基于协同过滤和雷达图法的信息运维技能培训平台研究

发布时间：2018-08-19 14:11

【摘要】：随着电网企业“三集五大”体系建设的全面推进,许多应用和管理系统不断上线。信息通信系统作为电力系统的重要组成部分,其安全可靠运行对维持电力系统的稳定性起着非常重要的作用。拥有一批适应新形势要求、技能过硬的高素质信息运维人员是保证信息通信系统安全可靠运行的基础,信息运维技能培训是提高信息运维人员专业技能水平的有效途径。针对电网企业信息运维人员专业技能培训中存在的若干问题,本文从三个方面对基于协同过滤和雷达图法的信息运维技能培训平台进行了研究。针对信息运维技能学习时不能满足信息运维人员对信息运维技能知识的个性化需求的问题,本课题提出了基于协同过滤的信息运维技能知识个性化推荐模型。该模型的基本思想是:首先设计一种隐式评分机制,将信息运维人员的在线学习行为转换为其对信息运维技能知识的隐式评分;然后提出一种基于项目特征和用户兴趣的协同过滤算法,即在用户相似性计算时引入项目相似性,并在未评分项目预测时引入时间函数对传统协同过滤算法进行改进;最后将改进的协同过滤算法应用于信息运维技能知识个性化推荐。通过实验验证了个性化推荐模型的有效性。针对信息运维技能评价时不能形象直观地反映信息运维人员的信息运维技能综合水平的问题,本课题提出了基于雷达图法的信息运维技能综合评价模型。该模型的基本思想是:首先根据调研情况和指标体系的构建原则建立信息运维技能评价指标体系;然后提出一种基于组合赋权的雷达图综合评价方法,即运用组合赋权法计算评价指标的权重,将权重转换为指标在雷达图中对应扇形区域的圆心角,并以扇形区域的角平分线作为指标轴对传统雷达图法进行改进;最后将改进的雷达图法应用于信息运维技能综合评价。通过实验验证了综合评价模型的可靠性。结合以上理论研究,本课题开发了一套电网企业信息运维人员专业技能培训平台,实现信息运维技能学习、训练、考试、评价一体化管理。该平台采用了“学、练、考、评”的培训模式,为信息运维人员提供一个集信息运维技能学习、训练、考试、评价等为一体的专业技能培训环境,提高了信息运维人员的专业技能水平,从而保证电网企业信息通信系统的安全可靠运行。
[Abstract]:With the overall development of the "three sets and five" system in power grid enterprises, many application and management systems have been put on line. As an important part of power system, information communication system plays an important role in maintaining the stability of power system. It is the foundation of ensuring the safe and reliable operation of information communication system to have a group of high quality information operators who can meet the requirements of the new situation. The training of information operation and maintenance skills is an effective way to improve the level of professional skills of information operators and maintainers. Aiming at some problems existing in the training of information operation and maintenance personnel in power grid enterprises, this paper studies the training platform of information operation and maintenance skills based on collaborative filtering and radar map method from three aspects. In order to solve the problem that information operation and maintenance skills can not meet the individual needs of information operations and maintenance personnel, a personalized recommendation model of information operational skills knowledge based on collaborative filtering is proposed in this paper. The basic idea of the model is as follows: firstly, an implicit scoring mechanism is designed to convert the online learning behavior of information operators into their implicit scoring of information operational skills; Then, a collaborative filtering algorithm based on item feature and user interest is proposed, that is, the item similarity is introduced when the user similarity is calculated, and the time function is introduced to improve the traditional collaborative filtering algorithm. Finally, the improved collaborative filtering algorithm is applied to the personalized recommendation of information operation and maintenance skills. The effectiveness of the personalized recommendation model is verified by experiments. Aiming at the problem that the comprehensive level of information operation and maintenance skills can not be visually reflected in the evaluation of information operation and maintenance skills, a comprehensive evaluation model of information operation and maintenance skills based on radar map method is proposed in this paper. The basic ideas of the model are as follows: firstly, according to the investigation situation and the principle of constructing index system, the evaluation index system of information operational skills is established, and then a comprehensive evaluation method of radar map based on combinatorial weight is proposed. That is, the weight of the evaluation index is calculated by using the combined weighting method, and the weight is converted into the center angle of the sector region corresponding to the index in the radar map, and the traditional radar map method is improved by using the angular bisection line of the sector region as the index axis. Finally, the improved radar map method is applied to the comprehensive evaluation of information operation and maintenance skills. The reliability of the comprehensive evaluation model is verified by experiments. Based on the above theoretical research, this paper develops a set of training platform for information operation and maintenance personnel in power grid enterprises, which realizes the integrated management of information operation and maintenance skills learning, training, examination and evaluation. The platform adopts the training mode of "learning, training, testing and evaluation", which provides a training environment for information operations and maintainers to learn, train, test, evaluate and so on. The level of professional skills of information operators is improved to ensure the safe and reliable operation of information communication system in power grid enterprises.
【学位授予单位】：东北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.52

【参考文献】