摘要
当前,电力设备选型主要依靠一些有经验的工程师根据其经验和国家标准来确定,然而在经验和标准都难以依靠的时候,可能会直接选择最昂贵或配置最顶级的设备,因此导致某种程度的资源浪费。文章基于大数据技术,运用随机森林算法对历史数据进行多元分类得到推荐模型,进而对电力设备选型给出智能推荐。基于本文形成的推荐模型准确率可达到75%以上,没有经验的相关人员也可以借助此模型进行设备选型,能够减少人力投入和对经验的过度依赖,且可以得到比经验选型更准确的结果,最终达到电力设备精细化管理的目标。
At present, the selection of power equipment mainly depends on some experienced engineers to determine the final selection according to their experience and national standards. However, when both experience and standards are difficult to rely on, they may directly choose the most expensive or top-level equipment, while some equipment may not need such a high configuration, which leads to a certain degree of waste of resources. In this paper, based on big data technology, the random forest algorithm is used to classify the historical data to obtain the recommendation model, and then the intelligent recommendation for the selection of power equipment is given. The accuracy of the recommended model based on this paper can reach more than 75%, and the inexperienced personnel can also use this model to select equipment, which can reduce the human investment and excessive dependence on experience, and may get more accurate results than the experience selection, and ultimately achieve the goal of fine management of power equipment.
作者
刘伟
徐文峰
LIU Wei;XU Wenfeng(Hubei Huazhong Electric Power Technology Development Co.,Ltd.,Wuhan 430000,China)
出处
《电力信息与通信技术》
2019年第5期19-24,共6页
Electric Power Information and Communication Technology
关键词
大数据
机器学习
电力设备选型
随机森林
智能推荐
big data
machine learning
power equipment selection
random forest
intelligent recommendation
