Multi-objective model predictive control for microgrids

被引：14

作者：

Schmitt, Thomas ^{[1
]}

Rodemann, Tobias ^{[2
]}

Adamy, Juergen ^{[1
]}

机构：

[1] Tech Univ Darmstadt, Control Methods & Robot Lab, Landgraf Georg Str 4, D-64283 Darmstadt, Germany

[2] Honda Res Inst Europe GmbH, Carl Legien Str 30, D-64283 Offenbach, Germany

来源：

AT-AUTOMATISIERUNGSTECHNIK | 2020年 / 68卷 / 08期

关键词：

Pareto; MOO; MPC; optimization; weight tuning; linear programming reformulation; smart grid; ENERGY MANAGEMENT; OPTIMIZATION METHODS; COMPROMISE SOLUTIONS; OBJECTIVES;

D O I：

10.1515/auto-2020-0031

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Economic model predictive control is applied to a simplified linear microgrid model. Monetary costs and thermal comfort are simultaneously optimized by using Pareto optimal solutions in every time step. The effects of different metrics and normalization schemes for selecting knee points from the Pareto front are investigated. For German industry pricing with nonlinear peak costs, a linear programming trick is applied to reformulate the optimization problem. Thus, together with an efficient weight determination scheme, the Pareto front for a horizon of 48 steps is determined in less than 4s.

引用

页码：687 / 702

页数：16

共 41 条

[1] A review on simulation-based optimization methods applied to building performance analysis [J].