Seite - 185 - in Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources
Bild der Seite - 185 -
Text der Seite - 185 -
formance of the distributed microwave heating system can be sig-
nificantly improved by the advanced MIMO control system. The
global temperature window (temperature window of the entire
heat load) achieved by the RLC system (5◦C - 6.5◦C) is also com-
parabletonormalcommercial industrialheatingequipments,such
as thestandard in [Ind13](±2.5◦C).
• The control concept that combines the real time thermal picture
and the intelligent control system (Q(λ) controller) is unique. It is
able to control the complete temperature distribution of the entire
workpiece, and more straightforward than conventional methods
basedonindividual temperaturevalues. Itseffectivenesshasbeen
provedbyexperimental results.
• The temperature control system and software developed in this
dissertation have a great potential to be implemented in practical
industrial applications, to promote the use of microwave heating
to broader areas. They can also be transferred to other distributed
microwavefeedingsystems,suchastheoctagonalmicrowavecav-
ity (proposedin[LLHG14])andotherrelatedapplications.
Based on the work done in this dissertation, the MIMO temperature
control system could be further optimized in order to improve its re-
liability and control performance. For example, more experiment pa-
rameters could be monitored and taken into account in the modeling
process, e.g. the air flow speed in the oven and the real-time tem-
perature of the metal table. The involvement of such parameters will
increasetheaccuracyof thecontrolmodel,aswellas thereusabilityof
the controller. With a high reusability, control performance could be
improved by reusing the former well-trained controller (such as the
results of repeated trials shown in figures 5.34 and 5.37), especially
for the reinforcement learning controller. In this case, a more power-
fulreinforcementlearningcontrollercouldbebuiltbasedonacompli-
cated function approximator (nonlinear function or neural network)
and long time practical experiment training. Another research direc-
tionwouldbethecombinationofmicrowaveandhotairheating. The
hot air flow could be controlled using an independent PID controller
toheat thecoldareaof the load,whichwillbeofgreathelptoachieve
amorehomogeneoustemperaturedistributionbetweentheinnercore
andsurfaceof theentire load.
185
Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources
- Titel
- Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources
- Autor
- Yiming Sun
- Verlag
- KIT Scientific Publishing
- Ort
- Karlsruhe
- Datum
- 2016
- Sprache
- englisch
- Lizenz
- CC BY-SA 3.0
- ISBN
- 978-3-7315-0467-2
- Abmessungen
- 14.8 x 21.0 cm
- Seiten
- 260
- Schlagwörter
- Mikrowellenerwärmung, Mehrgrößenregelung, Modellprädiktive Regelung, Künstliches neuronales Netz, Bestärkendes Lernenmicrowave heating, multiple-input multiple-output (MIMO), model predictive control (MPC), neural network, reinforcement learning
- Kategorie
- Technik