A hybrid algorithm for mixed integer nonlinear programming in residential energy management

Demand response plays an important role in helping us achieve clean energy and climate goals. It offers a variety of environmental benefits, such as reducing energy usage, offsetting the need for fossil-fueled power plants, and helping to manage system challenges from increased wind and solar energy...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of cleaner production Ročník 226; s. 940 - 948
Hlavní autoři: Huang, Yantai, Wang, Wanliang, Hou, Beiping
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Ltd 20.07.2019
Témata:
algorithms
clean energy
climate
ecosystem services
energy conservation
fossil fuels
MINLP
power plants
PSO
quadratic programming
Residential energy management
Scheduling
solar energy
SQP
wind
wind power
Residential energy management
MINLP
Scheduling
SQP
PSO
ISSN:0959-6526, 1879-1786
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Demand response plays an important role in helping us achieve clean energy and climate goals. It offers a variety of environmental benefits, such as reducing energy usage, offsetting the need for fossil-fueled power plants, and helping to manage system challenges from increased wind and solar energy. However, DR suffers from difficulties associated with optimization model complexity. Especially, when the DR problem is formulated as a mixed integer nonlinear programming (MINLP) problem, which is NP-hard problem. In order to solve this problem, a global optimization approach that combines a particle swarm optimization (PSO) algorithm with a sequential quadratic programming (SQP) local optimizer is proposed in this paper. This allows us to incorporate the global search characteristics of PSO algorithms with the local search capabilities of SQP algorithms. This unique combination increase the possibility of a successful global solution by balancing the exploration-exploitation trade-off and is particularly effective atsolving MINLP problems. Application of the approach in residential energy management indicates that the proposed algorithm is effective atresolving the MINLP DR problem. •A complex mixed nonlinear integer models is proposed and applied to a residential scenario.•A hybrid algorithm was implemented to solve the non-convex MINLP, and can obtain near optimal solution.•The performance of proposed algorithm is better for solving demand response related MINLP than commercial solver Knitro.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0959-6526
1879-1786
DOI:10.1016/j.jclepro.2019.04.062