PREDICTOR: Improving the energy performance of closed-loop heating and cooling systems.

研究假设，基于生命周期成本，决策支持工具能够在闭环HVAC系统中实现最佳维护干预，可以显著降低建筑物的能耗。该项目的目的是开发决策支持工具，以预测何时及时的维护干预可以经济有效地优化建筑物中闭环HVAC系统的运行和维护。我们的目标是通过实现以下五个目标来验证这一假设：1)通过分析调试和维护数据与设计的能源性能，调查不同闭环HVAC系统的生命周期性能。2)开发数据驱动的暖通空调性能模型，使用与“实时”传感器数据相关的机器学习技术，为系统运行效率的决策支持工具提供建议3)开发暖通空调系统的决策支持工具（基于调试和运行数据，以及创新的生命周期成本效益分析），与建筑能源管理框架相关联，实现实时干预，以优化维护和能源效率。4)测试和评估拟议的非住宅建筑能源管理决策支持系统的原型，使整个行业能够采用，并利用工业的商业化机会。5)传播研究结果，开发一个简单、互动的教育工具，并为暖通空调系统的设计、调试和维护制定标准建议和指导方针。博士时间表：第一年：建立一个实验平台，并将能源消耗与维护和设计数据联系起来第二年：使用数据驱动模型人工神经网络和生命周期成本开发暖通空调性能第三年：评估和测试原型系统，开发交互式教育工具，撰写并提交两篇期刊论文

The research hypothesis is that decision support tools enabling optimum maintenance interventions in closed loop HVAC systems, based on life cycle cost, can deliver significant energy reduction in buildings. The aim of this project is to develop decision support tools to predict when timely maintenance intervention can cost-effectively optimize the operation and maintenance of closed loop HVAC systems in buildings. We aim to test the hypothesis by achieving the following five objectives: 1) Investigate the life cycle performance of different closed loop HVAC systems by analysing commissioning and maintenance data against designed energy performance. 2) Develop data driven HVAC performance models, using machine learning techniques linked to 'live' sensor data, to advise the decision support tool regarding system operational efficiency 3) Develop a decision support tool for HVAC systems (based on a commissioning and operational data, and innovative lifecycle cost benefit analysis), linked to a Building Energy Management framework, enabling real-time intervention to optimise maintenance and energy efficiency. 4) Test and evaluate a prototype of the proposed decision support system for energy management in non-domestic buildings, to enable adoption across the sector and to exploit commercialisation opportunities with industry. 5) Disseminate the results, develop a simple, interactive educational tool and create standards recommendations and guidelines for design, commissioning and maintenance of HVAC systems. PhD timetable: Year 1: Setting up an experimental rig and relating energy consumption to maintenance and design dataYear 2: Developing HVAC performance using data driven models ANN and life cycle costYear 3: Evaluate and test prototype system and develop interactive educational tools, writing up and submitting two journal papers