Development of an AI-Based regulatory control for an Energy Management Information System (EMIS) to improve occupants health and comfort

迫切需要将数据驱动模型与能源管理信息系统（EMIS）和室内空气质量（IAQ）控制工具联系起来，以便在实时设置中利用它们并评估它们对整个建筑物的影响。这些影响可能与居住者相关（例如，热舒适和健康），与气候相关（例如，能源使用和排放），与设备相关（例如，暖通空调系统老化）以及与成本相关（例如，能源账单，前期成本，病假等）。EMIS的目标是收集、分析、优化并以建议调整、问题和报告的形式向建筑运营商/控制人员提供有用的信息。这在医院、学校和工厂等具有洁净室的敏感建筑物中尤其重要（提供COVID-19环境），以便清楚地了解建筑物的操作/控制。该项目的主要目标是将人工智能技术，特别是预测分析任务整合到建筑暖通空调系统中，以保持建筑的健康、舒适和节能，同时降低运营成本。对这种集成的需求来自：1)基于云的EMIS中记录的与BMS系统和外部资源（天气数据和水电费）相关的大量数据的可用性；2)行业对分析的需求日益增长，以提高效率，但目前受到严格预算和资源的限制；3)大数据分析工具、机器学习算法和强大的计算资源的可用性，以从数据中提取模式，用于基于预测的分析。学术和工业团队将在未来两年内合作，为这种基于云的EMIS配备基于人工智能的分析工具，并验证/测试其提高效率的能力。

There is a vital need to link data-driven models with Energy Management Information System (EMIS) and Indoor Air Quality (IAQ) control tools to leverage them in real-time settings and evaluate their effects in buildings as a whole. These effects could be occupant related (e.g., such as thermal comfort and health), climate related (e.g., energy usage and emissions), equipment related (e.g., aging of HVAC systems), and cost related (e.g., energy bills, upfront costs, sick leaves, etc.). The EMIS goal is to collect, analyze, optimize, and deliver useful information to the building operators/controls in forms of recommended adjustments, issues, and reports. This is particularly of growing importance in sensitive buildings such as hospitals, schools, and factories with clean rooms (provided the COVID-19 circumstances) to have a clear view of the building operation/controls. The main objective of the proposed project is to establish R&D into the integration of AI technologies, specifically predictive analytics tasks into building HVAC system in order to keep the building healthy, comfortable, and energy efficient with low operating costs. The need for this integration comes from: 1) Availability of massive data related to BMS system and external resources (weather data, and utility bills) recorded in the cloud-based EMIS, 2) The growing need of the industry to have analytics in place to improve efficiency which are currently limited by stringent budgets and resources, and 3) Availability of big data analytics tools, machine learning algorithms, and powerful computational resources to extract patterns from the data for prediction based analytics. The academic and industrial teams will collaborate over the next 2 years to equip this cloud-based EMIS with AI-based analytics tool and validate/test its efficiency-enhancement capability.