S&AS: INT: Autonomous Battery Operating System (ABOS): An Adaptive and Comprehensive Approach to Efficient, Safe, and Secure Battery System Management

S

• 批准号: 1724227
• 负责人: Nathan Fisher
• 金额: $ 125万
• 依托单位: Wayne State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-09-01 至 2023-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1724227&HistoricalAwards=false
• 关键词: S& INT; Autonomous; Battery; Operating

Effective battery systems are a cornerstone of the next wave ofinnovation. Emerging technologies including electric-drive vehicles(EVs) and stationary energy storage systems require integratedbattery systems to provide sustained and reliable power. A uniqueopportunity exists for the software-controlled battery managementsystem (BMS) to play a crucial role in enabling continuedinnovation. However, there are several unaddressed challenges tocurrent BMS design. A battery's decreasing capacity over time dueto aging necessitates that BMS automatically adapt to batterychanges. Furthermore, future BMS systems will require autonomousreasoning capabilities to make economically-sound decisions onusers' behalf (e.g., scheduling battery charging times in apersonalized fashion). This project injects intelligencecapabilities into BMS design with the development of the AutonomousBattery Operating System (ABOS). ABOS enables moreenergy-efficient, long-lasting, and secure battery-driven systems.Furthermore, the PIs incorporate computational and cyber-securityaspects of ABOS design into their undergraduate- and graduate-levelcourses.ABOS advances the science of autonomous system design withsimultaneously introspective and extrospective learning. ABOS willlearn and adapt to user-initiated charging/discharging patterns,reason about how these patterns affect the battery'sstate-of-health, and respond to potential faults or attacks. ThePIs and their external collaborators will develop a hybrid EVsimulation environment to test the ability of ABOS to control aphysical battery system. The simulation environment will evaluatethe effectiveness of ABOS in predicting battery state, minimizingcost of operation, and handling failures and threats.

高效的电池系统是下一波智能化浪潮的基石。包括电动汽车（EV）和固定式储能系统在内的新兴技术需要集成电池系统来提供持续可靠的电力。软件控制的电池管理系统（BMS）在实现持续创新方面发挥关键作用的独特机会。然而，目前的BMS设计存在一些未解决的挑战。电池的容量随着时间的推移，由于老化而下降，BMS需要自动适应电池的变化。此外，未来的BMS系统将需要严格的推理能力来代表用户做出经济合理的决策（例如，以个性化的方式调度电池充电时间）。该项目通过自主电池操作系统（ABOS）的开发将智能能力注入BMS设计。ABOS能够实现更节能，更持久，更安全的电池驱动系统。此外，PI将ABOS设计的计算和网络安全方面纳入其本科和研究生课程。ABOS通过同时进行内省和内省学习来推进自主系统设计科学。ABOS将学习和适应用户发起的充电/放电模式，推理这些模式如何影响电池的健康状态，并对潜在的故障或攻击做出响应。PI和他们的外部合作者将开发一个混合EV模拟环境，以测试ABOS控制物理电池系统的能力。模拟环境将评估ABOS在预测电池状态、最大限度地降低运营成本以及处理故障和威胁方面的有效性。

项目成果

Energy-Aware Resource Management in Vehicular Edge Computing Systems

• DOI: 10.1109/ic2e48712.2020.00012
• 发表时间: 2020-04
• 期刊: 2020 IEEE International Conference on Cloud Engineering (IC2E)
• 作者: Tayebeh Bahreini;Marco Brocanelli;Daniel Grosu

Explaining Deep Neural Network Models with Adversarial Gradient Integration

用对抗梯度积分解释深度神经网络模型

• DOI: 10.24963/ijcai.2021/396
• 发表时间: 2021
• 期刊: Thirtieth International Joint Conference on Artificial Intelligence (IJCAI
• 作者: Pan, Deng;Li, Xin;Zhu, Dongxiao
• 通讯作者: Zhu, Dongxiao

A comparison study of intel SGX and AMD memory encryption technology

• DOI: 10.1145/3214292.3214301
• 发表时间: 2018-06
• 期刊: Proceedings of the 7th International Workshop on Hardware and Architectural Support for Security and Privacy
• 作者: S. Mofrad;Fengwei Zhang;Shiyong Lu;W. Shi

DexLego: Reassembleable Bytecode Extraction for Aiding Static Analysis

• DOI: 10.1109/dsn.2018.00075
• 发表时间: 2018-03
• 期刊: 2018 48th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN)
• 作者: Zhenyu Ning;Fengwei Zhang

Pythia-MCS: Enabling Quarter-Clairvoyance in I/O-Driven Mixed-Criticality Systems

• DOI: 10.1109/rtss49844.2020.00015
• 发表时间: 2020-12
• 期刊: 2020 IEEE Real-Time Systems Symposium (RTSS)
• 作者: Zhe Jiang;Kecheng Yang;N. Fisher;N. Audsley;Zheng Dong