Page 80 - 水利学报2025年第56卷第4期

（2）：231-238.
［ 27］ HOLLAND J H. Adaptation in Natural and Artificial Systems：an Introductory Analysis with Applications to Biology，

.
Control，and Artificial Intelligence［M］ MIT press，1992.
.
［ 28］ XUE J，SHEN B. A novel swarm intelligence optimization approach：sparrow search algorithm［J］ Systems Science &
Control Engineering，2020，8（1）：22-34.
.
［ 29］ GAJGHATE P W， MIRAJKAR A. Irrigation pipe network planning at tertiary level： an Indian case study［J］
KSCE Journal of Civil Engineering，2020，24：322-335.
.
［ 30］ 王训俭，王增义 . 论给水管网压力监测点的选择［J］ 中国给水排水，1989，5（3）：9-12.
［ 31］ 郑书闽，颜建国，郭鹏程，等 . 基于 VMD 及深度学习的供水管道小尺度泄漏检测研究［J］ 水利学报，2024，
.
55（8）：999-1008.

Research on leakage positioning model of self-pressure irrigation pipe
network based on SSA-BP

ZHANG Hui 1，2，3，4 ，LIU Ningning 1，2，3，4 ，WANG Zhenhua 1，2，3，4 ，ZHANG Jinzhu 1，2，3，4 ，

1，2，3，4 1，4，5
LI Miao ，YIN Feihu

（1. College of Water Conservancy & Architectural Engineering，Shihezi University，Shihezi 832000，China；2. Key Laboratory of Modern

Water-Saving Irrigation of Xinjiang Production & Construction Corps，Shihezi 832000，China；3. Technology Innovation Center for

Agricultural Water and Fertilizer Efficiency Equipment of Xinjiang Production & Construction Corps，Shihezi 832000，China；

4. Laboratory of Northwest Oasis Water-saving Agriculture，Ministry of Agriculture and Rural Affairs，Shihezi 832000，China；

5. Research Institute of Farmland Water Conservancy and Soil-fertilizer，Xinjiang Academy of Agricultural

Reclamation Sciences，Shihezi 832000，China）

Abstract：Given the limitations of the existing self-pressure pipe network leakage location methods，this paper ana⁃

lyzes the impact of the spatial distribution of leakage points on the pressure changes in the pipe network under various
leakage conditions by constructing a hydraulic model of self-pressure irrigation pipe network. A self-pressure irriga⁃
tion pipe network leakage location model based on the SSA-BP neural network is proposed. The nonlinear relationship
between the leakage point position and the pressure change rate of the monitoring point is established and compared
with the traditional BP neural network and GA-BP neural network. The results show that the SSA-BP model has

higher prediction accuracy for the horizontal and vertical coordinates of the predicted leakage position，and the deter⁃
2
mination coefficients R reach 0.991 and 0.993，respectively，which are 0.90%，1.71% and 3.32%，3.12% higher

than those of the BP model and the GA-BP model，respectively. The root mean square error （RMSE）and the mean

absolute percentage error （MAPE）are 29.45 and 0.88%，and 26.76 and 0.74%，respectively，obviously lower than

those of the latter two. The prediction error is reduced dramatically，showing better generalization ability. In the

random simulation leakage location of the case pipeline network，the average prediction deviation of the SSA-BP
model under large-scale leakage conditions is only 39.93 m，which is 67.66% and 26.99% lower than that of the BP

model and the GA-BP model，respectively. The average prediction deviation of the SSA-BP model under small-scale

leakage conditions is only 66.17 m，which is 53.70% and 37.54% lower than that of the BP model and the GA-BP

model，respectively，which further proves that the SSA-BP model has higher accuracy and stability. This paper is not

only essential for studying the spatial distribution of leakage points in response to the pressure changes in the pipe net⁃

work and for selecting pressure monitoring points reasonably，but also for providing a reliable basis for the leakage
location of the self-pressure irrigation pipe network.

Keywords：self-pressure irrigation pipe network；leak location；SSA-BP neural networks；hydraulic model；pres⁃
sure change rate
（责任编辑：鲁 婧 韩 昆）
— 498 —