EcoHouse Research

Michael Fallows (MEng, 2013)

 

This project is in association with the EcoHouse Initiative (the Initiative), a Cambridge based organisation conducting research and design to enable sustainable development. One way the Initiative makes a difference is by improving the design of houses in Latin America to make them more sustainable and provide better conditions for the families who live in them. EcoHouse works with a Latin American Non Governmental Organisation (NGO), Un Techo para mi Pais (UTPMP), who construct simple housing for low income families across Latin America.

 

There is a proven link between housing conditions and health. After a literature review and discussion with EcoHouse members, temperature and humidity were determined as the most significant factors which should be measured over a period of months to help inform design. Other factors such as light levels, air quality, noise and structural performance were considered but not pursued.

 

The aim of this project is to develop a sustainable and affordable means of collecting necessary and appropriate data from houses in Latin America over a period of months and to present the data to selected EcoHouse members and partners, primarily in Cambridge, UK. In the short term, this will be achieved by developing a low cost wireless sensor network for deployment in Quito, Ecuador. Designated houses will have a set of sensors connected to a sensor hub, which will transmit the readings to a central gateway. The gateway will upload the data to an online database for viewing and downloading by authorised persons. In the long term, the objectives of this project are not purely technical. The project should also develop a strategy for longer term development within EcoHouse and outside the organization, enabling local people in Quito and others across the world to use the research and expand on it.

The sensor hubs use an XBee transceiver to communicate data wirelessly. This uses the open ZigBee protocol to provide long range and low power consumption compared to alternative methods. The XBee module and electronics will ideally be enclosed in a locally sourced plastic container, though alternatives have been specified. A battery pack is included which uses 4 AA batteries, estimated to last up to four months. Three temperature sensors are used to measure outdoor temperature, temperature inside at 1m and temperature inside at 2m. The sensors are low cost and an enclosure has been designed for weather and child proofing. A humidity sensor is located with the main hub and measures relative humidity in % RH.

 

The gateway aggregates data from the hubs and uploads it to a database online. Initially an Arduino microcontroller was used but ultimately a Raspberry Pi computer was chosen, delivering cost savings and expansion potential. Data is received using an XBee module connected to the Raspberry Pi using a USB adapter. The Raspberry Pi runs a variant of the Linux operating system which has been customised to run a specially developed program at start-up. This reads the data from the XBee module and then uploads it to a specified web address. The internet connection is provided by a mobile router which uses a SIM card from a local mobile network provider.

 

The online platform was built using standard web technologies. A single page is used to receive the data sent by the Raspberry Pi and store it in a database. To access the data, authorised persons must login with a unique username and password. The user can configure new hubs, download the data or view it graphically.

 

At two significant points in development, complete prototypes were built and tested. The first prototype was tested early in the project which demonstrated the feasibility of the proposed solution and identified key areas for improvement. The second was tested in a prototype of the latest EcoHouse design for Brazil at Jesus College, Cambridge, and in a residential property. This gave a more realistic trial of implementation and data was compared with that collected by significantly more expensive professional data loggers. Readings did differ but were generally within an acceptable range. A number of improvements were subsequently made to the gateway set up and sensor attachments. EcoHouse members were encouraged to try the monitoring website and feedback was used to design a more user-friendly graphical interface.

 

An important objective of the project was to ensure long term adoption and evolution. This will be achieved by sharing the research within the EcoHouse Initiative, within the communities where the system is deployed and online. By doing this, the people whose houses may be improved by the work of EcoHouse are empowered to contribute to the process themselves. The Initiative can use the data not only to help design but also to present the case for reforms and improved housing conditions to politicians. Finally, making resources available online means that anyone can learn from what has been achieved and in turn use the system themselves or contribute to its further development.

 

The system is to be deployed in Quito in summer 2013 by the researcher and the EcoHouse Information Systems team. They will work with UTPMP and the families who will benefit from the project to install hubs at several properties. Members of the EcoHouse student society and academic research cluster have already shown enthusiasm for the project and will continue to be encouraged and enabled to use the data.

The real achievement of this project will not be a technical one; it will be changing lives by improving housing and encouraging politicians and other organisations to act where housing conditions are threatening the health of the families who live in them.