【作者】 李媛媛；

【导师】 乞建勋； 牛东晓；

【作者基本信息】 华北电力大学（北京） ， 技术经济及管理， 2008， 博士

【摘要】 电能作为一种优质、清洁能源,提高其在终端能源中的比重,对于优化终端能源消费结构,提高电气化水平具有十分重要的意义。通过对电能消费特性及竞争力的分析研究,为电能竞争力的有效提升提供基础依据,有助于终端能源中电能比重的提高,既可以提高电力企业经济效益,又具有综合的社会效益。分析论述了电力产品和产业特性以及电能在能源消费中的地位和作用,并对国内外能源消费现状和特点及北京地区终端能源消费现状和特点进行了分析,从而为电能在终端能源中竞争力系统要素选择和分析评价指标体系构建提供基础依据。针对电能消费序列呈现复杂变化规律且受环境影响显著的特点,提出基于经验模式分解和因素影响的电能消费序列分析方法。首先,利用经验模式分解方法将电能消费序列分解为若干独立分量;然后,利用多个指标从不同方面分析各分量特性;再通过各分量与影响因素的相关分析,分析环境因素对电能消费的作用;最后,结合各分量自身特性及环境因素对它们的影响分析,归纳出电能消费序列的构成成分,它们规律性显著且包含特征信息集中。实例证明该方法不仅可以挖掘电能消费序列的自身构成及变化特点,且能全面利用环境因素信息。对电能消费需求进行预测建模研究。首先,提出一种基于因散经验模式分解的混合预测方法,通过对电能需求序列的分解和重构过程,将变动规律的序列解析为三个规律性较强的序列分量,继而对各分量分别进行建模预测。同时,还提出一种基于趋势拟合评判的多模型筛选方法,该方法既考虑到中长期电能需求预测侧重趋势预测的要求,又考虑到多模型有效性的判断问题,通过对灰色关联分析方法的改进,统一解决两方面要求,方法简单易行且有较好的预测效果。针对电能在终端能源中竞争力要素涉及多个层面与环节,且相互之间影响关系复杂的特点,在确定了电能竞争力要素系统结构的23个宏观微观要素及其关联关系的基础上,利用解释结构模型化方法,得到电能在终端能源中竞争力要素的系统结构模型,描绘出各要素及子系统之间的层次关系,为研究的进一步深入提供了理论分析框架。基于电能在终端能源中竞争力要素系统结构框架,创建了电能竞争力分析评价指标体系。从竞争结果、竞争实力和竞争潜力三个层面分别展开指标体系创建及评价工作,通过竞争结果评价把握电能消费的整体状况,对竞争实力与竞争潜力的分析评价则可以找到不同层面上影响电能竞争力的要素,从而针对这些要素制定提升策略,以有效提升电能竞争力。在指标体系构建的基础上,给出了分析评价方法与评价流程。更多还原

【Abstract】 It is very important to increase proportion of electric power among the end-use energy market for optimization of end-use energy consumption structure. The research on consumption characteristic and competitive capacity of electric power can provide strong support for upgrading electric power competitive capacity, which can increase the electric power proportion among end-use energy, and the power enterprises’economic benefit and social effect can be increased.The characteristic of electric power and its importance among the energy market were summarized. Meanwhile, the consumption actuality and characteristic of global world was analyzed, especially for Beijing, which can buildup the foundation for choosing of principle components and indexes for competitive capacity evaluation system about electric power in the end-use energy.To counter the complexity of electric power consumption serial, a novel approach for load analysis was presented based on empirical mode decomposition (EMD) and influencing factors. The load sequence was decomposed into several intrinsic modes intuitive, the local characteristic information of load can be reflected by them respective. And several measures were adopted to analyze the regularity of the intrinsic modes. Based on the above procedures, the correlations between the intrinsic modes and environmental factors were studied to mine the degree of different factor influence on the modes. The components of the load were concluded with their characteristics discussed in detail. A case studied shows that with the proposed method the characteristics and influential factors of load can be probed deeper.The model research was take for electric power requirement forecasting. Firstly, an ensemble empirical mode decomposition (EEMD) based hybrid power load forecasting method was proposed. Through the method, the objective series can be decomposed to several independent intrinsic modes adaptively, and these intrinsic modes are reconstructed as three components based on fine-to-coarse. On the basis of analyzing the features of these components, which are forecasted by respective models. Meanwhile, A novel method of multi-model sifting based on evaluation of trend comparability was presented. The degree of trend comparability between each forecasting curve and actual curve was calculated through the improved gray relation analysis. Thus, the gray relationship degree was got which reflects the optimality of an individual forecasting model. Reference to the value of each gray relationship degree, the better models were chosen from multiple candidate ones. Finally, the integrated optimum model was built with the models sorted out.23 principle components were chosen which belong to the system of competitive capacity about electric power in the end-use energy market, and the relationship among them were defined. Adopting interpretative structural modeling(ISM), the layered structure diagram of system above-mentioned was constructed to support further investigation.The evaluation index system of competitive capacity about electric power in the end-use energy market is constructed included three parts, i.e., competitive results, competitive capacity and competitive potential. From competitive result, the whole picture of electric power consumption can be figured out; and from other two, the factors can be picked out which pivotal influence the competitive capacity of electric power. According to the picked factor, strategy can be established to improve competitive capacity. Based on the building of index system, the evaluation method and flow were provided at the end.更多还原