Seite - 11 - in Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources

1.3. Objectiveandcontent perature distributions for different heating setups or workpieces. In other words, HEPHAISTOS has the potential for systematically alter- ingheatingpatterns toachievethedesiredheatingratesandtempera- turedistributionsondifferent typesofheatedworkpieces. The key to realize this potentiality is to develop an innovative and effective power control system. By means of advanced modeling and control methods, the temperature changing trends of different parts of the workpiece can be predicted and the power allocated to each heating source can be adjusted accordingly, to guarantee the heating stabilityandenhance the temperaturehomogeneity. System Estimation and Identification Measure separate temperatures + ++ ++ + + + +++ T1(k) T2(k) ... Tn(k) Value/Action Function Approximation Abstract different kinds of information Tmean(k) Thigh(k) Tlow(k) Phigh(k) Plow(k) ... + Predictive/Adaptive Control (MPC, NNC) Intelligent Control (RLC) (a)Adaptive temperaturecontrol System Estimation and Identification Measure separate temperatures + ++ ++ + + + +++ T1(k) T2(k) ... Tn(k) Value/Action Function Approximation Abstract different kinds of information Tmean( ) high( ) Tlow( ) high Plow(k) ... + Predictive/Adaptive Control (MPC, NNC) Intelligent Control (RLC) (b) Intelligent temperaturecontrol Figure1.2. Diagramsof twodifferent temperaturecontrolapproaches. Due to the demands of different heating scenarios, two different con- trol strategies are developed in this dissertation (see figure 1.2). Both of these two strategies aim to control the heating rate and achieve a desired temperature distribution, but they control based on different types of models. In the first adaptive temperature control approach (figure 1.2a), themodelingisbasedontemperaturesmeasuredatsep- arate points. During the heating process a system model based on the temperature-power relation is recursively estimated, and the power 11