基于神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)的高強(qiáng)高韌鉆桿用無(wú)縫鋼管的開(kāi)發(fā)

Neural Network Prediction-based R&D of High Strength/High Toughness Seamless Steel Drill Pipe

摘要: 利用BP算法和試驗(yàn)數(shù)據(jù)，構(gòu)建了神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)模型；用該神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)模型研究了各種主要化學(xué)成分對(duì)材料強(qiáng)度和沖擊韌性的影響；通過(guò)神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)，設(shè)計(jì)開(kāi)發(fā)出了一種高強(qiáng)高韌鉆桿用無(wú)縫鋼管。分析結(jié)果顯示：該高強(qiáng)高韌鉆桿用無(wú)縫鋼管在屈服強(qiáng)度達(dá)到1 100 MPa以上的同時(shí)，-20℃縱向沖擊功達(dá)到了100 J以上；該神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)模型的數(shù)據(jù)擬合度好，預(yù)測(cè)精度高。

Abstract: The neural network prediction model is set up with the BP algorithm and relevant test data.And this model is used to study the influences by various major chemical elements on strength and impact toughness of the material.Thanks to the neural network prediction technique, a type of high strength/high toughness seamless steel drill pipe is designed and developed.The result from relevant analysis shows that the said drill pipe has its yield strength above 1 100 MPa, and its longitudinal impact energy at -20℃ above 100 J.The neural network prediction model has satisfactory data fitness and high predictive accuracy.