This paper proposes a combined model of the multi-microgrid optimal operational problem and the peer-to-peer (P2P) energy trading model with the integration of an energy management scheme for the multi-microgrid communities. The primary objective of the proposed model is to maximize the financial benefits and minimize the imbalance between the total power generation and load demand of the multi-microgrid system. In this model, the daily energy operation problem of each microgrid in the multi-microgrid community is formulated as a mixed integer linear programming problem and solved by using the MILP solver in the MATLAB toolbox. The energy management scheme is also integrated into the daily operation problem to maximize the local renewable energy utilization. After local microgrid optimization, the optimal power values (either surplus or deficit) of each microgrid are transferred to the P2P energy trading process as input data. The sellers and buyers in the trading process can be determined based on the availability of net energy. An efficient hybrid deep learning model, Bidirectional Long Short-Term Memory with Gated Recurrent Units (BiLSTM-GRU) has also been considered in this paper to predict the uncertain input parameters such as solar output power, wind output power, and load demand values. Finally, the blockchain technology concept is incorporated into the proposed model to improve the security and trust levels in the trading process without any involvement of a third-party entity. A multi-microgrid community test system consisting of five different microgrids is considered to assess the performance and effectiveness of the proposed model. The simulation results demonstrate that the proposed BiLSTM–GRU model offers significantly more accurate predictions of renewable energy resource (RER) parameters and load demand, outperforming the benchmark LSTM–GRU model across all performance metrics. The integration of P2P energy trading process with an energy management scheme reduces the total daily operational cost by 19.51 %, daily curtailed load by 23.54 %, and unutilized RER energy by 18.36 %. Furthermore, increasing the level of participation in the energy management scheme leads to additional significant reductions in daily operational cost, curtailed load, and unutilized RER energy. Finally, it is concluded that the proposed optimal energy management model offers both economic and technical advantages for multi-microgrid systems.