Dissertation > Excellent graduate degree dissertation topics show

Application of Rough Set and Flex in Mid-long Term Runoff Forecasting

Author: SuAnJun
Tutor: YuanXiaoHui
School: Huazhong University of Science and Technology
Course: System Analysis and Integration
Keywords: Medium - and long - term runoff forecast Rough Set Flex Support vector regression
CLC: P338
Type: Master's thesis
Year: 2011
Downloads: 16
Quote: 0
Read: Download Dissertation


Water resources are the most basic natural resources that support human survival and sustainable development of society, the medium-and long-term runoff forecasting can provide important reference for the protection of water resources development, management. Medium-and long-term runoff forecast the largest convergence time of basin runoff forecast period is over, based on a mathematical model to predict the results of the prediction can provide an important basis for the development of water resources in the medium and long run program, and is widely used in reservoir operation hydropower run, flood and drought control, water resources allocation and shipping management. Foresee a longer period, however, many influencing factors led to the runoff formation process is very complex, the deepening of human activities, so that the long-term runoff prediction results are often not high precision. This paper attempts to introduce the rough set theory to the study of the long-term runoff forecast, from the most influential factor in the historical runoff data mining, delete the less impact factors, and then based on the core factor set for runoff prediction using support vector regression model. Based on Flex technology, design, and development of long-term runoff forecasting system with powerful interactive capabilities and rich page rendering capabilities, support for hydropower optimal scheduling. Work and research results of this paper is mainly reflected in the following aspects: 1) the rough set theory is introduced to the study of the long-term runoff forecast annual runoff and combined with of Xinjiang Ili Yamaha crossing station data modeling and application. Conditional attributes discretization based on information entropy algorithm, using genetic algorithms on condition attribute reduction runoff core factors set. Due to the rough set theory without providing any additional information required for the processing of the data collection beyond description and processing of medium-and long-term runoff forecast uncertainty is more objective. Rough set theory, attribute reduction and eliminating redundant information and greatly reduced SVM training data to improve the running speed of the system. 2) As the rough set theory can only deal with discrete data, and the anti-interference ability, this paper introduces SVM theory to get the runoff predicted values. SVM as rough rear set pretreatment system, fault tolerance and anti-jamming capability. Using genetic algorithm and cross-validation to the optimization of the relevant parameters of SVM, to get rid of the previous parameter selection blindness and subjectivity, improve forecast accuracy. The forecast results show that the combination of rough set theory and SVM algorithm has good generalization and anti-jamming capabilities for the medium-and long-term runoff forecast, prediction system, and get a good forecast performance. 3) based on the the the Flex technology development and long-term runoff forecast system, make full use of its powerful component library, as well as flexible custom components to build a WEB application with rich interactive capabilities chart, with a good user experience. WEB application using Flex technology development, thoroughly implement the MVC architecture, control user operation logic is completely separated from the server-side code, clear system hierarchy. Flex technology data and interface components closely linked, and asynchronously communicate with the server, users do not need to interrupt the operation to wait for the data to refresh the data transfer process. 4) in the server-side use of long-term runoff forecast system the JNI technology forecasting algorithm is called dynamic link library files. Prediction algorithm library files are already old module, using C language, which to some extent make up for the deficiencies of the Java language in a program run performance, improve the computing speed of the algorithm, and use the dynamic link library package algorithm, also largely protect the intellectual property rights of the forecast algorithm so that the guarantee on the basis of the system functions, shorten development cycles, reduce development costs, and protect the historical investment.

Related Dissertations

  1. Fault Diagnosis Method Based on Support Vector Machine,TP18
  2. Research on Clustering Algorithm Based on Genetic Algorithm and Rough Set Theory,TP18
  3. Based on Rough Set of Urban Areas When Traffic Green Control System Research,TP18
  4. Incremental rough set attribute reduction,TP18
  5. Calculation of Knowledge Granulation and Study of Its Application in Attribute Reduction,TP18
  6. Research on NO-reference Image Quality Assessment Based on HVS,TP391.41
  7. Research of License Plate Recognition Based on Rough Sets and Fuzzy SVM,TP391.41
  8. Importance of the study of the physical and chemical indicators based on rough set theory Daqu,TS262.3
  9. The software design and implementation of the the clothing quality prediction system,TP311.52
  10. And trusted high compression video surveillance network-related research,TP277
  11. Cooperative Optimization Scheduling with Application to Multi-Reservoir System During Non-Flood Period,TV697.11
  12. Intelligent algorithms based carbon fiber spinning process monitoring and optimization,TQ342.742
  13. Water quality time series data processing and Early Warning System Construction Research Database,TP274
  14. Study on the Decision Tree Classification Algorithm and Its Application Based on Rough Set Theory,TP18
  15. Application of Flex Technology in Display Cube Control System,TP273
  16. Support Vector Regression in chemical pesticides QSAR Application,S48
  17. Personalized desktop ordering system design and implementation,TP311.52
  18. Based on the combined effect of the rough planning model,O221
  19. Based on the core set of examples of attribute reduction method,O159
  20. WebGIS-based geographic information support technology in the water safety warning systems applied research,P208

CLC: > Astronomy,Earth Sciences > Geophysics > Hydrological Sciences (water sector physics) > Hydrological forecasting
© 2012 www.DissertationTopic.Net  Mobile