Using deep neural networks coupled with principal component analysis for ore production forecasting at open-pit mines

Ore production is usually affected by multiple influencing inputs at open-pit mines.Nevertheless,the complex nonlinear relationships between these inputs and ore production remain unclear.This becomes even more challenging when training data(e.g.truck haulage information and weather conditions)are massive.In machine learning(ML)algorithms,deep neural network(DNN)is a superior method for processing nonlinear and massive data by adjusting the amount of neurons and hidden layers.This study adopted DNN to forecast ore production using truck haulage information and weather conditions at open-pit mines as training data.Before the prediction models were built,principal component analysis(PCA)was employed to reduce the data dimensionality and eliminate the multicollinearity among highly correlated input variables.To verify the superiority of DNN,three ANNs containing only one hidden layer and six traditional ML models were established as benchmark models.The DNN model with multiple hidden layers performed better than the ANN models with a single hidden layer.The DNN model outperformed the extensively applied benchmark models in predicting ore production.This can provide engineers and researchers with an accurate method to forecast ore production,which helps make sound budgetary decisions and mine planning at open-pit mines.

Software Vulnerability Mining and Analysis Based on Deep Learning

In recent years,the rapid development of computer software has led to numerous security problems,particularly software vulnerabilities.These flaws can cause significant harm to users’privacy and property.Current security defect detection technology relies on manual or professional reasoning,leading to missed detection and high false detection rates.Artificial intelligence technology has led to the development of neural network models based on machine learning or deep learning to intelligently mine holes,reducing missed alarms and false alarms.So,this project aims to study Java source code defect detection methods for defects like null pointer reference exception,XSS(Transform),and Structured Query Language(SQL)injection.Also,the project uses open-source Javalang to translate the Java source code,conducts a deep search on the AST to obtain the empty syntax feature library,and converts the Java source code into a dependency graph.The feature vector is then used as the learning target for the neural network.Four types of Convolutional Neural Networks(CNN),Long Short-Term Memory(LSTM),Bi-directional Long Short-Term Memory(BiLSTM),and Attention Mechanism+Bidirectional LSTM,are used to investigate various code defects,including blank pointer reference exception,XSS,and SQL injection defects.Experimental results show that the attention mechanism in two-dimensional BLSTM is the most effective for object recognition,verifying the correctness of the method.

Solution to the Horizontal Crossing Problem of Transportation System at Key Points in Open-Pit Mines

In order to reduce the horizontal crossing transportation problems between coal trucks and stripping trucks,large and small vehicles,and transport trucks and belt conveyors at key points of open pit mine in production,the separate transportation mode of underpass bridge and overpass steel trestle is proposed to optimize the open pit development transportation system,so as to solve the practical problems that the horizontal cross of transport vehicles causes vehicle blockage,affects production schedule and production safety.The results show that the horizontal crossing road can be changed into a separate type of overpass steel trestle,which can realize the classified transportation of large and small vehicles,reduce the traffic density,make vehicles with different functions go their own way,eliminate the hidden danger of traffic accidents,and improve the production efficiency.

Deep seabed mining:Frontiers in engineering geology and environment

Ocean mining activities have been ongoing for nearly 70 years,making great contributions to industrialization.Given the increasing demand for energy,along with the restructuring of the energy supply catalyzed by efforts to achieve a low-carbon economy,deep seabed mining will play an important role in addressing energy-and resource-related problems in the future.However,deep seabed mining remains in the exploratory stage,with many challenges presented by the high-pressure,low-temperature,and complex geologic and hydrodynamic environments in deep-sea mining areas,which are inaccessible to human activities.Thus,considerable efforts are required to ensure sustainable,economic,reliable,and safe deep seabed mining.This study reviews the latest advances in marine engineering geology and the environment related to deep-sea min-ing activities,presents a bibliometric analysis of the development of ocean mineral resources since the 1950s,summarizes the development,theory,and issues related to techniques for the three stages of ocean mining(i.e.,exploration,extraction,and closure),and discusses the engineering geology environment,geological disasters,in-situ monitoring techniques,envi-ronmental protection requirements,and environmental effects in detail.Finally,this paper gives some key conclusions and future perspectives to provide insights for subsequent studies and commercial mining operations.

Artificial Fish Swarm Optimization with Deep Learning Enabled Opinion Mining Approach

Sentiment analysis or opinion mining(OM)concepts become familiar due to advances in networking technologies and social media.Recently,massive amount of text has been generated over Internet daily which makes the pattern recognition and decision making process difficult.Since OM find useful in business sectors to improve the quality of the product as well as services,machine learning(ML)and deep learning(DL)models can be considered into account.Besides,the hyperparameters involved in the DL models necessitate proper adjustment process to boost the classification process.Therefore,in this paper,a new Artificial Fish Swarm Optimization with Bidirectional Long Short Term Memory(AFSO-BLSTM)model has been developed for OM process.The major intention of the AFSO-BLSTM model is to effectively mine the opinions present in the textual data.In addition,the AFSO-BLSTM model undergoes pre-processing and TF-IFD based feature extraction process.Besides,BLSTM model is employed for the effectual detection and classification of opinions.Finally,the AFSO algorithm is utilized for effective hyperparameter adjustment process of the BLSTM model,shows the novelty of the work.A complete simulation study of the AFSO-BLSTM model is validated using benchmark dataset and the obtained experimental values revealed the high potential of the AFSO-BLSTM model on mining opinions.

Theoretical design and dynamic simulation of new mining paths of tracked miner on deep seafloor

With comprehensive considerations of the operational safety and collection efficiency for the tracked miner collecting the seafloor poly-metallic nodules, two new improved mining paths for the miner on the deep seafloor were proposed. Compared to the conventional mining path, the design principles and superiorities of the two new paths are that the miner turning with relative long radius should avoid large sinkage and high slip, so as to ensure its operational safety, while the space between its straight-line trajectories before and after the turning is optimum, which is designed as the total width of the miner, and collect nodules as more as possible, so as to ensure its collection efficiency. To realize the new mining paths, theoretical designs and quantitative calculations were carried out to determine the exact positions for the speed controls of the miner during its whole operation process. With the new dynamic model of the miner, and through regulations of the speeds of the left and right tracks of the miner on the exact motion positions according to the theoretical calculations, the two new improved mining paths for the miner on the seafloor were successfully simulated, thus the turning radius of the miner in the simulation is about 21.8 m, while the distance between the straight-line trajectories before and after the turning is about 5.2 m. The dynamic simulation results preliminarily prove the feasibility of these two new mining paths, and further can provide important theoretical guidance and useful technical reference for the practical tracked miner operation and control on the seafloor.

Optimization of mining method in subsea deep gold mines: A case study

The mining method optimization in subsea deep gold mines was studied. First, an index system for subsea mining method selection was established based on technical feasibility, security status, economic benefit, and management complexity. Next, an evaluation matrix containing crisp numbers and triangular fuzzy numbers(TFNs) was constructed to describe quantitative and qualitative information simultaneously. Then, a hybrid model combining fuzzy theory and the Tomada de Decis?o Interativa Multicritério(TODIM) method was proposed. Finally, the feasibility of the proposed approach was validated by an illustrative example of selecting the optimal mining method in the Sanshandao Gold Mine(China). The robustness of this approach was demonstrated through a sensitivity analysis. The results show that the proposed hybrid TODIM method is reliable and stable for choosing the optimal mining method in subsea deep gold mines and provides references for mining method optimization in other similar undersea mines.

Study on Critical,Modern Technology for Mining in Gassy Deep Mines

To achieve safe and highly efficient mining in the gassy, deep mines of the Huainan collieries simultaneous coal and gas extraction, and the corresponding ventilation methods were developed. This includes a set of mining procedures and principles which help insure safe and efficient production. Furthermore, green mining, meaning the comprehensive use of emitted gas, proper treatment of the environment and appropriate mine temperature control, is now standard. The concepts of modem mining and the principles of pressure relief are described. Coal-gas simultaneous ex- traction and multi-pressure relief techniques were developed which require a combination of surface and underground gas extraction. The application of Y-ventilation systems, of roadways retained along goafs, of stress control techniques for highly fragile mine roofs and of powerful, automatic and reliable mining equipment contributes to safe operation of modem deep mines. Operating parameters for these techniques are described and the results of their use discussed.

Key factor analysis and model establishment of variation of rock face temperature in a deep open-pit mine

In recent years, with the increase of the depth of open-pit mining, the pollution level has been on the rise due to harmful gases and dust occurring in the process of mining. In order to accelerate the diffusion of these air pollutants, the distributed regularity of the rock face temperature which is directly related to the air ventilation in deep open-pit mines should be studied. Here, we establish the key factors influencing the rock face temperature in a deep open-pit mine. We also present an empirical model of the rock face temperature variation in the deep open-pit mine, of which the performance is interestingly high compared with that of the field test. This study lays a foundation to study the ventilation thermodynamic theory in the deep open-pit mine, which is of great importance for theoretical studies and engineering applications of solving air pollution problem in deep open-pit mines.

Ground behaviour analysis,support system design and construction strategies in deep hard rock mining-Justified in Western Australian’s mines

Development of deep underground mining projects is crucial for optimum extraction of mineral deposits.The main challenges at great depth are high rock stress levels,seismic events,large-scale deformation,sudden failures and high temperatures that may cause abrupt and unpredictable instability and collapse over a large scale.In this paper,a ground control and management strategy was presented corresponding to the three stages of projects:strategic design,tactical design and operational design.Strategic design is results in preparing a broad plan and primary design for mining excavations.The tactical design is to provide detail design such as stabilisation methods.Operational design stage is related to monitoring and updating design parameters.The most effective ground control strategies in this stage are maintenance,rehabilitation,monitoring and contingency plan.Additionally,a new procedure for design of ground support systems for deep and hard rock was proposed.The main principles are:static and/or dynamic loading types,determination of loading sources,characterisation of geological conditions and the effects of orientation of major structures with openings,estimation of ground loading factor,identification of potential primary and secondary failures,utilisation of appropriate design analysis methods,estimation of depth failure,calculation of the static and/or dynamic demand ground support capacity,and selection of surface and reinforcement elements.Gravitational force is the dominant loading force in low-level stresses.In high stress level failure mechanism becomes more complex in rock mass structures.In this condition,a variety of factors such as release of stored energy due to seismic events,stress concentration,and major structures influence on ground behaviour and judgement are very complicated.The key rock engineering schemes to minimise the risk of failures in high-stress levels at great depth involve depressurisation and quality control of materials.Microseismic and blast monitoring throughout the mining operations are required to control sudden failures.Proper excavation sequences in underground stopes based on top-down,bottom-up,centre-out and abutment-centre were discussed.Also,the performance of a ground support system was examined by field observation monitoring systems for controlling and modifying ground support elements.The important outcome of the research is that the proposed procedure of selecting ground support systems for static and dynamic situations was applied in several deep underground mines in Western Australia.Ground behaviour modes and failure mechanism were identified and assessed.Ground demand for static and dynamic conditions was estimated and an appropriate ground support system was selected and evaluated in site-specific conditions according to proposed method for ground support design at great depth.The stability of rock masses was confirmed,and the reliability of the design methodology for great depth and hard rock conditions was also justified.

Prediction of Rock Burst with Deep Mining Excavation in Linglong Gold Mine

To predict rock burst in deep mining excavation in Linglong gold mine, systematical laboratory tests of mechanical properties of rock, in situ stress measurement and 3-D FEM analysis on energy distribution in rock mass surrounding deep mining rooms were carried out. According to various prediction criteria of rock burst, it is concluded that rock burst is liable to occur during deep mining excavation in the mine.

Human Verification over Activity Analysis via Deep Data Mining

Human verification and activity analysis(HVAA)are primarily employed to observe,track,and monitor human motion patterns using redgreen-blue(RGB)images and videos.Interpreting human interaction using RGB images is one of the most complex machine learning tasks in recent times.Numerous models rely on various parameters,such as the detection rate,position,and direction of human body components in RGB images.This paper presents robust human activity analysis for event recognition via the extraction of contextual intelligence-based features.To use human interaction image sequences as input data,we first perform a few denoising steps.Then,human-to-human analyses are employed to deliver more precise results.This phase follows feature engineering techniques,including diverse feature selection.Next,we used the graph mining method for feature optimization and AdaBoost for classification.We tested our proposed HVAA model on two benchmark datasets.The testing of the proposed HVAA system exhibited a mean accuracy of 92.15%for the Sport Videos in theWild(SVW)dataset.The second benchmark dataset,UT-interaction,had a mean accuracy of 92.83%.Therefore,these results demonstrated a better recognition rate and outperformed other novel techniques in body part tracking and event detection.The proposed HVAA system can be utilized in numerous real-world applications including,healthcare,surveillance,task monitoring,atomic actions,gesture and posture analysis.

Study of a low-disturbance pressure-preserving corer and its coring performance in deep coal mining conditions

With the increasing depth of coal mining,the requirements for coring devices that maintain pressure are increasing.To adapt to the special environment in deep coal seams and improve the accuracy of testing gas content,a low-disturbance pressure-preserving corer was developed.The measurement of gas content using this corer was analyzed.The coring test platform was used to complete a coring function test.A pressurized core with a diameter of 50 mm was obtained.The pressure was 0.15 MPa,which was equal to the pressure of the liquid column of the cored layer,indicating that the corer can be successfully used in a mud environment.Next,a pressure test of the corer was conducted.The results showed that under conditions of low pressure(8 MPa)and high pressure(25 MPa),the internal pressure of the corer remained stable for more than 1 h,indicating that the corer has good ability to maintain pressure.Therefore,the corer can be applied at deep coal mine sites.The results of this research can be used to promote the safe exploitation of deep coal mines and the exploitation of methane resources in coalbeds.

Feasibility Study on Continuous Mining Method in Deep Position of Jinchuan Nickel Mine, China

Jinchuan nickel mine is the largest nickel mine in China. Cut-and-fill mining method with high density cementing materials is used in the mine. The original mining design divided the mining operation into two steps. The first step stopped the mining rooms and the second step stopped the pillars. Because the two-step method made big trouble for finally mining pillars and strongly limited the mining speed and production, it was successfully changed to a continuous cut-and-fill method without pillars. However, the mining operation in the mine has been down to 800 m and the mining condition is getting worse and more complicated. Through systematical field investigations and 3-D FEM analysis, it is proved that the mining method without pillars is feasible for mining deeper orebodies in Jinchuan nickel mine.

Failure mechanism of pump chambers and their optimized design in deep mining at Qishan Coal Mine

Pump chambers, normally used as dominant structures in mining engineering to insure the safety and production of un-derground coal mines, become generally deformed under conditions of deep mining. Given the geology and engineering condition of Qishan Coal Mine in Xuzhou, the failure characteristics of pump chambers at the –1000 m level show that the main cause can be attributed to the spatial effect induced by intersectional chambers, where one pump is constructed per well. We developed an opti-mized design of the pump room, in which the pump wells in the traditional design are integrated into one compounding well. We suggest that the new design can limit the spatial effect of intersectional chambers during construction given our relevant numerical simulation. The new design is able to simplify the structure of the pump chamber and reduce the amount of excavation required. Based on a bolt-mesh-anchor with a rigid gap coupling supporting technology, the stability of pump chamber can be improved greatly.

Forecasting the Academic Performance by Leveraging Educational Data Mining

The study aims to recognize how efficiently Educational DataMining(EDM)integrates into Artificial Intelligence(AI)to develop skills for predicting students’performance.The study used a survey questionnaire and collected data from 300 undergraduate students of Al Neelain University.The first step’s initial population placements were created using Particle Swarm Optimization(PSO).Then,using adaptive feature space search,Educational Grey Wolf Optimization(EGWO)was employed to choose the optimal attribute combination.The second stage uses the SVMclassifier to forecast classification accuracy.Different classifiers were utilized to evaluate the performance of students.According to the results,it was revealed that AI could forecast the final grades of students with an accuracy rate of 97%on the test dataset.Furthermore,the present study showed that successful students could be selected by the Decision Tree model with an efficiency rate of 87.50%and could be categorized as having equal information ratio gain after the semester.While the random forest provided an accuracy of 28%.These findings indicate the higher accuracy rate in the results when these models were implemented on the data set which provides significantly accurate results as compared to a linear regression model with accuracy(12%).The study concluded that the methodology used in this study can prove to be helpful for students and teachers in upgrading academic performance,reducing chances of failure,and taking appropriate steps at the right time to raise the standards of education.The study also motivates academics to assess and discover EDM at several other universities.

Study on ascending mining roadway layout of close distance coal seams in deep mine

To solve the problems appeared in mining process of No.2 seam, the ascending stress-releasing mining method was adopted. Studying on the reasonable layout of actual mining roadway in upper coal seams is the precondition of successful ascending mining. By using ＂device of leak measuring by blocking up double ends＂, it detected the height of overburden water flowing fractured zone originated from sub-coal seams mining. Thus it proved that the actual mining roadway of No.2 upper ascending seam was located in the smooth sagging zone. On the basis of analyzing the stress-releasing effect of sub-coal seams mining to upper coal seams by using RFPA software, it analyzed the stability of up-face coal seams and the reasonable location of starting cut in up-face coal seams. It also analyzed the reasonable gateway location in upper coal seams, which ensured the crossheading in upper coal seams out of the effect of sub-coal work face mining by using theory of underground pressure. Meanwhile, the reasonable pillars dimensions in upper coal seams by building the structure mechanics model of stope were researched. It can make the roadway driven along next goaf to be located in low stress zone, and be beneficial to keeping roads stable owing to less stress of surrounding rock. Finally, it tested the rationality of the layout method of roads in upper coal seams by engineering field measurement in 3221 working face.

基于改进DeepLabv3+和高分辨率影像的露天矿区土地覆盖分类

深度学习技术和高分辨率遥感影像为矿区的土地覆盖分类提供了高效的技术手段,针对原始DeepLabv3+深度学习模型在矿区土地覆盖分类中的地物边缘分割不准确和错分、漏分等问题,本文通过在编码层进行特征融合、骨干网络添加监督分支模块和解码层融合浅层特征的方式对原始的DeepLabv3+模型进行改进。实验结果表明,改进后的DeepLabv3+模型的mIoU值平均为80.32%,相比原始的DeepLabv3+模型提升了5.71%,能够有效地进行露天矿区的土地覆盖分类。

Failure characteristics and its influencing factors of talus-derived rock mass during open-pit mining

The failure characteristic of talus-derived rock mass continues to challenge quantitative hazard assessments in open-pit mining. Physical model test was used to assess the failure modes and mechanisms on talus-derived rock mass. The different types of failure modes of the talus-derived rock mass were introduced and a possible failure mechanism relation between the failure zone and the structure of the talus-derived rock mass was also shown. The physical model test results indicate that the rainfall has significant influence on the stability and failure modes of talus-derived rock mass during open-pit mining. The development of the seepage area caused by rainfall initiates the localized failure in that particular area, and the initiation of localized instability is mainly induced by stress changes concentrated in the seepage area.

Failure mechanism and coupled static-dynamic loading theory in deep hard rock mining: A review

Rock failure phenomena,such as rockburst,slabbing(or spalling) and zonal disintegration,related to deep underground excavation of hard rocks are frequently reported and pose a great threat to deep mining.Currently,the explanation for these failure phenomena using existing dynamic or static rock mechanics theory is not straightforward.In this study,new theory and testing method for deep underground rock mass under coupled static-dynamic loading are introduced.Two types of coupled loading modes,i.e.'critical static stress + slight disturbance' and 'elastic static stress + impact disturbance',are proposed,and associated test devices are developed.Rockburst phenomena of hard rocks under coupled static-dynamic loading are successfully reproduced in the laboratory,and the rockburst mechanism and related criteria are demonstrated.The results of true triaxial unloading compression tests on granite and red sandstone indicate that the unloading can induce slabbing when the confining pressure exceeds a certain threshold,and the slabbing failure strength is lower than the shear failure strength according to the conventional Mohr-Column criterion.Numerical results indicate that the rock unloading failure response under different in situ stresses and unloading rates can be characterized by an equivalent strain energy density.In addition,we present a new microseismic source location method without premeasuring the sound wave velocity in rock mass,which can efficiently and accurately locate the rock failure in hard rock mines.Also,a new idea for deep hard rock mining using a non-explosive continuous mining method is briefly introduced.