本文選題：綜合集成賦權(quán)　切入點：多參數(shù)灰色預測模型　出處：《青島科技大學》2017年碩士論文

【摘要】：電力變壓器是電力系統(tǒng)中極其重要的設備之一,保障它的安全、穩(wěn)定運行對人們的生產(chǎn)、生活具有很重要的意義。為了使變壓器能夠得到合理的檢修,論文對電力變壓器的狀態(tài)評估及預測方法做了深入研究。不僅建立了比較系統(tǒng)的狀態(tài)評估指標體系,而且針對電力變壓器狀態(tài)的評估和預測都做了大量的仿真和實例論證,最終確定了改進的評估和預測方法真實有效,具有科學性。具體的工作內(nèi)容如下:(1)論文對電力變壓器的狀態(tài)評估指標體系做了研究�？茖W有效的評估指標體系是狀態(tài)評估的基礎,論文在搜集整理大量技術(shù)標準、規(guī)程導則、專家經(jīng)驗以及變壓器實際運行狀態(tài)數(shù)據(jù)的基礎上,深入研究了電力變壓器狀態(tài)評估的指標體系。最終從油色譜試驗、電氣試驗、油化試驗三個方面建立了完整的電力變壓器狀態(tài)評估指標體系,保證了變壓器狀態(tài)評估和預測的準確性。(2)論文對電力變壓器的狀態(tài)評估做了研究。論文將變壓器的狀態(tài)等級進一步細化,將其劃分為5個等級。在此基礎之上論文不僅將集對分析理論和模糊理論應用到了電力變壓器狀態(tài)評估中,而且從變壓器狀態(tài)評估的每一步仔細剖析,提出了一種主客觀綜合集成賦權(quán)的算法。該算法綜合考慮主觀專家經(jīng)驗和客觀事實,使變壓器狀態(tài)評估的指標權(quán)重更加可靠、合理。運用主觀賦權(quán)法中的5級標度法充分體現(xiàn)了專家經(jīng)驗對變壓器狀態(tài)評估的作用;運用客觀賦權(quán)法中的熵值法充分體現(xiàn)了各指標數(shù)據(jù)的變化情況在權(quán)重確定中的作用,符合客觀事實,最后采用最小二乘法的原理綜合集成了主客觀權(quán)重,得出的指標權(quán)重從理論上講最科學。最終評估出來的變壓器狀態(tài)等級結(jié)果與其他方法相比最為準確。這一改進的評估算法為變壓器的狀態(tài)檢修提供了有力保障。(3)論文對電力變壓器的狀態(tài)預測做了研究。本文主要對狀態(tài)預測過程中的數(shù)據(jù)預處理和預測模型方面進行了研究。在數(shù)據(jù)預處理方面的創(chuàng)新性在于使用了平均弱化算子處理方式,電力變壓器的特征氣體數(shù)據(jù)經(jīng)過平均化處理和弱化處理之后,它的平滑性更加符合預測模型的要求,預測的精度相對而言就有所提高。在預測模型方面的改進在于論文運用了多參數(shù)灰色預測模型MGM(1,N)模型對電力變壓器的特征氣體進行分組預測。分組方式是以電力變壓器常用故障診斷方法三比值診斷法為基礎進行分組,這個方法可以避免多參數(shù)灰色預測模型預測所需要的關(guān)聯(lián)度分析,從而簡化了模型的計算步驟,計算時間也得到了節(jié)省。通過對實驗結(jié)果的分析可以看出采用了基于三比值法分組的平均弱化多參數(shù)灰色預測模型在預測方面準確度更高,更能把握系統(tǒng)的規(guī)律性。最后利用理想點解理論和預測出來的數(shù)據(jù)對電力變壓器的狀態(tài)進行了評估,結(jié)果表明真實、可靠。
[Abstract]:Power transformer is one of the most important equipments in power system. It is very important to ensure its safety and stable operation for people's production and life.In order to enable the transformer to get reasonable overhaul, the paper makes a deep research on the state evaluation and prediction method of power transformer.Not only a systematic evaluation index system is established, but also a large number of simulations and examples are made for the evaluation and prediction of power transformer status. Finally, it is determined that the improved evaluation and prediction method is true, effective and scientific.The main work is as follows: (1) the paper studies the power transformer condition evaluation index system.A scientific and effective evaluation index system is the basis of state evaluation. Based on the collection and arrangement of a large number of technical standards, rules and guidelines, expert experience and actual operation status data of transformers,The index system of power transformer condition evaluation is studied in depth.Finally, a complete power transformer condition evaluation index system is established from three aspects: oil chromatographic test, electric test and oil test.It ensures the accuracy of transformer state evaluation and prediction.In this paper, the state of the transformer is further refined and divided into five grades.On this basis, the paper not only applies the set pair analysis theory and fuzzy theory to the power transformer state evaluation, but also from each step of the transformer state evaluation, proposes a subjective and objective integrated weighting algorithm.This algorithm synthetically considers the subjective expert experience and objective facts, and makes the index weight of transformer condition evaluation more reliable and reasonable.The effect of expert experience on transformer condition evaluation is fully reflected by the five-grade scale method in subjective weighting method, and the function of the change of index data in determining the weight of transformer is fully reflected by the entropy value method in objective weighting method.In accordance with the objective facts, the principle of least square method is used to synthesize the subjective and objective weights, and the index weights are theoretically the most scientific.The result of the final evaluation is the most accurate compared with other methods.This improved evaluation algorithm provides a strong guarantee for the condition maintenance of transformers.) the paper studies the state prediction of power transformers.In this paper, data preprocessing and prediction model in the process of state prediction are studied.The innovation in data preprocessing is that the average weakening operator is used to process the characteristic gas data of power transformer, and the smoothness of the characteristic gas data of power transformer is more in line with the requirement of prediction model.The accuracy of the prediction is relatively improved.The improvement of the prediction model is that the multi-parameter grey prediction model MGM1N) is used to predict the characteristic gases of power transformers in groups.The grouping method is based on the three-ratio diagnosis method commonly used in fault diagnosis of power transformers. This method can avoid the correlation analysis needed for the prediction of multi-parameter grey prediction model and simplify the calculation steps of the model.Computational time is also saved.Through the analysis of the experimental results, it can be seen that the average weakening multi-parameter grey prediction model based on the three-ratio method is more accurate in forecasting and can grasp the regularity of the system better.Finally, the state of power transformer is evaluated by using the theory of ideal point solution and the predicted data. The results show that it is true and reliable.
【學位授予單位】：青島科技大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TM41

【參考文獻】

相關(guān)期刊論文 前10條

1 戚玉璽;祝華遠;段剛;;基于綜合集成賦權(quán)法的飛行機務保障能力評估指標權(quán)重確定[J];兵工自動化;2016年04期

2 陳金強;李群湛;;基于灰色TOPSIS和DGA的變壓器狀態(tài)預測[J];高壓電器;2015年09期

3 閆敏慧;姚秀萍;王蕾;張金峰;;用層次分析法確定氣象服務評價指標權(quán)重[J];應用氣象學報;2014年04期

4 楊杰;翁文國;;基于改進無偏灰色模型的燃氣供氣量的預測[J];清華大學學報(自然科學版);2014年02期

5 楊映滋;羅日成;黃知明;黃彪;張f;;模糊集對分析在大型變壓器狀態(tài)評價中的應用[J];電氣技術(shù);2013年01期

6 廖瑞金;張鐿議;黃飛龍;鄭含博;楊麗君;;基于可拓分析法的電力變壓器本體絕緣狀態(tài)評估[J];高電壓技術(shù);2012年03期

7 楊華龍;劉金霞;鄭斌;;灰色預測GM(1,1)模型的改進及應用[J];數(shù)學的實踐與認識;2011年23期

8 吳龍山;;基于緩沖算子的灰色瓦斯涌出量預測研究[J];科技情報開發(fā)與經(jīng)濟;2011年30期

9 王清源;潘旭海;;熵權(quán)法在重大危險源應急救援評估中的應用[J];南京工業(yè)大學學報(自然科學版);2011年03期

10 杜林;袁蕾;熊浩;唐綱;李剛;孫才新;;電力變壓器運行狀態(tài)可拓層次評估[J];高電壓技術(shù);2011年04期

相關(guān)博士學位論文 前3條

1 鄭含博;電力變壓器狀態(tài)評估及故障診斷方法研究[D];重慶大學;2012年

2 吳立增;變壓器狀態(tài)評估方法的研究[D];華北電力大學（河北）;2005年

3 朱建軍;層次分析法的若干問題研究及應用[D];東北大學;2005年

相關(guān)碩士學位論文 前6條

1 伊弘博;基于電力變壓器故障特征氣體的預測方法研究[D];青島科技大學;2015年

2 段侯峰;基于遺傳算法優(yōu)化BP神經(jīng)網(wǎng)絡的變壓器故障診斷[D];北京交通大學;2008年

3 喬俊玲;人工神經(jīng)網(wǎng)絡在變壓器油中溶解氣體分析中的應用研究[D];北京交通大學;2008年

4 廖玉祥;一種電力變壓器運行狀態(tài)綜合評估模型的研究[D];重慶大學;2006年

5 王謙;基于模糊理論的電力變壓器運行狀態(tài)綜合評估方法研究[D];重慶大學;2005年

6 吳欣;基于改進貝葉斯網(wǎng)絡方法的電力系統(tǒng)故障診斷研究[D];浙江大學;2005年

，

本文編號：1713982