Abstract
Anti-jamming system has been widely used in rock drill rigs, in order to avoid uncontrolled jamming of the bit. An anti-jamming valve is the core component of the anti-jamming system. The backpropagation neural network (BPNN) was used in this study to construct an estimating model for the prediction of feed pressure, so as to find the key factors of jamming problems. Because the parameters of BPNN have a significant influence on results, and genetic algorithm (GA) is capable of quickly finding optimal solutions. We integrated GA and BPNN so that the convergence rate was improved and precision was relatively enhanced. The results showed that, BPNN realised a reasonable prediction for the feed pressure of anti-jamming valve. Moreover, when the parameters for GA are set between reasonable ranges, GA optimisation for such neural network is feasible, and the obtained results are satisfying, which has relatively higher accuracy of prediction, non-linear mapping and better network performance.
Disclosure statement
No potential conflict of interest was reported by the authors.