Portrait of Dr Gang Lu

Dr Gang Lu

Reader in Electronic Instrumentation
Programme Chair of MSc Advanced Electronic Systems Engineering (taught)

About

Dr Gang Lu is a Reader in Electronic Instrumentation with the School of Engineering and Digital Arts, University of Kent. He received his BEng degree in Mechanical Engineering from Central South University, China in 1982, and worked as an Engineer, Senior Engineer and Department Director in the Institute of Metallurgical Equipment Design and Research, Beijing, and Teesside Technology Centre, British Steel for fourteen years. Dr Lu returned to academia in 1996, and obtained his PhD degree in Electronic Engineering from the University of Greenwich, UK in 2000. He was a Post-Doctoral Research Fellow/Associate with the University of Greenwich and the University of Kent before becoming a Lecturer in Electronic Instrumentation in the University of Kent in 2006 and Senior Lecturer in 2012.
Dr Lu’s main areas of expertise are in sensors, instrumentation, measurement, digital image/signal processing, 2-D and 3-D visualisation and characterisation for combustion systems, condition monitoring, machine learning for engineering solutions. He has been a principal investigator or co-investigator for a range of research projects on advanced monitoring and characterisation of flames in both lab- and industrial-scale fossil/biomass fuel combustion systems, supported by external and internal funding bodies including EPSRC (Engineering and Physical Sciences Research Council), ERC (European Research Council), UKCCSRC (UK Carbon Capture & Storage Research Centre), BF2RA (Biomass and Fossil Fuel Research Alliance), Innovate UK, KIE (Kent Innovation & Enterprise), and UK industry (e.g., EDF, RWE Generation, EON UK, Drax Power Station, Doosan Babcock, British Sugar). Dr Lu has published more than 100 papers in peer reviewed journals and conference proceedings. His h-index is 25 with over 2100 citations. He was awarded the Engineering Innovation Prize (Energy) by the IET (Institution of Engineering and Technology) in 2006 in recognition of his contributions to flame visualisation and characterisation.
Dr Lu is a Chartered Engineer, Senior Member of IEEE, Member of the Energy Institute, and Fellow of Higher Education Academy.

Research interests

Current and past research projects

  • A condition-based monitoring and advisory tool for utility boilers, BF2RA, 2018-21
  • In-situ monitoring and characterisation of agglomeration and defluidisation in a biomass FB combustor through digital imaging and acoustic sensing, UKCCSRC, 2019-20
  • Advanced flame monitoring and emission predication through digital imaging and spectrometry, BF2RA, 2015-19
  • INCASE- Towards industry 4.0 via networked control applications and sustainable engineering, ERC (Interreg V), 2016-19
  • Experimental investigation into oxy-combustion behaviour of single biomass/coal particles through digital imaging techniques, UKCCSRC, 2017-18
  • Imaging of coal fired flames on the Doosan Babcock’s 40MW CTF (Enterprise project), 2017
  • Advanced burner flame monitoring through digital imaging, Innovate UK, 2016-17.
  • Investigation into the characterisation of oxy-coal/biomass flames through digital imaging, UKCCSRC, 2015
  • Intelligent flame detection incorporating burner condition monitoring and on-line fuel tracking, BCURA (British Coal Utilisation Research Association), 2010-13
  • EPSRC‐E.ON strategic partnership, carbon capture and storage, EPSRC, 2009-14
  • In-depth studies of oxy-coal combustion processes through numerical modelling and 3D flame imaging, EPSRC, 2009-14
  • Quantitative characterisation of flame radical emissions for combustion optimisation through spectroscopic imaging, EPSRC, 2009-11
  • Optimization of biomass/coal co-firing processes through integrated measurement and computational modelling, EPSRC, 2008-11
  • A stereoscopic fire detector, KIE, 2011
  • GOSE- Optimisation of combustion plant through advanced measurement and computer modelling, ERC (Interreg III), 2006-10

Teaching

  • EL896 Computer and Microcontroller Architectures (Msc)
  • EL890 MSc Project
  • EL844 Image Analysis with Security Applications (MSc)
  • EL875 Advanced Sensors and Instrumentation Systems (MSc)
  • EL600 Project (Stage 3, Module convenor)
  • EL565 Electronic Instrumentation and Measurement Systems (Stage 2)
  • EL515 Physiological Measurement (Stage 2, Module convenor)
  • EL562 Computer Interfacing Group Project (Stage 2)
  • EL318 Engineering Mathematics (Stage 1)
  • EL027 Semiconductor and Digital Electronics (Stage 0, Module convenor)

Supervision

Dr Lu has supervised more ten PhD students to successful completion. 

Selected current and past PhD projects

  • A condition-based monitoring and advisory tool for utility boilers
  • Advanced flame monitoring and emission predication through digital imaging and spectrometry
  • Flame stability and burner condition monitoring through optical sensing and digital imaging
  • Tomographic characterisation of burner flames through digital imaging and image processing
  • Digital imaging based characterisation of biomass particles in power generation
  • 3-D visualisation and quantitative characterisation of flames using tomography and digital imaging techniques
  • 3-D visualisation and quantitative characterisation of burner flames
  • Profiling of single coal particle combustion and monitoring of combustion process through digital imaging and soft-computing
  • NOx prediction in a biomass-fired CTF through flame radical imaging and machine learning
  • Contactless temperature measurement of stored biomass

Potential PhD projects

  • Combustion stability monitoring through flame imaging and machine learning
  • Emission profiling through flame radical imaging and soft-computing
  • Monitoring and characterisation of biomass-fired flames based on emission imaging and spectrogram analysis
  • 3-D reconstruction of flame radical emissions of through stereoscopic tomography
  • Flame chemiluminescence reconstruction through high-speed spectroscopic and acoustic sensing
  • 3-D monitoring and characterisation of flames in gas turbine systems using stereoscopic imaging techniques
  • Non-invasive measurement and characterisation of avian eggs through digital imaging and biological analysis

Professional

  • Chartered engineer
  • Senior member of IEEE
  • Member of the Energy Institute
  • Fellow of Higher Education Academy
  • Grant reviewer for EPSRC and STFC
  • Regular reviewer of more than 30 international Journals from IEEE, Elsevier, IET, IOP, ACS, OSA, and Taylor & Francis, etc
  • External examiner for PhD viva voce examinations at Edinburgh, Sheffield, Manchester, Cardiff, South Wales, Strathclyde Universities, etc
  • External examiner for BEng (Hons) and MEng (Hons) Instrumentation and Control Engineering, University of Teesside

Publications

Showing 50 of 128 total publications in the Kent Academic Repository. View all publications.

Article

  • Narushin, V., Romanov, M., Lu, G., Cugley, J. and Griffin, D. (2020). Digital imaging assisted geometry of chicken eggs using Hügelschäffer’s model. Biosystems Engineering [Online] 197:45-55. Available at: http://dx.doi.org/10.1016/j.biosystemseng.2020.06.008.
    Geometrical description of the egg shape is of a great importance in a variety of studies and can be instrumental in predicting quality traits of table and hatching poultry eggs. Importantly, developments of non-destructive oomorphological models can drive novel insights in engineering and physical science and lead to new egg-related technologies and egg sorting systems for poultry industry. We attempted to test the Hügelschäffer’s egg model according to which an egg profile curve can be transformed from an ellipse using a specific parameter w. For this purpose, two-dimensional digital imaging and follow-up image processing techniques of chicken eggs were employed. The formulae for recalculation of the egg volume and surface area were consequently deduced from the Hügelschäffer’s equation. Eventually, we refined the Hügelschäffer’s egg model and proved its applicability for defining the contours of hen’s eggs. For practical use in poultry industry and food engineering, the proposed non-destructive methodology can be contributory in defining accurately the contour of any avian egg and determining such characteristics of the egg shape as volume, surface area, etc., with an expected potential in designing automated systems in poultry industry and in egg-related applications in biology, physical science, engineering and other areas.
  • Narushin, V., Lu, G., Cugley, J., Romanov, M. and Griffin, D. (2020). A 2-D imaging-assisted geometrical transformation method for non-destructive evaluation of the volume and surface area of avian eggs. Food Control [Online]. Available at: https://dx.doi.org/10.1016/j.foodcont.2020.107112.
    Egg volume and surface area are reliable predictors of quality traits for both table and hatching chicken eggs. A new non-destructive technique for the fast and accurate evaluation of these two egg variables is addressed in the present study. The proposed method is based on the geometrical transformation of actual egg contour into a well-known geometrical figure which shape most of all resembles the examined egg. The volume and surface area of an examined egg were recomputed using the formulae appropriate for three figures including sphere, ellipsoid, and egg-shape ovoid. The method of the geometrical transformation includes the measurements of the egg length and the area of the examined eggs. These variables were determined using two-dimensional (2-D) digital imaging and image processing techniques. The geometrical transformation approach is proven to be reliable to turn the studied chicken eggs into the three chosen ovoid models, with the best prediction being shown for the ellipsoid and egg-shape ovoid, whilst the former was slightly more preferable. Depending on the avian species studied, we hypothesise that it would be more suitable to use the sphere model for more round shaped eggs and the egg-shaped ovoid model if the examined eggs are more conical. The choice of the proposed transformation technique would be applicable not only for the needs of poultry industry but also in ornithological, basically zoological studies when handling the varieties of eggs of different shapes. The experimental results show that the method proposed is accurate, reliable, robust and fast when coupled and assisted with the digital imaging and image processing techniques, and can serve as a basis for developing an appropriate instrumental technology and bringing it into the practice of poultry enterprises and hatcheries.
  • Ge, H., Li, X., Li, Y., Lu, G. and Yan, Y. (2019). Biomass Fuel Identification Using Flame Spectroscopy and Tree Model Algorithms. Combustion Science and Technology [Online]. Available at: https://doi.org/10.1080/00102202.2019.1680654.
    This paper presents an identification method for types of fuel such as biomass by combining flame spectroscopic monitoring and tree model algorithms. The features of the flame spectra are extracted, including the spectral intensity of flame radicals [OH* (310.85 nm),CN* (390.00 nm), CH* (430.57 nm) and C2* (515.23 nm, 545.59 nm)], flame radiation intensity and flame radiation energy (integration of spectral intensity). The identification models are built using four tree model algorithms, i.e., decision tree, random forest, extremely randomized trees and gradient boost decision tree. The different type biomass and spectra features of combustion flames are composed of sample pairs to train identification models. Experiments are carried out on a laboratory-scale biomass-air combustion test rig. Four different biomass fuels, including corncob, willow, peanut shell and wheat straw are burnt. The results demonstrate that the identification models proposed is capable of identifying types of biomass fuels correctly with the average identification success rate of 98% in ten trials.
  • Hu, Y., Guo, M., Yan, Y., Lu, G. and Cheng, X. (2019). Temperature Measurement of Stored Biomass of Different Types and Bulk Densities Using Acoustic Techniques. FUEL [Online]. Available at: https://www.sciencedirect.com/science/article/pii/S0016236119313389.
    The internal temperature of stored biomass needs to be measured to suppress the possible self-ignition at biomass-fired power stations. Acoustic sensing has been proven to be a promising approach to measuring the temperature of stored wood pellets online and non-intrusively. In such a temperature measurement system, a characteristic factor is defined to derive the sound speed from measured time of flight of sound waves. The characteristic factor is updated based on its experimental relationship with the biomass temperature during temperature measurement. When the type, particle size, particle density and bulk density of stored biomass change, whether the relationship between the characteristic factor and biomass temperature needs to be recalibrated needs investigation. Therefore, the relationship between the characteristic factor and biomass property is modelled using the empirical equation of Miki. Then the model is used to analyse the impact of the particle size, particle density and bulk density of stored biomass on the relationship. An acoustic sensing system is constructed to investigate the influence of bulk density for different types of biomass. The system is also applied to measure the temperature of four biomass fuels, including wood blocks, wood pellets, wood chips, and wheat straws. Results show that the relative error of temperature measurements for the four types of biomass is within 3.5%, 5.7%, 6.8% and 2.5%, respectively, over the temperature range from 22.1℃ to 74.2℃. The relationship between the characteristic factor and biomass temperature should be re-established for different types of biomass and different particle size distributions.
  • Guo, M., Yan, Y., Hu, Y., Lu, G. and Zhang, J. (2018). Temperature Measurement of Stored Biomass Using Low-frequency Acoustic Waves and Correlation Signal Processing Techniques. Fuel [Online] 227:89-98. Available at: https://dx.doi.org/10.1016/j.fuel.2018.04.062.
    As a substitute of traditional fossil fuels, biomass is widely used to generate electricity and heat. The temperature of stored biomass needs to be monitored continuously to prevent the biomass from self-ignition. This paper proposes a non-intrusive method for the temperature measurement of stored biomass based on acoustic sensing techniques. A characteristic factor is introduced to obtain the sound speed in free space from the measured time of flight of acoustic waves in stored biomass. After analysing the relationship between the defined characteristic factor and air temperature, an updating procedure on the characteristic factor is proposed to reduce the influence of air temperature. By measuring the sound speed in free space air temperature is determined which is the same as biomass temperature. The proposed methodology is examined using a single path acoustic system which consists of a source and two sensors. A linear chirp signal with a duration of 0.1 s and frequencies of 200-500 Hz is generated and transmitted through stored biomass pellets. The time of flight of sound waves between the two acoustic sensors is measured through correlation signal processing. The relative error of measurement results using the proposed method is no more than 4.5% over the temperature range from 22? to 48.9?. Factors that affect the temperature measurement are investigated and quantified. The experimental results indicate that the proposed technique is effective for the temperature measurement of stored biomass with a maximum error of 1.5? under all test conditions.
  • Cugley, J., Lu, G., Hossain, M., Yan, Y. and Searle, I. (2018). Visualisation and measurement of flames in a gas-fired multi-burner boiler. Journal of Physics: Conference Series [Online] 1065. Available at: https://dx.doi.org/10.1088%2F1742-6596%2F1065%2F20%2F202009.
    The paper presents the development of an instrumentation system for the visualisation and measurement of flames in a gas-fired multi-burner boiler based on digital imaging and spectrometric techniques. The system consists of a rigid optical probe and an optical fibre, a digital camera, a spectrometer and an embedded computer with associated application software. The characteristic parameters of the flame, including size, temperature and oscillation frequency are quantitatively determined based on flame images obtained. The spectral characteristics of the flame are analysed over the spectral range from the ultraviolet to near infrared. The system was evaluated on a gas-fired heat recovery boiler under different operation conditions. Results obtained suggest the promising correlation between computed flame parameters and operation conditions.
  • Shan, L., Kong, M., Bennett, T., Archi, S., Carol, E., Sun, D., Lu, G., Yan, Y. and Liu, H. (2017). Studies on Combustion Behaviours of Single Biomass Particles Using a Visualization Method. BIOMASS & BIOENERGY [Online] 109. Available at: https://doi.org/10.1016/j.biombioe.2017.12.008.
    Combustion behaviours of single particles (125-150m) of eucalyptus, pine and olive residue were investigated by means of a transparent drop-tube furnace, electrically heated to 1073 K, and a high-speed camera coupling with a long distance microscope. All three types of biomass samples were found to have two evident combustion phases, i.e., volatile combustion in an envelope flame and subsequent char combustion with high luminance. Yet, due to differences in chemical compositions and properties, their combustion behaviours - were also seen somewhat discrepant. The volatile flame of the olive residue was fainter than that of pine and eucalyptus due to its high ash content. During the char combustion phase, fragmentation took place for most pine particles but only for a few particles of olive residue and eucalyptus. For all three types of biomass samples, the flame size and the average luminous intensity profiles were deduced from the captured combustion video images whilst the combustion burnout times of the volatile matter and char were also calculated and estimated. There were two peak values clearly shown on the profiles of both the flame size and the average luminous intensity during the volatile combustion process of pine and eucalyptus particles, which, according to literature, could not be observed by optical pyrometry. The observed peaks correspond to the devolatilisation of hemicellulose and cellulose. The ratio between the estimated char burnout time and volatile combustion time increases quadratically with the fixed carbon to volatile matter mass ratio, confirming char combustion is much slower than volatile combustion.
  • Zhou, H., Li, Y., Tang, Q., Lu, G. and Yan, Y. (2017). Combining flame monitoring techniques and support vector machine for the online identification of coal blends. Journal of Zhejiang University-SCIENCE A (Applied Physics & Engineering) [Online] 18:677-689. Available at: http://dx.doi.org/10.1631/jzus.A1600454.
    The combustion behavior of two single coals and three coal blends in a 300 kW coal-fired furnace under variable operating conditions was monitored by a flame monitoring system based on image processing and spectral analysis. A similarity coefficient was defined to analyze the similarity of combustion behavior between two different coal types. A total of 20 flame features, extracted by the flame monitoring system, were ranked by weights of their importance estimated using ReliefF, a feature selection algorithm. The mean of the infrared signal was found to have by far the highest importance weight among the flame features. Support vector machine (SVM) was used to identify the coal types. The number of flame features used to build the SVM model was reduced from 20 to 12 by combining the methods of ReliefF and SVM, and computational precision was guaranteed simultaneously. A threshold was found for the relationship between the error rate and similarity coefficient, which were positively correlated. The success rate decreased with increasing similarity coefficient. The results obtained demonstrate that the system can achieve the online identification of coal blends in industry.
  • Bai, X., Lu, G. and Yan, Y. (2017). Fractal Characteristics of Thin Thermal Mixing Layers in Coal-Fired Flame. Journal of Combustion Science and Technology [Online] 3:225-230. Available at: http://journals.tju.edu.cn/rs/oa/DArticle.aspx?type=view&id=R201606021.
    The images of turbulent flame were acquired by using a digital imaging system on an industry-scale pulverized coal-fired test rig?The fractal dimensions of thin thermal mixing layers in flame were computed through morphology-based flame image processing techniques?The effects of the ratios of primary air and secondary/tertiary air on fractal dimensions were characterized?The results presented in this work show that the variations of fractal dimension are closely related to the ratio changes of primary air and secondary/tertiary air. Therefore?the fractal dimensions of flame thin thermal mixing layers can be used as an important indicator for the control and optimization of a combustion process.
  • Bai, X., Lu, G. and Yan, Y. (2017). Flame image segmentation using multiscale color and wavelet-based texture features. Computer Engineering and Applications (Chinese) [Online] 53:213-219. Available at: http://dx.doi.org/10.3778/j.issn.1002-8331.1610-0083.
    Accurate and reliable segmentation of flame images are crucial in vision based monitoring and characterization of flames. It is, however, difficult to maintain the segmentation accuracy while achieving fast processing time due to the impact of the background noise in the images and the variation of operation conditions. To improve the quality of the image segmentation, a flame image segmentation method is proposed based on Multiscale Color and Wavelet-based Textures?MCWT? of the images. By combining the color and texture features, a characteristic matrix is built and then compressed using a local mean method. The outer contour of the flame image under the compressed scale is detected by a cluster technique. Subsequently, the flame edge region under the original scale is determined, following that, the characteristic matrix of the edge region is constructed and classified, and finally, the flame image segmentation is achieved. Flame images captured from an industrial-scale coal-firedtest rig under different operation conditions are segmented to evaluate the proposed method. The test results demonstrate that the performance of segmenting flame images of the proposed method is superior to other traditional methods. It also has been found that the proposed method performs more effectively in segmenting the flame images with Gaussian and pepper and salt noise.
  • Daood, S., Ottolini, M., Taylor, S., Ogunyinka, O., Hossain, M., Lu, G., Yan, Y. and Nimmo, W. (2017). Additive technology for pollutant control and efficient coal combustion. Energy and Fuels [Online] 31:5581-5596. Available at: http://dx.doi.org/10.1021/acs.energyfuels.7b00017.
    High efficiency and low emissions from pf coal power stations has been the drive behind the development of present and future efficient coal combustion technologies. Upgrading coal, capturing CO2, reducing emission of NOx, SO2 and particulate matter, mitigating slagging, fouling and corrosion are the key initiatives behind these efficient coal technologies. This study focuses on a newly developed fuel additive (Silanite™) based efficient coal combustion technology, which addresses most of the aforementioned key points. Silanite™ a finely milled multi-oxide additive when mixed with the coal without the need to change the boiler installation has proven to increase the boiler efficiency, flame temperature with reduction in corrosion, NOx and particulate matter (dust) emissions. The process has been developed through bench, pilot (100kW) and full scale (233 MWth). The process has been found to have a number of beneficial effects that add up to a viable retrofit to existing power plant as demonstrated on the 233MWth boiler tests (under BS EN 12952-15:2003 standard)
  • Bai, X., Lu, G., Hossain, M., Szuhánszki, J., Daood, S., Nimmo, W., Yan, Y. and Pourkashanian, M. (2017). Multi-mode Combustion Process Monitoring on a Pulverised Fuel Combustion Test Facility based on Flame Imaging and Random Weight Network Techniques. Fuel [Online] 202:656-664. Available at: https://dx.doi.org/10.1016/j.fuel.2017.03.091.
    Combustion systems need to be operated under a range of different conditions to meet fluctuating energy demands. Reliable monitoring of the combustion process is crucial for combustion control and optimisation under such variable conditions. In this paper, a monitoring method for variable combustion conditions is proposed by combining digital imaging, PCA-RWN (Principal Component Analysis and Random Weight Network) techniques. Based on flame images acquired using a digital imaging system, the mean intensity values of RGB (Red, Green, and Blue) image components and texture descriptors computed based on the grey-level co-occurrence matrix are used as the colour and texture features of flame images. These features are treated as the input variables of the proposed PCA-RWN model for multi-mode process monitoring. In the proposed model, the PCA is used to extract the principal component features of input vectors. By establishing the RWN model for an appropriate principal component subspace, the computing load of recognising combustion operation conditions is significantly reduced. In addition, Hotelling’s T2 and SPE (Squared Prediction Error) statistics of the corresponding operation conditions are calculated to identify the abnormalities of the combustion. The proposed approach is evaluated using flame image datasets obtained on a 250 kWth air- and oxy-fuel Combustion Test Facility. Variable operation conditions were achieved by changing the primary air and SA/TA (Secondary Air to Territory Air) splits. The results demonstrate that, for the operation conditions examined, the condition recognition success rate of the proposed PCA-RWN model is over 91%, which outperforms other machine learning classifiers with a reduced training time. The results also show that the abnormal conditions exhibit different oscillation frequencies from the normal conditions, and the T2 and SPE statistics are capable of detecting such abnormalities.
  • Bai, X., Lu, G., Bennet, T., Sarroza, A., Eastwick, C., Liu, H. and Yan, Y. (2017). Combustion behavior profiling of single pulverized coal particles in a drop tube furnace through high-speed imaging and image analysis. Experimental Thermal and Fluid Science [Online] 85:322-330. Available at: https://dx.doi.org/10.1016/j.expthermflusci.2017.03.018.
    Experimental investigations into the combustion behaviors of single pulverized coal particles are carried out based on high-speed imaging and image processing techniques. A high-speed video camera is employed to acquire the images of coal particles during their residence time in a visual drop tube furnace. Computer algorithms are developed to determine the characteristic parameters of the particles from the images extracted from the videos obtained. The parameters are used to quantify the combustion behaviors of the burning particle in terms of its size, shape, surface roughness, rotation frequency and luminosity. Two sets of samples of the same coal with different particle sizes are studied using the techniques developed. Experimental results show that the coal with different particle sizes exhibits distinctly different combustion behaviors. In particular, for the large coal particle (150-212 m), the combustion of volatiles and char takes place sequentially with clear fragmentation at the early stage of the char combustion. For the small coal particle (106-150 m), however, the combustion of volatiles and char occurs simultaneously with no clear fragmentation. The size of the two burning particles shows a decreasing trend with periodic variation attributed to the rapid rotations of the particles. The small particle rotates at a frequency of around 30 Hz, in comparison to 20 Hz for the large particle due to a greater combustion rate. The luminous intensity of the large particle shows two peaks, which is attributed to the sequential combustion of volatiles and char. The luminous intensity of the small particle illustrates a monotonously decreasing trend, suggesting again a simultaneous devolatilization/volatile and char combustion.
  • Bai, X., Hossain, M., Lu, G., Yan, Y. and Liu, S. (2016). Multimode Monitoring of Oxy-gas Combustion through Flame Imaging, Principal Component Analysis and Kernel Support Vector Machine. Combustion Science and Technology [Online] 189:776-792. Available at: http://dx.doi.org/10.1080/00102202.2016.1250749.
    This paper presents a method for the multimode monitoring of combustion stability under different oxy-gas fired conditions based on flame imaging, principal component analysis and kernel support vector machine (PCA-KSVM) techniques. The images of oxy-gas flames are segmented into premixed and diffused regions through Watershed Transform method. The weighted color and texture features of the diffused and premixed regions are extracted and projected into two subspaces using the PCA to reduce the data dimensions and noises. The multi-class KSVM model is finally built based on the flame features in the principal component subspace to identify the operation condition. Two classic multivariate statistic indices, i.e. Hotelling’s T2 and squared prediction error (SPE), are used to assess the normal and abnormal states for the corresponding operation condition. The experimental results obtained on a lab-scale oxy-gas rig show that the weighted color and texture features of the defined diffused and premixed regions are effective for detecting the combustion state and that the proposed PCA-KSVM model is feasible and effective to monitor a combustion process under variable operation conditions.
  • Li, N., Lu, G., Li, X. and Yan, Y. (2015). Prediction of NOx Emissions from a Biomass Fired Combustion Process Based on Flame Radical Imaging and Deep Learning Techniques. Combustion Science and Technology [Online] 188:233-246. Available at: http://dx.doi.org/10.1080/00102202.2015.1102905.
    This article presents a methodology for predicting NOx emissions from a biomass combustion process through flame radical imaging and deep learning (DL). The dataset was established experimentally from flame radical images captured on a biomass-gas fired test rig. Morphological component analysis is undertaken to improve the quality of the dataset, and the region-of-interest extraction is introduced to extract the flame radical part and rescale the image size. The developed DL-based prediction model contains three successive stages for implementing the feature extraction, feature fusion, and emission prediction. The fine-tuning based on the prediction is introduced to adjust the process of the feature fusion. The effects of the feature fusion and fine-tuning are discussed in detail. A comparison between various image- and machine-learning-based prediction models show that the proposed DL prediction model outperforms other models in terms of root mean square error criteria. The predicted NOx emissions are in good agreement with the measurement results.
  • Sun, D., Lu, G., Zhou, H., Yan, Y. and Liu, S. (2015). Quantitative Assessment of Flame Stability Through Image Processing and Spectral Analysis. IEEE Transactions on Instrumentation and Measurement [Online] 64:3323-3333. Available at: http://doi.org/10.1109/TIM.2015.2444262.
    This paper experimentally investigates two generalized methods, i.e., a simple universal index and oscillation frequency, for the quantitative assessment of flame stability at fossil-fuel-fired furnaces. The index is proposed to assess the stability of flame in terms of its color, geometry, and luminance. It is designed by combining up to seven characteristic parameters extracted from flame images. The oscillation frequency is derived from the spectral analysis of flame radiation signals. The measurements involved in these two methods do not require prior knowledge about fuel property, burner type, and other operation conditions. They can therefore be easily applied to flame stability assessment without costly and complex adaption. Experiments were carried out on a 9-MW heavy-oil-fired combustion test rig over a wide range of combustion conditions including variations in swirl vane position of the tertiary air, swirl vane position of the secondary air, and the ratio of the primary air to the total air. The impact of these burner parameters on the stability of heavy oil flames is investigated by using the index and oscillation frequency proposed. The experimental results obtained demonstrate the effectiveness of the methods and the importance of maintaining a stable flame for reduced NOx emissions. It is envisaged that such methods can be easily transferred to existing flame closed-circuit television systems and flame failure detectors in power stations for flame stability monitoring.
  • Li, N., Lu, G., Li, X. and Yan, Y. (2015). Prediction of Pollutant Emissions of Biomass Flames Through Digital Imaging, Contourlet Transform, and Support Vector Regression Modeling. IEEE Transactions on Instrumentation and Measurement [Online] 64:2409-2416. Available at: http://doi.org/10.1109/TIM.2015.2411999.
    This paper presents a method for the prediction of NOx emissions in a biomass combustion process through the combination of flame radical imaging, contourlet transform and Zernike moment (CTZM), and least squares support vector regression (LS-SVR) modeling. A novel feature extraction technique based on the CTZM algorithm is developed. The contourlet transform provides the multiscale decomposition for flame radical images and the selected operator based on Zernike moments is designed to provide the well-defined structure for the images. The resulted image features are a variable structure, which is originated from the CTZM. Finally, the variable features of the images of four flame radicals (OH*, CN*, CH*, and C*2) are defined. The relationship between the variable features of radical images and NOx emissions is established through radial basis function network modeling, SVR modeling, and the LS-SVR modeling. A comparison between the three modeling approaches shows that the LS-SVR model outperforms the other two methods in terms of root-mean-square error and mean relative error criteria. In addition, the structure of the image features has a significant impact on the performance of the prediction models. The test results obtained on a biomass-gas fired test rig show the effectiveness of the proposed technical approach for the prediction of NOx emissions.
  • Li, X., Wu, M., Lu, G. and Yan, Y. (2015). NOx emission prediction based on flame radical profiling and support vector machine. Proceedings of the CSEE [Online] 35:1413-1419. Available at: http://www.dx.doi.org/10.13334/j.0258-8013.pcsee.2015.06.016.
    Characteristics of reacting radicals in a flame are crucial for an in-depth understanding of the formation process of combustion emissions. An algorithm for the prediction of NOx( NO and NO2) Emissions in flue gas was presented through flame radical imaging, flame temperature monitoring and application of soft computing techniques, support vector machine. Radiation images of flame radicals OH *, CN *, CH *and C2* Were captured using an intensified multi-wavelength imaging system. Flame temperature was determined using a spectrometer and two-color pyrometry. Based on these images, the characteristic values ??of the flame radicals were extracted. These characteristic values ??(contours and ratios of radical intensities), together with the flame temperature, were then used to predict NOx emissions. Experimental results from a laboratory-scale gas-fired combustion rig show the effectiveness of the proposed method for the prediction of NOx emissions.
  • Li, X., Yan, Y., Liu, S., Wu, M. and Lu, G. (2015). On-line identification of biomass fuels based on flame radical imaging and application of radical basis function neural network techniques. IET Renewable Power Generation [Online] 9:323-330. Available at: http://doi.org/10.1049/iet-rpg.2013.0392.
    In biomass fired power plants a range of biomass fuels are used to generate electric power. It is desirable to identify the type of biomass fuels on-line continuously in order to achieve an improved combustion efficiency, and reduced pollutant emissions. This paper presents the recent investigations into the on-line identification of biomass fuels based on the combination of flame radical imaging and radical basis function (RBF) neural network (NN) techniques. The characteristic values of flame radicals (OH*, CN*, CH* and C2*), including the intensity ratio, intensity contour, mean intensity, area and eccentricity, are computed to reconstruct two types of RBF NN, that is, accurate and probabilistic RBF networks. Experimental results obtained for three types of biomass fuels (flour, willow sawdust and palm kernel shell) firing on a laboratory-scale combustion test rig are presented to demonstrate the effectiveness of the proposed method.
  • Chalmers, H., Al-Jeboori, M., Anthony, B., Balusamy, S., Black, S., Marincola, F., Clements, A., Darabkhani, H., Dennis, J., Farrow, T., Fennell, P., Franchetti, B., Gao, L., Gibbins, J., Hochgreb, S., Hossain, M., Jurado, N., Kempf, A., Liu, H., Lu, G., Ma, L., Navarro-Martinez, S., Nimmo, W., Oakey, J., Pranzitelli, A., Scott, S., Snape, C., Sun, C., Sun, D., Szuhánszki, J., Trabadela, I., Wigley, F., Yan, Y. and Pourkashanian, M. (2014). OxyCAP UK: Oxyfuel Combustion - academic Programme for the UK. Energy Procedia [Online] 63:504-510. Available at: http://dx.doi.org/10.1016/j.egypro.2014.11.055.
    The OxyCAP-UK (Oxyfuel Combustion - Academic Programme for the UK) programme was a £2 M collaboration involving researchers from seven UK universities, supported by E.On and the Engineering and Physical Sciences Research Council. The programme, which ran from November 2009 to July 2014, has successfully completed a broad range of activities related to development of oxyfuel power plants. This paper provides an overview of key findings arising from the programme. It covers development of UK research pilot test facilities for oxyfuel applications; 2-D and 3-D flame imaging systems for monitoring, analysis and diagnostics; fuel characterisation of biomass and coal for oxyfuel combustion applications; ash transformation/deposition in oxyfuel combustion systems; materials and corrosion in oxyfuel combustion systems; and development of advanced simulation based on CFD modelling.
  • Sun, D., Yan, Y., Carter, R., Gao, L., Lu, G., Riley, G. and Wood, M. (2014). On-Line Nonintrusive Detection of Wood Pellets in Pneumatic Conveying Pipelines Using Vibration and Acoustic Sensors. IEEE Transactions on Instrumentation and Measurement [Online] 63:993-1001. Available at: http://dx.doi.org/10.1109/TIM.2013.2292284.
    This paper presents a novel instrumentation system for on-line nonintrusive detection of wood pellets in pneumatic conveying pipelines using vibration and acoustic sensors. The system captures the vibration and sound generated by the collisions between biomass particles and the pipe wall. Time-frequency analysis technique is used to eliminate environmental noise from the signal, extract information about the collisions, and identify the presence of wood pellets. Experiments were carried out on an industrial pneumatic conveying pipeline to assess effectiveness and operability. The impacts of various factors on the performance of the detection system are compared and discussed, including different sensing (vibration sensor versus acoustic sensor), different time-frequency analysis methods (wavelet-based denoising versus classic filtering), and different system installation locations.
  • Sun, D., Yan, Y., Carter, R., Gao, L., Qian, X. and Lu, G. (2014). On-line automatic detection of wood pellets in pneumatically conveyed wood dust flow. AIP Conference Proceedings [Online] 1592:71-75. Available at: http://dx.doi.org/10.1063/1.4872088.
    This paper presents a piezoelectric transducer based system for on-line automatic detection of wood pellets in wood dust flow in pneumatic conveying pipelines. The piezoelectric transducer senses non-intrusively the collisions between wood pellets and the pipe wall. Wavelet-based denoising is adopted to eliminate environmental noise and recover the collision events. Then the wood pellets are identified by sliding a time window through the denoised signal with a suitable threshold. Experiments were carried out on a laboratory test rig and on an industrial pneumatic conveying pipeline to assess the effectiveness and operability of the system.
  • Zhou, H., Tang, Q., Yang, L., Yan, Y., Lu, G. and Cen, K. (2014). Support vector machine based online coal identification through advanced flame monitoring. Fuel [Online] 117:944-951. Available at: http://dx.doi.org/10.1016/j.fuel.2013.10.041.
    This paper presents a new on-line coal identification system based on support vector machine (SVM) to achieve on-line coal identification under variable combustion conditions. Four different coals were burnt in a 0.3 MW coal combustion furnace with different coal feed rates, total air flow rates and flow rate ratios of primary air and secondary air. The flame monitoring system was installed at the exit of the burner to acquire the coal flame images and light intensity signals. Spatial and temporal flame features were extracted for coal identification. The averaged prediction accuracy is 99.1%. The mean value of the infrared signal has the most significant influence on prediction accuracy. For “unstudied” operation cases, the prediction accuracy is 94.7%.
  • Hossain, M., Lu, G., Sun, D. and Yan, Y. (2013). Three-dimensional reconstruction of flame temperature and emissivity distribution using optical tomographic and two-colour pyrometric techniques. Measurement Science and Technology [Online] 24:74010. Available at: https://doi.org/10.1088/0957-0233/24/7/074010.
    This paper presents an experimental investigation, visualization and validation in the three-dimensional (3D) reconstruction of flame temperature and emissivity distributions by using optical tomographic and two-colour pyrometric techniques. A multi-camera digital imaging system comprising eight optical imaging fibres and two RGB charged-couple device (CCD) cameras are used to acquire two-dimensional (2D) images of the flame simultaneously from eight equiangular directions. A combined logical filtered back-projection (LFBP) and simultaneous iterative reconstruction and algebraic reconstruction technique (SART) algorithm is utilized to reconstruct the grey-level intensity of the flame for the two primary colour (red and green) images. The temperature distribution of the flame is then determined from the ratio of the reconstructed grey-level intensities and the emissivity is estimated from the ratio of the grey level of a primary colour image to that of a blackbody source at the same temperature. The temperature measurement of the system was calibrated using a blackbody furnace as a standard temperature source. Experimental work was undertaken to validate the flame temperature obtained by the imaging system against that obtained using high-precision thermocouples. The difference between the two measurements is found no greater than ±9. Experimental results obtained on a laboratory-scale propane fired combustion test rig demonstrate that the imaging system and applied technical approach perform well in the reconstruction of the 3D temperature and emissivity distributions of the sooty flame.
  • Sun, D., Lu, G., Zhou, H. and Yan, Y. (2013). Condition Monitoring of Combustion Processes Through Flame Imaging and Kernel Principal Component Analysis. Combustion Science and Technology [Online] 185:1400-1413. Available at: http://dx.doi.org/10.1080/00102202.2013.798316.
    This article presents a methodology for the diagnosis of abnormal conditions in a combustion process through flame imaging and kernel principal component analysis (KPCA). A digital imaging system is used to capture real-time flame images and radiation signals, from which flame characteristics such as flame area, brightness, non-uniformity, and oscillation frequency are quantified. These characteristics are used as the variables to establish the KPCA model of the combustion process. With the use of Hotelling's T2 and Q statistics, the monitoring of abnormal conditions of the combustion process is achieved. Unlike the traditional principal component analysis (PCA) method, the KPCA method is capable of dealing with nonlinear data via nonlinear mapping, which projects the original nonlinear input space into a high-dimensional linear feature space. The effectiveness of the methodology is demonstrated by applying the approach to processing the data obtained on a 9MWth heavy oil fired combustion test facility. Experimental results obtained show that the KPCA method outperforms the traditional PCA in discriminating between the normal and abnormal combustion conditions, even in cases where the number of training samples is limited.

Book section

  • Bai, X., Lu, G., Bennett, T., Peng, Y., Liu, H., Eastwick, C. and Yan, Y. (2016). Measurement of Coal Particle Combustion Behaviors in A Drop Tube Furnace Through High-speed Imaging and Image Processing. In: 2016 IEEE International Instrumentation and Measurement Technology Conference Proceedings. IEEE, pp. 1445-1450. Available at: https://dx.doi.org/10.1109/I2MTC.2016.7520582.
    This paper presents the measurement and characterization of single coal particles in a drop tube furnace through high speed imaging and image processing. A high speed camera coupling with a long distance microscope is employed to acquire the images of the particle during its residence time in the furnace. A set of physical quantities of the particle, including size, shape and boundary roughness, are defined and computed based on the images obtained, which are then used describe the combustion behaviors of the particle. Experimental results show that the combined high speed imaging and image processing technique has provided an effective means for measuring and quantifying the characteristics of single coal particles during combustion.
  • Hossain, M., Lu, G., Li, X. and Yan, Y. (2013). Three-dimensional temperature profiling of oxy-gas burner flames. In: 2013 IEEE International Conference on Imaging Systems and Techniques (IST). IEEE, pp. 91-94. Available at: http://dx.doi.org/10.1109/IST.2013.6729669.
    This paper presents the three-dimensional temperature measurement of an oxy-gas burner flame based on optical tomographic and two-color techniques. Eight two-dimensional (2-D) image projections of the flame are obtained concurrently by multiple imaging fiber bundles and imaging sensors. The LFBP (Logical Filtered Back-Projection) algorithm combined the SART (Simultaneous Algebraic Reconstruction Technique) is utilized to reconstruct the gray-level intensity of the flame based on its 2-D images. The flame temperature distribution is then determined using the reconstructed gray-levels through the two-color pyrometry. Experiments were undertaken under different oxy-fuel conditions. The results show that the flame temperature increases with the O2 concentration and that the temperature distribution of the oxy-fuel flame is more uniform at the root region than that at other regions.

Conference or workshop item

  • Hossain, M., Lu, G. and Chowdhury, W. (2020). Three-dimensional Reconstruction and Measurement of Avian Eggs through Digital Imaging. In: IEEE International Instrumentation and Measurement Technology Conference (I2MTC). IEEE. Available at: https://doi.org/10.1109/I2MTC43012.2020.9129156.
    This paper presents a computer vision-based method for the 3-D (three-dimensional) reconstruction and characterization of avian eggs. Two low-cost cameras are used to acquire images of eggs from top and side views, respectively. The image segmentation is performed using the image binarization technique. The contour-slice based method is employed for the 3-D reconstruction. The geometrical
    parameters of avian eggs, such as length, breadth, volume and surface area, are then computed based on the reconstructed model. The performance of the system is evaluated using eggs from different breeds and sizes. Comparative results between the physical measurement and the proposed approach suggest that the digital imaging approach has an overall accuracy of 98% for the geometrical parameter measurement of avian eggs
  • Hossain, M., Cugley, J., Lu, G., Smith, D., Kim, I. and Yan, Y. (2018). Investigations into the Impact of Coal Moisture on Burner Performance through Flame Imaging and Spectroscopic Analysis. In: 12th ECCRIA - The European Conference on Fuel and Energy Research and Its Applications. TFERF- The Fuel and Energy Research Forum.
    Despite increasing use of renewable energy worldwide, coal remains to be the primary energy resource to meet the increasing demand for electric power in many countries. However, coal-fired power plants have to cope with coals with different properties, including those with high moisture content. It is known that moisture content in coal does not only affect coal handling but also burner performance, and thus combustion efficiency and emission formation process. A study is recently carried out to investigate the impact of moisture content in coal on the burner performance through flame imaging and spectroscopic analysis. Experimental tests were conducted on a 40MWth coal-fired combustion test facility (CTF). A typical pulverised coal was fired in the study. The variation in evaporated coal moisture was replicated by injecting steam into the primary coal flow in the range of 7%-55% (PFM, primary flow moisture) under different operation conditions including variations in furnace load and fuel-to-air ratio. A flame imaging system and a miniature spectrometer were employed to acquire concurrently flame images and spetroscopic data (Fig. 1). The characteristic parameters of the flame such as spreading angle, temperature, oscillation frequency and spectral intensity are computed and their relationship with the operation conditions including PFM and emissions (NOx, CO) are quantified. Fig. 2 illustrates typical flame images under different steam injections. Detailed experimental results and analysis will be presented at the conference.
  • Lu, G. (2018). Experimental Investigation of Oxy-combustion Behaviour of Single Biomass Pellets using High-speed Imaging and Colour Processing Techniques. In: 12th ECCRIA - The European Conference on Fuel and Energy Research and Its Applications. TFERF- The Fuel and Energy Research Forum.
    Despite increasing use of renewable energy worldwide, coal remains to be the primary energy resource to meet the increasing demand for electric power in many countries. However, coal-fired Biomass fuel has been widely accepted as renewable energy in conventional power generation plants. Biomass fuels, however, can vary widely in properties, composition and structure, leading to drastically different 'fuel performance', particularly under oxy combustion conditions. Whilst considerable research has been carried out on the experimental studies and modelling of single biomass particle's ignition and combustion under conventional air combustion conditions, limited work has been undertaken in this area under oxy combustion conditions. This is largely due to the lack of a quantitative means to measure and characterise the combustion behaviours of biomass particles/pellets. In this study, a combination of high-speed and spectroscopic imaging, and image processing techniques is employed to investigate the combustion behaviours of single biomass pellets in a V-DTF (Visual Drop Tube Furnace) under oxy-fuel combustion conditions. Five different biomass pellets (i.e., wood, straw, peanut shell, miscanthus and terrified wood) combust under air and three oxy conditions (i.e., 21%O2/79%CO2, 25%O2/75%CO2, and 30%O2/70%CO2) for the pre-set furnace temperatures of 800 ?C and 900 ?C. Images of burning pellets are recorded using a high-speed camera (up to 900 fps) and an EMCCD camera for each test condition. Characteristic parameters of the burning pellet, such as flame size, temperature, are colour features, defined and computed based the images obtained, which are then used to study the impact of the oxy conditions on the combustion behaviour of the tested biomass fuels. The experimental results provide a useful reference for improved understanding in the fundamental aspects of physical and chemical behaviours of biomass fuels under oxy-firing conditions. Fig. 1 illustrates typical images of a miscanthus pellet under air and oxy conditions. Detailed experimental results will be presented at the conference.
  • Hossain, M., Cugley, J., Lu, G., Caesar, S., Cornwell, S., Riley, G. and Yan, Y. (2018). Burner Condition Monitoring based on Flame Imaging and Data Fusion Techniques. In: 12th ECCRIA - The European Conference on Fuel and Energy Research and Its Applications. TFERF- The Fuel and Energy Research Forum.
    Rapid growth in electricity generation from intermittent renewables has resulted in increasing demand in conventional fossil-fuel power stations for plant flexibility, load balancing and fuel flexibility. This has led to new challenges in plant monitoring and control, particularly securing combustion stability for optimizing combustion process in terms of furnace safety, fuel efficiency and pollutant emissions. An unstable combustion process can cause many problems including furnace vibration, non-uniform thermal distribution in the furnace, high pollutant emissions and unburnt carbon in the flue gas. The stability of burners should therefore be continuously monitored and maintained for the improved overall performance of the furnace. A study is carried out to investigate the burner stability based on flame imaging and data fusion techniques. Experiments were carried out on a 915 MWth coal-fired power station. A bespoke flame imaging system (Fig. 1) was employed to acquire flame images from 16 individual burners (4 mills each with 4 burners) with a frame rate up to 200 frames per second. The characteristic parameters of the flame, including temperature, non-uniformity, entropy, oscillation frequency and colour characteristics (hue, saturation and intensity), are computed. The relationship between the flame characteristics and burner inputs and flue gas emissions (e.g., NOx) is quantified. Stability index is then introduced as an indicator of the stability of individual burner. Fig. 2 illustrates typical flame images for different burners. Detailed test results and analysis will be presented at the conference.
  • Cugley, J., Lu, G., Yan, y and Searle, I. (2018). Flame monitoring and characterisation through digital imaging and spectrometry. In: IFRF 2018 Conference- Clean, Efficient and Safe Industrial Combustion. Institute of Measurement and Control.
    Fossil fuel fired boilers are often required to work under variable operation conditions. The variability in fuel diet and load conditions result in various problems in boiler performances. A methodology based on digital imaging and spectrometric techniques is proposed for flame monitoring and characterisation on utility boilers. The system developed consists of an optical probe/water jacket, a digital camera, a spectrometer covering a spectral range from 200nm to 900nm and an embedded computer with associated application software. Computer algorithms are established to determine flame characteristic parameters, including size, shape, temperature and spectral distributions. The spontaneous emissions of flame radicals (e.g., CH*and C2*) and alkali elements such as Sodium (Na) and Potassium (K) are characterised and their relationships with the combustion inputs (e.g., fuel, air-to-fuel ratio) and pollutant emissions (e.g., NOx) are studied. The methodology proposed are examined on a gas-fired heat recovery boiler under different operation conditions. The results obtained suggest there exist close correlations between flame parameters computed and boiler operation conditions. In particular, flame radicals (CH* and C2*) and their ratio show a close relationship with the air-to-fuel ratio. The spectral intensities of Na (589nm) and K (767nm) also illustrate a strong link to the type of fuel. Current work focuses on quantifying the relationship between the flame parameters and the boiler operation conditions and establishing a computational model for online prediction of emissions from flame characteristic parameters.
  • He, Y., Lu, G. and Yan, Y. (2016). 3-D reconstruction of an axisymmetric flame based on cone-beam tomographic algorithms. In: 10th International Conference on Sensing Technology (ICST). IEEE Xplore. Available at: http://dx.doi.org/10.1109/ICSensT.2016.7796314.
    This paper presents a method of 3-D (three-dimensional) reconstruction of an axisymmetric flame based on cone-beam tomographic algorithms. A FDK-based analytic tomographic algorithm is developed. Computer simulations are undertaken to evaluate the structural similarity between the template and the reconstructed volume so as to evaluate the effectiveness of the algorithm developed. Experimental tests are also conducted using a CCD camera to capture images of a candle flame. The 3-D reconstruction of the flame is then performed. The simulation and experimental results demonstrate the feasibility of the proposed cone-beam based tomographic algorithm for 3-D flame image reconstruction
  • Farias Moguel, O., Clements, A., Szuhánszki, J., Ingham, D., Ma, L., Hossain, M., Lu, G., Yan, Y. and Pourkashanian, M. (2016). Large eddy simulation of a coal flame: estimation of the flicker frequency under air and oxy-fuel conditions. In: 11th European Conference on Coal Research and Its Applications. Available at: http://www.coalresearchforum.org/conference.html.
    Fossil fuel combustion, such as coal combustion, currently meets the majority of the global energy demand; however, the process also produces a significant proportion of the worldwide CO2 greenhouse gas emissions. Further improvement in the efficiency and control of the combustion process is needed, as well as the implementation of novel technologies such as carbon capture and storage (CCS). Oxy-fuel combustion is a very promising CCS technology, where the air in the combustion process is replaced with a mixture of recycled flue gas and oxygen producing a high CO2 outflow that can effectively be processed or stored. The adjustment of the combustion environment within the boiler resulting from the high CO2 concentration will modify the flame characteristics. It is therefore important to evaluate properly the changes of the flame that occur with different flue gas recycle schemes.
    A coal flame is often characterised by its physical parameters, such as the flame size, shape, brightness and temperature, and it can be considered as a stable flame by the presence of ignition and the propagation of the flame. The oscillatory behaviour of a flame can be quantified by the flicker frequency obtained after the instantaneous variations of the flame parameters, and is used as a reference for flame stability.
    Computational Fluid Dynamics (CFD) is widely used to model coal combustion. This work compares the estimated flicker frequency taken from CFD calculations against measurements undertaken at the experimental facilities of the UKCCSRC Pilot-scale Advanced Capture Technology (PACT) located in South Yorkshire, UK. The 250 kW combustion test facility consists of a down-fired, refractory lined cylindrical furnace, which is 4 m in height with a 0.9 m internal diameter. The furnace is fitted with a scaled version of a commercially available Doosan Babcock low-NOx burner.
    The flame physical parameters are approximated from performing a Large Eddy Simulation (LES) using the CFD code ANSYS FLUENT v15. The flicker frequency obtained from the CFD approach is compared against the experimentally measured value from a 2D flame imaging system. A series of oxy-fuel cases are then examined in the same fashion in order to assess their flame stability and the boiler operational limit. The flicker frequency trend obtained from the computations and measurements helps to determine the dynamic response of the flame for different combustion environments, and the results will be applicable in determining the optimal recycle ratio applied in future oxy-fuel systems.
  • Xiaojing, B., Lu, G., Hossain, M., Szuhánszki, J., Daood, S., Yan, Y., Nimmo, W. and Pourkashanian, M. (2016). Multi-mode Combustion Process Monitoring through Flame Imaging and Soft-computing. In: 11th European Conference on Coal Research and Its Applications. UK: Coal Research Forum. Available at: http://www.coalresearchforum.org/conference.html.
    Reliable monitoring and diagnosis of combustion stability in combustion systems such as fossil-fuel fired boilers, gas turbines and combustion engines are crucial to maintain the system safety, combustion efficiency and low emissions, particularly under variable operation conditions. Considerable efforts have thus been made in developing techniques for online monitoring and diagnosis of the stability of a combustion process. Among those, flame imaging conjoined with image processing and soft computing techniques has been paid much attention for both laboratorial and industrial applications. Some imaging and soft computing techniques have been proposed for combustion state monitoring, but most of them can only detect a single-mode condition. However, modern combustion systems often operate under variable conditions (i.e., multi-mode process). Due to the dynamic nature of the combustion process, single-mode monitoring methods often mistakenly determine some normal combustion behaviours as abnormal ones. The recent trend of using a variety of fuels, including low quality coals, coal blends, and co-firing biomass and coal, has further deteriorated this issue.
    In this study, a method based on flame imaging and soft-computing techniques for multi-mode combustion process monitoring is proposed. Flame images are acquired using a flame imaging system. Mean intensity values of RGB image components and texture descriptors are extracted and computed from the grey-level co-occurrence matrix. Such features are then used as inputs to a combined PCA-KSVM (principle component analysis-kernel support vector machine) model for multi-mode process monitoring. In this method, the PCA serves for eliminating the impact of noise and instabilities on the mode recognition. The KSVM identifies the combustion mode by using the scores of the features in the principle component subspace. Finally, two multivariate statistic indices, T2 and SPE, are computed and used to assess the stabilities of the combustion process. The proposed approach has been examined by using flame images obtained on the UKCCSRC PACT 250kW PF (pulverised fuel) test rig under different operation conditions (e.g., variations in the primary air and secondary-territory air split). Test results have shown that the computed image features represent well the dynamic behaviours of the flame, and that the PCA-KSVM model has outperformed conventional methods in monitoring the multi-mode combustion process.
  • Cugley, J., Lu, G., Yan, Y. and Marshall, G. (2016). Flame Characterisation in a Multi-burner Heat Recovery Boiler through Digital Imaging and Spectrometry. In: 11th European Conference on Coal Research and Its Applications. Coal Research Forum. Available at: http://www.coalresearchforum.org/conference.html.
    Fossil fuel fired utility boilers fire a range of fuels under variable operation conditions. This variability in fuel diet and load conditions is linked to various problems in boiler performances, particularly the flame quality which is closely associated with furnace safety, combustion efficiency and pollutant emissions. Reliable flame monitoring is thus critical as the flame can fluctuate significantly in terms of size, shape, location, colour and temperature distribution. For instance, heat recovery water tube boilers are commonly used in industry to recover the energy in the exhaust gas from gas turbines. The boiler is fitted with multiple burners which allow flexibility with tuning of the boiler firing rates depending on process steam demand. It was reported that flame properties in such boilers had a direct impact on the flame stability and pollutant emissions (i.e., NOx and CO). There is, however, no technique available for online monitoring and quantifying the flame properties of individual burners. This has resulted in a lack of understanding in how each burner operates with regard to the overall performance of the boiler, particularly the emissions.
    Under the support of the BF2RA and EPSRC, an imaging and spectrometry based instrumentation system is being developed for flame monitoring and emission. Fig 1 shows the block diagram of the system. An optical probe, protected by the air-cooled jacket, transmits the light of flame to the camera house. The light of flame is then split into two beams. The first beam is captured by a camera to provide images for determining the physical parameters of the flame. The second beam is received by a miniature spectrometer for flame spectral analysis. Intelligent computing algorithms are developed for flame monitoring and emission prediction. The system, once fully developed, will be assessed under a range of operation conditions on a heat recovery water tube boiler at a British Sugar’s factory. More test results will be presented at the conference.
  • Hossain, M., Lu, G., Hatem, F., Valera-Medina, A., Marsh, R. and Yan, Y. (2015). Temperature Measurement of Gas Turbine Swirling Flames using Tomographic Imaging Techniques’. In: IEEE International Conference on Imaging Systems and Techniques 2015 (IST2’015). Available at: http://dx.doi.org/10.1109/IST.2015.7294541.
  • Li, N., Lu, G., Li, X. and Yan, Y. (2015). Prediction of NOx emissions from a biomass fired combustion process through digital imaging, nonnegative matrix factorization and fast sparse. In: IEEE International Instrumentation and Measurement Technology Conference (I2MTC 2015),. pp. 176-180.
  • Hossain, M., Lu, G. and Yan, Y. (2014). Soot volume fraction profiling of asymmetric diffusion flames through tomographic imaging. In: IEEE International Conference on Imaging Systems and Techniques 2014. pp. 427-431.
  • Li, G., Lu, G. and Yan, Y. (2014). Fire detection using stereoscopic imaging and image processing techniques’. In: IEEE International Conference on Imaging Systems and Techniques 2014 (IST2014). pp. 28-32.
  • Hossain, M., Sun, D., Lu, G., Yan, Y., Szuhánszki, J., Black, S., Nimmo, W., Ma, L. and Pourkashanian, M. (2014). 3-D reconstruction and characterisation of oxy-coal flames on a 250kW combustion test facility’. In: 10th European Conference on Coal Research and Its Applications.
  • Li, N., Lu, G., Li, X. and Yan, Y. (2014). A multiple linear regression approach to NOx emission prediction based on flame radical imaging and contourlet transform. In: 10th European Conference on Coal Research and Its Applications.
  • Li, N., Lu, G., Li, X. and Yan, Y. (2014). Prediction of pollutant emissions of biomass flames using digital imaging, contourlet transform and radial basis function network techniques. In: IEEE International Instrumentation and Measurement Technology Conference. pp. 697-700.
  • Hossain, M., Lu, G. and Yan, Y. (2014). Tomographic imaging based measurement of three-dimensional geometric parameters of a burner flame. In: Proceedings of IEEE International Instrumentation and Measurement Technology Conference 2014 (I2MTC 2014). pp. 1111-1114. Available at: http://dx.doi.org/10.1109/I2MTC.2014.6860915.
    This paper presents the measurement of 3-D (three-dimensional) flame geometric parameters based on optical fiber imaging and tomographic techniques. Two identical CCD (Charge-coupled Device) cameras coupled with eight imaging fiber bundles are used to capture the 2-D (two-dimensional) images of a burner flame concurrently from eight different directions around the burner. An optical tomographic algorithm LFBP-SART is utilized to reconstruct the cross-sections and generate a complete 3-D model of the flame. A set of geometric parameters, including length, volume, surface area and circularity, are then determined from the model generated and used for characterizing the flame. The proposed technical approach is firstly evaluated using an LED (light emitting diode) tube with known dimensions, and then on a gas-fired combustion rig. The results obtained demonstrate that the proposed algorithms are effective for measuring the 3-D geometric parameters of a burner flame over a range of combustion conditions.
  • Szuhánszki, J., Nimmo, W., Pourkashanian, M., Hossain, M., Lu, G. and Yan, Y. (2013). Experimental investigation of oxy-coal combustion at a 250 kW Combustion Test Facility. In: The 3rd Oxyfuel Combustion Conference.
    ABSTRACT Carbon Capture and Storage (CCS) technology has considerable potential to reduce CO2 emissions of the energy sector to near zero. Therefore it promises to make a major contribution in mitigating climate change, whilst enabling the continued use of fossil fuels over the coming decades. In addition, it will enhance the energy security of nations with significant fossil fuel reserves, and enable those relying on energy imports to maintain a more diverse range of supply (DECC, 2012). Oxy-fuel combustion is one of the most developed CCS technologies and is suitable for near-term deployment (Wall, 2011). However, in order to ensure the success of the first large scale plants, and thereby demonstrate the technical and economic feasibility of the technology, the fundamentals of the oxy-fuel combustion process have to be fully understood.
    In oxy firing atmospheric N2 is substituted with CO2 from the recycled flue gas, in order to increase the exit CO2 concentration and to moderate the flame temperatures within the process. This changes the fundamentals of the combustion process and, as a result, oxy-coal combustion differs from conventional air fired combustion in a number of ways, including coal reactivity, flame characteristics, heat transfer and emissions performance. This paper, which explores the combustion of coal under oxy-fuel conditions in a state of the art 250 kW Combustion Test Facility (CTF), focuses on flame characterisation and heat transfer performance.
  • Qiu, T., Yan, Y. and Lu, G. (2013). Flame stability monitoring through statistical analysis of the medial axis. In: Instrumentation and Measurement Technology Conference (I2MTC), 2013 IEEE International. pp. 1658-1661. Available at: http://dx.doi.org/10.1109/I2MTC.2013.6555695.
    On-line monitoring of flame stability is crucial as it is closely related to plant safety, combustion efficiency, and pollutant emissions. This paper presents a new method for the stability monitoring of a burner flame using its medial axis. Flame video clips are acquired on a combustion test rig under different combustion conditions. Each frame of the flame video clips is processed and the medial axis of the flame is extracted. The processed medial axes are then superimposed for statistical analysis. The test results show that, although the shape of the flame is randomly fluctuated even under the same combustion condition, some of the statistical shape parameters are relatively stable for the given burner condition. These parameters can be used to indicate the stability of the flame.
  • Zhang, Y., Lu, G. and Yan, X. (2013). Modeling and control for a class of singular forest resource system. In: Control Conference (CCC), 2013 32nd Chinese. pp. 2126-2131.
    Forest resources is not only an important global ecological system component but also the material basis of human life. The sustainable development of forest resources has become the key to maintain and improve the whole human environment. How to find the scientific and reasonable ways to manage the forest resources and ensure the profit of the manager and keep sustainable development of forest resources have become principal subjects in the research fields of forest development. The dynamic control models of forest resources are established in this paper, FTS controller and optimal performance index of the singular discrete systems will be obtained on the basis of the calculation for the models.
  • Sun, D., Yan, Y., Carter, R., Lu, G., Riley, G. and Wood, M. (2013). Detecting the presence of large biomass particles in pneumatic conveying pipelines using an acoustic sensor. In: Instrumentation and Measurement Technology Conference (I2MTC), 2013 IEEE International. pp. 1487-1490. Available at: http://dx.doi.org/10.1109/I2MTC.2013.6555661.
    This paper proposes a novel approach to online automatic detection of the presence of large biomass particles in a pneumatic conveying pipeline using an acoustic emission sensor and time-frequency analysis techniques. The acoustic sensor is used to capture the sound emitted from the collisions between biomass particles and pipe wall. Time-frequency analysis technique is used to eliminate environmental noise from the acoustic signal, extract the revealing information about the collisions, and identify the large particles. The acoustic sensor together with its signal conditioning unit is integrated into a compact enclosure, which can be easily attached to the outer face of a pneumatic pipeline. Experimental results obtained from an industrial pneumatic conveyor demonstrate the method works well and results are promising.

Thesis

  • Cugley, J. (2019). Advanced Flame Monitoring and Emission Prediction through Digital Imaging and Spectrometry.
    This thesis describes the design, implementation and experimental evaluation of a prototype instrumentation system for burner condition monitoring and NOx emissions prediction on fossil-fuel-fired furnaces.

    A review of methodologies and technologies for burner condition monitoring and NOx emissions prediction is given, together with the discussions of existing problems and technical requirements in their applications. A technical strategy, incorporating digital imaging, UV-visible spectrum analysis and soft computing techniques, is proposed. Based on these techniques, a prototype flame imaging system is developed. The system consists mainly of an optical and fibre probe protected by water-air cooling jacket, a digital camera, a miniature spectrometer and a mini-motherboard with associated application software. Detailed system design, implementation, calibration and evaluation are reported.

    A number of flame characteristic parameters are extracted from flame images and spectral signals. Luminous and geometric parameters, temperature and oscillation frequency are collected through imaging, while flame radical information is collected by the spectrometer. These parameters are then used to construct a neural network model for the burner condition monitoring and NOx emission prediction.

    Extensive experimental work was conducted on a 120 MWth gas-fired heat recovery boiler to evaluate the performance of the prototype system and developed algorithms. Further tests were carried out on a 40 MWth coal-fired combustion test facility to investigate the production of NOx emissions and the burner performance.

    The results obtained demonstrate that an Artificial Neural Network using the above inputs has produced relative errors of around 3%, and maximum relative errors of 8% under real industrial conditions, even when predicting flame data from test conditions not disclosed to the network during the training procedure. This demonstrates that this off the shelf hardware with machine learning can be used as an online prediction method for NOx.
  • Coombes, J. (2016). Development of Electrostatic and Piezoelectric Sensor Arrays for Determining the Velocity and Concentration Profiles and Size Distribution of Pneumatically Conveyed Bulk Solids.
    One way countries around the world are increasing the proportion of renewable fuels for electricity generation is to convert coal fired power stations to co-fired (biomass/coal fired) or converting coal fired power stations to burn only biomass fuels. This however has led to measurement challenges monitoring the complex multi-phase flow of the pulverised fuels entering the furnace due to the complex shape of biomass particles.
    To meet these measurement challenges a novel electrostatic sensor array and piezoelectric sensor array have been developed. The electrostatic sensor array consists of an array of electrostatic electrode pairs that span the diameter of the pipe. Consequently the electrostatic sensor array is capable of determining the particle velocity and concentration profiles as well as detecting specific flow regimes such as roping. The piezoelectric impact sensor array consists of an array of piezoelectric individual impact sensors that span the diameter of the pipe. The piezoelectric sensor array is capable of determining the particle concentration and size distribution profiles.
    Experimentation has been carried out on laboratory scale pneumatic conveying systems using a variety of materials such as coal, biomass, coal/biomass blends and plastic shot. Experiments using the electrostatic sensor array have shown that it is indeed capable of determining the particle velocity and concentration profiles in both dilute developed and undeveloped flows. Analysis of the standard deviation of the velocity profiles as well as the correlation coefficient profiles have indicated that parts of the pipe cross section have a more stable flow compared to others. Data obtained through on-line and off-line experimentation using the piezoelectric sensor array has shown that through selective frequency filtering of the impact signal particle size can be determined assuming the particle velocity and the mechanical properties of the conveyed pulverised materials are known. By using a threshold voltage to determine when an impact has occurred on each element of the piezoelectric sensor array the particle concentration profile has been determined. The concentration profiles measured by the piezoelectric sensor array were verified using the electrostatic sensor array.
Last updated