【正文】 ，得到訓(xùn)練誤差曲線，再在訓(xùn)練完成的神經(jīng)網(wǎng)絡(luò)上進(jìn)行測試 和仿真 ，得出仿真結(jié)果正確率 。本文以機械設(shè)備滾動軸承故障診斷問題為背景， 針對傳統(tǒng)的時頻分析方法難以全面反映故障信息的缺陷， 探討了 BP(Back Propagation,反向傳播 )神經(jīng)網(wǎng)絡(luò)技術(shù)在滾動軸承故障診斷中的應(yīng)用。蘭州交通大學(xué)畢 業(yè)設(shè)計（論文） I 基于 BP 神經(jīng)網(wǎng)絡(luò)的滾動軸承故障診斷方法初探 摘 要 滾動軸承是機械設(shè)備中最常見、應(yīng)用最廣泛的零部件之一，其運行狀態(tài)對整個設(shè)備的工作狀態(tài)、生產(chǎn)過程都有直接影響。 因此 對 軸承 的 故障診斷 具有非常重要的意義。 選取 滾動軸承三種故障類型（內(nèi)圈故障、外圈故障、滾動體故障）下的軸承振動數(shù)據(jù)，經(jīng)小波包三層分解后得到 8 組能量特征值，作為 人工神經(jīng)網(wǎng)絡(luò)的 輸入層的輸入 ，然后根據(jù)神經(jīng)網(wǎng)絡(luò)的原理，設(shè)置 BP 神經(jīng)網(wǎng)絡(luò)隱含層、輸出層的相關(guān)參數(shù)，設(shè)計完成神經(jīng)網(wǎng)絡(luò)的結(jié)構(gòu) 模型 。 通過一系列的訓(xùn)練、測試和仿真可以看出， 本文構(gòu)建的 BP 神經(jīng)網(wǎng)絡(luò)結(jié)合對隱含層神經(jīng)元參數(shù)的不同設(shè)置 ， 得到不同的訓(xùn)練誤差曲線，均具有良好的收斂性 ，在測試、診斷過程中，能夠根據(jù)輸入值快速、準(zhǔn)確地識別出滾動軸承的故障類型，且具有較高的正確率。 關(guān)鍵詞： 滾動軸承； 故障診斷 ； BP 神經(jīng)網(wǎng)絡(luò) ； 能量特征值 蘭州交通大學(xué)畢業(yè)設(shè)計（論文） II Abstract The rolling bearing is one of the most mon and widely used ponents in the mechanical equipment. Its operating state has a direct impact on the entire working status of equipment and the production process. Therefore, the monitoring and diagnosis of the rolling bearing has a very important significance. The bearing fault diagnosis technology is often based on timefrequency analysis. These methods are restricted in many ways, which causes a lot of state detecting missed. This paper is based on the research of the rolling bearing fault diagnosis of the mechanical equipment, and focus on the BP neural work technology application in the problem. The rolling bearing vibration data of three fault patterns (innerrace fault, outrace fault and rolling element fault) are chosen in this paper, and it is adopted that taking eight energy ponents deposed by wavelet packet as the ANN (artificial neural work) input vector. Then, according to the ANN theory, set hidden layer and output layer parameters of the BP neural work and design the structure of the neural work model for rolling bearing fault diagnosis. At last, train the work on Matlab and get the training error curve, then test and simulate the work and calculate the correct rate of the simulation results. Through a series of training, testing and simulation process, it can be seen that the BP neural work method, which is applied to the rolling bearing fault diagnosis, can get different training error curves, bined with different set of parameters of the neurons in the hidden layer. All the curves have good convergence. In the test and diagnostic procedures, the work can identify different fault patterns quickly and accurately depending on the input data, at the same time it has a higher accuracy rate. The BP neural work method is more accurate, practical and it has a higher diagnostic accuracy rate pared with ordinary methods. So it surely has broad application prospects. Key Words： Rolling bear, Fault diagnosis, BP neural work, Energy ponents 蘭州交通大學(xué)畢 業(yè)設(shè)計（論文） III 目 錄 摘 要 ..................................................................................................................................... I Abstract ...................................................................................................................................... II 目 錄 ...................................................................................................................................III 1 緒論 ........................................................................................................................................ 1 論文背景與意義 ......................................................................................................... 1 論文研究現(xiàn)狀 ............................................................................................................. 1 論文的研究內(nèi)容與目標(biāo) ............................................................................................. 1 2 滾動軸承故障特征 ................................................................................................................ 2 滾動軸承的基本結(jié)構(gòu) ................................................................................................. 2 滾動軸承的失效形式和故障類型 ............................................................................. 2 3 BP 神經(jīng)網(wǎng)絡(luò) .......................................................................................................................... 4 人工神經(jīng)網(wǎng)絡(luò)概述 ...............................................