基于SCADA数据的风电机组性能分析及健康状态评估

发布时间：2018-05-03 15:25

本文选题：风电机组 + SCADA数据　；参考：《华北电力大学(北京)》2017年硕士论文

【摘要】：风力发电在众多新能源领域中以其技术和经济优势日益受到重视,然而高昂的运行和维护成本严重制约着风力发电行业的发展。风电机组的维修方式已经从最初的故障维修和预防性维修,发展到如今的预知性维修。准确可靠的设备健康状态评估是实施预知性维修方法的前提和基础。针对风电机组开展健康状态评估,掌握风机健康状态衰退趋势,合理调整运行并开展有针对性的维修决策,可以提高风电机组运行的安全性和可靠性。本文围绕风电机组运行性能和健康状态问题,开展了如下研究工作:1)输出功率是风电机组最具代表性的性能指标之一,风功率曲线也是机组发电能力最直观的表述。在对风电机组SCADA系统风速、功率数据清洗和筛选的基础上,通过描绘出机组正常工作状态下的风功率散点图,采用比恩法建立风电机组实际风功率曲线;统计分析不同风速区间的输出功率,利用逆向云发生器建立不同风速下的输出功率云模型,得到不同机组的整体功率云;通过对比分析功率云的特征值,实现输出功率大小、波动范围和离散程度的量化分析;同时计算风速、功率相关系数反映和评价机组响应的灵敏度。云模型的应用,把机组状态从定性评价拓展到定量评价,从宏观综合评价深入到风速区间段精准评价,提高风电机组性能分析的准确性和全面性。2)风电机组运行环境恶劣多变,故障机理复杂,健康状态评估存在明显的模糊性和随机性。为解决当前模糊评价隶属函数的确定具有主观性且未考虑随机性的问题,提出了一种基于组合赋权和云模型的风电机组健康状态评估方法。在构建健康状态评估指标体系及等级划分基础上,通过组合赋权法确定各指标的权重,应用云模型确定定量指标对各状态等级的隶属度,采用优化的模糊综合评判方法评估风电机组健康状态。最后,通过实例分析验证了本文所提方法的有效性、可行性和准确性。
[Abstract]:Wind power generation has been paid more and more attention in many new energy fields because of its technical and economic advantages. However, the development of wind power industry is seriously restricted by the high operating and maintenance costs. The maintenance mode of wind turbine has developed from original fault maintenance and preventive maintenance to predictive maintenance. Accurate and reliable assessment of equipment health condition is the premise and foundation of implementing predictive maintenance method. The safety and reliability of wind turbine operation can be improved by evaluating the health status of wind turbine, mastering the decline trend of fan health state, adjusting operation reasonably and carrying out pertinent maintenance decision. This paper focuses on the operational performance and health state of wind turbine units. The following research work is carried out: 1) output power is one of the most representative performance indexes of wind turbines, and the wind power curve is also the most intuitive expression of generating capacity of wind turbines. On the basis of cleaning and screening wind speed and power data of SCADA system of wind turbine, the wind power divergence plot under normal working condition of wind turbine is described, and the actual wind power curve of wind turbine unit is established by using Bian method. The output power of different wind speed range is analyzed statistically, the output power cloud model under different wind speed is established by using reverse cloud generator, the whole power cloud of different units is obtained, and the output power is realized by comparing and analyzing the eigenvalue of power cloud. At the same time, the wind speed and power correlation coefficient are calculated to reflect and evaluate the sensitivity of the unit response. The application of cloud model extends the unit status from qualitative evaluation to quantitative evaluation, from macroscopic comprehensive evaluation to accurate evaluation of wind speed interval, improves the accuracy and comprehensiveness of wind turbine performance analysis, and improves the poor operating environment of wind turbine. The fault mechanism is complex, and the assessment of health status has obvious fuzziness and randomness. In order to solve the problem that the determination of membership function of fuzzy evaluation is subjective and stochastic, a new method of wind turbine health assessment based on combination weight and cloud model is proposed. On the basis of constructing index system of health state evaluation and grading, the weight of each index is determined by combination weighting method, and the membership degree of quantitative index to each state grade is determined by cloud model. The optimal fuzzy comprehensive evaluation method is used to evaluate the health status of wind turbine units. Finally, the effectiveness, feasibility and accuracy of the proposed method are verified by an example.
【学位授予单位】：华北电力大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM315

【参考文献】