| Water is the most important resource for human being exist and development;it is also the basic element of the environment. However, when we enter 21st century, many countries face a serious challenge, i.e. water resource crisis. In China, water resource shortage, water pollution, flood disaster etc seriously jeopardize the advance of the economy and society;and also cause a series of social problem. So, it is meaningful for us to research water resource system to solve water resource crisis, then we can realize water resource sustainable utilization and the sustainable development of society, environment. The exploitation, utilization and management of water resource are rational activities that human being action on nature water system, Generally speaking, water resource system management is a multipurpose, multi-attribution, multi-layer, multi-function process that involves decision making, evaluation, prediction etc, where we can not ignore the fuzziness. Project fuzzy sets theory is a powerful tool to deal with those issues. This paper mainly research on decision making, evaluation and predicition issues in water resource system, including model development and their application. The main contents and research results are as follows:(1) Water pollution problem has been becoming a serious issue in our country, which affects the human being's survival and the development of society. Water quality evaluation is the initial work for further treatment of water pollution. In this paper, a multi-output fuzzy optimization artificial neural recognition network is developed, where training data is constructed based on standard water quality index, level relative membership degree is taken as expectation output. It is an efficiency method to obtain surface water quality level feature value. Case study of Tuo river water quality evolution shows that Fuzzy Neural Network Recognition (FNNR) has excellent applicability and practicability.(2) Water resource is an important resource constitution for a country or region;it bears a county's society and economy development. In this paper, the definition and characteristics of water resource bearing capability is introduced in details. Considering the fuzziness in water resource bearing capability analysis, we propose to use fuzzy recognition theory to deal this issue. We firstly obtain objective contribution ration of the factor through iteration computation. Meanwhile, experts' experience, knowledge and experience should be also taken into consideration. We apply complementary pair-wised comparison method to decide subjective weights of the factors, thereafter, synthesizing objective contribution ration and subjective weights of the factors, we use fuzzy recognition model to obtain the water bearingcapability feature value. Case study of 30 regions presents the water bearing capability condition of China.(3) Fuzzy clustering (FC) model is an efficiency tool for water resource system analysis, however, fuzzy clustering (FC) based on objective function have several difficulties such as clustering number selection, local minimum points in iteration etc. In this paper, firstly, reasonable clustering number selection is researched;a practicable method to decide reasonalble clustering number is developed. Secondly, in order to solve local minimum problem, this paper introduces Particle Swarm Optimization algorithm (PSO), which is a biology intelligence ialgorithm, and it is easy to be operated compared with Genetic Algorithm. Thereafter, PSO-FC (Fuzzy Clustering model Particle Swarm Optimization algorithm) is established. Finally, PSO-FC is used to deal with water resource partition and reservoir flood clustering, which shows that PSO-FC has good practicability.(4) Fuzzy optimum artificial neural network is a efficient prediction method in water resource system, however, the training algorithm (BP? Back Propagate) has local minimum problem. In this paper, Particle Swarm Optimization algorithm (PSO) is used to optimize the training process of the artificial neural network trainning, which can reduce the probability of local minimum problem occurrence. Then, a model of Fuzzy optimum artificial neural network based on PSO is developed, and it is used to predict the ice flood in Inner Mongolia in our country, the result shows that this model is a new predicition method for water resource system.(5) There is a kind of multi-criteria decision making problem in which criteria are all qualitiatative. in this paper, we propose to use fuzzy number theory to solve this issue, first, the definitions and properties of fuzzy number are introduced, next, fuzzy linguistic variables are established, we take consideration of experts' experience, knowledge and preferences to make decisions from viewpoints of quantitative analysis, then, a decision model base on fuzzy number theory is developed to convert the quantitative judgements into quantatitative analysis. A case study of project prequalification in water resource system is presented to illustrate this model.(6) Tender evaluation is a complex multi-objectives group decision issue in water resource project construction, a successful bid evaluation decision making will not only ensure the overall quality of the project, but also have the opportunity of saving on costs. Especially, as for the large investment, complex, high risk water resource project, so, an efficient multi-objectives decision making methodology is badly needed to meet our requirement. In this paper, a Fuzzy Pattern Recognition Group decision-making framework for selecting contractors is developed to solve bid evaluation issue. Thereafter, considering consensus ofgroup decision making, we propose a method to analyze the consensus degree of group decision making. A case study of water resource project bid evaluation shows the framework can support the decision making;and it will benefit the standardization of water resource system decision making.Finally, a summary is given and some problems to be further studied are discussed.
|
PDF Full Text Request