Data-Driven Predictive Control of Energy Storage in Thermal Inertia

为了热舒适而控制建筑物内的温度，以及为了安全地储存产品而控制冷链中的温度，占全球二氧化碳排放量的15%。在为调节温度而部署的供暖和制冷设备中，空调和电风扇占全球用电量的10%--其中太多是由于控制不力而浪费的。这种设备大多采用不灵活的控制体系结构运行，不能适应人、产品或电网的时变要求。创建更好的、数据驱动的算法来优化对这种设备的控制，以减少能源使用或使用时的换挡，将被证明是一种具有成本效益的减排战略，并构成这个项目的基础。该项目涉及几个EPSRC研究主题，特别是：能源效率、人工智能技术和能源储存。

Controlling temperature in buildings for thermal comfort, and in the cold-chain for safe product storage, accounts for 15% of global CO2 emissions. Of the heating and cooling equipment deployed to regulate temperature, air conditioners and electric fans account for 10% of global electricity use - too much of which is wasted through inefficient control. This equipment is mostly operated with inflexible control architecture that does not adapt to the time-varying requirements of people, products or the electricity grid. Creating better, data-driven algorithms that optimise the control of this equipment to reduce energy-use, or shift when it is used, would prove a cost-effective emission mitigation strategy, and forms the basis of this project.The project cuts across several EPSRC research themes, notably: energy efficiency, artificial intelligence technologies, and energy storage.