Long Zhao We propose a new robust actionable prescriptive analytics framework that leverages past data and side information to minimize a risk-based objective function under distributional ambiguity. Our framework aims to find a policy that directly transforms the side information into implementable decisions. Specifically, we focus on developing actionable response policies that offer the benefits of interpretability … Read more
At the heart of supervised learning is a minimization problem with an objective function that evaluates a set of training data over a loss function that penalizes poor fitting and a regularization function that penalizes over-fitting to the training data. More recently, data-driven robust optimization based learning models provide an intuitive robustness perspective of regularization. … Read more
This paper introduces the Stochastic Look Ahead Commitment (SLAC) software prototyped and tested for the Midcontinent Independent System Operator (MISO) look ahead commitment process. SLAC can incorporate hundreds of wind, load and net scheduled interchange (NSI) uncertainty scenarios. It uses a progressive hedging method to solve a two-stage stochastic unit commitment. The first stage optimal … Read more
Generation uncertainties, especially during the unit startup and shutdown (SU/SD) processes, pose uncertainties for the real-time market clearing process, and they are often underestimated. This paper proposes two approaches to predict generator SU/SD trajectories in the real-time operations of independent system operators or regional transmission organizations (ISO/RTOs). We first collect and pre-process raw market data … Read more
Power grid vulnerability is a major concern of modern society, and its protection problem is often formulated as a tri-level defender-attacker-defender model. However, this tri-level problem is compu- tationally challenging. In this paper, we design and implement a Column-and-Constraint Generation algorithm to derive its optimal solutions. Numerical results on an IEEE system show that: (i) … Read more
In this paper, we propose a set of two-stage robust optimization models to design reliable p-median facility location networks subject to disruptions. A customized column-and- constraint generation approach is implemented and shown to be more effective than Benders cutting plane method. Numerical experiments are performed on real data and management insights on system design are … Read more
Power grid vulnerability analysis is often performed through solving a bi-level optimization problem, which, if solved to optimality, yields the most destructive interdiction plan with the worst loss. As one of the most effective operations to mitigate deliberate outages or attacks, transmission line switching recently has been included and modeled by a binary variable in … Read more
In this paper, we consider a linear two-stage robust optimization model with a mixed integer recourse problem. Currently, this type of two-stage robust optimization model does not have any exact solution algorithm available. We first present a set of sufficient conditions under which the existence of an optimal solution is guaranteed. Then, we present a … Read more
In this paper, we consider a job scheduling problem with random local generation, in which some jobs must be scheduled day-ahead while the others can be scheduled in a real time fashion. To capture the randomness of the local distributed generation, we develop a two-stage robust optimization model by assuming an uncertainty set without probability … Read more
To improve the efficiency in power generation and to reduce the greenhouse gas emission, both Demand Response (DR) strategy and intermittent renewable energy have been proposed or applied in electric power systems. However, the uncertainty and the generation pattern in wind farms and the complexity of demand side management pose huge challenges in power system … Read more