A reliable and robust communication network is essential to exchange information between distributed generators (DGs) and accurately calculate their control actions in microgrids (MGs). However, the integration of the communication network and MGs poses challenges related to the flexibility, availability, and reliability of the system. Furthermore, random communication disorders such as time delays and packet loss can negatively impact the system performance. Therefore, it is essential to design a suitable secondary controller (SC) with a fast dynamic response to restore voltage and appropriate power-sharing, while ensuring that the effects of random communication disorders are eliminated. In this regard, an optimal distributed hybrid model predictive secondary control method is presented in this paper. Realistic simulations are carried out in a mixed simulation environment based on MATLAB and OMNET++, by considering IEEE 802.11 (WiFi) using the recently developed Internet networking (INET) framework. In the implemented application layer, the recoveryUnit is responsible for reducing the impact of random communication disorders. The effectiveness and performance of the proposed method in comparison with a conventional model predictive control are verified by simulation results.