【摘要】：針對目前數(shù)控機床故障復雜、診斷困難的問題,提出基于人工神經網絡的故障診斷方法。在研究傳統(tǒng)BP神經網絡故障診斷模型基礎上,引入改進的BP算法-LM算法,建立機床主軸系統(tǒng)LM-BP神經網絡故障診斷模型,對機床主軸系統(tǒng)故障進行分析與診斷,再通過Matlab仿真與傳統(tǒng)BP神經網絡相對比,仿真結果表明:傳統(tǒng)BP神經網絡存在較難實現(xiàn)快速、準確的故障定位問題,而BP神經網絡LM算法作為故障診斷的核心算法收斂速度快、識別準確。該方案設計合理可行,有較好的應用前景,并給出應用了實例。
[Abstract]:A fault diagnosis method based on artificial neural network (Ann) is proposed to solve the problem of complex fault and difficult diagnosis of CNC machine tool. On the basis of studying the traditional BP neural network fault diagnosis model, an improved BP algorithm-LM algorithm is introduced to establish the LM-BP neural network fault diagnosis model of machine tool spindle system, and to analyze and diagnose the machine tool spindle system fault. By comparing Matlab simulation with traditional BP neural network, the simulation results show that the traditional BP neural network is difficult to realize fast and accurate fault location problem, and LM algorithm of BP neural network, as the core algorithm of fault diagnosis, converges quickly. The identification is accurate. The design of this scheme is reasonable and feasible, and it has a good application prospect, and an example is given.
【作者單位】： 四川理工學院自動化與電子信息學院;
【基金】：四川理工學院學科建設項目(2014JC02) 人工智能四川省重點實驗室重點項目(2012RZY22) 四川理工學院學科特色培育項目(2013PMG04)
【分類號】：TP183;TG659

【參考文獻】

相關博士學位論文 前1條

1 任錕;高速數(shù)控加工的前瞻控制理論及關鍵技術研究[D];浙江大學;2008年

【共引文獻】

相關博士學位論文 前10條

1 沈洪W，

本文編號：2273295