Grid Computing Methodology In Large-Sgale Power System

The power system in China is now under the way of distensible scale and complicated model. Some bottle-neck problems, such as shortage of memory and slow-footed convergence, will be come forth if we adopt traditional means to analyze the large-scale power system. Further more grid computing can exert the advantage of multi computer and accomplish the job with lower cost and faster speed which would be finished by super computer previously. This dissertation focuses on the study of grid computing methodology in large-scale power system and the innovation of task scheduling and resource distributing. Moreover we apply the innovative algorithm to power system and significant strategy will be put forward.As the key, task scheduling and resource allocating are the bottle-neck of grid which slowed the computing speed and efficiency. A new grid task scheduling algorithm (SMM: Stable Matching Mechanism) based on matching problem is proposed and the quantification matching method founded on each condition value and its proportion value is put forward. The user and the resource can dynamically preset the benchmark values of each condition follow along with its character and obtain the matching degree via the comparison with the actual values. According to the rank of the matching degree, SMM can circularly match until stabilization. The demand of both the users and the resource is considered in SMM and the relevant stable matching scheduling strategy can be obtained. The simulation demonstrates that SMM can implement grid task scheduling effectively and a strategy will be provided to the scheduling centre.The heterogeneous resource is the difficulty of grid. A new greedy algorithm based on benefit first of grid resource distributing (BFG: Benefit First Greedy Algorithm) is proposed and the construction of benefit function founded on the proportion value of time and cost is put forward. Taking benefit function as the object of resource distributing, BFG can select the maximum of benefit function via greedy algorithm to achieve the best benefit.To overcome the disadvantage in centralized processor of power system, a scheme of grid application in power system is brought forward. The span-new subject has the character of transplantable and patulous. It can solve the inextricable problem in power system all the while, and is the indispensably important tool for analyzing and computing in power system.The original aspect of this paper is that a concept of the application of grid in power system is put forward. The object is study of task scheduling and resource distributing in power grid. The simulation in Matlab demonstrates that these algorithms will implement grid task scheduling and resource distributing effectively. Ultimately we make use of SMM and BFG in power system and describe significant blue print of the application in power grid. Finally, this paper summarizes the important contents and conclusions, and points out development research in the future.