煤与瓦斯突出强度预测的IGABP方法

针对传统松散式(Genetic Algorithm Based Back Propagation Neural Network,GABP)模型应用于复杂煤与瓦斯突出预测时,存在GA自身性能及模型间相对孤立等不足,提出二者优势互补的IGABP一体化模型。IGABP首先在自适应交叉、变异率等方面进行改进,以提高GA自身的性能。其次,将BP导向性训练以算子的形式引入到标准GA进化过程,实现了GA寻优搜索的随机性向自主导向性转变。BP对GA搜索到的近似最优值进行微调,GA算法的收敛速度得到提升,确定精确解的位置能力加强,同时,又避免了单一BP网络本论文易陷入局部极小值的缺点,实现了两者一体化结合。仿真实验表明,构造出的进化神经网络更能反映煤与瓦斯突出强度样本的复杂非线性关系,有效克服了传统模型的不足,其运行效率、预测精度、可靠性等方面均优于传统模型,为瓦斯智能预测提供了新的解决方案。

Prediction of coal and gas outburst intensity with Incorporate Genetic Algorithm Based Back Propagation Neural Network(IGABP)

For the prediction of coal and gas outburst intensity,Incorporate Genetic Algorithm Based Back Propagation Neural Network(IGABP)is proposed to solve the limitations in the traditional GABP such as time-consuming,optimal stop condition of GA pretreatment indeterminacy,independency and complex task of great importance etc.IGABP addresses some improvements in adaptive crossover and mutation probability to promote GA performance.And with the introduction of BP operator in the evolution of GA operations,the stan...