基于改进粒子群算法的大坝监控加权统计模型

发布时间：2018-04-25 05:19

本文选题：土石坝 + 加权统计模型　；参考：《长江科学院院报》2017年08期

【摘要】：用于大坝安全监控的加权统计模型主要依据工程经验确定各因子的权重,这种求解方式易导致部分因子信息的缺失。根据大坝安全监测数据,应用粒子群算法可优化确定加权统计模型中各参数的最优解,但对于高维度优化问题,该算法存在收敛速度慢、易陷入局部最小等不足。针对这些不足,考虑粒子种群平均位置信息的影响,提出一种新的改进粒子群算法,利用单体与种群平均位置的距离信息确定两者之间的学习因子。土石坝工程实例分析结果表明:改进粒子群算法加强了种群跳出局部最小的能力,所得加权统计模型的权重符合工程实际情况。尤其在大坝运行初期,监测资料较少的情况下,基于改进粒子群算法的大坝监控模型具有较高的预测精度和预报能力,可为大坝监控领域提供一种新的数据分析方法。
[Abstract]:The weighted statistical model used for dam safety monitoring is mainly based on the engineering experience to determine the weight of each factor. This method can easily lead to the loss of some factor information. According to the dam safety monitoring data, the particle swarm optimization algorithm can be used to optimize the optimal solution of the parameters in the weighted statistical model, but the algorithm is a high dimension optimization problem. In view of these deficiencies, considering the influence of the average position information of the particle population, a new improved particle swarm optimization algorithm is proposed to determine the learning factor between the two and the distance information between the monomer and the population average position. The analysis results of the earth rock dam project show that the improved particle swarm optimization is used. The method strengthens the ability of the population to jump out of the local minimum. The weight of the weighted statistical model is in line with the actual situation of the project. Especially in the case of the early stage of the dam operation, the dam monitoring model based on the improved particle swarm optimization algorithm has a higher prediction accuracy and prediction ability, and can provide a new number for the dam monitoring field. According to the method of analysis.

【作者单位】：南京水利科学研究院岩土工程研究所;
【基金】：国家自然科学基金项目(51379129) 水利部公益性行业科研经费项目(sg315002)
【分类号】：TV698.1

【相似文献】