With “Data-Driven Prediction of Plasma Performance”, General Fusion looks to the global community for hidden patterns in experimental results
VANCOUVER — (December 10, 2015) —Innovative clean tech company General Fusion has decided to come back to the wisdom of the crowd – this time in a search for significant patterns in data from its plasma experiments. Understanding the complex behaviour of plasma is a key to achieving fusion, a potential source of abundant, clean safe and affordable energy.
This is not the first time General Fusion, a leader in developing commercial fusion energy, has teamed up with InnoCentive of Waltham, MA and its roster of 355,000 registered “solvers” from nearly 200 countries, for a solution to a thorny high-tech problem. Earlier this year, General Fusion asked the solvers to design a seal for its fusion system that could withstand withering impacts, pressure, heat and rapid repetition.
Kirby Meacham, an MIT-trained mechanical engineer from Cleveland, OH with over 30 years of experience, came up with the winning “Metallic Pressure-Balanced Anvil Seal” design and claimed the $20,000 winner’s prize. Running for 30 days this spring, the challenge attracted over 60 submissions from 17 different countries.
In this challenge, titled “Data-Driven Prediction of Plasma Performance”, General Fusion is asking solvers to apply statistical techniques and computational tools to identify new patterns in hundreds of gigabytes of data from plasma experiments.
Plasma is the super-heated hydrogen gas that fuels General Fusion’s Magnetic Target Fusion process. General Fusion is eager to see if the solvers can identify patterns in the data that will allow the company to further improve the quality and performance of its plasma.
“We will tap into the global wealth of knowledge in data mining for ideas from a wide range of other industries,” says Brendan Cassidy, the company’s crowdsourcing project leader.
“Participants do not need to be plasma physicists,” says Mr. Cassidy, “but people who have a knack for finding statistical patterns and correlations in large amounts of complex data.”
As with the first crowdsourcing project, there’s prize money to be won – $20,000 for the best prediction methodology. To make the challenge even more interesting, General Fusion will post a weekly list of competitors’ top scores, giving the solvers a target to surpass.
This challenge will again be hosted by InnoCentive, who have conducted similar successful challenges with NASA and Procter & Gamble, among others. All details of this challenge and eligibility requirements can be found on InnoCentive’s website. The challenge opens today and runs until Mar 9, 2016.