In this work, we model the energy management problem in
smart grids as a multi-objective optimization problem subject
to a constraint on the generation capacity of the smart grid.
Efficient energy management involves tradeoffs between the
cost associated with energy consumption and a utility function.
The utility function can represent the living comfort of users or
gross income of the utility company. The utility function is non
decreasing with respect to total utilized power. Hence, it is
important to understand the tradeoffs between energy
consumption and utility.
The main goal of a multi-objective optimization algorithm is
to discover a set of mutually non-comparable solutions called
the Pareto-front which characterizes the tradeoff between
multiple objectives. Multi-objective evolutionary algorithms
simultaneously pursue the search for multiple solutions with
varying emphasis on different objective functions. They have
recently been applied to solve various multi-objective
optimization problems [5-7]. This paper employs a recently
developed algorithm called the Evolutionary Multi-Objective
Crowding algorithm (EMOCA) for solving the the energy
management problem in smart grids. EMOCA has
outperformed state-of-the-art multi-objective optimization
algorithms on several benchmark test functions and also
successfully used in sensor network optimization problems [8,
9]. Simulation results show that EMOCA successfully obtains
several Pareto-optimal solutions and outperforms a well-known
optimization algorithm called Strength Pareto Evolutionary
Algorithm II (SPEA-II) [10]. The rest of the paper is organized
as follows. Section II formulates our problem and describes the
objectives to be optimized. Section III explains the principles
of multi-objective optimization. Section IV describes the
proposed evolutionary algorithm for efficient energy
management. Simulation results and conclusions are presented
in sections V and VI respectively.
978-1-4673-7263-3/15/$31.00 ©2015 IEEE
978-1-