Search Documents

By C Xu, B Li

Multi-sensor fusion visual perception technology significantly enhances visual perception capabilities in complex and harsh environments by combining visual data from different sensors. This technology is especially useful for scenarios where traditional single visible light sensors, such as standard cameras, struggle due to poor lighting, adverse visual conditions, or other visual obstructions. In environments with low light, smoke, or dust, the performance of conventional visible light sensors degrades, limiting their application. By integrating data from various sensors like radar, infrared thermal imaging, and others, multi-sensor fusion technology provides a more comprehensive and complementary visual solution. This fusion not only enhances the robustness of visual systems but also greatly expands their potential applications in fields like visual surveillance. In recent years, deep learning methods have become the mainstream technology for addressing multi-sensor fusion issues, demonstrating significant advantages in adaptive feature selection compared to traditional machine learning methods. However, deep learning-based multi-sensor fusion algorithms still face several challenges. The design of network structures is often redundant, lacking effective screening of useful components within multimodal information. Mainstream fusion algorithms focus excessively on improving display effects without adequately considering the needs of downstream applications and tasks. Existing fusion perception algorithms are generally designed for open visual scenes and lack targeted design for complex visual degradation factors present in tunnel environments of mines. Current deep learning methods for designing multi-sensor fusion networks heavily rely on manual experience to create fusion modules. This reliance increases the complexity of network design and can lead to redundancy. Redundant network modules are difficult to identify in end-to-end learning tasks at this stage, significantly slowing down network inference and potentially interfering with the output of fusion perception. Additionally, training these fusion modules typically requires large amounts of well-annotated multimodal data, further increasing implementation difficulty and cost. For instance, the fusion network design proposed by Li and Wu (2018) focuses on extracting multimodal fusion features, while Zhao et al (2020) introduced a feature decomposition mechanism in the feature fusion and extraction modules. Researchers like Zhang and Ma (2021), Xu et al (2020), and Liu et al (2017) have attempted to improve network structures through dense cascades and the introduction of residual connections. Although these methods have improved the handling of multimodal information to some extent, they still face limitations in distinguishing between useful and redundant information, with redundancy persisting in network designs. Ideally, the design of fusion networks should be more intelligent, guiding network structure design heuristically based on the importance of different modal information for the current task. This would reduce information redundancy and network weights, requiring the development of new algorithms or frameworks capable of automatically identifying and optimising fusion strategies, thereby reducing dependence on manual experience while improving fusion efficiency and network performance. Current deep learning designs for multi-sensor fusion algorithms often overlook the

Sep 1, 2024

Mines in the future deliver a paradigm shift in the way mining, material handling and processing are performed, unlocking significant value through production efficiency, flexibility and lower costs and enabling substantial safety and environmental benefits (V Schweikart). Mines in the future are highly productive, with low production costs and are significantly safer. They operate with a skeleton crew and use intelligent systems that fuse information from field instrumentation and other sources to generate the most efficient operating environment. Automated systems are seamlessly integrated across the mine and across the corporate enterprise. Geological models, long-term and short-term plans create optimised production schedules, suitably generated by the mine simulator. Production schedules are shared with the operations management system for deployment mainly by remotely operated and autonomous equipment. As the schedule is implemented, information is collected and shared with various plans, schedules, models and optimisation programs. This production environment is managed in real time from a æCentral Operations CentreÆ and it is visible across the business providing fast and accurate information about productivity, machine condition and environmental conditions. Central Operations Centres also house teams of subject matter experts that monitor and manage the mines and are able to communicate directly with the skeleton crew in the field. Constraints across the production line are displayed and managed appropriately from this location to maximise throughput. Maintenance is interactive and predictive. Reliability specialists in Central Operations Centres monitor the status of machinery and equipment and adjust maintenance plans and strategies to suit. Mining businesses rely heavily on knowledge workers û people with considerable knowledge and learning û and by manual workers. This new knowledge workforce is a team of associates rather than of bosses and subordinates. Field experts become leaders with tremendous autonomy regarding their work. The former line hierarchy becomes a support structure, intervening only in extreme emergencies; instead of giving orders, they are now removing barriers, expediting resources, conducting studies and acting as consultants.

Jan 1, 2009

By N Robson, E Stephens

Australian mining companies are actively exploring for and developing mineral deposits throughout the world with interests in emerging markets such as Africa, Asia and South America. It is important to understand emerging market risk as this will affect the willingness of mining companies and financial institutions to invest in a country, especially with reference to their key asset, the mining licence or operating/concession agreement. Mining companies looking to develop new projects in emerging markets have different risk profiles than projects in countries such as Australia and Canada. Examples of key risks faced in emerging markets include: political û expropriation, nationalisation, contract frustration, currency inconvertibility, war, terrorism, strikes, riots and civil commotion; and legal û mining legislation, arbitration and dispute resolution, trade agreements. There is a problem in the difference in risk perceptions between mining companies, financial institutions and insurers. This can cause different stakeholders to carry out different evaluations of the risks faced by a project, resulting in confusion about the risk margin attached to any project. In a time where resource nationalisation and energy security are prompting governments to become more involved in the mining sector we run through some recent examples of mining companies that have had problems due to the key risks highlighted earlier. Importance of carrying out the right due diligence to replace perceptions with reality: early identification of risks û detailed due diligence on emerging market risks that may arise will help with their identification and mitigation, make use of companies that can help identify risks and suggest solutions, control the cost of financing û risk margin applied to a project finance loan will be influenced by the risks faced, and risk management and using political risk insurance. There are plenty of opportunities in emerging markets but it is important to recognise the associated risks. Detailed due diligence at an early stage will provide long-term benefits.

Jan 1, 2007

Currently, mining in Australia underpins the economic vitality of the nation, generating massive revenue from commodity exports and providing substantial infrastructure and employment. In addition, it is predicted that mining related services will in the future develop a global market of their own that could eventually grow to rival, in value, the export trade in mineral commodities. There is however little room for complacency. In the short- to medium-term, escalating, vigorous competition can be expected from many previously underdeveloped countries that are reforming their own mining and mineral export industries. In addition, most mining companies are facing a modem-day renaissance in terms of new expectations, changing social norms, evolving employee relations and statutory obligations. Thus, along with whatever promise the future might hold, the mining industry is currently being subjected to a growing number of largely unprecedented problems and pressures, both economic and social, that are both highly complex and demanding. In particular, it remains a matter of on-going concern that relations between mining companies and their employees still, in spite of recent reforms, continue to constrain productivity, perpetuate inefficiencies, affect quality of work life and retard progress towards mutually beneficial goals and achievements. If mining in Australia is to attain and hold on to a commanding position on the global scene, then further changes will clearly need to be made. Realisation of this appears to be growing, with most mining companies having attempted in recent years to improve workplace culture. Based on a wide ranging study, this paper presents an overview of what has been learned to-date with regard to the `people factor' when it comes to implementing and progressing positive change and reform in mining. From this, a number of practical suggestions, arising from hard won lessons and experiences in the field, are offered. Attention to the nine `key principles' given should help to ensure that many of the common problems encountered to-date with change and reform in the mining industry are in future avoided or at least minimised.

Jan 1, 1997

At the outset I want to thank you for inviting a representative of the Chamber of Mines of South Africa to address you on the subject of mining and its future at this, the 13th Congress of the Council of Mining and Metallurgical Institutions. It is indeed a pleasure and an honour to be here. As many of you will recall, four years ago I, as the then President of the South African Institute of Mining and Metallurgy, had the privilege of hosting and welcoming you to the 12th Congress, the third such congress hosted by South Africa. The late Sir Ernest Oppenheimer, President of the 1930 Congress - the first to be held in South Africa - in his opening address foresaw the mining industry as "the most potent factor in spreading the benefits of civilisation over the sub-continent." We have come a long way since those halcyon days but, as I noted four years ago, it still remains true that mines have been, and will continue to be, the missionaries of development in many, if not most, of the Third World countries. I have been asked to address you on the subject of "mining" and to let my imagination roam widely so that delegates can leave the congress with a broader vision than they brought on arrival. It is a tall order and in my attempt to fulfil it I ask you to cast your minds back 120 years to a sultry, rustic day when the first diamond was discovered in South Africa at the now famous town of Kimberley. At that time the South African economy, predominantly pastoral, was still at subsistence level. Almost overnight the ensuing diamond rush turned the then British- owned northern Cape Colony's semi-desert scene into a hive of mining activity, every prospector bent on making his instant fortune. It was very little different, I am sure, to the gold rushes that hit California in 18+8, Australia in 1851, New Zealand in the 1860s. and the Yukon in the 1890s.

Jan 1, 1986

By T Sirinawin, R H. Murfitt, W Shywolup

The Bong Mieu Project lies within the Pre-Palaeozoic Kontum Craton near the southern margin of the regional Phuoc Son û Tam Ky Suture. The Suture and accompanying Truongson Fold Belt of Indosinian age host significant gold and base-metal bearing districts including OlympusÆ other project at Phuoc Son to the west of Bong Mieu and OxianaÆs Sepon project in Laos. The Bong Mieu Project is located some 90 km south of Da Nang and some 30 km south-west of Tam Ky, the provincial capital of Quang Nam province in central Vietnam. The project area is covered by an Investment Licence of some 30 km2 which was granted in 1991 and a mining permit of some 365 ha granted in 1992. Several mesothermal, gold-silver deposits with assigned resources are present on the property. Deposits with assigned resources contain some 615 000 ounces of gold in situ. These are at Ho Gan and Ho Ray which are potentially exploitable by open-pit mining and Nui Kem which is potentially exploitable by underground mining. Newly discovered extensions to the known deposits at Ho Gan, Ho Ray and Nui Kem, and several other new prospects, are currently being investigated by drilling. The Ho Gan mineralisation is of mesothermal, shear and shear breccia quartz vein, as well as intrusive breccia hosted style. Nui Kem mineralisation is of mesothermal, shear and shear breccia quartz vein style. The Ho Ray mineralisation is of skarn (gold-bismuth-tungsten) and lesser quartz vein style. The deposits straddle and flank the east-west trending Bong Mieu Anticline which trends through the central portion of the Investment Licence. In this paper the geology of the deposits, details of resources, results of recent exploration on the Investment Licence and recent engineering and mining studies at Ho Gan are briefly reviewed.

Jan 1, 2004

By Parr JM, Scott SD, Herzig PM, Gemmell JB

PACMANUS is an actively-forming seafloor analogue of volcanic-hosted massive sulfide orebodies, situated on the crest of an andesite-dacite-rhyodacite ridge overlying rifted older arc crust in a back-arc basin. Hydrothermal deposits range from Mn-oxide crusts to Cu-Zn-rich massive sulfide mounds and chimneys containing significant Au and Ag. Preliminary data suggest a substantial magmatic component to fluids and metal sources, and we speculate that PACMANUS is the cap to a complete hydrothermal system including subvolcanic and intrusive-related mineralisation.

Jan 1, 1995

In the Massif du Sud ophiolite complex platinum group elements occur predominantly as ultrafine-grained inclusions of rutheniridosmine, laurite and iridosmine. These form primary inclusions less than 5 to 10 um in diameter within chromite in podiform chromitite fringed by dunite, and emplaced subsequent to solidification of the host harzburgite. There is no specific association with trace inclusions of transition metal sulphides. The crystallization of then: Ir. Os and Ru minerals may have occurred concomitant with hromite, perhaps related to more reducing conditions produced by crystallization of the chromite. The Ir, Os. and Ru show pronounced fractionation from Pd, Pt and Rh expected to be also present in the mantle source rock. The dunite shows evidence of metasomatic origin from adjoining harzburgite, and the possibility of selective crystallization of the Ir sub-group elements suggests that Pd, Pt and Rh may have been transported elsewhere.

Jan 1, 1989

By Mauk JL, Arehart GB, Mathews SJ

The banded nature of the quartz veins which form the Empire lode at the Golden Cross mine, is perhaps their most striking feature. Petrographic studies show that significant differences occur between the bands in terms of textures, quartz grain size, calcite, adularia and kaolinite content, and the form, concentration and mineralogy of sulfide phases. The 818Oqu i values of individual bands within the vein material range from 7.3-10.9%x. Calculations show that the 818Oqäa,u values can be adequately explained by temporal variations in the system temperature, boiling and/or variations in the systems fluid/rock ratios.

Jan 1, 1995

The main asset of a mining company is the mineral resource; without resources there is no mining business and the infrastructure is useless. In this context the role of the mining geologist is to always keep the resource inventory to an optimum level for the mining business and to maximise the transformation of the resources into reserves in a multi- disciplinary work environment. In short, a ægood mining geologist has to always provide the best ore to the millÆ.To do his/her work properly, several technical skills are needed: to know how to observe and describe nature in order to optimise the data acquisition process, particularly when mapping; to have a good understanding of natural variability in order to develop proper sampling strategies and sample preparation protocols to minimise all the errors generated by the sampling process; to have a good understanding of assaying practices: their limitations as well as their assaying errors; to carefully and properly store the geological data and information and to preserve the geological patrimony; to fully understand ore forming processes to generate the best possible geoscientific models, which are fundamental for the mining business; to fully understand the consequences of the geoscientific models along the value chain from rock to cathodes; and to continuously validate and improve geological practices and models by continuous data collection and observation of geo- mining-metallurgy processes. Some very important behavioural competencies are also needed:long-term vision, passion for excellence and value generation, strategic influence, systemic and innovative thinking, and knowledge and experience transfer. Last but not least, it is important to point out that a deep knowledge of the mining business is essential to add value by using geology best practices.

Jan 1, 2006

By Yu H, Wan B

40 percent of the coal reserves and 40 percent of the ouput of coal in China is of thick coal seams. The sub-level caving in Yangguan has evolved together with the hydraulic supports, to provide recovery rates today of 80 to 85 percent, waste ratios less than 10 percent and daily outputs exceeding 7,000 tonnes. Matching of rate of shearing bottom coal to caving of top coal is most important to achieve "simultaneous" operations. Spontaneous combustion in the goat, gas, including in the goal, and dust, are particular safety hazards which are combatted with various stringent measures.

Jan 1, 1991

By Brathwaite RL

The gold-scheelite-quartz lodes of the Wakamarina valley were the most productive of the mesothermal gold-bear in the Marlborough Schist. Most of the production came from the Empire City and Golden Bar mines which worked tug, Golden Bar lode. Only 134 m of its 1800 m strike length are accessible, in the No. 2 level of the Golden Bar Mine. A fau I has downthrown the Golden Bar Mine section by 2 47 m and displaced it left-laterally by 18 m. The lode is a typical ribbon-banded mesothermal vein, averaging c. 1.75 m wide, and enclosing narrow horses of schist country rock. It i composed of bands of milky "buck" quartz with thin laminae of schist. Scheelite occurs in short strings of grains withi quartz on the footwall side of the lode and also in small quartz-carbonate veinlets in the foot wall and hanging wall schist The lode quartz contains only localised traces of pyrite and arsenopyrite. They are more abundant in the wall-rock schist.

Jan 1, 1995

By Quemard D, Troncoso J

The Gordonstone Project with a capital cost in excess of $500 million will become the largest underground coal mine in Australia. The project development decision was taken on 31 July, 1990 and at 31 May, 1991 construction was approximately 35 per cent complete. The first train of processed product coal will be railed to Gladstone during December 1992. The general theme of this paper therefore will address the planning and development processes at Gordonstone particularly the interrelatedness of technical, production, construction and environmental disciplines to ensure success during operations.

Jan 1, 1991

By D Craw

Interactions between structure and lithology in the Hyde-Macraes Shear Zone controlled the style of gold mineralisation at the Macraes mine. Large volumes of rock in some mine pits have most gold disseminated in hydrothermally altered schist, rather than in quartz veins. Mineralisation was initiated in the latter stages of metamorphism. Pods of massive schist rich in quartz and feldspar are surrounded by more fissile schist consisting of interlayered micaceous and quartzofeldspathic schist. Later shears have been partly controlled by the micaceous schist, and anastomose around pods of more massive schist. Micaceous schist was replaced by pyrite and arsenopyrite, accompanied by recrystallisation of phengitic muscovite under lower greenschist facies conditions. Gold occurs in sulfides and with microcrystalline quartz formed during minor silicification. Massive schist is locally mineralised along sulfidic microshears which have phengitic muscovite largely illitised under sub-greenschist facies conditions. Mineralised massive and micaceous schists were overprinted by cataclastic shears, with addition of graphite and more sulfides. Mineralisation occurred during a transition from ductile to brittle deformation regimes as the schist was uplifted in the late Jurassic.

Jan 1, 2004

Lake Kopiago is located some 71km northwest of Tari in the Southern Highlands Province and lies on the northern edge of the Papuan Fold Belt. The area is underlain by middle Miocene carbonate and fine grained clastic sediments intruded by late Miocene hornblende gabbros.Two mineral deposits occur to the south of Lake Kopiago. These are the Bisamu gold prospect and the Wunaki zinc-lead gossan.The Bisamu deposit consists of a calcic skarn assemblage which is geochemically anomalous in gold, silver, copper, zinc, cobalt, bismuth and tellurium. The prograde mineral assemblage includes garnet, clino-pyroxene and epidote whereas the retrograde assemblage includes epidote, quartz, actinolite and chlorite. Pyrrhotite is the dominant sulphide with minor pyrite and marcasite and trace chalcopyrite and sphalerite. Minor magnetite is also present. The Bisamu skarn shows many of the features typical of gold-skarn type mineral-isation.

Jan 1, 1994

By Espinosa-Gomez R, Johnson NW

The Copper Concentrator at Mount Isa Mines Limited treats on average 18 000 tonnes/day of chalcopyrite ore containing typically three per cent copper. The main gangue minerals are iron sulphides (-13 per cent) and non sulphide gangue (-78 per cent). In August, 1991, the comminution flowsheet was changed from a conventional circuit (crusher/rod mill/two stage ball milling) to a modem circuit (fully autogenous grinding/one stage ball milling). The 80 per cent passing size of the flotation feed with the conventional circuit was 125 microns. The Copper Concentrator has two parallel grinding lines, two roughing and scavenging flotation lines, and common cleaning, recleaning and retreatment sections. During the first six weeks of the installation phase, one of the two flotation lines was fed by the conventional grinding circuit and the other by the autogenous grinding circuit. Data from parallel surveys and historical data are used to analyse the rougher/scavenger flotation performance before and after the change. The solid and mineral distributions of the flotation feed, the pulp chemistry and the reagent requirements are compared. The mineral liberation before and after the installation of autogenous grinding is compared on a sized basis using data from the 14 consecutive months before, and the seven months after the change. Liberation was measured using the QEM*SEM(*) instrument. The overall effect of the new grinding circuit on the plant metallurgy is also assessed using the copper recovery values, the dilution of the final concentrate and the distribution of copper losses in the final tailings. (*) Quantitative Evaluation of Materials by Scanning Electron Microscopy

Jan 1, 1993

Reconnaissance geochemical exploration in the vicinity of low angle detachment faults which bound the core of the Paparoa metamorphic core complex yielded Au up to 14 ppm, with 35% of greywacke, granitoid and quartz vein samples in the northern range exceeding the 0.01 ppm detection limit. Au is associated with Ag, As, Sb and Cu anomalies consistent with epithermal-type mineralisation. The Ag and Cu anomalies also distinguish this mineralisation from older, mesothermal Au deposits widespread in the Ordovician 'slate belt' greywackes, which comprise much of the upper plate of the core complex. Baryte, molybdenite, fluorite and stratiform U mineralisation in the northern Paparoa Range may also be related to hydrothermal activity along the detachment fault zones.

Jan 1, 1995

The Permian rocks at Gympie were first studied as far back as 1868 in conjunction with the discovery of gold. The relationship of the Gympie Group to other sequences in the Gympie Composite Terrane is an important subject of investigations in a new mapping project of the Geological Survey Program of the Department of Resource Industries.In two areas the Gympie Group has been divided into four defined formations. They are the Highbury Volcanics (base), Ram mutt Formation, South Curra Limestone and Tamaree Formation (top) in the Gympie area, and the Mant basalt (base), Kolbar Formation, Gigoomgan Limestone and Teebar Formation (top) in the Gigoomgan area. Geochemical and lithological evidence from the basal volcanics of the Gympie Group (Highbury Volcanics, Mant basalt) indicate that they developed in an intraoceanic island arc setting. The overlying volcaniclastic sediments (Rammutt Formation, Kolbar Formation), which contain some continentally-derived material, are mainly of local derivation and shallow-water marine. A hiatus is present at the top of the Early Permian Rammutt Formation and Kolbar Formation. Following the hiatus, limestone sedimentation characteristic of a basin of restricted circulation developed in the Late Permian. There is some evidence of a hiatus between the limestone deposition of the Gigoomgan and South Curra Limestones and the overlying probable turbidite deposits of the Teebar and Tamaree Pnrmatinnc

Jan 1, 1990

By Donchak PJ, Lang SC, Hammond RL, Bultitude RJ

The Hodgkinson and Broken River Provinces form the northern part of the Tasman Orogen. They are separated by a broad belt of Late Palaeozoic igneous rocks, but were probably originally continuous. They are bounded to the west by a system of major faults (Fig. 1) which separate them from the Georgetown Province. Although largely Precambrian, the Georgetown Province includes probable Siluro- Devonian granitoids, and in the south adjacent to the Broken River Province, metamorphosed Early Palaeozoic(?) volcanics and sediments of the Greenvale Subprovince. South of the Broken River Province, the Lolworth-Ravenswood Province includes probable Precambrian rocks, Cambro-Ordovician volcanics and sediments, and voluminous Ordovician to Early Devonian granitoids. The first systematic studies of the provinces were the 1:250 000 scale mapping programs by the Bureau of Mineral Resources and Geological Survey of Queensland (White, 1965; de Keyser and Lucas, 1968). These and subsequent regional studies were reviewed by Arnold and Fawckner (1980). Remapping by the Geological Survey of Queensland since 1982, and a concurrent regional study of the deformation by Hammond (1986) have provided much new data. We acknowledge the contribution of other members of the GSQ field parties - L. Cranfield, J. Domagala, B.G. Fordham, P. Green, R. Halfpenny, R. Hegarty, L. Hutton, S. Law, D. Macansh, T. McLennan and M. Scott.

Jan 1, 1987

By Allen RL, Clifford BA, Wright JV, Bull S

The Bunga Beds of the Late Devonian - Boyd Volcanic' Complex is a well exposed succession of rhyolite, sedimentary rocks and minor basalt outcropping along the south coast of New South Wales. The sedimentary rocks vary from shallow water basin margin facies, to deep-water basin centre facies, and the majority of .the volcanics are associated with the deep water facies association. The volcanics are dominated by rhyolite domes, lavas and cryptodomes, whose margins are extensively autobrecciated and quench fragmented. In situ pyroclastics represent a very minor volcanic component and are restricted to the remnants of two small doors- top tuff cone successions. Pyroclastic debris is common in the sedimentary succession enclosing the volcanics and was clearly derived from shallow to emergent basin margin areas through mass-flow redeposition. The succession has bearing on the origin of the fragmental component of, the volcanics that host Kuroko- type massive sulphide deposits. The characteristics of these fragmental volcanics are not well documented. We suggest that based on the Bunga Beds,'Kuroko-type volcanic successions are dominated by coherent rhyolite/ felsic lavas and intrusive domes, autobreccias and hyaloclastites, that in situ pyroclastics are rare or lacking, and that redeposited volcanic debris is abundant. Pyroclastic debris is most likely to have been derived from shallow areas marginal to the deep-water volcanic centre that produce the Kuroko-type volcanics and sulphides.

Jan 1, 1987