Seite - 133 - in Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources

5. ExperimentalResults In this chapter, the results of practical experiments will be presented. In the beginning, a number of preliminary trials are performed in or- der toverifythevalidityof the linearandthenonlinearmodels (given bytheequation 3.41and 3.46respectively). Thenthesystemidentifi- cationalgorithmsintroducedinchapter 3aretestedandcompared, to identify themostappropriatesystemidentificationmethodforpracti- cal heating experiments. In the end, both the adaptive and the intelli- gent control methods are applied, and the corresponding experimen- tal resultsarepresentedandcompared. 5.1. VerificationinHEPHAISTOS Due to its complicated structure, some fundamental characteristics of HEPHAISTOS are difficult to determine theoretically, e.g. the proper- ties of the EM field distributions. In order to understand the capabili- tiesandlimitsoftheHEPHAISTOSsystembetter,anumberofprelim- inary tests have been done with respect to one fundamental question, which is thepracticalpowersuperpositionprinciple. According to the introductions in chapter 3, there are two different powersuperpositionprinciples,definedasthescalaradditionandthe vector addition principles. These two principles are the basis of the linear 3.41 and the nonlinear 3.46 state-space models, respectively, which are also the foundations of the entire MPC system. From this pointofview,toverifythepowersuperpositionprincipleisequivalent to theverificationofvaliditiesof these twogrey-boxmodels. The entire verification procedures contain three aspects. The first as- pect is to check if each individual feeding source can provide a sta- ble heating pattern. For each microwave source of HEPHAISTOS, 133