【正文】 畢業(yè)論文 艦用燃氣輪機某關鍵部件故障診斷方法研究 系 別 信息工程系 專 業(yè) 測控技術與儀器 班 級 學 號 姓 名 指導教師 負責教師 沈陽航空 航天大 學 北方科技學院 2021 年 6 月沈陽航空航天大學北方科技學院畢業(yè)設計（論文） I 摘 要 燃氣輪機的關鍵部件之一滾動軸承是機械設備運行過程中產(chǎn)生最易產(chǎn)生故障的零件，它運行的正常與否直接影響到整臺機器的性能。防止故障升級，發(fā)生災難性事故。所以對滾動軸承故障診斷技術進行探討和學習就具有十分重要的意義。 本文主要以燃氣輪機的滾 動軸承為研究對象，利用測量的軸承振動信號參數(shù)來進行故障診斷，利用神經(jīng)網(wǎng)絡技術對某一動態(tài)的模擬原理，應用到對滾動軸承故障診斷的具體方面，設計并構(gòu)建了基于 BP 神經(jīng)網(wǎng)絡和 自適應模糊神經(jīng)網(wǎng)絡 （ Adaptive Network Fuzzy Inference System） 的滾動軸承故障診斷系統(tǒng) ， 在 MATLAB 軟件里對構(gòu)造的訓練樣本進行訓練，利用訓練完成后的神經(jīng)網(wǎng)絡我們就可以對滾動狀態(tài)故障進行診斷。 關鍵詞 ： 滾動軸承； BP 神經(jīng)網(wǎng)絡；模糊神經(jīng)網(wǎng)絡 沈陽航空航天大學北方科技學院畢業(yè)設計（論文） II Abstract Rolling bearing is one of the most ordinary parts in mechanical machine, its running state can influence the performance of the whole machine directly, the aircraft stabilizer health status need to be monitored in real time to ensure the aircraft fly safety. so it is important to study the technology of fault diagnosis for rolling bearing. On the basis of analyzing the fault mechanism and vibration signal characteristics of rolling bearing systematically, and after analyzing and processing the vibration signals of right and fault state of rolling bearing, partial appropriate feature parameters are selected as the input of the neural work according to the time and frequency domain characteristics of parameters in this thesis. and the fault diagnosis system for rolling bearing based on BP neural work is built up. Finally,and fuzzy artificial neural work diagnosis technique the training set of right and fault states of rolling bearing is built up by using the measuring data of rolling bearing from former research, the neural work model is trained on the platform of Matlab operating state of rolling bearing has been diagnosed by using the above work which has been trained well. Keywords: rolling bearing。 BP neural work。 fuzzy artificial neural work 沈陽航空航天大學北方科技學院畢業(yè)設計（論文） III 目 錄 1 緒 論 ............................................................................................................................... 1 課題背景和研究的意義 ............................................................................................. 1 研究背景 .............................................................................................................. 1 研究意義 .............................................................................................................. 2 國內(nèi)外故障診斷技術研究現(xiàn)狀 ................................................................................. 3 國外現(xiàn)狀 .............................................................................................................. 3 國內(nèi)現(xiàn)狀 .............................................................................................................. 4 研究內(nèi)容 ..................................................................................................................... 5 2 燃氣輪機關鍵部件故障診斷的總體方案設計 ............................................................... 6 滾動軸承故障的總體方案設計 ................................................................................. 6 滾動軸承機理及故障類型 ......................................................................................... 6 滾動軸承基本結(jié)構(gòu) .............................................................................................. 7 滾動軸承故障類型 .............................................................................................. 8 滾動軸承參數(shù)選取及處理 ......................................................................................... 9 3 故障診斷方法 .................................................................................................................. 12 BP 神經(jīng)網(wǎng)絡 .............................................................................................................. 12 BP 神經(jīng)網(wǎng)絡原理 ............................................................................................... 12 BP 神經(jīng)網(wǎng)絡診斷方法 ....................................................................................... 13 自適應模糊神經(jīng)網(wǎng)絡（ ANFIS） ............................................................................. 16 自適應 模糊神經(jīng)網(wǎng)絡（ ANFIS）結(jié)構(gòu) ............................................................. 17 自適應模糊神經(jīng)網(wǎng)絡（ ANFIS）原理 ............................................................. 18 自適應模糊神經(jīng)網(wǎng)絡（ ANFIS）診斷方法 ..................................................... 20 4 滾動軸承 BP 神經(jīng)網(wǎng)絡故障診斷試驗研究 .................................................................. 23 故障診斷模型建立 ................................................................................................... 23 各層節(jié)點數(shù)確定 ................................................................................................ 23 初始權(quán)值的選擇 ................................................................................................ 25 期望誤差和學習率選取 .................................................................................... 25 軸承故障診斷的仿真試驗研究 ............................................................................... 25 樣本選取 ............................................................................................................ 25 神經(jīng)網(wǎng)絡的訓練 ................................................................................................ 27 沈陽航空航天大學北方科技學院畢業(yè)設計（論文） IV 神經(jīng)網(wǎng)絡的測試 ................................................................................................ 28 5 基于自適應模糊神經(jīng)網(wǎng)絡（ ANFIS）的故障診斷試驗研究 ...................................... 30 診斷模型的建立 ....................................................................................................... 30 樣本的選取 ........................................................................................................ 30 輸入樣本的模糊化處理 .................................................................................... 31 輸出結(jié)果的去模糊化處理 ................................................................................ 32 軸承故障 診斷仿真與結(jié)果 ..........