A Prediction-Based Hierarchical Delay Compensation (PHDC) Technique Enhanced by Increment Autoregression Prediction for Wide-Area Control Systems

In a wide-area control system (WACS) of power systems, delays will affect the control performance. Dealing with randomly distributed delays besides achieving desired control performance via the WACS control strategy will make the control strategy design more challenging. Moreover, delays from the centralized controller to actuators in WACS are unknown for the controller when calculating the control strategy. A prediction-based hierarchical delay compensation (PHDC) technique enhanced by an increment autoregressive prediction (IAR) algorithm is proposed in this paper to solve these problems. With PHDC, the WACS inputs and outputs are approximated by IAR and thus available in real time for the control strategy by taking advantage of the WACS synchronization mechanism. Meanwhile, the IAR algorithm is proposed with characteristic analysis by which IAR is proved to be an enhancement to PHDC. The combination of PHDC and IAR thus achieves the ideal compensation performance. A Real-Time Digital Simulator (RTDS) hardwarein-the-loop simulation of a wide-area damping control system in an actual power grid is carried out to verify the performance of PHDC under the actual conditions and extreme conditions, respectively. The results demonstrate that PHDC compensates the randomly distributed delay effectively as well as mitigates the packet congestion and packet loss.

I. Introduction

In modern interconnected power systems, wide-area measurement system (WAMS) provides synchronous measurement data for the dynamic wide-area monitoring and control to mitigate stability problems. Though the wide-area control system (WACS) is widely studied [1], especially the wide-area damping control (WADC) system for low frequency oscillation (LFO) instability problems [2]–[4], time delay along with other networked communication issues is still one of the major problems affecting the control performance of WACS.