Page 134 - 2025年第56卷第1期
P. 134
(4): 461-473.
[ 23 ] MIRJALILI S, MIRJALILI S M, LEWIS A. Grey Wolf Optimizer[J]. Advances in Engineering Software, 2014,
69(3): 46-61.
[ 24 ] LI X, FU Q, LI Q, et al. Multi-objective binary grey wolf optimization for feature selection based on guided mu⁃
tation strategy[J]. Applied Soft Computing, 2023, 145: 11058.
[ 25 ] KOHLI M, ARORA S. Chaotic grey wolf optimization algorithm for constrained optimization problems[J]. Compu⁃
tational Design & Engineering, 2018, 5(4): 458-472.
[ 26 ] BREIMAN L. Random Forests[J]. Machine Learning, 2001, 45(1): 5-32.
[ 27 ] 吕菲, 钟登华, 余佳, 等. 迁移学习框架下高心墙堆石坝施工仿真参数 IGOA-MLP 动态预测模型[J]. 水
利学报, 2023, 54(10): 1151-1162.
[ 28 ] 王佳俊. 心墙堆石坝坝面碾压智能监控方法与应用研究[D]. 天津; 天津大学, 2019.
[ 29 ] ZADEH L A. Fuzzy sets[J]. Information & Control, 1965, 8(3): 338-353.
[ 30 ] 陈文苑, 张明, 赵伟. 不确定条件下水资源系统多目标优化配置研究[J]. 水力发电学报, 2021, 40(12):
12-24.
[ 31 ] YAGER R R. On the Dempster-Shafer framework and new combination rules[J]. Information Sciences, 1987, 41
(2): 93-137.
[ 32 ] 朱小飞, 王永君, 李大军. 模糊评价中最大隶属度原则有效性检验[J]. 测绘与空间地理信息, 2016, 39
(5): 135-137.
[ 33 ] 张剑亭, 郭生练, 陈柯兵, 等. 基于信息熵的梯级水库联合优化调度增益分配法[ J]. 水力发电学报,
2020, 39(2): 94-102.
A comprehensive evaluation method for rolling quality of earth-rock
dam based on improved evidence theory
LIN Weiwei, ZHANG Jun, WANG Xiaoling, WANG Jiajun, YU Hongling
(Tianjin University, SKL of Hydraulic Engineering Intelligent Construction and Operation, Tianjin 300350, China)
Abstract: The quality of compaction during rolling significantly influences whether dam settlement and deformation
meet requirements. Current research on evaluation models relies on limited random test pit data, leading to issues
of randomness, fuzziness, grey areas, and uncertainty. To address these challenges, this study proposes a rolling
quality evaluation and feed control method based on the ACGWO-RF algorithm and evidence theory. Initially, a⁃
daptive factor and chaos theory enhance the search capability of the GWO algorithm, while the Ntree and Mtry pa⁃
rameters of the RF algorithm, suitable for small sample datasets, are optimized based on the proposed ACGWO.
Subsequently, a rolling quality prediction model employing the ACGWO-RF algorithm is established to elucidate
the intricate nonlinear mapping relationship between rolling quality and influencing variables such as rolling parame⁃
ters, material source parameters, and meteorological parameters. Moreover, to address challenges in rolling
quality evaluation related to randomness, fuzziness, grey areas, and uncertainty, as well as mitigating the one-
sidedness and poor accuracy of a single evaluation indicator, a comprehensive evaluation method integrates the con⁃
tinuous monitoring index (CV) and prediction results (dry density and compaction degree) from the ACGWO-RF
algorithm. Fuzzy membership degree assessment is employed for the aforementioned aspects, and the D-S evidence
theory, capable of managing multiple uncertainties, is utilized for evidence fusion. Finally, a multi - level
feedback control mechanism is proposed within the rolling intelligent monitoring feedback control framework to pro⁃
vide on-site feedback control for rolling operations. Engineering examples demonstrate that compared to existing e⁃
valuation algorithms, the proposed method exhibits high precision (R = 0.839), strong generalization ability (R =
0.793), and robustness. The comprehensive evaluation method based on evidence theory can account for the un⁃
certainty of limited random test pit data while reducing CV sensitivity to noise by 69.8%. Additionally, the multi-
level feedback control mechanism effectively ensures on-site rolling quality.
Keywords: earth-rock dam construction; rolling quality evaluation; uncertainty; ACGWO-RF algorithm; im⁃
proved evidence theory
(责任编辑: 韩 昆)
— 1 2 9 —