Research On Power System Stability Intelligent Optimization Control With Multi-type Controllers

Numerous critical issues in power system stability control remain unsolved because of characteristics of electric power systems such as high-dimension,non-linearity,tight-coupling and uncertainty,etc.With various uncertainty factors and rapid development of modern interconnected power grids in terms of both scale and complexity,the performance limitation of the originally designed stability controller is emerging.Methods to realize the coordination optimization design of multi-type stability controllers that can enable the power system to possess desirable dynamic performance under various disturbances both inside and outside the system,are critical to safe and stable operation of lager-scale power grids.Advance in the field of computer technology,control theory and artificial intelligence technology in recent years,intelligent optimization control(IOC)is widely used in several practical engineering fields and it provides a new perspective to realize optimization control of power system stability.In this thesis,the combination of intelligent optimization technology and power system stability is investigated,and research of intelligent optimization control of power system stability is conducted.The recently proposed Grey Wolf Optimizer(GWO)is utilized as a parameter optimization tool.For load frequency control(LFC)system controller optimization design for power system frequency stability,and damping controller coordination optimization design for small signal stability,the optimal control performance of power system stability is obtained by the simultaneous setting and coordinated optimization of multiple controller parameters.Research conducted in this thesis is mainly concerned with:Firstly,for the two-area interconnected power grid LFC PI/PID controller parameter optimization design problem,a controller parameter optimization design method based on intelligent optimization algorithm is proposed.GWO,with insensitivity to initial values,high efficiency and global search ability,is introduced and utilized as optimization tool throughout the whole thesis.The time domain output response index of Integral of Time multiplied Absolute Error(ITAE)is considered as the objective function,and then GWO is utilized to find the optimal LFC controller parameters.Analysis of controller performance degradation caused by parameter uncertainty,and the non-fragility problem of the controller is discussed.Simulation results show that the optimization design method proposed in this thesis is superior to other methods in terms of searching ITAE index and dynamic response control performance.The designed optimal controller possesses more robustness and non-fragility under parameter uncertainty of both the system and the controllers.Secondly,for communication time-delay LFC problems in two-area interconnected power grids,an optimal design strategy of fractional-order PID controller for interconnected power grid LFC considering time-delay is proposed in this thesis.Fractional-order PID controller is adopted for time-delay LFC system controller,time domain output response ITAE index is used to form the objective function,and GWO is utilized to optimize parameters of the controller.Simulation results show that the fractional-order PID controller designed in this way has more desirable index of ITAE,better time-domain dynamic response performance,robustness to time delay and stability,compared with integer-order PID controllers.Then,for simultaneous optimization of multiple Power System Stabilizer(PSS)parameters,a design method based on GWO algorithm is proposed in the thesis.The parameter design of traditional lead-lag PSS is transformed to a quadratic optimization problem based on eigenvalues.By using GWO algorithm,the PSS parameter values are optimized for the optimal damping properties under various operating conditions.The electromechanical oscillation modes with weak damping ratios in multimachine power system are shifted to the strongly damped region in the right half complex plane,where damping ratio is greater than a given value,by the optimized PSS.Simulation results demonstrate the efficacy and robustness of the PSS obtained by GWO in terms of suppressing electromechanical oscillation,whose control performance is much better than that designed by traditional phase compensation method,under various operating conditions.And the performance analysis of the algorithm indicates that GWO has merits like insensitivity to initial values and robustness.Furthermore,for the parameter optimization design problem of multiple types of damping controllers in a power system,a parameter design scheme to coordinate and optimize multiple parameters of PSS,Static Var Compensator(SVC)Power Oscillation Damper(POD),and Doubly-Fed Induction Generator(DFIG)is proposed in this thesis to suppress both local and inter-area oscillations in multimachine power systems.By coordinately optimizing the local PSS and the wide-area signal based SVG and DFIG POD,the scheme adopts the joint modal controllability/observability index to choose the most suitable POD wide-area feedback signal,with the coordinated design scheme modeled as a quadratic optimization problem based on eigenvalues.The simulation results show the feasibility and efficacy of the proposed method in terms of suppressing electromechanical oscillation in multimachine power systems.The damping controllers designed by the proposed approach also demonstrates robustness under various operating conditions.In the last chapter,for probabilistic small signal stability problem considering load correlation in power systems,a probabilistic small signal stability analysis method based on Latin Hypercube Sampling(LHS)technique with variance reduction property combined with Cholesky decomposition is proposed in the thesis.Probabilistic distribution of the eigenvalues is obtained by LHS Monte-carlo simulation.The simulation examples show that,compared with simple random sampling techniques,the LHS Monte-carlo simulation provides higher sampling rates,which makes it a superior probabilistic small signal stability analysis method.The correlation between loads has significant impact on the results of power system probabilistic small signal stability analysis.The risk of instability of system critical oscillation mode increases when the uncertainty and correlation of loads grow.