Design and optimization of a natural gas-fired thermoelectric generator by computational fluid dynamics modeling. (1st October 2017)

• Record Type: Journal Article
• Authors: Bargiel, Paweł; Kostowski, Wojciech; Klimanek, Adam; Górny, Krzysztof
• Is Part Of: Energy conversion and management. Volume 149(2017)
• Page Start: 1037
• Page End: 1047
• Publication Date: 2017-10-01
• Subjects: Thermoelectricity -- Heat transfer -- Numerical modeling -- Natural gas -- Island mode supply -- CFD -- Optimization; Direct energy conversion -- Periodicals; Energy storage -- Periodicals; Energy transfer -- Periodicals; Énergie -- Conversion directe -- Périodiques; Direct energy conversion; Periodicals; 621.3105
• DOI: 10.1016/j.enconman.2017.05.030
• Languages: English
• ISSNs: 0196-8904

Abstract

Highlights: A natural gas fired thermoelectric generator with 50 W output power is designed. Vertical layout with three functional regions: heating, combustion and active. Original, simplified approach for modeling the global effects of combustion. Five final design variables for optimization (max. output power). Three candidate points on the response surface: simulated output 49.0–49.6 W. Abstract: The paper reports the results of the computational part of an ongoing research and development project aimed at designing and optimization of a thermoelectric generator fired with natural gas. The idea of the project is to provide an independent and reliable source of power in the power range of 50–100 W for island-mode supply, tailored for remote objects of the natural gas distribution infrastructure which, for economic reason, are not connected to the electric grid. The device includes a burner as a heat source, flow channels for air and for exhaust gases and a set of thermoelectric modules generating electric current if subject to temperature difference. The pre-selected design configuration (type E) was subject to parametric optimization. A corresponding numerical model based on the Ansys/Fluent computational fluid dynamics code was built. The optimization procedure aims to maximize the temperature difference across the thermoelectric module in order to maximize the generated power output. Simultaneously, it is essential to maintain the hot-side temperature below the required technical limit of 320 °C. First, thirteen design parameters (most of them concerning geometry) influencing the target function were identified. A central value and a possible range for each parameter have been set. Next, the partial impact of each parameter on the target function was found, and the set of the design parameters was limited to eight. Finally, only five parameters were estimated to be non-monotonically related to the generated power. The impact of the selected five parameters on the target function was investigated using a central composite design optimization plan. The set of points obtained in the multi-dimensional response surface was then limited to points meeting the imposed limitation of maximum temperature. Eventually, three candidate points (i.e. three sets of design parameters) have been proposed for the next stage of the project involving an experimental set-up for the design of the prototype.