Interval-affine Analysis Method In Power System With Consideration Of Intermittent Generation Uncertainty

The power outputs of photovoltaic systems and wind generators have theproperty of intermittence and randomness for the uncertainty of weather conditions.The integration of intermittent power generations in power systems has introducedmore and more uncertainty. Conventional deterministic methods are unable to analyzethe influence of wind or photovoltaic power generation systems uncertainty on powergrid, this dissertation focuses on the study of the increasing uncertainty caused byintermittent power generations in power systems and proposes a novel interval-affineanalysis method in power system uncertainty analysis. The main contributions of thisdissertation are as follows:1. This dissertation proposes a DC Power flow method based on affine arithmetic(DCA) to quantify risks due to uncertainties in loads, power generations. The DCA isable to calculate the power flow with consideration of uncertainty compared to theconventional DC power flow method, and more accurate than interval power flow. Anautomatic contingency selection method under uncertainty (ACSU) is proposed basedon DCA method. The ACSU method selects every case from the contingency listwhich includes all credible line and generator outages, and run DCA solution tocalculate the active power performance interval index. The ranking results of indexprovide the system planner or operator with quantitative and fast decision support toolunder uncertainty environments.2. An uncertain model of wind turbine and photovoltaic based on affinearithmetic is developed, and then a Complex Affine arithmetic-based Three-phaseForward-Backward Sweep power flow (CATFBS) is proposed to deal withuncertainties and unbalanced distribution power system. The CATFBS method has theproperty of comprehensive, and can explore the impacts of uncertainties (introducedby distribution generations or loads) on the operation of distribution systems. Theproposed method is much more efficient than the Monte Carlo method whenconsidering uncertainties.3. Based on the affine arithmetic, an index, Relative Influence of UncertainVariables on Outcome (RIUVO), is proposed to quantitatively analyze the influenceof uncertain variables on the result. The concept of RIUVO has been expanded toquantitatively analyze the influence of distribution generations’ uncertainties on eachbus voltage.4. A complex affine arithmetic-based constant-current source model for harmonic source is proposed to analysis the uncertainties of distribution generation on theharmonic power flow, then a complex affine algorithm-based three-phaseforward-backward sweep harmonic power flow method is proposed to study theimpacts of uncertainties on unbalanced three-phase distribution systems.5. To improve the insufficiency of decision-making ability of existing distributedgeneration planning method, a model and a method of optimal planning of distributedgenerations are proposed which consider the uncertain information. Thesub-objectives in this model include the least investment cost, the highest earning, thehighest environment benefits and the least loss in distribution network. Then, amethod combined by TOPSIS algorithm based on interval mathematics and geneticalgorithm is proposed in this dissertation to solve the proposed model. The resultshows the validity and practicability of the proposed model and method.