Control of the risk of dyke failure caused by contact erosion. geophyconsult

Transcription

1 Control of the risk of dyke failure caused by contact erosion Rémi Beguin 1, J.-J. Fry 2, C. Picault 3, J.-R. Courivaud 2, Y.-H. Faure 4, P. Philippe 5 1, Bourget-du-Lac, France 2 edf-cih, Bourget-du-Lac, France 3 Compagnie Nationale du Rhône, Lyon, France 4 LTHE - Grenoble University, France 5 IRSTEA Aix-en-Provence, France

2 Outline 2 1. Contact erosion problematic 2. First large scale test program ( ) : Main results and conclusions 3. Second large scale test program (2012) : Results of the first experiment 4. Conclusions and perspectives

3 Outline 3 1. Contact erosion problematic 2. First large scale test program ( ) : Main results and conclusions 3. Second large scale test program (2012) : Results of the first experiment 4. Conclusions and perspectives

4 1 Context (Schaefer et al., 2011) 4 River or canal Sink-hole Core : silt or clay Fine particles Foundation : gravel with sand Fine soil layer Contact erosion? Coarse soil layer Pipe initiation

5 Outline 5 1. Contact erosion problematic 2. First large scale test program ( ) : Main results and conclusions 3. Second large scale test program (2012) : Results of the first experiment 4. Conclusions and perspectives

6 2.1 Objectives Objectives o Study the scale effect on the contact erosion process o What will be the consequences of the development of contact erosion on the structure stability? o What are the best methods to detect contact erosion? 6 8m

7 2.1 Objectives Objectives o Study the scale effect on the contact erosion process o What will be the consequences of the development of contact erosion on the structure stability? o What are the best methods to detect contact erosion? 7 8m

8 2.2 First campaign scheme Scheme and instrumentation 8 o Base layer : Gravel (20/40mm, Cu 1 =1,5) o Core : Silty-Sand (d 50 = 0,19 mm, Cu 1 =25) 1 Cu : Coefficient of uniformity : d 60 /d 10

9 2.3 Results o Pipe formation and propagation for 4 tests out of 5 conducted with this geometry 9 Test n 9 Test n 4 Test n 6

10 Outline Contact erosion problematic 2. First large scale test program ( ) : Main results and conclusions 3. Second large scale test program (2012) : Results of the first experiment 4. Conclusions and perspectives

11 3.1 Objectives 11 Objectives : Experiment shell geometry and material grading more similar to Rhône and Rhine dike sections Can the erosion process be control by the gravelly shell? Understand the shell failure mechanism

12 3.2 Modifications 12 Reinforced shell : 1,3m thick and made of a mix of sand and gravels

13 3.2 Modifications 13 Reinforced shell : 1,3m thick and made of a mix of sand and gravels Reinforced monitoring : 30 new interstitial pressure data logger

14 3.3 Results : Main parameters Upstream level (m) - Concentration (g/l) Flow-rate (l/s) Evolution of the main parameters during the test Upstream level (m) Flow-rate (l/s) Concentration (g/l) Time (h) : Artificial burst of turbidity due to the manual collect of the deposited sediment in the downstream reservoir

15 3.3 Results : Main parameters Upstream level (m) - Concentration (g/l) Flow-rate (l/s) Evolution of the main parameters during the test Upstream level (m) Flow-rate (l/s) Concentration (g/l) Time (h) : Artificial burst of turbidity due to the manual collect of the deposited sediment in the downstream reservoir

16 3.4 Results : Interstitial pressure Hydraulic head (m) Q23 Q24 Q25 Q26 Q27 Q28 Upstream level Hydraulic head in the foundation Time (h)

17 3.4 Results : Interstitial pressure Hydraulic head (m) Q23 Q24 Q25 Q26 Q27 Q28 Upstream level Hydraulic head in the foundation Time (h) Q26

18 3.5 Results : Deformation X (m) Deformation along Y axis (µε) 18 2 T = 28,9h Upstream tank Y(m) Downstream tank Top view

19 3.6 FO deformation measurement T = 28,9h T = 36,1h T = 50,6h 19 T = 58,5h T = 62,7h T = 64,0h T = 70,4h T = 74,6h

20 3.7 Results 20 Signs of a pipe formation and propagation : o Leakage flow-rate increase o Interstitial pressure increase o Total eroded estimated mass ~ 1000 kg o High deformation of the core detected Hopefully: o o After 20 hrs, decrease of the leakage flow-rate and the pore pressure Finally, the shell remained stable and the experiment was stopped after 95 h at the maximum hydraulic head

21 End of the test 21 After shell removal During core excavation

22 Outline Contact erosion problematic 2. First large scale test program ( ) : Main results and conclusions 3. Second large scale test program (2012) : Results of the first experiment 4. Conclusions and perspectives

23 Conclusions The 2010/11 campaign demonstrates the possibility of a pipe formation due to contact erosion initiation. This pipe formation lead to the failure of the shell in a few hours for 4 tests out of 5. A new test during the 2012 campaign concluded that the use of a thicker shell made of a widely graded soil enables to control the pipe progression, at least for some days, and preserve the structure stability. 23

24 Perspectives 24 Two new large scale tests are planned in 2012 and will give more data to confirm, or not, the results of this first test. We expect to determine rules for the dike stability based on: the global uplift stability of the shell (after clogging), the local stability of shell particles when the shell is submitted to a large leakage.

25 25 thanks you for your attention and is looking forward to hearing your questions

26 Contacts Large-scale contact erosion tests : Rémi Beguin (remi.beguin@.com) 26 Headquarters and Fibre Optics Surveillance department administration@.com Bât. Passerelle 6 30, allée du Lac d Aiguebelette B.P Le Bourget-du-Lac cedex France Tel. VoIP : Fax : Erosion Tests Laboratory Patrick Pinettes (pinettes@.com) Sup Agro SAS - 2, place P. Viala Montpellier cedex 2 France Mob. : Tel. : Fax :

27 27

28 1.2 Small scale tests Sample scale contact erosion tests 28 Observation of pipe initiation at the granular interface Thursday, October 04, 2012

29 2.2 First campaign scheme 29 Silty-sand layer d 50 = 0,19 mm Gravel layer D 50 = 28,3 mm Titre de la présentation jeudi 4 octobre 2012

30 30 Shell grading curve D50~12mm Cu~30 Titre de la présentation Thursday, October 04, 2012

31 H (%) Sol testé Limite stabilité Kenney and Lau Limite stabilité Kezdi Instable F (%)