Reserve Estimation Research Articles

Assessment of Mine Plans for Production and Variations at Ibese Limestone Mines

This research investigated the mine plans used for the production target and annual production of limestone for the past seven (7) years. The actual amount of limestone produced per year in tons was computed to determine the variations in the mine's production and the life span of the reserve of limestone deposits. A new mine plan was proposed, and a targeted amount of limestone production per year was suggested. The mine plans for the mine's production were generated from 2014 to 2020 using a complete data collection system for Real Time (RTK) GPS and Total Stations with in-field coordinate geometry called SurCE by Carlson Software. The total production target for the seven (7) years was 87,017,146 tons and the actual production was 76,313,056 tons. The total variations for the years under this assessment were -10,704,090 tons except for the year 2018 when the actual production was higher than the production target of 511,179 tons. The volumetric reserve estimation revealed that the total volume of the limestone reserve is 5,052,236,090.186 tons. The expected production per week was computed to be 356,400 tons/week, expected monthly production, (EMP) was estimated to be 1,425,600 tons/month. The expected production year, (EPY) was 17,107,200 tons/year and the life span of the limestone deposit, (LS) was computed to be 295.34 years. It can be suggested that the company should target a production rate of 1,425,200 monthly and 17,104,200 per year. This suggestion can be achieved if there is enough investment in terms of skilled labour, heavy-duty machinery, and equipment which will cost-effectively boost production.

Dynamic factor models for claim reserving

This study presents a new approach to claim reserving in the insurance industry using dynamic factor models (DFMs). Traditional methods often struggle to adapt to temporal variations in loss development, a gap that DFMs can effectively address. By employing DFMs on a multivariate time series of loss development factors (LDFs), we offer a more sensitive adaptation and understanding of loss development over time. Our methodology not only facilitates adjustment to trends in loss development but also provides clear explanations for the underlying reasons behind these trends. This aspect is crucial for actuaries, whose responsibilities include offering transparent and understandable reserve estimates. We apply the proposed DFMs to datasets from two different lines of business, demonstrating their ability to capture the temporal evolution of factors influencing loss development. The results indicate that our approach enhances fitting ability and provides deep insights into the dynamics of claim reserving. Furthermore, we assess the uncertainty in the ultimate loss amounts required for risk management to ensure financial stability and compliance with insurance regulatory requirements. This study contributes to the field of actuarial science by highlighting the potential of DFMs in enhancing the accuracy and reliability of claim reserving processes.

Evaluation of functional significance of coronary arteries stenoses with estimation of fractional flow reserve based on computed tomography angiography data

Evaluation of functional significance of coronary arteries stenoses with estimation of fractional flow reserve based on computed tomography angiography data

Physico-Mechanical and Petrographic Appraisal of Carbonate Rocks as Construction Aggregate: A Case Study from Lesser Himalaya, Pakistan

Owing to the exceptional role of crushed rock aggregates in construction, their prospective reserves fulfilling qualitative and economic parameters should be identified to meet their ever-increasing demand. This research aims to unlock new sources of high-quality construction materials, thereby optimizing concrete performance and meeting the requirement for sustainable building materials. The study investigated the physicomechanical and petrographic properties of Jurassic, Cretaceous, and Eocene limestones of the Kawagarh Formation (KF), Samana Suk Formation (SSF), and Margalla Hill Limestone (MHL) along the Abbottabad–Nathiagali roadside section, as per the American Society for Testing and Materials (ASTM) and British Standards Institute (BSI). The results of geotechnical investigations are flakiness and elongation index (19.6–34.85% and 17.53–31.2%), specific gravity (2.64 to 2.67), water absorption (0.34–0.63%), unconfined compressive strength (41.72–80.65 MPa), loss angeles abrasion value (15.63–19.08%), impact value (15.38–19.36%), and the soundness (0.9–1.97%). These results showed values within the acceptable range according to ASTM and BSI standards. Petrographically, the KF is classified as mud and wackestone, the SSF was mainly classified as wacke-packstone and grainstone, while the MHL was classified as bioclastic wacke and packstone. The dominancy of micrite matrix favors the strength of aggregate, whereas porosity, microfractures, stylolites, and allochems adversely affect the mechanical properties of rock aggregates. Based on reserve estimation, the studied sections possess significant exploitable reserves and are feasible for quarry operations. The studied rocks can be utilized to produce high-strength concrete, optimizing its performance and quality.

Calculation and Comparison of Water Saturation in Carbonate Formation by Different Methods

Water saturation is an important governing quality and quantity of the hydrocarbon evaluation, as reserve estimation. The challenge of the reservoir engineer is to resolve and understand the differences distribution of water saturation content by using different techniques. However, though out this work illustrate variety models to calculate water saturation which is compared with core measured. Therefore, water saturation is calculated using different models; Archie, Simandoux, Indonesia which are compared with Leverett J-Function (capillary pressure measurements). Archie's parameters of the studied reservoir rock; Cementation Factor (m) and Saturation Exponent (n) are derived from the log data used. Error of the water saturation estimation may significantly impose bias in the estimation of both quantities. In spite of, there are various factors affecting potential error in water saturation estimation, such as, average permeability, average porosity and Archie's parameters are used. Consequently, the comparison between results indicates that, Indonesia and Simandoux models are show the smallest average absolute error values, where Archie's model common one used for water saturation in Carbonate rocks. Three wells are selected to present his comparison.

Analiza strukturalna i zróżnicowanie pegmatytów w starożytnym twierdzeniu, Giraúl IV, dystrykt pegmatytów Namibe, Angola

Since the early 60´s, in the past XX century, the Giraúl pegmatites have been known for their resources of beryl, mica and feldspars, which were exploited in a regular basis from Giraúl claims I to IV till 1974, during the Portuguese administration of Angolan territory. A broader exploration of this pegmatite field was performed by the ancient Lobito Mining Company (LMC) engaged in detailed geological mapping of the granitic pegmatites and the structural constraints of their location. A structural map of the region was than elaborated, combining the interpretation of aerial photographs with field work performed by the LMC geologists. Recently, a growing economic interest is attributed to these claims, in the region of Bulamucolocai, Pitau and Muvero desert-dry rivers (locally known as “Mulolas”), considering the Li, Cs, and Ta (LCT) metallic specialization of some pegmatite bodies and the occurrence of beryl and tourmaline gemstones, mainly, morganite (Cs-beryl), aquamarine and also elbaite-liddicoatite. Giant crystals of spodumene, up to 6 m in length, define individualized quartz + spodumene units inside some of the more typical LCT pegmatite bodies. Pollucite was identified in the main pegmatites of Giraúl IV claim and not in the adjacent igneous leucocratic breccias. These, in turn, correspond to a complex pegmatite assemblage, very peculiar in what concerns its selective metasomatic effect over some surrounding rocks, with the formation of rims of holmquistite amphibole in contact with gabbro and norite and schorl-dravite tourmaline in contact with gneissic to metapelitic hosts. The breccia-like granitic rock combines clasts of spodumene an K-feldspar with a matrix mainly composed of some quartz, albite and mica including tourmaline, garnet and F-apatite, as accessory minerals. In the same area, huge potassic pegmatites hold giant crystals of microcline and orthoclase and very little quartz, being unusual due to its high content of triplite – zwieselite and triphylite – lithiophilite primary phosphates. The overall composition of these pegmatites is more likely syenitic (low quartz content) than truly granitic. A high-resolution structural analysis of the LCT ensemble (pegmatite plus related lithotypes) is now proposed enhancing the unusual relations between granite breccia plugs, sill-like more typical pegmatites, irregular shaped isodiametric bodies and products of metasomatism. This approach will lead to the understanding of the true dimension, anatomy and inner fraccionation of the different LCT facies and rare-metal deposits with obvious consequences regarding mineral detection and resource – reserve estimation, through the proposal of a more suitable conceptual model to rule its exploration.

Integrated Approach to Reserve Estimation and Optimal Open Pit Design for Hydrothermal Gold Deposit

Integrated Approach to Reserve Estimation and Optimal Open Pit Design for Hydrothermal Gold Deposit

Review about evaluation methods of recoverable reserves of deep water drive gas reservoirs in China

It has been widely accepted that China is one of the biggest natural gas consumers. Related to the imports of LNG, China stands in a very uncomfortable situation. Most domestic gas reservoirs fall within deep water drive gas reservoirs inordinately, which has entered the production depletion stage. Accurate estimation of SEC recoverable reserves of deep water drive gas reservoirs is of great significance for gas consumption planning and peak shaving. The existing calculation methods of recoverable reserves mainly consist of static methods and dynamic methods. In the early stage of exploration and development, the volumetric method has often been utilized to calculate the recoverable reserves. With the continuous development of gas reservoirs, the main methods for evaluation are dynamic methods, including the successive subtraction of production method, water drive curve method, prediction model method, attenuation curve method, improved virtual curve method, and material balance method for deep gas reservoirs.

Estimation of renal functional reserve in patients with preserved renal function and coronary microvascular dysfunction

Estimation of renal functional reserve in patients with preserved renal function and coronary microvascular dysfunction

Right ventricular outflow tract diameter change with exercise: a prospective exercise echocardiography and invasive CPET study.

While cardiac output reserve with exercise predicts outcomes in cardiac and pulmonary vascular disease, precise quantification of exercise cardiac output requires invasive cardiopulmonary testing (iCPET). To improve the accuracy of cardiac output reserve estimation with transthoracic echocardiography (TTE), this prospective study aims to define changes in right ventricular outflow tract diameter (RVOTd) with exercise and its relationship with invasively measured haemodynamics. Twenty subjects underwent simultaneous TTE and iCPET, with data collected at rest, leg-raise, 25 W, 50 W (n = 16), 75 W (n = 14), and 100 W (n = 6). This was followed by a second exercise study with real-time RV pressure-volume loops at similar stages (except leg-raise). The overall cohort included heart failure with preserved ejection fraction (n = 12), pulmonary arterial hypertension (n = 5), and non-cardiac dyspnoea (n = 3). RVOTd was reverse engineered from the TTE-derived RVOT velocity time integral (VTI) and iCPET-derived stroke volume, using the formula: Fick stroke volume = RVOT VTI × RVOT area (wherein RVOT area = π × [RVOTd/2]2). RVOTd increased by nearly 3-4% at every 25 W increment. Using linear regression models, where each subject is treated as a categorical variable and adjusting for subject intercept, RVOTd was correlated with haemodynamic variables (cardiac output, heart rate, pulmonary artery and RV pressures). Of all the predictor haemodynamic variables, cardiac output had the highest r2 model fit (adjusted r2 = 0.68), with a unit increase in cardiac output associated with a 0.0678 increase in RVOTd (P < 0.001). Our findings indicate that RVOTd increases by 3-4% with every 25 W increment, predominantly correlated with cardiac output augmentation. These results can improve the accuracy of cardiac output reserve estimation by adjusting for RVOTd with graded exercise during non-invasive CPET and echocardiogram. However, future studies are needed to define these relationships for left ventricular outflow tract diameter.

Solutions to Complex Well-Testing Challenges Aid Offshore Black Sea Gasfield Development

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 215123, “Frontiering Ultradeep Water Gas Field Development, Offshore Black Sea, Turkey: Solutions to Complex Well-Testing Challenges and Proving Production Potential,” by Coşan Ayan, SPE, Suat Aktepe, and Koscal Cig, Turkish Petroleum, et al. The paper has not been peer reviewed. _ Increasing demand for reliable energy resources has led to an increased exploration activity for untapped hydrocarbon resources in deep water. The recently discovered, fast-tracked Sakarya offshore natural gasfield development is a prime example. This paper describes dynamic reservoir characterization considerations, challenges, and engineering solutions to derisk field-development decisions, confirmed by a well-testing campaign in a complex setting with no tolerance for failure. Sakarya Field The Sakarya field is approximately 170 km north of the Turkish coast in the Black Sea at 2117 m (Fig. 1). The natural gas field was discovered in August 2020 and is estimated to have potential natural gas reserves of 11 Tcf of lean gas, making it the largest gas reserve discovered in the Turkish Exclusive Economic Zone as well as the Black Sea. In October 2020, a second discovery was made in the lower sections of Tuna 1, which increased the potential recoverable reserve estimate to 14.3 Tcf of lean gas. The discovery was found at the deeper part of the well, where an additional 30 m of gas pay was encountered in the sandstone reservoir of the Late Miocene. Further, the drilling of the exploratory well Amasra 1, 40 km north of the Sakarya field, led to the discovery of 4.7 Tcf of gas in June 2021, and increased the potential recoverable cumulative natural gas reserves to 19 Tcf. The first gas production from the Sakarya gas field began in 2023 into the Turkish grid from the completed wells of the first phase of the project. To accelerate the time to first gas production, it was deemed necessary to flow-test the reservoirs to acquire critical reservoir information, assess production potential and completions efficiency, and reduce the number of wells. This was achieved by drilling and testing several appraisal wells with multiple target intervals that will be used later as producers. Well-Completion Strategy Affecting Well-Test Program Formation-failure studies and wireline formation testing data have shown that these weak reservoirs are prone to sand production. As a result, downhole sand-exclusion completions are necessary. Gas permeabilities are good, with core-measured air permeabilities in the range of 1–1000 md. The reservoir pressure varies from 4,200 to 5,800 psi, depending on depth. Reservoir temperatures are relatively low, in the range of 25–60°C. Lower Completion Installation Sequences. After drilling, the main section was completed with 9⅞-in. 62.8-lbm/ft production casing. The selected interval from openhole logs and formation test results was perforated and overbalanced by individual runs of a tubing-conveyed perforation (TCP) system. On the second run, sump packers with direct wrap screens across the interval were installed and gravel was pumped. After the lower completion was in place and the drillstem test (DST) had been conducted, another part of the lower completion across the upper interval was installed and flow-tested. After completing all DST well tests, the well was suspended with the lower completion system in place, ready for installation of the upper completion with intelligent flow-control valves and permanent gauges.

Occurrence and lithogeochemical composition of iron ore in Ngororero District, Rwanda

This research aimed to investigate the occurrence and lithogeochemical composition of iron ore in the Ngororero District of Rwanda, with a specific focus on assessing its suitability as a raw material for iron and steel production. The objective was to determine the elemental composition, mineral species, and morphology of the iron ore samples collected from the study area. Eleven samples were collected and analyzed using Inductively Coupled Plasma-Mass Spectrometry (ICP-MS), X-ray diffractogram (XRD), and Optical Light Microscopy. The study shows that Ngororero iron ore is a medium-grade hematite (Fe2O3) ore with iron content ranging from 58.3 % to 65.8 %. Importantly, it contains minimal impurities (less than 3.7 % SiO2 and less than 0.4 % Al2O3) and low levels of detrimental elements (P less than 0.02 % and S less than 0.19 %). While the sulfur (S) content slightly exceeds standard limits, the phosphorus (P) content falls within the acceptable range for commercial iron ores. These findings, marking the first report of iron ore occurrence in the Ngororero District, have significant implications for the potential exploitation of Ngororero iron ore as a valuable raw material in iron and steel production and its suitability for industrial applications. To confirm its economic potential and assess its viability for iron and steel production in Rwanda, the study recommends that additional thorough exploration activities be conducted in this area using both geology and geophysical investigation as well as exploration drilling for iron ore reserve estimation in the study area.

Inter-Repetition Rest Impact on the Estimated Repetitions in Reserve at Various Loads and Proximities to Neuromuscular Failure.

Ruiz-Alias, SA, Ramirez-Campillo, R, Leando Quidel-Catrilelbún, ME, García-Pinillos, F, and Pérez-Castilla, A. Inter-repetition rest impact on the estimated repetitions in reserve at various loads and proximities to neuromuscular failure . J Strength Cond Res 38(8): 1379-1385, 2024-The repetitions in reserve (RIR) estimation method allows for monitoring resistance training set volume. However, a significant bias is commonly observed when ending the set further from neuromuscular failure and using lower relative loads. Is unclear if implementing an IRR could improve its estimation validity. This study aimed to determine if the RIR estimation is influenced by the IRR at different relative loads and proximities to neuromuscular failure during the back squat (BQ) and bench press (BP) exercises. After a familiarization session, 19 male subjects (age, 21-26 years) completed an experimental session for each IRR configuration: IRR0 (without rest), IRR3 (3 seconds of rest), and SSIRR (self-selected rest [up to 5 seconds]). In each session, single sets to neuromuscular failure were performed at 3 relative loads (65% 1 repetition maximum [1RM], 75% 1RM, 85% 1RM) for BQ and BP exercises. Using the Estimated Repetitions to Failure scale, subjects estimated the RIR associated with a score of "5" and "2" (i.e., RIR-5 and RIR-2). The results revealed no interactions between the RIR estimation and IRR for BQ (RIR-5: p = 0.812; RIR-2: p = 0.084) or BP (RIR-5: p = 0.884; RIR-2: p = 0.944). Subjects provided valid estimations, with overestimation (BQ RIR-5: 0%; BQ RIR-2: 2.9%; BP RIR-5: 1.1%; BP RIR-2: 2.3%) or underestimation (BQ RIR-5: 14.9%; BQ RIR-2: 4%; BP RIR-5: 15.2% BP RIR-2: 8.2%) in few sets. In conclusion, the RIR estimation method seems valid, and the IRR (0-5 seconds) does not influence its validity at 65-85% 1RM loads nor proximities to neuromuscular failure (RIR-2, RIR-5).

Rare earth elements-A review of exploration, processing, environmental impact, recycling and prospect analysis

In high-tech and low-carbon economies, rare earth elements are becoming increasingly important. Global demand for REEs is increasing as society transitions to a more sustainable model with the goal of reducing reliance on fossil fuels. The study on rare earth elements (REEs) is reviewed in this document, including estimations of the world's reserves, categorization of REE deposits, key REE production paths and processing techniques, environmental effects, and the possibility for recycling in the rare earth business. This paper's main goal is to present a comprehensive environmental and socioeconomic perspective on REEs, emphasizing the environmental effects of various REE-related activities and looking at how REEs can be used more effectively and scientifically, such as by investigating the viability of recycling REEs. and offer an outlook for the future. In order to address the issues with their demand and lower the dangers involved with supplying rare earths, researchers are further developing the methods for mining and using rare earth elements as well as recycling and reusing them.

Linear Geostatistical Estimating Methods Compared Using Annual Average Daily Traffic Data from Low-Volume Roads in Minnesota

Geostatistical methods are utilized to analyze and predict values associated with spatial or spatiotemporal phenomena. These techniques incorporate both the spatial and, in some cases, temporal coordinates of the data. The mining industry prefers these methods for ore reserve estimation, and they have also gained popularity in other specialized fields. Notably, the transportation industry is gradually adopting them to estimate annual average daily traffic (AADT). Although researchers use various predictive techniques to estimate AADT, many of them are experiential or based on simple subjective assessments, especially for low-volume roadways. The shortcomings of these assessment techniques include high margins of error, non-integrated spatial and temporal variability in the datasets, and inadequate consideration of total coverage. Geostatistical techniques are superior predictive tools in many fields because their algorithms account for these limitations. Moreover, geostatistical methods can predict data at both sampled and unsampled locations with minimal available data. Therefore, the crucial question arises as to which method best estimates AADT. This paper focuses on investigating linear univariate geostatistical methods, namely empirical Bayesian Kriging (EBK), ordinary Kriging (OK), simple Kriging (SK), and universal Kriging (UK), along with proportional valuations, to estimate AADT on low-volume roadways and provides a recommendation based on the findings. The study utilizes AADT data from 1,486 and 2,731 locations on low-volume roads throughout Minnesota in 2009 and 2016, respectively. The results obtained from this investigation hold significant systematic and practical implications for estimating AADT on low-volume roads and can be adapted for high-volume roadways as well.

Spatial coal seam modeling and reserve estimation using geostatistical methodologies by GIS technique in a selected coalfield in Ermenek Basin, Turkey

The accurate and reliable calculation of tonnage and quality of coal is vital for the success of mining projects. The Ermenek coal basin is located in the Western Taurus Mountains in the southern part of Turkey, which covers an area of approximately 620 km2 and is a Neogene coal basin. This paper generates spatial modeling of Pamuklu-Tepebaşı coal seams in the Ermenek Basin and the prediction of coal reserve by using geostatistical methods with Ordinary Kriging (OK) based on the GIS technique. The data used in the study were obtained from 82 boreholes belonging to a private company that has actively produced coal in the Pamuklu-Tepebaşı coalfield. In this study, the OK method was applied to spatial coal seam modeling for comparison of the performance of 11 distinct models. Using the rational quadratic models better explained the spatial structure of the coal seam depth data. The nugget-sill ratio indicated high spatial dependency with 0.01 and 0.02 for coal seam upper-lower surfaces, respectively. The kriged coal depth maps for upper and lower coal seam surfaces have been created using ArcGIS 10.2 software. As a result of these analyses, it was calculated that there is a coal reserve of 3.734.017 m3 using the determined surfaces with OK in the study area. Furthermore, developed spatial distribution maps in the study area could be helpful to plan further exploration activities.

Measurement and modeling of moisture equilibrium and methane adsorption in shales from the southern Sichuan Basin

The adsorption characteristic of methane in water-bearing shale is essential for the accurate estimation of shale gas reserves. The sorption mechanisms of moisture and methane in shale remain poorly understood. In this work, moisture equilibrium measurements in shales were conducted to study moisture absorption characteristics. The governing factors for moisture absorption were determined, and the suitability of moisture absorption models was evaluated. An innovative water saturation apparatus and an improved methane adsorption system were built to measure methane adsorption in high water-bearing shales. The suitability of the Langmuir model in describing methane adsorption in water-bearing shale was studied. A predictive model for methane adsorption in water-bearing shale was developed based on improved BP neural network algorithm, which was further utilized to determine the primary governing factors for methane adsorption. The results revealed that the equilibrated moisture content in shale presents a three-stage and S-type ascending trend with increasing relative humidity. The methane adsorption amount shows a slide-type downward trend with the increase of water content. The Peleg model demonstrates the superior performance in describing the hygroscopic isotherms, and the Langmuir model performs poorly in characterizing methane adsorption in shales with high water contents. The controlling factors for the ultimate moisture content mainly include specific surface area, average pore diameter, carbonate and pyrite content. The methane adsorption in water-bearing shale is mainly governed by water content and specific surface area. The results of this work enrich the understanding of the adsorption mechanisms of shale gas and provide a solid basis for the assessment of shale gas reserves.

THE IMPACT OF MINING EXCAVATIONS (VOIDS) ON THE WATER QUALITY MANAGEMENT IN THE NORTH KALIMANTAN PROVINCE

The presence of a coal mining business permit not only has an economic impact but can also cause environmental problems. The objectives of this research are (1) to analyze the impact of PT XXX's overflowing voids on water management and water quality, and (2) to carry out spatial analysis, mapping and estimation of coal reserves that cannot be extracted by PT YYY. The research method is a case study selected by purposive sampling at two companies in North Kalimantan Province using water quality test analysis with grab sampling and estimating the potential loss of coal reserves using the rule of gradual changes and rule of nearest point methods as well as aerial photography. The research results show that in general the water use meets Class 3, namely: Water that can be used for cultivating freshwater fish, animal husbandry, water for irrigating crops, and/or other uses that require the same water quality as that use. Apart from that, PT XXX has built an overflow water channel into the PT YYY area without management approval, thus having a negative impact on PT YYY's mining plans which are dominated by peat land. Map of PT YYY's lost coal potential of 714,688 MT with an affected area of 15.16 hectares with details of 6.4 ha being the void area included in PT YYY's IUP area, 1.86 ha being the road and buffer zone area and 6.9 ha is a potential lost area of PT YYY.

Estimating the natural gas compressibility factor using a statistical correlations and machine learning approaches

Gas compressibility factor plays an critical role in petroleum engineering applications such as gas metering, pipeline design, reserve estimation, gas flow rate, material balance calculations, and many other significant tasks. Therefore, it is crucial to accurately estimate the gas compressibility factor. There have been a lot of studies on calculating the gas compressibility factor from laboratory data, which can be summarized into two main approaches: statistical correlations and machine learning algorithms. In this study, on statistical correlations the authors implement explicit and implicit method while on machine learning algorithms, we use Artificial Neural Network (ANN) and Least-Squares Support Vector Machine (LS-SVM). The data was collected from open literature. Implementing the two approaches mentioned above and comparing statistical parameters such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Coefficient of Determination (R2 ) found that machine learning algorithms give much more accurate estimation results than statistical correlations, and besides, the ANN algorithm has the most accurate prediction results with the lowest MSE and RMSE (0.000002 and 0.0016) and the highest R2 (0.9999). The high-precision calculation results show that the ANN algorithm mentioned above can be applied to estimate other real gas compressibility factor data sets. On the other hand, this study can be extended to another subset of machine learning algorithms, such as deep learning and ensemble learning.

A New Workflow for Estimating Reservoir Properties With Gradient Boosting Model and Joint Inversion Using MWD Measurements

Triple-combo logs are important measurements for estimating geological, petrophysical, and geomechanical properties. Unfortunately, wireline and advanced logging-while-drilling (LWD) logs are typically dropped from the formation evaluation plan for unconventional wells due to economic constraints or borehole instability risks. Available measurements are typically measurement-while-drilling (MWD) gamma ray (GR) logs, along with surface measurements such as weight on bit (WOB), rate of penetration (ROP), torque, rotation per minute (RPM), and differential pressure. The development of a robust and rapid model for predicting reservoir properties using this limited data set would be of high value for geological evaluation. Estimating such properties is a challenging task due to the nonlinear relationship between the available log data and unknown reservoir properties. A novel workflow that combines two sequential models is presented. First is a machine-learning (ML) algorithm to predict triple-combo logs from drilling dynamics and GR logs. To train the ML algorithm, well logs obtained from multiple wells located in the Eagle Ford and Permian Basins are scrutinized to identify important features. This process includes depth shifting, outlier detection, and feature selection, which allows for strategic hyperparameter tuning. Several regression algorithms are investigated, and it is found that gradient boosting algorithms yield superior prediction performance. Unlike random forest methods, boosting algorithms train predictors sequentially, each trying to correct its predecessor. After triple-combo logs are predicted from MWD logs, a physics-based joint inversion model is applied to estimate various reservoir properties. The trained model is deployed on a blind test well, and the predicted logs show excellent agreement compared to the corresponding triple-combo measurements. The multimineral inversion using predicted triple-combo logs yields a geologic model that is validated with elemental capture spectroscopy (ECS) measurements. Additionally, reconstructed logs from the forward model closely match measured logs by minimizing the cost function. Therefore, real-time estimated geological, petrophysical, and geomechanical properties can reveal complex geologic information and be used to mitigate uncertainty related to drilling optimization, reservoir characterization, development planning, and reserve estimation. Using the MWD logs to predict triple-combo logs followed by a joint inversion is an innovative approach for geological evaluation with a limited data set. The developed workflow can successfully provide (1) geologic lithofacies identification and rock typing, (2) more confidence in real-time drilling operation, (3) reservoir properties prediction, (4) missing log imputations and pseudo-log generation with forward modeling, (5) guidance for future logging and perforation, (6) reference for seismic quantitative interpretation (QI) and well tie, and (7) potentially massive computation time saving from days to minutes.