A practical approach for hybrid distributed MPC. (July 2017)

• Record Type: Journal Article
• Authors: Mendes, Paulo R.C.; Maestre, Jose M.; Bordons, Carlos; Normey-Rico, Julio E.
• Is Part Of: Journal of process control. Volume 55(2017)
• Page Start: 30
• Page End: 41
• Publication Date: 2017-07
• Subjects: Hybrid systems -- Distributed model predictive control -- Networked microgrids; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2017.01.001
• Languages: English
• ISSNs: 0959-1524

Abstract

Abstract : Highlights: The method solves distributed problems with continuous and binary variables. Binary variables in each subsystem are obtained by a binary search algorithm. A distributed coordination algorithm obtains the iterative binary solution. The proposed method is applied to the problem of networked microgrids. Abstract: This paper presents a framework to deal with distributed optimization problems composed by binary and continuous variables. Instead of using a mixed integer quadratic programming (MIQP), the approach proposed here transforms the MIQP into a set of quadratic programming's (QP) that are easier to solve. In this way an instance of the controller related to each feasible combination of binary variables is created. The distributed controller performs an iterative process where the set of agents must agree on the value of continuous interconnection variables, while each agent must decide the values of local binary variables. During the iteration procedure the instances are rated according to a performance index and the instances with best performance are selected until the best one is obtained. The proposed methodology is applied to economic optimization of networked microgrids.