【摘要】：綜合評價是人們對事物的一種全面認識,其研究對象包括指標體系的選取、權重的計算、評價方法的選擇以及評價結(jié)果的可靠性分析等。經(jīng)過幾十年的發(fā)展,開發(fā)了近百余種經(jīng)典的綜合評價模型,其理論基礎已經(jīng)相對成熟,但是在評價方法的優(yōu)化改進、權重的客觀計算和評價方法的合理選擇等方面還存在一定的不足。 根據(jù)目前綜合評價方法存在的主要問題,本文做了如下研究工作： 1.基于對層次分析法(AHP法)、模糊綜合評價法(FCE法)、數(shù)據(jù)包絡分析法(DEA法)、灰色綜合評價法(GRAP法)和人工神經(jīng)網(wǎng)絡評價法(ANN法)的分析研究,詳細地總結(jié)了各方法的優(yōu)缺點、適用范圍以及模塊設計。 2.改進了層次分析方法。在傳統(tǒng)層次分析法的判斷矩陣中,專家只考慮了矩陣元素αij最可能出現(xiàn)的值,忽視了人對事物判斷的模糊性和主觀性。有學者對判斷矩陣的構(gòu)造做了一些改進,但是改進的模型不具有廣泛代表性。本文從兩個方面對AHP法做了有效的改進：(1)用統(tǒng)計分析的方法證明了更具一般性的改進模型,構(gòu)造了判斷函數(shù),對改進的模型進行了一定的討論,得出了更合理的判斷矩陣構(gòu)造方法；(2)引入了幾種新的權重計算方法,省略了一致性檢驗的復雜過程。 3.通過對AHP法、FCE法、DEA法等綜合評價方法的優(yōu)缺點分析,采用取長互補的原則,分別用模糊綜合評價法、數(shù)據(jù)包絡分析法、灰色綜合評價法和人工神經(jīng)網(wǎng)絡評價法與層次分析法進行合理集成,形成更加客觀、合理的綜合評判模型。 4.通過對油藏經(jīng)營管理理論基礎和運營模式的深入分析研究,選擇了具有一定代表性的評價指標,構(gòu)建了油藏經(jīng)營管理層次結(jié)構(gòu)�？偨Y(jié)并建立了篩選綜合評價方法的原則,結(jié)合油藏經(jīng)營管理實例,采用模糊層次分析法,從本文兩兩集成的方法中,首次分析了如何選擇相對適宜的評價方法過程。 5.建立了適合油藏經(jīng)營管理的非線性模糊層次分析評價模型。為了更好地反映指標對評價結(jié)果的突出影響,結(jié)合油藏經(jīng)營管理指標體系的模糊性、不確定性和非線性的特點,建立了適合油藏經(jīng)營管理的非線性模糊層次分析評價模型。該模型既有傳統(tǒng)模糊綜合評價模型的優(yōu)點,又能夠反映個性指標的突出影響,克服了傳統(tǒng)線性模糊綜合評價模型對突出影響因素的不足,有效地決策出了油藏經(jīng)營管理最優(yōu)方案。
[Abstract]:Comprehensive evaluation is a comprehensive understanding of things. Its research objects include the selection of index system, the calculation of weights, the selection of evaluation methods and the reliability analysis of evaluation results. After decades of development, more than 100 classical comprehensive evaluation models have been developed. Their theoretical basis has been relatively mature, but the evaluation method has been optimized and improved. The objective calculation of weights and the reasonable selection of evaluation methods still have some shortcomings. According to the main problems existing in the current comprehensive evaluation method, this paper has done the following research work: 1. Based on the analysis of Analytic hierarchy process (AHP), Fuzzy Comprehensive Evaluation (FCE), data Envelopment Analysis (DEA), Grey Comprehensive Evaluation (GRAP) and artificial Neural Network Evaluation (ANN). The advantages and disadvantages, application scope and module design of each method are summarized in detail. 2. The analytic hierarchy process is improved. In the traditional analytic hierarchy process (AHP) judgment matrix, experts only consider the most likely value of matrix element 偽 ij, and ignore the fuzziness and subjectivity of human judgment on things. Some scholars have improved the construction of judgment matrix, but the improved model is not widely representative. In this paper, the AHP method is improved from two aspects: (1) A more general improved model is proved by statistical analysis, a judgment function is constructed, and the improved model is discussed. A more reasonable method of constructing judgment matrix is obtained. (2) several new weight calculation methods are introduced, and the complicated process of consistency checking is omitted. 3. By analyzing the advantages and disadvantages of AHP method, FCE method, DEA method and so on, the principle of length complementarity is adopted, and fuzzy comprehensive evaluation method and data envelopment analysis method are used, respectively. The grey comprehensive evaluation method and the artificial neural network evaluation method are reasonably integrated with the analytic hierarchy process to form a more objective and reasonable comprehensive evaluation model. 4. Through the deep analysis of the theoretical basis and operation mode of reservoir management, the representative evaluation index is selected and the hierarchical structure of reservoir management is constructed. The principle of screening comprehensive evaluation method is summarized and established. Combining with an example of reservoir management and using fuzzy analytic hierarchy process, this paper analyzes for the first time how to select a relatively suitable evaluation method from the method of pairwise integration in this paper. 5. A nonlinear fuzzy analytic hierarchy process evaluation model suitable for reservoir management is established. In order to better reflect the prominent influence of indicators on the evaluation results, a nonlinear fuzzy analytic hierarchy process evaluation model suitable for reservoir management and management was established by combining the fuzziness, uncertainty and nonlinearity of the index system of reservoir management and management. This model not only has the advantages of the traditional fuzzy comprehensive evaluation model, but also can reflect the prominent influence of the personality index, which overcomes the deficiency of the traditional linear fuzzy comprehensive evaluation model on the outstanding influencing factors. The optimal scheme of reservoir management is determined effectively.
【學位授予單位】：西南石油大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：F224;F426.22

【參考文獻】

相關期刊論文 前10條

1 駱正清;層次分析法中判斷矩陣構(gòu)造的新方法[J];電子科技大學學報;1999年05期

2 左軍;層次分析法中判斷矩陣的間接給出法[J];系統(tǒng)工程;1988年06期

3 張曉慧,馮英浚,白莽;一種反映突出影響因素的評價模型[J];哈爾濱工業(yè)大學學報;2003年10期

4 陳衍泰,陳國宏,李美娟;綜合評價方法分類及研究進展[J];管理科學學報;2004年02期

5 樊為剛,侯麗紅;層次分析法的改進[J];科技情報開發(fā)與經(jīng)濟;2005年04期

6 侯翔;馬占新;趙春英;;數(shù)據(jù)包絡分析模型評述與分類[J];內(nèi)蒙古大學學報(自然科學版);2010年05期

7 王賢琳;張華;宋佳佳;謝助新;;綠色制造企業(yè)一體化管理體系及其評價指標體系構(gòu)建[J];機械工業(yè)標準化與質(zhì)量;2010年03期

8 魯柳利;謝祥俊;;基于DEA的油藏經(jīng)營管理有效性評價研究[J];西南石油大學學報(社會科學版);2009年05期

9 郭竹梅;;AHP中判斷矩陣一致性改進的一種新方法[J];齊齊哈爾大學學報(自然科學版);2010年06期

10 張輝;高德利;;基于模糊數(shù)學和灰色理論的多層次綜合評價方法及其應用[J];數(shù)學的實踐與認識;2008年03期

，

本文編號：2373895