Application improved particle swarm algorithm in parameter optimization of hydraulic turbine governing systems

The dynamic characteristics of a hydraulic turbine governing system is determined by the parameters of the hydraulic turbine governor. There are several drawbacks of the conventional particle swarm algorithm in parameter optimization, such as low speed of convergence, low accuracy and being inclined to result in partial optimization during the process of optimization. This paper introduced concave function form as the inertia weight into the conventional particle swarm algorithm and established a mathematical model for a Francis hydraulic turbine governing system. The index of ITAE was chosen as the objective function in the model and the modified particle swarm algorithm was applied into the parameter optimization of the hydraulic turbine governing system. Meanwhile, the performance of the optimization process of the modified particle swarm algorithm was compared with the conventional parameter optimization methods by means of simulation experiment. The results show superior performance of control system can be obtained from the optimization results of the modified particle swarm algorithm.