A thermostatic control strategy based on multi-sensor data fusion and fuzzy-PID method

The steady temperature is vital to organ-saving out of body in a hypothermic machine perfusion (HMP) system. A thermostatic control strategy based on multi-sensor data fusion and fuzzy-PID method is proposed in this paper to improve the accuracy. Firstly, the basic frame of HMP system and the installation of sensors are expounded. Then the data fusion based on modified Bayes estimation is carried out to weaken the possible measurement error, resulting from the sensor faults and noise interference. Specially, the better error-recovery ability is proved in the cascaded Bayes algorithm. Secondly, the fuzzy and the fuzzy-PID (Proportion Integration Differentiation) controller are adopted respectively according to the difference of temperature-deviation. Here the former is designed to offer the control variation of compressor needed while the latter is to gain three control coefficients of PID algorithm. The dynamic and static tests indicate that the thermostatic control result meets the need of patients although it is also affected by some extra factors, such as the external temperature, flow speed of solution and working modes.