Seite - 6 - in Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources

1. Introduction heatingratesaswellas thetemperaturedistributionof theworkpiece. On the other hand, the temperature distribution also affects the prop- erties of the workpiece, such as thermal conductivity and dielectric permittivity, which is reciprocated to change the EM field distribu- tion. An uniform temperature distribution of microwave heating re- quires optimizations from a number of different parts, including the design of the microwave applicator, adjustment of microwave power, temperaturemeasurementapproachandothers. 1.2. Overviewof theStateof theArt In general, there are three different types of approaches to control the temperature and improve the temperature homogeneity under mi- crowaveheating. 1.2.1. Moderncontrolmethods The idea of using modern control methods to control the heat trans- fer process in microwave heating started from the 1980s. In the be- ginning it was used simply as an approach to keep the thermal sta- bility and prevent the thermal runaway phenomenon. In simple sce- narios,e.g. one-dimensionalslabsorsingle-modemicrowavecavities, the two coupling PDEs (Maxwell’s equations and heat transfer equa- tion) can be solved analytically, and the required microwave power is directly calculated based on the analytical solution, such as the idea usedin[Smy90]. But thiscontrolschemeisonlyvalidforsimpleheat- ingscenarios,anditalsoassumespropertiesoftheheatedmaterialare perfectlyknowninadvance,whichdoesnotholdinmostcases. Besides the analytical control solution method, another more general control approach, which is the model based temperature control, was proposed in [RMTV85]. A transfer function [Dor95] was used to de- scribe the relationship between one measured temperature and the microwave power of one source. A simple single-input single-output (SISO)feedbackcontrollerwasconstructedtocontrol thepowerofthe 6