Dissertation > Excellent graduate degree dissertation topics show

Study on Prediction of Runoff and Sedimentation and Multi-objective Optimal Operation of Reservoir

Author: LiHui
Tutor: LianJiJian
School: Tianjin University
Course: Water Resources and Hydropower Engineering
Keywords: River generalized model Water flow propagation time Runoff Prediction Wavelet decomposition Sediment concentration prediction Reservoir sedimentation Water and Sediment electric multi - objective optimization Improved multi-objective particle swarm optimization
CLC: TV697.1
Type: PhD thesis
Year: 2008
Downloads: 472
Quote: 2
Read: Download Dissertation


Sustainable economic and social development of water conservancy, the ever-increasing demands. Reservoir to regulate runoff, development and utilization of water resources effectively, and how to become an important research topic to achieve its short-term benefits and long-term benefits, economic and social benefits of the unified multi-objective optimization of reservoir scheduling. Reservoir multi-objective optimization scheduling involves a number of related issues, including the natural evolution of river flow and water flow propagation time, the prediction of multi-day runoff prediction, sediment concentration, sediment erosion and deposition in the reservoir area of ??intelligent forecasting, reservoir water and sediment electric multi-objective optimization scheduling, etc. The main conclusions are as follows: (1) the form of a single slot form and beach slot the two river generalized model technology to achieve the major cross-section of the abstract generalization partition constructed river generalized model. Then this article and take advantage of the side of the stream, and to consider the Saint-Venant equations and sidestream Preissmann implicit difference method simulated the evolution of river flow. An example shows that consistent the considered interval inflow and tributaries flow evolution of simulation results with the actual process trends, peak and valley corresponds to a good relationship, and can basically reflect the actual flow conditions. Finally, we will apply genetic algorithm to improve BP neural network model to forecast river flow propagation time and flow propagation time prediction model. Calculation examples show that our model can better reflect changes in the propagation time under different input trend, the propagation time of the water in the case of the smaller tributaries flow more accurately forecast, optimal for rivers utilization and reservoir scheduling run to provide a reference. (2) run from the reservoir actual scheduling needs, the short warehousing day runoff forecast forecast period, forecast accuracy and low, stepwise regression prediction model based on wavelet decomposition daily runoff. With daily runoff prediction model, this paper will predict hydrological stations upstream hydrological stations daily runoff sequence introduction of predictive models, while taking advantage of the wavelet decomposition and reconstruction sequence of daily runoff forecasting hydrological stations and upstream hydrological stations in different scales overview of components is determined on the basis of the correlation analysis candidate impact factor and the use of stepwise regression analysis to determine the optimal regression forecasting model. An example shows the stepwise regression of the wavelet decomposition daily runoff prediction model based prediction accuracy is higher than multivariate autoregressive model, able to predict the next one to three days of non-ice flood season and ice flood season the average daily flow rate of 1 to 7 days able to provide a scientific basis for the development of hydropower generating plan. (3) in order to The Yellow River Toudaoguai hydrological stations for example the prediction of the average daily flow sediment concentration studied in different runoff mode of non Ling of the Yellow River flood season and ice flood season sediment concentration factor analysis the basis of on, respectively, the neural network model for predicting sediment concentration in runoff mode. Calculation examples show that, compared with the linear multiple regression model, the silt content of the neural network prediction model can be more effective handling nonlinear relationship between sediment concentration and various influencing factors, for the river water and sediment regulation reservoir scheduling run to provide some reference. (4) study the problem of reservoir sedimentation, composite model predictions of reservoir sediment erosion and deposition. Considering the muddy water of suspended sediment siltation and density current dual role of siltation, non-saturated equilibrium sediment transport model and density current motion model coupling and the establishment of a one-dimensional coupled sediment mathematical model, and its The basic equations and solving method of study, and the virtual flow method is applied to non-constant density flow model solution to accurately forecast the evolution of density current flow time. The model of this paper were applied to the the Wanjiazhai reservoir sediment erosion and deposition calculation and Xiaolangdi Reservoir density flow simulation calculation results show that the model can well reflect the changes in the development process of reservoir sedimentation, and be able to accurately simulate gravity flow The head of the evolution and predict the flow of time. In addition, in order to overcome the water and sediment dynamics model parameters more calculated burdensome shortcomings, one-dimensional coupled sediment mathematical model and BP neural network model of the composite and the establishment of the the composite reservoir sediment erosion and deposition BP network model, the model The input and output unit has a clear physical meaning. The Wanjiazhai reservoir sediment erosion and deposition Quick predict examples show that the the composite BP network model has a simple, rapid, high precision, the calculation speed is 250 times higher than the one-dimensional coupled sediment mathematical model, solve the reservoir of multi-objective scheduling the speed bottleneck problems, and provide a forecast for reservoir management reservoir sedimentation easy way. (5) the reservoir water and sediment electric multi-objective optimization studies, the establishment of multi-objective optimization model of reservoir water and sediment electric. Meanwhile, in view of the lack of the traditional constraint method and weighting method chapter combined with the concept of Pareto optimal solution introduced particle swarm optimization inertia weight adaptive adjustment mechanism and Pareto optimal solution library mechanism for the formation of improved multi-objective particle swarm optimization (IMOPSO), then With this algorithm, the multi-objective optimization model to directly solve the optimization problem to get the multi-objective reservoir Pareto optimal front. The method of calculation examples show that this chapter to find out the Pareto optimal front with good dispersion properties, shows the results of the optimization is very intuitive, providing more effective support for the decision-makers. In addition, in contrast with the NSGA-II, IMOPSO algorithm has excellent global optimization performance and dispersion, ideally suited for multi-objective optimization problems.

Related Dissertations

  1. Inspection systems for metal abrasive noise cancellation algorithm,TP391.41
  2. Based on wavelet analysis method for fast simulation of fluid,TP391.41
  3. Network Traffic Model Based on Wavelet Decomposition and Arima,TP393.06
  4. Wavelet-based multi-resolution image de-noising method of nonlocal,TP391.41
  5. Content-based Image Retrieval Methodology and Implementation,TP391.41
  6. Study on Wavelet-Based Damage Detection in Pipes Using Ultrasonic Longitudinal Guided Wave,TP274.53
  7. Based on wavelet transform image fusion technology research,TP391.41
  8. Research on Image Fusion Algorithm Based on Wavelet Analysis,TP391.41
  9. The Analysis of Multi-point Input about Long-span Structures under the Vertical Seismic,TU311.3
  10. Study of Palmprint Feature Extraction Algorithm Based on Multi-resolution Analysis and Gray Level Co-Occurence Matrix,TP391.41
  11. Tire -prints image enhancement method,TP391.41
  12. Image Hiding Based on Wavelets and Hexagonally Sampling,TP391.41
  13. Research and Application of statistical regularities of natural images in digital image forensics,TP391.41
  14. Surface Defect Detection of HGA Using Wavelet Method,TP391.41
  15. EEG-EMG Synergistic Analysis and Correlation Study Based on External-Source Load Inspiration,R87
  16. The longitudinal erosion and deposition , and the research and application of numerical simulation of lateral deformation of alluvial river,TV147.3
  17. Whitehead use of reservoir sediment,TV145
  18. The Sedimentation Deformation Calculation by Quasi-2D Numerical Simulation in the Qingtongxia Reservoir,TV145
  19. Numerical Simulation of Hydraulic Scouring of Silt in Front of Qingtongxia Reservoir Dam,TV145
  20. Sediment-laden river Dispatching Optimization,TV697.1

CLC: > Industrial Technology > Hydraulic Engineering > Water control,hydraulic structures > Reservoir Management > Reservoir operation management
© 2012 www.DissertationTopic.Net  Mobile