Rezaei Zare, AfshinRahmani, Soleiman2019-11-222019-11-222019-042019-11-22http://hdl.handle.net/10315/36644This thesis developed an advanced assessment tools to recover the power system voltage margin to the acceptable values during the disturbance. First, the effect of disturbance in islanded microgrids are analyzed using power factor-based power-voltage curves and a comprehensive under voltage-frequency load shedding(UVFLS) method is proposed as a last resort in order to restore the system voltage and frequency. The effect of disturbance in conventional power system is investigated by introducing a phenomenon called fault induced delayed voltage recovery(FIDVR) and comprehensive real-time FIDVR assessments are proposed to employ appropriate emergency control approaches as fast as possible to maintain the system voltage margins within the desired range. Then, polynomial regression techniques have been used for predicting the FIDVR duration. Next, advanced FIDVR assessment is implemented which simultaneously predicts whether the event can be classified as FIDVR or not and also predicts the duration of FIDVR with high accuracy.Author owns copyright, except where explicitly noted. Please contact the author directly with licensing requests.EngineeringElectronic Thesis or Dissertation2019-11-22Electrical engineeringComputer sciencePower systemMicrogridDistributed generationInverter-based microgridsLoad sheddingVoltage recoveryFrequency recoveryUnder frequency load sheddingUnder voltage load sheddingUnder voltage frequency load sheddingPower system stabilityPower-voltage curvePower factor-based power-voltage curveFault induced delayed voltage recoverySingle-phase induction motorAir conditionerMachine learningSupervised learningRegressionClassificationTime-series decision makingLinear regressionPolynomial regressionDecision treeRandom forestPrediction