Key points are not available for this paper at this time.
Many industrial sectors built cogeneration plants to secure their power supplies reliably and to efficiently produce the plant demand of steam through the associated heat. Due to the rise of fuel cost and tightening environmental regulations, the number of cogeneration plants will increase in lieu to individual boilers and steam turbine generators. Most of the recent cogeneration plants are equipped with hardware-based analyzer which is a continuous emission monitoring system (CEMS) to monitor the NOx emissions from the plant stack as per U.S. Environmental Protection Agency (EPA) regulations. The CEMS is unreliable due to high failure rates and requires high capital cost, high maintenance cost, high operational cost in addition to being subject to long lag time and having slow response. In this work, a software-based analyzer is designed by applying artificial neural networks (ANNs) on process data collected from cogeneration plant (156 MW X 2 combustion gas turbine generators (CGTGs)) equipped with CEMS for NOx monitoring. The developed soft analyzer will be used to verify the existing CEMS readings and used as a reliable tool to monitor the NOx emissions that will eventually replace the CEMS. By providing a relationship between the process and the emissions, the soft analyzer will also assist in understanding the NOx behavior in reference to the process variations and thus enables better emission control.
Al-Malak et al. (Fri,) studied this question.