Integrated modelling of contemporaneous multi-utility demand data from intelligent meters

Abstract

With the advent of smart metering in recent years where water and energy consumption data could be recorded at high resolution, several studies have been undertaken all over the world to unpack various benefits for both consumers and suppliers. Separate analysis and modelling of water or energy data has shown valuable state-of-the-art applications to inform single and multi-utilities and regulatory agencies. This paper suggests a new concept where high resolution multi-utility data is concurrently collected and modelled to allow for enhanced pattern recognition of other resources (e.g. having electricity data assists pattern recognition of water), deeper insight into customer demand and optimal opportunities to manage it. Through using a smart device to capture concurrent water and energy consumption in near real-time, and exploring the correlation between these two consumption activities, the proposed system has helped avoid the need of using Hidden Markov Model and Dynamic Time Warping algorithms in several analysis stages, thus allowed the classification process to be undertaken much faster with higher achieved accuracy. Once finished, the system will result in a wide range of benefits for utilities and regulatory agencies, especially allowing them to have a unique single platform to monitor all water and energy consumption of any particular household or region in near real-time to immediately identify faulty issue with the power system or pipe leakage if it happens. For customer, they will also be immediately alerted when there is any single problem occurring to any water or energy device, or be informed about the current efficiency status of water and power appliance in the house, as well as receiving various incentives when they follow instruction to improve the current supply network.