An understanding of the local weather is very useful to a contextual smart home. Whilst there are main types of rainfall sensor readily available, we also wanted to add a real-time understanding of whether it was raining or not. This would then allow our smart home to make voice announcements, send notifications or take necessary action (close roof vents, etc.). We do this using an optical rain sensor connected to our smart home. Having tried many other types of sensors, this proved to be the most accurate, reliable and responsive sensor we have found.
The RG-11 rain sensor manufactured by the Hydreon Corporation is connected to our home using an Arduino and our digital input board as it requires 12V dc power. It is very responsive to rain and has adjustable sensitivity. We often get notifications of it raining before we have noticed it. It works by sensing water on its surface using beams of infrared light and this is the same principle used by in cars by their automatic windscreen wiper controls.
You can't really use a flood sensor for real-time detection of rain. Rain drops would need to pool before a flood sensor would be any use and this means it could be many minutes before rain is actually detected. Rain water is also very pure and no flood sensors on the market will work reliably with pure water like this.
Our contextual smart home can easily be configured to send us real-time notifications when it starts raining or stops raining. This allows us to bring in washing from the washing line before it gets wet.
Our @smartest_home will occasionally tweet in real-time about the optical rain sensor.
Our contextual smart home will make a voice announcement when it starts raining, based on our optical rain sensor. It doesn't do this every time though as this would quickly become very annoying. It only does this if it hasn't rained for at least 24 hours.
