Improving and maintaining the efficiencies of power plant is critical. Continuous on-line heat rate monitoring systems cannot make up for the loss in heat rate caused by poor quality fuel, poor maintenance, and even the equipment design itself.
Being able to predict and simulate the leading causes of variation in heat rates which are key factors in measuring the overall efficiency of the power plant.
We acquire, clean, transform, and integrate power plant equipments from sources like inspection data points, sensors data (caloric value, excess O2, ash content, CO content, flow rate, voltage), and maintenance post-event data (damage classes, maintenance types, planned v.s. unplanned events, maintenance actions). Then we use machine learning to do Root Cause Analysis to identify “important features” or variables that affected the heat rate to find the best most important parameters that gives the highest overall efficiency.