基于移动检测平台的藻类水华短期预测方法研究

发布时间：2018-06-13 04:04

本文选题：藻类水华 + 水华形成机理　；参考：《浙江大学》2015年硕士论文

【摘要】：近年来,随着水体富营养化程度加剧,藻类水华频繁暴发。藻类水华不仅破坏水体生态环境,还威胁人类身体健康,并且缺少短期内的有效治理手段,因此,对水体藻类浓度进行实时预测,在水华暴发之前采取应急措施,降低治理成本,具有重要意义。本文在分析了国内外预测方法的基础上,根据生长机理和通用信号处理方法各自的特点,提出了一种基于藻类水华形成机理的分段短期预测方法。同时针对国内外水质移动检测系统的不足,改进了一种低成本、便携性好、操作方便灵活、可在线检测部分水质参数的移动检测系统。并将基于藻类水华形成机理的分段短期预测模型与水质移动检测系统结合,实现对饮用水源地(地表水)的不同水域进行藻类水华实时预测。本文主要工作和特色如下：(1)建立了基于藻类水华形成机理的分段短期预测模型。根据藻类水华形成的过程,确定了藻类水华形成关键影响因素,即水温、光照和营养盐；参考藻类水华形成“四阶段理论”,本文将全年按月分为3个阶段,选择不同阶段的影响因子,建立藻类水华形成分段机理模型；并根据藻类短期生长趋势预测未来某个时刻的生长情况。(2)完成了藻类分段短期机理预测模型的仿真实验。选用德国易北河2000年3月到10月的监测数据用于模型验证,并用粒子群优化算法动态率定模型参数。从率定数据时间跨度、率定参数组合、参数动态率定预测序列和未来三日预测序列四个角度分析了预测模型对叶绿素a浓度的预测情况,初步结论为：①选取率定数据时间跨度为7天时,预测结果最优；②光半饱和常数K,的率定对预测结果的影响要优于光系数(?)；③预测序列误差约为10%,说明该预测模型能够很好的应用于德国易北河叶绿素a浓度预测；④从未来一日到未来三日,预测误差依次增大。(3)完成了水质移动检测系统的硬件和软件改进。该系统由移动检测平台、监控中心和手持终端三部分组成,移动检测平台用于对目标水域水质信息在线检测,并将检测结果发送到监控中心和手持终端,本文实现了水温和光照强度的远程检测；监控中心用于存储水质历史数据和实现藻类水华预测；手持终端用于发送相应控制命令。改进后的系统支持多种水质参数在线检测,并在移动检测平台上搭载图像传感模块,采集水样图像,通过水样颜色来快速判断藻类水华暴发情况。
[Abstract]:In recent years, algae Shui Hua outbreaks frequently with the increase of eutrophication. Algal Shui Hua not only destroys the ecological environment of water body, but also threatens human health, and lacks effective control measures in the short term. Therefore, the concentration of algae in water is predicted in real time, and emergency measures are taken before the outbreak of Shui Hua. Reduce management cost, have important meaning. Based on the analysis of the prediction methods at home and abroad and according to the characteristics of the growth mechanism and the general signal processing method, a segmented short-term prediction method based on algal Shui Hua formation mechanism is proposed in this paper. At the same time, a mobile detection system with low cost, good portability, convenient and flexible operation and on-line detection of some water quality parameters is improved in view of the shortage of domestic and foreign mobile water quality detection system. Combining the segmented short-term prediction model based on algal Shui Hua formation mechanism with the water quality moving detection system, the real-time prediction of algae Shui Hua in different water areas of drinking water source (surface water) is realized. The main work and characteristics of this paper are as follows: (1) A segmented short-term prediction model based on algal Shui Hua formation mechanism is established. According to the process of algal Shui Hua formation, the key influencing factors of algal Shui Hua formation, namely, water temperature, light and nutrient, are determined. With reference to the "four-stage theory" of algal Shui Hua formation, this paper divides the whole year into three stages by month. A segmental mechanism model of algal Shui Hua formation was established by selecting the influence factors of different stages, and the simulation experiment of algal segmented short-term mechanism prediction model was completed according to the short-term growth trend of algae at a certain time in the future. The monitoring data from March to October 2000 of the Elbe River in Germany were used to verify the model and the model parameters were determined by using the particle swarm optimization algorithm (PSO). The prediction model for the concentration of chlorophyll a was analyzed from four angles: the time span of rate data, the combination of rate and parameter, the prediction sequence of parameter dynamic rate and the prediction sequence of future three days. The preliminary conclusion is that when the data span is 7 days, the optimal optical half-saturation constant K is obtained, and the effect of the ratio determination on the prediction results is better than that on the optical coefficient. (3) the error of prediction sequence is about 10, which indicates that the prediction model can be applied to the prediction of chlorophyll a concentration in the Elbe River of Germany from the next day to the next three days. The prediction error increases in turn. 3) the hardware and software of the mobile water quality detection system are improved. The system consists of three parts: mobile detection platform, monitoring center and handheld terminal. The mobile detection platform is used for on-line detection of water quality information in target waters, and the results are sent to the monitoring center and handheld terminal. In this paper, the remote detection of water temperature and light intensity is realized; the monitoring center is used to store water quality history data and realize algal Shui Hua prediction; and the handheld terminal is used to send corresponding control commands. The improved system supports on-line detection of various water quality parameters and carries image sensing module on the mobile detection platform to collect water sample images and quickly judge algae Shui Hua outbreak by water sample color.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：X84;X832

【参考文献】