Problem wyboru właściwej architektury sieci neuronowej

• Autorzy: Tadeusiewicz R.

Warianty tytułu

EN

Problem of selection of proper architecture for neural network

• Języki publikacji: PL

Abstrakty

PL

W pracy dokonano przeglądu zagadnień związanych ze stosowaniem sieci neuronowych w zadaniach związanych z szeroko rozumianą metalurgią, wskazując na uwarunkowania, jakie wskazane zadania narzucają na wybór struktury używanej sieci. W powiązaniu z wcześniejszymi pracami Autora, dyskutującymi zagadnienia metod uczenia sieci oraz zasady wyboru danych tworzących zbiór uczący - tworzy to zasób wiedzy wystarczający do tego, by podjąć samodzielne eksperymenty z użyciem sieci neuronowych jako narządzi w dalszych zagadnieniach metalurgicznych. Właśnie takie jest przeznaczenie tej pracy - tworzącej naukowe przesłanki dla dalszych praktycznych zastosowań neurokomputingu w metalurgii.

EN

Review of problems connected with applications of artificial neural networks in generally understood metallurgy is discussed in the paper. Conditions, which particular tasks impose on the selection of the network architecture, are pointed out. These suggestions, connected with earlier Author's research on training the networks, are a useful guide for further experiments aiming at an application of the artificial neural networks as a tool for solving of various problems in metallurgy. Scientific basis for practical metallurgical applications is given in the paper

Słowa kluczowe

PL

sieci regresyjne; sieci neuronowe; sieci klasyfikacyjne; architektura sieci; sieci liniowe; metalurgia; neurokomputing

• Wydawca: Wydawnictwo Naukowe AKAPIT
• Czasopismo: Informatyka w Technologii Materiałów
• Rocznik: 2001
• Tom: T. 1, Nr 1
• Strony: 3--22
• Opis fizyczny: Bibliogr. 57 poz., rys.

Twórcy

autor

Tadeusiewicz R.

• Katedra Automatyki, Akademia Górniczo-Hutnicza

Bibliografia

• Aistleitner K., Mattersdorfer L.G., Haas W., Kugi A., 1996, Neural network for identification of roll eccentricity in rolling mills, J. Mat. Proc. Techn., 60, 387-392.
• Cerqueira J.J.F., Palhares A.G.B., Madrid M.K., 2000, A complement to the back-propagation algorithm: an upper bound for the learning rate, Proc. IEEE-INNS-ENNS Int. Joint Conf. on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium. IEEE Comput. Soc., 4, 517-522 Chidambaram P.R., Jones J., Cao X.Y., Rohdes, V., 1995, Model¬ling of weld metal properties as a function of weld metal composition. Welding and Weld Automation in Shipbuil-ding, Cleveland, 123-133.
• Cios K.J., Baaklini G.Y., Vary A., Tjia R.E., 1994, Radial basis function network learns ceramic processing and predicts related strength and density, ASTM J. Testing and Evaluation, 22, 343-350.
• Cios K.J., Berke L., Vary A., Sharma S., 1996, Soft Computing Methods in Design of Superalloys, Report NASA Nr. TM- 106888.
• Colla V., Reyneri L.M., Sgarbi M., 2000, Neuro-wavelet parame-tric modeling. Proceedings of the IEEE-INNS-ENNS In-ternational Joint Conference on Neural Networks. IJCNN 2000, Neural Computing: New Challenges and Perspectives for the New Millennium, IEEE Comput. Soc., 4, 499-504.
• Cooper D.J., Woli S.L.B., Souder B.V., 1996, Online pattern-based part quality monitoring of the injection molding pro-cess, Polymer Eng Sci., 36, 1477-1488.
• Cuo Z. i in., 1995, Neural network approach in research of metal corrosion in soil. Corrosion Science and Protection Tech-nique, 1, 258-262.
• Demirci H.H., Coulter J.P., Guceri S.I., 1996, A numerical and experimental investigation of neural network-based intelligent control of molding processes, J. Manuf. Sei. Eng. 119, 88-94.
• Di S., Thomson P.F., 1997, Neural network approach for prediction of wrinkling limit in square metal sheet under diago-nal tension. ASTM, J. Testing and Evaluation, 25, 74-81.
• Draeger A., Gesthuisen R., 1996, Neuronal network for estimation of the degree of polymerization in the polycondensa-tion ofpolyethyleneterephthalate, Chemie Ingenieur Tech-nik, 68, 1320-1323.
• Fujii H., Mackay D.J.C., Bhadeshia H.K.D.H., 1996, Bayesian neural network analysis of fatigue crack growth rate in nickel base superalloys, ISIJ Int., 36, 1373-1382.
• Gavard L., Bhadeshia H.K.D.H., MacKay D.J.C., Suzuki S., 1996, Bayesian neural network model for austenite formation in steels, Mat. Sei. Techn., 12,453-463.
• Giles A.E., Aldrich C., van Deventer J.S.J., 1996, Modelling of rare earth solvent extraction with artificial neural nets. Hydrometallurgy, 43, 241-255.
• Gradisek J., Govekar E., Grabec I., 1996, A chaotic cutting process and determining optimal cutting parameter values using neural networks, Int. J. Mach. Tools & Manuf act., 36, 1161-1172.
• Hattori M. i in, 1996, Estimation of thermal-deformation in machine tools using neural network technique, J. Mat. Proc. Techn., 56, 765-772.
• Hill E.v.K, Walker J.L., Rowell G.H., 1994, Neutral network burst pressure prediction in graphite/epoxy pressure vessels from acoustic emission amplitude data, NDE, An Answerfor an Aging World, New Orleans, 161-163.
• Hill E.v.K, Walker J.L., Rowell G.H., 1996, Burst pressure prediction in graphite/epoxy pressure vessels using neural ne-tworks and acoustic emission amplitude data, Materials Evaluation, 54, 744-748, 754.
• Isalgue A. i in., 1994, Meso-scale model of a Cu-Zn-Al single crystal SMA, Mechanics of Phase Transformations and Shape Memory Alloys, Chicago, 71-84.
• Izworski A., Tadeusiewicz R., 2000, Classification of auditory brainstem response signals by application of neural ne-tworks, in: Halici U., Leblebicioglu K., Atalay V., Nalcaci E. (eds.), Brain Machine, METU, Ankara, 146-156.
• Jin B., Soeda M., Oshima K., 1996, Control of weldpool width and cooling time in TIG welding using a neural network model, Welding Int., 10, 614-621.
• Johansen B.S., Lilholt H., Liisberg C., 1995, Design for durability of polymeric composites, Tenth Int. Conf. on Composite Materials, I. Fatigue and Fracture, Whistler, 683-690.
• Jung J -Y, Im Y-T., Lee-Kwang H., 1996, Fuzzy-control simulation of cross-sectional shape in six-high cold-rolling mills, J. Mat. Proc. Techn. 62, 61-69.
• Keshavaraj R., Tock R.W., Narayan R.S., Bartsch R.A., 1995, Neural-network-based model approach for density of high- molecular-weight esters used as plasticizers. Adv. Polym. Technol., 14, 215-225.
• Kim B.H., Hwang T.W., Farooqi S., Witkus, S., 1995a, Prediction of injection molded plaque properties using artificial neu-ral networks and flow simulation, ANTEC ’95. Vol. I-Processing, Boston, 714-721.
• Kim B.H., Hwang T.W., Park, H J, 1995b, Predicting mechanical properties of acrylonitrile-butadiene-styrene terpolymer in injection molded plaque and box, Polym. Eng. Sci., 35, 1252-1259. 
• Konishi M., 1993, Development of intelligent processes in iron and steel industry, Met. Techn. (Japan), 63, 11-15.
• Larkiola J., Myllykoski P., Nylander J., Korhonen A.S., 1996, Prediction of rolling force in cold rolling by using physical models and neural computing, J. Mat. Proc. Techn., 60, 381-386.
• Lee B.H., Kim B.H., 1995, Automated selection of gate location based on desired quality of injection molded part, ANTEC '95. Vol. I-Processing, Boston, 554-560.
• Lee, L.J., Twu J.T., 1995, Application of artificial neural networks for the optimal design of sheet molding compound (SMC) compression molding, Polym. Compos., 16, 400-408.
• Liao T.W., Chen L.J., 1994, A neural network approach for grin¬ding processes: modelling and optimization. Int. J. Mach. Tools Manuf, 34, 919-937.
• Matsuda K. i in., 1994, Sintering process control using fuzzy inference and neural network. 6th Int. Symp. on Agglomeration, Nagoya, 421-426.
• Middle J.E., Khalaf, G.H., 1995, Neural network modeling of temperature distribution for control of gas metal are welding. Modeling of Casting, Welding and Advanced Solidification Processes VII, London, 397-405.
• Myllykoski P., Larkiola J., Nylander J., 1996, Development of prediction model for mechanical properties of batch annealed thin steel strip by using artificial neural network modelling, J. Mat. Proc. Techn., 60, 399-404.
• Payne R.D., Rebis R.E., Moran A.L., 1993, Spray forming quality predictions via neural networks, J. Mat. Eng. Perform., 2, 693-701.
• Rademan J.A.M. i in., 1996, Neural net based knowledge extraction from the historical data of an industrial leaching process, Hydrometallurgy, 43, 95-116.
• Rao H.S., Mukherjee A., 1996, Artificial neural networks for predicting the macromechanical behaviour of ceramic-matrix composites, Comp. Mat. Sci., 5, 307-322.
• Rojas I., Pomares H., Gonzalez J., Ros E., Salmeron M., Ortega J., Prieto A., 2000, A new radial basis function networks structure: application to time series prediction, Proc. 1EEE- INNS-ENNS Int. Joint Conf. on Neural Networks, IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium, IEEE Comput. Soc., 4, 449¬454.
• Samways N.L., 1996, In-line gage enhances galvannealing, Iron Steel Eng., 73, 169-170.
• Sayeh M.R., Viswanathan R., Dhali S.K., 1990, Neural networks for the assessment of impact on composite materials, Composite-Technology, 6th Annual Conf, Carbondale, 77-86.
• Sergi V., 1996, A neural network for optimization of laser welding of sheet steel, LAM I ERA, 33, 72-77.
• Shen Y., Moon S., 1996, Error compensation of coordinate measurements in computer-integrated manufacturing using neural networks, J. Mat. Proc. Techn., 61, 12-17.
• Slattery K.T., 1995, Randon-damage finite element for studying micromechanical failure mechanisms in advanced composites, J. Thermoplast. Compos. Mater., 8, 260-271.
• Stone R., Krishnamurthy K., 1996, A neural network thrust force controller to minimize delamination during drilling of graphite-epoxy laminates, Int. J. Mach. Tools Manuf, 36,985¬1003.
• Tadeusiewicz R., 1993, Sieci Neuronowe, Akademicka Oficyna Wydawnicza, Warszawa.
• Tadeusiewicz R., 1997, Możliwości i problemy wykorzystania sieci neuronowych w prognozowaniu procesów gospodarczych, Zeszyty Naukowe Akademii Ekonomicznej w Krakowie, nr 493, 5-18.
• Tadeusiewicz R., 1998, Badanie funkcjonowania mózgu za po-mocą sieci neuronowych, w pracy zbiorowej: Kognitywi- styka i Media w Edukacji, 1, 169 - 203.
• Tadeusiewicz R., Izworski A., Wszołek W., Wszołek T., 1999, Processing and Classification of Deformed Speech Using Neural Networks, w: Fouke J.M., Nerem R.M., Blanchard S.M., Yoganathan A.P., (eds.): Serving Humanity, Advancing Technology, Proc. 1st Joint BMES/EMBS Conf (IEEE Catalog Number 99CH37015C), Atlanta, 927-928.
• Tadeusiewicz R., Wszołek T., Izworski A., Wszołek W., 2000, Recognition of defects in High Voltage Transmission Lines Using the Acoustic Signal of Corona Effect, in: Widrow B. (i in.. eds.): Neural Networks for Signal Processing X, IEEE, New York, 869-875.
• Teshima T. i in., 1993, Estimation of cutting tool life by proces-sing tool image data with neural network, Ann. CIRP, Manuf. Techn., 42, 59-62.
• Thavasimuthu M., Rajagopalan C., Kalyanasundaram P., Raj B., 1996, Improving the evaluation sensitivity of an ultrasonic pulse echo technique using a neural network classifier, NDT&E Int., 29, 175-179.
• Torres V.M. Chaves A.P. Meech J.A., 2000, Intelligold-an expert system for gold plant process design, Cybernetics & Sys-tems, 31, 591-610.
• Walker J.L., Hill E.v.K., 1994, Back propagation neural networks for predicting ultimate strengths of unidirectional graphi¬te/epoxy tensile specimens, NDE: An Answer for an Aging World, New Orleans, 158-160.
• Workman G.L., Walker J.L., 1996, Materials characterization of powder metallurgy products using acousto-ultrasonics [and neural nets]. Mat. Sci. Forum, 210-213, 679-86.
• Xia Z.N., Lai S.G., Sun Y.Z., Lu Y.W., 1996, Study on property prediction for sealing alloys. Acta Metall. Sinica (English Letters), 9, 307-309.
• Yo K., Okigata K., 1993, Application of expert systems in metal working processes. Met. Techn. (Japan), 63, 38-43.
• Zgonc K., Achenbach J.D., 1996, A neural network for crack sizing trained by finite element calculations, NDT&E Int., 29, 147-155.