基于多目标投影寻踪模型的电力发展水平评价方法  

Evaluating method for power development level based on multi‑objective projection pursuit model

作　　者：李小双 冷亚军 吴坚 LI Xiaoshuang;LENG Yajun;WU Jian(College of Economics and Management,Shanghai University of Electric Power,Shanghai 200090,China;East China Electric Power Design Institute Co.,Ltd.,China Electric Power Consulting Group,Shanghai 200000,China;School of Economics and Management,Shanghai Maritime University,Shanghai 201306,China)

机构地区：[1]上海电力大学经济与管理学院,上海200090 [2]中国电力工程顾问集团华东电力设计院有限公司,上海200000 [3]上海海事大学经济管理学院,上海201306

出　　处：《电力科学与技术学报》2024年第5期46-57,共12页Journal of Electric Power Science And Technology

基　　金：国家自然科学基金项目(71971135);教育部人文社会科学研究青年基金项目(No.22YJCZH073)。

摘　　要：电力生产活动所释放的碳排放在全球碳排放总量中占比较高。因此,电力行业成为推动“碳减排”目标达成的关键责任主体。借助量化综合评价的方式对国家电力发展水平进行评估,不仅可以清晰地描绘出各国在电力领域的发展脉络,而且有助于更准确地识别出中国与其他国家在电力发展上的差距。该文对国家电力发展水平进行综合评价与研究,提出一种基于改进多目标粒子群算法优化投影寻踪模型的电力发展水平评价方法。首先,提出改进的多目标粒子群优化算法;其次,建立2种投影寻踪模型,利用改进的多目标粒子群算法优化构建2种投影寻踪模型,得到投影向量的最优Pareto解集;再次,利用模糊综合评价,得到最优权重折中解,将最优权重代入前景理论模型,得到各个国家电力发展水平的综合评分值,并基于此评分,对各国电力发展水平优劣进行客观排序,最后,利用实际的国家电力发展水平数据集对该文所提方法进行验证。研究结果表明:该方法可实现对国家电力发展水平的有效排序,其评价准确性优于现有的电力发展水平评价方法的。Carbon emissions from electricity production activities account for a significant proportion of total global emissions.Therefore,the power industry has become a key stakeholder in achieving"carbon emission reduction"goals.By employing a quantitative comprehensive evaluation approach to assess national power development levels,we can not only clearly delineate the development trajectory of various countries in the power sector but also more accurately identify the gaps between China and other countries in power development.This paper conducts a comprehensive evaluation and research on national power development levels and proposes an evaluation method for power development levels based on an improved multi-objective particle swarm optimization(MOPSO)algorithm to optimize the projection pursuit model.Firstly,an improved MOPSO algorithm is proposed.Secondly,two projection pursuit models are established,and further optimized using the improved MOPSO algorithm to obtain the optimal Pareto solution set for the projection vectors.Finally,a fuzzy comprehensive evaluation is used to obtain the optimal weight compromise solution,which is then substituted into the prospect theory model to derive comprehensive scores for the power development levels of various countries.Based on these scores,an objective ranking of the power development levels of various countries is conducted.The proposed method is validated using an actual dataset of national power development levels.The experimental results demonstrate that this method can effectively rank national power development levels,with evaluation accuracy superior to existing evaluation methods for power development levels.

关 键 词：电力发展水平 改进的多目标粒子群算法 投影寻踪模型 评价方法 

分 类 号：TM73[电气工程—电力系统及自动化]