The Design Of Power System Stabilizer Based Nonlinear Predictive Control

The Chinese power grid has been becoming bigger now, and its operation mode is alsobecoming more and more complex. These make the power system stability problems becomeincreasingly prominent. Among these problems, the low frequency oscillation problem is oneof the most common problems. The most common method to solve the problem is installingthe power system stabilizer (PSS). But the conventional PSS is designed basing on the linearmodel of the stably operating system. Once the operating point is changed, the conventionalPSS cannot work efficiently.The predictive control theory has been widely used in the industry, and it has beenapplicated in the electric power industry too; however, most of these applications are based onthe linear model of the system. The linear predictive control theory is mature, within a certainrange, it can make the system stable and reliable. But because the model is linearized, so thecontrol accuracy is not accurate enough. Therefore, this paper proposes a nonlinear predictivecontrol algorithm which bases on the particle swarm optimization (PSO). PSO is used toimprove rolling optimization processes of the nonlinear predictive control, so as to improvethe precision and response speed of the system’s output. But considering the population isgive randomly at the beginning time, the input of the system may need a long time to optimizewhich will affect the dynamic response performance; or the input of the system is changedviolently which can make the system’s output swing. So, this paper decides to use the maturelinear predictive control algorithm at the beginning time, and use the nonlinear predictivecontrol algorithm based on PSO when PSO population is stable. By this way, the newnonlinear predictive control strategy has the advantages of both linear and nonlinearpredictive control algorithm: the system is more precise; the response is faster and it’s morestable. In this paper, the algorithm is applied to the design of PSS and has carried on thesimulation by MATLAB.The simulation results prove that the designed PSS improves the stability of the systemwhen the single machine infinite system has a two phase short circuit fault, a three-phasecircuit fault, a step increase of the input mechanical power.