Research and Strengthening of Sensitive Power Grid

Based on distributed algorithm, in order to make it more effective, we hope to optimize certain steps of the algorithm by the introduction of PSO. PSO is one of the swarm intelligence algorithm that has been developed in recent years aiming at multi-objective optimization, which based on the hypothesis that every particle in the swarm could acquire information from its or other particle's origin experiences[1]. PSO possess high rate of convergence, and few parameters are required. When it applied to the decision of the leader in distributed algorithm, we hope to find one or severe proper leaders more quickly. The introduction of PSO under this circumstance will make the task easy to complete, therefore enhance the effective of distributed algorithm. The paper contains the principle and mathematical formulation of standardized PSO, and present program of PSO when applying to distributed algorithm for the sake of algorithm's nature testing. Introduction In the context of power systems, the trend is society's increasing demand for power supply performance, power users will pay more, will encounter a more time-varying load conditions, this requires network controller can effectively and quickly react to the changes anywhere in the network [2]. In the case of total output to meet demand, minimize cost, and economic factors are taken into account on the grid controller processing power requirements [3] is in the power system network, with new ways of distributing control, as interaction networks in power system communication method to improve performance of the electric power system. Distributed Algorithm Basic concepts of distributed algorithm and algorithm. Distributed algorithms are applied to systems network, distributed algorithm for network topology structure must be the basic factor for consideration. We use a simple matrix of the adjacent to describe the structure of the network, which is an NxN matrix. After various nodes of the network are numbered 1 to n and off-diagonal element represents the number of connections from node I to node j. In undirected networks presented in this chart, this matrix is a diagonal elements is 0 (0,1)-matrices, and is symmetric.Array L= [ i j l ]. In this array, i j l = ij i ja ≠ ∑ for the diagonal elements, i j l = ij a for non-diagonal elements. L matrix is the n n × matrix. For an undirected graph, L matrix is called the Laplacian matrix. Traditional control methods with Lagrange multipliers method to solve economic problems, based on the assumption that each motor has no power limit is reached, and each motor in the optimum operation point have the same marginal cost of IC. Consistency algorithm ensures that all variables are converging on a common set of values, thus the marginal cost of IC were selected as consistent variable [4]. Hypothesis generator unit cost of a quadratic equation: 2nd International Conference on Automation, Mechanical Control and Computational Engineering (AMCCE 2017) Copyright © 2017, the Authors. Published by Atlantis Press. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/). Advances in Engineering Research, volume 118