https://orcid.org/0000-0001-5923-554X
![Sample our Built Environment journals, sign in here to start your access, Latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5926/TOC-Subject-Sample-2021-Built-Environment.jpg"})
ABSTRACT
This paper presents the realistic optimization model for wind farm (WF) design. The investigated WF is located in the southern region of Turkey and consists of 10 WTs. In this modeling process, the real wind measurement data, the experimental power curve of WT, variable thrust coefficient, and the real WT characteristic features are taken into account. Also, the wake effect calculations are realized considering the wind direction, topographic effect, and roughness. Wind Atlas Analysis and Application Program (WAsP) is used in calculating WF annual energy production (AEP) and verifies the results of the WF optimization model. ABC algorithm is used in the solution of the developed model. Optimal WT coordinates are determined. The real AEP value of the existing WF is 51.570 GWh. In the WAsP software, the AEP value of the existing WF is found to be 50.665 GWh. With the solution of the developed model, the AEP value of WF is increased to 53.454 GWh. The WTs are positioned on determined coordinates in the WAsP software. The AEP value is found to be 53.542 GWh. As a result, the AEP of WF is increased by 5.68%. The AEP of WTs has been increased except for one WT.
Acknowledgments
This work was financially supported by the Kahramanmaras Sutcu Imam University, Scientific Research Projects Unit, the project entitled “Developing a model for optimal turbine placement in wind farm installation” under Project No: 2019/1-15D.
Nomenclature
| GA | = | Genetic Algorithm k Shape parameter [m/s] |
| PSO | = | Particle Swarm Optimization c Scale parameter [-] |
| EA | = | Evolutionary Algorithm P(v) Power curve of WT [MW] |
| ACA | = | Ant Colony Algorithm P Total wind power installed in the WF [kW] |
| GRA | = | Greedy Algorithm Ps Sector frequency [–] |
| HSA | = | Harmony search algorithm Indicator [–] |
| EPS | = | Extended Pattern Search ∆v Decreasing wind speed [m/s] |
| MAS | = | Multi-Agent System z WT rotor elevation [m] |
| T-ABC | = | Teacher-Artificial Bee Colony z0 Terrain roughness height [m] |
| GPSO | = | Gaussian Particle Swarm Optimization i Upstream WT(s) |
| DEA | = | Differential Evolution Algorithm j Downstream WT(s) |
| SAA | = | Simulated annealing algorithm ∆h Hub height difference between the WTs [m] |
| GRNN | = | general regression neural network rj,0 Rotor radius [m] |
| SOM | = | Self-organizing map rw Wake radius [m] |
| WF | = | Wind farm Xij Crosswind direction difference between the WTs [m] |
| AEP | = | Annual energy production Aoverlap Overlap area between wake and rotor [m2] |
| WT | = | Wind turbine Apartial Overlap area created by partial wake [m2] |
| WFLO | = | Wind farm layout optimization Aj Wind turbine rotor swept area [m2] |
| ABC | = | Artificial Bee Colony θw, θj Angles of the wake and rotor intersection arc [°] |
| Obj | = | Objective function Ct Thrust coefficient [-] |
| lb | = | Lower bound kw Wake expansion coefficient of upstream WT [–] |
| ub | = | Upper bound h Hub height of the wind turbine [m] |
| (xi, yi) | = | Location of the ith WT Enet Net AEP [GWh] |
| x | = | Crosswind direction Ewake Annual wake losses [GWh] |
| y | = | Prevailing wind direction ∆v Wind speed deficit [m/s] |
| f(.) | = | Probability of wind speed [-] v1 Wake velocity at a downstream distance x. [m/s] |
| v0 | = | Incoming wind speed of the WF [m/s] vi Wind speed before the hub of the ith WT [m/s] |
| θ | = | Wind direction [°] vj Wind speed before the hub of the jth WT [m/s] |
Additional information
Notes on contributors
İbrahim Celik
İbrahim Çelik received his B.S. degree in Electrical and Electronic Engineering from Firat University in 2012, his M.Sc. degree in Electrical and Electronics Engineering from Kilis 7 Aralik University in 2016. He is currently working toward to a Ph.D. degree in Electrical and Electronic Engineering and he is a lecturer at Kahramanmaras Istiklal University. His research areas include power electronics and renewable energy systems.
Ceyhun Yildiz
Ceyhun Yildiz received a Ph.D. degree in Electrical and Electronic Engineering from Kahramanmaras Sutcu Imam University in 2017. He is a lecturer in the Department of Electricity and Energy at Kahramanmaras İstiklal University. His research interests are electrical machine drivers, renewable power, artificial intelligence, and forecasting methods.
Mustafa Sekkeli
Mustafa Sekkeli received B.Sc., M.Sc., and Ph.D. degrees in electrical and electronics engineering from Istanbul Technical University, 1986, 1989, and 2005, respectively. Between 1999 and 2007, he worked as a lecturer in the Electrical and Electronics Department, Kahramanmaras Sutcu Imam University. Since 2007, he has been with the Electrical and Electronics Engineering Department, Kahramanmaras Sutcu Imam University as Professor. His research interests include power quality, power electronics, electrical machine control, reactive power compensation, and renewable energy systems.