Historically,landslides have been the primary type of geological disaster worldwide.Generally,the stability of reservoir banks is primarily affected by rainfall and reservoir water level fluctuations.Moreover,the stab...Historically,landslides have been the primary type of geological disaster worldwide.Generally,the stability of reservoir banks is primarily affected by rainfall and reservoir water level fluctuations.Moreover,the stability of reservoir banks changes with the long-term dynamics of external disastercausing factors.Thus,assessing the time-varying reliability of reservoir landslides remains a challenge.In this paper,a machine learning(ML)based approach is proposed to analyze the long-term reliability of reservoir bank landslides in spatially variable soils through time series prediction.This study systematically investigated the prediction performances of three ML algorithms,i.e.multilayer perceptron(MLP),convolutional neural network(CNN),and long short-term memory(LSTM).Additionally,the effects of the data quantity and data ratio on the predictive power of deep learning models are considered.The results show that all three ML models can accurately depict the changes in the time-varying failure probability of reservoir landslides.The CNN model outperforms both the MLP and LSTM models in predicting the failure probability.Furthermore,selecting the right data ratio can improve the prediction accuracy of the failure probability obtained by ML models.展开更多
The instability of soil bank slopes induced by freeze-thaw cycles at the northern foot of Tianshan Mountain is very common.The failure not only caused a large amount of soil erosion,but also led to serious reservoir s...The instability of soil bank slopes induced by freeze-thaw cycles at the northern foot of Tianshan Mountain is very common.The failure not only caused a large amount of soil erosion,but also led to serious reservoir sedimentation and water quality degradation,which exerted a lot of adverse effects on agricultural production in the local irrigation areas.Based on field investigations on dozens of irrigation reservoirs there,laboratory tests were carried out to quantitatively analyze the freeze-thaw effect on the soil engineering characteristics to reveal the facilitation on the bank slope instability.The results show that the softening characteristics of the stressstrain curves gradually weaken,the effective cohesions decline exponentially,the seepage coefficients enlarge,and the thermal conductivities decrease after 7 freeze-thaw cycles.The freeze-thaw effect on the specimens with low confining pressures,low dry densities and high water contents is more significant.The water migration and the phase transition between water and ice result in the variations of the soil internal microstructures,which is the main factor affecting the soil engineering characteristics.Sufficient water supply and the alternation of positive and negative temperatures at the reservoir bank slopes in cold regions make the water migration and phase transition in the soil very intensely.It is easy to form a large number of pores and micro cracks in the soil freezing and thawing areas.The volume changes of the soil and the water migration are difficult to reach a dynamic balance in the open system.Long-term freeze-thaw cycles will bring out the fragmentation of the soil particles,resulting in that the micro cracks on the soil surfaces are developing continuously.The soil of the bank slopes will fall or collapse when these cracks penetrate,which often happens in winter there.展开更多
After water is impounded in a reservoir, rock mass in the hydro-fluctuation belt of the reservoir bank slope is subject to water saturation- dehydration circulation （WSDC）. To quantify the rate of change of rock mec...After water is impounded in a reservoir, rock mass in the hydro-fluctuation belt of the reservoir bank slope is subject to water saturation- dehydration circulation （WSDC）. To quantify the rate of change of rock mechanical properties, samples from the Longtan dam area were measured with uniaxial compression tests after different numbers （1, 5, 10, 15, and 20） of simulated WSDC cycles. Based on the curves derived from these tests, a modified Hock- Brown failure criterion was proposed, in which a new parameter was introduced to model the cumulative damage to rocks after WSDC. A case of an engineering application was analyzed, and the results showed that the modified Hock-Brown failure criterion is useful. Under similar WSDC-influenced engineering and geological conditions, rock mass strength parameters required for analysis and evaluation of rock slope stability can be estimated according to this modified Hoek-Brown failure criterion.展开更多
In most cases, the slope stability of reservoir bank is analyzed on the premise that the location of phreatic surface is obtained. But many designers generalize a line as the phreatie surface through their experience ...In most cases, the slope stability of reservoir bank is analyzed on the premise that the location of phreatic surface is obtained. But many designers generalize a line as the phreatie surface through their experience to analyze the stability, which is unsafe in the project. To find a solution of the phreatic surface which is convenient to put into use and in accordance with the practice, the article, based on Boussinesq equation, infers analytic solutions suitable to the water level at different ratios and achieves an analytic solution equation through fitting curves. The correctness of the equation is also proved by the experiments of sand and sand-clay models and the inaccuracy of empirical generalization is analyzed quantitatively. The calculation results show that the inaccuracy through the method of experiential generalizing is so large that the designers should be awake to it.展开更多
基金supported by the National Natural Science Foundation of China(Grant No.52308340)the Innovative Projects of Universities in Guangdong(Grant No.2022KTSCX208)Sichuan Transportation Science and Technology Project(Grant No.2018-ZL-01).
文摘Historically,landslides have been the primary type of geological disaster worldwide.Generally,the stability of reservoir banks is primarily affected by rainfall and reservoir water level fluctuations.Moreover,the stability of reservoir banks changes with the long-term dynamics of external disastercausing factors.Thus,assessing the time-varying reliability of reservoir landslides remains a challenge.In this paper,a machine learning(ML)based approach is proposed to analyze the long-term reliability of reservoir bank landslides in spatially variable soils through time series prediction.This study systematically investigated the prediction performances of three ML algorithms,i.e.multilayer perceptron(MLP),convolutional neural network(CNN),and long short-term memory(LSTM).Additionally,the effects of the data quantity and data ratio on the predictive power of deep learning models are considered.The results show that all three ML models can accurately depict the changes in the time-varying failure probability of reservoir landslides.The CNN model outperforms both the MLP and LSTM models in predicting the failure probability.Furthermore,selecting the right data ratio can improve the prediction accuracy of the failure probability obtained by ML models.
基金supported by the National Key Research and Development Program of China(Grant No.2018YFC0809605,2018YFC0809600)the Key Research Program of Frontier Sciences of Chinese Academy of Sciences(Grant No.QYZDY-SSWDQC015)+2 种基金the National Natural Science Foundation of China(Grant No.41230630)the National Science Fund for Distinguished Young Scholars(Grant No.41825015)the Key Research Program of the Chinese Academy of Sciences(Grant No.ZDRW-ZS-2020-1)。
文摘The instability of soil bank slopes induced by freeze-thaw cycles at the northern foot of Tianshan Mountain is very common.The failure not only caused a large amount of soil erosion,but also led to serious reservoir sedimentation and water quality degradation,which exerted a lot of adverse effects on agricultural production in the local irrigation areas.Based on field investigations on dozens of irrigation reservoirs there,laboratory tests were carried out to quantitatively analyze the freeze-thaw effect on the soil engineering characteristics to reveal the facilitation on the bank slope instability.The results show that the softening characteristics of the stressstrain curves gradually weaken,the effective cohesions decline exponentially,the seepage coefficients enlarge,and the thermal conductivities decrease after 7 freeze-thaw cycles.The freeze-thaw effect on the specimens with low confining pressures,low dry densities and high water contents is more significant.The water migration and the phase transition between water and ice result in the variations of the soil internal microstructures,which is the main factor affecting the soil engineering characteristics.Sufficient water supply and the alternation of positive and negative temperatures at the reservoir bank slopes in cold regions make the water migration and phase transition in the soil very intensely.It is easy to form a large number of pores and micro cracks in the soil freezing and thawing areas.The volume changes of the soil and the water migration are difficult to reach a dynamic balance in the open system.Long-term freeze-thaw cycles will bring out the fragmentation of the soil particles,resulting in that the micro cracks on the soil surfaces are developing continuously.The soil of the bank slopes will fall or collapse when these cracks penetrate,which often happens in winter there.
基金supported by the National Natural Science Foundation of China under No. 41630639the National Basic Research Program of China (2014CB744703)+1 种基金Natural Science Basic Research Plan in Shaanxi Province of China (No. 2016JQ4014)China Postdoctoral Science Foundation (2016M602743)
文摘After water is impounded in a reservoir, rock mass in the hydro-fluctuation belt of the reservoir bank slope is subject to water saturation- dehydration circulation （WSDC）. To quantify the rate of change of rock mechanical properties, samples from the Longtan dam area were measured with uniaxial compression tests after different numbers （1, 5, 10, 15, and 20） of simulated WSDC cycles. Based on the curves derived from these tests, a modified Hock- Brown failure criterion was proposed, in which a new parameter was introduced to model the cumulative damage to rocks after WSDC. A case of an engineering application was analyzed, and the results showed that the modified Hock-Brown failure criterion is useful. Under similar WSDC-influenced engineering and geological conditions, rock mass strength parameters required for analysis and evaluation of rock slope stability can be estimated according to this modified Hoek-Brown failure criterion.
文摘In most cases, the slope stability of reservoir bank is analyzed on the premise that the location of phreatic surface is obtained. But many designers generalize a line as the phreatie surface through their experience to analyze the stability, which is unsafe in the project. To find a solution of the phreatic surface which is convenient to put into use and in accordance with the practice, the article, based on Boussinesq equation, infers analytic solutions suitable to the water level at different ratios and achieves an analytic solution equation through fitting curves. The correctness of the equation is also proved by the experiments of sand and sand-clay models and the inaccuracy of empirical generalization is analyzed quantitatively. The calculation results show that the inaccuracy through the method of experiential generalizing is so large that the designers should be awake to it.
