智能科学与技术学报 ›› 2020, Vol. 2 ›› Issue (1): 62-71.doi: 10.11959/j.issn.2096-6652.202007

• 学术论文 • 上一篇    下一篇

基于改进bin算法的风电机组风速-功率数据清洗

王新,王政霞()   

  1. 重庆交通大学,重庆 400074
  • 修回日期:2020-02-25 出版日期:2020-03-20 发布日期:2020-04-10
  • 作者简介:王新(1991-),男,重庆交通大学硕士生,主要研究方向为机器学习与数据挖掘、故障检测|王政霞(1977-),女,重庆交通大学教授、硕士生导师,主要研究方向为机器学习与数据挖掘、图像处理、医学影像诊断
  • 基金资助:
    重庆市科委自然科学基金资助项目(2018jcyjAX0398);重庆市研究生科研创新项目(2019S0104)

Wind speed-power data cleaning of wind turbine based on improved bin algorithm

Xin WANG,Zhengxia WANG()   

  1. Chongqing Jiaotong University,Chongqing 400074,China
  • Revised:2020-02-25 Online:2020-03-20 Published:2020-04-10
  • Supported by:
    Natural Science Foundation Project of CQCSTC(2018jcyjAX0398);Graduate Student Research Innovation Project in Chongqing(2019S0104)

摘要:

风速-功率是风电机组发电性能的重要指标,对风电场的运行管理具有重要意义。风速-功率数据是通过安装在风电场的监视控制与数据采集(SCADA)系统采集得到的,原始数据存在大量噪声,给后续应用研究带来了很大的挑战。基于风速-功率数据的空间分布特征,将风速-功率数据分为3类,并改进了数据预处理方法bin算法,提出了基于分区域(dbin)算法的异常数据识别清洗方法及流程。实验结果表明,dbin算法识别异常数据的效率比传统算法更高,具有较强的通用性。

关键词: 风电机组, 风速功率, dbin, 监视控制与数据采集

Abstract:

Wind power is an important indicator of the generating performance of wind turbines,which is of great significance to the operation and management of wind farms.The wind-speed and power data were collected through the monitoring and control (SCADA) system installed in the wind farm.There are a lot of noises in the original data,which brings great challenges to the subsequent application research.Based on the spatial distribution characteristics of wind-speed and power data,wind-speed and power data was divided into three categories,the data preprocessing method bin algorithm was improved,and the method and process of abnormal data identification and cleaning based on district bin(dbin) algorithm were proposed.The experimental results show that the dbin algorithm proposed in this paper is more efficient than the traditional algorithm in identifying abnormal data,and has strong universality.

Key words: wind turbines, wind-speed and power, dbin, supervisory control and data acquisition

中图分类号: 

No Suggested Reading articles found!