高心墙堆石坝大场景视频监控网络覆盖NCHHO优化

覆盖率和成本是衡量高心墙堆石坝视频监控网络部署优劣的重要指标。然而，现有研究缺乏对建设成本的综合考虑，且常用的视频监控网络覆盖优化求解方法存在收敛速度慢、易陷入局部最优等不足。针对上述问题，本文提出一种高心墙堆石坝大场景视频监控网络覆盖改进哈里斯鹰优化模型——非线性混沌哈里斯鹰优化（nonlinear chaotic Harris hawks optimization, NCHHO）模型。首先，提出表征视频网络部署成本的单位摄像头重复采样率指标，并基于集合覆盖理论构建以覆盖率和单位摄像头重复采样率最大为目标的视频监控网络覆盖优化模型。其次，利用混沌序列和非线性能量更新策略改进哈里斯鹰算法的种群初始化和搜索过程，提高算法的收敛速度、避免陷入早熟，并利用其求解视频监控网络覆盖优化模型。实例验证了改进哈里斯鹰算法在视频监控网络部署优化中的有效性和优越性，本研究得到的优化方案覆盖率和重复采样点比例分别为99.98%和60.3%，相比经验方案提高了13.8%和23.2%，显著优化了视频监控效果。

Nonlinear chaotic Harris hawks optimization model for large-scene video monitoring network coverage of high core rockfill dams

Coverage rate and cost are two important measures of a video surveillance network for high core rockfill dams, but previous studies in the literature lack a comprehensive consideration of deployment cost, and the commonly-used optimization methods have shortcomings such as slow convergence and easy falling into local optimization. To solve these problems, this paper develops a nonlinear chaotic Harris hawks optimization model for large-scene video monitoring network coverage of a high core rockfill dam. First, a resampling rate per camera is employed to represent deployment cost, and an optimization model is constructed to maximize the coverage rate and resampling rate per camera based on the set covering theory. Then, a chaotic sequence and the nonlinear energy update strategy are adopted to optimize the population initialization and search process of the Harris hawks algorithm, which improves the convergence and avoids falling into prematurity. Finally, an improved algorithm is used to solve this new model. Application to a dam construction project has verified our improved Harris hawks algorithm is effective and superior in the deployment optimization of the surveillance network, achieving a coverage rate and resampling points proportion of up to 99.98% and 60.3% respectively, or 13.8% and 23.2% higher than the empirical scheme, with the video surveillance effect improved significantly.