A hybrid cross-entropy guided genetic algorithm for scheduling multi-energy systems

This investigation develops a novel hybrid fast converging Cross-Entropy Genetic Algorithm technique for scheduling of multi-energy system (MES). The scheduling problem is a mixed integer non-linear programming problem with non-linear and non-convex constraints, due to switching and non-linear dynam...

Full description

Saved in:
Bibliographic Details
Main Authors: Dan, Mainak, Srinivasan, Seshadhri, Sundaram, Suresh
Other Authors: School of Computer Science and Engineering
Format: Conference or Workshop Item
Language:English
Published: 2020
Subjects:
Online Access:https://hdl.handle.net/10356/142997
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:This investigation develops a novel hybrid fast converging Cross-Entropy Genetic Algorithm technique for scheduling of multi-energy system (MES). The scheduling problem is a mixed integer non-linear programming problem with non-linear and non-convex constraints, due to switching and non-linear dynamics exhibited by the MES devices. A hybridization of cross entropy and genetic algorithm termed as CE-mGA is proposed for the betterment of search space exploration as well as exploitation with fast convergence. In addition, a constraint-driven mutation strategy is also introduced in GA framework for tackling the non-linear and non-convex constraints. The investigation illustrates that the proposed algorithm is able to provide a stand-0ff between exploration and exploitation with an improvement in convergence speed than hybrid real-coded genetic algorithm upon validation at Cleantech building, Singapore.