Stability Constrained Economic Load Dispatch for Droop Controlled AC Microgrids

Abstract

Cost-based droop schemes for microgrids have been developed to achieve cost reduction; meanwhile, the stability of microgrids is highly dependent on the droop control design and its parameters. This work proposes a hybrid cost-based droop control that achieves both optimal economical operation and stability preserving for autonomous microgrids. The effect of involving the economics in the control structure is analyzed utilizing a modified cost-based small-signal linearized model accompanied by a modified cost-based power flow approach. Achieving optimality in cost reduction might introduce nonlinearity in the active power droop as well as dispatching the power unproportionally between the distributed generators in the microgrid. Such nonlinear and disproportionate power-sharing causes significant variation in the operating condition of the MG when the load varies, and accordingly, conducting the small signal stability analysis around a single operating point is not valid. Thus, the cost-based power flow approach is proposed to ease the finding of the operating points of the MG for all loading conditions. The outcomes of the power flow method are fed to the cost-based small-signal linearized model to perform the analysis. Low-frequency eigenvalues tend to migrate towards instability when the load increases if a cost-based droop scheme is adopted. The proposed hybrid droop control manages to achieve the optimal generation by incorporating the incremental cost in the active power droop while ensuring stable performance by utilizing active and reactive power derivative controllers. The reactive power derivative controller is utilized to suppress the migration of the low-frequency eigenvalues towards instability, whereas the active power derivative controller is utilized to dampen the oscillations of the active power-sharing among the distributed generators and accordingly enhance the stability margin of the MG even further. The effectiveness of the proposed droop control to ensure an optimal and stable operation is validated on Matlab/Simulink and through OPAL-RT real-time simulation. Although the proposed controller performs satisfactorily well with manual tuning, the droop parameters, including the derivative gains, are optimized to boost the MG's stability even further and enhance the control performance without sacrificing cost reduction. This is done by using the particle swarm optimization method to find the optimal droop parameters to push the low-frequency eigenvalues as far as they can be to the left of the

Date of Award	May 2022
Original language	American English