Application of decentralized fuzzy inference method for the inverse geometry design of radiative enclosures

The decentralized fuzzy inference method (DFIM) is applied to solve the inverse geometry design of a 2D radiative enclosure filled with gray participating medium. The heat transfer process in the irregular enclosure is solved using the discrete ordinate method in body-fitted coordinate. The simulated surface heat flux is employed as measurement signals. The design task is formulated as an inverse problem which is solved by the DFIM. Retrieval results illustrate that the design purpose of producing a uniform distribution of radiative heat flux on the design surface can be successfully satisfied using the present method. The DFIM is proved to be more efficient than other inverse techniques. Moreover, the effects of the number of measurement points and radiative properties of participating medium on the inverse design results are investigated. The DFIM is effective in designing the geometry of radiative enclosures for different kinds of participating media.