Page 76 - 2022年第53卷第3期

Mitigation of the lake hydraulic regulation on the accompanied flood risk
in an Interconnected River System Network

ZHANG Chen，ZHENG Yunhe，LIU Yinzhu，YU Ruolan，GAO Xueping
（State Key Laboratory of Hydraulic Engineering Simulation and Safety，Tianjin University，Tianjin 300350，China）

Abstract：The Interconnected River System Network （IRSN） is an important and complex project for water
resources management，which is prone to be accompanied by a series of risks. It is difficult to quantitative⁃
ly assess the impact of lake hydraulic regulation on the risk of river-lake connectivity. To address this is⁃
sue， we investigated the quantitative impact of Lake Luoma hydraulic regulation on the accompanied flood
risk in the lower reaches of Yishusi IRSN under different flood return periods， by coupling Vine-Copula
function and hydrodynamic model. The results show that the medium risks are exhibited for the flow，water
level，and velocity in the Middle Canal，respectively，during a flood with a 50-year return period. Without
Lake Luoma hydraulic regulation， more than 62% channel sections of the downstream Xinyi River are at
3
medium or high risks when the inflow reaches 11，046 m ·/s. While after regulation （i.e.， decrease in the
original regulating water level of Lake Luoma from 23.5 m to 22.5 m）， 80% of the downstream river
3
cross-sections could still be at low risks even if the inflow reaches 12，672.6 m ·/s. Consequently， about
42% of the cross-sections at medium or high risks shifts to the low risks. However， the regulation effect
of Lake Luoma will fail under a 100-year flood return period throughout the Yishusi basin. It is found that
lake hydraulic regulation can effectively improve the risk safety threshold of the water depth and flow veloci⁃
ty in the lower reaches of Yishusi IRSN and has a mitigative effect on the accompanied flood risk in IRSN.
Keywords： Interconnected River System Network； hydraulic regulation； flood risk； risk mitigation； risk
analysis model；Yishusi IRSN
（责任编辑：韩 昆）
（上接第 315 页）
Research on ensemble surrogate models
of dam seepage parameters inversion under Bayesian framework

1 1 1 1 1 2
YU Hongling ，WANG Xiaoling ，WANG Cheng ，ZENG Tuocheng ，YU Jia ，GE Shicong
（1. State Key Laboratory of Hydraulic Engineering Simulation and Safety，Tianjin University，Tianjin 300072，China；
2. Tianjin Municipal Engineering Design and Research Institute，Tianjin 300051，China）

Abstract： The key to Bayesian inversion of seepage parameters is to solve the problem of time-consuming
calculations caused by the large number of calls to seepage forward models. Most of the existing researches
on improving the computational efficiency of Bayesian inversion adopt surrogate models based on a single
machine learning algorithm，which has the problem of low computational accuracy. In response to the above
problems， an ensemble surrogate model is proposed for the inversion of dam seepage parameters under the
Bayesian framework in this paper. This method integrates Support Vector Regression （SVR）， Kriging and
Multivariate Adaptive Regression Splines （MARS） machine learning algorithms under the Bayesian frame⁃
work. Among them， the advantage of parallel sampling in the Differential Evolution Adaptive Metropolis
（DREAM ZS） algorithm is used to calculate the random distribution function of the weight coefficients， and
the model weight coefficients are obtained under the consideration of uncertainty. Case analysis shows that
compared to the seepage numerical model that takes at least 4 hours to run once， the ensemble surrogate
model proposed in this paper takes only a few seconds，which significantly improves the computational effi⁃
ciency of Bayesian inversion. In addition， compared with Bayesian inversion methods based on SVR， Krig⁃
ing， and MARS， the Bayesian inversion method based on the ensemble surrogate model proposed in this
paper can obtain more accurate inversion results，and its average accuracy has been increased by 13.78%，
19.34%，and 12.27% respectively，which provides a new idea for the inversion of dam seepage parameters.
Keywords： seepage parameter； Bayesian inversion； ensemble surrogate model； DREAM ZS algorithm； SVR；
Kriging；MARS
（责任编辑：李福田）
— 324 —