Cognitive Prediction-Enabled Online Intelligent Fault Diagnosis and Prognosis for Wind Energy Systems

支持认知预测的风能系统在线智能故障诊断和预测

• 批准号: 1308045
• 负责人: Wei Qiao
• 金额: $ 35.99万
• 依托单位: University of Nebraska-Lincoln
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-06-01 至 2017-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1308045&HistoricalAwards=false
• 关键词: Cognitive; Prediction; Enabled; Online; Intelligent

The objective of this research is to study the use of cognitive prediction paradigms for online fault diagnosis and prognosis to enable condition-based smart maintenance for wind energy systems. The approach is to: (1) study the use of time and frequency domain data mining methods to effectively extract the features of faults in a wind turbine from the signals acquired from the wind turbine condition monitoring system; and (2) study the use of artificial neural networks and machine learning for intelligently diagnosing and prognosing faults, predicting the lifetime, and quantitatively evaluating the physical condition of the wind turbine using the extracted fault features.Intellectual Merit: This project will create innovative cognitive prediction-based models and computational algorithms to enhance condition awareness of geographically distributed wind energy systems. The findings of this research are highly transformable and will provide capabilities for enabling condition-based intelligent maintenance for other energy conversion and engineered systems.Broader Impacts: The outcome of this project will further exploit the benefits of wind power by successfully reducing cost and improving reliability of wind energy systems and, therefore, will make wind energy a reliable, cost-competitive source of clean electricity. The increasing use of wind power will benefit various sectors of the nation's economy and contribute to sustainable development of society. Multiple fields covered by this project are areas where a talent shortage is projected in the United States, particularly in the Midwest. The proposed activities will provide a unique learning platform for young individuals to become skilled professionals.

本研究的目的是研究使用认知预测范式进行在线故障诊断和预测，使基于状态的智能维护风能系统。 办法是：（1）研究利用时域和频域数据挖掘方法，从风力涡轮机状态监测系统采集的信号中有效提取风力涡轮机故障特征;（2）研究人工神经网络和机器学习在智能故障诊断和排除中的应用，预测寿命，以及使用所提取的故障特征来定量地评估风力涡轮机的物理状况。这个项目将创造创新的认知预测-的模型和计算算法，以增强地理分布的风能系统的状况意识。 该研究成果具有高度可转换性，将为其他能源转换和工程系统提供基于状态的智能维护能力。更广泛的影响：该项目的成果将通过成功降低成本和提高风能系统的可靠性，进一步利用风能的优势，从而使风能成为可靠的、具有成本竞争力的清洁电力来源。 风力发电的增加将使国家经济的各个部门受益，并有助于社会的可持续发展。 该项目涵盖的多个领域是美国预计人才短缺的领域，特别是在中西部。 拟议的活动将为年轻人成为熟练的专业人员提供一个独特的学习平台。