The copper mineralized bodies and orebodies of the Pulang copper deposit in Yunnan Province are mainly distributed in the Pulang complex porphyry body and were formed through complex multi-stage development. This study aims to detail the geophysical response and fractures of copper reservoirs and provide detailed orebody characteristics, fractures, and horizon burial depth to be referenced in the exploration and exploitation of the Pulang copper deposit. First, the borehole-log data in the Pulang copper deposit were sampled for comprehensive evaluation. Then, in combination with the drilling reports and data on partial core samples, this study analyzed the log response characteristics and fractures and identified the lithology of the Pulang copper deposit using mathematical statistics, three-dimensional cross plots, convolutional neural networks（CNNs）, and fracture parameter calculation. The log response characteristics of the three major strata of quartz monzonite porphyries, quartz diorite porphyrites, and hornstones in the study area are as follows. The hornstone strata have relatively high resistivity, followed by the quartz diorite porphyrite strata and the quartz monzonite porphyry strata in sequence. The resistivity decreases significantly at the intervals with fractures occurring or at the relatively fractured intervals. The quartz monzonite porphyry strata have a relatively high charge rate（polarization rate） of up to about 10%. The hornstone strata have relatively high radioactive intensity than the quartz diorite porphyrite strata and the quartz monzonite porphyry strata. CNNs were used to identify and analyze the lithology of the three major types of strata based on log data, with an accuracy rate of 97.94%. Finally, this study identified fractures in these strata using dual laterolog data. The resistivity significantly decreases at intervals with fractures occurring and differs greatly between deep and shallow lateral resistivity. The quartz monzonite porphyry strata with a high copper grade have relatively low resistivity and relatively well-developed high-angle fractures. The results of this study are of significance for the identification of ore body characteristics and the exploitation of ore bodies in the Pulang copper deposit.

电力文本语料稀缺造成了训练数据的不平衡,使得主流的双向长短期记忆网络和条件随机场方法表现较差.为此引入一个分类模型,将语句分为强弱两类,分别训练优化网络模型.通过在公开语料数据上进行实验,证明了在对电力语料数据进行命名实体识别时,该方法比传统的聚类方法和原双向长短期记忆模型分别高12％和4％.

科学准确地开展矿区岩体质量评价研究,对于矿区的可崩性评价、地下空间工程结构稳定性分析具有重要意义.以普朗铜矿区为研究区,通过收集普朗铜矿勘探区的地质资料、钻孔信息和岩心信息,建立了普朗铜矿区地质模型,并采用钻孔超声波成像和光学成像测井测试方法,结合岩心实物资料和室内岩石试验所获取的岩体质量评价指标参数,基于提出的模糊岩石工程系统(RES)理论-可拓学理论评价模型对矿区单个钻孔及区域岩体质量进行了评价与预测研究.结果表明:普朗铜矿区17个钻孔中15个钻孔整体岩体质量为Ⅱ级,仅2个钻孔的岩体质量分别为Ⅱ～Ⅲ级和Ⅲ级;矿区3种岩性岩体质量均为Ⅱ级,其中石英闪长玢岩和石英二长斑岩岩体质量良好(Ⅰ、Ⅱ级)段岩体占比更高,角岩岩体质量劣质(Ⅲ、Ⅳ、Ⅴ级)段岩体占比更高;工作区整体岩体质量为Ⅱ级;与RES-云模型和简单关联函数-可拓学模型相比,模糊RES-可拓学理论评价模型对某工程岩体质量的评价结果与实际工程岩体质量等级能够较好吻合,进一步验证了本文所提出的模糊RES-可拓学理论评价模型在岩体质量评价中具有较强的适用性.

在当前我国"双碳"政策下,矿井余热回收再利用是矿山向绿色低碳转型途径之一.基于此背景,本文针对某高寒矿山通风现状,设计了通风余热回收方案,用以解决寒冷季节矿山井口防冻问题,通过采集系统六个月的运行数据,对余热回收效果进行分析.结果表明,余热回收各个机组总体的运行状态比较平稳,余热回收效果显著,相比于电加热方案,极大的减少了能源的消耗,降低了运行成本,存在巨大的推广潜力.

为支撑电力物联网数据共享,发挥电网数据价值,文章针对当前电力物联网数据平台中存在的技术组件多样、应用难度大、检索数据困难、数据应用门槛高和数据模型管控机制不完善等问题,优化电力物联网数据平台整体框架,提出基于孤立森林的量测类实时数据质量异常检测及改进算法,通过抢修故障研判和停电故障研判2个场景仿真验证数据质量检测算法的可行性,仿真试验结果表明该模型能够有效提升数据质量检测准确率.

由于普朗铜矿近年来井下多次发生不同规模的泥石流,严重影响到矿山井下的安全生产.因此,泥石流是普朗铜矿采用自然崩落法开采的一个较大安全隐患.通过对塌陷坑的实地观察及相关的资料收集,分析泥石流的成因及造浆运动过程的变化,并总结相应规律,同时结合塌陷坑的现状、地理位置、气候环境、地质特性及生产实际等客观条件,采取相应的防范治理措施,对地表塌陷坑及周围区域进行安全治理,避免井下泥石流的再次发生.

针对主数据的科学识别,通过设计识别工作流程,对识别评判标准的量化分析,解决以往业务人员对主数据定义理解的不一致而造成的主数据识别分歧与风险.通过对某大型电网企业主数据应用现状进行调研,结合企业实际业务条件,设计符合企业运行的主数据识别工作流程,分清责任实体,做到数据谁生产、谁识别、谁维护.再根据业界流行的分析方法,设计出符合需求的主数据量化识别模式,准确科学地识别企业内部的主数据,从而达到解决业务分歧与风险的效果.

Due to the high computational cost, the application of deep neural networks (DNNs) to the real-time tasks has been limited. A possible solution is to compress the size of the model so that the demand for computation resources can be decreased. A popular method is called knowledge distillation (KD). The basic philosophy behind KD is to transfer the information extracted from the larger teacher network to the smaller student network. The general knowledge transfer strategy is to match the one-to-one logit or intermedia layers between the teacher and student networks, correspondingly. This objective may neglect the informative relationship information between different samples. In this paper, we borrow the idea of metric learning to transfer the contrastive relationship information learned from teacher network to the student. Specifically, we use the well-known Triplet loss to regularize the training of the student network. By the modified negative selection strategy, our Contrastive Knowledge Distillation (CKD) method can efficiently improve the performance of the student network compared with the traditional KD methods. Empirical experiments on KD benchmarks and real-world datasets also demonstrate the superiority of CKD.

针对当前电网行业产生的海量数据,提出采用规范化元数据管理等方式来实现对电力行业数据的统一存储与管理方案.首先通过数据预处理,将不同格式的电力数据统一转换为XML格式数据,然后采用中间件技术实现对XML数据抽取与访问;其次,针对大规模数据存储问题,提出基于哈希分桶算法对数据进行存储,以提高数据存储的效率;再次采用MapRe-duce框架对数据进行查询;通过对电力行业的部分数据进行查询试验,结果表明在查询时间方面,具有优势.

目的 探讨2,4-二硝基苯肼(2,4-dinitrophenylhydrazine,DNPH)柱前衍生高效液相色谱法测定水中微量乙醛的方法.方法 将样品在60 ℃水浴中用DNPH衍生化20 min,由正己烷萃取,以乙腈∶水(52∶48,V∶V)为流动相,于365 nm波长下进行测定.结果 乙醛在0.015～1.00 mg/L浓度范围内线性关系良好,相关系数r为0.9999,加标回收率为80.2％～99.8％,相对标准偏差为4.6％～7.4％,方法最低检出浓度为0.003 mg/L.结论 该方法样品前处理简单快捷、检出限低、灵敏度高,能够满足生活饮用水卫生规范的要求,并且能够同时测定大批量样品,为解决国家标准中检测方法最低检出限与其乙醛的限值相矛盾的问题提供参考.

In this study,we intended to develop a method for the determination of microcystin in lake water by high performance liquid chromatographic(HPLC). The trichloroacetic acid solution with 0.5mL/L was applied as mobile phase. Microcystin was separated by Agilent Eclipas XDB C18 column and determined at 238nm wavelength by VWD. The detection limit of the method reached 0.4μg/L. There was a good linearity within the range of 0.4μg/mL to 2μg/mL with the correlation coefficient of 0.998. The recoveries were between 84.2% and 92.0% at the spiking level of 0.4μg/mL to 1.8μg/mL,and the relative standard deviation ranged from 0.2% to 3.2%. The method built in this study is sensitive,accurate and easy to operate.

A method for the determination of active pentanol and furfural in wine samples by chromatography was studied. The ZB-WAX column and hydrogen flame ionization detector was selected. The results showed that the recovery of active pentanol ranged between 97.9%-109.4% and the recovery of furfural ranged between 2.1%-4.9%. The detection limits were 5mg/L. The relative standard deviations of active pentanol ranged from 2.6% to 3.2% and the relative standard deviations of furfural ranged from 4.1% to 4.9%. The method is simple,quick,accurate and safe and it can be applied for the determination of active pentanol and furfural in wine samples.

目的 建立食品中6种工业染料的QuEChER前处理高效液相色谱-串联质谱同时测定方法.方法 均匀样品用水分散溶解后,加乙腈提取,经改进的QuEChERS方法净化,0.22 μm微孔滤膜过滤,用0.1％甲酸-5mmol/L乙酸铵溶液和乙腈为流动相梯度洗脱,C18柱分离,分别在电喷雾正、负离子模式下以多反应监测方式检测,外标法定量.结果 结果表明,6种工业染料在0.5～40 μg/L浓度范围内与其呈良好的线性关系(r＞0.999),检出限为1.0～2.0 μg/kg,在添加水平为1.0、5.0、20μg/kg时,平均回收率为76.1％～114.8％,相对标准偏差(RSD)为2.0％～9.7％(n=6).结论 该方法分析速度快,灵敏可靠,重复性好,适用于调味品、豆制品、卤肉制品、红糖等食品中6种工业染料的测定.

Objective To establish a mathematical model of uncertainty evaluation of sulfapyridine, sulfathiazole and sulfamethoxazole in milk by high performance liquid chromatography-tandem mass spectrometry method. Methods To establish a mathematical model for determination of sulfapyridine, sulfathiazole and sulfamethoxazole uncertainty in milk. The sources of uncertainty of the whole determination procedure were analyzed and the combined uncertainty was finally obtained. Results The expanded uncertainty of 5.09 μg/kg (k=2) with 0.62 μg/kg of the content of sulfapyridin was derived. The expanded uncertainty of 8.62 μg/kg (k=2) with 0.71 μg/kg of the content of sulfathiazole was derived. The expanded uncertainty of 10.15 μg/kg (k=2) with 0.81 μg/kg of the content of sulfamethoxazole was derived. The major sources of uncertainty of measurement include the repeatability of detection and the curve fitting. Conclusion The major uncertainty sources were standard curve fitting;measurement repeatability and sample pretreatment.

A new method for the sensitive determination of Aflatoxins B1 in tea by high performance liquid chromatography with photoelectric reactor and fluorescence detector was established.A solution of V(acetonitrile)∶ V(H2O)=86∶ 14 was used to extract Aflatoxins B1 from tea.The extracted solution was then purified by a multifunctional and immuneaffinity column,respectively.The peak area and the concentration of Aflatoxins B1 showed a good linear relationship within the range from 0.591 μg/L to 5.91 μg/L with a linear correlation coefficient(r) of 0.9994.The recoveries at the concentrations studied [low level(7.090 μg/kg),high level(14.180 μg/kg)] were between 85.6% and 98.9% with a relative standard deviations ranging from 1.7% to 1.8%.The limit of detection(LOD) is 0.1 μg/kg(S/N=3).The limit of quantitation(LOQ) is(0.591 μg/kg).The new method was used to analyze eight tea samples collected from the local markets and negative results were obtained.The meth od is suitable for detection of Aflatoxins B1 in tea with high selectivity and sensitivity.