Adaptive Decoupling Control and Energy Efficient System for Meat Drying Processing based on Neuro-Fuzzy Approaches

Relative humidity is an important control parameter in food manufacturing. The performance of relative humidity control can directly impact product quality and energy consumption. There are weaknesses in the current drying process control, such as the stability of humidity measurement for a smoked drying environment and the coupling issue of temperature and relative humidity in a drying room control system. The objective of this research is to develop an effective measurement and drying process control system with the aid of advanced control technology for food safety and energy consumption. The distinct feature of this research is the combination of theory and practice. The research is directed towards the real issue of the drying process in meat manufacturing. All information regarding the measurement data and system characteristics included in this thesis were obtained from current drying rooms. The conclusions of the research can easily be applied in current drying room process control systems. Several topics are discussed in the thesis. An adapted psychrometer is proposed for a smoked meat drying environment. The measurement accuracy of the proposed method is evaluated in an actual smoked drying environment. The coupling relationship between temperature and relative humidity and its consequences are investigated based on the measured data from a current meat drying room. The coupling model of temperature and relative humidity for the meat drying room is proposed based on fuzzy logic and neuro-fuzzy algorithms. A neuro-fuzzy based control method is proposed for improving the tracking performance of the drying room control system. Experiments are performed to compare the control performance of the proposed decoupling methods to existing systems. The reduction in energy consumption with decoupling temperature and relative humidity in the meat drying process is confirmed based on the real-time measured results.