Katherine L. Silversides

Automated recognition of stratigraphic marker shales from geophysical logs in iron ore deposits

The mining of stratiform ore deposits requires a means of determining the location of stratigraphic boundaries. A variety of geophysical logs may provide the required data but, in the case of banded iron formation hosted iron ore deposits in the Hamersley Ranges of Western Australia, only one geophysical log type (natural gamma) is collected for this purpose. The information from these logs is currently processed by slow manual interpretation. In this paper we present an alternative method of automatically identifying recurring stratigraphic markers in natural gamma logs from multiple drill holes.Our approach is demonstrated using natural gamma geophysical logs that contain features corresponding to the presence of stratigraphically important marker shales. The host stratigraphic sequence is highly consistent throughout the Hamersley and the marker shales can therefore be used to identify the stratigraphic location of the banded iron formation (BIF) or BIF hosted ore.The marker shales are identified using Gaussian Processes (GP) trained by either manual or active learning methods and the results are compared to the existing geological interpretation. The manual method involves the user selecting the signatures for improving the library, whereas the active learning method uses the measure of uncertainty provided by the GP to select specific examples for the user to consider for addition.The results demonstrate that both GP methods can identify a feature, but the active learning approach has several benefits over the manual method. These benefits include greater accuracy in the identified signatures, faster library building, and an objective approach for selecting signatures that includes the full range of signatures across a deposit in the library. When using the active learning method, it was found that the current manual interpretation could be replaced in 78.4% of the holes with an accuracy of 95.7%. We apply Gaussian Processes to identify marker shales in iron ore mines.We examine manual and active learning methods of training Gaussian Processes.The active learnings use of uncertainty increases the accuracy of results.The active learning method can replace more than 3/4 of the manual interpretations.

Detection of geological structure using gamma logs for autonomous mining

This work is motivated by the need to develop new perception and modeling capabilities to support a fully autonomous, remotely operated mine. The application differs from most existing robotics research in that it requires a detailed world model of the sub-surface geological structure. This in-ground geological information is then used to drive many of the planning and control decisions made on a mine site. This paper formulates a method for automatically detecting in-ground geological boundaries using geophysical logging sensors and a supervised learning algorithm. The algorithm uses Gaussian Processes (GPs) and a single length scale squared exponential covariance function. The approach is demonstrated on data from a producing iron-ore mine in Australia. Our results show that two separate distinctive geological boundaries can be automatically identified with an accuracy of over 99 percent. The alternative approach to automatic detection involves manual examination of these data.

A Dynamic Time Warping based covariance function for Gaussian Processes signature identification

Modelling stratiform deposits requires a detailed knowledge of the stratigraphic boundaries. In Banded Iron Formation (BIF) hosted ores of the Hamersley Group in Western Australia these boundaries are often identified using marker shales. Both Gaussian Processes (GP) and Dynamic Time Warping (DTW) have been previously proposed as methods to automatically identify marker shales in natural gamma logs. However, each method has different advantages and disadvantages. We propose a DTW based covariance function for the GP that combines the flexibility of the DTW with the probabilistic framework of the GP. The three methods are tested and compared on their ability to identify two natural gamma signatures from a Marra Mamba type iron ore deposit. These tests show that while all three methods can identify boundaries, the GP with the DTW covariance function combines and balances the strengths and weaknesses of the individual methods. This method identifies more positive signatures than the GP with the standard covariance function, and has a higher accuracy for identified signatures than the DTW. The combined method can handle larger variations in the signature without requiring multiple libraries, has a probabilistic output and does not require manual cut-off selections. A DTW based covariance function is developed for Gaussian Processes.We apply this method to identify marker shales in iron ore mines.The results are compared to those using only Gaussian Processes or Dynamic Time Warping.The new function combines the flexibility of the DTW with the probabilistic framework of the GP.

Robust Library Building for Autonomous Classification of Downhole Geophysical Logs Using Gaussian Processes

Machine learning techniques such as Gaussian Processes can be used to identify stratigraphically important features in geophysical logs. The marker shales in the banded iron formation hosted iron ore deposits of the Hamersley Ranges, Western Australia, form distinctive signatures in the natural gamma logs. The identification of these marker shales is important for stratigraphic identification of unit boundaries for the geological modelling of the deposit. Machine learning techniques each have different unique properties that will impact the results. For Gaussian Processes (GPs), the output values are inclined towards the mean value, particularly when there is not sufficient information in the library. The impact that these inclinations have on the classification can vary depending on the parameter values selected by the user. Therefore, when applying machine learning techniques, care must be taken to fit the technique to the problem correctly. This study focuses on optimising the settings and choices for training a GPs system to identify a specific marker shale. We show that the final results converge even when different, but equally valid starting libraries are used for the training. To analyse the impact on feature identification, GP models were trained so that the output was inclined towards a positive, neutral or negative output. For this type of classification, the best results were when the pull was towards a negative output. We also show that the GP output can be adjusted by using a standard deviation coefficient that changes the balance between certainty and accuracy in the results.

Identification of marker shale horizons in banded iron formation: linking measurements of downhole natural gamma-ray with measurements from reflectance spectrometry of rock cores

Marker shale horizons are used in stratigraphic interpretation and for modelling and mining of the banded-iron-formation-hosted iron ore deposits in the Hamersley Province in the Pilbara region of Western Australia. These deposits contain marker shale horizons that are highly consistent throughout the Hamersley Province and are used to define lithology and provide context to enable the separation of visually similar rock units. The locations of these shales are normally determined from natural gamma-ray logs in exploration holes, which provide a coarse resolution boundary due to their wide spatial separation. Hyperspectral imagery can provide detailed information at high spatial resolution on the location and strike of shale horizons as they are presented on benches and walls in open-pit mines. The relationship between measurements of downhole gamma and shale horizons as detected by hyperspectral imagery must be understood if these very different kinds of measurements are to be used in a complementary way. The ability to identify marker shale horizons using their spectral features would allow them to be mapped using hyperspectral imagery. Multivariate analysis of shales from a typical Marra Mamba deposit showed absorption features relating to ferric iron, OH, H2O, and Al-OH, which allowed different marker shales to be spectrally separated. The kaolinite and Al2O3 abundance was estimated from the hyperspectral data using the intensity of the feature at 2202 nm to within 5.8% and 2.6%, respectively. Comparison of measurements of downhole gamma-ray with a proportion of kaolinite estimated from hyperspectral data showed both similarities and differences that required examination. Peaks in the gamma were located at the same depth in the core as peaks in the proportion of kaolinite. However, the relative magnitudes of these peaks were not consistent. Additionally, the gamma-ray baseline measurements between peaks varied at different depths in the core, with some sections of the core having a much higher baseline than others. This was not true for estimated kaolinite. These differences would lead to different interpretations of the distribution of material types down the hole. Similarities in the location of peaks in both types of data provide a basis to link these very different types of data to improve the resolution and accuracy of ore body boundary models.

Mineralogy identification through linear unmixing of blast hole geochemistry

ABSTRACT Validating and updating geological models is important in mining to reduce the errors between the predicted and mined values. In the Banded Iron Formation-hosted iron ore deposits in the Hamersley Ranges Western Australia, geological models are based on widely spaced exploration data, resulting in errors on smaller scales. Closely spaced chemical assays from production blast hole drilling become available for updating models, but do not routinely provide mineralogy. This study identifies mineralogy by applying linear unmixing methods to chemical assays. Linear spectral mixture analysis (LSMA), extended linear mixing model (ELMM) and spectral unmixing within a multi-task GP framework (SUGP) successfully identified the mineralogy in synthetic mixtures, with RMSEs of 13.75%, 11.78% and 11.97%. LSMA had consistently poorer results. SUGP was more accurate when using fewer endmember examples. ELMM was easier to train and required less processing power. Therefore either SUGP or ELMM should be chosen to identify the mineralogy depending on the application.

Gaussian Processes Based Fusion of Multiple Data Sources for Automatic Identification of Geological Boundaries in Mining

Mining stratified ore deposits such as Banded Iron Formation (BIF) hosted iron ore deposits requires detailed knowledge of the location of orebody boundaries. In one Marra Mamba style deposit, the alluvial to bedded boundary only creates distinctive signatures when both the magnetic susceptibility logs and the downhole chemical assays are considered. Identifying where the ore to BIF boundary occurs with the NS3-NS4 stratigraphic boundary requires both natural gamma logs and chemical assays. These data sources have different downhole resolutions. This paper proposes a Gaussian Processes based method of probabilistically processing geophysical logs and chemical assays together. This method improves the classification of the alluvial to bedded boundary and allows the identification of concurring stratigraphic and mineralization boundaries. The results will help to automatically produce more accurate and objective geological models, significantly reducing the need for manual effort.