A study on on-line learning neural network for prediction for rolling force in hot-rolling mill

Using artificial neural networks to represent a diesel–biodiesel engine

“…For the last few years, NNs have been successfully applied in diverse fields such as finance, medicine, engineering, geology, and physics. The application of NNs in several industrial processes can be found in the extensive literature ( [42][43][44]46], among others). In [19,22,29], NNs were proposed as an efficient model to predict mechanical properties in industrial processes.…”

Prediction model of rolling force for electrical steel based on finite element method and neural network

“…The model of rolling force is a key model in rolling process [1][2].The force can be calculated through mathematic model determined by the plastic mechanics with lots of difficulties in getting accurate solution in plastic deformation zone [3]; the models can also be built by FEA methods for hot and cold rolling process [4][5][6][7] , unsuitable for on-line prediction because of plenty of running time. With the advantages of selflearning and the fast computing speed, the artificial neural network (ANN) is widely used [8]: Li [9] predicted rolling forces of each finishing mills using BP-ANN by the measured rolling data; Joon-Sik Son [10] use an online self-learning ANN to predict the rolling force in the hot rolling process; Zhang [11] use RBF-ANN method to forecast the strip yielding stress in order to improve the precision of the rolling force model. The ANN has been extend to the predication of temperature field, stress field and strain field in rolling process by Shahani [12].…”

A NR–BFGS method for fast rigid-plastic FEM in strip rolling

“…With the development of computer technology, the artificial intelligence methods including fuzzy theory and control [3][4], neural networks [5][6] and expert system [7] had been discussed and used for predicting the rolling parameters and mechanical properties of the rolled material during hot strip rolling. However, the credibility and successful application of neural network must be based on the accuracy and amount of measured learning data, method and a good understanding process [8].…”