In DBA, none of the nodes knows the total demand, yet the demand and supply balance is guaranteed by the algorithm. No master or leader node aware of the total power demand is needed in DBA, whereas such a node is required in. In, each node needs to know some parameters of all other nodes, which implies that the computation and communication package size grow at least linearly with the network size, while in DBA each node only needs to know its local parameters. In, a distributed algorithm is proposed, but global parameters including network topology and generators’ parameters are needed to design an appropriate learning gain to guarantee convergence. Compared with other algorithms, DBA has the following features.ĭBA requires no global information of the system. In this chapter, a distributed bisection algorithm (DBA) based on a consensus-like iterative method is presented to solve the EDP. In, the authors propose an algorithm for the EDP with a quadratic cost function, which can be treated as a distributed implementation of the standard Lambda-iteration method, without requiring other nodes’ parameters. In, an algorithm based on a consensus and innovation framework is proposed, where the consensus term makes all the nodes agree with each other to realize their common goal of estimating the global price index, while the innovation term makes all the nodes estimate the index according to the local knowledge of loads. In, the authors present a ratio consensus-based decentralized algorithm to find the optimal incremental cost, under the assumption that each node (i.e., generator) knows the parameters of all the nodes. In, the authors propose a consensus-based algorithm to realize decentralized economic dispatch, where a master node aware of the total power demand is required to ensure the equality between the total power supply and demand. In fact, a lot of such work has been done so far. In a smart grid integrating distributed generation, renewable power sources, and a communication network, it is desirable to solve the EDP in a distributed fashion. To meet environmental targets, to accommodate a greater emphasis on demand response, and to support plug-in hybrid electric vehicles, distributed generation, and storage capabilities, traditional power grids need to become “smart grids.” This is an area that has been heavily studied in recent years. In general, compared with centralized algorithms, distributed algorithms have many advantages, including enhanced robustness, reduction in communication between agents, and uniform power consumption for each agent. A lot of work has been done about distributed optimization using the distributed gradient method, distributed subgradient method, alternating direction method of multipliers (ADMM), and so on. A smart grid with distributed renewable power generation is a typical such large-scale system. Spatially distributed large-scale systems interconnected by a communication network are ubiquitous in the real world, where the traditional centralized control algorithms are inefficient. A parallel microgenetic algorithm is employed in to solve the ramp-rate constrained EDP with nonmonotonically and monotonically increasing incremental cost functions.ĭistributed algorithms for control, estimation, and optimization have been intensively investigated for large-scale systems. In, an algorithm based on evolutionary programming, tabu search, and quadratic programming methods are proposed to solve the nonconvex EDP. In, a strategy based on a direct search method with multilevel convergence is proposed to solve the EDP with transmission capacity constraints. In, the conventional Lagrangian relaxation approach and first-order gradient method are given. Many centralized solutions have been proposed to solve the EDP. A convex and piecewise linear cost function is used in, but a quadratic cost function is usually preferred. Many types of cost functions are available. In this scenario, the operation and planning for power generation systems can be done by one or several central decision makers. The classic EDP is mainly concerned with the economic dispatch of fossil-fired power generation systems to achieve minimum operational costs within capacity limits. This problem is usually formulated as an optimization problem. The economic dispatch problem (EDP) has been actively studied in the electric power industry for optimal operation and planning of energy resources. Zhiyun Lin, in Distributed Control Methods and Cyber Security Issues in Microgrids, 2020 1 Introduction