Study of an innovative non intrusive load monitoring system for energy emancipation of domestic users: hardware and ICT optimized solutions

Non-intrusive appliance load monitoring (NIALM) is the process of disaggregating a household s total electricity consumption into its contributing appliances. Smart meters are currently being deployed on national scales, providing a platform to collect aggregate household electricity consumption data. Existing approaches to NIALM require a manual training phase in which either sub-metered appliance data is collected or appliance usage is manually labelled. This training data is used to build models of the household appliances, which are subsequently used to disaggregate the household s electricity data. Due to the requirement of such a training phase, existing approaches do not scale automatically to the national scales of smart meter data currently being collected. In this thesis an unsupervised disaggregation method is presented which, unlike existing approaches, does not require a manual training phase. A NIALM system reads real-time data from a smart meter, usually positioned at the point on the public electricity network at which the customers is connected, and uses algorithms not only to quantify how much energy is used in the home, but also to determine what main devices are being operated. NIALM algorithms need a complete load signature and complex optimization algorithms to find the right combination of single loads that fits the real electrical measurements. It is practically impossible to get the detailed signature of all appliances inside a house/building and sophisticated optimization algorithm are not suitable for on-line applications. To do so, we address the following topics. First, a straightforward NIALM algorithm is proposed, it is based on both a simple load signature, rated active and reactive power and a heuristic disaggregation algorithm. Second, on real applications, this approach cannot reach very high performances; this is the reason why an active involvement of users is considered. The users feedback aims to: correct the load signatures, reduce the error of disaggregation algorithm and increase the active participation of users in saving energy politics. Third, the NIALM algorithm has been accurately tested numerically using as input load curves generated randomly but under given constraints. In this way, the causes of inefficiency of the proposed approach are quantitatively analyzed both separately and in different combinations. The above contributions provide a solution which satisfies the requirements of a NIALM method which is both unsupervised (no manual interaction required during training) and uses only smart meter data (no installation of additional hardware is required). When combined, the contributions presented in this thesis represent an advancement in the state of the art in the field of non-intrusive appliance load monitoring, and a step towards increasing the efficiency of energy consumption within households.