動(dòng)態(tài)自適應(yīng)遺傳算法在樁錨支護(hù)結(jié)構(gòu)優(yōu)化設(shè)計(jì)中的應(yīng)用*
丁敏1，張永興1,2，王輝3
- 摘 要
-
丁敏1，張永興1,2，王輝3（1 重慶大學(xué)土木工程學(xué)院，重慶 400045；2 重慶大學(xué)山地城鎮(zhèn)建設(shè)與新技術(shù)教育部重點(diǎn)實(shí)驗(yàn)室，重慶400030；3 重慶高速公路集團(tuán)有限公司，重慶 401121）［摘要］在保證基坑穩(wěn)定性的條件下，為了提高樁錨支護(hù)結(jié)構(gòu)經(jīng)濟(jì)性，建立了樁錨支護(hù)結(jié)構(gòu)的優(yōu)化設(shè)計(jì)模型，并對(duì)其支撐道數(shù)、樁徑、嵌固深度和混凝土強(qiáng)度等級(jí)等設(shè)計(jì)參數(shù)做了敏感性分析，在此基礎(chǔ)上確定各個(gè)優(yōu)化設(shè)計(jì)參數(shù)的合理取值范圍，為樁錨支護(hù)優(yōu)化提供了約束條件。針對(duì)傳統(tǒng)的遺傳算法在迭代過程中出現(xiàn)過早地收斂到局部最優(yōu)解和在最優(yōu)值附近收斂速度慢等缺點(diǎn)，提出了動(dòng)態(tài)自適應(yīng)技術(shù)改進(jìn)遺傳算法的新算法（動(dòng)態(tài)自適應(yīng)遺傳算法，DAGA）。工程實(shí)例分析表明，動(dòng)態(tài)自適應(yīng)遺傳算法不受設(shè)計(jì)空間可微性、連續(xù)性等限制，適合于求解具有離散設(shè)計(jì)變量的樁錨支護(hù)工程；不僅有較好的收斂性，而且在求解時(shí)間上具有一定的優(yōu)勢(shì)。［關(guān)鍵詞］動(dòng)態(tài)自適應(yīng)遺傳算法；樁錨支護(hù)；優(yōu)化設(shè)計(jì)；敏感性分析；動(dòng)態(tài)自適應(yīng)技術(shù)；自適應(yīng)懲罰函數(shù)中圖分類號(hào)：TU473 文獻(xiàn)標(biāo)識(shí)碼：A 文章編號(hào)：1002-848X(2013)02-0083-05Application of dynamic adaptive genetic algorithm in the optimization design of pile-anchor retaining structuresDing Min1, Zhang Yongxing1,2, Wang Hui3(1 College of Civil Engineering, Chongqing University, Chongqing 400045, China; 2 Key Laboratory of New Technology for Construction of China in Mountainous Area, Chongqing University, Chongqing 400030, China; 3 Chongqing Expressway Group Co., Ltd., Chongqing 401121, China)Abstract: In order to reduce the cost of pile-anchor retaining structures in the condition of ensuring stability of foundation pit, the optimization model was established. Sensitivity analysis of design parameters were carried out,such as layer number of bracing, pile diameter, concrete strength grade, and etc. On the basis of above analyses, the reasonable ranges of preliminary values of optimization design parameters were given, which provided constraint conditions for the optimization of the pileanchor retaining. In view of disadvantages of traditional genetic algorithm, such as early converge to local optimal solution and slow convergent speed near optimal value, and etc, a new algorithm (dynamic adaptive genetic algorithm, DAGA) was put forward, and the algorithm was established by applying dynamic adaptive technology to improve traditional genetic algorithm. Engineering example analysis shows that the dynamic adaptive genetic algorithm not only solves the problems of the differentiability and continuity of design space, but also is particularly suitable for optimization design of the pile-anchor retaining structures with discrete design variables. The dynamic adaptive genetic algorithm presented has a better convergence as well as the advantages of the solution time.Keywords: dynamic adaptive genetic algorithm (DAGA); pileanchor retaining; optimization design; sensitivity analysis; dynamic adaptive technology; adaptive penalty function*國(guó)家自然科學(xué)基金資助項(xiàng)目（510270004），“十一五”國(guó)家科技支撐計(jì)劃項(xiàng)目（2008BAJ06B04-2）。參考文獻(xiàn)［1］ 馮仲仁, 王雄江, 姚愛民, 等. 基于遺傳算法的深層攪拌樁優(yōu)化設(shè)計(jì)［J］. 巖土力學(xué), 2003, 24(3): 420-422.［2］ 劉勇鍵，夏繼君，鄧浩. 自適應(yīng)小生境遺傳算法在土釘支護(hù)結(jié)構(gòu)整體穩(wěn)定性分析中的應(yīng)用［J］. 工業(yè)建筑，2007，37（S1）：740-744.［3］ 吳恒, 周東,李陶深, 等. 深基坑樁錨支護(hù)協(xié)同演化優(yōu)化設(shè)計(jì)［J］. 巖土工程學(xué)報(bào), 2002, 24(4): 465-470.［4］ 周東. 基坑支護(hù)工程遺傳優(yōu)化設(shè)計(jì)研究［D］. 南寧: 廣西大學(xué), 2002.［5］ 秦四清,等.深基坑工程優(yōu)化設(shè)計(jì)［M］.北京: 地震出版社, 1998.［6］ 陳昌富, 吳子儒, 曹佳, 等. 水泥土墻支護(hù)結(jié)構(gòu)遺傳進(jìn)化優(yōu)化設(shè)計(jì)方法［J］. 巖土工程學(xué)報(bào), 2005, 27(2): 224-229.［7］ 楊云，馮亞. 自適應(yīng)遺傳算法在場(chǎng)強(qiáng)傳播損耗預(yù)測(cè)中的應(yīng)用［J］. 微電子學(xué)與計(jì)算機(jī)，2010，27(11)：121-127.［8］ SIMPAON A R, PRIEST S D. The application of genetic algorithms to optimization problems in geotechnics ［J］. Computers and Geotechnics, 1993, 15(1): 1-19.［9］ 田豐，邊婷婷. 基于自適應(yīng)遺傳算法的交通信號(hào)配時(shí)優(yōu)化［J］. 計(jì)算機(jī)仿真，2010，27(6)：305-308.［10］ 雷英杰，張善文，李續(xù)武，等. Matlab 遺傳算法工具箱及應(yīng)用［M］. 西安: 西安電子科技大學(xué)出版社, 2005.
- 下載地址
-
你還沒注冊(cè)？或者沒有登錄？這篇論文要求至少是本站的注冊(cè)會(huì)員才能閱讀！
如果你還沒注冊(cè)，請(qǐng)趕緊點(diǎn)此注冊(cè)吧！
如果你已經(jīng)注冊(cè)但還沒登錄，請(qǐng)趕緊點(diǎn)此登錄吧！
