Swarm Optimization Improved BP Algorithm for Microchannel Resistance Factor

In this paper, a new swarm optimization improved BP (Back Propagation) algorithm, combination of PSE (Particle Swarm Evolution) and BP, called PSE-BP algorithm, is introduced to train ANN (Artificial Neural Network) for the purpose of microchananel resistance factor prediction. The PSE algorithm was firstly proposed by comprehensively learning the principle of gradient descent, genetic algorithm and particle swarm optimization. Then, the search capability of PSE was analyzed by a standard cost function. Its appropriate control parameters were also determined by the same time. By utilizing the global search ability and high search efficiency of PSE algorithm, the improved BP (PSE-BP) algorithm was proposed. The training efficiency of PSE-BP algorithm was tested and compared with BP algorithm using Iris dataset. Finally, the resistance factor of rectangular cross-section microchannel was established using ANN, and trained with PSE-BP and BP algorithm, respectively. The results show that the PSE-BP algorithm can greatly improve the training efficiency of ANN, compare with BP algorithm. And the microchananel resistance factors, which predicted by ANN model and trained with PSE-BP algorithm, are in good agreement with the simulation samples.

Abstract licence: CC BY-NC-ND

35 more references, whose wording the paper’s licence keeps private.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.