水基動力無桿抽油機關鍵部件故障診斷方法研究
[Abstract]:The water-based rodless pumping unit is a new type of pumping equipment which uses hydraulic power to promote the underground pumping pump work. It has the characteristics of high efficiency and low power consumption, and can completely eliminate the problem of rod and pipe bias wear existing in the traditional pumping unit. Therefore, it has a broad application prospect. The research of state detection and fault diagnosis can ensure the safety and stability of the equipment and improve the production efficiency of the equipment. It has important theoretical significance and engineering application value. The main contents and innovations of this paper are as follows: (1) the dynamics of the power cylinder, the key component of the water-based rodless pumping unit, is studied, and the kinematics equation of the cylinder is derived. The mapping relationship between the variation of kinematics combined parameters of power cylinder and the fault under different faults is established, and the method of judging the fault type by using the variation of combined parameters of power cylinder is put forward. This method can reveal the fault types of the power cylinder of the water-base rodless pumping unit and provide a theoretical basis for its fault diagnosis. (2) the feasibility of the application of HMM in the fault diagnosis of the water-base rodless pumping unit is discussed. The research on fault diagnosis method of water-based rod-less pumping unit based on HMM is carried out, and the flow chart of fault diagnosis is given. In this paper, HMM method is used to diagnose four kinds of faults, such as normal production of water base power rod pumping unit, leakage of power hydraulic pipe, swam plug, counterweight wear, fixed fan clogging, etc. The results show that this method is suitable for fault diagnosis of water-based rodless pumping units, and can achieve high accuracy. (3) A fault diagnosis method for water-based rodless pumping units based on higher-order cumulant spectrum is studied. A method for fault diagnosis of water-based rodless pumping units is proposed, which combines 2.5 dimension spectrum with HMM. The comparison between this method and the single application of HMM in fault diagnosis shows that the fault diagnosis method of hydrodynamic rod less pumping unit based on HMM with 2.5 D spectrum has higher fault diagnosis accuracy. And the speed of model training is also greatly improved. (4) the test bench of water-base rodless pumping unit is constructed, and the circuit control system and man-machine interface of the test bed are designed. The test rig can be used to simulate the normal working state and various fault states of water-based rodless pumping units. A simulation system for condition monitoring and fault diagnosis of water-based rodless pumping unit is developed. The system can simulate the downhole working state and can be used to observe the running state of the key components directly. It provides a visual method for condition detection and fault diagnosis of water-based rod-less pumping unit.
【學位授予單位】:北京信息科技大學
【學位級別】:碩士
【學位授予年份】:2015
【分類號】:TE933.1
【參考文獻】
相關期刊論文 前10條
1 李志華;;高階統(tǒng)計量方法在地球物理學中的應用[J];工程地球物理學報;2006年05期
2 胡勁松,楊世錫,吳昭同,嚴拱標;基于EMD的旋轉(zhuǎn)機械振動信號Winger分布分析[J];機床與液壓;2003年05期
3 周云龍;柳長昕;宋延宏;趙鵬;孫斌;洪文鵬;;基于AR的二維隱Markov模型離心泵故障診斷方法[J];流體機械;2008年10期
4 劉偉;;水基液壓雙作用無桿采油系統(tǒng)[J];中國石油和化工標準與質(zhì)量;2013年08期
5 許寶杰 ,韓秋實 ,徐小力;基于隱馬爾可夫模型的機器狀態(tài)趨勢預測方法[J];精密制造與自動化;2003年S1期
6 吳偉;何焱;陳國定;;基于MATLAB的神經(jīng)網(wǎng)絡有桿抽油系統(tǒng)故障診斷[J];石油礦場機械;2006年04期
7 魏秦文;張茂;郭詠梅;;潛油電機驅(qū)動采油技術的發(fā)展[J];石油礦場機械;2007年07期
8 沐峻丞;檀朝東;孟祥芹;曾霞光;;應用MATLAB中的BP神經(jīng)網(wǎng)絡診斷抽油機井工況[J];中國石油和化工;2010年04期
9 李建華;;在線設備狀態(tài)監(jiān)測與故障診斷技術的應用[J];石油化工設備;2010年03期
10 王國鋒;李玉波;秦旭達;喻秀;李啟銘;;基于TVAR-HMM的滾動軸承故障診斷[J];天津大學學報;2010年02期
相關博士學位論文 前2條
1 肖文斌;基于耦合隱馬爾可夫模型的滾動軸承故障診斷與性能退化評估研究[D];上海交通大學;2011年
2 顏秉勇;非線性系統(tǒng)故障診斷若干方法及其應用研究[D];上海交通大學;2010年
相關碩士學位論文 前2條
1 王文莉;基于高階譜的齒輪故障診斷與識別[D];武漢科技大學;2007年
2 沙亮;潛油電機設計方法研究及實驗驗證[D];哈爾濱理工大學;2012年
本文編號:2281568
本文鏈接:http://sikaile.net/kejilunwen/shiyounenyuanlunwen/2281568.html