论文部分内容阅读
为解决黄河冰层厚度测量的误差大、精度低等问题,设计了一套针对黄河冰情特点的钻式冰层厚度自动测量系统,并研究了基于径向基神经网络的非线性数据融合处理技术.该测量系统首先通过电机(或破冰机)推动钻头在冰上钻孔,用压力传感器来判断钻孔动作的开始和停止;同时用钻头来带动编码器旋转,进行脉冲计数;当钻孔动作结束后,用钻头钻行速度和启停时间来计算冰层厚度,同时也通过编码器脉冲计数值来计算冰层厚度;最后以高斯基函数建立二元径向基的数学模型,将启停时间和脉冲计数值进行融合处理,得出冰厚测量结果.测量结果可在现场显示,也可通过GPRS无线传输到观测站.该测量系统不仅减少了人工操作带来的误差,而且克服了黄河水中杂质多、含沙量大等因素的影响,提高了系统的测量精度和稳定性,较适合在黄河及黄河流域库区冰厚测量进行推广使用.
In order to solve the problem of large error and low precision of ice thickness measurement in the Yellow River, a set of automatic measurement system of ice thickness for drilling ice is designed, and the nonlinear data fusion based on RBF neural network Technology.The measurement system first drives the drill bit to drill on the ice through the motor (or ice-breaker), and uses the pressure sensor to judge the start and stop of the drilling action; at the same time, the drill is used to drive the encoder to rotate for pulse counting; After the action, drill speed and start-stop time are used to calculate the thickness of ice layer, and the ice thickness is also calculated by encoder pulse count value. Finally, the mathematical model of binary radial basis is established with Gauss basis function, Stop time and pulse count value fusion processing, obtained ice thickness measurement results can be displayed in the field measurements can also be wirelessly transmitted to the GPRS observation station through the measurement system not only reduces the error caused by manual operation, but also to overcome Yellow River water impurities, a large amount of sediment and other factors, to improve the measurement accuracy and stability of the system, more suitable for the Yellow River and the Yellow River Basin ice thickness measurement push Use.