Jure Vogrinc

John Moriarty, Jure Vogrinc, Alessandro Zocca

The frequency stability of power systems is increasingly challenged by various types of disturbances. In particular, the increasing penetration of renewable energy sources is increasing the variability of power generation and at the same time reducing system inertia against disturbances. In this paper we are particularly interested in understanding how rate of change of frequency (RoCoF) violations could arise from unusually large power disturbances. We devise a novel specialization, named ghost sampling, of the Metropolis-Hastings Markov Chain Monte Carlo method that is tailored to efficiently sample rare power disturbances leading to nodal frequency violations. Generating a representative random sample addresses important statistical questions such as "which generator is most likely to be disconnected due to a RoCoF violation?" or "what is the probability of having simultaneous RoCoF violations, given that a violation occurs?" Our method can perform conditional sampling from any joint distribution of power disturbances including, for instance, correlated and non-Gaussian disturbances, features which have both been recently shown to be significant in security analyses.

Jure Vogrinc, Samuel Livingstone, Giacomo Zanella

We study the class of first-order locally-balanced Metropolis--Hastings algorithms introduced in Livingstone & Zanella (2021). To choose a specific algorithm within the class the user must select a balancing function $g:\mathbb{R} \to \mathbb{R}$ satisfying $g(t) = tg(1/t)$, and a noise distribution for the proposal increment. Popular choices within the class are the Metropolis-adjusted Langevin algorithm and the recently introduced Barker proposal. We first establish a universal limiting optimal acceptance rate of 57% and scaling of $n^{-1/3}$ as the dimension $n$ tends to infinity among all members of the class under mild smoothness assumptions on $g$ and when the target distribution for the algorithm is of the product form. In particular we obtain an explicit expression for the asymptotic efficiency of an arbitrary algorithm in the class, as measured by expected squared jumping distance. We then consider how to optimise this expression under various constraints. We derive an optimal choice of noise distribution for the Barker proposal, optimal choice of balancing function under a Gaussian noise distribution, and optimal choice of first-order locally-balanced algorithm among the entire class, which turns out to depend on the specific target distribution. Numerical simulations confirm our theoretical findings and in particular show that a bi-modal choice of noise distribution in the Barker proposal gives rise to a practical algorithm that is consistently more efficient than the original Gaussian version.