In this ongoing research, a variety of sustainable energy sources, decentralized power generation entities, energy storage solutions, and hybrid electric vehicles with plug-in capability are taken into account. The aim is to present a dependable framework for managing large-scale energy, applicable to both isolated and network-connected modes of renewable hybrid microgrids (HMGs). This framework employs a novel adaptation of Cultural Algorithms (CA) with a two-step modification to minimize the operational expenses of the network. Additionally, it incorporates an innovative intrusion detection system (IDS) based on a deep learning model. The IDS is designed to identify cyber threats related to identity, such as Sybil attacks and masquerading attacks, within the cyber layer. The proposed IDS utilizes the real-time data gathered from the digital twin of the HMG and then uses the LSTM model to differentiate between various signal sources and identify instances of cyberattacks. The data transaction security over the messaging is first preserved by the blockchain technology and later is assessed by the proposed IDS system. To assess the reliability and effectiveness of the proposed framework, a real-time hybrid microgrid is simulated partly in physical layer and partly by digital twin of renewable units to check the model performance.