Page 137 - 2025年第56卷第11期

Regulation mechanism of automatic flushing valve on emitter clogging characteristics during
drip irrigation with Yellow River water

1 1 1 2 1 1

GAO Hao ，MO Yan ，LI Zaiyu ，LI Wenjun ，ZHANG Yanqun ，LI Hao

（1. State Key Laboratory of Water Cycle and Water Security in River Basin，China Institute of Water Resources and Hydropower Research，

Beijing 100038，China；2. Law School of Shihezi University，Shihezi University，Shihezi 832000，China）

Abstract：Automatic flushing valves （AFVs）periodically discharge sediment from driplines，enhancing clogging
resistance in Yellow River drip irrigation systems. This study evaluated three AFV specifications：ND-15 （FD = 15

s），FA-40 （FD = 40 s），and FA-80 （FD = 80 s），against a no-valve control （CK）over 400 h under 1.0 g/L sedi‐
ment concentration. We investigated AFV hydraulic performance and its impact on average relative emitter flow

.
（Dra） FA-40 and FA-80 showed minimal performance changes （-3.3% to 5.2%），while ND-15 varied by 33.3%

-66.6%，indicating poor stability. Dra fell to 75% after 40 h （ND-15），70 h （FA-40），210 h （FA-80），and 30 h

（CK）；at 170 h and 270 h，Dra for ND-15 and CK dropped to zero. During a single flush，FA-40 and FA-80 trans‐

ported sediment distances equal to 72.9% and 160% of the 48 m tape length，respectively，and discharged on aver‐

age 95.0% of input sediment，raising Dra by 559.5%-607.1% over CK；clogging was most severe in front > middle >

s
rear. ND-15’ short FD yielded low discharge efficiency （21.2%），reducing Dra by 11.3% versus CK. Although sedi‐

ment flushing efficiency （R ）first increased and then decreased with FD，FA-80 offers the best balance of anti-

sf
clogging and water use for field irrigation. For the Zuncun irrigation district in Shanxi Province with 1.0 g/L sediment，
FA-80 maintained Dra ≥ 75% during 3-4 years. These findings support AFV deployment in sediment-rich drip irriga‐
tion.

Keywords：automatic flushing valve；flushing duration；average relative flow；sediment discharge；flushing effi‐
ciency
（责任编辑：王 婧）
（上接第 1516 页）
Distributed hybrid flood forecasting model coupling physical mechanisms and deep learning
1，2 1，2 1，2 3 4 4

HE Miao ，JIANG Shanhu ，REN Liliang ，YANG Bang ，WANG Jianping ，ZHOU Le

（1. National Key Laboratory of Water Disaster Prevention，Hohai University，Nanjing 210098，China；

2. College of Hydrology and Water Resources，Hohai University，Nanjing 210098，China；

3. Hydrology Bureau，Haihe Water Conservancy Commission，Ministry of Water Resources，Tianjin 300170，China；

4. Nanjing Nanrui Water Conservancy and Hydropower Technology Co.，LTD.，Nanjing 211106，China）

Abstract：Accurate flood forecasting plays a crucial role in flood prevention and mitigation，optimal allocation of

water resources，and protection of water environments. To address the limitations of process-driven models regarding

physical mechanism representation and the lack of interpretability in data-driven models for flood forecasting，this

study proposes a Distributed Hybrid Flood Modeling （DHFM）framework that effectively integrates physical mecha‐

nisms with deep learning methodologies. By integrating physical constraints with neural networks on a unified deep
learning platform，the framework enables collaborative encoding of hydrological processes and synchronous optimiza‐

tion of physical parameters and network weights via backpropagation. Building on this foundation，the DHFM frame‐
work seamlessly incorporates differentiable Diffusion Wave （DW）and Convolutional Neural Network （CNN）-based

river routing methods，while integrating the differentiable Muskingum （MK） method as a comparative baseline.

Taking the Mishui River Basin in Hunan Province as a typical study area，we systematically evaluate the performance

and interpretability of the three differentiable river routing methods within the DHFM framework under both gauged
and ungauged scenarios. The results indicate that the DHFM framework demonstrates robust capabilities in simulating
daily streamflow and flood events across all scenarios. With other model structures and input conditions held con‐

stant，the differentiable CNN-based river routing method slightly outperforms the differentiable DW method，and
both outperform the differentiable MK method. Furthermore，the neural networks embedded within the DHFM frame‐

work can capture complex mapping relationships between static river attributes and physical parameters，with the dif‐
ferentiable CNN-based river routing method also showing potential for adaptively learning unit hydrographs based on
static river attributes. Beyond improving simulation accuracy，the DHFM framework introduces a novel paradigm for

integrating physical mechanisms and data-driven modeling approaches in flood research.

Keywords：hybrid model；deep learning；differentiable modeling；flood forecasting

（责任编辑：耿庆斋）
— 1528 —