In this paper we consider periodic optimal operation of constrained periodic linear systems. We propose an economic model predictive controller based on a single layer that unites dynamic real time optimization and control. The proposed controller guarantees closed-loop convergence to the optimal periodic trajectory that minimizes the average operation cost for a given economic criterion. A-priori calculation of the optimal trajectory is not required and if the economic cost function is changed, recursive feasibility and convergence to the new periodic optimal trajectory is guaranteed. The results are demonstrated with two simulation examples, a four tank system, and a simplified model of a section of Barcelona's water distribution network.

Abstract-In this paper a new model predictive controller for tracking arbitrary periodic references is presented. The proposed controller is based on a single layer that unites dynamic trajectory planning and control. A design procedure to guarantee that the closed loop system converges asymptotically to the optimal admissible periodic trajectory while guaranteeing constraint satisfaction is provided. In addition, the constraints of the optimization problem solved by the controller do not depend on the reference, allowing for sudden changes in the reference without loosing feasibility. The properties of the proposed controller are demonstrated with a simulation example of a ball and plate system.

This paper presents the application of economic predictive control to minimize the cost of operating a non-isolated micro-grid connected to an electric utility subject to a periodic internal demand. The micro-grid considered is made of a set of photovoltaic panels, two storage systems and can buy and sell energy to a electric utility. The first storage system is made of a cluster of batteries of lead acid and the second storage system is based on hydrogen storage. A function that describes the economic cost of operating the plant taking into account aspects such as electric market costs, degradation of the microgrid and amortization costs is proposed. Based on this cost and considering the periodic nature of the plant, an economic predictive controller capable of adapting to sudden changes on the cost function while guaranteeing stability and recursive feasibility has been successfully tested on a realistic nonlinear model of an experimental configurable test-bed located at the laboratories of the University of Seville.Index Terms-Model predictive control, renewable energy, economic control, periodic control.

No abstract

No abstract