A new approach to rotational speed control structures based on an optimized intelligent recurrent wavelet-based Elman neural network (RWENN) controller used for the integration of offshore wind and wave energy conversion systems driven by a doubly fed induction generator. The nodes connecting the weights of the RWENN are trained online using a backpropagation method. A modified gravitational search algorithm (MGSA) is developed to adjust the learning rates and improve learning capability. The proposed control scheme has improved the real power regulation and dynamic performance of a combined wind and ocean wave energy scheme over a wide range of operating conditions. The performance of this control scheme is assessed by comparing it to a traditional proportional-integral based control scheme in a series of case studies representative of maximum power generation. Simulations are carried out using PSCAD/EMTDC software to verify the robustness of the power electronics converters and the efficiency of the proposed controller under steady state and transient conditions. (C) 2018 Elsevier Ltd. All rights reserved.