Adaptive frequency control support of a DFIG based on second-order derivative controller using data-driven method

Nowadays, wind turbines can contribute to the frequency control of the power network. This article demonstrates that the second derivative of the power system frequency contains useful information for controlling the network frequency. Therefore in this paper, the use of the proportional, derivative, and second-order derivative terms (PDD2) controller for frequency control of the power network is proposed. A parabolic sliding-mode filter is also used to remove measured frequency noise. Adjusting the PDD2 gains significantly influences the performance of power system frequency control, but due to the dynamic properties of the power system and the wind speed, it is very difficult to adjust the PDD2 coefficients in the frequency controller which performs best in controlling the network frequency. In this article, a new algorithm for adaptive tuning of PDD2 gains in doubly fed induction generator (DFIG) is presented. Data-driven approach updates PDD2 gains only based on the input and output data of the network frequency controller in DFIG. In the proposed adaptive frequency control method, the power system frequency of the next step is approximated via the K-vector nearest neighborhood (K-VNN). Then the PDD2 coefficients were updated using the descending gradient method. Simulation results indicate the good performance of the adaptive PDD2 controller in wind turbine for supporting grid frequency. According to the simulations, the proposed control method will improve the FN (frequency nadir) by minimum 18% compared to the traditional network frequency control methods. In some cases, the implementation of the proposed control method will improve FN up to of 80%.