Neuronski model performansi hladne kočnice motornih vozila
A neural model of automotive cold brake performance
Апстракт
Performanse kočnica motornih vozila su rezultat složenih međusobno povezanih fenomena koji se javljaju u kontaktu frikcionog para. Ovi složeni kočni fenomeni su uglavnom određeni fizičko-mehaničkim osobinama sirovina frikcionog materijala, uslovima njegove proizvodnje i radnim uslovima kočnice. Uspostavljanje analitičkih modela kočnih performansi je vrlo teško gotovo nemoguće, usled složenih i izraženo nelinearnih fenomena koji se javljaju u toku procesa kočenja. Zbog toga su u ovom radu obuhvaćeni svi relevantni uticaji na performanse hladne kočnice pomoću veštačkih neuronskih mreža. Uticaji 26 ulaznih parametara određeni sastavom frikcionog materijala (18 sirovina), njegovim proizvodnim uslovima (5 parametara) i radnim uslovima kočnice (3 parametra) su modelirani u odnosu na promenu C karakteristike kočnice. Neuronski model performansi hladne kočnice je razvijen na osnovu obuke i testiranja 90 različitih neuronskih modela. Ovi neuronski modeli su dobijeni obukom 18 različitih arhitek...tura neuronskih mreža koje su obučavane sa pet algoritama učenja.
The automotive brake's performance results from the complex interrelated phenomena occurring in the contact of the friction pair. These complex braking phenomena are mostly affected by the physicochemical properties of friction materials’ ingredients, its manufacturing conditions, and brake’s operation regimes. Analytical models of brakes performance are difficult even impossible to obtain due to complex and highly nonlinear phenomena involved during braking. That is why in this paper all relevant influences on the cold brake performance have been integrated by means of artificial neural networks. The influences of 26 input parameters defined by the friction material composition (18 ingredients) its manufacturing conditions (5 parameters') and brake's operation regimes (3 parameters) have been modeled versus changes of the brake factor C. The neural model of the cold brake performance has been developed by training and testing of 90 neural models. These neural models have been obtained... by training of 18 different architectures of neural networks with the five learning algorithms.
Кључне речи:
neural model / cold brake / brake performanceИзвор:
FME Transactions, 2007, 35, 1, 9-14Издавач:
- Univerzitet u Beogradu - Mašinski fakultet, Beograd
Колекције
Институција/група
Mašinski fakultetTY - JOUR AU - Aleksendrić, Dragan AU - Duboka, Čedomir PY - 2007 UR - https://machinery.mas.bg.ac.rs/handle/123456789/727 AB - Performanse kočnica motornih vozila su rezultat složenih međusobno povezanih fenomena koji se javljaju u kontaktu frikcionog para. Ovi složeni kočni fenomeni su uglavnom određeni fizičko-mehaničkim osobinama sirovina frikcionog materijala, uslovima njegove proizvodnje i radnim uslovima kočnice. Uspostavljanje analitičkih modela kočnih performansi je vrlo teško gotovo nemoguće, usled složenih i izraženo nelinearnih fenomena koji se javljaju u toku procesa kočenja. Zbog toga su u ovom radu obuhvaćeni svi relevantni uticaji na performanse hladne kočnice pomoću veštačkih neuronskih mreža. Uticaji 26 ulaznih parametara određeni sastavom frikcionog materijala (18 sirovina), njegovim proizvodnim uslovima (5 parametara) i radnim uslovima kočnice (3 parametra) su modelirani u odnosu na promenu C karakteristike kočnice. Neuronski model performansi hladne kočnice je razvijen na osnovu obuke i testiranja 90 različitih neuronskih modela. Ovi neuronski modeli su dobijeni obukom 18 različitih arhitektura neuronskih mreža koje su obučavane sa pet algoritama učenja. AB - The automotive brake's performance results from the complex interrelated phenomena occurring in the contact of the friction pair. These complex braking phenomena are mostly affected by the physicochemical properties of friction materials’ ingredients, its manufacturing conditions, and brake’s operation regimes. Analytical models of brakes performance are difficult even impossible to obtain due to complex and highly nonlinear phenomena involved during braking. That is why in this paper all relevant influences on the cold brake performance have been integrated by means of artificial neural networks. The influences of 26 input parameters defined by the friction material composition (18 ingredients) its manufacturing conditions (5 parameters') and brake's operation regimes (3 parameters) have been modeled versus changes of the brake factor C. The neural model of the cold brake performance has been developed by training and testing of 90 neural models. These neural models have been obtained by training of 18 different architectures of neural networks with the five learning algorithms. PB - Univerzitet u Beogradu - Mašinski fakultet, Beograd T2 - FME Transactions T1 - Neuronski model performansi hladne kočnice motornih vozila T1 - A neural model of automotive cold brake performance EP - 14 IS - 1 SP - 9 VL - 35 UR - https://hdl.handle.net/21.15107/rcub_machinery_727 ER -
@article{ author = "Aleksendrić, Dragan and Duboka, Čedomir", year = "2007", abstract = "Performanse kočnica motornih vozila su rezultat složenih međusobno povezanih fenomena koji se javljaju u kontaktu frikcionog para. Ovi složeni kočni fenomeni su uglavnom određeni fizičko-mehaničkim osobinama sirovina frikcionog materijala, uslovima njegove proizvodnje i radnim uslovima kočnice. Uspostavljanje analitičkih modela kočnih performansi je vrlo teško gotovo nemoguće, usled složenih i izraženo nelinearnih fenomena koji se javljaju u toku procesa kočenja. Zbog toga su u ovom radu obuhvaćeni svi relevantni uticaji na performanse hladne kočnice pomoću veštačkih neuronskih mreža. Uticaji 26 ulaznih parametara određeni sastavom frikcionog materijala (18 sirovina), njegovim proizvodnim uslovima (5 parametara) i radnim uslovima kočnice (3 parametra) su modelirani u odnosu na promenu C karakteristike kočnice. Neuronski model performansi hladne kočnice je razvijen na osnovu obuke i testiranja 90 različitih neuronskih modela. Ovi neuronski modeli su dobijeni obukom 18 različitih arhitektura neuronskih mreža koje su obučavane sa pet algoritama učenja., The automotive brake's performance results from the complex interrelated phenomena occurring in the contact of the friction pair. These complex braking phenomena are mostly affected by the physicochemical properties of friction materials’ ingredients, its manufacturing conditions, and brake’s operation regimes. Analytical models of brakes performance are difficult even impossible to obtain due to complex and highly nonlinear phenomena involved during braking. That is why in this paper all relevant influences on the cold brake performance have been integrated by means of artificial neural networks. The influences of 26 input parameters defined by the friction material composition (18 ingredients) its manufacturing conditions (5 parameters') and brake's operation regimes (3 parameters) have been modeled versus changes of the brake factor C. The neural model of the cold brake performance has been developed by training and testing of 90 neural models. These neural models have been obtained by training of 18 different architectures of neural networks with the five learning algorithms.", publisher = "Univerzitet u Beogradu - Mašinski fakultet, Beograd", journal = "FME Transactions", title = "Neuronski model performansi hladne kočnice motornih vozila, A neural model of automotive cold brake performance", pages = "14-9", number = "1", volume = "35", url = "https://hdl.handle.net/21.15107/rcub_machinery_727" }
Aleksendrić, D.,& Duboka, Č.. (2007). Neuronski model performansi hladne kočnice motornih vozila. in FME Transactions Univerzitet u Beogradu - Mašinski fakultet, Beograd., 35(1), 9-14. https://hdl.handle.net/21.15107/rcub_machinery_727
Aleksendrić D, Duboka Č. Neuronski model performansi hladne kočnice motornih vozila. in FME Transactions. 2007;35(1):9-14. https://hdl.handle.net/21.15107/rcub_machinery_727 .
Aleksendrić, Dragan, Duboka, Čedomir, "Neuronski model performansi hladne kočnice motornih vozila" in FME Transactions, 35, no. 1 (2007):9-14, https://hdl.handle.net/21.15107/rcub_machinery_727 .