Effective Management of Operating and Maintenance Activities for Wind Turbines

该奖项旨在开发分析和计算决策支持工具，以有效管理风能系统的维护活动。 将根据专门的扩散过程开发基于退化的可靠性模型，以明确说明随机变化的环境条件对关键风力涡轮机组件、状态监测系统和时间或天气约束的维护计划的影响。随后，将开发随机优化模型，以通过在不确定条件下自适应地安排维护活动来最小化风力涡轮机生命周期内的运营和维护成本。 分析工具将：（1）确定检查、维修或更换的最佳时机;（2）在计划停机期间优先考虑服务活动;（3）确定订购哪些备件以及订购时间;以及（4）决定应如何使用运行和环境数据来评估关键风力涡轮机部件的当前和未来健康状况。 专门的马尔可夫决策过程（MDP）和部分可观察马尔可夫决策过程（POMDP）模型与混合（离散和连续）的状态空间将开发沿着与其相关的解决方案技术。 将使用真实的风力涡轮机运行数据验证所开发的基于退化的可靠性模型，并将使用优化模型获得的长期平均维护成本与计算机仿真模型进行比较。如果成功，本研究的结果将导致风能系统的运行和维护的改进，风能系统将很快提供美国相当大一部分的电力。风能生产成本的很大一部分直接归因于定期和非定期维护活动产生的运营和维护成本。 这项工作的主要目标是提供更好地管理环境和风力涡轮机数据的技术，以制定有关维护活动的成本效益决策。 降低风能成本将加快新的陆基和海上风力涡轮机的设计、分析和安装。 增加风力资源发电的份额将有助于稳定能源价格，减轻温室气体的影响，并大大减少国家对外国自然资源的依赖。 此外，发展一个可行的风能市场将有助于刺激美国经济，创造与风能直接相关的就业机会，包括涡轮机制造和安装，以及风力涡轮机操作员和维修工人的永久性就业机会。

The objective of this award is to develop analytical and computational decision support tools to effectively manage the maintenance activities of wind energy systems. Degradation-based reliability models, based on specialized diffusion processes, will be developed to explicitly account for the influence of randomly-varying environment conditions on critical wind turbine components, condition-monitoring systems, and time- or weather-constrained maintenance scheduling. Subsequently, stochastic optimization models will be developed to minimize operating and maintenance costs over the life cycle of a wind turbine by adaptively scheduling maintenance activities under uncertain conditions. The analytical tools will: (1) determine the optimal timing of inspections, repairs or replacements; (2) prioritize service activities during scheduled outages; (3) determine which spare parts to order and when; and (4) decide how operating and environmental data should be used to assess the current and future health of critical wind turbine components. Specialized Markov decision process (MDP) and partially observable Markov decision process (POMDP) models with mixed (discrete and continuous) state spaces will be developed along with their associated solution techniques. The developed degradation-based reliability models will be validated using real wind turbine operating data, and the long-run average maintenance costs obtained using the optimization models will be compared with those of a computer simulation model.If successful, the results of this research will lead to improvements in the operation and maintenance of wind energy systems that will soon supply a significant portion of electric power in the United States. A large fraction of the cost of producing wind energy is directly attributable to operating and maintenance costs stemming from scheduled and unscheduled maintenance activities. The primary goal of this work is to provide techniques that better manage environmental and wind turbine data to make cost-effective decisions about maintenance activities. Reducing the cost of wind energy will accelerate the design, analysis and installation of new land-based and offshore wind turbines. Increasing the share of electricity generated by wind resources will help stabilize energy prices, mitigate the effects of greenhouse gases and substantially reduce the nation's dependence on foreign natural resources. Furthermore, the development of a viable wind energy market will help stimulate the U.S. economy by creating jobs directly related to wind energy including turbine manufacturing and installation, as well as permanent jobs for wind turbine operators and maintenance workers.