Please use this identifier to cite or link to this item:
http://hdl.handle.net/11375/26112
Title: | Development of a Cloud-Based Dual-Objective Nonlinear Programming Model for Irrigation Water Allocation |
Authors: | Yan, Zehao |
Advisor: | Baetz, Brian Li, Zoe |
Department: | Civil Engineering |
Keywords: | Uncertainty;Optimization model;Irrigation;Cloud model;Water resource;Agriculture management |
Publication Date: | 2020 |
Abstract: | Irrigation water allocation is essential to the management of agricultural water use in irrigation districts. Many irrigation optimization models were proposed from previous studies to provide decision support for water managers. In order to capture the complex nonlinear relationships and meet different water demands, more advanced multi-objective nonlinear programming models were developed in the past decade. However, it is still a challenging task to address varies uncertainties associated with irrigation optimization. Fuzzy programming, interval programming, and chance-constrained programming can be used to quantify uncertainties in simplified formats, but none of them can represent complex uncertainty in a composite format. In this thesis, a cloud-based dual objective nonlinear programming (CDONP) model is developed by implementing a cloud modeling method in an irrigation model to address the uncertainties of reference evapotranspiration (ET0) and surface water availability (SWA). The cloud modeling method is used to generate 2,000 data samples from historical data. The results show that the generated samples are consistent with historical data. Optimized allocation schemes are provided, and the performance of the CDONP model are discussed. This is the first Canadian study that used the cloud modeling method in irrigation water allocation. This method provides a solution to quantify composite uncertainties based on limited data, which represents a unique contribution to irrigation water allocation modeling. This study provides valuable decision support for agriculture management to improve water use efficiency. |
URI: | http://hdl.handle.net/11375/26112 |
Appears in Collections: | Open Access Dissertations and Theses |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
ZEHAO_YAN_2020-12-18_Master of Applied Science.pdf | Agricultural water management has become an essential problem in recent years due to the increasing water demands. Irrigation water resources allocation is a dynamic decision making process associated with various uncertainties, which often exist in a complex and composite format. In this study, a new uncertainty quantification technique, the cloud model, is introduced to a dual-objective nonlinear programming (DONP) framework, and a cloud-based dual-objective nonlinear programming (CDONP) model is developed to support irrigation water allocation and agricultural water planning under composite uncertainties. The cloud model is applied to address the complex composite uncertainties associated with reference evapotranspiration (ET0) and surface water availability (SWA). A case study of the Yingke irrigation district (YID) in Northwest China is conducted to demonstrate the applicability of the developed model. The results show that the net economic profit (ENP) and irrigation system efficiency (ISE) are influenced by ET0 more than SWA. The obtained results are also compared to those of a traditional dual-objective nonlinear programming model to illustrate the advantages of the proposed CDONP model. In addition, four water shortage scenarios are built and discussed for risk analysis. | 1.81 MB | Adobe PDF | View/Open |
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.