EXPERIMENTAL DATA REGARDING THE EFFECTS OF UREA ADDITION INTO LIQUID FUEL TO COMBUSTION ENHANCEMENT OF A LOW NOX GAS TURBINE COMBUSTOR
Description
• Combustion performance data: gas flue temperature, NOx, CO2, CO, O2. The gas flue temperature was acquired at 4 Hz: the average value and the standard deviation are reported in the file COMBUSTION_PERFORMANCE.xlsx – sheet named OPERATING_CONDITIONS, in combination with the controlled data inlet gas temperature of 410 K, the fuel mass flow rate of 0.001 kg/s and the total air mass flow rate varied between 0.04 kg/s and 0.08 kg/s depending on the overall equivalence ratio. Instead, pollutant emissions were simultaneously acquired by means of the portable gas analyser model PG-350E Horiba, at sampling frequency of 1 Hz. The average value and the standard deviation of each specie concentration are reported in the file COMBUSTION_PERFORMANCE.xlsx - sheet named EMISSIONS. The files rawData_test#.xlsx, with # denoting the test case number, report the raw signal data. • Luminosity matrix of the VIS images. 100 images are provided sampled at 1 kHz corresponding to 0.1 s of investigated time. For each test case, a 3D matrix of 1024×1240x100 elements are provided in Matlab workspace data format: test1_BB.mat, test2_BB.mat, test3_BB.mat, test4_BB.mat, test5_BB.mat, test6_BB.mat. Into each Matlab data file, three variables have been stored: 1) YIC: 3D single-precision data type tensor, size 1024x1240x100 (calibration factor: 0.041 mm/pixel), representing the y-coordinate (units: pixels), the x-coordinate (units: pixels), and the number of time points, respectively. 2) nImages: integer scalar, number of images. 3) Fs: double-precision scalar, sampling frequency. • Luminosity matrix of the OH* images. 100 images are provided sampled at 1 kHz corresponding to 0.1 s of investigated time. For each test case, a 3D matrix of 966×1574x100 elements are provided in Matlab workspace data format: test1_OH.mat, test2_OH.mat, test3_OH.mat, test4_OH.mat, test5_OH.mat, test6_OH.mat. Into each Matlab data file, three variables have been stored: 1) YIC: 3D single-precision data type tensor, size 966x1574x100 (calibration factor: 0.067 mm/pixel). representing the y-coordinate (units: pixels), the x-coordinate (units: pixels), and the number of time points, respectively. 2) nImages: integer scalar. number of images. 3) Fs: double-precision scalar. sampling frequency. • Luminosity matrix of the CH* images. 100 images are provided sampled at 1 kHz corresponding to 0.1 s of investigated time. For each test case, a 3D matrix of 966×1574x100 elements are provided in Matlab workspace data format: test1_CH.mat, test2_CH.mat, test3_CH.mat, test4_CH.mat, test5_CH.mat, test6_CH.mat. Into each Matlab data file, three variables have been stored: 1) YIC: 3D single-precision data type tensor, size 966x1574x100 (calibration factor: 0.067 mm/pixel), representing the y-coordinate space (units: pixels), the x-coordinate space (units: pixels), and the number of time points, respectively. 2) nImages: integer scalar, number of images. 3) Fs: double-precision scalar, sampling frequency.
Files
Steps to reproduce
The data were acquired during experiments that were carried out at the Green Engine laboratory of the University of Salento in Lecce – Italy, where a 300 kW liquid-fuelled swirling combustor was used in a non-premixed mode. The burner has the inner diameter equal to 14 cm and the length equal to 29 cm. The air passage consists of an inner annular air tube that is equipped with eight-septa, 45° swirler, and an outer coaxial tube. Liquid fuel is supplied at 7 bar and it is injected by a single-hole nozzle by Monarch 1.20 45° R with 45° injection angle. The operating control parameters are monitored by the National Instruments LabVIEW® integrate platform. The VSE 0.02 flow meter is used to measure the mass flow rate of fuel, while the air mass flow is evaluated using the pressure measurements given by the pressure sensor Nuova Fima ST18 and the temperature measurements by T-type thermocouples. Pollutant emissions were acquired by the PG-350E Horiba analyzer system. High-speed flame visualisations were performed in both visible and UV spectral ranges, by means of the Memrecam GX-1F camera, equipped with a Sigma Macro lens 105 mm set-up at f/0.5 aperture, and an intensified CCD camera ( Phantom M320S camera equipped with a Lambert intensifier and UV lens 78 mm with f/3.8 aperture). The Memrecam GX-1F acquired flame images for each experimental test at 1000 Hz acquisition rate, with resolution of 1280 pixels x 1024 pixels. For the UV acquisition narrow optical band-pass filters were used: a 307.1 nm center-wavelength (CWL) filter for the OH* chemiluminescence emission and a 436 nm CWL for the methylidyne radical radicals (CH*) chemiluminescence emission. For each test condition, the ICCD camera performed acquisitions at 1000 Hz, with 1574 pixels x 966 pixels resolution. The ICCD intensifier gain was set to 800 and 900 for CH* and OH* acquisitions, respectively. Three equivalence ratios were investigated, i.e. 0.36 (test cases 1 and 2), 0.24 (test cases 3 and 4) and 0.18 (test cases 5 and 6). The total air flow rate ranged between 40·10−3 kg/s and 80·10−3 kg/s with a ratio of 1:11 between primary and secondary air flow, respectively. Instead, the fuel flow rate was kept constant at 1·10−3 kg/s. The inlet air was pre-heated at 410 K, and the combustion occurred at atmospheric pressure. The water-urea emulsified fuels were prepared by the Institute of Nanotechnology (NANOTEC) in Lecce (Italy). Concerning production process, water and solid urea were added into a proper tank containing the specified amount of Jet-A1. Therefore, the resulting mixture was homogenized by using a UltraTurrax T25 homogenizer operating at 14,000 rpm for at least 5 min. Instead, during experiments the emulsion homogeneity was ensured by continuous agitation by means of the magnetic stirring plate model Buo Qua SH3.