本文關(guān)鍵詞：一種用于磨粒識(shí)別的基于改進(jìn)PSO算法的支持向量機(jī)模型　出處：《潤滑與密封》2016年02期 　論文類型：期刊論文

  更多相關(guān)文章： 油液檢測(cè) 磨粒識(shí)別 粒子群優(yōu)化算法 支持向量機(jī)

【摘要】：為提高磨粒智能識(shí)別的準(zhǔn)確率,以傳統(tǒng)支持向量機(jī)和粒子群優(yōu)化(PSO)算法為基礎(chǔ),提出一種基于改進(jìn)PSO算法的支持向量機(jī)(SVM)識(shí)別模型。該識(shí)別模型的懲罰參數(shù)和核函數(shù)參數(shù)可同時(shí)得到最佳優(yōu)化,從而可建立模型參數(shù)最優(yōu)的自適應(yīng)SVM識(shí)別模型。采用該識(shí)別模型對(duì)油液中的磨粒進(jìn)行智能識(shí)別,結(jié)果表明該模型識(shí)別準(zhǔn)確率高達(dá)98%,明顯優(yōu)于BP神經(jīng)網(wǎng)絡(luò)模型。
[Abstract]:In order to improve the accuracy of intelligent wear particle recognition, it is based on the traditional support vector machine and particle swarm optimization (PSO) algorithm. A support vector machine (SVM) recognition model based on improved PSO algorithm is proposed. The penalty parameters and kernel function parameters of the recognition model can be optimized at the same time. Thus an adaptive SVM recognition model with optimal model parameters can be established. The recognition model is used to identify the wear particles in oil. The results show that the recognition accuracy of the model is as high as 98%. It is obviously superior to BP neural network model.
【作者單位】： 軍械工程學(xué)院七系;武漢軍械士官學(xué)校四系;軍械工程學(xué)院軍械技術(shù)研究所;
【基金】：國家自然科學(xué)基金項(xiàng)目(51205405;51305454)
【分類號(hào)】：TH117;TP18
【正文快照】： j縥縥縥縥，

本文編號(hào)：1418976