基于模糊贴近度理论的负荷密度指标方法的模型和评估因素影响研究

论文总字数：30570字

摘 要

以往负荷密度指标的得出主要靠经典预测方法包括时间序列法和回归分析法，精确度比较差。而现在海内外有了很多负荷预测的新方法，比如用回归模型法、负荷密度指标法、依赖于人工神经网络的总体负荷预测技术和灰色预测法等办法。本文利用了模糊贴近度理论来求取负载密度标准。模糊贴近度理论是一种能有效确定负荷密度指标的方法，但是具体实施中由于模糊集合关系的贴近度表示模型多样，其评价指标的影响因素的构成不一，为了更好地构建评估指标集合需要对贴近度模型和主要评估因素的影响作用进行评估。基于模糊贴近度理论的负载密度标准求取需要先依据城市用地规划将不同用地类型的负荷划分成不同的区域，接着研究待测对象，在待测区内把相邻的相同类型小区合并，并划分出不同类型的小区的边界，生成元胞。然后对于不同性质的负载分开进行预测计算，得出不同性质负载的密度，接着乘以占地大小求出小区负载，加起来得到待测对象的整体负载。这种方法是先分类，再求和，科学合理。本文把低水平样本作为基准样本，求出不同等级标准样本对其的贴近度，以此为依据来划分出几个负荷密度等级，之后利用六种贴近度模型，即加权海明贴近度公式、最大最小贴近度公式、最小平均贴近度公式、海明贴近度公式、欧几里得贴近度公式与兰伯特贴近度公式计算待测区对低水平样本的贴近度，并给出待测区参考负荷密度指标。接着，本文将用各个公式得出的结果进行比较，找出精度最高的几个公式，并分析用不同公式得出结果不同的原因。本文最终得出基于距离的贴近度公式精度更高的结论。

关键词：负荷预测；评估因素；模糊贴近度；负荷密度指标

ABSTRACT

Previous load density is derived mainly depending on classical prediction methods, including time series and regression analysis,and the accuracy is relatively poor. At present, there are many new methods to forecast the load at home and abroad, such as regression model method,load density index method,general load forecasting technique which depends on the artificial neural networks and gray prediction method and so on. In this paper,the fuzzy proximity theory is used to get the load density index. The fuzzy proximity theory is a method to determine the load density index effectively. However,because the representation models of the proximity of fuzzy set relation are that diverse,the Influencing factors of the evaluation index are different. In order to better construct the evaluation index set,the impact of the proximity model and the main assessment factors needs to be evaluated.The load density index standard based on the fuzzy close degree theory needs to first divide the load of different land types into different units according to the urban land plan, then study the area to be measured,merge the adjacent same types of cells in the area to be measured, and divide out the boundaries of the different types of units, generating cells. And then predict and calculate separately for the load of different nature,get the density of the load , and then multiplied by the size of the district to get the load of the district,add all to get the overall load of the object to be measured.This method first makes classification, and then sum.It's scientific and reasonable.In this paper, the low level of sample is used as the reference sample to obtain the close degree of different levels of standard sample on it.Then the close degree is used as a basis for the breakdown of several load density level. Then we use six Proximity models, namely Hamming proximity formula,the maximum and minimum proximity formula,the minimum average proximity formula,the Hamming proximity formula, the Euclid proximity formula and the Lambert proximity formula to calculate the proximity of the forecast area to the low level sample and give the reference load density index of the forecast area. Then,this paper will use the results obtained by the various formulas to compare in order to find several formulas which have the highest precision and analysis why different formulas have different results.This paper finally concludes that the accuracy of distance-based approach is higher.

KEY WORDS: load forecasting;assessment factors;fuzzy proximity;load density index

目 录

摘要..............................................................................................................................................I

Abstract ......................................................................................................................................II

目录.............................................................................................................................................III

第一章 绪论................................................................................................................................1

1.1 研究背景与意义.......................................................................................................1

1.2负荷预测概述............................................................................................................2

1.3 负荷预测研究方法综述...........................................................................................2

1.4本文主要工作............................................................................................................2

第二章 模糊贴近度理论概述..................................................................................................... 4

2.1模糊贴近度理论介绍................................................................................................ 4

2.2模糊模式识别的方法................................................................................................ 4

2.2.1模糊模式识别的直接方法................................................................................ 4

2.2.2模糊模式识别的间接方法................................................................................ 6

2.3常用贴近度求取公式................................................................................................ 6

2.4小结............................................................................................................................ 8

第三章 基于模糊贴近度理论的空间负荷密度指标预测......................................................... 9

3.1空间负荷预测概述.................................................................................................... 9

3.2基于模糊贴近度理论的空间负荷密度指标............................................................. 9

3.2.1基于负荷密度指标的空间负荷预测........................................................ .......9

3.2.2负荷分类分析................................................................................................... 10

3.2.3平均指标和标准样本........................................................................................ 10

3.2.4隶属函数........................................................................................................... 10

3.3利用模糊贴近度理论的负荷密度指标求取过程.................................................... 11

3.4小结............................................................................................................................11

第四章 算例分析........................................................................................................................ 12

4.1对安徽凤台县算例的分析........................................................................................12

4.1.1构造标准样本集和标准样本指标集................................................................ 12

4.1.2构造隶属函数................................................................................................... 13

4.1.3计算贴近度，确定模糊集的范围.................................................................... 14

4.2对上海黄浦区算例的分析......................................................................................... 16

4.2.1构造标准样本集和标准样本指标集 .............................................................. 16

4.2.2构造隶属函数................................................................................................... 18

4.2.3计算贴近度，确定模糊集的范围.................................................................... 19

4.3对苏州工业园区算例的分析..................................................................................... 20

4.3.1构造标准样本集和标准样本指标集................................................................ 22

4.3.2构造隶属函数................................................................................................... 23

4.3.3计算贴近度，确定模糊集的范围................................................................... 23

4.4结果分析.................................................................................................................... 25

剩余内容已隐藏，请支付后下载全文，论文总字数：30570字