Surrogate-Assisted Optimization for Multi-stage Optimal Scheduling of Virtual Power Plants

Abstract

This paper presents a comparison between two surrogate-assisted optimization methods dealing with two-stage stochastic programming. The Efficient Global Optimization (EGO) framework is challenging a method coupling Genetic Algorithm (GA) and offline-learnt kriging model for the lower stage optimization. The objective is to prove the good behavior of bayesian optimization (and in particular EGO) applied to a real-world two-stage problem with strong dependencies between the stages. The problem consists in determining the optimal strategy of an electricity market player participating in reserve (first stage) as well as day-ahead energy and real-time markets (second stage). The decisions optimized at the first stage induce constraints on the second stage so that both stages can not be dissociated. One additional difficulty is the stochastic aspect due to uncertainties of several parameters (e.g. renewable energybased generation) that requires more computational power to be handled. Surrogate models are introduced to deal with that additional computational burden. Experiments show that the EGO-based approach gives better results than GA with offline kriging model using smaller budget.

Identification of the Parameter Values of the Constitutive and Friction Models in Machining Using EGO Algorithm: Application to Ti6Al4V

“…In a PC, parallel computing is already used inside the simulator (Abaqus) in order to reduce the computation time of a single simulation. Then, sequential EGO is preferred to parallel versions [27,50].…”

Parallel Bayesian Optimization for Optimal Scheduling of Underground Pumped Hydro-Energy Storage Systems

“…where K = Φ T ΣΦ is known as the covariance matrix and K (1)(2) = Φ(x (1) ) T ΣΦ(x (2) ) = k x (1) , x (2) . The k function is referred to as the covariance kernel and K = (k(x i , x j )) i,j∈{1,...,n} and is chosen as an hyper-parameter.…”

“…However, in modern competitive energy networks, individual actors rely on efficient operational strategies, enabling them to hedge the uncertainty of renewable energy resources. It is thus essential to dispose of efficient tools to take informed decisions at the different time steps of the energy markets (e.g., from long-term towards real-time) [2]. Let us assume that for a decision x ∈ R d , the expected profit of a UPHES operator is given by f : R d → R; x → y = f (x).…”

Batch Acquisition for Parallel Bayesian Optimization—Application to Hydro-Energy Storage Systems Scheduling

“…However, in modern competitive energy networks, individual actors rely on efficient operational strategies, enabling them to hedge the uncertainty of renewable energy resources. It is thus essential to dispose of efficient tools to make informed and fast decisions at the different time steps of the energy markets (e.g., from long-term towards real-time) [2]. From the operator's point of view, the quality of a decision is measured as a profit, so let us assume that for a decision x ∈ R d , the expected profit of a UPHES operator is given by f : R d → R; x → y = f (x).…”

“…To the best of our knowledge, surrogate-based optimization has never been applied to the UPHES scheduling problem. In our previous work [2], we demonstrated that Bayesian Optimization (BO) can be handy in management problems in electrical engineering.…”