【摘要】 传统径向基神经网络在全局搜索能力和收敛速度上存在不足,本文将遗传算法和模拟退火算法相结合,利用构造的GA-SA混合算法优化RBF网络结构和权值参数,通过对兖州矿业集团济三煤矿提升机故障数据进行的仿真结果表明,优化网络在有效防止传统网络陷入局部最优的同时显著提高了网络收敛速度和搜索效率,网络故障诊断速度和精度均得到显著提高,为矿井提升机故障诊断提供了一种新方法。更多还原

【Abstract】 Traditional RBF neural network in the global search and convergence speed deficiencies,this paper genetic algorithms and simulated annealing algorithm combined to form a hybrid algorithm(GA-SA) optimization of RBF network structure and weight parameters.Optimized network algorithm in the effectively prevent local optimum at the same time improve the convergence speed and search efficiency.Yanzhou Mining Group on Mine Hoist Jisan fault simulation experiments show that the optimized network speed and accuracy of diagnosis are improved remarkably,with a high practical value for fault diagnosis of mine hoist provides a new approach.更多还原