Research On The Grid Resource Scheduling Mechanism For The Power Flow Computation

All aspects of power system have characteristics such as a wide resource distribution, large amount of calculation and the high requirements for computing, that has a complementary links with the idea of grid design. In particular, as the promotion of constructing a modern power system information platform, the existing department including power dispatch, safety production and management need to be integrated urgently, and the demand of distributed computing and real-time electricity monitoring need to be satisfied too. The integration of grid technology becomes a trend, and its related technologies are under study currently as the focus of attention.In this paper, the characteristics, architecture and the key technologies required to application development of the grid system is discussed firstly, the design concept and functional differences of several architectures are compared, and the grid resource scheduling technology and development environment is carried out some detailed analysis. Secondly, combining with the current problems in the power flow calculation and based on the grid technology, the electricity demand is analyzed, a grid platform based parallel load flow algorithm is build up, and take a detailed design of the base model and middle ware. A corresponding configuration is carried for operating environment.Finally, based on the smart particle swarm optimization algorithm and careful consideration of many cardinal variables that affect the grid scheduling, a scheduling algorithm in grid computing environment is proposed, which introducing the crossover and mutation of the genetic selection algorithm and dynamic parameter auto-adapted adjustment method. The simulation and performance evaluation are taken for the improved algorithm. Simulation results show the algorithm can speed up the iteration rate, reduce the probability of falling into local optimum, get better optimization results and significantly improve the computational efficiency of the system. The effect is obvious in solve problems which have complex feature, large amount of calculation and many influence factors.