13-5 |
Improving food safety by fault detection and diagnosis in batch retort operations |
X. Dong, Department of Agricultural and Biological Engineering, Purdue University, 1146 Agricultural and Biological Engineering Building, West Lafayette, IN 47907 and T. A. HALEY, Department of Food Science, Purdue University, 1160 Food Science Bldg, West Lafayette, IN 47907. Batch food sterilization carried out in static retorts heated with steam is widely used for processing canned foods. Computer-based intelligent on-line control systems have been developed to accommodate process temperature deviations by calculating accumulated F values in real-time. However, sensor reliability is essential for accurate feedback control and numeric calculation of sterility. When a temperature sensor fault occurs, small deviations or drift in true process temperature will be hidden by the feedback actions and remain unnoticed until it's too late to take corrective action, Fault detection and diagnosis (FDD) based on analytical redundancy makes use of intrinsic system dynamics to detect sensor or component faults that are difficult or impossible to be detected by the operator. The objective was to develop FDD algorithms using analytical redundancy to detect and accommodate temperature sensor fault in both traditional and intelligent thermal processing, and therefore to ensure food safety. A detailed dynamic model for batch retort operation was developed, and different levels of functional redundancy were used to detect sensor fault and other process abnormalities. When a temperature sensor fault was detected and identified (determining fault type and magnitude), the temperature output was corrected, or replaced by model estimation. Simulation results showed that different sensor faults can be detected, and the processing can be continued under faulty situation using an estimated temperature as the feedback signal. The proposed FDD system can be implemented on-line without need of extra hardware investment. It can ensure food safety, improve record-keeping and assist in equipment maintenance.
Session 13, Food Engineering: Thermal processes
|