应用离散粒子群算法的大距离超声信号LMS自适应时延估计  

Least Mean Square Time Delay Estimation of Long Distance Ultrasonic Echoes based on Discrete Particle Swarm Optimization Algorithm

作　　者：王柯渊 谷立臣[1] 寇雪芹[1] 郭佳[1] 

机构地区：[1]西安建筑科技大学机械电子研究所,西安710055

出　　处：《机械科学与技术》2017年第7期1099-1104,共6页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家自然科学基金项目(50975218);陕西省教育厅专项基金项目(2013JK1011)资助

摘　　要：针对塔机安全预警系统中远距离障碍物检测对精度和实时性的需求,以超声回波时延估计为研究对象,提出了一种基于离散粒子群算法(DPSO)的最小均方误差自适应时延估计(LMSTDE)方法。该方法引入DPSO进行LMSTDE的寻优规划,解决了LMSTDE计算量庞大的问题;通过引入步长可变的LMSTDE算法和加速因子可变的DPSO,解决了算法过早收敛易陷入局部最优的问题。实验对比表明:改进后的算法保留了原有算法的高精度及抗噪性强等优点,且运算速度提升了25倍左右,可以实现中远距离障碍物的实时检测,且可靠性较高。In order to meet the need of accuracy and real-time in long distance obstacle detection of tower crane warning system, a new least mean square adaptive time delay estimation （LMSTDE） in ultrasonic echo time delay estimation is proposed, which is based on discrete particle swarm optimization（DPSO） algorithm. The method can reduce computation amount greatly with DPOS and overcome immature constringency in the optimization algorithm with variable step-size LMS and variable acceleration coefficients. The experimental results show that： this method not only keeps the high accuracy and good anti-noise ability of the original algorithm, but also increases the computation speed by 25 times. It can be used in the real time detection of middle and long distance obstacles with higher reliability.

关 键 词：塔式起重机 超声波测距 实时性 自适应时延估计 离散粒子群算法 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]