Search Documents

By D W. Evans

Research into the dynamic properties of ground support elements continues to grow at a global level, given the increasing importance of this field of application within underground hard rock mining. Measuring the mechanical response of such elements under the rapid onset of simulated rock burst loads forms the basis for this experimental work, typically utilising gravitational drop testing methods. This field of research is fundamentally driven by an increasing need for dynamic performance data, given the constant evolution of rock bolting product designs, materials selection, combination testing and specific application needs based on ground types and loading scenarios. A smaller number of globally recognised dynamic test facilities have predominantly lead research within this arena, typically in association with academic institutes, larger industry groups and government research institutes. However, due to the increasing requirement for dynamic performance data, additional facilities have now emerged in order to satisfy the overall demand for test work. These newer facilities are more typically associated with suppliers to the underground hard rock industry, with experimental designs looking to replicate, at least in part, the work of the primary facilities. For standardisation, repeatability and calibration purposes, the cross validation of test data between primary and secondary facilities becomes an obvious and important consideration across this field of research. This paper outlines the development of a new dynamic drop test facility for testing rock bolts, including the structural design and associated high speed data capture system. Based on the utilisation of identical bolt types, material grades, anchoring systems and decoupled ‘free-lengths’, comparative data has been generated to provide cross validation of the new test facility against data generated by a globally recognised primary facility. The cross validation of data provides increasing confidence in the repeatability and accuracy of the new facility as activity continues towards the provision of further experimental test work.

Nov 29, 2022

By C Mark, Z Agioutantis

This paper presents an empirical design methodology that is based on successful US roof support systems using both primary and secondary support in coal mining entries and intersections. The recent nationwide dramatic reduction of roof fall rates, especially in areas with weaker roof, can be directly correlated to the effectiveness of the roof support systems. A large database of roof fall histories at several mines is used to define the approximate boundaries of three modes of roof support with respect to roof strength as classified by the Coal Mine Roof Rating (CMRR) and in situ vertical stress as indicated by the depth of cover. These roof support classifications are: • Suspension, where roof bolts mainly provide skin control for strong roof. • Beam building or reinforcement, where a roof of moderate strength can be supported by roof bolts only. • Supplemental (secondary) support, where the primary support by roof bolts may be augmented by several secondary support systems, eg cable bolts, trusses, additional roof bolts or standing supports. The Primary Support Rating index (PSUP) captures the capacity of the primary roof bolt system as a function of the actually implemented roof control plan, ie roof bolt length, roof bolt density, and roof bolt capacity, while it does not consider the type of bolt. The actual PSUP value is compared against a suggested PSUP value derived from statistical analysis of US case histories. Where secondary support is indicated, the expected rock load determines the necessary capacity of the secondary support system. A new (free) software package, Analysis of Mine Roof Support (AMRS), implements the design methodology and also allows comparison with other international indices such as the Reinforcement Density Index (RDI), the PRSUP, etc.

Nov 29, 2022

By P Mikula, B Darlington, J Graham, P Young

In underground mines experiencing dynamic events, there is an ever-present challenge to design and maintain a capable ground support system. One key component of ground support systems is surface support, which serves the essential roles of retaining failed rock and transferring forces generated by surface displacement to rock bolts or other support elements. Observations underground show that in many instances, standard mesh panels alone do not provide sufficient dynamic support to retain substantial bulking failures imposing displacement onto mesh, and in such cases Mesh straps and W-straps may be employed. Determining the performance of surface support elements proves to be difficult, with a multitude of variables at play in the underground environment. Quite comprehensive work has been completed on laboratory testing of surface support systems; however, very little work has been conducted on in situ dynamic testing of Mesh straps and W-straps. In 2019 in Western Area’s Flying Fox Mine, in situ dynamic testing was conducted on a series of Mesh straps and W-straps using Sandvik’s in situ Dynamic Test Rig (DTR). The testing regime applied impact energies from 5.1 to 15.3 kJ to Mesh straps and W-straps. The performance of the straps that did not fail was a maximum absorbed energy of 9.3 kJ and 14.2 kJ for the Mesh straps and W-straps respectively. The testing successfully demonstrated the ability of Sandvik’s DTR to test not only rock bolts, but also surface support elements, creating new testing and research possibilities. The testing also provided important design guidance data that is not practicable to obtain using laboratory dynamic testing facilities.

Nov 29, 2022

By T M. Nester, D J. Reardon, A N. Page, B J. Barsanti

The load capacity of a friction bolt has long been regarded as 4 t/m of embedment for ground support design analysis and supported by in situ pull test results. In recent years, the authors have observed lower pull test results or inconsistent results which were rarely sighted in the early 2000s. The lower pull test results indicated in some cases that reliable friction bolt capacity (≥90 per cent of results) is closer to 3 t/m of embedment with some geotechnical engineers adjusting the ground support parameters in the design work accordingly. This can be in contrast with other mines which reliably obtain 4 t/m (or greater) using the same suppliers bolt in similar ground conditions. This raised several questions: • What is the effect of poor installation (bolt twist or bolt hole deviation)? • Does the performance of friction bolts vary in different rock masses? • Does bolt performance vary between suppliers? • What effect does increased jumbo feed/percussion/rotation settings have? • Have the bolt dimensions or steel strip width changed? • Have the physical bolt properties changed? • Have the steel properties changed? This paper was aimed to answer as many of these questions as possible and discuss each variable in detail. This project included friction bolt trials at various mine sites across Australia, multiple friction bolt manufacturers, steel manufacturers and laboratory analysis. The paper also includes discussion on what the authors consider best practice for pull testing. It should be noted, the terms ‘friction stabiliser’ and ‘split set’ used in this paper are used in a generic manner and is not linked to any specific ground support supplier. Whilst its acknowledged within the industry the term split set and friction stabiliser is used interchangeably, the authors acknowledge that the term split set is a trademark of Split Set Mining Systems and has been used in context in this paper.

Nov 29, 2022

By C G. Sweeney

The underground mining industry is seeing unprecedented shortages in skilled labour. Migration to Australia being at an all-time low is a leading concern, with pandemic-induced border closures and lockdowns only exacerbating the skills shortage, not causing it. The skills shortage is forecast to peak in 2023 and remain high beyond 2025. With this skills shortage in mind, mining companies throughout Australia are trying to deal with stricter protocols around Health and Safety (H&S) legislation, as well as more stringent controls on quality assurance, particularly as underground mines are reaching greater depths and extending into ‘unchartered territory’. The cumulation of this is proving to be a major hurdle in ‘effective’ operational geotechnical management. The drive for automation in mining is bringing another level of complexity. However, ‘once automation is up and running, it presents the opportunity to access talent pools that may not normally be available. There is also a change in skillset that will be required to drive automation. ‘The top two emerging roles that will be critical for the future of the mining industry are artificial intelligence and machine learning (AI/ML) specialists, and process-automation specialists’ (Writer, 2021). If anything positive has come from the pandemic, it has shown that remote working can be effective and that the drive for automation and critical evaluation of skills required on-site should be given greater priority. UQ acknowledges that the industry is changing, and that digital technology is required to be offered in mining related disciplines. Industry and academia need to do more to promote geotechnical engineering in mining and balance the curriculum to suit the future of mining and make it more enticing. The average cost of a geotechnical failure is about $5 million in coalmines and can be into the hundreds of millions of dollars if production is affected (Cartledge Mining and Geotechnics, 2020). The cost of a good geotechnical design far outweighs the potential losses if a failure were to occur. This alludes to the growing importance of experienced geotechnical consultancies playing a pivotal role in managing mine site geotechnical hazards, while the industry is developing the next generation of geotechnical engineers.

Nov 29, 2022

By P Mikula, R Geranmayeh, R Carlton, M Ferguson, R Whiting

The concept of using specially designed slot blasts to mine through stress abutments that intersect faults, while controlling fault-slip seismicity, has come of age at Mt Charlotte Mine. Due to the high and adverse stress field and the presence of major faults, the mine is vulnerable to significant fault-slip seismicity. As mining proceeded, advancing stress abutments around various mined areas encroached on several major subvertical faults. If the sequence formed a diminishing pillar between stope and fault, the removal or failure of that pillar daylighted the fault, and often resulted in significant fault-slip seismicity. A new strategy called ‘Tight Slot Blasting’ (TSB) was developed to restrain or dampen this style of fault-slip movement. The TSB has geometric similarities to a conventional cut-off slot used to create stress shadows, but its function is different. Its objective is to trigger and control a sudden shift of the stress state on a fault from an initial stable state, through an unstable transition, to a second stable state. A TSB firing advances a long tall narrow slot through a stressed abutment between a stope void and a nearby fault, using a rapid blast with a low swell ratio, so that the slot substantially and rapidly fills with blasted rock fragments. During the blast process, the stressed rock is rapidly removed and transformed into the fragmented rock fill in the slot. The strategy deliberately allows a fault-slip episode to manifest at a chosen time and place. The tight-packed broken rock in the slot performs an essential shock-absorbing function by controlling the slip amount and rate on the fault, while at the same time the fill becomes compacted by the movement. This controls the dissipation of fault-slip energy, and greatly reduces the seismic response in time and space. A small closure across the slot occurs, with consequent relaxation and destressing of the adjacent rock mass. Stress monitoring data shows that slot creation lowers the abutment stress in the nearby rock mass, while the newly placed fragmented rock fill acts as a shock absorber, providing considerable resistance to substantial fault-slip movement that otherwise could deliver a significant seismic event. First used in 2007, the strategy has been refined to become a routine specialised practice for management of seismically hazardous situations at Mt Charlotte, with 12 successful TSBs conducted to date. Despite mining through multiple high stress abutments, seismic reactions have not exceeded ML 2.1 with any TSB firing, all remaining below the site tolerable maximum of ML 2.3. This paper presents an overview of the strategy, and five case examples of the strategy in critical stressed areas in July 2016, Sept 2017, Dec 2017 and two in Oct 2020. The strategy has successfully managed the seismicity and is expected to remain an essential part of the seismic management plan at Mt Charlotte Mine.

Nov 29, 2022

By E Villaescusa, C Drover, I Onederra

Development face destress blasting is a construction technique in deep mining, which aims to prevent, reduce the frequency, or manipulate the timing of violent, stress-driven rock mass instability at the face. This paper presents a destress blasting design concept that was optimised for the specific rock mass conditions of a deep mine using the Hybrid Stress Blasting Model (HSBM) software. The design was then implemented in an area of the mine where stress-driven instability of the face regularly occurred during development. An adjacent tunnel was constructed in parallel without destressing, within the same rock mass conditions, as a control. The rock mass response to each excavation was quantitatively compared using high-resolution local seismic monitoring. Visual observations of the fractures created by destress blasting were also made. The analysis indicated that destressing reduced both seismicity and the rock mass stiffness ahead of the face. This was inferred from the spatial broadening of the development seismogenic zone and the typically significant reduction in the spatial density of seismic events. Seismic source mechanism analysis also indicated that destress blasting generated a much wider variety of failure modes than conventional blasting. Most seismic source mechanisms following destressing were associated with natural joint structures. A relatively small portion were consistent with newly created blast-induced fractures. The greater diversity of rock mass failure modes associated with destressing implied a lower potential for stain-energy accumulation within the rock mass at the face, and thus a lower risk of hazardous instability. Observations of visible rock fracturing indicated that the major principal stress influenced the damage intensity on natural geological structures within the destressed zone, and that numerical predictions of radial fracturing limits made by HSBM analysis were realistic representations of the real blasting performance.

Nov 29, 2022

By O Kirtel, A A. Faizan

With the rapid increase and development of high-speed trains (HSTs), the problems such as train-induced ground vibration and its dynamic effects on nearby structures have become a major environmental concern in urban areas. Ground-born vibrations produced by high-speed trains pose a great challenge for engineers to build structures in such areas that is applicable for residents. In order to study the effect of train-induced ground vibration on nearby buildings, an advanced three-dimensional finite element model for train-induced ground vibrations has been developed using PLAXIS 3D Ultimate. The results of the in situ measurements which was previously carried out by the authors were used to validate the numerical model. The field test for ground vibration due to the passage of a high-speed train was carried out on the Istanbul-Ankara high-speed railway in Turkey with a train speed of 250 km/h. An in-depth analysis was achieved to investigate the effect of different soil properties on the train-induced ground vibrations by using the verified model. According to the result of the vibration analysis, the relative vibration response curves in all directions (x, y, z) for different soil conditions were obtained comparatively. The obtained results of the vibration analysis are discussed in order to figure out some useful conclusions.

Nov 29, 2022

By M Revell, B Sainsbury, D Sainsbury

This paper extends the existing knowledge associated with the design and placement of prefabricated pastefill barricades through a consideration of asymmetric drive geometry and conditions. The load capacity of prefabricated barricades and their failure modes are considered in relation to abutment geology and barricade geometry that includes curvature and span and drive geometry that includes abutment angles, haunch angles and misalignment. Design charts are provided for typical barricade sizes to provide general guidance for design capacities and optimising location placement based on asymmetric drive geometry. A recent barricade failure is considered in relation to the drive geometry and the potential impact it had on the installed capacity.

Nov 29, 2022

By J J. Van Wijk, A R. Penney

Successful management of geotechnical risk at a mining operation is more than just having a good ground control management plan or using advanced numerical modelling and the latest monitoring technologies. The fundamental requirements for effective geotechnical risk management include identification of potential failure modes, identifying and implementing appropriate controls, and demonstrating the effectiveness of those controls. Effective geotechnical risk management requires an integrated approach by confirming that appropriate risk management systems have been identified and implemented (management plans, procedures etc), and that these systems are executed consistently in the field across all working areas. Development of a thorough integrated auditing process executed by an experienced team of technical subject matter experts with substantial operational knowledge, provides a more complete picture of catastrophic hazard management. Following consistent evaluation criteria allows benchmarking of systems and implementation compliance between the sites being audited, and the results can then be ranked against other operations in an industry database.

Nov 29, 2022

By E Shilov, R Dixon, L Tovey

BHP Western Australian Iron Ore (WAIO) operations use a suite of slope monitoring systems which are deployed to operational areas based on the ground control risk. Slope Radars are one system used within WAIO, having recently deployed and integrated a network of 22 radar systems across the operations. The integration required the development of supporting network architecture, infrastructure, virtualised hosting, security protocols and a technology support model. This has enabled data transfer between the physical radar systems located in open pits, to on-site offices, a vendor supported remote control room in Indonesia and the WAIO Integrated Remote Operations Centre for data analysis and monitoring of system and displacement alarms 24/7. The typical technical skillset of a Geotechnical Engineer does not include the knowledge and capability to understand and develop the supporting IT infrastructure required for establishing an integrated radar monitoring network. This paper will provide a discussion of the technology considerations for integrating radar systems and highlight the importance of the developed partnerships between internal geotechnical and technology teams and the external vendor team to ensure a reliable, robust and scalable slope monitoring solution.

Nov 29, 2022

By C Jacobsen, N Harries

Assessing and managing instability hazards is an essential activity when working with unstable natural slopes and excavated open pit mining slopes. Slope monitoring has become the standard technique for the management of geotechnical risks associated with mine slope instability hazards. In recent decades, a proliferation of new monitoring techniques such as terrestrial radar, satellite InSAR and LIDAR have become standard operational tools, complementing conventional survey-based monitoring solutions. A reliable monitoring system should identify and record incipient anomalous slope movement. Although the most obvious purpose of a monitoring system is safety related, slope deformation and performance monitoring also enhance the understanding of slope behaviour and assists in improving system design and implementation. For a monitoring system to be considered an effective early warning tool, it needs to operate in real-time and should be linked to the mine’s emergency communication systems. Terrestrial radar systems have been developed that can provide near real-time measurements, at a high level of precision and covering broad areas in all weather conditions. This has had a positive impact on the management of geotechnical risks in operational conditions. The technology adoption of RADAR in mine monitoring has significantly progressed in the last 20 years from technology enthusiasts, to visionaries, to pragmatists to conservative users. However, the increasing adoption of RADAR cannot reduce or remove other monitoring techniques that often have important benefits; even if these technologies cannot be used as a real-time operational tool. An increasing awareness on the importance of combining data from different monitoring techniques to develop a holistic picture of all possible geotechnical hazards is a current industry trend. This is proving to be an excellent foundation for managing hazards associated with slope instability in open pit mines, and is only expected to improve with time with the adoption of new technologies based on big data and Artificial Intelligence (AI).

Nov 29, 2022

By G Guy, J Summerville

Over the past five years in coalmine slope evaluation the application of 3D modelling has increased significantly as software has become more user friendly but also areas requiring evaluation have become increasingly complex. Often simplistic 2D Limit Equilibrium modelling and analysis has been poorly used to solve slope stability issues by back analysis and subsequent forward predictive analysis. A plane strain condition with slope movement perpendicular to the slope strike is assumed where in reality the problem is more complex. This often leads to a misinterpretation of the failure mechanism because the failure mechanism being modelled in 2D does not generally represent the actual failure mechanism observed in the pit, with 2D modelling often oversimplifying a slope. This paper presents a case study where the initial slope assessment and remedial design evaluation was undertaken as a 2D Limit Equilibrium model which led to an overly conservative and costly remedial solution being implemented. Subsequent modelling of the slope using both 3D Limit Equilibrium and Finite Element methods were able to provide a better understanding of the structural complexities associated and correlate well with slope monitoring data to create a more realistic outcome.

Nov 29, 2022

By S Grimsey, E Janetzki, J Emery

Floor failure on the longwall face is a complex problem due to the dynamic interaction between the surrounding strata and powered roof supports involving multiple variables. It can also be highly problematic and costly when not well understood as there are significantly less remedies available post event. This was the early experience at the case study mine where significant lost time was incurred due to unplanned delays in the first two panels. To detect and prevent such occurrences is clearly the best approach, hence a tool to predict both floor failure occurrence and magnitude prior to mining was developed. Several parameters were determined as the key drivers for the failure events that had been experienced at this site. These parameters were, in list of influence from highest to lowest, material strength, retreat rate and the presence of water. The parameters were evaluated using a semi quantitative risk rating system ‘Heave Risk’ and given a weighting per their apparent influence. This ‘expert system’ methodology was selected as the use of risk rating systems in underground coal mining is well established. The heave risk model was first applied to previously mined panels. It accurately identified all areas of known major floor heave, with clearly defined areas of low and high risk apparent. The heave risk model was then used prior to mining and evaluated post mining based on observed conditions. It was found to also be highly accurate for prediction. The variability of each parameter and its influence on the overall risk rating is illustrated and discussed.

Nov 29, 2022

By B Liu, W J. Zhang, S Saydam, B Li

Rock bolts are commonly used to reinforce ground in underground mining and civil tunnel environments. Rock bolts serve two main functions: i) they suspend large, loose blocks of ground, and ii) provide a protective pressure arch to restrain the deformation in an excavated void. Automated and precise tracking of rock bolt positions can assist with the operational success of ground support by reducing the amount of labour required in current manual practices. A point cloud is a large list of x, y and z coordinates in three-dimensional space, that when visualised, show a precise 3D representation of a physical environment. Point cloud data can be collected using accurate LiDAR scanners. These scanners can be mounted on a stationary tripod or on a moving vehicle. A key way point cloud data is used is in the bolt detection project being completed in conjunction with UNSW Sydney, which aims to provide an automated location of underground rock bolts from LiDAR scans. The immediate challenge with identifying rock bolts is that they constitute a very small portion of the data set amounting to approximately 0.01 per cent of some of the collected civil tunnel data. The process of detecting rock bolts from point clouds therefore requires a quick elimination of large background data while still preserving the bolt data. A further challenge is that the geometry of underground environments can often be very complex and non-uniform. To overcome these challenges, a two-step coarse to fine deep learning approach is taken. The first step is a coarse background elimination step aimed at removing as much of the background data as possible. The second step is a finer classification method, which uses deep learning to further segment the bolts from the remaining background. Additionally, a computer vision module is being implemented, which is expected to increase the detection rate by adding another dimension to the training data set.

Nov 29, 2022

By M Colwell

The Coal Mine Roof Rating (CMRR) is a rock (roof) mass classification (RMC) system/index that was originally developed by the United States Bureau of Mines (USBM) for aspects of underground coalmine geotechnical design and was first introduced to Australia as a part of the original Analysis of Longwall Tailgate Serviceability (ALTS) project conducted in 1997/1998. In subsequent years further refinement to its calculation was undertaken, such that the CMRR is now an industry standard for quantifying the structural competence of the primary bolted roadway roof interval and has been routinely/widely used by Australian collieries over the last 25 years with great success for roof support/chain pillar design. There have been many publications discussing the use and calculation of RMC systems and this paper also provides an important update and recommendations in how best to calculate the CMRR. However, it is rare to find an explanation as to why and how a RMC system/index satisfactorily takes account of the mechanics and rock mass behavioural issues associated with ground reinforcement. This paper provides such an explanation and insight with respect to the CMRR so it is: 1) better understood, 2) to allay any concerns, and therefore 3) will continue to be confidently used by and for the collieries’ benefit.

Nov 29, 2022

By Nikhil Dhawan, Shrey Agrawal

India’s most extensive renewable energy expansion program targets 280 GW of solar energy by 2030. Due to the massive generation of photovoltaic waste (expected 34,600 T by 2030), stringent recycling effort to recover metal resources from endof- life PVs is required for resource recovery, circular economy, and subsequent reduction in the environmental impact. The present study investigates the recycling potential of end-of-life silicon solar modules. The feed characterization reveals that most of the values in PV modules are concentrated in the frame (pure Al), Si wafer (72–84% Si, 9–12% Al, and 0.13–0.71% Ag), and electrical connections (Cu wire coated with Sn). The Si and Ag values in the feed were beneficiated using water fluidization, and concentrate with 90% Si and 2% Ag was attained in the underflow concentrate. Ag values were recovered in the form of AgCl by leaching in nitric acid and precipitation. After the purification step, the residue containing 99% Si can be further utilized to fabricate Si wafer and Al-Si alloy. The preliminary cost estimations suggest that the proposed process is economically viable for the recycling of solar panels with the maximum value generated from the AgCl precipitate followed by Al panels and Si wafers.

Nov 14, 2022

By JESSICA W. A. AZURE, Kansake A. Bruno, Samuel Frimpong, AZUPURI G. A. KABA, FORSYTH A. KADINGDI, PROSPER E. A. AYAWAH

Blasting generates undesirable ground vibration with adverse effects on humans, the environment, and structures. Numerous studies have used various techniques including machine learning (ML) to predict the peak particle velocity (PPV) caused by blasting in surface mines. These studies used the validation-set approach to evaluate the performance of their predictive models. In this study, Random Forest (RF), Gaussian Process (GP), and Gradient Boosting Machine (GBM) were applied for the prediction of PPV using data from a large open-pit gold mine in Ghana. An ensemble model based on a stacked generalization approach was then built on these three base models to significantly improve the PPV prediction performance. The model evaluation was done using repeated tenfold cross-validation which resolves the challenges with the validationset approach for small datasets. The dataset used in this study comprised a total of 196 measurements taken from different blasting operations. R-squared ( R2 ), mean absolute error (MAE), and root mean square error (RMSE) were used to evaluate the model performance. The results obtained from these models revealed that all the base models did not significantly outperform each other. However, the stacked generalization approach had a better performance, with RMSE = 0.12, R2 = 0.96, and MAE = 0.07. This approach presented an improvement over the existing methods and will be useful for highly accurate predictions of ground vibration during blast design. The application of this model will be useful for efficient and effective blast design.

Nov 8, 2022

By Zhonglin Ye, Guangping Dai, Ba Zhang, Yonggang Wei, Shiwei Zhou, Bo Li, Hua Wang

Decreasing copper content in slag is critical for improving copper recovery, where apparent viscosity is the main factor affecting the separation of matte/copper from the slag. The viscous behavior of copper converter slag was investigated during reduction process in this paper, and its non-Newtonian viscous behavior and shear thickening characteristics were clarified. The effects of silicate structure and solid particles in the melt on the slag apparent viscosity were investigated. As slag cleaning proceeded in an electric furnace, the magnetite gradually transformed into ferrous oxide, increasing the FeO/SiO2 ratio. With this ratio increased from 1.2 to 2.0, the slag structure was simplified, and its apparent viscosity decreased from 0.26 to 0.12 Pa·s at 1523 K. The iron phase was excessively reduced to metallic state; the slag contained metallic copper-iron alloy particles, increasing apparent viscosity. The apparent viscosity and phase structure of the slag corresponding to the iron phase reduction process were analyzed in detail; the results obtained in the present study are of great significance to the reasonable control of the converter copper slag cleaning process.

Nov 7, 2022

By Jianbo Hu, Tao Zheng, Zhao Qihua, Rui Su

Rock mass discontinuities play a significant role in evaluating the stability of rock slopes. It is necessary to collect and analyze these discontinuities to fully understand the mechanical and deformational behaviors of rock masses. Based on the digital images of some slope faces, this paper proposes a framework based on machine learning for detecting and extracting discontinuities. First, digital images and photo poses are combined, the original scene is reconstructed, and 3D point cloud data are obtained. All captured point clouds can be mapped with each image by the direct linear transformation algorithm. Next, the detection machine learning algorithm is used to segment rock mass discontinuities on 2D digital images. Finally, based on the mapping relationship, the 2D digital images that were segmented are transformed into 3D point cloud data. The geometric parameters of the corresponding discontinuities can be extracted by the DBSCAN and the Skmeans algorithm and the characterization of specific rules. The proposed framework was evaluated utilizing a quantity of slope digital images in the Yebatan hydropower under construction, located in Sichuan Province, China, as a case study. The results illustrate that most deviations are less than 10° for the dip direction and dip angle, which verifies the reliability of the proposed framework and maintains acceptable measurement accuracy.

Nov 6, 2022