Modelovanje i predikcija performansi flotacije korišćenjem metode potpornih vektora za regresiju

Autori

DOI:

https://doi.org/10.5937/ror1701031D

Ključne reči:

deinking, flotacija, reciklaža papira, mašinsko učenje, regresija - metoda nosećih vektora

Apstrakt

Poslednjih godina vrše se značajna ulaganja u poboljšanje procesa reciklaže papira kako bi se uticalo na očuvanje šumskih, vodnih i energetskih resursa. Flotacija predstavlja najznačajniji postupak za separaciju čestica tonera od vlakana celuloze iz suspenzije papira. Modelovanje procesa flotacije je vrlo složeno, a dobijeni modeli su često komplikovani za implementaciju i upotrebu. U ovom radu predstavljen je model za predikciju performansi procesa flotacije koji je zasnovan na upotrebi metode potpornih vektora za regresiju (SVR). Reprezentativni uzorci za učenje modela dobijeni su u laboratorijskim uslovima, ispitivanjem uticaja različitih promenljivih kojima se kontroliše proces flotacije, poput tipa i koncentracije surfakanta, pH vrednosti i vremena flotiranja. Dobijeni rezultati potvrđuju da se metod potpornih vektora za regresiju može uspešno koristiti za predikciju performansi procesa flotacije čak i kada je skup podataka za učenje modela ograničen.

Reference

Ali T., McLellan F., Adiwinata J., May M., Evants, T., Functional and perfomance characteristics of solube silicate in deinking. Part I: alkaline deinking of newsprint/magazine. Journal of Pulp and Paper Science, 20 (1), 1994, 3-8,

Azevedo M.A.D., Drelich J., Miller J.D., The Effect of pH On Pulping and Flotation of Mixed Office Wastepaper, Journal of Pulp And Paper Science, 25 (9), 1999, 317-320,

Bajpai P., Recycling and Deinking of Recovered Paper, Elsevier insights, 1, 2014, 304.,

https://doi.org/10.1016/B978-0-12-416998-2.00001-5

Behin J., Vahed S., Effect of alkyl chain in alcohol deinking of recycled fibers by flotation process, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 297, 2007, 131-141,

https://doi.org/10.1016/j.colsurfa.2006.10.037

Borchardt J. K., Mehanistic insights into de-inking, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 88 (1), 1994, 13-25,

https://doi.org/10.1016/0927-7757(94)80081-2

Carre B., Galland G., Overview of deinking technology, 8th CTP/PTS Deinking Technology, Grenoble, 2007, 1-21,

Chang C. C., Lin C. J., LIBSVM: A Library for Support Vector Machines, ACM Transactions on Intelligent Systems and Technology, 2 (3), 2011, Article No. 27,

https://doi.org/10.1145/1961189.1961199

Chehreh Chelgania S., Shahbazib B., Hadavandi E., Support vector regression modeling of coal flotation based on variable importance measurements by mutual information method, Measurement, 114, 2018, 102-108,

https://doi.org/10.1016/j.measurement.2017.09.025

Costa C. A., Rubio J., Deinking flotation: influence of calcium soap and surface-active substance, Minerals Engineering, 18 (1), 2005, 59-64,

https://doi.org/10.1016/j.mineng.2004.05.014

Cortes C., Vapnik V., Support-vector networks, Machine Learning, 20, 1995, 273-297,

https://doi.org/10.1023/A:1022627411411

Dorris G. M., Sayegh N. N., The role of print layer thickness and cohesiveness on deinking of toner printed papers, Pulping Conference Proceedings, TAPPI PRESS, 1994, 1273-1289,

Dorris G., Ben Y., Richard M., Overview of flotation Deinking, Progress in paper recycling, 20 (1), 2011, 3-43,

Gong R., New Approaches on Deinking Evaluations, (Dissertations), Western Michigan University, 2013, Paper 184,

Jiang C., Ma J., Deinking of waste paper: flotation, Encyclopedia of Separation Science, Academic press, 2000, 2537-2544,

https://doi.org/10.1016/B0-12-226770-2/05881-6

Labidi J., Pelach M. A., Turon X., Mutje P., Predicting flotation efficiency using neural networks, Intensification, 46 (4), 2007, 314-322,

https://doi.org/10.1016/j.cep.2006.06.011

Laperrière L., Wasik L., Modeling and simulation of pulp and paper quality characteristics using neural networks, Tappi J., 84, 2001, 1-15,

Liphard M., Schereck B., Hornfeck K., Surface chemical aspects of filler flotation in waste paper recycling, Pulp and Paper Canada, 98 (8), 1993, 218-222,

Luo Q., Deng Y., Zhu J., Shin W. T., Foam control using a foaming agent spray: A novel concept for flotation deinking of waste paper, Industrial & Engineering Chemistry Research, 15 (42), 2003, 3578-3583,

https://doi.org/10.1021/ie021018g

Nakhaei F., Irannajad M., Application and comparison of RNN, RBFNN and MNLR approaches on prediction of flotation column performance, International Journal of Mining Science and Technology, 25 (6), 2015, 983-990,

https://doi.org/10.1016/j.ijmst.2015.09.016

Pathak P., Bhardwaj N.K., Singh A.K., Optimization of Chemical and Enzymatic Deinking of Photocopier Waste Paper, BioResources 6 (1), 2011, 447-463,

https://doi.org/10.15376/biores.6.1.447-463

Pauck W. J., Venditti R., Pocock J., Andrew J., Using statistical experimental design techniques to determine the most effective variables for the control of the flotation deinking of mixed recycled paper grades, Tappsa Journal, 2, 2012, 28-41,

Pauck W. J., Venditti R., Pocock J., Andrew J., Neural network modelling and prediction of the flotation deinking behaviour of industrial paper recycling processes, Recycling Nordic Pulp & Paper Research Journal, 29 (3), 2014, 521-532,

https://doi.org/10.3183/npprj-2014-29-03-p521-532

Pelach Serra, M. A., Proces de destintatge del paper per flotacio. Avaluacio de l'eficacia d'eliminacio de tinta, (Dissertation), Universitet de Girona, Girona, 1997, 278,

Svensson R., The influence of surfactant chemistry on flotation deinking, (Dissertation), University of Technology, Sweden, 2011, 12-31,

Trumić M. S., Antonijević M., Trumić M. Ž., Bogdanović G., The application of mineral processing techniques for the printed paper recycling, Recycling and Sustainable Development 9 (1), 2016, 47-57,

https://doi.org/10.5937/ror1601047T

Trumić M. Ž., Trumić M. S., Marković Z., Separation of ink particles from waste newspaper by deinking flotation, Journal of Mining and Metallurgy, 43 (A), 2007, 33-41,

Vapnik V., The Nature of Statistical Learning Theory, Springer, New York, 1995,

https://doi.org/10.1007/978-1-4757-2440-0

Vapnik V., Golowich S., Smola A., Support vector method for function approximation, regression estimation, and signal processing, Advances in Neural Information Processing Systems, 9, 1997, 281-287,

Verikas F., Malmqvist K., Bacauskiene M., Bergman L., Monitoring the de-inking process through neural network-based coulor image analysis, Neural Computing & Applications, 9, 2000, 142-151.

https://doi.org/10.1007/s005210070025

##submission.downloads##

Objavljeno

2017-12-15

Broj časopisa

Rubrika

Articles