Grantee Research Project Results
Final Report: Development of a Real-Time Flare Combustion Efficiency Monitor
EPA Contract Number: EPD15012Title: Development of a Real-Time Flare Combustion Efficiency Monitor
Investigators: Zeng, Yousheng
Small Business: Providence Photonics, LLC
EPA Contact: Richards, April
Phase: II
Project Period: November 1, 2014 through October 31, 2016 (Extended to October 31, 2017)
Project Amount: $299,877
RFA: Small Business Innovation Research (SBIR) - Phase II (2014) Recipients Lists
Research Category: Small Business Innovation Research (SBIR) , SBIR - Air
Description:
There are thousands of flares operating at petroleum refineries, chemical plants, and other industrial facilities in the U.S, and many more in other countries. Flares are used as a safety and emission control device to combust process vent gases that are otherwise difficult to control. Because flares operate in open air, currently there is no good method to measure or monitor the combustion efficiency (CE) or destruction and removal efficiency (DRE) of a flare. In the state of Texas alone, there were 1,130 flares in 2006 that emitted 13,078 tons per year of volatile organic compounds (VOC), a portion of which was air toxics, based on an assumed DRE of 98%1 . Research has shown that the assumed 98% CE or DRE is not reliable and actual VOC emissions could be drastically different from estimates using current estimation method.
For many years, both regulating and regulated communities have been searching for a practical method to directly measure or monitor flare CE. Currently available technologies include 1) extractive sampling followed by conventional flue gas analyzers, and 2) Passive Fourier Transform Infrared (PFTIR). The extractive sampling method is a "point measurement" method. It is used in research only and is not practical for routine monitoring. The PFTIR technology is a "path measurement" method and it has several technical shortcomings. The instrument must be aimed at a specific region of the flare plume and must assume that the optical path length during a data acquisition cycle remains constant. Flare plume dynamics and long data acquisition cycles (> 1 second per cycle) make both of these requirements unreliable. The PFTIR also has a limited dynamic range due to a one sensor configuration and must operate at lower temperatures where atmospheric interferences can be significant. In contrast, the Video Imaging Spectral Radiometer (VISR) technology developed by Providence 2 is based on 3- dimensional (3-D) measurement with information in one dimension being compressed into a 2-D image (see Figure ES-1), and it has at least 20-30 times faster data rate than PFTIR. Therefore, it can overcome the shortcomings associated with PFTIR.
Figure ES-1. Illustration of Three Types of Flare CE Measurement Approaches
The purpose of the SBIR Phase II research was to continue the development of the VISR technology demonstrated successfully in the SBIR Phase I feasibility study. The encouraging Phase I results showed the technical feasibility of this innovative technology for real-time monitoring of flare CE. At the core of this technology is a multi-spectral Infrared (IR) imager that is designed to measure relative concentrations of unburned hydrocarbons and product of combustion (i.e., carbon dioxide, or CO2). Providence built a prototype VISR system and conducted a series of tests in two field campaigns in 2014 and 2016 to establish the performance of this technology. In these experiments, the flare plume was captured by an extractive system and the true CE values were determined by conventional analyzers. The flare was imaged by the VISR prototype and CE was calculated and compared to the true CE values. The results of these experiments are reported below.
Summary/Accomplishments (Outputs/Outcomes):
During the 2014 field campaign, a total of 39 tests were conducted on three of the most common types of industrial flares: steam assisted, air assisted, and pressure assisted flares. 28 tests were designed to test the accuracy of VISR technology. Eleven paired duplicates were tested for precision or repeatability. The results of the field campaign are summarized below.
- Accuracy: Average difference between the extractive method (considered as the reference method) and VISR is 0.5% in CE measurement
- Linearity: CE measured by VISR correlated well with CE measured by extractive method in a CE range from low 60% to 100% (see Figure ES-2). The correlation coefficient (r2) was 0.985.
- Precision: The average difference over the 11 paired tests was 0.20% in CE measurement, indicating an excellent repeatability.
Figure ES-2. Correlation between Reference CE Measured by Analyzers and CE Determined by the New Method without Calibration.
During the 2016 field campaign, a total of 28 tests were conducted on a steam assisted flare and an air assisted flare. As with the previous test, an extractive sampling system was used as the reference method. The average difference between the extractive sampling method and VISR is -0.7% in CE measurement, confirming the validity of the VISR method.
Conclusions:
This Phase II study has further validated the VISR technology and resulted in the development of a commercial product which will have a profound impact on the flare industry. For the first time, a product is available which can provide a direct measurement of flare CE in real time. This product will provide the flare operator with the tools to optimize flare performance and minimize emissions. VISR can also reduce the cost of compliance with new and existing regulations.
SBIR Phase I:
Development of Real-Time Flare Combustion Efficiency Monitor | Final ReportThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.