Multi-site and multi-objective optimization for wind turbines based on the design of virtual representative wind farm

• Published in: Energy (Oxford) 2022-08, Vol.252, p.123995, Article 123995
• Main Authors: Song, Dongran, Xu, Shanmin, Huang, Lingxiang, Xia, E., Huang, Chaoneng, Yang, Jian, Hu, Yang, Fang, Fang
• Format: Article
• Language: English
• Subjects: Competitiveness; Design; Design optimization; Energy costs; Fuzzy membership function; Multi-objective harmony algorithm; Multi-site and multi-objective optimization; Multiple objective analysis; Production costs; Renewable energy; Sorting algorithms; Turbines; Virtual representative wind farm; Wind farms; Wind power; Wind turbines
• ISSN: 0360-5442; 1873-6785
• DOI: 10.1016/j.energy.2022.123995

The design optimization of wind turbines is an effective solution for reducing the generation cost of wind power and enhancing its market competitiveness. Because wind power projects often require a long construction period and large investment, designing a wind turbine suitable for multiple sites is economically feasible. In this study, a multi-site and multi-objective optimization framework was proposed. Based on the modified energy cost model at single wind farm level, the concept of a virtual representative wind farm (VRWF) was presented, wherein a weight calculation method to design the weights of multiple wind farms was proposed. To simultaneously maximize the annual energy production and minimize the annual production cost of the VRWF, the non-dominated sorting multi-objective harmony algorithm was improved and employed for conducting the optimization, using which the non-dominated solutions were obtained. Subsequently, the subjective and objective combined fuzzy membership function method was presented to determine the best parameters from non-dominated solutions. The proposed method was applied to a case involving three wind farms, with the results confirming its applicability. Compared with the single-site optimization, the proposed multi-site optimization reduced the overall energy cost by 0.4%, 0.8%, and 1.9%, respectively, indicating the necessity of employing multi-site and multi-objective optimization. •The Virtual Representative Wind Farm is proposed to integrate optimization objects.•Subjective and objective factors are considered when designing weights.•Non-dominated Sorting Multi-objective Harmony algorithm is presented.•The decision weights are presented to implement the design preferences.