Real-time monitoring of appliances can be achieved in an intrusive way by installing smart plugs in every target appliance, or non-intrusively, by detecting appliance-specific patterns on smart meter data. The latter option is more cost-efficient, although it demands high-quality software. This white paper focuses on real-time monitoring of washing machines on edge.

The value that a real-time monitoring use case offers for end-users is that they can check the status of their washing machine. Based on research findings, one of the most desired use cases for customers is to get a push notification when the washing machine is finished. This is a popular feature for connected washing machines. However,  most washing machines are not connected yet and the alarm signal produced, when the wash is finished, cannot be heard everywhere in the house. This is why NET2GRID has invested in a simple but accurate detection method that works with any type of washing machine. 

From a technical point of view, such services usually run on high-end microprocessors or in the cloud, thus increasing the cost. NET2GRID has developed a lightweight model for real-time washing machine monitoring using state-of-the-art deep learning techniques that can be efficiently integrated with low-cost chipsets.