Analysis On Load Flow Calculationand Operation Characteristic For Microgrid

With the rapid development of industry and economy, the problem of exhaustion ofenergy resources and environmental pollution has become increasingly prominent. It hasalso been exposed that large power system brings the grid vulnerability. In order to solvethe above problems, the distributed generation technology based on renewable energy hasbeen rapidly developed. As the main component of smart girds, microgrid has emerged. Thecontradictions between distributed generation and the large power grid have beencoordinated by microgrid, and the power supply reliability of power system has beenimproved. Load flow calculation, studying the operational status of power system, is a kindof basic electric calculation. Because of the stochastic output power of the microsourcesand loads, it has been difficult for deterministic load flow calculation method to meet thedemands. So it is very necessary to study stochastic load flow calculation method, whichcan consider the influence of the uncertainty random factors. The stochastic load flowcalculation method of microgrid and analyzing its operating characteristics are mainlystudied in this thesis.Firstly, it is analyzed that the fundamentals of the three general methods of stochasticload flow calculation, and their advantages and disadvantages are compared. Themathematical foundations of stochastic load flow calculation are also researched, includingprobability density function, cumulative distribution function, expectation, variance centralmoment, cumulant and Gram-Charlier series.The random models of energy storage, loads,wind generator and fuel cell are established, which have laid the foundation for microgridstochastic load flow calculation.Secondly, aiming at the deficiencies of that three methods, the improved method basedon the cumulant method of Monte Carlo simulation is put forward. The cumulant iscalculated in normal method when the distribution function of input power obeys normaldistribution or discrete distribution in this method. However, the cumulant is calculated inMonte Carlo sampling method when it is unknown or obeys other distribution. This methodhas solved the problem that the normal method is difficult to solve the cumulant in thesituation that the distribution function of input power is unknown, while maintaining theaccuracy and rapidity. Program of this method is written, and the MATLAB running result of a concrete example is compared with Monte Carlo simulating result. The accuracy,reliability and rapidity of this method have been verified.Finally, microgird system model is established. Combined with the proposed methodin this thesis, it is analyzed that the operating characteristics of load flow in microgirdsystem, and mainly the influence of different microsources with same capacity on thevoltage of microgird system. The simulation results of MATLAB show that the influence ofdifferent microsources with same capacity on the voltage of microgird system isproportional to the randomness of them. Meanwhile, accessing energy storage can decreasethe voltage fluctuation of microgrid system, which has certain effect on the improvement ofpower quality.