论文部分内容阅读
为了实现对煤矿瓦斯浓度的准确预测,提出了一种基于改进的人工蜂群算法(Artificial Bee Colony Algorithm,ABC)优化广义回归神经网络(General Regression Neural Network,GRNN)的瓦斯浓度预测方法。将基于无线传感器网络的瓦斯监控系统收集到的数据作为原始样本,经过具有伸缩性的自适应阈值函数对小波系数进行修正的方式对其进行去噪滤波预处理。采用ABC算法对GRNN网络参数进行优化,建立了瓦斯浓度的预测模型,并通过Matlab使用相关数据来训练和测试该模型的性能。仿真研究表明,所建模型能很快找到合适的平滑参数并对瓦斯浓度进行有效地预测。与其他预测模型相比较,该模型的预测精度更高,泛化能力更强,有较高的实用价值。
In order to accurately predict the gas concentration in coal mine, a gas concentration prediction method based on the improved Artificial Bee Colony Algorithm (ABC) for generalized regression neural network (GRNN) is proposed. The data collected by the gas monitoring system based on wireless sensor network is taken as the original sample, and the filter coefficients are preprocessed by adaptive scalar adaptive threshold function to modify the wavelet coefficients. The ABC algorithm is used to optimize the parameters of GRNN network, and the prediction model of gas concentration is established. The relevant data are used to train and test the performance of the model. The simulation results show that the model can find suitable smoothing parameters and predict the gas concentration effectively. Compared with other prediction models, this model has higher prediction accuracy, more generalization ability and higher practical value.