Intelligent model to categorise mechanised water end uses

Abstract

The current population boom has pushed demand on residential land in most world metropolitan cities to its extreme limit, and also imposed significant pressure on water security. To ease this problem, construction of residential apartment building has been the top priority of the governments. With the rapid advancement in technology, especially in water metering area, the next generation of these buildings is expected to not only offer their residents a comfortable living area but also make the water supply and management become much more effectively controllable. To contribute to this goal, the aim of this study was to develop an autonomous and intelligent system for residential water end-use classification that could interface with customers and water business managers via a user-friendly smartphone and webbased applications. The authors recently developed an intelligent application called Autoflow© which served as a prototype tool to solve the complex problem of autonomously categorising residential water consumption data into a registry of single and combined events. This model was developed using data collected in several cities in Australia, and when applied on single residential property, the achieved accuracy ranged from 86% - 96%. The only limitation of Autoflow© is that when being deployed overseas, a new water end use dataset must be collected to retrain the model to adapt to the new regions. The ultimate goal of this study is to develop the next Autoflow© generation, called AutoflowAB, which can be applied in all future apartment buildings to help disaggregate water consumption autonomously into the six main categories including: shower, toilet, tap, clothes washer, dishwasher and evaporative air cooler without relying on any previously collected data for training. The key employed techniques are Dynamic Time Warping (DTW) and Decision Tree. The model has been tested on 150 residential properties in Australia where the accuracy ranges from 87 – 94 %, and also planned to be applied on 30 new 7-story buildings in the Commonwealth Games Village located in Gold Coast Australia.