Identity preservation of grain in elevators.

摘要

Grain commingling is the process brought about by handling different types or varieties of grain. The increase in the use of specialty grain and the proliferation of genetically modified grain have heightened the need for identity preservation in grain handling industry. Development of a rule-based decision support system is a vital management tool that would facilitate programs designed to address commingling and ensure identity preservation of grain during grain handling. This research was conducted to: (1) characterize the impact of handling equipment and facility design on grain commingling during grain transfer; and (2) develop a decision support system for identity preservation of grain in elevators.;Experiments were conducted at the elevator facility of the USDA-ARS, Grain Marketing and Production Research Center to determine the effects of handling equipment on grain commingling. Commingling was defined as the amount of grain from a specific load mixed into the next load during the grain transfer process. Commingling was highest at the grain cleaned at 0.24%, followed by inline scale at 0.22%, combined pit and bucket elevator at 0.15%, and grain scalper at 0.01%. At grain flow rates of 43 to 50 t/h, commingling levels approached the initial impurity level after moving 2 to 3 t of grain.;Experiments were also conducted at a country elevator to quantify the effects of the receiving configuration on grain commingling. The gravity-type pit and bucket elevator produced the highest commingling of 1.25%. The effect of using the same leg and pit with drag conveyor resulted in 0.30%, of which 0.23% was from the effect of the leg. Simulation results using SIMIAN/ARENA showed that for a 10 t load, commingling was about 0.28% for the combined effect of leg and pit with drag conveyor and 0.27% for the effect of leg alone. Predicted values revealed that 50:50 load ratio combination generated the highest potential of grain mixing.;Commingling Advisor (COMMA), a rule-based decision support system, was developed to assist grain handlers in determining the possible commingling during grain transfer and obtaining alternatives in handling multiple-type grain to ensure purity. The domain knowledge was built based on the two experiments, previous studies on handling operations and processing times, and information gathered from literatures and interviews, while the inference engine was implemented using Java Expert System Shell (JESS). The system was verified by comparing the commingling values with actual field test data.*.;*This dissertation is a compound document (contains both a paper copy and a CD as part of the dissertation).