中国机械工程

• 智能制造 • 上一篇    下一篇

基于超声时序神经网络目标识别的塔机安全预警

李西平;谷立臣;寇雪芹   

  1. 西安建筑科技大学,西安,710055
  • 出版日期:2016-08-25 发布日期:2016-08-17
  • 基金资助:
    国家自然科学基金资助项目(50975218);陕西省教育厅专项基金资助项目(2013JK1011);陕西省重点实验室开放基金资助项目(2014G1502043)

Target Recognition of Tower Crane Safety Pre-warning Based on Ultrasound Timing Sequence Neural Network

Li Xiping;Gu Lichen;Kou Xueqin   

  1. Xi'an University of Architecture and Technology, Xi'an, 710055
  • Online:2016-08-25 Published:2016-08-17
  • Supported by:

摘要: 为实现塔机失稳监测和防碰撞的安全预警功能,改善塔机被动安全模型存在的低成本、主动性、灵活性、快速性和及时性的不足,通过分析目标物特性与超声时序信号相关特征及测距值特征的关系,结合Elman、SOM网络,构建了基于超声时序神经网络目标识别的塔机安全预警系统,可以实现对扭转角和障碍物的信号采集、数据融合、主动预警功能,试验结果表明该系统可以达到预期的低成本、高速度、高精度的塔机工作要求。

关键词: 神经网络, 塔机;目标识别;安全预警;超声时序

Abstract: Aiming at tower crane safety pre-warning of instability and collision prevention, to improve the passive security model of tower crane which lacked low cost, initiative, flexibility, rapidity and simultaneity, the relationships among the target characteristics, ultrasound timing sequence relevent characteristics and distance characteristics were analyzed. Combined with Elman and SOM network, a system of target recognition of tower crane safety pre-warning was developed based on ultrasound timing sequence neural network. The functions such as sampling, data fusion, initiative pre-warning of twist angles and obstacles were achieved. Experimental results verify that the system can satisfy the tower crane working requirements with low cost, high speed and high precision.

Key words: neural network, tower crane, target recognition, safety pre-warning, ultrasound timing sequence

中图分类号: