Control of hydrogen based metal hydride reactor coupled with thermoelectric module via neuro-fuzzy adaptive PID control

Thanana Nuchkrua*, Thananchai Leephakpreeda

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

Abstract

This paper presents a control scheme for improving the dynamic characteristic of the thermodynamics of hydrogen based metal hydride (MH) reactor by heating/cooling, thermal management, with a thermoelectric module. Normally, the uncertainty of the parameters obtained by mathematical model are deviated from the nominal parameters due to such as unexpected outside temperature of the reactor, or uncertainty of the time-delay of the reactor process. Moreover, the complicated problem of the integrated system model, thermoelectric module and MH reactor, make the system to be nonlinear which makes it hard to use the conventional controllers like proportional-integral-derivative, PID, or conventional fuzzy control to improve dynamic response. To obtain a high performance of those thermodynamics of hydrogen gas within the MH reactor, the Neuro-fuzzy network is employed to be control scheme for that thermodynamics non-linear system in real-time implementation. A back-propagation technique is applied to finely adapt the parameters of the Neuro-fuzzy PID. The results show the effectiveness of the proposed technique compared to conventional control approaches.

Original languageEnglish
StatePublished - 2013
Externally publishedYes
Event26th International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, ECOS 2013 - Guilin, China
Duration: 16 Jul 201319 Jul 2013

Conference

Conference26th International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, ECOS 2013
CountryChina
CityGuilin
Period16/07/1319/07/13

Keywords

  • Fuzzy system
  • Metal hydride
  • Thermoelectric module

Fingerprint Dive into the research topics of 'Control of hydrogen based metal hydride reactor coupled with thermoelectric module via neuro-fuzzy adaptive PID control'. Together they form a unique fingerprint.

Cite this