| The Three Gorges Reservoir has been constructed and has officially scheduled to run. It has achieved great comprehensive benefits in power generation, flood control, navigation and so on. But it also causes the hydrological condition variations to many secondary rivers. Due to the impounding of the Three Gorges Reservoir, many backwater areas with different length and scope formed, at the convergence between the secondary rivers and the Yangtze River. In the backwater areas, the average flow velocity dropped sharply. Besides, the nutrient concentration in the backwater areas was high, such as nitrogen, phosphate and so on, together with appropriate sunlight, it provided favorable conditions for algal's growing and reproduction. In recent years, many secondary rivers backwater areas break out different degrees of "bloom" phenomenon. For the higher eutrophication potential and less study of secondary rivers backwater areas, this paper chose the heavily polluted tributaries Linjiang River backwater area of the Yangtze River as the research object, and had conducted the following studies on the base of the continuous monitoring of the backwater area water quality:â‘ Based on the analysis of eight indicators including CODMn , TN , TP , Chla , SD , DO , TW and flow velocity, the author used trophic state index method to evaluate the eutrophic conditions of the backwater area. The results indicated the eutrophication degrees of backwater areas were very high, especially during the spring which had suitable temperature, light and sufficient nutrients in river, the comprehensive nutrition indexes were more than 80. Eutrophication of the backwater area was very serious.â‘¡Analyzed the correlation between each of the sever indicators and Chla with Grey relational analysis method, which are TN, TP, CODMn, Tw, SD, DO, flow velocity. Results showed that TP had the closest correlation with Chla, while TN had the minimal correlation with Chla, which also showed that the chlorophyll-a concentration was limited by TP to some extent. Though the TN concentration was very high, the small changes of TN concentration had smaller effect on growth and reproduction of algae.â‘¢Based on the advantages and disadvantages of genetic algorithm and BP neural network, this paper had established a GA-BP neural network for the short-term prediction of eutrophication in backwater area. Chose the network's input variables by the results of Grey relational analysis method, and chose Chla to be the output of the network, then use Matlab 2010 to program. The results showed that the model had a higher accuracy, that predicted values had the high correlation coefficient of 0.9892 and lower average relative error of 9.8% to measured values. Also the model had a good generalization performance that could be used to predict the eutrophication of the backwater area in a short time. Meanwhile, the comparison results between the GA-BP model and the BP neural network showed that the training effect and the prediction accuracy of the GA-BP neural network model were both better than the single BP neural network.â‘£This paper combined the GA-BP neural network mode with research achievement of the previous ecological floating bed technology using in heavily polluted Linjiang river water, and used GA-BP model to predict the concentration of chlorophyll-a after setting a range of different sizes canna floating beds in the Linjiang River backwater area and its upstream. Then it analysised the eutrophication potential in the backwater area, and evaluated the Chla's control effect under different hydraulic loading by canna ecological floating bed. The results showed that ecological floating bed technology can effectively control the eutrophication state of Linjiang River backwater area.
|
PDF Full Text Request