Research On AGC Unit Dispatch In Power Market

The automatic generation control (AGC) is an important component of ancillary service in power market and it is crucial to the stable operation of power system. With the development of socialist market economy, the objective of AGC ancillary service has changed from making the generation cost lowest of all the AGC units before deregulation to making the ancillary service cost lowest of power dispatching centre pays power plant in power market.As to the AGC unit dispatching problem in power market, the traditional preferential order method considers the performance index such as regulating speed more, but it neglects the factor of capacity bid. So it may result in the unit with higher bid becoming the marginal unit with running up the market clearing price and increasing the capacity purchasing cost. Using the Genetic Algorithm can realize the global optimization, but it has slow convergence speed, tends to get into local optimization and has difficulty in solving the constraints.A new AGC unit dispatch method is developed using the particle swarm optimization (PSO) algorithm in this paper in view of the deficiency of traditional methods.Aiming at the AGC unit dispatch mathematic model of power market, this paper analyses the unit constraints of regulating capacity and regulating speed, firstly. Then several optimization functions are tested and the influences of different values of parameter are analyzed. The testing result shows that adjusting the value of control coefficient can realize the aim of enlarging searching space and increasing convergence speed. Case studies on an actual power system show that this method could overcome the shortcoming in using the integer programming method. And compared with the genetic algorithm, the PSO algorithm can avoid the complicated parameter setting and has good convergence and high convergence speed. The PSO is an effective method in solving the problem of AGC unit dispatch.