Data-Driven Fuzzy C-Means Equivalent Turbine-Governor for Power System Frequency Response

This research paper proposes a turbine-governor modelling technique based on equivalent FCM (Fuzzy C-Means) for a control area of an equivalent power system used for frequency response analysis. The FCM algorithm implementation is proposed to find an equivalent Fuzzy model of n turbine-governors that are in an area of the electric power system (EPS). The FCM algorithm is mainly used to generate the rules for the fuzzy model; this algorithm uses input-output data, deviation of frequency, velocity and its derivatives, these are numerical data of a control area of the electrical system that contains n turbine-governors. Two cases are used to test the equivalent FCM model: (i) model of three areas simulink model, where area two has been modified by adding four turbine-governors to verify that it is possible to define an equivalent FIS model based on data, and (ii) the multi-area system, that is extracted from a reduced system frequency response model of an electric area of Great Britain Power System (GBPS), which contains three different types of turbine-governors, the data for this model was obtained from DIgSILENT-PowerFactory. The equivalent fuzzy model is tested under the same conditions as the original system with n turbine-governors, and they are compared against each other. The simulation results and performance analysis show it is possible to find an equivalent model with excellent performance with FCM and that the parameters of the FIS model can be adjusted if necessary, with ANFIS.