Distributed Connectivity Control In Power System Based On Hybrid System Theory

The network connections of local power systems, access of distributed powergeneration and new energy power generation increase the complexity of the operationand control of the power system and the scale of power system which presentsdecentralized and centralized properties. The risk of occurring blackouts throughout theworld becomes more and more frequent. So, more and more attention is paid to safetytransmission and reliable supply of power system. Many problems that power systemfaces has been difficult to be solve by the traditional control theory. The power systempromiscuous characteristic and topological properties are analyzed combining hybridsystem theory and algebra graph theory in this paper. A distributed control algorithm isproposed to guarantee connectivity of the overall power system and realize the powersystem security and reliable operation.Basic theory knowledge of hybrid system is introduced, including its concept,characteristics and models. This paper has analyzed the power system promiscuouscharacteristic and established a power network model combining hybrid system theorykonwlgde.A new judging approach to the network connectivity is proposed, it is based on theproperties of Laplacian matrix eigenvalues in spectral graph theory. In order to improveoperation efficiency of the judgment of network connectivity, a polynomial matrix isconstructed based on the polynomial acceleration methods, the limited eigenvalues weneeded are computed through matrix-vector multiplication and real backward FFT.Finally, the network connectivity is judged by the size of eigenvalues. This method issuitable for the judgment of network connectivity for large-scale power systemrequiring O (n)operations.According to the hybrid properties of power system, a distributed control algorithm is proposed to guarantee connectivity of the overall power system, identify critical linesand establish a stable connected network framework. Because the power generationprocess has dynamic continuous properties, nearest neighbor potential fields could beused to maintain the normal operation of the power system. Simultaneously, all kinds ofemergencies and accidents happening in power system have highly discrete properties intime and space, in order to avoid a power system becoming several isolated individualswhen multiple links were deleted at one time, real-time connectivity control of the powersystem in dynamic operating is achieved by means of gossip algorithms and distributedmarket-based control. IEEE30nodes test system is selected as the test object. Simulationresults demonstrate that the overall network is connected when links are deleted andfinally critical line of the power system is obtained, this algorithm is more reasonable.