An improved particle swarm optimisation algorithm applied to battery sizing for stand-alone hybrid power systems

•Battery-integrated dispatch model with analysis of system non-differentiability.•Dispatch-coupled battery sizing which prevents over- and under-sizing.•Numeric analysis on different levels of solar and wind energy penetration.•An improved particle swarm optimisation which outperforms the standard o...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International journal of electrical power & energy systems Ročník 74; s. 104 - 117
Hlavní autoři: Shang, Ce, Srinivasan, Dipti, Reindl, Thomas
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Ltd 01.01.2016
Témata:
Dispatch
Hybrid power system
PSO
Sizing
Stand-alone
Storage
Dispatch
Hybrid power system
Storage
Stand-alone
Sizing
PSO
ISSN:0142-0615, 1879-3517
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•Battery-integrated dispatch model with analysis of system non-differentiability.•Dispatch-coupled battery sizing which prevents over- and under-sizing.•Numeric analysis on different levels of solar and wind energy penetration.•An improved particle swarm optimisation which outperforms the standard one. Stand-alone hybrid power systems with renewable energies are an economic alternative to the main electricity grid where the extension of the grid is too costly or the small local consumption would not justify it. Properly sizing the battery of the systems is an important step to guarantee their reliability and low cost. This paper accepts the dispatch-coupled sizing method by integrating the battery into the operation of the generation units in the system, and formulates this application problem using optimal control. Two major renewable energy sources – solar photovoltaic panels and wind turbines – are considered, together with traditional diesel generators. Penetration level is used as the lever to indicate the different integration degrees of renewables in the system. A particle swarm optimisation (PSO) algorithm is adapted and improved for this specific application. The numeric results of the system planning are benchmarked using an economic indicator, the levelised cost of electricity, to address the real-world system.
ISSN:0142-0615
1879-3517
DOI:10.1016/j.ijepes.2015.07.009