考虑决策单元异质性的中国省域能源效率研究

发布时间：2018-06-04 13:39

本文选题：数据包络分析(DEA) + 异质性　；参考：《山西大学》2017年硕士论文

【摘要】：目前的现实背景是:世界各国都面临着生态环境问题愈发严重的能源资源危机。2009年后,中国成为了全世界消耗能源最多的国家,也是发展中国家中排放CO2最多的,节省能源消耗和减少废气排放是长时间以来留给中国的非常重要的历史任务和使命,这个是我国必须肩负起的。所以这就使得合理系统得测算我国的省域能源利用效率以及在结果之上提出有针对性的能源政策,给我国有智慧地走未来道路赋予了非同一般的现实意义。先测算能源效率然后提出能够提升能源效率的方法是目前大量文献的主要关注点。本文不同于传统的DEA方法,考虑能源现实环境中决策单元的投入产出指标的非同质现象,先将决策单元内部元素拆分为一系列和决策单元有着相同的投入产出指标的互斥的子决策单元;然后采纳DEA交叉效率方法计算子决策单元效率;最后通过子决策单元加权平均效率得出决策单元综合效率,以此来测算中国31个省份2010年到2014年的能源效率排名。最后得出5年来依然是中西部被东部远远领先的结论。虽然三个区域能源效率值有所起伏,但整体上能源效率都有所增加。本论文的创新之处有以下3个方面:(1)在多流程DEA模型中测算子决策单元的流程效率时,采用DEA交叉效率模型代替传统的CCR模型,交叉效率模型采用相对被评价单元的自我评价和他人评价两种方法融合的办法来测算决策单元的相对效率,减弱了经典DEA模型结论有过多一致的效率值从而无法准确判断效率大小的弊端。(2)在选用多流程DEA模型计算出2010年到2014年中国各省域能源效率后,采用灰色关联TOPSIS法对31个省份的能源效率值进行排序,最终得出综合排序。由于TOPSIS法有着需要决策的方案和最佳值的贴近度不能从形状上反映的缺陷,灰色关联刚好是反映变化形势的贴近程度从而弥补了缺陷,所以通过灰色关联与TOPSIS相结合,可以弥补灰色关联和TOPSIS法各自的缺点,提升决策结果的全面性。(3)区别于传统方法的投入指标选取,前人的文献中,在测算能源效率时,能源资源是作为一个投入指标测算的,而本文将能源拆分为煤、天然气和石油三个投入指标进行测算,虽测算结果与前人文献有所差异,但可以从更具体的能源资源使用情况分析各省域的能源效率问题,得出更具针对性的结论。
[Abstract]:At present, the realistic background is that every country in the world is facing an increasingly serious energy and resource crisis of ecological and environmental problems. After 2009, China has become the most energy-consuming country in the world and the largest emitter of CO2 among developing countries. Saving energy consumption and reducing exhaust emissions is a very important historical task and mission for China for a long time, which our country must shoulder. Therefore, this makes it reasonable and systematic to measure the energy efficiency of our country and put forward the targeted energy policy on the basis of the results, which gives our country an extraordinary practical significance to take the future path intelligently. First measuring energy efficiency and then proposing methods to improve energy efficiency is the main focus of a large number of literatures. In this paper, different from the traditional DEA method, we consider the input-output index of the decision making unit in the energy realistic environment. First, the internal elements of the decision making unit are divided into a series of mutually exclusive decision units with the same input-output index as the decision making unit, and then the efficiency of the sub-decision unit is calculated by using the DEA cross-efficiency method. Finally, the comprehensive efficiency of Decision-making units is obtained by weighted average efficiency of sub-Decision-making units, and the energy efficiency rankings of 31 provinces in China from 2010 to 2014 are calculated. Finally came to the conclusion that the central and western regions are still far ahead of the east in the past five years. Although the energy efficiency of the three regions fluctuates, the overall energy efficiency has increased. The innovations of this paper are as follows: 1) when measuring the process efficiency of sub-decision units in the multi-process DEA model, the DEA cross-efficiency model is used instead of the traditional CCR model. The cross-efficiency model measures the relative efficiency of the decision making unit by combining the self-evaluation of the relative evaluated unit and the evaluation of others. It weakens the drawback that the classical DEA model has too many consistent efficiency values so that it can not accurately judge the efficiency. (2) after selecting the multi-process DEA model to calculate the energy efficiency in the provinces of China from 2010 to 2014, The energy efficiency value of 31 provinces is sorted by grey correlation TOPSIS method, and finally the comprehensive ranking is obtained. Because the TOPSIS method has the defect that the scheme and the best value's closeness cannot be reflected from the shape, the grey relation is just the close degree to reflect the changing situation, so it combines the grey correlation with TOPSIS. It can make up for the shortcomings of grey correlation and TOPSIS method, and improve the comprehensive decision result. It is different from the traditional method in selecting input index. In previous literature, energy resources are calculated as an input index in energy efficiency measurement. In this paper, energy is divided into three input indexes, coal, natural gas and petroleum. Although the calculation results are different from the previous literatures, we can analyze the energy efficiency in each province from the more specific situation of energy resources use. Draw more targeted conclusions.
【学位授予单位】：山西大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F426.2

【参考文献】