Hierarchical Coordinated Fast Frequency Control Using Inverter-Based Resources

Abstract

We develop and test a data-driven and areabased fast frequency control scheme, which rapidly redispatches inverter-based resources to compensate for local power imbalances within the bulk power system. The approach requires no explicit system model information, relying only on historical measurement sequences for the computation of control actions. Our technical approach fuses developments in low-gain estimator design and data-driven control to provide a model-free and practical solution for fast frequency control. Theoretical results and extensive simulation scenarios on a three area system are provided to support the approach.

Hierarchical Coordinated Fast Frequency Control using Inverter-Based Resources

“…We have proposed and validated through detailed simulations a robust data-driven disturbance estimator that allows us to reliably compute the real-time power imbalance in a highly nonlinear power system, and in the presence of measurement noise. This data-driven estimate has been integrated in the hierarchical frequency control architecture initially proposed in [13], to provide a completely model-free approach to provide fast, localized frequency regulation in the power system. An important direction for future research is further investigation into the design of improved excitation input signals for data collection, and integration of this approach with transmission and distribution coordination schemes.…”

“…2 In Scenario #1, we have compared the ODDE against all the alternatives listed above, and demonstrate its performance premium relative to the LDDE. In the remainder of the scenarios, we focus the plots on comparing the better estimator (ODDE) against the modelbased approach presented in [13]. Finally, the data collection and real-time simulation steps include measurement noise, modelled as zero-mean white noise of standard deviation 10 −6 p.u.…”

Spectral clustering and model reduction for weakly-connected coherent network systems

“…Then (9) implies λ 3 (L A ) ∼ Ω p (n), showing that T 2 (s) will be a good approximation to T yu (s), by Theorem 1. Also, (10) shows that…”

“…The coherency or synchrony [4] (a generalized notion of coherency) is identified by studying the first few slowest eigenmodes (eigenvectors with small eigenvalues) of M −1 L, the analysis can be carried over to the case of uniform [3] and non-uniform [5] damping. However, such state-space-based analysis is limited to very specific node dynamics (second order) and do not account for, more complex dynamics or controllers that are usually present at a node level; e.g., in the power systems literature [8]- [10]. Therefore, we need a coherence identification procedure that works for more general node dynamics.…”

Fast hierarchical coordinated controller for distributed battery energy storage systems to mitigate voltage and frequency deviations

“…Dynamics in traditional power systems are primarily dominated by the actions of synchronous generators (SGs) [1,2]. However, the increasing spread of distributed energy resources (DERs), renewable energy systems and the connection of nonlinear loads with fast timeconstants triggers undesired dynamics that droop control and automatic generation control (AGC) systems are unable to handle, impacting the stability and reliability of modern power grids [3][4][5][6]. Thus, this fact motivates the suitable combination of both SGs and battery energy storage systems (BESSs) acting synergistically through novel control schemes that prioritise, in an accurate and a fast way, the power dispatch.…”