EDI-Net: A smarter approach to city-scale energy management

摘要

Energy metering data is becoming 'smart' - sub-hourly and downloaded at regular intervals in a day - but the software which is used to display such data remains 'dumb'. Such software relies on the experience of the analyst to identify instances where consumption is outside of a typical pattern. The analyst must trawl through data looking for problems. Though alarms can be set in 'dumb' software - these alarms are also dumb -typically an energy manager sets maximum and minimum thresholds and is notified when they are exceeded. This paper presents 'smart' software and advocates for a more innovative approach to energy management. The software maintains a detailed model of consumption patterns across a whole portfolio of buildings and communicates energy performance in a user friendly way. Rather than simply alarming on peaks and troughs it can track unexpected levels of consumption and manage a list of exceptions. The approach opens up energy management to a wider community of stakeholders through the use of simple visualisation, coherent information architecture and a flexible communication platform. It is now possible to analyse large volumes of consumption data automatically with software, notifying the energy manager only when something unexpected occurs. With a continually evolving model of consumption patterns across the entire portfolio smart software can present stakeholders with interactive, high-level views on the performance of the portfolio as a whole whilst allowing a user to 'drill' down to see the detail on demand. Reports can be designed for energy experts or for other professionals such as finance managers and decision makers enabling better communication across an organisation and more effective energy management. The Energy Data Innovation Network (EDI-Net) project is developing an approach to energy management which engages a community using modern, scalable communication tools with an ambitious approach to modelling, analysis and visualisation of large datasets.