Several Key Technology Research In The Decision Model Management Of Optimal Operation For The Pumping Stations

Water is the important and indispensable resource to the human being’s production and life. China is a nation that has the shortage water resource per capita. In our country, the water resource is generally poor, and its distribution is uneven in time and space. The level of the development and utilization is relatively low. With the strong development of industry and agriculture and the outstanding improvement of people’s living standards, the demand to water is increasing. Water stress is exacerbated by the regional difference of its distribution and the unreasonable water supply quantity. The water shortage has become an important factor to hinder Chinese economic development and social progress. Therefore, in order to effectively alleviate and solve the problem of water shortage, it is very urgent and necessary to construct the inter-basin water transfer project and optimize its operation scientifically. According to the idea of "energy-saving and cost-reducing" proposing by China’s "Eleventh Five-Year Plan", we should set up Decision Support Systems for the use the inter-basin water transfer project, using the modern decision theory and combining with the project’s actual situation. It is of great significance to the optimal scheduling for the inter-basin water transfer project.In this respect, on the basis of reading and consulting a lot of literature at home and abroad, the paper combined the theoretical knowledge of Decision Support System with the optimal operation of the pumping stations. It gave the theoretical framework of the Model-Driven Decision Support System of the optimal operation of the pumping stations. And as a starting point, the previous studies of the decision model of the optimal operation of the pumping stations were summed up and sorted out. We researched the several key technology in the model management of the optimal operation of the pumping stations, including research on Hidden Markov Model-based model classification, model knowledge representation and method base building, and the model selection. The paper did the following work:First, we analyzed the necessity for optimal operation of the pumping stations to use DSS, as well as the necessity for the model as the focus on. We gave the decision-making process of the system.Second, we researched the modeling principle of the decision model of the optimal operation for the pumping stations, summarizing the existing models. We set up the Model Base Management System. The model management is mainly reflected in the model’s properties base management, model’s storage management, model’s operation management, and model combination etc.Third, in the model classification, we used Hidden Markov Model to research the process of the model classification theoretically. We gave the classification process, training process and the performance evaluation setup. And the method also had the inadequacies and the improving places.Fourth, we separately researched explicit knowledge and implicit knowledge of the decision model in the model’s knowledge representation. We used model dictionary to store the properties knowledge, and create a model directory tree to facilitate model management. Then we analyzed the implicit knowledge’s origin in the field of the optimal operation for the pumping stations, and transform the implicit knowledge using SECI knowledge spiral model.Fifth, in the model selection, we stored the model attributes and attribute values by constructing two-stage model dictionary base. Then, we set up a search tree and form the model selection process, proposing the evaluating indicator of the model selection. As the background of North Water Transfer Eastern Source Project, we developed the model selection system of the optimal operation for the pumping stations, which can demonstrate the method’s validity.Finally, we summarized the findings of the paper, and pointed out the follow-up research and development direction in the decision model management of the optimal operation for the pumping stations.