The smart morphing wing aircraft (SMWA) is a highly adaptable platform that can be widely used for intelligent warfare due to its real-time variable structure. The flexible conformal array (FCA) is a vital detection component of SMWA, when the deformation parameters of FCA are mismatched or array elements are mutually coupled, detection performance will be degraded. To overcome this problem and ensure robust beamforming for FCA, deviations in array control parameters (ACPs) and array perturbations, the effect of mutual coupling in addition to looking-direction errors should be considered. In this paper, we propose a robust adaptive beamforming (RAB) algorithm by reconstructing a multi-domain interference plus noise covariance matrix (INCM) and estimating steering vector (SV) for FCA. We first reconstruct the INCM using multi-domain processing, including ACP and angular domains. Then, SV estimation is executed through an optimization procedure. Experimental results have shown that the proposed beamformer outperforms existing beamformers in various mismatch conditions and harsh environments, such as high interference-to-noise ratios, and mutual coupling of antennas.