摘要
提出一种改进粒子群算法和反向传播算法相结合的混合算法进行煤岩识别的方法。改进的粒子群算法提高了粒子群的联系,修正个体行动策略,加快局部搜索速度,保证搜索的全局有效性。研究结果表明,该煤岩识别方法不仅提高识别率,而且收敛速度和计算精度均有较大的改善,是一种有效和可行的煤岩识别方法。
A new method of coal-rock interface identification was proposed, which was based on a neural network trained by a hybrid algorithm combining PSO （Particle Swarm Optimization） algorithm with BP （Back Propagation） algorithm. In the improved PSO algorithm, the connection of each particle was reinforced, the individual movement strategy was corrected, the speed of local search was quickened and the effectiveness of global search was guaranteed. Study results showed that the method could not only improve the identification efficiency, but also greatly increased the convergence speed and computing precision. It was an effective and feasible method of coal-rock interface identification.
出处
《矿山机械》
北大核心
2013年第6期12-15,共4页
Mining & Processing Equipment
基金
宿迁市教改基金(2011YJG19)
