API SMART LEAK-2004 Smart Leak Detection and Repair (LDAR) for Control of Fugitive Emissions《气体挥发控制的灵活探测和修复泄漏》.pdf

1、 Smart Leak Detection and Repair (LDAR) for Control of Fugitive Emissions Regulatory Analysis of which there are hundreds-of-thousands at an industrial plant. A Smart LDAR program that focuses on finding and repairing this minority of high “leakers” could achieve equivalent or better environmental p

2、rotection at a lower cost. Emerging optical imaging technologies provide a tool to more quickly identify high leaking components. Laser-based optical imagers have been identified. Remote sensing and instantaneous detection capabilities of these laser-based optical imaging technologies allow an opera

3、tor to quickly scan large areas containing tens to hundreds of potential leaks. Significant leaks are identified immediately, allowing quicker repair, and ensuring efficient use of resources. Monte Carlo Analyses have been performed to determine control equivalence for the optical imaging technology

4、 compared to current methods (i.e. EPA Reference Method 21). Environmental benefit equivalent to the current work practice is demonstrated when Monte Carlo simulations show that emission reduction for an alternative technology is the same as, or larger than, the current work practice emission reduct

5、ion. In current fugitive emission control programs, quarterly monitoring is usually required for most components with leak definitions of 10,000 ppmv, 1,000 ppmv or 500 ppmv. Pumps are monitored monthly. The Monte Carlo analyses showed that for valves, optical imaging used at bi-monthly monitoring f

6、requency, provides greater environmental protection than the current Method 21 quarterly monitoring. Field and laboratory tests of optical imaging technologies have been conducted to demonstrate that the technologies could detect fugitive emissions at refineries and chemical plants under normal oper

7、ating conditions and to determine detection limits. The project has been a cooperative effort of the petroleum industry, government funded laboratories, the U.S. EPA, and technology vendors. 1ES-1.0 The Basis of Smart LDAR In 1997, the American Petroleum Institute (API) conducted a study1to identify

8、 opportunities for conducting LDAR programs in a more cost-effective manner. The study evaluated data collected over more than 5 years at 7 Los Angeles, California refineries in the South Coast Air Quality Management District (SCAQMD). The data were examined to help determine if there were any desig

9、n or operational characteristics that influence fugitive emissions, and whether a focused LDAR program could be more cost effective at controlling these emissions compared to the current method. The API Study showed that 84 percent of the refinery fugitive emissions were from high leakers (10,000 pp

10、mv), which were only 0.13 percent of the total number of components (See Figure ES-1)2. Of the remaining 16 percent of the estimated emissions, 9.5 percent were from non-leakers (screening =100,0000%10%20%30%40%50%60%70%80%90%PercentagePPMV Range% of Total Count % of Total EmissionsThe study also fo

11、und that there were no chronic leakers and only 5.4 percent of all emissions were from repeat leakers. Instead, the high leakers were found to occur randomly. No systematic explanation for their occurrence was apparent. The Study concluded that a more cost effective LDAR program would be one that em

12、phasizes the location and repair of high leakers. The API has named such a program Smart LDAR. ES-2.0 Optical Imagers for Locating Leaking Components Two technologies have been tested at plants by the API led work group and have successfully found leaking components: 1American Petroleum Institute, “

13、Analysis of Refinery Screening Data,” Publication # 310, Washington, DC, November 1997. 2The overall percentage of high leakers (screening10,000) in any of the seven refineries was less than 0.2 percent. 2 A CO2laser imager. This is a commercially available instrument, manufactured and marketed by L

14、aser Imaging Systems (LIS) under the brand “Gas Vue.” Gas Vue utilizes a CO2laser. The Gas Vue was successfully tested at two chemical plants and is referred to as a CO2laser imager throughout this report. A “fiber” laser imager. This instrument, developed by Sandia National Laboratorys (SNL) Lawren

15、ce Livermore facility, utilizes a backscatter technique patented by LIS. It is referred to as a “fiber” laser in reference to its optical fiber laser amplifier. It was successfully tested at two refineries and a chemical plant. Each laser is tuned to emit a specific wavelength of infrared light that

16、 provides specific compound or compound type detection. The CO2laser is discreetly tunable in the 8-10 micron spectral region. The fiber laser is continuously tunable in the 3 micron spectral region. ES-2.1 Backscatter Absorption Gas Imaging (BAGI) The principle of operation of the CO2laser and fibe

17、r laser is Backscatter Absorption Gas Imaging (BAGI). In BAGI, a live video image is produced by illuminating the view area with laser light in the infrared frequency range. The reflected (backscattered) laser light is detected with a camera sensitive to that light. When the chosen laser wavelength

18、is strongly absorbed by the gas of interest, a cloud of that gas is revealed as a dark image as shown in Figure ES 2-1. A video camera-type scanner both sends out the laser beam and picks up the backscattered infrared light. The camera converts this backscattered infrared light to an electronic sign

19、al, which is displayed in real-time as an image on both the viewfinder and a video monitor. The same image will be seen whether the scanning is done in daylight or at night because the scanner is only sensitive to illumination coming from the infrared light source, not the sun. The imager can be swi

20、tched between visible and infrared views. Figure ES 2-1. Schematic Description of BAGI Process Incident infraredlaser lightBackscattered laser lightIncident infraredlaser lightGas PlumeBackscatteredlaser lightFigures ES 2-2 and ES 2-3 show the visible light and infrared views of leaking components v

21、iewed with the CO2and fiber lasers. Source: As Adapted from McRae, Tom, GasVue: A Rapid Leak LocationTechnology or Large VOC Fugitive Emissions. (Presentation at the CSIPetroleum Refining Sector Equipment Leaks Group, Washington, DC, Sept. 9,1997). See U.S. Patent # 4,555,627. f Note: Although this

22、Figure shows the gas in contact with the background material, it is not a requirement that the gas be in contact with the background. The gas plume need only be between the background and the infrared camera. 3Figure ES 2-2. CO2Laser Views of a Leaking Connector in Visible and Infrared Light ice eth

23、yleneleak tag connector Figure ES 2-3. Fiber Laser Views of Leaking Flange in Visible and Infrared Light Visible light view of leaking flange Infrared view of leaking flangehydrocarbon plumeflange flangeES-3.0 Variability in Method 21 There is significant variability in EPA Reference Method 21. As s

24、hown in Figure ES 3-1, for a fixed mass rate, the screening value can range over several orders of magnitude. This uncertainty in Method 21 leads to bottom false positives and false negatives when compared to regulatory leak limits. False negatives from Method 21 can result in significant emissions

25、because these components would not be repaired and would continue to leak under a Method 21 based program. Since these components would be identified as leakers by the optical imaging instrument, the reduction of these emissions, which were “missed” by Method 21, is a major advantage for using the o

26、ptical imaging technology. Thus, the new Smart LDAR approach using optical imaging allows a much higher mass leak definition than when using Method 21 since these missed leaks (the false negatives) are found and repaired more frequently. 4In the current fugitive emission control program required by

27、U.S.EPA regulation, monthly monitoring is required for pumps and quarterly monitoring is required for other components with leak definitions for repair of 500, 1,000 and 10,000 ppmv. Less frequent monitoring is allowed if the percent of leaking components remains below a specified level for a specif

28、ied number of periods. Lower leak definitions for repair do not necessarily lead to better emissions control since, as the leak level is decreased, few additional leaking components are added to the repair group and these contribute very little to the overall mass emissions. Figure ES 3-1. Variabili

29、ty of Method 21 Results for Equivalent Mass Emission Rates Note: In a box plot, boxes enclose the middle half of the data spread, from the 25thto the 75thpercentiles, with a horizontal line drawn inside the box at the median (50thpercentile) value; whiskers extend below the box to the 5thpercentile

30、value and above the box to the 95thpercentile value; and “dots” indicate values smaller than the 5thpercentile value or larger than the 95thpercentile value. This illustration shows box plots for 1993-94 Petroleum Industry bagging data-set (American Petroleum Institute, 1993a; 1993b; and 1994) depic

31、ting reported screening value ranges (ppmv) for different “levels” of mass emission rates, denoted by mass magnitude bins (e.g., mass emission rate magnitude “1E-8” indicates mass emission rates with units of 10-8kg/hr; i.e., values 110-9kg/hr or larger, and less than 110-8kg/hr, because the integer

32、 portion of the base-10 logarithm for values between these bounds is “-8”) The Smart LDAR approach focuses on identifying and repairing the highest leakers since these are the source of almost all the mass emissions. Equivalence is obtained by more quickly finding and repairing these large leaks, wh

33、ich more than off-sets the emission rates from components with low ppmv readings that leak for longer periods under current guidelines. Figure ES 3-2 illustrates the concept that Smart LDAR (on average) finds large leaks in shorter time, while the current approach (on average) finds smaller leaks ov

34、er longer time. The total emission is equal under both approaches but more cost-effective 5Figure ES 3-2. Equivalence From Quicker Repair of Highest LeakersTimeLeak RateCurrent ApproachSmart LDARNote: Not drawn to scale. using the Smart LDAR approach. The use of optical imaging provides a more cost

35、effective approach to more quickly find the high leakers. This, combined with the potential to find the false negatives (missed high leakers) from the Method 21 approach, results in the benefits from the Smart LDAR technique using the optical imaging technology. ES-4.0 Testing and Demonstrating Appl

36、icability of Optical Imaging Several field and laboratory studies have been conducted to demonstrate the use of optical imaging for fugitive emissions monitoring. These studies, their purpose and overall outcomes are summarized in Table ES 4-1. 6Table ES 41. Laboratory and Field Tests and Demonstrat

37、ions of the Fiber- and CO2Lasers Test/Year Purpose Outcome Chapter for Details Refinery Demonstration of Van-mounted Fiber Laser Technology, 1999 Determine if fiber laser could detect aliphatic emissions at a refinery under normal operating conditions. Successfully detected leakers. Chapter 6 Labora

38、tory Tests of Portable Fiber Laser, 2000 Explore influence of viewing distance, wind speed, leak rate, reflective background, on the performance of the instrument in a controlled environment. Data obtained on mass rate detection threshold for controlled wind speed, distance and reflective background

39、. The lowest detected leak was at 8 g/hr3seen from 10 ft away at wind speed of 1 m/s. Chapter 7 Laboratory Tests of Portable Fiber Laser, 2001 Determine the influence of viewing distance, wind speed, mass leak rate, reflective background, on the detection thresholds in: (1) a controlled wind-tunnel

40、and (2) in out-door test with “blind” elements Test determined the leak detection threshold of the fiber laser. Lowest detected leak was about 0.2g/hr. Chapter 8 Ethylene Chemical Plant Demonstrations of CO2Laser, 2002 Test whether the CO2laser can detect fugitive emissions at ethylene plants under

41、normal operating conditions. CO2laser successfully detected leaks from components with a mass leak rate greater than 1 g/hr. Chapter 10 Refinery Test of Fiber laser, 2003 Test whether the fiber laser can detect fugitive emissions at a refinery under normal operating conditions Fiber laser successful

42、ly detected leaks from components with a mass leak greater than 20 g/hr. Chapter 9 ES 4-2. Laboratory Testing of Fiber Laser The results of the laboratory testing (wind tunnel and out-door roving tests) of the fiber laser for all conditions are shown in Figure ES 4-1. As shown in the plot, the mass

43、rate detection threshold exceeded 10 g/hr in only 10 cases. Seventy-three percent of the mass detection thresholds determined in the lab test were below 5 g/hr. 3Leak rate set at test component. 7Figure ES 4-1. Minimum Detected Mass Flow Ratefor all Leak Conditions under Laboratory Testing Wind Tunn

44、elThresholds and Roving TestResults0.20.350.350.430.480.60.70.70.751.051.171.21.21.251.31.41.41.61.61.71.751.751.751.7522222.253.23.253.353.354.34.54.759.29.2101117213045505082109150020406080100120140160sandpaper,3m,0m/ssandpaper,3m,1m/scurved metal,3m,0m/scurved metal,3m,1m/scurved metal,3m,5m/ssan

45、dpaper,6.1m,0m/sroving paved ground,3m,.16m/sroving sandpaper,3m, .54m/ssandpaper,6.1m,1m/sroving sandpaper,3m, .79m/sno background, 3m,0m/ssandpaper,3m,10m/ssandpaper,9.1m,0m/sno background, 6.1m,0m/scurved metal,3m,10m/sno background, 3m,1m/srovingsky,6.1m,.44m/scurved metal,9.1m, 0m/sno backgroun

46、d, 9.1m,0m/sroving,curvedmetal, 3m, 2.5m/ssandpaper,3m,5m/sno background, 3m,5m/sroaming,sky,4.6m,2m/sroving,sandpaper,3m, 3.2m/scurved metal,6.1m, 0m/sno background, 6.1m,1m/sroving,sky, 3m, .21m/sroving,paved ground,3m,1.55m/sno background, 9.1m,1m/sno background, 9.1m,5m/sroving,curvedmetal, 3m,

47、1.56m/scurved metal,6.1m, 1m/sroving,curvedmetal, 3m, .53m/ssandpaper,9.1m,1m/sno background, 6.1m,5m/sroving,paved ground,3m,2.4m/ssandpaper,6.1m,5m/sroving,sandpaper,6.1m,2.8m/scurved metal,30ft, 1m/sroving,sky, 3m, 2.2m/ssandpaper,6.1m,10m/sroving,paved ground,7.6m,2.1m/ssandpaper,9.1m,5m/ssandpa

48、per,9.1,10m/scurved metal,6.1m, 5m/sroving,curvedmetal, 9.1m, 4m/scurved metal,9.1m, 5m/scurved metal,9.1m, 10m/scurved metal,6.1m, 10m/sgas flow(g/hr)8Results of refinery testing with the fiber laser showed that the imager detected all leaks above 20 g/hr (see Figure ES 4-2). The mass leak rates we

49、re determined with bagging analysis. Below 20 g/hr, the fiber laser detected roughly 50 percent of the leaks. Some leaks at lower mass rates were detected when a background was added behind the gas cloud. Figure ES 4-2. Fiber Laser Performance at Refinery during 2003 Testing40.1 1.0 10.0 100.0 1000.0Total Hydrocarbons Mass Leak Rate, (g/hr)seen w ith added background seen not seenStyrofoam USGP White Sign SGP ISOM SGP USGP Styrofoam SGP Figure ES 4-3. Scanning Refinery Components with Fiber Laser fiber laser camera backpack with batteries 94SGP saturated ga