Prognostic Methods for Battery Management Systems

电池管理系统的预测方法

• 批准号: 1234451
• 负责人: Michael Pecht
• 金额: $ 24万
• 依托单位: University of Maryland, College Park
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-09-01 至 2015-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1234451&HistoricalAwards=false
• 关键词: Prognostic; Methods; Battery; Management; Systems

The objective of this research is to develop prognostic methods for improving the safety and availability of battery-powered systems such as electric vehicles and unmanned aerial vehicles. Battery-powered systems suffer from two problems. The first is "range anxiety" problem, which refers to the fear of running out of battery power during vehicle operation. The second problem is related to the safety of a battery pack, which can rupture or even explode under certain conditions. To address these problems, an approach for predicting the (1) end-of-discharge (the time at which a battery will run out of electrical charge), and (2) remaining-useful-performance of a battery with a known level of confidence will be developed. This approach involved machine learning algorithms to address future loading conditions, unit-to-unit variations, and modeling uncertainties. This research will improve the operation readiness and safety of battery-powered systems that are used in applications ranging from commercial (electric vehicles) to defense (unmanned aerial vehicles) sectors. In 2011, President Obama announced his goal of having one million electric vehicles on the road by 2015. The proposed research will make significant contributions to reaching this target, since it can ease user concern about the safety and reliability of electric vehicles by providing robust battery state and health information in real-time, thus encouraging their widespread use. As a result, it will also decrease US dependence on foreign oil and reduce the emission of greenhouse gases. The content of this research will assist in the advancement of prognostics and health management techniques which will be disseminated to the engineering community through online graduate courses, seminars, workshops, and short courses thereby benefiting people from academia, industry and the military.

本研究的目的是开发预测方法，以提高电动汽车和无人驾驶飞行器等电池供电系统的安全性和可用性。 电池供电系统有两个问题。首先是“里程焦虑”问题，指的是车辆运行过程中担心电池电量耗尽。第二个问题与电池组的安全性有关，电池组在某些条件下可能破裂甚至爆炸。为了解决这些问题，将开发一种用于预测（1）放电结束（电池将耗尽电荷的时间）和（2）具有已知置信水平的电池的充电有用性能的方法。这种方法涉及机器学习算法，以解决未来的负载条件，单元之间的变化和建模的不确定性。这项研究将提高电池供电系统的操作准备和安全性，这些系统用于从商业（电动汽车）到国防（无人机）等领域的应用。2011年，奥巴马总统宣布了到2015年拥有100万辆电动汽车的目标。拟议的研究将为实现这一目标做出重大贡献，因为它可以通过实时提供强大的电池状态和健康信息来缓解用户对电动汽车安全性和可靠性的担忧，从而鼓励其广泛使用。因此，它还将减少美国对外国石油的依赖，并减少温室气体的排放。这项研究的内容将有助于推进生物学和健康管理技术，这些技术将通过在线研究生课程，研讨会，讲习班和短期课程传播给工程界，从而使学术界，工业界和军方的人们受益。