The Research Of Modeling Hydraulic Turbine Governing System Based On Enhanced Particle Swarm Optimization

Our power system has made a great progress with the step of our modernization construction, its scale is bigger, and its structure is more complicated, this presents new challenges to power system's stable operation while making a great contribution to economic construction. As a great part of power system, hydraulic turbine units play an important role. Because of the duty of peaking and frequency regulation, its running state has a great impact on power system's stable operation. People usually simplify the models of hydraulic turbine unit and its control system in power system's stabilization analysis and calculation, this indeed provides some convenience, but, as time goes on, these models are more and more unsuitable, we should build more precise models to satisfy the demands of power system stable and economical operation.This paper combines current research situation of home and abroad, builds a detailed model of the hydraulic turbine and its control system on the basis of deep research. The model contains important parts of hydraulic turbine control device and controlled object, called hydraulic turbine governing system, it can be simulated dynamically after setting up its parameters. For the control device, its parameters can be calculated through real measured data because of having definite physical meaning. For the controlled object, because it has many parameters and its parameters have not definite physical meaning, we identify its parameters with different identification methods.There are many methods of system identification, such as the lease squares, linear regression and so on, compared with some intelligent methods, these methods are easy to use, but, they can't avoid the problem of producing local optimization. To resolve this contradiction, this paper introduces an enhanced particle swarm optimization method. This method is a global search method, it's easy to understand and convenient to be operated, and it has lower computational cost, it's a good parameter identification method. By inputting real measured data, identifying hydraulic turbine unit and its conduit system with this method, we found that the simulation output data shows in good agreement with the real measured data. After obtaining the parameters of each subsystem model, this paper verifies the model with different real measured data. The result proves the rationality of the model parameters and its construction. Finally, we compare this model with models in BPA and PSASP power system stability calculation software's, the result also shows that it is precise and valuable to apply.