This paper presents a novel modeling and control approach for the aggregation of large numbers of heterogeneous thermostatically controlled loads, such as refrigerators, electric water heaters, and air conditioners, and their usage for Demand Response. Unlike traditional Demand Response methods that act on time scales of hours, this approach is able to provide short-term (e.g., second-to-second...

This paper derives 2-norm performance bounds for the feedback control of discrete-time MIMO nonminimum phase (NMP) systems with arbitrary delay structure. Also, the associated optimal controller, in Youla parameterization form, is explicitly obtained. The derivation of those results uses a special interactor matrix to extract the delays. It is shown that this interactor is unique and a building...

In this paper, we present a robust output-feedback model predictive control (MPC) design for a class of open-loop stable systems with hard inputand soft state constraints. The proposed output-feedback design is based on a linear state estimator and a novel parameterization of the soft state constraints that has the advantage of leading to optimization problems of prescribable size. Robustness a...

Control of distributed systems has various industrial applications, as it is often desired to keep complex multi-disciplinary systems in some given state. Definition or parameterization of control space is the first main issue we face when formulating a control problem. Usually, one wishes to keep the parameterization space dimension as small as possible to limit the complexity of the problem. ...

In this paper we are concerned with the factorization approach to control systems, which has the advantage that it embraces, within a single framework, numerous linear systems such as continuous-time as well as discrete-time systems, lumped as well as distributed systems, 1-D as well as n-D systems, etc.[1]. The factorization approach was patterned after Desoer et al.[2] and Vidyasagar et al.[1...

Control parameterization technique is an effective method to solve optimal control problems. It works by approximating the function piece-wise constant (or linear) function. In this way, problems are approximated parameter selection problems, which can be regarded as finite-dimensional optimization where heights and switching times of taken decision variables. They solved using gradient-based m...

An important challenge when using Reinforcement Learning for learning motions in robotics is the choice of parameterization for the policy. We use Gaussian Mixture Regression to extract a parameterization with relevant non-linear features from a set of demonstrations of a motion following the paradigm of Learning from Demonstration. The resulting parameterization takes the form of a non-linear ...

A parameterization for the restratification by finite-amplitude, submesoscale, mixed layer eddies, formulated as an overturning streamfunction, has been recently proposed to approximate eddy fluxes of density and other tracers. Here, the technicalities of implementing the parameterization in the coarseresolution ocean component of global climate models are made explicit, and the primary impacts...