智能投影寻踪方法在能效分级测评中的应用  

Application of Intelligent Projection Pursuit Method in Classification of Energy Efficiency Evaluation

作　　者：马立新[1] 原晓琴 项庆[1] 

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093

出　　处：《电力系统及其自动化学报》2016年第10期8-11,共4页Proceedings of the CSU-EPSA

摘　　要：为了深入研究电力能效测评系统采集海量数据并精细量化节能方案,根据电能质量国际标准,引入能效等级概念从多维度评测电力能效状态。首先建立能效的投影寻踪等级评价模型,采用人工鱼群算法寻求最佳投影方向,并将该模型应用于电力能效等级的评价。研究表明,基于人工鱼群算法的投影寻踪等级评价模型对用电单位进行能效分析及评级,能精细量化反映用电单位的能效状态。评测过程无需人为确定权重,避免了传统评价方法因主观原因造成的误差。实例计算表明,评价结果具有较高的准确性和可行性。测评方法简捷高效,为能效分级与测评提供了新算法。To deeply research the electric power energy efficiency evaluation system which collects a vast amount of data ,and quantify the energy-saving scheme in detail ,the concept of energy efficiency grade is introduced to multi-dimensionally evaluate the power state of energy efficiency according to the national standard of power quality. A projection pursuit grade evaluation model of energy efficiency is established first ,then artificial fish school algorithm （AFSA）is used to seek the optimal projection direction. This model is applied to the evaluation of electric power energy efficiency grade. Results show that energy efficiency analysis and classification of power units using projection pursuit grade evaluation model based on AFSA reflects the state of energy efficiency of power units quantitatively. The evaluation is processed without artificial weights to avoid the traditional evaluation method of error due to subjective reasons. Example calculation shows that the evaluation result has high accuracy and feasibility. The evaluation method is simple and efficient ,and provides a new algorithm for the classification and evaluation of energy efficiency.

关 键 词：用电单位 能效等级 人工鱼群算法 投影寻踪 

分 类 号：TM73[电气工程—电力系统及自动化]