基于遗传算法优化神经网络瓦斯浓度预测

Prediction of coal gas density by optimized neural network based on genetic algorithm

针对煤矿安全生产,将遗传神经网络应用到煤矿瓦斯浓度预测中,BP神经网络初始权值和阈值通过遗传算法优化。用MATLAB进行仿真测试,并与单纯使用BP网络预测进行了对比。结果表明,优化后的网络具有训练时间短、精度高等特点,对瓦斯浓度的预测有效。

In order to ensure safe production in collieries, the genetic neural network was applied to the prediction of the coal gas density, and initial weight and threshold of BP neural network were optimized by the genetic algorithm. MALTAB was applied to simulation, which was contrasted with the prediction by BP network. It was showed that the optimized network was characterized by short training duration, high precision and effectiveness in prediction of coal gas densitv.