Blockchain Enabled Intelligence of Federated Systems (BELIEFS): An attack-tolerant trustable distributed intelligence paradigm

In this article, a Blockchain Enabled Intelligence of Federated Systems (BELIEFS) is proposed to conduct cooperative control for the multi-regional large-scale power system with multi-agents (MAS). By establishing two levels blockchain, each regional AI agent can simultaneously manage intra-regional controllers and cooperate other agents. Under consensus mechanism, agents, which respectively conducted distributed deep reinforcement learning (DDRL) algorithm in multi-regions, have tolerant capability malicious attacks their training process. The demonstration approach within interconnected system. mode “centralized dispatching hierarchical management”, article aims definite mathematical model deal problem systems. With comparison experiments, effectiveness efficiency our method process are verified. addition, set on main chain shard chains verify attack-tolerant capability. We expect that such results suggest new paradigm trustable deployment.