BP neural network based GPSA used in tandem cold rolling force prediction

This paper established a back propagation (BP) neural network tandem cold rolling force prediction model, and optimized by genetic particle swarm algorithm (GPSA). Genetic particle swarm algorithm has the advantage of both genetic algorithm (GA) and particle swarm algorithm (PSO) algorithm, integrates global searching ability with high convergence speed. Taking neural network weights and threshold values as independent variables, and neural network prediction error as target function, through GPSA operations, and find out the prediction error global minimum, then the corresponding weights and thresholds are used as the initial weights and thresholds of neural network train the neural network determine the neural network model of rolling force with highest forecast accuracy. Using field data of some tandem cold rolling mill, the off-line computation result showed that this method has better convergence speed and prevent into the local optimal value, can be used in practice as a new method for tandem cold rolling force prediction.