Sensewave is supported by Paris&Co
Make our cities smarter! Sensewaves’ technology aims at giving us a better understanding of our cities. The two founders studied sound waves for years at the French Scientific Center IRCAM. They developed a self-learning technology to predict behaviors and system failures in various fields…
You started with research on sound waves, which you are now applying to IoT – what are the highlights of Sensewaves’ technology ? To which fields can it be applied?
Sensewaves developed two offers, which are getting increasing traction on the market.
Adaptix.Core is our solution analysing data streams from any type of sensor. Our technology relies on extensive modeling and retrieval originally applied to sound data. The model we developed based on our research studies is now fully adapted to IoT applications. Its key benefit is to be able to analyse a raw data set as a series of events with specific duration and location. Cross analytics on IoT data streams may have many applications : to explain, to predict, to spot anomalies, to control, … Sometimes we need to take an extra step to provide specialized analysis tools, tailored to a specific industry. Our machine learning technology will adapt to the context it is launched into, by learning from scratch, even no dataset available a priori : it learns from its environment to become more and more reliable with time. It will even be capable of informing you know when your sensor becomes less reliable. Information gathered through these analyses can be fed into dashboards or integrated directly into our clients’ through API.
A key differentiating point of our technology is its transparency, there is no « black box » phenomenon : thanks to our adaptix approach, analysis comes together with a justification, so everything is clear !
Adaptix.Grid is our solution for electric grid dynamic optimisation. It wouldn’t be affordable to place sensors everywhere on the grid. Adaptix.Grid thus interprets smart meters and other sparse grid data available in order to accurately reconstruct the model of the complete electricity network (Picture 1) and its energy flows.
It can advise if some areas are ‘under stress’, as well as if there is equipment that faces outage and black out risks or premature wear. Our “virtual metering” accuracy is above 90%. Through a more profound, root-cause analysis, one can identify the reason behind the congestion and suggest corrective scenarios (Picture 2 & 3). It can for instance advise to modify a grid connection or to replace some devices. Our solution goes even further by advising the network manager on future grid expansion. It is miles away from information electricity providers have today – they often have to resort to their intuition to identify fragile zones, or just react once the issue has arisen.
Ex. Electricity Grid Modeling, focus on areas under pressure, Adaptix Grid.
You collaborate with Danish DSOs, is the French market not ready yet?
We mainly work with Scandinavian DSOs. The smart meters roll-out and the grid’s documentation level is more advanced there. We also collaborate with English DSOs, who aim at improving their grid development planning methodology. In Europe our market will gradually grow: by 2020, 80% of regular meters should be replaced by smart meters, according to a European directive. We are also starting to look into the North-American market.
What kind of services do you provide to make our buildings smarter, and who are your customers?
At building-level, our technology significantly improves the wellbeing of occupants in both homes and offices. Despite increasing automation of heating and cooling systems, most smart buildings do not provide the comfort of living they are expected to. Our systems automatically detect all irregularities such as under- or over-consumption of energy as well as any failure of equipment. Cisco and SETEC supported us to test this software through a proof of concept with the Datacity program: we installed it in Paris City Hall and at SETEC headquarters. Our technology prevents equipment breakdowns and predicts electrical consumption. It is scalable to a building stock. We aim to collaborate with energy managers who already have their own data management system and require smarter tools to get more accurate insights, easily and fastly. We are currently working with customers / early adopters in Portugal and Denmark. We plan to further develop the technology to improve the wellbeing of occupants. We already work with the Austrian luminaire leader Zumtobel on space optimization in offices, analysing smart lighting sensors data.
What are your next milestones?
Our goal this year is to accelerate Adaptix Grid roll-out in Denmark, by increasing our end customer reporting points from 40 000 to 500 000. As far as Adaptix Core is concerned, we are working on new partnerships with specialised IoT databases to extend their current functionalities through the Adaptix intelligent layer. We can address so many markets with this technology: smart lighting, shipping, industrial applications such as ventilation, heating and production chains…