Our core technology NILM
The energy consumption measured at the source (typically at the breaker box) is an aggregation of all appliances that are switched on. Our AI algorithm estimates the combination of unique electric waveforms each appliance has, from the ever changing aggregated waveform. The technology of understanding how much power is consumed by the major home appliances at home is known as NILM (Non-intrusive load monitoring), and researched by many companies around the world. However due to its complexity only a handful of companies have managed to monetise with it. It enables to grasp the breakdown of energy (*1) without installing numerous sensors around the home, or r to upgrade to expensive smart appliances.
*1 Up to 10 types of home appliances are subject to disaggregation: refrigerator, standby power appliances (including appliances that constantly consume electricity), air conditioner, rice cooker, microwave oven, washing machine, TV, IH cooking heater and high-heat home appliances such as heaters, dryers, kettles, etc. (As of August 2024)
Basic principles of disaggregation technology
Informetis' disaggregation technology is characterized by capturing and analyzing the characteristics of the current waveform. When home appliances are used, they generate current waveforms that have characteristics unique to each type of appliances. Since the current measured at the breaker box is the sum of the currents flowing through each home appliance operating at any given time, the AI algorithm matches the ever-changing main current waveform. By identifying the combination of current waveforms of home appliances, we can estimate which home appliances are operating at that moment.
However, there are dozens of home appliances used in the home, and the current state changes depending on the operating state, such as the strength of the vacuum cleaner, so the number of combinations is almost infinite. Informetis' disaggregation technology analyzes this huge combination of home appliances in real time and estimates which home appliances are in use, using the current waveform data of home appliances we have collected for over 10 years with our AI algorithms based on extensive research.
Example of disaggregation
The graph below shows the results of disaggregation of a typical day. We can visualise what types of home appliances were used, when, and for how long during the day.
24H View of a house
