本文關(guān)鍵詞： 粒子群算法 RBF神經(jīng)網(wǎng)絡(luò)數(shù) 燒結(jié)終點(diǎn)　出處：《鑄造技術(shù)》2016年11期 　論文類型：期刊論文

【摘要】：針對燒結(jié)生產(chǎn)過程中多變量、強(qiáng)耦合的特點(diǎn)和RBF神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)參數(shù)選取依據(jù)經(jīng)驗(yàn)的問題,為提高燒結(jié)終點(diǎn)預(yù)報(bào)模型的精度,提出粒子群算法優(yōu)化RBF神經(jīng)網(wǎng)絡(luò)的燒結(jié)終點(diǎn)預(yù)測方法。在標(biāo)準(zhǔn)PSO算法的基礎(chǔ)上,優(yōu)化RBF神經(jīng)網(wǎng)絡(luò)隱層節(jié)點(diǎn)中心和寬度2個(gè)結(jié)構(gòu)參數(shù),并建立燒結(jié)終點(diǎn)預(yù)測模型;在此基礎(chǔ)上利用UCI數(shù)據(jù)庫中的Computer Hardware和Concrete Slump Test標(biāo)準(zhǔn)數(shù)據(jù),驗(yàn)證了方法的有效性,并以某鋼廠265 m2燒結(jié)機(jī)的實(shí)際生產(chǎn)數(shù)據(jù),建立燒結(jié)終點(diǎn)的預(yù)報(bào)模型。結(jié)果表明,與標(biāo)準(zhǔn)BP,RBF相比,基于PSO優(yōu)化RBF的燒結(jié)終點(diǎn)預(yù)測模型精度高、泛化能力強(qiáng)。
[Abstract]:Aiming at the characteristics of multivariable and strong coupling in sintering process and the experience of selecting parameters of RBF neural network structure, the precision of the prediction model of sintering end point is improved. Particle swarm optimization (PSO) algorithm is proposed to predict the sintering endpoint of RBF neural network. Based on the standard PSO algorithm, two structural parameters of the hidden layer node center and width of RBF neural network are optimized. The prediction model of sintering end point is established. On this basis, the validity of the method is verified by using the standard data of Computer Hardware and Concrete Slump Test in UCI database. Based on the actual production data of 265 m2 sintering machine in a steel plant, a prediction model of sintering end point is established. The results show that the model is compared with the standard BPU RBF. The prediction model of sintering end point based on PSO optimization RBF has high precision and strong generalization ability.
【作者單位】： 內(nèi)蒙古科技大學(xué)機(jī)械工程學(xué)院;包鋼集團(tuán)煉鐵廠;內(nèi)蒙古科技大學(xué)材料與冶金學(xué)院;
【基金】：國家自然科學(xué)基金(21366017) 內(nèi)蒙古自然科學(xué)基金(2015MS0512) 內(nèi)蒙古高等學(xué)�？茖W(xué)研究項(xiàng)目(NJZY146) 內(nèi)蒙古科技大學(xué)創(chuàng)新基金(2015QDL12)
【分類號(hào)】：TF046.4;TP183
【正文快照】： 我國冶金行業(yè)中,燒結(jié)礦己占高爐爐料的90%以上[1],其質(zhì)量和產(chǎn)量直接關(guān)聯(lián)到煉鐵及煉鋼的產(chǎn)量和質(zhì)量指標(biāo)。燒結(jié)生產(chǎn)是一個(gè)非常復(fù)雜的物理、化學(xué)反應(yīng)過程,作為判斷燒結(jié)生產(chǎn)的重要質(zhì)量指標(biāo),燒結(jié)終點(diǎn)與燒結(jié)機(jī)產(chǎn)量、質(zhì)量和成本戚戚相關(guān);其優(yōu)化對提高燒結(jié)礦質(zhì)量、降低成本至關(guān)重要。，

本文編號(hào)：1452269