mdh.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Publications (10 of 18) Show all publications
Ahmed, M. U., Andersson, P., Andersson, T., Aparicio, E. T., Baaz, H., Barua, S., . . . Zambrano, J. (2019). A Machine Learning Approach for Biomass Characterization. In: Yan, J Yang, HX Li, H Chen, X (Ed.), INNOVATIVE SOLUTIONS FOR ENERGY TRANSITIONS: . Paper presented at 10th International Conference on Applied Energy (ICAE), AUG 22-25, 2018, Hong Kong, HONG KONG (pp. 1279-1287). ELSEVIER SCIENCE BV
Open this publication in new window or tab >>A Machine Learning Approach for Biomass Characterization
Show others...
2019 (English)In: INNOVATIVE SOLUTIONS FOR ENERGY TRANSITIONS / [ed] Yan, J Yang, HX Li, H Chen, X, ELSEVIER SCIENCE BV , 2019, p. 1279-1287Conference paper, Published paper (Refereed)
Abstract [en]

The aim of this work is to apply and evaluate different chemometric approaches employing several machine learning techniques in order to characterize the moisture content in biomass from data obtained by Near Infrared (NIR) spectroscopy. The approaches include three main parts: a) data pre-processing, b) wavelength selection and c) development of a regression model enabling moisture content measurement. Standard Normal Variate (SNV), Multiplicative Scatter Correction and Savitzky-Golay first (SGi) and second (SG2) derivatives and its combinations were applied for data pre-processing. Genetic algorithm (GA) and iterative PLS (iPLS) were used for wavelength selection. Artificial Neural Network (ANN), Gaussian Process Regression (GPR), Support Vector Regression (SVR) and traditional Partial Least Squares (PLS) regression, were employed as machine learning regression methods. Results shows that SNV combined with SG1 first derivative performs the best in data pre-processing. The GA is the most effective methods for variable selection and GPR achieved a high accuracy in regression modeling while having low demands on computation time. Overall, the machine learning techniques demonstrate a great potential to be used in future NIR spectroscopy applications.

Place, publisher, year, edition, pages
ELSEVIER SCIENCE BV, 2019
Series
Energy Procedia, ISSN 1876-6102 ; 158
Keywords
Artificial Neural Network, Chemometrics, Gaussian Process Regression, Near Infrared Spectroscopy, Multiplicative Scatter Correction, Standard Normal Variate, Support Vector Regression, Partial Least Squares, Savitzky-Golay derivatives
National Category
Energy Engineering
Identifiers
urn:nbn:se:mdh:diva-44835 (URN)10.1016/j.egypro.2019.01.316 (DOI)000471031701100 ()
Conference
10th International Conference on Applied Energy (ICAE), AUG 22-25, 2018, Hong Kong, HONG KONG
Available from: 2019-07-11 Created: 2019-07-11 Last updated: 2019-07-11Bibliographically approved
Ahmed, M. U., Andersson, P., Andersson, T., Tomas Aparicio, E., Baaz, H., Barua, S., . . . Zambrano, J. (2019). Real-time Biomass Characterization in Energy Conversion Processes using Near Infrared Spectroscopy: A Machine Learning Approach. In: Elsevier (Ed.), “Innovative Solutions for Energy Transitions”: . Paper presented at International Conference on Applied Energy, 2018 (pp. 1279-1287). , 158
Open this publication in new window or tab >>Real-time Biomass Characterization in Energy Conversion Processes using Near Infrared Spectroscopy: A Machine Learning Approach
Show others...
2019 (English)In: “Innovative Solutions for Energy Transitions” / [ed] Elsevier, 2019, Vol. 158, p. 1279-1287Conference paper, Published paper (Refereed)
Abstract [en]

The aim of this work is to apply and evaluate different chemometric approaches employing several machine learning techniques in order to characterize the moisture content in biomass from data obtained by Near Infrared (NIR) spectroscopy. The approaches include three main parts: a) data pre-processing, b) wavelength selection and c) development of a regression model enabling moisture content measurement. Standard Normal Variate (SNV), Multiplicative Scatter Correction and Savitzky-Golay first (SG1) and second (SG2) derivatives and its combinations were applied for data pre-processing. Genetic algorithm (GA) and iterative PLS (iPLS) were used for wavelength selection. Artificial Neural Network (ANN), Gaussian Process Regression (GPR), Support Vector Regression (SVR) and traditional Partial Least Squares (PLS) regression, were employed as machine learning regression methods. Results shows that SNV combined with SG1 first derivative performs the best in data pre-processing. The GA is the most effective methods for variable selection and GPR achieved a high accuracy in regression modeling while having low demands on computation time. Overall, the machine learning techniques demonstrate a great potential to be used in future NIR spectroscopy applications.

Series
Energy Procedia, ISSN 1876-6102
Keywords
Artificial Neural Netwrok; Chemometrics; Gaussian Process Regression; Multiplicative Scatter Correction; Standard Normal Variate; Support Vector Regression; Partial Least Squares; Savitzky-Golay derivatives
National Category
Environmental Engineering Environmental Biotechnology Industrial Biotechnology
Research subject
Energy- and Environmental Engineering
Identifiers
urn:nbn:se:mdh:diva-40396 (URN)2-s2.0-85063865772 (Scopus ID)
Conference
International Conference on Applied Energy, 2018
Projects
FUDIPO
Funder
EU, Horizon 2020, 723523
Available from: 2018-08-21 Created: 2018-08-21 Last updated: 2019-04-25Bibliographically approved
Skvaril, J., Kyprianidis, K. & Dahlquist, E. (2017). Applications of near-infrared spectroscopy (NIRS) in biomass energy conversion processes: A review. Applied spectroscopy reviews (Softcover ed.), 52(8), 675-728
Open this publication in new window or tab >>Applications of near-infrared spectroscopy (NIRS) in biomass energy conversion processes: A review
2017 (English)In: Applied spectroscopy reviews (Softcover ed.), ISSN 0570-4928, E-ISSN 1520-569X, Vol. 52, no 8, p. 675-728Article in journal (Refereed) Published
Abstract [en]

Biomass used in energy conversion processes is typically characterized by high variability, making its utilization challenging. Therefore, there is a need for a fast and non-destructive method to determine feedstock/product properties and directly monitor process reactors. The near-infrared spectroscopy (NIRS) technique together with advanced data analysis methods offers a possible solution. This review focuses on the introduction of the NIRS method and its recent applications to physical, thermochemical, biochemical and physiochemical biomass conversion processes represented mainly by pelleting, combustion, gasification, pyrolysis, as well as biogas, bioethanol, and biodiesel production. NIRS has been proven to be a reliable and inexpensive method with a great potential for use in process optimization, advanced control, or product quality assurance.

Keywords
Biodiesel, bioethanol, biogas, biomass, chemometrics, near-infrared spectroscopy, NIRS, near infrared, NIR, instrumentation
National Category
Analytical Chemistry Energy Systems Chemical Process Engineering Bioenergy Bioprocess Technology
Research subject
Biotechnology/Chemical Engineering; Energy- and Environmental Engineering
Identifiers
urn:nbn:se:mdh:diva-34992 (URN)10.1080/05704928.2017.1289471 (DOI)000412218800001 ()2-s2.0-85029956089 (Scopus ID)
Available from: 2017-03-03 Created: 2017-03-03 Last updated: 2018-07-25Bibliographically approved
Kyprianidis, K. & Skvaril, J. (Eds.). (2017). Developments in Near-Infrared Spectroscopy. Rijeka, Croatia: InTech
Open this publication in new window or tab >>Developments in Near-Infrared Spectroscopy
2017 (English)Collection (editor) (Refereed)
Abstract [en]

Over the past few decades, exciting developments have taken place in the field of near-infrared spectroscopy (NIRS). This has been enabled by the advent of robust Fourier transform interferometers and diode array solutions, coupled with complex chemometric methods that can easily be executed using modern microprocessors. The present edited volume intends to cover recent developments in NIRS and provide a broad perspective of some of the challenges that characterize the field. The volume comprises six chapters overall and covers several sectors. The target audience for this book includes engineers, practitioners, and researchers involved in NIRS system design and utilization in different applications. We believe that they will greatly benefit from the timely and accurate information provided in this work.

Place, publisher, year, edition, pages
Rijeka, Croatia: InTech, 2017. p. 150
National Category
Medical Laboratory and Measurements Technologies Chemical Process Engineering Energy Systems Renewable Bioenergy Research
Identifiers
urn:nbn:se:mdh:diva-35026 (URN)10.5772/62932 (DOI)978-953-51-3018-5 (ISBN)978-953-51-3017-8 (ISBN)
Available from: 2017-03-15 Created: 2017-03-15 Last updated: 2018-01-13Bibliographically approved
Skvaril, J., Kyprianidis, K., Avelin, A., Odlare, M. & Dahlquist, E. (2017). Fast Determination of Fuel Properties in Solid Biofuel Mixtures by Near Infrared Spectroscopy. Paper presented at 8th International Conference on Applied Energy, ICAE 2016; Beijing; China; 8 October 2016 through 11 October 2016. Energy Procedia, 105, 1309-1317
Open this publication in new window or tab >>Fast Determination of Fuel Properties in Solid Biofuel Mixtures by Near Infrared Spectroscopy
Show others...
2017 (English)In: Energy Procedia, ISSN 1876-6102, E-ISSN 1876-6102, Vol. 105, p. 1309-1317Article in journal (Refereed) Published
Abstract [en]

This paper focuses on the characterization of highly variable biofuel properties such as moisture content, ash content and higher heating value by near-infrared (NIR) spectroscopy. Experiments were performed on different biofuel sample mixtures consisting of stem wood chips, forest residue chips, bark, sawdust, and peat. NIR scans were performed using a Fourier transform NIR instrument, and reference values were obtained according to standardized laboratory methods. Spectral data were pre-processed by Multiplicative scatter correction correcting light scattering and change in a path length for each sample. Multivariate calibration was carried out employing Partial least squares regression while absorbance values from full NIR spectral range (12,000–4000 cm-1), and reference values were used as inputs. It was demonstrated that different solid biofuel properties can be measured by means of NIR spectroscopy. The accuracy of the models is satisfactory for industrial implementation towards improved process control. 

Place, publisher, year, edition, pages
Amsterdam: Elsevier, 2017
Keywords
Ash content; biofuels; higher heating value, moisture content, Near infrared spectroscopy, NIRS.
National Category
Energy Systems Analytical Chemistry
Research subject
Energy- and Environmental Engineering; Biotechnology/Chemical Engineering
Identifiers
urn:nbn:se:mdh:diva-33988 (URN)10.1016/j.egypro.2017.03.476 (DOI)000404967901061 ()2-s2.0-85020707357 (Scopus ID)
Conference
8th International Conference on Applied Energy, ICAE 2016; Beijing; China; 8 October 2016 through 11 October 2016
Available from: 2016-11-27 Created: 2016-11-27 Last updated: 2018-07-25Bibliographically approved
Winn, O., Sivaram, K. T., Aslanidou, I., Skvaril, J. & Kyprianidis, K. (2017). Near-infrared spectral measurements and multivariate analysis for predicting glass contamination of refuse-derived fuel. Paper presented at 9th International Conference on Applied Energy, ICAE2017, 21-24 August 2017, Cardiff, UK. Energy Procedia, 142, 943-949
Open this publication in new window or tab >>Near-infrared spectral measurements and multivariate analysis for predicting glass contamination of refuse-derived fuel
Show others...
2017 (English)In: Energy Procedia, ISSN 1876-6102, E-ISSN 1876-6102, Vol. 142, p. 943-949Article in journal (Refereed) Published
Abstract [en]

This paper investigates how glass contamination in refuse-derived fuel can be quantitatively detected using near-infrared spectroscopy. Near-infrared spectral data of glass in four different background materials were collected, each material chosen to represent a main component in municipal solid waste; actual refuse-derived fuel was not tested. The resulting spectra were pre- processed and used to develop multi-variate predictive models using partial least squares regression. It was shown that predictive models for coloured glass content are reasonably accurate, while models for mixed glass or clear glass content are not; the validated model for coloured glass content had a coefficient of determination of 0.83 between the predicted and reference data, and a root- mean-square error of validation of 0.64. The methods investigated in this paper show potential in predicting coloured glass content in different types of background material, but a different approach would be needed for predicting mixed type glass contamination in refuse-derived fuel. 

National Category
Energy Engineering
Identifiers
urn:nbn:se:mdh:diva-37507 (URN)10.1016/j.egypro.2017.12.151 (DOI)000452901601017 ()2-s2.0-85041551965 (Scopus ID)
Conference
9th International Conference on Applied Energy, ICAE2017, 21-24 August 2017, Cardiff, UK
Available from: 2017-12-19 Created: 2017-12-19 Last updated: 2019-01-03Bibliographically approved
Mirmoshtaghi, G., Skvaril, J., Campana, P. E., Li, H., Thorin, E. & Dahlquist, E. (2016). The influence of different parameters on biomass gasification in circulating fluidized bed gasifiers. Energy Conversion and Management, 126, 110-123
Open this publication in new window or tab >>The influence of different parameters on biomass gasification in circulating fluidized bed gasifiers
Show others...
2016 (English)In: Energy Conversion and Management, ISSN 0196-8904, E-ISSN 1879-2227, Vol. 126, p. 110-123Article in journal (Refereed) Published
Abstract [en]

The mechanism of biomass gasification has been studied for decades. However, for circulating fluidized bed (CFB) gasifiers, the impacts of different parameters on the gas quality and gasifiers performance have still not been fully investigated. In this paper, different CFB gasifiers have been analyzed by multivariate analysis statistical tools to identify the hidden interrelation between operating parameters and product gas quality, the most influencing input parameters and the optimum points for operation. The results show that equivalence ratio (ER), bed material, temperature, particle size and carbon content of the biomass are the input parameters influencing the output of the gasifier the most. Investigating among the input parameters with opposite impact on product gas quality, cases with optimal gas quality can result in high tar yield and low carbon conversion while low tar yield and high carbon conversion can result in product gas with low quality. However using Olivine as the bed material and setting ER value around 0.3, steam to biomass ratio to 0.7 and using biomass with 3 mm particle size and 9 wt% moisture content can result in optimal product gas with low tar yield.

National Category
Chemical Process Engineering
Identifiers
urn:nbn:se:mdh:diva-32432 (URN)10.1016/j.enconman.2016.07.031 (DOI)000385326400011 ()2-s2.0-84982682364 (Scopus ID)
Available from: 2016-08-04 Created: 2016-08-04 Last updated: 2018-12-18Bibliographically approved
Skvaril, J., Kyprianidis, K., Avelin, A., Odlare, M. & Dahlquist, E. (2015). Application of Near Infrared Spectroscopy for Rapid Characterization of Feedstock Material in Pulp and Paper Industry. In: Book of abstracts: . Paper presented at 17th International Conference on Near Infrared Spectroscopy - NIR 2015.
Open this publication in new window or tab >>Application of Near Infrared Spectroscopy for Rapid Characterization of Feedstock Material in Pulp and Paper Industry
Show others...
2015 (English)In: Book of abstracts, 2015Conference paper, Oral presentation with published abstract (Refereed)
Abstract [en]

Pulp digesters can be continuous or batch reactors with significant residence time which are fed with woodchips and cooking chemicals. They deliver the pulp-fibers that are used in the production of paper, as well as black liquor that is combusted in the chemical recovery boiler. The possibility to measure what is happening inside the digester is limited. The most important quality properties of the feedstock material is content of lignin, which is being dissolved during the process, and related material reactivity. Pulp quality after the process is measured by Kappa number which is a measure of residual lignin in the pulp. One of the biggest challenges in pulp production process is the great variability in feedstock material properties. If the process is not adjusted by well-timed and appropriate operational control measures i.e. control of inlet and outlet flows and setting of the cooking recipe, it will result in the large variations in Kappa number, lower fiber quality or excess use of environmentally harmful cooking chemicals. This becomes particularly important during the swing between softwood and hardwood as part of meeting the final paper product quality requirements. Therefore, a rapid method that is capable of continuous feedstock material characterization is required.Near infrared (NIR) spectroscopy can be used for non-destructive characterization of the feedstock material. In this study, both Fourier transform and grating NIR spectrophotometers were used for NIR absorbance spectra acquisition. Each spectrum was recorded in the range between 700 and 2500 nm. During the calibration of spectra of various wood species with known lignin content, wood samples were placed on a tray so that the tray may move horizontally in a reciprocating manner underneath the sensor while maintaining the constant distance between the sensor and sample. This was done in order to simulate the movement of a real conveyor belt as used for transporting feedstock to the digester. In the on-line application the NIR meter is situated above the conveyor belt that wood up to the digester.Spectral data were pretreated with different methods such as normalization, scatter correction, smoothing, first and second derivative (Savitzky-Golay algorithm), selection of different spectral ranges and its combinations. Mathematical models to estimate lignin content were constructed using Partial Least Square Regression (PLS-R) and Principle component regression (PCR) statistical methods. Response data for model build-up were determined in the chemical laboratory according to standardized procedures including test repetitions. Different combinations of NIR instrument used, pre-treatment methods and statistical methods were evaluated in order to find the model with the best prediction performance.Results are promising and demonstrate that it is possible to characterize the lignin content and reactivity of the feedstock material by NIR spectrophotometers with reasonable prediction model performance. Improved prediction can be obtained if only selected spectral ranges are included as an input for statistical modelling; similarly using derivatives is better than using the raw spectrum. In the next step, developed statistical models for rapid lignin content prediction will be used as a feed-forward input for dynamic process control.

National Category
Energy Systems
Identifiers
urn:nbn:se:mdh:diva-29848 (URN)
Conference
17th International Conference on Near Infrared Spectroscopy - NIR 2015
Projects
SPECTRA
Available from: 2015-12-07 Created: 2015-12-07 Last updated: 2016-02-25Bibliographically approved
Skvaril, J., Kyprianidis, K., Avelin, A., Odlare, M. & Dahlquist, E. (2015). Fast Determination of Lignin Content in Feedstock Material for Pulping Process Monitoring and Optimization. In: ICAVS 8 - Abstracts poster: . Paper presented at 8th International Conference on Advanced Vibrational Spectroscopy, 12 July 2015 to 17 July 2015, Vienna University of Technology, Vienna, Austria (pp. 556-557).
Open this publication in new window or tab >>Fast Determination of Lignin Content in Feedstock Material for Pulping Process Monitoring and Optimization
Show others...
2015 (English)In: ICAVS 8 - Abstracts poster, 2015, p. 556-557Conference paper, Poster (with or without abstract) (Refereed)
Abstract [en]

Pulping process is delivering pulp fibers which are further used in the production of paper. The reactor is fed with feedstock material in the form of wood chips. Moreover, cooking chemicals are brought at several points into the reactor. Previous studies have shown that the knowledge of the feedstock material properties which are highly variable is limited. One of the most important parameters is the lignin content, which has to be dissolved, this requires a significant residence time. The residual lignin in the resulting pulp after the process is measured in the form of Kappa number. Inappropriate application of cooking chemicals could lead to large variations in the Kappa number, low fiber quality and other issues. Therefore continuous characterization of the feedstock material is required. One of the available methods for nondestructive characterization of feedstock material is NIR spectroscopy. Presented study is conducted in order to assess the possibility of determining lignin content using NIR method. The spectroscopy workflow consist of four major steps i.e. sample preparation, spectral data acquisition, data pre-processing and multivariate calibration. We used test samples from 13 different tree species, which were tested in the form of wood chips, pulverized wood and mixture of both. Acquired spectral data were pre-processed mainly by second derivative and standard normal variate transformation. PLS regression with full cross validation was used for the development of a calibration model based on selected wavelengths. Acquisition of reference variable has been done according to standardized procedures and it represents the total amount of lignin in the sample.

The results of lignin characterization in feedstock material by NIR are very promising. The resulting PLS regressionmodel includes 2-factors and uses 16 predicting variables, resulting in R2 = 0,975, RMSE = 0,885 wt%. In the next step, presented work will be improved by applying large amount of samples, independent validation data set and by simulation of conveyor belt movements. The objective of this research is to test the NIR method at a real pulp digester, in order to improve monitoring andoptimization of the process. Furthermore, continuous characterization of the feedstock materials is intended to be used for the improvement of the control process. The measured lignin content will be compared to the content calculated within the pulp digester physical model and the Kappa number. This will be used for improving the digester physical model accuracy and as an input to advanced model based control, where the correlation will be made not only to lignin content but also with the feedstock material reactivity.

Keywords
lignin content, near infrared (NIR) spectroscopy, data pre-processing, multivariate calibration, partial least squares (PLS) regression, process optimization and control
National Category
Energy Systems
Identifiers
urn:nbn:se:mdh:diva-29850 (URN)978-3-200-04205-6 (ISBN)
Conference
8th International Conference on Advanced Vibrational Spectroscopy, 12 July 2015 to 17 July 2015, Vienna University of Technology, Vienna, Austria
Available from: 2015-12-07 Created: 2015-12-07 Last updated: 2016-01-14Bibliographically approved
Mirmoshtaghi, G., Skvaril, J., Li, H., Thorin, E. & Dahlquist, E. (2015). INVESTIGATION OF EFFECTIVE PARAMETERS ON BIOMASS GASIFICATION IN CIRCULATING FLUIDIZED BED GASIFIERS. In: : . Paper presented at 2015 AIChE Annual Meeting, November 8-13, 2015, Salt Lake City, UT, USA.
Open this publication in new window or tab >>INVESTIGATION OF EFFECTIVE PARAMETERS ON BIOMASS GASIFICATION IN CIRCULATING FLUIDIZED BED GASIFIERS
Show others...
2015 (English)Conference paper, Oral presentation with published abstract (Refereed)
National Category
Energy Engineering
Identifiers
urn:nbn:se:mdh:diva-31582 (URN)978-0-8169-1094-6 (ISBN)
Conference
2015 AIChE Annual Meeting, November 8-13, 2015, Salt Lake City, UT, USA
Available from: 2016-05-12 Created: 2016-05-12 Last updated: 2016-12-27Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0002-5341-3656

Search in DiVA

Show all publications