength can be well interpreted by an increase in dislocation densities and formation of subgrains.

The microstructure evolution of Al-3wt.%Mg reinforced with 10¿vol% B4C during isothermal compression at temperatures ranging 300–500¿°C at strain rates of 0.001–10 s-1 was investigated by electron backscatter diffraction (EBSD). According to the results, at strain rates lower than 0.01 s-1 and temperatures higher than 400¿°C, the grain size distribution in the microstructure is uniform, dynamic recovery is the predominant softening mechanism and continues recrystallization through lattice rotation is responsible for grain refinement. However, during deformation at higher strain rates or lower temperatures, deformation zones appeared in special locations around particles where microstructure is formed by recovered and hardened grains, and particle stimulating nucleation leaded to partially discontinues dynamic recrystallization which in turns promoted finer average grain and sub-grain size than those in single phase Al-Mg alloy. Moreover, it was found that the variation of grain and sub-grain size with deformation parameters (Zener-Holloman parameter (Z)) can be described by a power law type equation rather than by an initially expected exponential expression.

Dynamic recrystallization (DRX) grain size of a Nb-Ti microalloyed steel is investigated at various strain rates and temperatures. Electron microscopy reveals that Nb preferably precipitates on TiN particles at temperature range of T

The Behabad district, located in the central Iranian microcontinent, contains numerous epigenetic stratabound carbonate-hosted Zn-Pb ore bodies . The mineralizations formed as fault, fracture and karst fillings in the Permian-Triassic formations, especially in Middle Triassic dolostones , and comprise mainly non-sulfides zinc ores . These are all interpreted as Mississippi Valley-type (MVT) base metal deposits. From an economic geological point of view, it is imperative to recognize the processes that have plausibly controlled the emplacement of MVT Zn-Pb mineralization in the Behabad district. To address the foregoing issue, analyses of the spatial distribution of mineral deposits comprising fry and fractal techniques and analysis of the spatial association of mineral deposits with geological features using distance distribution analysis were applied to assess the regional-scale processes that could have operated in the distribution of MVT Zn-Pb deposits in the district. The obtained results based on these analytical techniques show the main trends of the occurrences are NW–SE and NE-SW, which are parallel or subparallel to the major northwest and northeast trending faults, supporting the idea that these particular faults could have acted as the main conduits for transport of mineral-bearing fluids. The results of these analyses also suggest that Permian-Triassic brittle carbonate sedimentary rocks have served as the lithological controls on MVT mineralization in the Behabad district as they are spatially and temporally associated with mineralization.

The hot flow behavior of an Nb-Ti microalloyed steel is investigated through hot compression test at various strain rates and temperatures. By the combination of dynamic recovery (DRV) and dynamic recrystallization (DRX) models, a phenomenological constitutive model is developed to derive the flow stress. The predefined activation energy of Q = 270 kJ/mol and the exponent of n = 5 are successfully set to derive critical stress at the onset of DRX and saturation stress of DRV as functions of the Zener–Hollomon parameter by the classical hyperbolic sine equation. The remaining parameters of the constitutive model are determined by fitting them to the experiments. Through substitution of a normalized strain in the DRV model and considering the interconnections between dependent parameters, a new model is developed. It is shown that, despite its fewer parameters, this model is in good agreement with the experiments. Accurate analyses of flow data along with microstructural analyses indicate that the dissolution of NbC precipitates and its consequent solid solution strengthening and retardation of DRX are responsible for the distinguished behaviors in the two temperature ranges between T

Recognition of significant ore-forming processes, which control mineralization, improves the efficiency of mineral prospectivity modeling. In this study, controlling processes of skarn copper mineralization in Varzaghan district, northwestern Iran, were distinguished by a series of spatial and numerical analyses comprising point pattern, fractal, fry and distance distribution methods. The recognized processes were then translated to a set of exploration criteria of the deposits in the area. Based on the accomplished exploration criteria, two data-driven models of skarn copper prospectivity were generated using logistic regression and random forest techniques. The comparison of two generated models demonstrated that the targets derived by the latter technique were more reliable for further exploration than those created by the former one.

This study intends to (i) identify the potential sources of heavy metal (HM) pollution in an industrial town situated in the northwestern part of Iran and (ii) and assess whether the pollution levels discharged by HMs are significant. A suite of geochemical samples collected from the topsoil horizon of the study area were analyzed for HMs, namely Cr, Ni, Co, Cu, V, Zn, Mn, and Pb, to attain the objectives mentioned above. Compositional data analysis (CODA) techniques were employed for pinpointing the potential source of HM pollution. Initially, the centered log-ratio (clr) transformation was applied to the dataset containing raw values of chemically analyzed soil samples to address the data closure problem. Multifractal moving average interpolation was then applied to the clr-transformed data to model the spatial distribution of HMs in the study area. Maps of the spatial distribution of HMs in the study area were further classified by the concentration-area (C-A) fractal technique for identifying the potential hotspots of HM pollution in the area. It was followed by using the K-means clustering technique for classifying samples into distinct classes. This method was selected for geochemical pattern recognition owing to its simplicity and robustness in identifying subtle patterns embedded in geochemical data. The above hybrid methodology points to three anthropogenic sources discharging HM pollution into the area: leather factories, brickworks, and highways. Finally, pollution indices, namely the geo-accumulation index, pollution load index, and potential ecological risk, revealed that the elevated levels of Ni, Cr, and Co can pose serious impediments to the soil quality of the study area.

Data-driven mineral prospectivity mapping (MPM) with random forests (RF) has been documented in various brownfield zones hosting known mineral deposits of the type being sought. However, the practical implementation of RF-based MPM is crippled by uncertainties stemming from the exploration targeting criteria used as predictors for this algorithm; some exploration targeting criteria are poorly linked to known mineral deposits. Aiming to modulate the effects of this type of uncertainty in RF-based MPM, this study employs three parameters, namely the Gini impurity index, least-squared errors, and receiver operating characteristics curves, for sensitivity analysis of exploration targeting criteria. These were used to assess whether the exclusion or inclusion of specific targeting criteria can improve the results of RF-based MPM. Sensitivity analysis was applied to a set of mineral systems-derived exploration targeting criteria in the Ahar-Arasbaran area, located in the Alborz–Azarbaijan magmatic belt (AAMB) of northern Iran. It is demonstrated here that the exploration targeting criteria that are poorly linked to known deposit sites deteriorate the performance of RF-based MPM, the exclusion of which can significantly improve the performance of MPM. The sensitivity analysis adopted in this paper reduces the search space for complimentary surveys and improves the correlation of models generated with known mineralized sites; consequently, one may achieve an improved performance of data-driven MPM by applying the strategy employed in this study.

Mineral Prospectivity Mapping (MPM) is a multifaceted process relying heavily on experts' judgments. Notwithstanding the importance of human interpretations and cognitive knowledge in the success of exploration projects, human input, granted, is material to cognitive heuristics, subjectivity, and mental traits of geologists involved in the various stages of MPM. Knowledge-driven MPM used in greenfield areas – where few or no mineral deposits are known – use experts' opinions for assigning weights to exploration targeting criteria. This issue introduces a systemic uncertainty that eventually propagates to the generated targets. This study, therefore, intends to propose a methodology for modulating the effects of this type of uncertainty in knowledge-driven MPM. Aiming to attain this objective, a procedure combining Monte Carlo simulation with fuzzy logic was articulated and applied to a suite of mineral systems-derived targeting criteria derived by geochemical and geological data in a case study. The proposed procedure returns three components, namely (a) the modulated prospectivity model, (b) uncertainty, and (c) confidence. In this method, plots of uncertainty and confidence versus prospectivity values are used for target generation; low-risk targets are those marked by low uncertainty, high confidence, and high prospectivity values. In this study, low-risk targets occupy merely 0.5% of the study area, showcasing the applied framework's efficacy for reducing the search space in greenfield exploration.

This study intends to showcase the application of Extreme Gradient boosting (XGboost), a state-of-the-art ensemble-learning technique, for district-scale mineral potential mapping (MPM) of carbonate-hosted Zn Pb mineral systems in the Emarat district, W Iran. Notwithstanding the demonstrated effectiveness of XGboost in addressing a range of classification and regression problems, the inadequate number of deposit locations employed as labeled samples, which is intrinsic to district-scale MPM, impedes the practical application of this method. Given this notion, an iterative window-based data augmentation technique was proposed and applied to generate additional labeled samples. The predictive models derived by XGboost were compared to those generated by random forests (RF). Both methods were applied to a set of mineral systems-derived exploration targeting criteria representing critical ore-forming processes. Setting aside the similarities of predictive models derived by both methods, it was recognized that RF is much less sensitive to insufficient labeled samples. It was also recognized that, unlike RF, many parameters require tuning for generating an optimum XGboost-based predictive model. Nevertheless, given the achieved results, XGboost slightly outperformed RF in generating an optimum predictive model. Therefore, XGboost coupled to data augmentation can be deemed a viable alternative for district-scale, data-driven MPM.

In this study, fractal/multifractal modeling methods have been applied for preliminary hydrothermal Ni prospection using a set of stream sediment geochemical data of Ahangaran district, which is an arid terrain in eastern Iran. The study area has a complex geological and morphological setting, which is dominated by the presence of aeolian sands. Cretaceous ophiolitic rocks hosting nickel endowment are influenced by extensive listwanitic alteration. Hydrothermal nickel mineralization occurs as veinlets and disseminated forms within the listwanite units. Among the analyzed elements, Ni, Cr and, Co have been recognized as being associated with listwanitic alteration and Ni mineralization, through robust factor analysis of compositional data. The regional background patterns of these mineralization-related elements were delineated via the concentration – area (C-A) fractal modeling method. Different multifractal characteristics of the selected elements were recognized using the multifractal spectrum f(α) curves and different multifractal indices. The enrichment and depletion patterns of the related elements in different parts of the study area were also identified using the local singularity mapping technique. The application of local singularity mapping manifested different geochemical patterns that were not recognized via the C-A fractal modeling method. The superiority of the local singularity mapping technique in delineating geochemical populations over the C-A fractal modeling method, was revealed by quantification of the correlation between geochemical anomalies and geological evidence of the mineralization.

The stability of reversely formed austenite and related mechanism of transformation were investigated against temperature and time in an Fe-9.6Ni-7.1Mn (at. pct) martensitic steel during intercritical annealing at a dual-phase (α + γ) region. Dilatometry, electron backscattering diffraction (EBSD), atom probe tomography (APT), and X-ray diffraction (XRD) were used to characterize the mechanism of reverse transformation. It was found that under intercritical annealing at 853 K (580 °C), when the heating rate is 20 K/s (20 °C/s), reverse transformation takes place through a mixed diffusion control mechanism, i.e., controlled by bulk diffusion and diffusion along the interface, where Ni controls the diffusion as its diffusivity is lower than that of Mn in the martensite and austenite. Increasing the intercritical annealing to 873 K (600 °C) at an identical heating rate of 20 K/s (20 °C/s) showed that reverse transformation occurs through a sequential combination of both martensitic and diffusional mechanisms. The transition temperature from diffusional to martensitic transformation was obtained close to 858 K (585 °C). Experimental results revealed that the austenite formed by the diffusional mechanism at 853 K (580 °C) mainly remains untransformed after cooling to ambient temperature due to the enrichment with Ni and Mn. It was also found that the stability of the reversely formed austenite by martensitic mechanism at 873 K (600 °C) is related to grain refinement.