Please use this identifier to cite or link to this item: http://hdl.handle.net/10125/58840

Management-Oriented Modeling: Optimizing Nitrogen Management using Computerized Artificial Intelligence

File Size Format  
Li.pdf 5.78 MB Adobe PDF View/Open

Item Summary

Title:Management-Oriented Modeling: Optimizing Nitrogen Management using Computerized Artificial Intelligence
Authors:Li, Mengbo
Date Issued:1997
Abstract:Increasing nitrate levels in groundwater have caused growing public health concern in recent years. This has prompted research on precision nitrogen management to understand and control nitrogen impact on the environment. Many nitrogen (N) models have been developed to describe the N status and behavior in soil-plant systems, but they are uniformly weak in finding optimal management strategies. To model nitrogen management, Management-Oriented Modeling (MOM), a dynamic simulation model using artificial intelligence (AI) optimization techniques, was developed in this study. MOM was designed as a tool to find optimal solutions for N management to minimize nitrate leaching and maximize production and profits.
MOM consists of a generator, a simulator, and an evaluator. In searching for optimal management strategies, the generator produces a group of nodes (management choices). The evaluator uses the built-in knowledge and communication with users to analyze the outputs of the simulator and to guide the generator’s work. A mixed search method that combines hill-climbing method for a global, strategic search with best-first method for a local, tactical search was developed to find the shortest path from start nodes to goals. In this manner, MOM searches for user-weighted goals by simulating the N cycle and management effects on the fate of N in a soil-plant system. In addition to general simulation and evaluation of N fertilization, MOM provides real time decision-aid for within-season management. MOM-guided within-season management uses weather forecasting to estimate rainfall in the near future and simulates the consequences in soil-plant systems. It gives users daily “snapshots” of the N status in soil-plant systems without within-season sampling and testing. Scenarios show that MOM can provide precision nitrogen management that maximizes profits and yields while minimizing nitrate leaching by updating management of irrigation and fertilization within-season. MOM-guided within-season management is a precision tool with high efficiency, low cost and “transparency” for nitrogen management. MOM simulator was evaluated with 11 datasets from Hawaii and Brazil. Calibration and validation results suggest that the model prediction accuracy was acceptable for the field N management.
URI/DOI:http://hdl.handle.net/10125/58840
Appears in Collections: Ph.D. - Agronomy and Soil Science


Please email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.

Items in ScholarSpace are protected by copyright, with all rights reserved, unless otherwise indicated.