摘要
能效测评可为节能减排、提高能源使用率提供有力的数据支撑和解决方案。针对传统电力用户所采用能效指标单一且评估过程易受人为因素影响等问题,提出了一种高效的能效评级量化方法。首先选择科学合理的能效指标并确定分级标准,之后通过交叉检验（cross-validation）选择适合的近邻参数K,建立基于K最近邻（KNN）算法的能效评级模型,最后根据投票策略评估电力用户的能效等级。研究及仿真结果表明,所提出的KNN能效评级方法具有操作简捷、符合内在机理、分级准确等优越性,在该领域具有较强的实用价值和推广前景。
Energy efficiency assessment can provide strong data support and solutions for energy saving and emission reduction and energy efficiency.Aiming at the problem that the energy efficiency index of traditional electric power users is single and the evaluation process is susceptible to human factors,presents a efficient energy efficiency rating method.First,select the scientific and rational energy efficiency indicators and determine the classification criteria,Then through the Cross-Validation to select the appropriate neighbor parameters K,an energy efficiency rating model based on KNN algorithm was established,Finally,according to the voting strategy to assess the energy efficiency of power users.The research and simulation results show that the proposed method of KNN energy efficiency rating has the advantages of simple operation,in line with the internal mechanism,accurate classification and other advantages,and has strong practical value and promotion prospects in this field.
作者
马立新
徐曼如
吕梦圆
Ma Lixin;Xu Manru;Lv Mengyuan(School of Optieal-Electrical and Computer Engineering,University of Shanghai for Science and Teehnology,Shanghai 200093,Chin)
出处
《电子测量技术》
2018年第13期10-13,共4页
Electronic Measurement Technology
基金
国家自然科学基金(61205076)项目资助
关键词
电能质量
能效评级
KNN算法
power quality
energy efficiency rating
KNN algorithm
