The application of pre-training language model in the field of sentiment analysis is analyzed,the traditional pre-training language model according to the actual needs of sentiment classification of public opinion information in the college network environment is integrated,to further improve the accuracy of model classification and prediction. A system design scheme for emotional analysis of the college network public opinion is put forward,which forms a complete closed-loop from data collection,data processing to data analysis and can effectively assist in public opinion supervision on the college network environment in ideological and political education.

The prevailing intelligent recommendation algorithms are introduced,the current situation of courses on an online learning platform is taken into consideration,and two recommendation algorithms based on course label association rules and course similarity collaborative filtering are proposed.Based on the advantages of the two algorithms,a course recommendation system with multi-channel alternative set is constructed,the weights are assigned to each type of recommendation mode and the weight parameters can be adjusted according to the actual needs to flexibly recall recommendation results.

受新冠肺炎疫情的影响，企业线下培训面临极大的挑战，培训方式已从线下转为线上线下相融合，从广播式到互动式转变。企业直播从时间和远距离培养进化到全场景、多维度的培训模式。本文从企业在线直播培训入手，探讨企业直播培训特点，分析企业直播培训模式与流程，形成直播培训数据运营评估体系，促进企业线上知识沉淀和实现组织人才培养目标。

针对目前国网员工培训的需求,此文对泛在学习理念下国网网络培训系统进行探析,面向国网员工的网络培训系统从学习目的,学习任务,网络平台,学习资源,教师资源的配置5个方面进行设计.该模式实现了以学习者为中心持续、交互、主动的学习形式,使泛在学习、网络培训、自主学习、终身学习成为一个系统的工程.

企业网络大学是能够提升企业竞争优势的知识服务机构.知识经济时代企业大学面临的重要课题是如何实现知识的传递、共享、进化和增值,从而更好地整合和利用企业的各类资源.此文基于网络学习生态系统理论,分析了网络学习生态系统的构成要素,研究并构建了企业网络学习生态系统.

国家电网公司以建设运营电网为核心业务,是关系国民经济命脉和国家能源安全的特大型国有重点骨干企业,始终致力于实现"建设世界一流电网,建设国际一流企业"的宏大愿景.随着人力资源和知识资产越来越成为提升企业战略竞争力的核心源泉,自2013年起,国家电网公司开展了网络大学建设工作,快速建成了基于内网和ERP系统的国内规模最大、范围最广、资源丰富、技术先进的企业网络学习与考试平台.同时,借助"互联网+"的最新理念,全面构建基于知识资产管理的电力企业网络学习生态圈,以提升全员网络培训服务能力为目标,深化网络大学应用,提升平台支撑能力,打造移动互联网学习模式,构建学习运营体系,加速知识资产在企业内部的有效流转,打造覆盖公司全部单位、全部岗位、全部人员,公平开放、共生共享的学习环境.

Enterprise network university is a knowledge service institution that can enhance the competitive advantage of enterprises. In the era of knowledge economy, the important task that enterprise universities face is how to realize the transfer, sharing, evolution and proliferation of knowledge, so as to better integrate the various resources of the enterprise. Based on the theory of network learning ecosystem, this paper analyzes the components of the network learning ecosystem, studies and constructs the enterprise network learning ecosystem.

近年来,世界政治和经济环境都发生了前所未有的变化,经济全球化的趋势加大,企业不再是单一的存在,而是向国际化、多元化、联盟化的企业集团形式发展,本文从省公司子改分后经营管理可能遇到的困境出发,引入阿米巴管理哲学,提出了解决问题的思路和目标.

为克服传统距离保护中存在的超越或拒动问题,利用六序分量法,分析同杆双回线中反序负序分量特征;通过对分量的图示分析,提出了一种基于反序负序分量的距离保护判据,进而结合纵联电气量和纵联逻辑信号两种保护原理,得出同杆双回线纵联距离保护新方法.该纵联保护同时适用于同杆线路背端电源弱馈的特殊运行方式.通过电磁暂态程序EMTP(electro-magnetic transient program)仿真表明,此保护综合性能良好,具有明显的方向性.

Based on image processing of SF6 gas leakage on-line pattern recognition method, this paper achieves gas leakage feature extracting, on-line identification of gas leakage and leakage points, SF6 gas leakage can be on-line automatic identification. The simulation results show the feasibility of the algorithm. Compared with the traditional method, paper provides a more intuitive discrimination basis for field staff , as well as for the depth of the late testing data mining provides a research way of thinking

Application of big data techniques in power system will contribute to the sustainable development of power industry companies and the establishment of strong smart grid. This article introduces a universal framework of electric power big data platform, based on the analysis of the relationships among the big data, cloud computing and smart grid. Then key techniques of electric power big data is discussed in four aspects, including big data management techniques, big data analysing techniques, big data processing techniques and big data visualization techniques. Finally, the article presents three typical application examples of electric power big data techniques which are new and renewable energy integration, wind turbine condition monitoring and assessment and data base integrative backup for electric power enterprises.

Based on multi-source substation equipment inspection data, this article achieve fault diagnosis in the case of incomplete understanding of the mechanism of substation equipment, by establishing fault diagnosis model.This article selected a substation equipment failures and operational information in different conditions as simulation study. The simulation results show the feasibility of the algorithm. Compared with the traditional fault diagnosis model, this method is more flexible , have a stronger ability to handle noisy data and good prospects.

Based on image processing of SF 6 gas leakage on-line pattern recognition method, this paper achieves gas leakage feature extracting, on-line identification of gas leakage and leakage points, SF 6 gas leakage can be on-line automatic identification. The simulation results show the feasibility of the algorithm. Compared with the traditional method, paper provides a more intuitive discrimination basis for field staff , as well as for the depth of the late testing data mining provides a research way of thinking

本文提出了一种基于RFID无线射频技术、无线局域网技术、物联网技术以及支持移动终端的配网智能巡检培训系统,创建了电力培训的一种全新的工作模式,可以有效提高配网巡检工作效率,规范配网的巡检作业,提高配网巡检的培训效果和配网管理水平,从而满足配网精益化管理的需求.

This paper deeply researches on successive action zone and dead zone of transverse differential protection in parallel lines, and improves the existing analysis software for relay protection action. First determine which zone the fault point belongs to, and then analyze the protection action according to the different zone concretely. This improvement will greatly increase the accuracy to analyze the relay protection action, and will be significant to the safe and stable operation of regional networks.

This paper presents the key technologies of big data based condition monitoring of power apparatuses. Firstly, the characteristics of big data and big data of power system are discussed and the application prospects of big data based condition monitoring of power apparatuses is presented and the key technologies of the system is discussed, in terms of big data analyzing technologies, big data management technologies, big data processing technologies and big data visualization technologies. Thirdly, big data based condition assessment techniques of power apparatuses are discussed, including data fusion of signal collected from different sensors, historical trending analysis and association analysis of combined equipments. Finally, to further introduce big data techniques, an integrated condition monitoring system of transformer, GIS and power cable is presented, including system hardware structures and big data based condition assessments.

A micro grid simulation model is built according to the distributed generation characteristic. Based on the simulation model, the impact from generation characteristic to the power distribution network relay protection after micro grid connection is analyzed. The properties of the failure current from distributed generation of micro grid are discussed. Based on the characteristic of distributed generation failure current, the effect on the distribution network relay protection can be got.

In this paper, the transient characteristics of the permanent magnet synchronous motor based on the space vector pulse width modulation technology are researched in detail. Firstly, the mathematical model of the control system is established. Secondly, the simulation model based on MATLAB/Simulink is built. Thirdly, the transient simulation is implemented by the constructed model. It is proved that the mathematical model is valid and effective, which provides a theoretic base for the design of an actual permanent magnet synchronous motor control system.

By analyzing the two lightning accidents happening to a 500kV substation situated in the lightning-prone northeastern Guangdong Province, I conclude as follows, there should be some problem in the equipotential earthing of the secondary system of this substation. Specifically, the existing potential difference in the secondary equipment gave rise to the meltdown of it which finally caused the accident. In the context,I build the one-point earthing parallel system simulation model according to ATP graphic pretreatment program, then by using slope-ramp simulation I study the lightning current which leaked into the grounding grids of the substation.In conclusion, the application of high frequency signal mixed low frequency signal earth system in the secondary system and the equipotential bonding is key to preventing accidents.

This paper applies the theory of expert system to analysis for relay protection action based on PAS. And it also summarizes various rules according to the experience of operating personals and relay protection personals for several years. All these rules are suitable for 110kV and 66kV regional networks. This paper not only analyzes three-phase current protection, zero-sequence current protection, distance protection etc, but also discusses some special situations such as under voltage lockout over-current protection for transformer, remote backup protection etc.