Three years with my smart home – a self-test

Changes in life open up new opportunities. When I started working for Bosch in Automotive Electronics, relocating into my new, rented apartment in Reutlingen, Germany, gave me a chance to install my private smart home from scratch – leveraging all my prior existing household appliances and devices. Open source software and cheap hardware provided a low entry level for me, and after having worked for 15 years in consumer electronics, it was a fun, distracting project. Since then my smart home and I have been living together for three years, so I can come to a few longer-term conclusions. Let me first say it’s important to understand that the results are individual (rental apartment, daily life) and cannot be generalized. Now let me briefly introduce you to my set-up and what I’ve learned.

First things first: moving into a new environment allowed me to install two separate power circuits per room, one for the devices that need to be powered all the time such as refrigerator, router or landline phone, and one for those that can be switched off.

Yearly electrical power consumption

Yearly electrical power consumption

Then, to use electricity as efficiently as possible, I created three different modes for energy management: night shutdown, wake-up alarm, and presence detection. Night shutdown starts every day at 1 a.m. and ends with the (too early) ring of my daily wake-up alarm. Connecting the alarm clock wirelessly allows me to personalize the system on a daily basis. I used the same principle for presence detection. My smart home system detects whether my mobile phone is logged in to my local WLAN and switches devices on or off accordingly.

Did you know that houses are the number one consumers of energy? So heating efficiency was next on my agenda. For managing my heating consumption, I needed to analyze the heating and cooling parameters. I think there is lots of potential for future smart home systems to automate this data mining step. Based on the three abovementioned modes of power consumption management, I switch to night temperature three hours before starting the night mode, e.g. at 10 p.m., and start heating to my desired room temperature again 90 minutes before my alarm clock rings. Finally, when leaving my apartment (as detected by the WLAN presence detection), my heating system automatically switches to the ‘out of home’ temperature after 15 minutes.

The missing service for heating management in a rented apartment is what I often call the ‘Feierabend mode.’ How can I forecast my daily arrival time at home after work? The most straightforward approach was to track the WLAN logins and logoffs and calculate a usual quitting time from the data collated for each working day.

Feierabend” means quitting time in German. It is composed of the nouns “Feier” (celebration) and “Abend” (evening), and was in former times used to denote the “Vorabend” – the eve, or evening before, a “Feiertag” (holiday). There is no direct literal translation in English.

Yearly heating consumption

Yearly heating consumption

 

On the one hand, I wanted the Feierabend mode to increase my comfort, by making sure my apartment is at a comfortable temperature when I get home from a busy day at work. On the other hand, it soon was very clear that the Feierabend mode is also my biggest driver of heating costs: the better my Feierabend mode is defined, the more positive the impact on my heating costs, which split into 40% on weekends and 60% on weekdays. I am currently experimenting with different options using connected devices to improve this mode. For example, I’ve installed a dedicated smart home calendar to sync private calendar events that affect my usual quitting time. Another approach I’m currently testing is to implement a geofence application into my system to track my departures and trigger the heaters accordingly.

The four morning, Feierabend and night  modes for heating consumption management

The four morning, Feierabend and night modes for heating consumption management

 

Despite power and heating management, the four modes defined can be used for additional comfort and security applications: heating shutdown triggered when opening windows, push messages for forgotten open windows, reminders to open windows at high humidity in rooms, and of course remote control of all devices via my mobile phone, ….

There are several lessons to be learned from the past three years:

  • A smart home installation driven by personal events is able to compensate the increased cost of electrical power and heating by optimizing consumption. As a tenant, I benefit from stable utilities charges and from the knowledge that my budget for them will not have to change.
  • For me, the most time-consuming activity was to analyze and model the heating behavior that fits my lifestyle and work habits.
  • Looking at my installation from an ROI (Return on Investment) perspective, it became positive in the third year.
Breakdown of additional property expenses

Breakdown of additional property expenses

 

As an engineer and smart home fan, I hope that in the future all device-related activities will be automated by different intelligent learning algorithms: all controlling events could be triggered by people’s activities and related to connected devices already in use e.g. carrying a mobile phone to work or setting an alarm clock at night. The combination of an intelligent environment at home, work or on the go, triggered by people’s activities and leveraged by connected devices, is what I think of as a real benefit of the Internet of Things.

Consider a smart home you want to live in – what would be your desired modes to make it more energy efficient, comfortable or secure? 

 

About the author

Andreas Schaller

Andreas Schaller

Dr. Andreas Schaller joined Robert Bosch’s Automotive Electronics Division in 2011 working in Technology Management. Since 2016 he manages the Robert Bosch Center of Competence “Connectivity”, a Network of Expert across different Robert Bosch business divisions. In 2009 he founded ASC – Technology Consulting to support the European businesses in getting active in the Internet of Things. He previously worked for Motorola Labs, Germany, managing the Motorola Labs research area ‘Short Range RF Communications’ in Europe. Andreas received a PhD in Manufacturing Engineering from the University of Erlangen-Nuremberg, Germany.