The Secondary Voltage Control Partition And The Identify Of Central Node In Power Systerm Based On Cloud Theory

As the interconnection of foreign and domestic regional power grid, the construction of strong smart grid and also the constantly construction of the large capacity of the EHV (Extra-High Voltage) power transmission by AC and DC over a long distance, bring huge benefits to economic and social development with energy support. At the same time, it also put forward severe challenges for the safety and stability operation of power system especially put forward higher requirements to the reactive voltage control system.Tertiary voltage control model firstly put forward by the French electricity company, which contains three control models. It comes to the partitions of large power grid and identify the central node in the secondary voltage control by the principle of reactive power voltage control partition boundary weak coupling.The secondary voltage control is the core part of the tertiary voltage control model, mainly including the secondary voltage control partition and the determination of regional central node. The unity of the tertiary voltage control is responsible for the entire network’s optimization calculation. Primary voltage control is to implement the operation of the important load node, central substation, power plants, etc. As the based on polar coordinates or directly active power/reactive power decoupled power flow equations reactive power/voltage sensitivity in traditional has been difficult to adapt to the frequent changes of space truss structure and the parallel processing of the new energy unit in the secondary voltage control partition, a linear sensitivity based on the linearization of the equations of injection current which is closer to the actual physical characteristics of power grid was proposed, and combined with soft partition algorithm to realize secondary voltage control partition and determine the central node by artificial intelligence algorithm.By the linearization of the injective current power flow equation and then achieve the linear sensitivity between node’s injective reactive current/the real part of node voltage, with which form the reactive power source control space and the electric distance between every two nodes. Determine the reasonable partition number, introduce glowed class algorithm with whom the cloud model of each node in power system is established and implement the soft partition of power grid. After implementing the partition, set objective function by improved particle swarm optimization (PSO) algorithm, choice the top nodes of the selected fitness as preprocessing nodes, further by calculating the control sensitivity of the pretreatment nodes to the generators in partition and determine the central node. At the completion of the secondary voltage control partition and determine the central node, through calculate the reactive power reserve margin in the reactive voltage control partition and take break one power transmission line in chance to verify and analysis the rationality and reliability of the partition results.The simulation of IEEE14,30,39and 57 nodes power system verified the effectiveness of the proposed method in this paper.