new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0], j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src= '//'+i+dl;f.parentNode.insertBefore(j,f); })(window,document,'script','dataLayer','GTM-WGWP2C');

Using big data to revolutionize energy consumption 

August 28, 2015, 10:02am

We are currently producing more data in 48 hours than have been produced since mankind first walked the earth. The volume of what is known as “big data” is such that it is impossible, in practical terms, to process it with conventional databases. Once processed, however, it is very useful, with many benefits for society. The data validation institute (Institut de Valorisation des Données – IVADO) at Polytechnique Montréal is working to find ways of putting this valuable information to use. Its current priorities are the transport and logistics industries, energy, finance and productivity.

Professor Michel F. Anjos is searching for mathematical solutions to the problems of optimizing energy consumption. In Quebec’s particular case, the problems are posed by wide fluctuations in energy demand due to intense cold and, more generally, peak periods of energy consumption. The team led by Professor Anjos is therefore trying to discover how processing big data could help us regulate spikes in consumption, and spread energy demand more evenly throughout the day.

The public is already being asked to reduce consumption at peak periods, but the response is not something that can be controlled. The next step would be to bring about a shift in demand outside peak hours, with no impact on how consumers live their lives. This is where big data can prove useful. For example, we could program water heaters to warm water in advance of peak hours, so that they would only have to maintain its temperature until it was used. To avoid inconvenience to consumers, however, we need an algorithm to program the water heaters. At what time should they go to work? For how long? The algorithm also has to take into account the specific requirements of every home. The development of optimum mathematical solutions for these problems will be based on big data.

As a second step, we could support the development of a participatory system by encouraging small-scale local energy production. Using solar panels, small wind-power generating units and other systems, households could produce energy for their own consumption and make up any shortfall by drawing power from the grid. They could also sell any surplus to the grid. In order for such a win-win arrangement to have no negative impact on people’s daily lives, algorithms would have to be developed in order to automate and personalize the process.

Better yet, we could store energy cheaply in times of low demand, in order to avoid consuming more at peak times. We are not talking science fiction here: all the necessary technologies already exist. All we need is the algorithms to automate, personalize and optimize them, and that is exactly what Professor Anjos, his team at IVADO and other researchers* are working on.

The goal is not so much self-sufficiency as efficiency. By optimizing its energy demand, a community could:

  • reduce the cost of consumption;
  • make better use of resources;
  • avoid the environmental impact and cost of building the means of producing electricity, such as dams, powerhouses and so on;
  • export more energy; and
  • envisage mid- and long-term planning to address any lack of the necessary resources to produce energy.

In other words, big data can be employed for the benefit of consumers, and of society as a whole. Starting with energy makes sense for Quebec, and the resulting expertise is by its very nature exportable.

*Research is being done in close cooperation with other on-campus researchers, including Guchuan Zhu, Roland Malhamé, Frédéric Sirois, Brunilde Sansò, Michel Gendreau, Michel Bernier, Juan Gómez, Adham Ismail, Giuseppe Costanzo, François Gilbert, Patrice Marcotte and Gilles Savard.




Post a comment

0 of 1000 characters