The scientists looked at how well a coal-burning boiler was working by measuring leftover carbon in the ash. They used a special computer system called a neural network to predict the carbon levels in the ash. They then used this prediction to create a new system that could make the boiler work even better by adjusting how it burns coal more efficiently. This system was tested at the Fuxin Power Plant with a 200 MW unit and it improved the boiler's performance.