Page 234 - Mathematical Models and Algorithms for Power System Optimization

226 Chapter 7

complex disturbance, and only when an accurate model of such disturbance is determined, the
previously discussed controller can show its advantages. When studying the issue of load
frequency control, this chapter has noticed many useful ideas in the classical control theory and
tried to combine many new achievements in modern control theory, so as to improve control
performance, practically apply new methods, and be able to use the existing actual control
instruments at that same time. The basic ideas for modeling in this chapter are illustrated in the
following text.

7.2.1 Way of Formulating the Load Disturbance Model

The operation process of the power system is a random process, and the load
disturbance is present at any time. The law of the change is not easy to grasp. The
frequency of the power system can be used as an indicator of the balance of power system
operation. The load disturbances inevitably cause frequency changes. Conversely, if the
mathematical model of the load disturbance is known, the variation of the load disturbance
canbeestimatedbasedonthe frequencychange. Because the power system load
disturbance is a complex type of disturbance, the appropriate model problem for
establishing the load disturbance has not been properly solved in the previous load
frequency controller.

The mathematical model of the discrete state equation of the load disturbance is established in
this chapter, in which the time series correlation function analysis method is applied to
stabilize the nonstationary time series by using the high-order difference method, and the
model structure (moving or moving with autoregressive-moving) and its parameters
(autocorrelation function, partial autocorrelation function) of the load disturbance is
determined by autoregressive, moving, and autoregressive-moving that describes the
stationary random process.

7.2.2 Way of Constructing Estimator at All Levels

The successful control of the power system depends on knowing exactly its state at any time,
and some states are determined from the direct observation by distributed measuring
instruments. The speed and placement of the measuring device depends largely on the control
system’s ability on disturbances. The effective control by the application of computer control
requires reliable data, which can be provided by the online state estimation based on the future
system variables.
Because the power system is a complex large-scale system consisting of various power plants,
and each power plant is composed of a speed control, excitation, prime mover, and generator,
this chapter only considers the variation of the load frequency under small disturbances.