( A2) Proceedings of 9 th Conference on Current Researches in Geotechnical Engineering, Shihman Reservoir, Tai-Yuan, Taiwan, R.O.C. August 3-3 and September, 2 RIDO Construction Information System for Deep Excavation Hsih-Hsun Hung, S. Y. Chi, J. C. Chern Geotechnical Engineering Research Center, Sinotech Engineering Consultants, INC. ABSTRACT In this paper, the construction information system for deep excavation was developed, the lateral displacement of the retaining wall was taken into account to obtain the optimized soil parameters via back-analysis. According to the optimized soil parameters, we can predict the response of retaining system for next construction stage by RIDO program. In addition, we establish the database of the wall deflection of case histories. Through the learning and analysis processes of artificial neural network, we use the network to predict the lateral displacement of the retaining wall in the coming excavation, too. Furthermore, the stress of the retaining wall and the settlement of the ground can be predicted by theoretical calculation and empirical formula. With this prediction, the safety of the buildings adjacent to the construction site can be assessed. In the end of this paper, the case history analysis of a deep excavation site located at Taipei City was performed. From this analysis, it will prove that good agreements between the results of prediction and field measurement are obtained. Keywords: deep excavation, back-analysis, Artificial Neural Network, safety assessment. (Artificial Neural Network, ANN) [] A2-
A2-2 2. 2 Visual Basic 6. RIDO RIDO RIDO RIDO ANN C Basic (RIDORIDO ANN) 2 2.2 3 3 2.2. RIDO RIDO RIDO (ELASTOPLASTIC EQUILIBRIUM) ( ) RIDODOS RIDO 4 RIDO
(2) (Analysis) ANN 5 4 RIDO 2.2.2 RIDO RIDO [2] RIDO 5 ANN [3] RIDO RIDO 2.2.4 RIDO 2.2.3 ANN [4,5] [6,7,8,9] RIDO (Back-Propagation Neural Network, BPN) 6 () (Learning) (Node) (Link) A2-3
RIDO RIDO ANN RIDO RIDO ANN 6 2.2.5 RIDO 8 EXCEL (.xls) EXCEL EXCEL 7 3. 54m49m 9.8m 2.m 4.3m 7m 9.8m 2m 9 N (m) (m) (t/m 3 ) C (t/m 2 ) () C (t/m 2 ) () 7 2.3 6 6 CL 2.88.2 2.3.7 4.6.4 26.2 7 SM 4.85 26.9 2 3 CL 2.85.8 3.8.3 25.5 2 SM 6.89 28.5 26 5 CL 4.89 3.3 8 26 5M GM 5 44.5 A2-4
圖 9 案例基地監測儀器配置圖 第階 第2階 而 RIDO 土層輸入參數中 水平地盤反力係數 Kh (t/m3) 於 黏 土 採 用 Kh=2Su 於 砂 土 則 採 用 Kh=N [3] 詳細的參數值則如表 2 表 2 RIDO 土層輸入參數 深度 γ (m) (t/m3) 5.88.527 4.2.74.5 -.5 348.88.58 8.6 2.75.5 -.5 466 6.88.58 3.44 4.3.5 -.5 86 7.85.23.548 4 6.5 26.9 2.85 8.2 5.83 2.89 Ka Ko Kp σv' (t/m2).59.3.523 4.35 8.65 Su φ (t/m2) ( Da Db Kh (t/m3) ) 第三階開挖 第階 第2階 第3階 4.5 -.5 65 28.5 6 Ko=-SINφ(Sand) Su/σv'=.32 (NC) δ=2/3φ Kh=N(Sand) Ko=.95-SINφ(Clay) Su=3-.3σv' (OC) Kh=2Su(Clay) 第四階開挖 第階 第2階 第3階 第4階 針對開挖基地選取兩個斷面來作分析 分別為 S 及 S7 如圖 9 所標示 因第 階開挖尚無監測資料 可作反算分析 所以將針對第 2~4 階的開挖來作分析 斷面 S.預測壁體位移 RIDO 反算分析 將前幾階壁體位移監測資料加入反算 以預測下 一階之開挖結果 例如第二階開挖 需以第一階監測 資料加入反算 以預測第二階壁體位移 各階分析結 果如下 由上顯示 當監測資料階數越多時 則預測下一 階的壁體位移值越準確 ANN 預測分析 第二階開挖 為預測下一階之開挖結果 需將前三階壁體位移 監測資料加入分析 例如第二階開挖 前三階監測資 A2-5
2 2..54.75.46.8.4.5.3 +.56 +.56 -.7 -.4 -.45 -.37 -.44 -.38 -.5cm~+.5cm 2 3 3. 5 6 3 35 37 2 5 4. 2 2 3 4 3 4 ANN RIDO S S7 2. SM3 SM4 SM5 SM6 SM7 SM8 SM9 SM2.26.44.3.33.3.2.24.9.8.5.3.2.3.6.2. -.8 -.39 -.27 -.3 -.3 -.9 -.23 -.9. RIDO 2 SM3 SM4 SM5 SM6 SM7 SM8 SM9 SM2.86.5.4.36.32.2.24.9.74.46.26.2.2.6.25.9 -.2 -.4 -.4 -.6 -.2 -.4 -.2 -.8 SM3 SM4 SM5 SM6 SM7 SM8 SM9 SM2.44.98.82.87.63.4.45.38 2 3 A2-6
S7 S 2 3 4 ANN 2 2. R7 SM44 SM67 SM68 SM69 SM7 R2 SM4.33.26.28.8.4.34.6.3.7.5.2. +. -. -.5 -. +.5 +.2 +. R7 SM44 SM67 SM68 SM69 SM7 R2 SM4..53.48.34.32.48.5.8.63.23..3. +.5 +.27 +.5 -. -.22 +. R7 SM44 SM67 SM68 SM69 SM7 R2 SM4 3.7 3.6 2.66 2.38 2.27.33.5.36 2.57 2.6.4..39.7.5.2 -.6 -. -.26 -.28 -.88 -.6 -.45 -.34 3. 2 8 3 55 4 24 26 2 3 46 5 48 4. 2 2 3 2 3 4 3 4 A2-7
S7 (CEDM)(998) Form.HevPic(p). 2. 3... 347- (986) 2. Nelder, J. A. and R. Mead A simplex method for function minimization. Computer Journal, Vol. 7, pp. 38-33 (965). 3. RIDO 75 62-76 (999) 4. Hsieh, P. G. and C. Y. Ou, Shape of ground surface settlement profiles caused by excavation, Canadian Geotechnical Journal, Vol. 35, No. 5, pp. -4 (998). 5. Ou, C. Y., P. G. Hsieh and D. C. Chiou, Characteristic of ground surface settlement during excavation, Canadian Geotechnical Journal, Vol. 3, No. 5, pp. 758-767 (993). 6. Boone, S. J., Ground-movement-related building damage. Journal of Geotechnical Engineering, ASCE, Vol. 22, No., pp. 886-896 (996). 7. Boscardin, M. D. and E. J. Cording, Building response to excavation-induced settlement. Journal of Geotechnical Engineering, ASCE, Vol. 5, No., pp. -2 (989). 8. Burland, J. B. and C. P. Wroth, Settlement of buildings and associated damage. Conference on Settlement of structures, Cambridge, Pentech Press, pp. 6-654 (975). 9. A2-8