Dissertation > Excellent graduate degree dissertation topics show
A Parallel PCG Solver for Groundwater Model in the Area of Energy Base of Mongolia
Author: XuTeng
Tutor: HuFuSheng
School: Chinese Geology University (Beijing)
Course: Hydrology and Water Resources
Keywords: Energy base in Inner Mongolia Groundwater balance OpenMP Parallel computing
CLC: TV12
Type: Master's thesis
Year: 2010
Downloads: 82
Quote: 0
Read: Download Dissertation
Abstract
|
Energy base in Inner Mongolia is located in the arid - semi-arid climate zone , the annual evaporation is much larger than the annual rainfall . Resources development and utilization of the energy base in recent years , accelerated groundwater shortage , caused some impact on the local economy and ecological . Therefore, the understanding of groundwater occurrence , rational planning of groundwater exploitation has profound significance . This study on the basis of the hydrogeological information GMS software simulation of the 2000-2006 groundwater in the study area three-dimensional flow field simulation . The simulation results show that the the Research Chief recharge 465,866.66 * 104m3 / a . Total excretion 535356.34 * 104m3 / a, the amount of storage of the aquifer to release 69491.55 * 104m3 / a. Study area, the main recharge from rainfall recharge , recharge 456,903.71 * 104m3 / a 98.08% of the total groundwater recharge , the main channel of excretion diving evaporation , evaporation excretion 428,492.12 * 104m3 / a , the total excretion 80.04% . Level excretion elements artisanal mining and border excretion, or 11.21% of the total excretion . However , due to the larger research area 1000m * 1000m grid resolution to split the study area , the results are more rough , you can not fine carved of groundwater flow field , used to correct simulation of water level observation wells it appears relatively tasteless. Therefore need to characterize the study area is finer , and this increases the processor 's calculation of the amount of the processor 's performance is also higher . Today, however, a single processor can not avoid performance bottlenecks due to the restrictions on the processor , memory , and data channel system has encountered . Therefore, this study uses a parallel approach to the use of multi-core processors for solving the three - dimensional flow equations . The the serial version MODFLOW2000 source automatically generated flow equation coefficient matrix based on the OpenMP Application Programming Interface parallelization of the preconditioned conjugate gradient (PCG) iterative method for solving the flow equation . The the serial environment model simulation calculated simulation time of 20.705s, while in a parallel environment 1.05-1.11 times faster than analog computing time . Obviously , through the design of parallel programs can be a good resolve single processor performance bottlenecks , and improve the efficiency of the calculation .
|
Related Dissertations
- Research and Design of a High-Performance Scalable Public Key Cryptographic Coprocessor,TN918.1
- Research on Video Compression Algorithm Based on Multi-core Computing Platform,TN919.81
- The Algorithm Researches of Novel Wide Area Backup Protection for Power Grid,TM774
- Fault Tolerance for MapReduce in the Cloud Environment,TP302.8
- Image retrieval method and system for parallel computing,TP391.3
- GPU-accelerated particle filter PET image reconstruction algorithm,TP391.41
- CPU-based inverse algorithm source strength,TP18
- Large-scale approximation paragraph fingerprint - based page detection algorithm research,TP393.092
- Research on Fault-Tolerant Parallel Skyline Query Technology in Cloud Computing Environment,TP311.13
- Paralleling Genetic Annealing Algorithm with OpenMP,TP301.6
- Algorithm Study on Accelerate CV Image Segmentation and Exterior Industrial Image Reconstruction by CUDA,TP391.41
- FCM combines multi-resolution image cone multicore parallel image segmentation algorithm,TP391.41
- Application of Parallel FDTD and MPSTD Algorithm in EM Scattering,O441.4
- BSD protocol stack performance analysis and optimization of parallel,TP393.04
- The Design and Implementation of Intrusion Detection System for High-Speed Network,TP393.08
- Parallel Topological Analysis of Two-dimensional Flow Field Based on Physical Feature,TP391.41
- Software optimization techniques in software development projects,TP311.52
- Research on Transposing Matrices Using In-Place Method,TP301.6
- The Research of Three-dimensional FDTD Parallel Algorithm Based on MPI and OpenMP,TN011
- H.264/AVC Based Video Encoder Design and Optimization,TN762
- Application Research for Optimization Design of GM Detector with Mento Carlo Method,O242.2
CLC: > Industrial Technology > Hydraulic Engineering > The basic science of water resources project > Engineering and hydrology
© 2012 www.DissertationTopic.Net Mobile
|