2010 Annual Report
1a.Objectives (from AD-416)
Existing packing factor data are of unknown reliability and are widely mistrusted in the industry. Accurate data are required for government-mandated inventory control and are a crucial component of new quality management systems being developed to enable source verification in the grain handling industry. The current Farm Bill requires the Risk Management Agency (RMA) to determine the efficacy and accuracy of current pack factors and, as a result, they desire ARS to evaluate their existing packing factor data.
The new data and model developed in this research will improve the scientific basis for predicting pack factor in stored grain. We will define, for the first time, uncertainty in predicted pack factors from the old method as well as from the new model. We will produce a user-friendly, windows-based software that can be used by farmers, elevator managers, and government officials. The software will allow the user to enter needed measurements and materials for the bin and quality factors for the stored grain. This tool will calculate the average pack factor for the bin and will provide accurate estimates of the confidence intervals for those pack factors.
The objective of the project is to refine and validate a procedure with known accuracy, based on measurable physical parameters, for determining the packing of grains within upright storage structures. Factors identified for the study are:
1. structural shape and size
2. bin wall type
3. type of grain
4. time in storage
5. the impact of facility aeration systems
6. bulk density (test weight) of the incoming grain
7. moisture content of the grain
8. additional factors such as broken material and fines in the grain
1b.Approach (from AD-416)
The major variables affecting stored grain packing are grain type, moisture content, test weight, and bin geometry and dimensions. Variation across different regions of the U.S. must also be investigated as well as other minor factors. In order to avoid the excessive cost from experimentally determining pack factors for all grains under all conditions, we plan to use science-based modeling to reduce the total amount of data required to achieve valid results. Physical properties will be measured in the laboratory to use as inputs for modeling. A preliminary model for determining pack factor for a wide range of grains and bins has been developed and is currently being calibrated in limited experiments. We will calibrate and validate this model by measuring pack for selected grains in bins spread over the major grain producing regions of the U.S. Calibrating the model instead of developing pack factors from field measurements alone will allow us to reduce the number of bins measured from tens of thousands to several hundred. Confidence intervals will be established from the field measurements and used to characterize the predictions of the new model and will be compared to confidence intervals determined for the old method.
Field exercises were conducted at the University of Kentucky Animal Research Center and the USDA ARS research facility in Manhattan, Kansas. Five metal bins ranging in size from 10,000 bu to 27,000 bu and containing corn or wheat were measured in Kentucky. The weight, moisture content, and test weight of every load in and out was obtained from grain tickets to establish the true inventory. Inventory was determined following National Crop Insurance Services (NCIS) and Farm Service Agency (FSA) protocols by four trained personnel. The measurements showed a general trend of larger absolute differences for larger grain volumes and larger percentage differences for smaller grain volumes. In the bins with normal loading, the differences between measurements by different personnel was small in one case (1%), but fairly large (about 8%) in the other two bins. The average difference due to measurement with the cone versus the leveled measurement was less than 2% on three bins with normal loading.
Twenty-four popular HRW wheat variety samples from the leading HRW wheat producing states have been delivered to the University of Kentucky Granular Mechanics Laboratory to undergo compressibility tests. The selected varieties have been prioritized with a set of varieties that minimally covers all the fundamental variables and composite sample types for an initial analysis. The computer model for predicting compaction of granular materials stored in bins was modified with a more flexible system for entering surface conditions to determine cone volume. In addition, work has begun on porting the entire computer model code to Visual Basic.