Microseismic event waveform classification using CNN-based transfer learning models

The efficient processing of large amounts of data collected by the microseismic monitoring system(MMS),especially the rapid identification of microseismic events in explosions and noise,is essential for mine disaster prevention.Currently,this work is primarily performed by skilled technicians,which results in severe workloads and inefficiency.In this paper,CNN-based transfer learning combined with computer vision technology was used to achieve automatic recognition and classification of multichannel microseismic signal waveforms.First,data collected by MMS was generated into 6-channel original waveforms based on events.After that,sample data sets of microseismic events,blasts,drillings,and noises were established through manual identification.These datasets were split into training sets and test sets according to a certain proportion,and transfer learning was performed on AlexNet,GoogLeNet,and ResNet50 pre-training network models,respectively.After training and tuning,optimal models were retained and compared with support vector machine classification.Results show that transfer learning models perform well on different test sets.Overall,GoogLeNet performed best,with a recognition accuracy of 99.8%.Finally,the possible effects of the number of training sets and the imbalance of different types of sample data on the accuracy and effectiveness of classification models were discussed.

Construction of well logging knowledge graph and intelligent identification method of hydrocarbon-bearing formation

Based on the well logging knowledge graph of hydrocarbon-bearing formation(HBF),a Knowledge-Powered Neural Network Formation Evaluation model(KPNFE)has been proposed.It has the following functions:(1)extracting characteristic parameters describing HBF in multiple dimensions and multiple scales;(2)showing the characteristic parameter-related entities,relationships,and attributes as vectors via graph embedding technique;(3)intelligently identifying HBF;(4)seamlessly integrating expertise into the intelligent computing to establish the assessment system and ranking algorithm for potential pay recommendation.Taking 547 wells encountered the low porosity and low permeability Chang 6 Member of Triassic in the Jiyuan Block of Ordos Basin,NW China as objects,80%of the wells were randomly selected as the training dataset and the remainder as the validation dataset.The KPNFE prediction results on the validation dataset had a coincidence rate of 94.43%with the expert interpretation results and a coincidence rate of 84.38%for all the oil testing layers,which is 13 percentage points higher in accuracy and over 100 times faster than the primary conventional interpretation.In addition,a number of potential pays likely to produce industrial oil were recommended.The KPNFE model effectively inherits,carries forward and improves the expert knowledge,nicely solving the robustness problem in HBF identification.The KPNFE,with good interpretability and high accuracy of computation results,is a powerful technical means for efficient and high-quality well logging re-evaluation of old wells in mature oilfields.

RADAR TARGET IDENTIFICATION BY ADAPTIVE DISCRIMINATION WAVEFORM SYNTHESIS AND NEAREST NEIGHBOR NEURAL NETWORK

In this paper,a new radar target identification scheme is presented based on adaptivediscrimination waveform synthesis and a nearest neighbor neural network.It can directly use theimpulse response of the target to synthesize discrimination waveform,so the poles extractionprocedure is not required.Particularly,it can successfully operate on the case that the poles ofthe target are weakly dependent on the aspect angle.

An intelligent identification method of safety risk while drilling in gas drilling

In view of the shortcomings of current intelligent drilling technology in drilling condition representation, sample collection, data processing and feature extraction, an intelligent identification method of safety risk while drilling was established. The correlation analysis method was used to determine correlation parameters indicating gas drilling safety risk. By collecting monitoring data in the safety risk period of more than 20 wells, a sample database of a variety of safety risks in gas drilling was established, and the number of samples was expanded by using the method of few-shot learning. According to the forms of gas drilling monitoring data samples, a two-layer convolution neural network architecture was designed, and multiple convolution cores of different sizes and weights were set to realize the vertical and horizontal convolution computations of samples to extract and learn the variation law and correlation characteristics of multiple monitoring parameters. Finally, based on the training results of neural network, samples of different kinds of safety risks were selected to enhance the recognition accuracy. Compared with the traditional BP(error back propagation) full-connected neural network architecture, this method can more deeply and effectively identify safety risk characteristics in gas drilling, and thus identify and predict risks in advance, which is conducive to avoid and quickly solve safety risks while drilling. Field application has proved that this method has an identification accuracy of various safety risks while drilling in the process of gas drilling of about 90% and is practical.

A real-time intelligent lithology identification method based on a dynamic felling strategy weighted random forest algorithm

Real-time intelligent lithology identification while drilling is vital to realizing downhole closed-loop drilling. The complex and changeable geological environment in the drilling makes lithology identification face many challenges. This paper studies the problems of difficult feature information extraction,low precision of thin-layer identification and limited applicability of the model in intelligent lithologic identification. The author tries to improve the comprehensive performance of the lithology identification model from three aspects: data feature extraction, class balance, and model design. A new real-time intelligent lithology identification model of dynamic felling strategy weighted random forest algorithm(DFW-RF) is proposed. According to the feature selection results, gamma ray and 2 MHz phase resistivity are the logging while drilling(LWD) parameters that significantly influence lithology identification. The comprehensive performance of the DFW-RF lithology identification model has been verified in the application of 3 wells in different areas. By comparing the prediction results of five typical lithology identification algorithms, the DFW-RF model has a higher lithology identification accuracy rate and F1 score. This model improves the identification accuracy of thin-layer lithology and is effective and feasible in different geological environments. The DFW-RF model plays a truly efficient role in the realtime intelligent identification of lithologic information in closed-loop drilling and has greater applicability, which is worthy of being widely used in logging interpretation.

DSP Based Waveform Generator

The DSP Based Waveform Generator is used for CSR Control system to control special controlled objects, such as the pulsed power supply for magnets, RF system, injection and extraction synchronization, global CSR synchronization etc. This intelligent controller based on 4800 MIPS DSP and 256M SDRAM technology will supply highly stable and highly accurate reference waveform used by the power supply of magnets. The specifications are as follows:

Whale Optimization Algorithm-Based Deep Learning Model for Driver Identification in Intelligent Transport Systems

Driver identification in intelligent transport systems has immense demand,considering the safety and convenience of traveling in a vehicle.The rapid growth of driver assistance systems(DAS)and driver identification system propels the need for understanding the root causes of automobile accidents.Also,in the case of insurance,it is necessary to track the number of drivers who commonly drive a car in terms of insurance pricing.It is observed that drivers with frequent records of paying“fines”are compelled to pay higher insurance payments than drivers without any penalty records.Thus driver identification act as an important information source for the intelligent transport system.This study focuses on a similar objective to implement a machine learning-based approach for driver identification.Raw data is collected from in-vehicle sensors using the controller area network(CAN)and then converted to binary form using a one-hot encoding technique.Then,the transformed data is dimensionally reduced using the Principal Component Analysis(PCA)technique,and further optimal parameters from the dataset are selected using Whale Optimization Algorithm(WOA).The most relevant features are selected and then fed into a Convolutional Neural Network(CNN)model.The proposed model is evaluated against four different use cases of driver behavior.The results show that the best prediction accuracy is achieved in the case of drivers without glasses.The proposed model yielded optimal accuracy when evaluated against the K-Nearest Neighbors(KNN)and Support Vector Machines(SVM)models with and without using dimensionality reduction approaches.

Intelligent identification and real-time warning method of diverse complex events in horizontal well fracturing

The existing approaches for identifying events in horizontal well fracturing are difficult, time-consuming, inaccurate, and incapable of real-time warning. Through improvement of data analysis and deep learning algorithm, together with the analysis on data and information of horizontal well fracturing in shale gas reservoirs, this paper presents a method for intelligent identification and real-time warning of diverse complex events in horizontal well fracturing. An identification model for "point" events in fracturing is established based on the Att-BiLSTM neural network, along with the broad learning system (BLS) and the BP neural network, and it realizes the intelligent identification of the start/end of fracturing, formation breakdown, instantaneous shut-in, and other events, with an accuracy of over 97%. An identification model for "phase" events in fracturing is established based on enhanced Unet++ network, and it realizes the intelligent identification of pump ball, pre-acid treatment, temporary plugging fracturing, sand plugging, and other events, with an error of less than 0.002. Moreover, a real-time prediction model for fracturing pressure is built based on the Att-BiLSTM neural network, and it realizes the real-time warning of diverse events in fracturing. The proposed method can provide an intelligent, efficient and accurate identification of events in fracturing to support the decision-making.

Intelligent risk identification of gas drilling based on nonlinear classification network

During the transient process of gas drilling conditions,the monitoring data often has obvious nonlinear fluctuation features,which leads to large classification errors and time delays in the commonly used intelligent classification models.Combined with the structural features of data samples obtained from monitoring while drilling,this paper uses convolution algorithm to extract the correlation features of multiple monitoring while drilling parameters changing with time,and applies RBF network with nonlinear classification ability to classify the features.In the training process,the loss function component based on distance mean square error is used to effectively adjust the best clustering center in RBF.Many field applications show that,the recognition accuracy of the above nonlinear classification network model for gas production,water production and drill sticking is 97.32%,95.25%and 93.78%.Compared with the traditional convolutional neural network(CNN)model,the network structure not only improves the classification accuracy of conditions in the transition stage of conditions,but also greatly advances the time points of risk identification,especially for the three common risk identification points of gas production,water production and drill sticking,which are advanced by 56,16 and 8 s.It has won valuable time for the site to take correct risk disposal measures in time,and fully demonstrated the applicability of nonlinear classification neural network in oil and gas field exploration and development.

A road hypnosis identification method for drivers based on fusion of biological characteristics

Risky driving behaviors,such as driving fatigue and distraction have recently received more attention.There is also much research about driving styles,driving emotions,older drivers,drugged driving,DUI(driving under the influence),and DWI(driving while intoxicated).Road hypnosis is a special behavior significantly impacting traffic safety.However,there is little research on this phenomenon.Road hypnosis,as an unconscious state,is can frequently occur while driving,particularly in highly predictable,monotonous,and familiar environments.In this paper,vehicle and virtual driving experiments are designed to collect the biological characteristics including eye movement and bioelectric parameters.Typical scenes in tunnels and highways are used as experimental scenes.LSTM(Long Short-Term Memory)and KNN(K-Nearest Neighbor)are employed as the base learners,while SVM(Support Vector Machine)serves as the meta-learner.A road hypnosis identification model is proposed based on ensemble learning,which integrates bioelectric and eye movement characteristics.The proposed model has good identification performance,as seen from the experimental results.In this study,alternative methods and technical support are provided for real-time and accurate identification of road hypnosis.

Ground peak identification in dense shrub areas using large footprint waveform LiDAR and Landsat images

Large footprint waveform LiDAR data have been widely used to extract tree heights.These heights are typically estimated by subtracting the top height from the ground.Compared to the top height detection,the identification of the ground peak in awaveform is more challenging.This is particularly evident in ground detection in shrubareas,where the reflection of the shrub canopy may significantly overlap with theground reflection.To tackle this problem,a novel method based on Partial Curve-Fitting(PCF)of the shrub peak was developed to detect the ground peak.Resultsindicated that the PCF method improves ground identification by 32-42%,comparedto existing methods.To offer further improvement,a Multi-Algorithm IntegrationClassifier(MAIC)was built to fuse multiple ground peak algorithms and selectivelyapply the best method for each waveform plot.The PCF ground peak identificationmethod along with the MAIC-based fusion is expected to significantly improve grounddetection and shrub height estimation,thus assisting biodiversity,forest succession,and carbon sequestration studies,while offering an early example of future multiplealgorithm integration.

Intelligent optimization methods of phase-modulation waveform

With the continuous improvement of radar intelligence, it is difficult for traditional countermeasures to achieve ideal results. In order to deal with complex, changeable, and unknown threat signals in the complex electromagnetic environment, a waveform intelligent optimization model based on intelligent optimization algorithm is proposed. By virtue of the universality and fast running speed of the intelligent optimization algorithm, the model can optimize the parameters used to synthesize the countermeasure waveform according to different external signals, so as to improve the countermeasure performance.Genetic algorithm(GA) and particle swarm optimization(PSO)are used to simulate the intelligent optimization of interruptedsampling and phase-modulation repeater waveform. The experimental results under different radar signal conditions show that the scheme is feasible. The performance comparison between the algorithms and some problems in the experimental results also provide a certain reference for the follow-up work.

Rock thin-section analysis and identification based on artificial intelligent technique

Rock thin-section identification is an indispensable geological exploration tool for understanding and recognizing the composition of the earth.It is also an important evaluation method for oil and gas exploration and development.It can be used to identify the petrological characteristics of reservoirs,determine the type of diagenesis,and distinguish the characteristics of reservoir space and pore structure.It is necessary to understand the physical properties and sedimentary environment of the reservoir,obtain the relevant parameters of the reservoir,formulate the oil and gas development plan,and reserve calculation.The traditional thin-section identification method has a history of more than one hundred years,which mainly depends on the geological experts'visual observation with the optical microscope,and is bothered by the problems of strong subjectivity,high dependence on experience,heavy workload,long identification cycle,and incapability to achieve complete and accurate quantification.In this paper,the models of particle segmentation,mineralogy identification,and pore type intelligent identification are constructed by using deep learning,computer vision,and other technologies,and the intelligent thinsection identification is realized.This paper overcomes the problem of multi-target recognition in the image sequence,constructs a fine-grained classification network under the multi-mode and multi-light source,and proposes a modeling scheme of data annotation while building models,forming a scientific,quantitative and efficient slice identification method.The experimental results and practical application results show that the thin-section intelligent identification technology proposed in this paper does not only greatly improves the identification efficiency,but also realizes the intuitive,accurate and quantitative identification results,which is a subversive innovation and change to the traditional thin-section identification practice.

煤矸石井下原位智能分选充填技术研究进展

传统井下煤矸石需运输至地面处理,不仅占用地面国土空间、自燃或雨水淋滤造成大气与环境污染,而且长距离无效运输造成的能源消耗问题已成为制约煤矿低碳发展的关键瓶颈。为实现煤矿矸石不出井,从煤炭生产源头减少碳排放与单位产出能源资源消耗,实现煤炭的绿色低碳智能开采,回顾了煤矸石井下分选充填技术现状及智能化进展,并在此基础上展望煤矸石井下分选充填技术发展趋势,提出了煤矸石井下原位绿色智能分选、充填新方法,详细阐述了煤矸智能分选机及新型充填液压支架的结构及原理,以最大化缩短矸石无效运输距离。为处理采煤工作面矸石,预防煤岩动力灾害,提出了包含少矸化智能开采系统、原位智能分选系统、工作面矿压反演系统、精准科学充填系统四大子系统的煤矿井下采选充智能一体化系统,并探讨了各子系统间的新环逻辑关系,以形成采煤利于分选、分选利于充填、充填利于采煤的良性循环。为处理掘进工作面矸石,提出了包含智能快掘系统、智能分选系统、煤矸分运系统、智能充填系统四大子系统的煤矿井下掘选充智能一体化系统,并对各子系统所负责工作及智能化实现进行了阐述。所提出的新工艺有望实现煤矿矸石不出井,并为煤矸石原位智能分选充填方法及采选充一体化系统的研究提供新思路。

基于手机信令数据的城市居民动态OD矩阵提取方法

现有的城市居民出行调查周期较长,交通小区划分粒度粗糙,导致调查不能及时准确地获取居民出行信息。针对该问题,提出了一种基于手机信令数据的城市居民动态OD矩阵提取方法。首先,针对信令数据中的两种复杂噪声:乒乓切换和漂移数据,提出了基于窗口阈值的检测与等效位置替换方法,以及复杂漂移点的检测和标记处理方法;然后,提出一种改进的ST-DBSCAN聚类方法,引入一种等时化方法将时间信息与空间信息相结合,识别出行过程中的驻留点;最后,基于地理信息系统构建含有道路关键节点的路网,将居民出行OD与路网节点相匹配,有效推导出城市居民动态OD矩阵。实验结果表明:与ST-DBSCAN算法相比,所提改进的ST-DBSCAN算法在聚类效果和识别速度上分别提升了6.10%和5.26%;与统计方法和二阶统计量方法相比,基于改进的ST-DBSCAN算法的动态OD矩阵提取方法在均方误差(MSE)上分别降低了16.98%和21.55%。以北京市为例,运用提出的动态OD矩阵提取方法,能够及时有效地分析城市居民日常与高峰时段的出行特征。

多源车载数据驱动的地铁轨道不平顺智能识别方法

针对轨道不平顺检测成本高与时效性低等不足,从车辆动态响应与轨道不平顺之间的相关性为切入点,提出一种多源车载数据驱动的轨道不平顺智能识别方法。首先,建立地铁车辆系统动力学模型,获取车辆振动与运动姿态响应数据;其次,通过相关性分析算法,选取强相关性数据,制作网络模型数据集;最后,建立卷积神经网络-长短期记忆网络(CNN-LSTM),通过粒子群算法优化(PSO)神经网络模型参数,建立PSO-CNN-LSTM模型,实现对轨道不平顺的识别拟合。研究结果表明:在车辆动态响应信号中,与横向信号与轨道不平顺之间的相关性相比,垂向信号的更强,同时,车体的运动姿态如车体点头角速度与不平顺有明显的相关性。所提出的PSO-CNN-LSTM模型轨道垂向与横向不平顺识别拟合度分别达0.92和0.76。与经典的全连接神经网络FCNN和支持向量机SVR相比,PSO-CNN-LSTM有更好的识别效果与时效性。

矿物组分识别与智能解释在不同岩性之间的信息共享与迁移学习

在地球科学领域,岩石微观观测数据的采集过程繁琐且效率低下,这不仅增加了研究成本,降低了可靠性,同时也限制了数据的开源共享。此外,由于岩性的多样性和观测手段的差异,单一数据集的规模通常较小,这对于依赖大规模数据集的深度学习框架而言是一大挑战。为此,本研究探索迁移学习如何促进不同岩性间的信息共享,并通过此机制提高矿物组分识别与智能解释任务的模型性能。通过采集不同区域、岩性、矿物组分和偏光模式下的铸体薄片样本,本文深入研究了深度学习模型在不同观测对象和手段下的迁移学习机制,并聚焦于探索地质信息的深层表征。研究成果不但揭示了迁移学习在促进地质学领域信息共享与模型性能提升中的关键作用,还为自动化和智能化地质认识融合奠定了基础。实验结果显示,通过迁移学习,本文模型在智能解释任务中的准确率显著提高,从53.3%提高至98.73%,而在矿物组分识别任务中,准确率也实现了近10%的提升。这些成果证明了迁移学习在地质学领域内解决实际问题和提高模型泛化能力、性能和稳定性方面的巨大潜力。

基于红外光谱技术智能识别润滑油的研究进展

机器学习作为人工智能发展的核心,在各行业得到快速发展,近年来也成为润滑油领域研究的热点之一,标志着润滑油的研究不再局限于大规模的试验研究,高通量数据、机器学习、优化算法开始应用于润滑油的研究。文章介绍了基于红外光谱技术在润滑油种类鉴别、润滑剂筛选、润滑性能评估和润滑监测等方面的研究进展,并对未来基于红外光谱技术应用于智能识别润滑油的研究进行了展望。

千万吨级综放工作面智能化放煤理论及关键技术

针对综采放顶煤工作面在智能放煤理论、智能感知与识别关键技术、智能放煤综合决策技术和远程放顶煤智能控制技术方面存在的问题,依托“十三五”国家重点研发计划——千万吨级特厚煤层智能化综放开采关键技术及示范,开展了千万吨级综放工作面智能化放煤理论及关键技术研究,取得如下成果:(1)开展了顶板顶煤组合体压缩试验、煤岩组合体破碎块度分布试验和不同顶板-顶煤条件下综放开采放煤相似模拟试验,阐明了顶板顶煤的破碎-运移的相互作用过程;开发了采空区三维激光空间探测技术,证明了群组放煤过程中顶煤“面接触块体成拱现象;开展以采放协调、高回收率、低含矸率为约束条件的特厚煤层多口群组智能放煤数值模拟,确定了群组放煤口数量;综上,为智能放煤工艺优化提供了可靠理论基础。(2)对工作面煤矸地质信息及物性特征、顶煤放落过程全周期感知要素进行探索研究,形成了包含“顶煤厚度在线探测-煤矸精准识别-煤流动态测量”的综合感知技术体系,为智能化放煤决策技术提供了充分的数据信息支撑。(3)建立了综放面“人-机-环”多源信息数据库,构建了基于采放时间协调、采放空间协调、采放运能协调的特厚煤层综放面采放协调决策模型,开发了基于Q-learning算法的智能放煤决策软件,形成了基于煤矸识别、顶煤厚度探测和过煤量监测感知,结合煤矸运移时序特征和经验数据的人工智能决策技术。(4)研发了智能综放面三机位姿高精度惯导监测与控制技术,实现了采煤机、液压支架、刮板输送机的实时定位、姿态监测及动作控制;开发了智能化矿山融合通信调度系统,构建了智能综放远程综合控制平台,成功实现了“远程一键启动”模式的智能放煤。(5)在塔山矿8222工作面,开展了基于探地雷达的顶煤厚度在线探测技术、融合振动-音频-高光谱的煤矸精准识别技术、基于激光三维扫描的放煤量实时监测技术和综放工作面智能化放煤决策软件的应用,现场顶煤厚度探测、混矸率和放煤量的误差分别控制在10.71%、9.32%和7.8%以内,平均每个放煤循环节省时间约30 min,实现了年产1500万t的综放工作面智能高效放煤。

植物病害智能识别APP开发及实践教学应用

该文分析了传统植物病理实践教学存在的一些问题和弊端,利用卷积神经网络技术开发了植物病害症状和病原菌智能识别APP,并将之应用于实践教学,促进了学科交叉融合,丰富了教学内容,完善了实践教学评价体系。实践表明,植物病害智能识别APP在实践教学中效果显著,提高了学生的学习兴趣,便于学生个性化、精准化、自由化学习,有利于培养学生跨学科的创新思维和应用能力,为新农科复合型人才的培养提供了一种便利工具。