Intelligent decision support for marine oil spill response in harsh environments

恶劣环境下海上溢油应急响应智能决策支持

• 批准号: RGPIN-2020-05002
• 负责人: Chen, Bing
• 金额: $ 6.41万
• 依托单位: Memorial University of Newfoundland
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=717486
• 关键词: Intelligent; decision; support; marine; oil

Marine oil spills (e.g., 2010 BP spill) can not only cause significantly negative impacts on the environment and socio-economy but constitute a long-term hazard to marine life and human health. The response to a spill is a time-sensitive, dynamic, multifaceted and complex process, depending on various factors e.g., oil quantity/properties, location, environmental conditions, response resources, and human judgement/behaviours. The success much relies on how efficiently and effectively the information and resources can be utilized and how optimally and quickly the decision and actions can be made. It is further challenged by harsh marine environments (low temperature, rough sea, sea ice, etc.) prevailing in Canadian offshore areas such as the Arctic/Northern Atlantic oceans. The response window of opportunity is very short due to technical/logistic difficulties, safety, and rapid oil movement. This has been evidenced by three recent spills in Newfoundland offshore. Thus, dynamic and efficient response decision support is much desired but still lacking of research and practice. Furthermore, human factors/errors contribute 60-80% in spill accidents but without good understating of their effects on spill response. Therefore, the research is to help address these challenges by developing an intelligent Marine Oil Spill response Decision Support System (iMOS-DSS) for harsh environments with focuses on: 1) comprehensive review and development of marine oil spill i-Base (oil/spill databases and a response knowledge base); 2) intelligent agent and human factor analysis aided spill simulation; 3) deep learning aided response process simulation-optimization coupling supported by decanting experiments (skimming, separation, treatment, and disposal); 4) artificial intelligence aided marine oil spill response decision support; and 5) virtual reality aid visualization for marine oil spill response decision making. The most commonly used response option (and only legally allowed in Canada), decanting process, will be mainly targeted for lab experiment to support process simulation and optimization work. System validation and evaluation will be further conducted via local spill responders' classroom and at-sea training exercises in NL. The proposed program will 1) produce new/advanced knowledge and methodologies to address urgent needs in improving response capabilities in Canada and beyond; 2) enrich and improve the theories and technologies of environmental emergency management and decision making in general; 3) train next-generation professionals and researchers through a cross-disciplinary, equitable, diverse and inclusive (EDI) research/training environment; 4) help position my group and university globally in marine oil spill research to expand international collaborations/impacts and apply the outcomes worldwide; and 5) bring significant short-/long-term benefits by supporting responsible development of industry and protecting our marine environments.

海洋漏油事件（如2010年英国石油公司漏油事件）不仅会对环境和社会经济造成重大负面影响，而且对海洋生物和人类健康构成长期危害。对泄漏的响应是一个时间敏感、动态、多方面且复杂的过程，取决于多种因素，例如石油数量/性质、位置、环境条件、响应资源和人类判断/行为。成功在很大程度上取决于如何高效和有效地利用信息和资源，以及如何最佳和快速地做出决策和行动。北极/北大西洋等加拿大近海地区普遍存在的恶劣海洋环境（低温、波涛汹涌、海冰等）进一步挑战了这一点。由于技术/后勤困难、安全性和石油快速流动，响应机会窗口非常短。最近纽芬兰近海发生的三起泄漏事故就证明了这一点。因此，人们迫切需要动态、高效的响应决策支持，但仍缺乏研究和实践。此外，人为因素/错误在泄漏事故中占 60-80%，但尚未充分认识到其对泄漏响应的影响。因此，本研究旨在通过开发适用于恶劣环境的智能海洋溢油响应决策支持系统（iMOS-DSS）来帮助应对这些挑战，重点是：1）全面审查和开发海洋溢油i-Base（石油/溢油）数据库和响应知识库）； 2）智能代理和人为因素分析辅助泄漏模拟； 3）由倾析实验（撇取、分离、处理和处置）支持的深度学习辅助响应过程模拟-优化耦合； 4）人工智能辅助海上溢油应急决策支持； 5) 虚拟现实辅助海上溢油响应决策的可视化。最常用的响应选项（仅在加拿大合法允许），即倾析工艺，将主要针对实验室实验，以支持工艺模拟和优化工作。系统验证和评估将通过荷兰当地泄漏应急人员的课堂和海上培训演习进一步进行。拟议的计划将 1) 产生新的/先进的知识和方法，以满足提高加拿大及其他地区响应能力的迫切需求； 2）丰富和完善环境应急管理和决策的总体理论和技术； 3）通过跨学科、公平、多元化和包容性（EDI）的研究/培训环境培训下一代专业人员和研究人员； 4) 帮助我的团队和大学在海洋溢油研究领域在全球范围内定位，以扩大国际合作/影响并将成果应用于全球； 5) 通过支持负责任的工业发展和保护我们的海洋环境，带来显着的短期/长期效益。