Consensus-Based Decentralized Optimization for Distributed Generators Power Allocation Over Time-Varying Digraphs in Microgrids

In this article, a consensus-based decentralized optimization method for distributed generators (DGs) power allocation is proposed to ensure the economic and reliable operation of microgrids. According to the method, all parameters are designed in a fully distributed fashion, which does not depend on any centers or leaders. To further achieve decentralization and to meet the flexible requirement of DGs, the strategy is adaptable to time-varying communication topologies (time-varying digraphs) and is proved to be convergent. During the communication process, the stochastic noise is introduced to imitate a nonideal communication environment, and the convergence rate is accelerated via coupling designed gains of noise and feedback (mismatch). Meanwhile, the cost of reactive power generation and reactive power compensation are considered for a comprehensive optimization, which guarantees the locally reactive power balance and reduces the cost. Simulations based on a test microgrid are used to verify the effectiveness of the proposed method. Finally, the dynamic characteristics and robustness are evaluated and compared by numerical simulation results.