Evaluation And Optimization Of Hydrologic Network Based On The Theory Of Information Entropy

As the carrier of hydrologic information, Hydrologic network layout directly affects the representativeness and research value of hydrological data provided by the hydrologic network on the analysis of inherent law and characteristics for basin.However, the problem of "data fully but lack of information" appears. A large amount of manpower and resources is put into the construction of hydrologic network, but the hydrologic data produces redundant information, which has brought inconvenience and interference to the research and application. So we can optimize the hydrologic network to get the maximal information, which can not only save resource and investment for hydrological construction, but also provide scientific and systematic hydrological data for hydrology and water resources research.The traditional network optimization method has certain limits and disadvantages, and this paper introduced the information Entropy which is a measure of the uncertainty or information of system. As it based on information theory, information entropy method is differ from traditional method of network optimization. It can quantify the amount of information provided by a site and transmited between sites from the angle of systematicness and integrity, and establish evaluation model and network optimization model to evaluate and optimize the layout of the network. This paper researchs and analysis the network evaluation and optimization model based on information entropy combining with the instance, trying to provides a new feasible network optimization method for river basin. The main research contents and results are as follows:(1) Based on the related basic information entropy theory, and according to the characteristics of the watershed hydrologic network and data, which is the transferability, correlation and attenuation between site hydrology information, this paper demonstrate the reasonableness and feasibility of hydrometric network evaluation and optimization method based on information entropy, to lay the theoretical foundation for further exploration.(2)On the basis of two existing models which is used to describe the law of information transmission between sites:transinformation-distance model(T-D Model) and correlation-distance model(C-D Model), This paper put forward two new models: transinformation-distance along the river model (T-AD Model) and correlation-distance along the river model(C-AD Model). The experiment simulated the flow and water level information transmission rule of three subbasin network by using four model, and found the best model:transinformation-distance model. Then use this model to evaluate special scale controlled by the hydrological sites, under different information transfer standard. The result indicates that there are certain redundancy between sites for the two kinds of hydrological information, and network reduction should be feasible.(3)This paper introduces two kinds of site optimization model based on information entropy theory, namely MaxT model and ti model, to optimise the sites combination for three subbasin from the space. By comparing the experimental results of two models, it can be concluded that the method based on information entropy is more applicable to reducing the redundancy site in the network. By comparing the site optimization results of two kinds monitoring data, it can be concluded that different hydrological information has different transmission characteristics on the space. In the real work, site setting should consider the information of different hydrological monitoring data provided by a site synthetically.(4) Optimising the network from the time is the optimization of site monitoring frequency. The experiment choice the sites which has the maximum and minimum information entropy from the network to carry on the frequency optimization research respectively. Results show that the information entropy method can be well used for the monitoring frequency optimization. Because the method can quantify the information provided by new monitoring sequence, then corresponding monitoring frequency can be determined according to the requirements of the different monitoring accuracy.The innovation point of this article lies in the exploration of the network evaluation and optimization and site monitoring frequency optimization method based on information entropy theory, and in the analysis of the applicability of each method. There are only less research on evaluation and optimization of hydrologic network based on the information entropy theory domestically, this paper has some exploratory.