无线传感器网络下静态水体中的近岸污染源定位

发布时间：2018-10-07 18:19

【摘要】：排放于水库湖泊中的污染物的扩散易受到边界影响.本文首先对静态水体中的近岸污染源扩散进行理论分析,提出了一种分段浓度模型.然后,研究了静态水体中靠近不透水边界的污染源定位问题,指出该问题中的未知参数不仅有污染源位置,还包括质量流率和初始扩散时间,分别给出了通用模型法、近似函数法、基于无迹卡尔曼滤波的估计方法求解参数估计问题.通用模型法与近似函数法分别通过求解基于原始扩散模型和分段扩散模型的约束非线性最小二乘算法获取参数估计.通用模型法可快速获取目标源相关信息,近似函数法有更稳健的参数估计性能,但需要经历多个采样时刻后才可执行.基于无迹卡尔曼滤波(Unscented Kalman filter,UKF)的估计方法结合扩散过程,可有效权衡数值计算复杂度与估计性能.在仿真实验部分,对近岸污染源扩散过程进行了水文模拟,根据模拟数据对比了不同算法的实验性能,说明了各算法的优势和不足.
[Abstract]:The diffusion of pollutants discharged into reservoir lakes is easily affected by the boundary. In this paper, the diffusion of near-shore pollution sources in static water is analyzed theoretically, and a segmental concentration model is proposed. Then, the problem of pollution source location near the impermeable boundary in static water is studied. It is pointed out that the unknown parameters in the problem include not only the source location, but also the mass flow rate and initial diffusion time, and the general model method is given respectively. The approximate function method, based on unscented Kalman filter, is used to solve the parameter estimation problem. The general model method and the approximate function method obtain the parameter estimation by solving the constrained nonlinear least squares algorithm based on the original diffusion model and the piecewise diffusion model respectively. The general model method can obtain the target source correlation information quickly. The approximate function method has more robust parameter estimation performance, but it needs to go through several sampling times before it can be executed. The estimation method based on unscented Kalman filter (Unscented Kalman filter,UKF) combined with diffusion process can effectively balance the computational complexity and estimation performance. In the part of simulation experiment, the hydrological simulation of the diffusion process of near-shore pollution source is carried out. The experimental performance of different algorithms is compared according to the simulated data, and the advantages and disadvantages of each algorithm are explained.
【作者单位】：武汉科技大学冶金自动化与检测技术教育部工程研究中心;
【基金】：国家自然科学基金(60974012,61171160) 湖北省高等学校优秀中青年科技创新团队计划项目(T201302)资助~~
【分类号】：X52;TP212.9;TN929.5

【参考文献】