A steadily increased penetration of renewable power sources into the existing power grid has been shifting operation mode of existing coal-fired power plants from baseload to dynamic cycling operation. These changes create not only challenges for power utility companies but also opportunities in technical and economic of the plant. In fact, boiler leaks are still the primary cause of forced outages in the existing coal fired units. This work focuses on the detection and isolation of the boiler leaks from five different coal-fired power fleets with different types of boiler system (subcritical and supercritical). Five different plants were analyzed having boiler leaks in different locations and discriminant functions were compared and analyzed for similarities and differences.

Combined multivariable statistical analyses, including Principal Component (PCA), cluster, and Fischer Discriminant Analysis (FDA) were used to characterize the leak occurrence. Four operational states of the boiler were provided in the original process datasets including: 1) the normal operation (Set A), 2) the operation before the leak (set B), 3) the operation with the leaks (Set C), and 4) the operation after the leak repair (set B’). The original datasets were then classified into three different groups; one of the groups was assigned as training while the rest were reserved for validation. PCA was used to reduce dimensionality of the original datasets. The cluster and Fischer Discriminant Analysis (FDA) analyses were used to investigate the relationship between process variables with the loads and to discriminate among faults and categorize the processed data into two main groups: non-leak (0) and leak (1). The outcome of the analyses revealed that more than 40,000 observations were finalized while less than 5.0% of total observations were misclassified. The leaks were identified to occur at different sections of the boiler for each plant, including waterwalls, economizer and superheater.


Data Analytics Applied to Coal Fired Boilers for Detecting Leaks

Division / Sponsors: 
Power Division & Nuclear Engineering Division (NED)

Power Conference
Paper Proceedings: https://asmedigitalcollection.asme.org/POWER/POWER2020/volume/83747

International Conference on Nuclear Engineering
Paper Proceedings: https://asmedigitalcollection.asme.org/ICONE/ICONE2020/volume/83761

POWER-ICONE - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.