Power Generation Technology ›› 2024, Vol. 45 ›› Issue (4): 675-683.DOI: 10.12096/j.2096-4528.pgt.23040

• New Energy • Previous Articles     Next Articles

Onshore Wind Farm Collector Circuit Division and Topology Optimization Based on Improved Fuzzy C-Means Clustering

Hai YI1, Zhouan LÜ1, Lingli ZHANG1, Xi CHEN1, Dian LIU1, Yuwei HUANG2, Xingxing HAN2, Chang XU2   

  1. 1.China Three Gorges Renewables (Group) Co. , Ltd. , Xicheng District, Beijing 101100, China
    2.College of Energy and Electrical Engineering, Hohai University, Nanjing 211100, Jiangsu Province, China
  • Received:2023-12-04 Revised:2024-02-20 Published:2024-08-31 Online:2024-08-27
  • Contact: Chang XU
  • Supported by:
    National Natural Science Foundation of China(52106238);China Three Gorges Renewables Technology Project(2021-491);Bilateral Innovation Cooperation Projects Between Governments(BZ2021019)

Abstract:

Objectives Driven by the dual carbon goal and the accelerated transformation of China’s energy structure, the scale of the wind power industry continues to grow rapidly, and there is an urgent need to reduce costs and increase efficiency to cope with the pressure of grid parity. The cost of collector lines accounts for a relatively large proportion of investment, and there is considerable space for optimization. In order to reduce the investment cost, an improved fuzzy C-means (FCM) clustering algorithm was proposed. Methods The improved FCM clustering algorithm was used to divide the collector circuit of onshore wind farm. The algorithm comprehensively considered the azimuth angle and the Euclidean distance to ensure that the lines between the circuits were not crossed, and the adjacent units were gathered to the same circuit. The correction factor of the distance from the machine position to the cluster center was introduced, and the circuit capacity was limited by adjusting the distance correction factor. On the basis of circuit division, the dynamic Prim algorithm was used to optimize the line selection of each circuit. Finally, the effectiveness of the method was verified by an example of an onshore wind farm. Results Compared with the clustering method only considering the azimuth angle, the improved FCM algorithm considering azimuth angle and spacing has better optimization effect. The total cost of single-circuit and double-circuit collection lines is reduced by 2.6% and 5.4%, respectively. Conclusions The proposed algorithm can effectively reduce the total cost of collector lines, and has certain application value. It can provide a reference for the design of wind farm collector lines.

Key words: onshore wind farm, collector line, topology optimization, fuzzy C-means (FCM) clustering algorithm, dynamic Prim algorithm

CLC Number: