Consensus-Based Distributed Economic Dispatch with Optimized Transition Matrix

With an increasing penetration of distributed energy resources (DERs) in Energy Internet, there is a surge of interest in distributed economic dispatch (DED) problems. The existing literature mainly focuses on average consensus algorithms, the convergence of which remains a bottleneck in practical applications. To improve the convergence rate of DED, this paper proposes a consensus algorithm based on optimized transition matrix (TM) of Markov chain. A DED model is firstly developed under a consensus-based information exchange architecture, in which a generator only communicates with its neighbors. Thus DED is implemented in an iterative manner with generators exchanging information according to a predefined TM. In contrast to the averagely weighted TM, we minimize the Frobenius norm (F-norm) of the TM in order to limit the eigenvalues that greatly influence the convergence rate of DED. Additionally, a virtual power method is introduced to meet generators' power limits by reformulating the power allocation ratio. Case studies verify the effectiveness and efficiency of the proposed algorithm.