Page - 154 - in Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources
Image of the Page - 154 -
Text of the Page - 154 -
5. ExperimentalResults
hand, the EKF based NN approach has a much smaller MSE than the
nonlinear EKF algorithm in the validation part, which indicates the
black-box approach is a more powerful tool to model systems with
unknowndynamics.
5.3. ResultsofDifferentControlMethods
The performance of different control methods is the most important
result of this dissertation. In this section, the control performance of
all aforementioned control methods is presented, with respect to dif-
ferent setups. Before the experimental results of temperature control
are demonstrated, an important question that has to be discussed is
thecontrollabilityof thesystem.
Controllability isan importantpropertyofasystem. Accordingto the
definition in [Dor95], a system is controllable if any initial state of the
system can be moved to any other state in a finite time interval us-
ing the external input. Take HEPHAISTOS for example, it is control-
lable if any desired temperature distributions can be obtained within
a finite time interval using the control input from the initial tempera-
ture distribution. Compared with this complete controllability, what
ismoreimportant totheheatingapplications is thereachabilityofcer-
tain states in HEPHAISTOS. A particular state of the system is reach-
able if any initial states can be transfered to this state within a finite
time intervals using a corresponding control input sequence [Rug96].
In HEPHAISTOS, the control task is to achieve a homogeneous tem-
perature distribution for the whole workpiece, therefore the reacha-
bilityof thehomogeneoustemperaturedistributionbecomesthemost
importantpropertyof thesystem.
Although it has been proved in [WYT12] that any arbitrary tempera-
turedistributioncouldbeachievedbyacorrespondingEMfielddistri-
bution in microwave heating, it neglected that not all EM field distri-
butions can be realized within a microwave cavity. Moreover, the EM
fielddistributionofHEPHAISTOSissofarnotpossibletobeobtained
in neither analytical nor numerical ways. Due to these reasons, it is
more realistic to analyze the controllability quantitively via a number
of realexperiments.
154
back to the
book Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources"
Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources
- Title
- Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources
- Author
- Yiming Sun
- Publisher
- KIT Scientific Publishing
- Location
- Karlsruhe
- Date
- 2016
- Language
- English
- License
- CC BY-SA 3.0
- ISBN
- 978-3-7315-0467-2
- Size
- 14.8 x 21.0 cm
- Pages
- 260
- Keywords
- Mikrowellenerwärmung, Mehrgrößenregelung, Modellprädiktive Regelung, Künstliches neuronales Netz, Bestärkendes Lernenmicrowave heating, multiple-input multiple-output (MIMO), model predictive control (MPC), neural network, reinforcement learning
- Category
- Technik