NO{sub X} advisor: Intelligent software for combustion optimization

Technical Report · Mon Jan 01 04:00:00 EST 1996

OSTI ID:374516

Pfahler, J; Eskenazi, D; Levy, E ^[1]

---

1. Lehigh Univ., Bethlehem, PA (United States); and others

Under a Tailored Collaboration Project funded by PEPCO and EPRI, Lehigh University`s Energy Research Center has been developing software for use by plant personnel to optimize a pulverized coal-fired boiler to reduce NO{sub x} emissions and minimize heat rate. Based on expert systems, neural networks and a mathematical optimization algorithm, the NO{sub x} Advisor uses the expert system to safely guide the plant engineer through a series of parametric boiler tests, gathering a database to characterize the operation of the boiler over a wide range of conditions. The neural network develops non-linear mappings between NO{sub x} heat rate and the controllable parameters. These are then used by the mathematical optimization algorithm to identify optimal operating conditions. The Advisor can be used to adjust a boiler to reach minimum NO{sub x} or identify the conditions that give minimum heat rate subject to a target NO{sub x} value. The Advisor operates off-line on a PC platform, gathering data automatically from the plant data highway and/or manually as needed. Results of a field trial at PEPCO`s Potomac River Station are described.

• Research Organization: Electric Power Research Inst., Palo Alto, CA (United States)
• OSTI ID: 374516
• Report Number(s): EPRI-TR--105978-V3; CONF-9505150--Vol.3
• Country of Publication: United States
• Language: English

Related Subjects

01 COAL, LIGNITE, AND PEAT
20 FOSSIL-FUELED POWER PLANTS
AIR POLLUTION CONTROL
ALGORITHMS
BOILERS
BURNERS
COAL
COMBUSTION
COMPUTERIZED CONTROL SYSTEMS
DENITRIFICATION
FLUE GAS
FOSSIL-FUEL POWER PLANTS
HEAT RATE
NEURAL NETWORKS
NITROGEN OXIDES
OPERATION
OPTIMIZATION