基于耦合算法的煤矿瓦斯涌出量预测模型研究

摘要

A new CIPSO-ENN（Chaos Immune Particle Swarm Optimization-Elman Neural Network）coupling algorithm was presented to predict gas emission in the view of the coal mine characteristics such as complicated,time varying and nonlinear etc.CIPSO algorithm was merged with dynamic feedback Elman neural network to optimize weight and threshold.The network model of gas emission quantity prediction was established by CIPSO-ENN,with the historical data of mine actual monitoring to experiments.The results show that this model has faster convergence rate,higher prediction accuracy and robust characteristics than the other models.%针对煤矿回采工作面瓦斯涌出量系统的时变性、非线性、复杂性、不确定性等特点,提出了混沌免疫粒子群算法（CIPSO）与Elman神经网络的耦合算法（CIPSO-ENN）用于非线性动态绝对瓦斯涌出量预测。算法通过实时的对其权值、阈值寻优,建立了基于CIPSO和ENN的耦合算法的绝对瓦斯涌出量预测系统模型,并利用矿井监测到的各项历史数据进行试验,结果表明该模型较其他预测模型其辨识收敛速度、预测精度和鲁棒性等性能都有明显的提高。